[kernel] r11453 - in dists/sid/linux-2.6/debian/patches: features/all/vserver series
Bastian Blank
waldi at alioth.debian.org
Sun May 25 09:16:20 UTC 2008
Author: waldi
Date: Sun May 25 09:16:17 2008
New Revision: 11453
Log:
* debian/patches/series/1-extra: Remove old patches.
* debian/patches/series/4-extra: Add new patches.
* debian/patches/features/all/vserver/bindmount-dev.patch: Fix.
* debian/patches/features/all/vserver/vs2.2.0-rc5.patch: Remove.
* debian/patches/features/all/vserver/vs2.3.0.34.10.patch: Add.
Added:
dists/sid/linux-2.6/debian/patches/features/all/vserver/vs2.3.0.34.10.patch
- copied, changed from r11449, /dists/sid/linux-2.6/debian/patches/features/all/vserver/vs2.2.0-rc5.patch
dists/sid/linux-2.6/debian/patches/series/4-extra
Removed:
dists/sid/linux-2.6/debian/patches/features/all/vserver/vs2.2.0-rc5.patch
Modified:
dists/sid/linux-2.6/debian/patches/features/all/vserver/bindmount-dev.patch
dists/sid/linux-2.6/debian/patches/series/1-extra
Modified: dists/sid/linux-2.6/debian/patches/features/all/vserver/bindmount-dev.patch
==============================================================================
--- dists/sid/linux-2.6/debian/patches/features/all/vserver/bindmount-dev.patch (original)
+++ dists/sid/linux-2.6/debian/patches/features/all/vserver/bindmount-dev.patch Sun May 25 09:16:17 2008
@@ -4,7 +4,7 @@
if (!mnt)
goto out;
-+ if (!capable(CAP_SYS_ADMIN) && (old_nd.mnt->mnt_flags & MNT_NODEV))
++ if (!capable(CAP_SYS_ADMIN) && (old_nd.path.mnt->mnt_flags & MNT_NODEV))
+ mnt_flags |= MNT_NODEV;
+
mnt->mnt_flags = mnt_flags;
@@ -37,5 +37,5 @@
- if (!capable(CAP_SYS_ADMIN))
- mnt_flags |= MNT_NODEV;
flags &= ~(MS_NOSUID | MS_NOEXEC | MS_NODEV | MS_ACTIVE |
- MS_NOATIME | MS_NODIRATIME | MS_RELATIME);
+ MS_NOATIME | MS_NODIRATIME | MS_RELATIME| MS_KERNMOUNT);
Copied: dists/sid/linux-2.6/debian/patches/features/all/vserver/vs2.3.0.34.10.patch (from r11449, /dists/sid/linux-2.6/debian/patches/features/all/vserver/vs2.2.0-rc5.patch)
==============================================================================
--- /dists/sid/linux-2.6/debian/patches/features/all/vserver/vs2.2.0-rc5.patch (original)
+++ dists/sid/linux-2.6/debian/patches/features/all/vserver/vs2.3.0.34.10.patch Sun May 25 09:16:17 2008
@@ -1,163 +1,6 @@
---- linux-2.6.22/Documentation/vserver/debug.txt 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/Documentation/vserver/debug.txt 2007-06-15 02:37:02 +0200
-@@ -0,0 +1,154 @@
-+
-+debug_cvirt:
-+
-+ 2 4 "vx_map_tgid: %p/%llx: %d -> %d"
-+ "vx_rmap_tgid: %p/%llx: %d -> %d"
-+
-+debug_dlim:
-+
-+ 0 1 "ALLOC (%p,#%d)%c inode (%d)"
-+ "FREE (%p,#%d)%c inode"
-+ 1 2 "ALLOC (%p,#%d)%c %lld bytes (%d)"
-+ "FREE (%p,#%d)%c %lld bytes"
-+ 2 4 "ADJUST: %lld,%lld on %ld,%ld [mult=%d]"
-+ 3 8 "ext3_has_free_blocks(%p): %lu<%lu+1, %c, %u!=%u r=%d"
-+ "ext3_has_free_blocks(%p): free=%lu, root=%lu"
-+ "rcu_free_dl_info(%p)"
-+ 4 10 "alloc_dl_info(%p,%d) = %p"
-+ "dealloc_dl_info(%p)"
-+ "get_dl_info(%p[#%d.%d])"
-+ "put_dl_info(%p[#%d.%d])"
-+ 5 20 "alloc_dl_info(%p,%d)*"
-+ 6 40 "__hash_dl_info: %p[#%d]"
-+ "__unhash_dl_info: %p[#%d]"
-+ 7 80 "locate_dl_info(%p,#%d) = %p"
-+
-+debug_misc:
-+
-+ 0 1 "destroy_dqhash: %p [#0x%08x] c=%d"
-+ "new_dqhash: %p [#0x%08x]"
-+ "vroot[%d]_clr_dev: dev=%p[%lu,%d:%d]"
-+ "vroot[%d]_get_real_bdev: dev=%p[%lu,%d:%d]"
-+ "vroot[%d]_set_dev: dev=%p[%lu,%d:%d]"
-+ "vroot_get_real_bdev not set"
-+ 1 2 "cow_break_link(»%s«)"
-+ "temp copy »%s«"
-+ 2 4 "dentry_open(new): %p"
-+ "dentry_open(old): %p"
-+ "lookup_create(new): %p"
-+ "old path »%s«"
-+ "path_lookup(old): %d"
-+ "vfs_create(new): %d"
-+ "vfs_rename: %d"
-+ "vfs_sendfile: %d"
-+ 3 8 "fput(new_file=%p[#%d])"
-+ "fput(old_file=%p[#%d])"
-+ 4 10 "vx_info_kill(%p[#%d],%d,%d) = %d"
-+ "vx_info_kill(%p[#%d],%d,%d)*"
-+ 5 20 "vs_reboot(%p[#%d],%d)"
-+ 6 40 "dropping task %p[#%u,%u] for %p[#%u,%u]"
-+
-+debug_net:
-+
-+ 2 4 "nx_addr_conflict(%p,%p) %d.%d,%d.%d"
-+ 3 8 "inet_bind(%p) %d.%d.%d.%d, %d.%d.%d.%d, %d.%d.%d.%d"
-+ "inet_bind(%p)* %p,%p;%lx %d.%d.%d.%d"
-+ 4 10 "ip_route_connect(%p) %p,%p;%lx"
-+ 5 20 "__addr_in_socket(%p,%d.%d.%d.%d) %p:%d.%d.%d.%d %p;%lx"
-+ 6 40 "sk,egf: %p [#%d] (from %d)"
-+ "sk,egn: %p [#%d] (from %d)"
-+ "sk,req: %p [#%d] (from %d)"
-+ "sk: %p [#%d] (from %d)"
-+ "tw: %p [#%d] (from %d)"
-+ 7 80 "__sock_recvmsg: %p[%p,%p,%p;%d]:%d/%d"
-+ "__sock_sendmsg: %p[%p,%p,%p;%d]:%d/%d"
-+
-+debug_nid:
-+
-+ 0 1 "__lookup_nx_info(#%u): %p[#%u]"
-+ "alloc_nx_info(%d) = %p"
-+ "create_nx_info(%d) (dynamic rejected)"
-+ "create_nx_info(%d) = %p (already there)"
-+ "create_nx_info(%d) = %p (new)"
-+ "dealloc_nx_info(%p)"
-+ 1 2 "alloc_nx_info(%d)*"
-+ "create_nx_info(%d)*"
-+ 2 4 "get_nx_info(%p[#%d.%d])"
-+ "put_nx_info(%p[#%d.%d])"
-+ 3 8 "claim_nx_info(%p[#%d.%d.%d]) %p"
-+ "clr_nx_info(%p[#%d.%d])"
-+ "init_nx_info(%p[#%d.%d])"
-+ "release_nx_info(%p[#%d.%d.%d]) %p"
-+ "set_nx_info(%p[#%d.%d])"
-+ 4 10 "__hash_nx_info: %p[#%d]"
-+ "__nx_dynamic_id: [#%d]"
-+ "__unhash_nx_info: %p[#%d.%d.%d]"
-+ 5 20 "moved task %p into nxi:%p[#%d]"
-+ "nx_migrate_task(%p,%p[#%d.%d.%d])"
-+ "task_get_nx_info(%p)"
-+ 6 40 "nx_clear_persistent(%p[#%d])"
-+
-+debug_quota:
-+
-+ 0 1 "quota_sync_dqh(%p,%d) discard inode %p"
-+ 1 2 "quota_sync_dqh(%p,%d)"
-+ "sync_dquots(%p,%d)"
-+ "sync_dquots_dqh(%p,%d)"
-+ 3 8 "do_quotactl(%p,%d,cmd=%d,id=%d,%p)"
-+
-+debug_switch:
-+
-+ 0 1 "vc: VCMD_%02d_%d[%d], %d,%p [%d,%d,%x,%x]"
-+ 1 2 "vc: VCMD_%02d_%d[%d] = %08lx(%ld) [%d,%d]"
-+ 4 10 "%s: (%s %s) returned %s with %d"
-+
-+debug_tag:
-+
-+ 7 80 "dx_parse_tag(»%s«): %d:#%d"
-+ "dx_propagate_tag(%p[#%lu.%d]): %d,%d"
-+
-+debug_xid:
-+
-+ 0 1 "__lookup_vx_info(#%u): %p[#%u]"
-+ "alloc_vx_info(%d) = %p"
-+ "alloc_vx_info(%d)*"
-+ "create_vx_info(%d) (dynamic rejected)"
-+ "create_vx_info(%d) = %p (already there)"
-+ "create_vx_info(%d) = %p (new)"
-+ "dealloc_vx_info(%p)"
-+ "loc_vx_info(%d) = %p (found)"
-+ "loc_vx_info(%d) = %p (new)"
-+ "loc_vx_info(%d) = %p (not available)"
-+ 1 2 "create_vx_info(%d)*"
-+ "loc_vx_info(%d)*"
-+ 2 4 "get_vx_info(%p[#%d.%d])"
-+ "put_vx_info(%p[#%d.%d])"
-+ 3 8 "claim_vx_info(%p[#%d.%d.%d]) %p"
-+ "clr_vx_info(%p[#%d.%d])"
-+ "init_vx_info(%p[#%d.%d])"
-+ "release_vx_info(%p[#%d.%d.%d]) %p"
-+ "set_vx_info(%p[#%d.%d])"
-+ 4 10 "__hash_vx_info: %p[#%d]"
-+ "__unhash_vx_info: %p[#%d.%d.%d]"
-+ "__vx_dynamic_id: [#%d]"
-+ 5 20 "enter_vx_info(%p[#%d],%p) %p[#%d,%p]"
-+ "leave_vx_info(%p[#%d,%p]) %p[#%d,%p]"
-+ "moved task %p into vxi:%p[#%d]"
-+ "task_get_vx_info(%p)"
-+ "vx_migrate_task(%p,%p[#%d.%d])"
-+ 6 40 "vx_clear_persistent(%p[#%d])"
-+ "vx_exit_init(%p[#%d],%p[#%d,%d,%d])"
-+ "vx_set_init(%p[#%d],%p[#%d,%d,%d])"
-+ "vx_set_persistent(%p[#%d])"
-+ "vx_set_reaper(%p[#%d],%p[#%d,%d])"
-+ 7 80 "vx_child_reaper(%p[#%u,%u]) = %p[#%u,%u]"
-+
-+
-+debug_limit:
-+
-+ n 2^n "vx_acc_cres[%5d,%s,%2d]: %5d%s"
-+ "vx_cres_avail[%5d,%s,%2d]: %5ld > %5d + %5d"
-+
-+ m 2^m "vx_acc_page[%5d,%s,%2d]: %5d%s"
-+ "vx_acc_pages[%5d,%s,%2d]: %5d += %5d"
-+ "vx_pages_avail[%5d,%s,%2d]: %5ld > %5d + %5d"
---- linux-2.6.22/arch/alpha/Kconfig 2007-07-09 13:17:31 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/Kconfig 2007-06-15 02:37:02 +0200
-@@ -662,6 +662,8 @@ config DUMMY_CONSOLE
+--- linux-2.6.25.4/arch/alpha/Kconfig 2008-04-17 12:05:26.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/alpha/Kconfig 2008-04-19 15:14:51.000000000 -0400
+@@ -671,6 +671,8 @@ config DUMMY_CONSOLE
depends on VGA_HOSE
default y
@@ -166,27 +9,8 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/alpha/kernel/asm-offsets.c 2006-02-15 13:54:10 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/kernel/asm-offsets.c 2007-06-15 02:37:02 +0200
-@@ -36,6 +36,7 @@ void foo(void)
- DEFINE(PT_PTRACED, PT_PTRACED);
- DEFINE(CLONE_VM, CLONE_VM);
- DEFINE(CLONE_UNTRACED, CLONE_UNTRACED);
-+ DEFINE(CLONE_KTHREAD, CLONE_KTHREAD);
- DEFINE(SIGCHLD, SIGCHLD);
- BLANK();
-
---- linux-2.6.22/arch/alpha/kernel/entry.S 2007-07-09 13:17:31 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/kernel/entry.S 2007-06-15 02:37:02 +0200
-@@ -643,7 +643,7 @@ kernel_thread:
- stq $2, 152($sp) /* HAE */
-
- /* Shuffle FLAGS to the front; add CLONE_VM. */
-- ldi $1, CLONE_VM|CLONE_UNTRACED
-+ ldi $1, CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD;
- or $18, $1, $16
- bsr $26, sys_clone
-
+--- linux-2.6.25.4/arch/alpha/kernel/entry.S 2008-04-17 11:31:21.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/alpha/kernel/entry.S 2008-04-21 11:09:01.000000000 -0400
@@ -872,24 +872,15 @@ sys_getxgid:
.globl sys_getxpid
.ent sys_getxpid
@@ -219,8 +43,8 @@
ret
.end sys_getxpid
---- linux-2.6.22/arch/alpha/kernel/osf_sys.c 2007-07-09 13:17:31 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/kernel/osf_sys.c 2007-06-15 02:37:02 +0200
+--- linux-2.6.25.4/arch/alpha/kernel/osf_sys.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/alpha/kernel/osf_sys.c 2008-05-21 14:30:40.000000000 -0400
@@ -883,7 +883,7 @@ osf_gettimeofday(struct timeval32 __user
{
if (tv) {
@@ -230,8 +54,8 @@
if (put_tv32(tv, &ktv))
return -EFAULT;
}
---- linux-2.6.22/arch/alpha/kernel/ptrace.c 2006-04-09 13:49:39 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/kernel/ptrace.c 2007-06-15 02:37:02 +0200
+--- linux-2.6.25.4/arch/alpha/kernel/ptrace.c 2008-04-17 11:31:21.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/alpha/kernel/ptrace.c 2008-04-19 15:14:51.000000000 -0400
@@ -15,6 +15,7 @@
#include <linux/slab.h>
#include <linux/security.h>
@@ -240,28 +64,16 @@
#include <asm/uaccess.h>
#include <asm/pgtable.h>
-@@ -283,6 +284,11 @@ do_sys_ptrace(long request, long pid, lo
- goto out_notsk;
- }
-
-+ if (!vx_check(vx_task_xid(child), VS_WATCH_P | VS_IDENT)) {
-+ ret = -EPERM;
-+ goto out;
-+ }
-+
- if (request == PTRACE_ATTACH) {
- ret = ptrace_attach(child);
- goto out;
---- linux-2.6.22/arch/alpha/kernel/semaphore.c 2004-08-14 12:55:32 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/kernel/semaphore.c 2007-06-15 02:37:02 +0200
+--- linux-2.6.25.4/arch/alpha/kernel/semaphore.c 2008-04-17 11:31:21.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/alpha/kernel/semaphore.c 2008-04-19 15:14:51.000000000 -0400
@@ -68,8 +68,8 @@ __down_failed(struct semaphore *sem)
DECLARE_WAITQUEUE(wait, tsk);
#ifdef CONFIG_DEBUG_SEMAPHORE
- printk("%s(%d): down failed(%p)\n",
-- tsk->comm, tsk->pid, sem);
+- tsk->comm, task_pid_nr(tsk), sem);
+ printk("%s(%d:#%u): down failed(%p)\n",
-+ tsk->comm, tsk->pid, tsk->xid, sem);
++ tsk->comm, task_pid_nr(tsk), tsk->xid, sem);
#endif
tsk->state = TASK_UNINTERRUPTIBLE;
@@ -270,9 +82,9 @@
#ifdef CONFIG_DEBUG_SEMAPHORE
- printk("%s(%d): down acquired(%p)\n",
-- tsk->comm, tsk->pid, sem);
+- tsk->comm, task_pid_nr(tsk), sem);
+ printk("%s(%d:#%u): down acquired(%p)\n",
-+ tsk->comm, tsk->pid, tsk->xid, sem);
++ tsk->comm, task_pid_nr(tsk), tsk->xid, sem);
#endif
}
@@ -281,14 +93,14 @@
#ifdef CONFIG_DEBUG_SEMAPHORE
- printk("%s(%d): down failed(%p)\n",
-- tsk->comm, tsk->pid, sem);
+- tsk->comm, task_pid_nr(tsk), sem);
+ printk("%s(%d:#%u): down failed(%p)\n",
-+ tsk->comm, tsk->pid, tsk->xid, sem);
++ tsk->comm, task_pid_nr(tsk), tsk->xid, sem);
#endif
tsk->state = TASK_INTERRUPTIBLE;
---- linux-2.6.22/arch/alpha/kernel/systbls.S 2007-07-09 13:17:31 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/kernel/systbls.S 2007-06-15 02:37:02 +0200
+--- linux-2.6.25.4/arch/alpha/kernel/systbls.S 2008-04-17 12:05:26.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/alpha/kernel/systbls.S 2008-04-19 15:14:51.000000000 -0400
@@ -446,7 +446,7 @@ sys_call_table:
.quad sys_stat64 /* 425 */
.quad sys_lstat64
@@ -298,34 +110,34 @@
.quad sys_ni_syscall /* sys_mbind */
.quad sys_ni_syscall /* sys_get_mempolicy */
.quad sys_ni_syscall /* sys_set_mempolicy */
---- linux-2.6.22/arch/alpha/kernel/traps.c 2006-09-20 16:57:57 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/kernel/traps.c 2007-06-15 02:37:02 +0200
+--- linux-2.6.25.4/arch/alpha/kernel/traps.c 2008-04-17 11:31:21.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/alpha/kernel/traps.c 2008-04-19 15:14:51.000000000 -0400
@@ -182,7 +182,8 @@ die_if_kernel(char * str, struct pt_regs
#ifdef CONFIG_SMP
printk("CPU %d ", hard_smp_processor_id());
#endif
-- printk("%s(%d): %s %ld\n", current->comm, current->pid, str, err);
+- printk("%s(%d): %s %ld\n", current->comm, task_pid_nr(current), str, err);
+ printk("%s(%d[#%u]): %s %ld\n", current->comm,
-+ current->pid, current->xid, str, err);
++ task_pid_nr(current), current->xid, str, err);
dik_show_regs(regs, r9_15);
+ add_taint(TAINT_DIE);
dik_show_trace((unsigned long *)(regs+1));
- dik_show_code((unsigned int *)regs->pc);
---- linux-2.6.22/arch/alpha/mm/fault.c 2007-07-09 13:17:32 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/alpha/mm/fault.c 2007-06-15 02:37:02 +0200
-@@ -197,8 +197,8 @@ do_page_fault(unsigned long address, uns
+--- linux-2.6.25.4/arch/alpha/mm/fault.c 2008-04-17 11:31:21.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/alpha/mm/fault.c 2008-04-19 15:14:51.000000000 -0400
+@@ -193,8 +193,8 @@ do_page_fault(unsigned long address, uns
down_read(&mm->mmap_sem);
goto survive;
}
- printk(KERN_ALERT "VM: killing process %s(%d)\n",
-- current->comm, current->pid);
+- current->comm, task_pid_nr(current));
+ printk(KERN_ALERT "VM: killing process %s(%d:#%u)\n",
-+ current->comm, current->pid, current->xid);
++ current->comm, task_pid_nr(current), current->xid);
if (!user_mode(regs))
goto no_context;
- do_exit(SIGKILL);
---- linux-2.6.22/arch/arm/Kconfig 2007-07-09 13:17:32 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm/Kconfig 2007-06-15 02:37:02 +0200
-@@ -1042,6 +1042,8 @@ source "arch/arm/oprofile/Kconfig"
+ do_group_exit(SIGKILL);
+--- linux-2.6.25.4/arch/arm/Kconfig 2008-04-17 12:05:26.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/arm/Kconfig 2008-04-19 15:14:51.000000000 -0400
+@@ -1180,6 +1180,8 @@ source "fs/Kconfig"
source "arch/arm/Kconfig.debug"
@@ -334,8 +146,8 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/arm/kernel/calls.S 2007-07-09 13:17:32 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm/kernel/calls.S 2007-07-07 03:52:53 +0200
+--- linux-2.6.25.4/arch/arm/kernel/calls.S 2008-04-17 12:05:26.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/arm/kernel/calls.S 2008-04-19 15:14:51.000000000 -0400
@@ -322,7 +322,7 @@
/* 310 */ CALL(sys_request_key)
CALL(sys_keyctl)
@@ -345,107 +157,46 @@
CALL(sys_ioprio_set)
/* 315 */ CALL(sys_ioprio_get)
CALL(sys_inotify_init)
---- linux-2.6.22/arch/arm/kernel/process.c 2007-07-09 13:17:32 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm/kernel/process.c 2007-07-07 03:52:53 +0200
-@@ -261,7 +261,8 @@ void __show_regs(struct pt_regs *regs)
+--- linux-2.6.25.4/arch/arm/kernel/process.c 2008-04-17 12:05:26.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/arm/kernel/process.c 2008-04-21 11:09:01.000000000 -0400
+@@ -264,7 +264,8 @@ void __show_regs(struct pt_regs *regs)
void show_regs(struct pt_regs * regs)
{
printk("\n");
-- printk("Pid: %d, comm: %20s\n", current->pid, current->comm);
+- printk("Pid: %d, comm: %20s\n", task_pid_nr(current), current->comm);
+ printk("Pid: %d[#%u], comm: %20s\n",
-+ current->pid, current->xid, current->comm);
++ task_pid_nr(current), current->xid, current->comm);
__show_regs(regs);
__backtrace();
}
-@@ -423,7 +424,8 @@ pid_t kernel_thread(int (*fn)(void *), v
- regs.ARM_pc = (unsigned long)kernel_thread_helper;
- regs.ARM_cpsr = SVC_MODE;
-
-- return do_fork(flags|CLONE_VM|CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s, 0, NULL, NULL);
- }
- EXPORT_SYMBOL(kernel_thread);
-
---- linux-2.6.22/arch/arm/kernel/traps.c 2007-07-09 13:17:32 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm/kernel/traps.c 2007-07-09 13:11:53 +0200
-@@ -222,8 +222,8 @@ static void __die(const char *str, int e
+--- linux-2.6.25.4/arch/arm/kernel/traps.c 2008-04-17 12:05:26.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/arm/kernel/traps.c 2008-04-19 15:14:51.000000000 -0400
+@@ -214,8 +214,8 @@ static void __die(const char *str, int e
str, err, ++die_counter);
print_modules();
__show_regs(regs);
- printk("Process %s (pid: %d, stack limit = 0x%p)\n",
-- tsk->comm, tsk->pid, thread + 1);
+- tsk->comm, task_pid_nr(tsk), thread + 1);
+ printk("Process %s (pid: %d:#%u, stack limit = 0x%p)\n",
-+ tsk->comm, tsk->pid, tsk->xid, thread + 1);
++ tsk->comm, task_pid_nr(tsk), tsk->xid, thread + 1);
if (!user_mode(regs) || in_interrupt()) {
dump_mem("Stack: ", regs->ARM_sp,
---- linux-2.6.22/arch/arm/mm/fault.c 2007-07-09 13:17:39 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm/mm/fault.c 2007-06-15 02:37:02 +0200
-@@ -266,7 +266,8 @@ do_page_fault(unsigned long addr, unsign
+--- linux-2.6.25.4/arch/arm/mm/fault.c 2008-04-17 12:05:27.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/arm/mm/fault.c 2008-04-19 15:14:51.000000000 -0400
+@@ -292,7 +292,8 @@ do_page_fault(unsigned long addr, unsign
* happened to us that made us unable to handle
* the page fault gracefully.
*/
- printk("VM: killing process %s\n", tsk->comm);
+ printk("VM: killing process %s(%d:#%u)\n",
-+ tsk->comm, tsk->pid, tsk->xid);
- do_exit(SIGKILL);
++ tsk->comm, task_pid_nr(tsk), tsk->xid);
+ do_group_exit(SIGKILL);
return 0;
-
---- linux-2.6.22/arch/arm26/Kconfig 2007-07-09 13:17:40 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm26/Kconfig 2007-06-15 02:37:02 +0200
-@@ -243,6 +243,8 @@ source "drivers/usb/Kconfig"
-
- source "arch/arm26/Kconfig.debug"
-
-+source "kernel/vserver/Kconfig"
-+
- source "security/Kconfig"
-
- source "crypto/Kconfig"
---- linux-2.6.22/arch/arm26/kernel/calls.S 2005-03-02 12:38:19 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm26/kernel/calls.S 2007-06-15 02:37:02 +0200
-@@ -257,6 +257,11 @@ __syscall_start:
- .long sys_lremovexattr
- .long sys_fremovexattr
- .long sys_tkill
-+
-+ .rept 313 - (. - __syscall_start) / 4
-+ .long sys_ni_syscall
-+ .endr
-+ .long sys_vserver /* 313 */
- __syscall_end:
-
- .rept NR_syscalls - (__syscall_end - __syscall_start) / 4
---- linux-2.6.22/arch/arm26/kernel/process.c 2006-09-20 16:57:57 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm26/kernel/process.c 2007-06-15 02:37:02 +0200
-@@ -365,7 +365,8 @@ pid_t kernel_thread(int (*fn)(void *), v
- regs.ARM_r3 = (unsigned long)do_exit;
- regs.ARM_pc = (unsigned long)kernel_thread_helper | MODE_SVC26;
-
-- return do_fork(flags|CLONE_VM|CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s, 0, NULL, NULL);
- }
- EXPORT_SYMBOL(kernel_thread);
-
---- linux-2.6.22/arch/arm26/kernel/traps.c 2006-09-20 16:57:57 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/arm26/kernel/traps.c 2007-06-15 02:37:02 +0200
-@@ -185,8 +185,9 @@ NORET_TYPE void die(const char *str, str
- printk("Internal error: %s: %x\n", str, err);
- printk("CPU: %d\n", smp_processor_id());
- show_regs(regs);
-- printk("Process %s (pid: %d, stack limit = 0x%p)\n",
-- current->comm, current->pid, end_of_stack(tsk));
-+ printk("Process %s (pid: %d[#%u], stack limit = 0x%p)\n",
-+ current->comm, current->pid,
-+ current->xid, end_of_stack(tsk));
-
- if (!user_mode(regs) || in_interrupt()) {
- __dump_stack(tsk, (unsigned long)(regs + 1));
---- linux-2.6.22/arch/cris/Kconfig 2007-05-02 19:24:19 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/cris/Kconfig 2007-06-15 02:37:02 +0200
-@@ -200,6 +200,8 @@ source "drivers/usb/Kconfig"
+ }
+--- linux-2.6.25.4/arch/cris/Kconfig 2008-04-17 12:05:27.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/cris/Kconfig 2008-04-19 15:14:51.000000000 -0400
+@@ -679,6 +679,8 @@ source "drivers/usb/Kconfig"
source "arch/cris/Kconfig.debug"
@@ -454,42 +205,9 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/cris/arch-v10/kernel/process.c 2006-09-20 16:57:57 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/cris/arch-v10/kernel/process.c 2007-06-15 02:37:02 +0200
-@@ -103,7 +103,8 @@ int kernel_thread(int (*fn)(void *), voi
- regs.dccr = 1 << I_DCCR_BITNR;
-
- /* Ok, create the new process.. */
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s, 0, NULL, NULL);
- }
-
- /* setup the child's kernel stack with a pt_regs and switch_stack on it.
---- linux-2.6.22/arch/cris/arch-v32/kernel/process.c 2006-09-20 16:57:57 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/cris/arch-v32/kernel/process.c 2007-06-15 02:37:02 +0200
-@@ -120,7 +120,8 @@ kernel_thread(int (*fn)(void *), void *
- regs.ccs = 1 << (I_CCS_BITNR + CCS_SHIFT);
-
- /* Create the new process. */
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s, 0, NULL, NULL);
- }
-
- /*
---- linux-2.6.22/arch/frv/kernel/kernel_thread.S 2005-03-02 12:38:20 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/frv/kernel/kernel_thread.S 2007-06-15 02:37:02 +0200
-@@ -13,6 +13,8 @@
- #include <asm/unistd.h>
-
- #define CLONE_VM 0x00000100 /* set if VM shared between processes */
-+#define CLONE_KTHREAD 0x10000000 /* kernel thread */
-+#define CLONE_KT (CLONE_VM | CLONE_KTHREAD) /* kernel thread flags */
- #define KERN_ERR "<3>"
-
- .section .rodata
-@@ -37,7 +39,7 @@ kernel_thread:
+--- linux-2.6.25.4/arch/frv/kernel/kernel_thread.S 2007-02-04 13:44:54.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/frv/kernel/kernel_thread.S 2008-04-21 11:09:01.000000000 -0400
+@@ -37,7 +37,7 @@ kernel_thread:
# start by forking the current process, but with shared VM
setlos.p #__NR_clone,gr7 ; syscall number
@@ -498,9 +216,9 @@
sethi.p #0xe4e4,gr9 ; second syscall arg [newsp]
setlo #0xe4e4,gr9
setlos.p #0,gr10 ; third syscall arg [parent_tidptr]
---- linux-2.6.22/arch/h8300/Kconfig 2007-07-09 13:17:41 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/h8300/Kconfig 2007-06-15 02:37:02 +0200
-@@ -222,6 +222,8 @@ source "fs/Kconfig"
+--- linux-2.6.25.4/arch/h8300/Kconfig 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/h8300/Kconfig 2008-04-19 15:14:51.000000000 -0400
+@@ -233,6 +233,8 @@ source "fs/Kconfig"
source "arch/h8300/Kconfig.debug"
@@ -509,154 +227,8 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/h8300/kernel/process.c 2006-09-20 16:57:58 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/h8300/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -134,7 +134,7 @@ int kernel_thread(int (*fn)(void *), voi
-
- fs = get_fs();
- set_fs (KERNEL_DS);
-- clone_arg = flags | CLONE_VM;
-+ clone_arg = flags | CLONE_VM | CLONE_KTHREAD;
- __asm__("mov.l sp,er3\n\t"
- "sub.l er2,er2\n\t"
- "mov.l %2,er1\n\t"
---- linux-2.6.22/arch/i386/Kconfig 2007-07-09 13:17:41 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/i386/Kconfig 2007-07-09 13:11:53 +0200
-@@ -1230,6 +1230,8 @@ endmenu
-
- source "arch/i386/Kconfig.debug"
-
-+source "kernel/vserver/Kconfig"
-+
- source "security/Kconfig"
-
- source "crypto/Kconfig"
---- linux-2.6.22/arch/i386/kernel/process.c 2007-07-09 13:17:42 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/i386/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -302,8 +302,10 @@ void show_regs(struct pt_regs * regs)
- unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
-
- printk("\n");
-- printk("Pid: %d, comm: %20s\n", current->pid, current->comm);
-- printk("EIP: %04x:[<%08lx>] CPU: %d\n",0xffff & regs->xcs,regs->eip, smp_processor_id());
-+ printk("Pid: %d[#%u], comm: %20s\n",
-+ current->pid, current->xid, current->comm);
-+ printk("EIP: %04x:[<%08lx>] CPU: %d\n",
-+ 0xffff & regs->xcs,regs->eip, smp_processor_id());
- print_symbol("EIP is at %s\n", regs->eip);
-
- if (user_mode_vm(regs))
-@@ -355,7 +357,8 @@ int kernel_thread(int (*fn)(void *), voi
- regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;
-
- /* Ok, create the new process.. */
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s, 0, NULL, NULL);
- }
- EXPORT_SYMBOL(kernel_thread);
-
---- linux-2.6.22/arch/i386/kernel/syscall_table.S 2007-07-09 13:17:42 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/i386/kernel/syscall_table.S 2007-06-15 02:37:03 +0200
-@@ -272,7 +272,7 @@ ENTRY(sys_call_table)
- .long sys_tgkill /* 270 */
- .long sys_utimes
- .long sys_fadvise64_64
-- .long sys_ni_syscall /* sys_vserver */
-+ .long sys_vserver
- .long sys_mbind
- .long sys_get_mempolicy
- .long sys_set_mempolicy
---- linux-2.6.22/arch/i386/kernel/sysenter.c 2007-07-09 13:17:42 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/i386/kernel/sysenter.c 2007-06-15 02:37:03 +0200
-@@ -17,6 +17,7 @@
- #include <linux/elf.h>
- #include <linux/mm.h>
- #include <linux/module.h>
-+#include <linux/vs_memory.h>
-
- #include <asm/cpufeature.h>
- #include <asm/msr.h>
---- linux-2.6.22/arch/i386/kernel/traps.c 2007-07-09 13:17:42 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/i386/kernel/traps.c 2007-06-15 02:38:37 +0200
-@@ -56,6 +56,8 @@
- #include <asm/stacktrace.h>
-
- #include <linux/module.h>
-+#include <linux/vs_context.h>
-+#include <linux/vserver/history.h>
-
- #include "mach_traps.h"
-
-@@ -303,8 +305,8 @@ void show_registers(struct pt_regs *regs
- regs->esi, regs->edi, regs->ebp, esp);
- printk(KERN_EMERG "ds: %04x es: %04x fs: %04x gs: %04x ss: %04x\n",
- regs->xds & 0xffff, regs->xes & 0xffff, regs->xfs & 0xffff, gs, ss);
-- printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)",
-- TASK_COMM_LEN, current->comm, current->pid,
-+ printk(KERN_EMERG "Process %.*s (pid: %d[#%u], ti=%p task=%p task.ti=%p)",
-+ TASK_COMM_LEN, current->comm, current->pid, current->xid,
- current_thread_info(), current, task_thread_info(current));
- /*
- * When in-kernel, we also print out the stack and code at the
-@@ -375,6 +377,8 @@ void die(const char * str, struct pt_reg
-
- oops_enter();
-
-+ vxh_throw_oops();
-+
- if (die.lock_owner != raw_smp_processor_id()) {
- console_verbose();
- spin_lock_irqsave(&die.lock, flags);
-@@ -412,9 +416,9 @@ void die(const char * str, struct pt_reg
- if (nl)
- printk("\n");
- if (notify_die(DIE_OOPS, str, regs, err,
-- current->thread.trap_no, SIGSEGV) !=
-- NOTIFY_STOP) {
-+ current->thread.trap_no, SIGSEGV) != NOTIFY_STOP) {
- show_registers(regs);
-+ vxh_dump_history();
- /* Executive summary in case the oops scrolled away */
- esp = (unsigned long) (®s->esp);
- savesegment(ss, ss);
---- linux-2.6.22/arch/i386/mm/fault.c 2007-07-09 13:17:43 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/i386/mm/fault.c 2007-06-17 05:54:16 +0200
-@@ -586,7 +586,8 @@ out_of_memory:
- down_read(&mm->mmap_sem);
- goto survive;
- }
-- printk("VM: killing process %s\n", tsk->comm);
-+ printk("VM: killing process %s(%d:#%u)\n",
-+ tsk->comm, tsk->pid, tsk->xid);
- if (error_code & 4)
- do_exit(SIGKILL);
- goto no_context;
---- linux-2.6.22/arch/ia64/Kconfig 2007-07-09 13:17:43 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/Kconfig 2007-06-15 02:37:03 +0200
-@@ -592,6 +592,8 @@ endmenu
-
- source "arch/ia64/Kconfig.debug"
-
-+source "kernel/vserver/Kconfig"
-+
- source "security/Kconfig"
-
- source "crypto/Kconfig"
---- linux-2.6.22/arch/ia64/ia32/binfmt_elf32.c 2007-05-02 19:24:20 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/ia32/binfmt_elf32.c 2007-06-15 02:37:03 +0200
-@@ -233,7 +233,8 @@ ia32_setup_arg_pages (struct linux_binpr
- kmem_cache_free(vm_area_cachep, mpnt);
- return ret;
- }
-- current->mm->stack_vm = current->mm->total_vm = vma_pages(mpnt);
-+ vx_vmpages_sub(current->mm, current->mm->total_vm - vma_pages(mpnt));
-+ current->mm->stack_vm = current->mm->total_vm;
- }
-
- for (i = 0 ; i < MAX_ARG_PAGES ; i++) {
---- linux-2.6.22/arch/ia64/ia32/ia32_entry.S 2007-07-09 13:17:43 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/ia32/ia32_entry.S 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/ia64/ia32/ia32_entry.S 2008-04-17 10:37:14.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/ia32/ia32_entry.S 2008-04-19 15:14:51.000000000 -0400
@@ -446,7 +446,7 @@ ia32_syscall_table:
data8 sys_tgkill /* 270 */
data8 compat_sys_utimes
@@ -666,9 +238,9 @@
data8 sys_ni_syscall
data8 sys_ni_syscall /* 275 */
data8 sys_ni_syscall
---- linux-2.6.22/arch/ia64/ia32/sys_ia32.c 2007-05-02 19:24:20 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/ia32/sys_ia32.c 2007-06-15 02:37:03 +0200
-@@ -1182,7 +1182,7 @@ sys32_gettimeofday (struct compat_timeva
+--- linux-2.6.25.4/arch/ia64/ia32/sys_ia32.c 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/ia32/sys_ia32.c 2008-04-19 15:14:51.000000000 -0400
+@@ -1177,7 +1177,7 @@ sys32_gettimeofday (struct compat_timeva
{
if (tv) {
struct timeval ktv;
@@ -677,18 +249,19 @@
if (put_tv32(tv, &ktv))
return -EFAULT;
}
---- linux-2.6.22/arch/ia64/kernel/asm-offsets.c 2007-05-02 19:24:20 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/kernel/asm-offsets.c 2007-06-15 02:37:03 +0200
-@@ -192,6 +192,7 @@ void foo(void)
- /* for assembly files which can't include sched.h: */
- DEFINE(IA64_CLONE_VFORK, CLONE_VFORK);
- DEFINE(IA64_CLONE_VM, CLONE_VM);
-+ DEFINE(IA64_CLONE_KTHREAD, CLONE_KTHREAD);
-
- BLANK();
- DEFINE(IA64_CPUINFO_NSEC_PER_CYC_OFFSET,
---- linux-2.6.22/arch/ia64/kernel/entry.S 2007-07-09 13:17:43 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/kernel/entry.S 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/ia64/Kconfig 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/Kconfig 2008-04-19 15:14:51.000000000 -0400
+@@ -615,6 +615,8 @@ source "arch/ia64/hp/sim/Kconfig"
+
+ source "arch/ia64/Kconfig.debug"
+
++source "kernel/vserver/Kconfig"
++
+ source "security/Kconfig"
+
+ source "crypto/Kconfig"
+--- linux-2.6.25.4/arch/ia64/kernel/entry.S 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/kernel/entry.S 2008-04-19 15:14:51.000000000 -0400
@@ -1547,7 +1547,7 @@ sys_call_table:
data8 sys_mq_notify
data8 sys_mq_getsetattr
@@ -698,8 +271,8 @@
data8 sys_waitid // 1270
data8 sys_add_key
data8 sys_request_key
---- linux-2.6.22/arch/ia64/kernel/perfmon.c 2007-07-09 13:17:43 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/kernel/perfmon.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/ia64/kernel/perfmon.c 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/kernel/perfmon.c 2008-04-19 15:14:51.000000000 -0400
@@ -40,6 +40,7 @@
#include <linux/capability.h>
#include <linux/rcupdate.h>
@@ -708,7 +281,7 @@
#include <asm/errno.h>
#include <asm/intrinsics.h>
-@@ -2395,7 +2396,7 @@ pfm_smpl_buffer_alloc(struct task_struct
+@@ -2374,7 +2375,7 @@ pfm_smpl_buffer_alloc(struct task_struct
*/
insert_vm_struct(mm, vma);
@@ -717,30 +290,21 @@
vm_stat_account(vma->vm_mm, vma->vm_flags, vma->vm_file,
vma_pages(vma));
up_write(&task->mm->mmap_sem);
---- linux-2.6.22/arch/ia64/kernel/process.c 2007-07-09 13:17:43 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -105,7 +105,8 @@ show_regs (struct pt_regs *regs)
+--- linux-2.6.25.4/arch/ia64/kernel/process.c 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/kernel/process.c 2008-04-21 11:09:01.000000000 -0400
+@@ -105,8 +105,8 @@ show_regs (struct pt_regs *regs)
unsigned long ip = regs->cr_iip + ia64_psr(regs)->ri;
print_modules();
-- printk("\nPid: %d, CPU %d, comm: %20s\n", current->pid, smp_processor_id(), current->comm);
-+ printk("\nPid: %d[#%u], CPU %d, comm: %20s\n",
-+ current->pid, current->xid, smp_processor_id(), current->comm);
- printk("psr : %016lx ifs : %016lx ip : [<%016lx>] %s\n",
- regs->cr_ipsr, regs->cr_ifs, ip, print_tainted());
- print_symbol("ip is at %s\n", ip);
-@@ -696,7 +697,8 @@ kernel_thread (int (*fn)(void *), void *
- regs.sw.ar_fpsr = regs.pt.ar_fpsr = ia64_getreg(_IA64_REG_AR_FPSR);
- regs.sw.ar_bspstore = (unsigned long) current + IA64_RBS_OFFSET;
- regs.sw.pr = (1 << PRED_KERNEL_STACK);
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s.pt, 0, NULL, NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s.pt, 0, NULL, NULL);
- }
- EXPORT_SYMBOL(kernel_thread);
-
---- linux-2.6.22/arch/ia64/kernel/ptrace.c 2007-05-02 19:24:20 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/kernel/ptrace.c 2007-06-15 02:37:03 +0200
+- printk("\nPid: %d, CPU %d, comm: %20s\n", task_pid_nr(current),
+- smp_processor_id(), current->comm);
++ printk("\nPid: %d[#%u], CPU %d, comm: %20s\n", task_pid_nr(current),
++ current->xid, smp_processor_id(), current->comm);
+ printk("psr : %016lx ifs : %016lx ip : [<%016lx>] %s (%s)\n",
+ regs->cr_ipsr, regs->cr_ifs, ip, print_tainted(),
+ init_utsname()->release);
+--- linux-2.6.25.4/arch/ia64/kernel/ptrace.c 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/kernel/ptrace.c 2008-04-19 15:14:51.000000000 -0400
@@ -17,6 +17,7 @@
#include <linux/security.h>
#include <linux/audit.h>
@@ -749,44 +313,34 @@
#include <asm/pgtable.h>
#include <asm/processor.h>
-@@ -1443,6 +1444,9 @@ sys_ptrace (long request, pid_t pid, uns
- read_unlock(&tasklist_lock);
- if (!child)
- goto out;
-+ if (!vx_check(vx_task_xid(child), VS_WATCH_P | VS_IDENT))
-+ goto out_tsk;
-+
- ret = -EPERM;
- if (pid == 1) /* no messing around with init! */
- goto out_tsk;
---- linux-2.6.22/arch/ia64/kernel/traps.c 2007-07-09 13:17:43 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/kernel/traps.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/ia64/kernel/traps.c 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/kernel/traps.c 2008-04-21 10:33:04.000000000 -0400
@@ -60,8 +60,9 @@ die (const char *str, struct pt_regs *re
put_cpu();
if (++die.lock_owner_depth < 3) {
- printk("%s[%d]: %s %ld [%d]\n",
-- current->comm, current->pid, str, err, ++die_counter);
+- current->comm, task_pid_nr(current), str, err, ++die_counter);
+ printk("%s[%d[#%u]]: %s %ld [%d]\n",
-+ current->comm, current->pid, current->xid,
++ current->comm, task_pid_nr(current), current->xid,
+ str, err, ++die_counter);
- (void) notify_die(DIE_OOPS, (char *)str, regs, err, 255, SIGSEGV);
- show_regs(regs);
- } else
-@@ -313,8 +314,9 @@ handle_fpu_swa (int fp_fault, struct pt_
+ if (notify_die(DIE_OOPS, str, regs, err, 255, SIGSEGV)
+ != NOTIFY_STOP)
+ show_regs(regs);
+@@ -324,8 +325,9 @@ handle_fpu_swa (int fp_fault, struct pt_
if ((last.count & 15) < 5 && (ia64_fetchadd(1, &last.count, acq) & 15) < 5) {
last.time = current_jiffies + 5 * HZ;
printk(KERN_WARNING
- "%s(%d): floating-point assist fault at ip %016lx, isr %016lx\n",
-- current->comm, current->pid, regs->cr_iip + ia64_psr(regs)->ri, isr);
+- current->comm, task_pid_nr(current), regs->cr_iip + ia64_psr(regs)->ri, isr);
+ "%s(%d[#%u]): floating-point assist fault at ip %016lx, isr %016lx\n",
-+ current->comm, current->pid, current->xid,
++ current->comm, task_pid_nr(current), current->xid,
+ regs->cr_iip + ia64_psr(regs)->ri, isr);
}
}
}
---- linux-2.6.22/arch/ia64/mm/fault.c 2007-07-09 13:17:43 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ia64/mm/fault.c 2007-06-15 02:39:01 +0200
+--- linux-2.6.25.4/arch/ia64/mm/fault.c 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ia64/mm/fault.c 2008-04-19 15:14:52.000000000 -0400
@@ -10,6 +10,7 @@
#include <linux/interrupt.h>
#include <linux/kprobes.h>
@@ -795,36 +349,23 @@
#include <asm/pgtable.h>
#include <asm/processor.h>
---- linux-2.6.22/arch/m32r/kernel/process.c 2007-05-02 19:24:21 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m32r/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -211,8 +211,8 @@ int kernel_thread(int (*fn)(void *), voi
- regs.psw = M32R_PSW_BIE;
-
- /* Ok, create the new process. */
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL,
-- NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s, 0, NULL, NULL);
- }
-
- /*
---- linux-2.6.22/arch/m32r/kernel/traps.c 2006-11-30 21:18:28 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m32r/kernel/traps.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/m32r/kernel/traps.c 2008-04-17 11:31:23.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/m32r/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
@@ -195,8 +195,9 @@ static void show_registers(struct pt_reg
} else {
printk("SPI: %08lx\n", sp);
}
- printk("Process %s (pid: %d, process nr: %d, stackpage=%08lx)",
-- current->comm, current->pid, 0xffff & i, 4096+(unsigned long)current);
+- current->comm, task_pid_nr(current), 0xffff & i, 4096+(unsigned long)current);
+ printk("Process %s (pid: %d[#%u], process nr: %d, stackpage=%08lx)",
-+ current->comm, current->pid, current->xid,
++ current->comm, task_pid_nr(current), current->xid,
+ 0xffff & i, 4096+(unsigned long)current);
/*
* When in-kernel, we also print out the stack and code at the
---- linux-2.6.22/arch/m68k/Kconfig 2007-07-09 13:17:47 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m68k/Kconfig 2007-07-07 03:52:53 +0200
-@@ -678,6 +678,8 @@ source "fs/Kconfig"
+--- linux-2.6.25.4/arch/m68k/Kconfig 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/m68k/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -674,6 +674,8 @@ source "fs/Kconfig"
source "arch/m68k/Kconfig.debug"
@@ -833,20 +374,8 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/m68k/kernel/process.c 2006-11-30 21:18:28 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m68k/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -159,7 +159,8 @@ int kernel_thread(int (*fn)(void *), voi
-
- {
- register long retval __asm__ ("d0");
-- register long clone_arg __asm__ ("d1") = flags | CLONE_VM | CLONE_UNTRACED;
-+ register long clone_arg __asm__ ("d1") =
-+ flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD;
-
- retval = __NR_clone;
- __asm__ __volatile__
---- linux-2.6.22/arch/m68k/kernel/ptrace.c 2007-07-09 13:17:47 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m68k/kernel/ptrace.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/m68k/kernel/ptrace.c 2008-04-17 11:31:23.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/m68k/kernel/ptrace.c 2008-04-19 15:14:52.000000000 -0400
@@ -18,6 +18,7 @@
#include <linux/ptrace.h>
#include <linux/user.h>
@@ -855,7 +384,7 @@
#include <asm/uaccess.h>
#include <asm/page.h>
-@@ -278,6 +279,8 @@ long arch_ptrace(struct task_struct *chi
+@@ -269,6 +270,8 @@ long arch_ptrace(struct task_struct *chi
ret = ptrace_request(child, request, addr, data);
break;
}
@@ -864,22 +393,22 @@
return ret;
out_eio:
---- linux-2.6.22/arch/m68k/kernel/traps.c 2007-05-02 19:24:21 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m68k/kernel/traps.c 2007-06-15 02:37:03 +0200
-@@ -899,8 +899,8 @@ void show_registers(struct pt_regs *regs
+--- linux-2.6.25.4/arch/m68k/kernel/traps.c 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/m68k/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
+@@ -898,8 +898,8 @@ void show_registers(struct pt_regs *regs
printk("d4: %08lx d5: %08lx a0: %08lx a1: %08lx\n",
regs->d4, regs->d5, regs->a0, regs->a1);
- printk("Process %s (pid: %d, task=%p)\n",
-- current->comm, current->pid, current);
+- current->comm, task_pid_nr(current), current);
+ printk("Process %s (pid: %d[#%u], task=%p)\n",
-+ current->comm, current->pid, current->xid, current);
++ current->comm, task_pid_nr(current), current->xid, current);
addr = (unsigned long)&fp->un;
printk("Frame format=%X ", regs->format);
switch (regs->format) {
---- linux-2.6.22/arch/m68knommu/Kconfig 2007-07-09 13:17:47 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m68knommu/Kconfig 2007-06-15 02:37:03 +0200
-@@ -670,6 +670,8 @@ source "fs/Kconfig"
+--- linux-2.6.25.4/arch/m68knommu/Kconfig 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/m68knommu/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -722,6 +722,8 @@ source "fs/Kconfig"
source "arch/m68knommu/Kconfig.debug"
@@ -888,34 +417,23 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/m68knommu/kernel/process.c 2007-02-06 03:00:08 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m68knommu/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -122,7 +122,7 @@ void show_regs(struct pt_regs * regs)
- int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
- {
- int retval;
-- long clone_arg = flags | CLONE_VM;
-+ long clone_arg = flags | CLONE_VM | CLONE_KTHREAD;
- mm_segment_t fs;
-
- fs = get_fs();
---- linux-2.6.22/arch/m68knommu/kernel/traps.c 2007-05-02 19:24:21 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/m68knommu/kernel/traps.c 2007-06-15 02:37:03 +0200
-@@ -80,8 +80,9 @@ void die_if_kernel(char *str, struct pt_
+--- linux-2.6.25.4/arch/m68knommu/kernel/traps.c 2008-04-17 10:37:14.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/m68knommu/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
+@@ -78,8 +78,9 @@ void die_if_kernel(char *str, struct pt_
printk(KERN_EMERG "d4: %08lx d5: %08lx a0: %08lx a1: %08lx\n",
fp->d4, fp->d5, fp->a0, fp->a1);
- printk(KERN_EMERG "Process %s (pid: %d, stackpage=%08lx)\n",
- current->comm, current->pid, PAGE_SIZE+(unsigned long)current);
+ printk(KERN_EMERG "Process %s (pid: %d[#%u], stackpage=%08lx)\n",
-+ current->comm, current->pid, current->xid,
++ current->comm, task_pid_nr(current), current->xid,
+ PAGE_SIZE+(unsigned long)current);
- show_stack(NULL, (unsigned long *)fp);
+ show_stack(NULL, (unsigned long *)(fp + 1));
+ add_taint(TAINT_DIE);
do_exit(SIGSEGV);
- }
---- linux-2.6.22/arch/mips/Kconfig 2007-07-09 13:17:47 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/Kconfig 2007-07-07 03:52:53 +0200
-@@ -1959,6 +1959,8 @@ source "arch/mips/oprofile/Kconfig"
+--- linux-2.6.25.4/arch/mips/Kconfig 2008-04-17 12:05:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -2099,6 +2099,8 @@ source "fs/Kconfig"
source "arch/mips/Kconfig.debug"
@@ -924,9 +442,9 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/mips/kernel/linux32.c 2007-07-09 13:17:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/kernel/linux32.c 2007-06-17 05:54:16 +0200
-@@ -229,7 +229,7 @@ sys32_gettimeofday(struct compat_timeval
+--- linux-2.6.25.4/arch/mips/kernel/linux32.c 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/kernel/linux32.c 2008-04-19 15:14:52.000000000 -0400
+@@ -209,7 +209,7 @@ sys32_gettimeofday(struct compat_timeval
{
if (tv) {
struct timeval ktv;
@@ -935,24 +453,12 @@
if (put_tv32(tv, &ktv))
return -EFAULT;
}
---- linux-2.6.22/arch/mips/kernel/process.c 2007-05-02 19:24:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -236,7 +236,8 @@ long kernel_thread(int (*fn)(void *), vo
- #endif
-
- /* Ok, create the new process.. */
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s, 0, NULL, NULL);
- }
-
- /*
---- linux-2.6.22/arch/mips/kernel/ptrace.c 2007-07-09 13:17:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/kernel/ptrace.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/mips/kernel/ptrace.c 2008-04-17 11:31:23.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/kernel/ptrace.c 2008-04-19 15:14:52.000000000 -0400
@@ -25,6 +25,7 @@
- #include <linux/user.h>
#include <linux/security.h>
- #include <linux/signal.h>
+ #include <linux/audit.h>
+ #include <linux/seccomp.h>
+#include <linux/vs_base.h>
#include <asm/byteorder.h>
@@ -967,8 +473,8 @@
switch (request) {
/* when I and D space are separate, these will need to be fixed. */
case PTRACE_PEEKTEXT: /* read word at location addr. */
---- linux-2.6.22/arch/mips/kernel/scall32-o32.S 2007-07-09 13:17:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/kernel/scall32-o32.S 2007-06-17 05:54:16 +0200
+--- linux-2.6.25.4/arch/mips/kernel/scall32-o32.S 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/kernel/scall32-o32.S 2008-04-19 15:14:52.000000000 -0400
@@ -619,7 +619,7 @@ einval: li v0, -EINVAL
sys sys_mq_timedreceive 5
sys sys_mq_notify 2 /* 4275 */
@@ -978,8 +484,8 @@
sys sys_waitid 5
sys sys_ni_syscall 0 /* available, was setaltroot */
sys sys_add_key 5 /* 4280 */
---- linux-2.6.22/arch/mips/kernel/scall64-64.S 2007-07-09 13:17:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/kernel/scall64-64.S 2007-06-17 05:54:16 +0200
+--- linux-2.6.25.4/arch/mips/kernel/scall64-64.S 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/kernel/scall64-64.S 2008-04-19 15:14:52.000000000 -0400
@@ -434,7 +434,7 @@ sys_call_table:
PTR sys_mq_timedreceive
PTR sys_mq_notify
@@ -989,8 +495,8 @@
PTR sys_waitid
PTR sys_ni_syscall /* available, was setaltroot */
PTR sys_add_key
---- linux-2.6.22/arch/mips/kernel/scall64-n32.S 2007-07-09 13:17:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/kernel/scall64-n32.S 2007-06-17 05:54:16 +0200
+--- linux-2.6.25.4/arch/mips/kernel/scall64-n32.S 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/kernel/scall64-n32.S 2008-04-19 15:14:52.000000000 -0400
@@ -360,7 +360,7 @@ EXPORT(sysn32_call_table)
PTR compat_sys_mq_timedreceive
PTR compat_sys_mq_notify
@@ -1000,8 +506,8 @@
PTR compat_sys_waitid
PTR sys_ni_syscall /* available, was setaltroot */
PTR sys_add_key
---- linux-2.6.22/arch/mips/kernel/scall64-o32.S 2007-07-09 13:17:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/kernel/scall64-o32.S 2007-06-17 05:54:16 +0200
+--- linux-2.6.25.4/arch/mips/kernel/scall64-o32.S 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/kernel/scall64-o32.S 2008-04-19 15:14:52.000000000 -0400
@@ -482,7 +482,7 @@ sys_call_table:
PTR compat_sys_mq_timedreceive
PTR compat_sys_mq_notify /* 4275 */
@@ -1011,23 +517,23 @@
PTR sys32_waitid
PTR sys_ni_syscall /* available, was setaltroot */
PTR sys_add_key /* 4280 */
---- linux-2.6.22/arch/mips/kernel/traps.c 2007-07-09 13:17:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/kernel/traps.c 2007-07-09 13:11:53 +0200
-@@ -302,8 +302,9 @@ void show_registers(struct pt_regs *regs
+--- linux-2.6.25.4/arch/mips/kernel/traps.c 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
+@@ -313,8 +313,9 @@ void show_registers(const struct pt_regs
{
- show_regs(regs);
+ __show_regs(regs);
print_modules();
- printk("Process %s (pid: %d, threadinfo=%p, task=%p)\n",
-- current->comm, current->pid, current_thread_info(), current);
+- current->comm, task_pid_nr(current), current_thread_info(), current);
+ printk("Process %s (pid: %d:#%u, threadinfo=%p, task=%p)\n",
-+ current->comm, current->pid, current->xid,
++ current->comm, task_pid_nr(current), current->xid,
+ current_thread_info(), current);
show_stacktrace(current, regs);
- show_code((unsigned int *) regs->cp0_epc);
+ show_code((unsigned int __user *) regs->cp0_epc);
printk("\n");
---- linux-2.6.22/arch/mips/mm/fault.c 2007-07-09 13:17:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/mips/mm/fault.c 2007-06-15 02:37:03 +0200
-@@ -179,7 +179,8 @@ out_of_memory:
+--- linux-2.6.25.4/arch/mips/mm/fault.c 2008-04-17 11:31:24.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/mips/mm/fault.c 2008-04-19 15:14:52.000000000 -0400
+@@ -178,7 +178,8 @@ out_of_memory:
down_read(&mm->mmap_sem);
goto survive;
}
@@ -1035,11 +541,11 @@
+ printk("VM: killing process %s(%d:#%u)\n",
+ tsk->comm, tsk->pid, tsk->xid);
if (user_mode(regs))
- do_exit(SIGKILL);
+ do_group_exit(SIGKILL);
goto no_context;
---- linux-2.6.22/arch/parisc/Kconfig 2007-05-02 19:24:26 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/parisc/Kconfig 2007-06-15 02:37:03 +0200
-@@ -271,6 +271,8 @@ source "arch/parisc/oprofile/Kconfig"
+--- linux-2.6.25.4/arch/parisc/Kconfig 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/parisc/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -278,6 +278,8 @@ source "fs/Kconfig"
source "arch/parisc/Kconfig.debug"
@@ -1048,29 +554,19 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/parisc/kernel/entry.S 2007-07-09 13:17:50 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/parisc/kernel/entry.S 2007-06-17 05:54:16 +0200
-@@ -761,6 +761,7 @@ END(fault_vector_11)
-
- #define CLONE_VM 0x100 /* Must agree with <linux/sched.h> */
- #define CLONE_UNTRACED 0x00800000
-+#define CLONE_KTHREAD 0x10000000
-
- .import do_fork
- ENTRY(__kernel_thread)
---- linux-2.6.22/arch/parisc/kernel/process.c 2007-07-09 13:17:50 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/parisc/kernel/process.c 2007-06-17 05:54:16 +0200
-@@ -173,7 +173,7 @@ pid_t kernel_thread(int (*fn)(void *), v
- * kernel_thread can become a #define.
- */
-
-- return __kernel_thread(fn, arg, flags);
-+ return __kernel_thread(fn, arg, flags | CLONE_KTHREAD);
- }
- EXPORT_SYMBOL(kernel_thread);
-
---- linux-2.6.22/arch/parisc/kernel/sys_parisc32.c 2007-07-09 13:17:50 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/parisc/kernel/sys_parisc32.c 2007-06-17 05:54:16 +0200
+--- linux-2.6.25.4/arch/parisc/kernel/syscall_table.S 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/parisc/kernel/syscall_table.S 2008-04-19 15:14:52.000000000 -0400
+@@ -361,7 +361,7 @@
+ ENTRY_COMP(mbind) /* 260 */
+ ENTRY_COMP(get_mempolicy)
+ ENTRY_COMP(set_mempolicy)
+- ENTRY_SAME(ni_syscall) /* 263: reserved for vserver */
++ ENTRY_DIFF(vserver)
+ ENTRY_SAME(add_key)
+ ENTRY_SAME(request_key) /* 265 */
+ ENTRY_SAME(keyctl)
+--- linux-2.6.25.4/arch/parisc/kernel/sys_parisc32.c 2008-04-17 11:31:24.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/parisc/kernel/sys_parisc32.c 2008-04-19 15:14:52.000000000 -0400
@@ -204,11 +204,11 @@ static inline long get_ts32(struct times
asmlinkage int
sys32_gettimeofday(struct compat_timeval __user *tv, struct timezone __user *tz)
@@ -1085,57 +581,46 @@
if (put_compat_timeval(tv, &ktv))
return -EFAULT;
}
---- linux-2.6.22/arch/parisc/kernel/syscall_table.S 2007-07-09 13:17:50 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/parisc/kernel/syscall_table.S 2007-06-17 05:54:16 +0200
-@@ -363,7 +363,7 @@
- ENTRY_COMP(mbind) /* 260 */
- ENTRY_COMP(get_mempolicy)
- ENTRY_COMP(set_mempolicy)
-- ENTRY_SAME(ni_syscall) /* 263: reserved for vserver */
-+ ENTRY_DIFF(vserver)
- ENTRY_SAME(add_key)
- ENTRY_SAME(request_key) /* 265 */
- ENTRY_SAME(keyctl)
---- linux-2.6.22/arch/parisc/kernel/traps.c 2007-07-09 13:17:50 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/parisc/kernel/traps.c 2007-06-17 05:54:16 +0200
-@@ -218,8 +218,9 @@ void die_if_kernel(char *str, struct pt_
+--- linux-2.6.25.4/arch/parisc/kernel/traps.c 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/parisc/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
+@@ -237,8 +237,9 @@ void die_if_kernel(char *str, struct pt_
if (err == 0)
return; /* STFU */
- printk(KERN_CRIT "%s (pid %d): %s (code %ld) at " RFMT "\n",
-- current->comm, current->pid, str, err, regs->iaoq[0]);
+- current->comm, task_pid_nr(current), str, err, regs->iaoq[0]);
+ printk(KERN_CRIT "%s (pid %d:#%u): %s (code %ld) at " RFMT "\n",
-+ current->comm, current->pid, current->xid,
++ current->comm, task_pid_nr(current), current->xid,
+ str, err, regs->iaoq[0]);
#ifdef PRINT_USER_FAULTS
/* XXX for debugging only */
show_regs(regs);
-@@ -251,8 +252,8 @@ KERN_CRIT " || |
+@@ -270,8 +271,8 @@ KERN_CRIT " || |
pdc_console_restart();
if (err)
- printk(KERN_CRIT "%s (pid %d): %s (code %ld)\n",
-- current->comm, current->pid, str, err);
+- current->comm, task_pid_nr(current), str, err);
+ printk(KERN_CRIT "%s (pid %d:#%u): %s (code %ld)\n",
-+ current->comm, current->pid, current->xid, str, err);
++ current->comm, task_pid_nr(current), current->xid, str, err);
/* Wot's wrong wif bein' racy? */
if (current->thread.flags & PARISC_KERNEL_DEATH) {
---- linux-2.6.22/arch/parisc/mm/fault.c 2007-05-02 19:24:26 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/parisc/mm/fault.c 2007-06-15 02:37:03 +0200
-@@ -209,8 +209,9 @@ bad_area:
+--- linux-2.6.25.4/arch/parisc/mm/fault.c 2008-04-17 11:31:24.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/parisc/mm/fault.c 2008-04-19 15:14:52.000000000 -0400
+@@ -210,8 +210,9 @@ bad_area:
#ifdef PRINT_USER_FAULTS
printk(KERN_DEBUG "\n");
- printk(KERN_DEBUG "do_page_fault() pid=%d command='%s' type=%lu address=0x%08lx\n",
-- tsk->pid, tsk->comm, code, address);
+- task_pid_nr(tsk), tsk->comm, code, address);
+ printk(KERN_DEBUG "do_page_fault() pid=%d:#%u "
+ "command='%s' type=%lu address=0x%08lx\n",
-+ tsk->pid, tsk->xid, tsk->comm, code, address);
++ task_pid_nr(tsk), tsk->xid, tsk->comm, code, address);
if (vma) {
printk(KERN_DEBUG "vm_start = 0x%08lx, vm_end = 0x%08lx\n",
vma->vm_start, vma->vm_end);
-@@ -260,7 +261,8 @@ no_context:
+@@ -261,7 +262,8 @@ no_context:
out_of_memory:
up_read(&mm->mmap_sem);
@@ -1143,11 +628,11 @@
+ printk(KERN_CRIT "VM: killing process %s(%d:#%u)\n",
+ current->comm, current->pid, current->xid);
if (user_mode(regs))
- do_exit(SIGKILL);
+ do_group_exit(SIGKILL);
goto no_context;
---- linux-2.6.22/arch/powerpc/Kconfig 2007-07-09 13:17:50 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/Kconfig 2007-06-15 02:37:03 +0200
-@@ -906,6 +906,8 @@ endmenu
+--- linux-2.6.25.4/arch/powerpc/Kconfig 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/powerpc/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -706,6 +706,8 @@ source "lib/Kconfig"
source "arch/powerpc/Kconfig.debug"
@@ -1156,64 +641,32 @@
source "security/Kconfig"
config KEYS_COMPAT
---- linux-2.6.22/arch/powerpc/kernel/asm-offsets.c 2007-07-09 13:17:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/kernel/asm-offsets.c 2007-06-15 02:37:03 +0200
-@@ -250,6 +250,7 @@ int main(void)
-
- DEFINE(CLONE_VM, CLONE_VM);
- DEFINE(CLONE_UNTRACED, CLONE_UNTRACED);
-+ DEFINE(CLONE_KTHREAD, CLONE_KTHREAD);
-
- #ifndef CONFIG_PPC64
- DEFINE(MM_PGD, offsetof(struct mm_struct, pgd));
---- linux-2.6.22/arch/powerpc/kernel/irq.c 2007-07-09 13:17:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/kernel/irq.c 2007-07-09 13:11:53 +0200
+--- linux-2.6.25.4/arch/powerpc/kernel/irq.c 2008-04-17 12:05:29.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/powerpc/kernel/irq.c 2008-04-19 15:14:52.000000000 -0400
@@ -53,6 +53,7 @@
- #include <linux/mutex.h>
#include <linux/bootmem.h>
#include <linux/pci.h>
+ #include <linux/debugfs.h>
+#include <linux/vs_context.h>
#include <asm/uaccess.h>
#include <asm/system.h>
---- linux-2.6.22/arch/powerpc/kernel/misc_32.S 2007-07-09 13:17:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/kernel/misc_32.S 2007-06-15 02:37:03 +0200
-@@ -745,7 +745,7 @@ _GLOBAL(kernel_thread)
- mr r30,r3 /* function */
- mr r31,r4 /* argument */
- ori r3,r5,CLONE_VM /* flags */
-- oris r3,r3,CLONE_UNTRACED>>16
-+ oris r3,r3,(CLONE_UNTRACED|CLONE_KTHREAD)>>16
- li r4,0 /* new sp (unused) */
- li r0,__NR_clone
- sc
---- linux-2.6.22/arch/powerpc/kernel/misc_64.S 2007-05-02 19:24:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/kernel/misc_64.S 2007-06-15 02:37:03 +0200
-@@ -434,7 +434,7 @@ _GLOBAL(kernel_thread)
- mr r29,r3
- mr r30,r4
- ori r3,r5,CLONE_VM /* flags */
-- oris r3,r3,(CLONE_UNTRACED>>16)
-+ oris r3,r3,(CLONE_UNTRACED|CLONE_KTHREAD)>>16
- li r4,0 /* new sp (unused) */
- li r0,__NR_clone
- sc
---- linux-2.6.22/arch/powerpc/kernel/process.c 2007-07-09 13:17:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -422,8 +422,9 @@ void show_regs(struct pt_regs * regs)
- trap = TRAP(regs);
- if (trap == 0x300 || trap == 0x600)
+--- linux-2.6.25.4/arch/powerpc/kernel/process.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/powerpc/kernel/process.c 2008-04-19 15:14:52.000000000 -0400
+@@ -464,8 +464,9 @@ void show_regs(struct pt_regs * regs)
+ #else
printk("DAR: "REG", DSISR: "REG"\n", regs->dar, regs->dsisr);
+ #endif
- printk("TASK = %p[%d] '%s' THREAD: %p",
-- current, current->pid, current->comm, task_thread_info(current));
+- current, task_pid_nr(current), current->comm, task_thread_info(current));
+ printk("TASK = %p[%d,#%u] '%s' THREAD: %p",
-+ current, current->pid, current->xid,
++ current, task_pid_nr(current), current->xid,
+ current->comm, task_thread_info(current));
#ifdef CONFIG_SMP
- printk(" CPU: %d", smp_processor_id());
---- linux-2.6.22/arch/powerpc/kernel/sys_ppc32.c 2007-07-09 13:17:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/kernel/sys_ppc32.c 2007-07-07 03:52:53 +0200
+ printk(" CPU: %d", raw_smp_processor_id());
+--- linux-2.6.25.4/arch/powerpc/kernel/sys_ppc32.c 2008-04-17 11:31:24.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/powerpc/kernel/sys_ppc32.c 2008-04-19 15:14:52.000000000 -0400
@@ -205,7 +205,7 @@ asmlinkage long compat_sys_gettimeofday(
{
if (tv) {
@@ -1223,22 +676,22 @@
if (put_tv32(tv, &ktv))
return -EFAULT;
}
---- linux-2.6.22/arch/powerpc/kernel/traps.c 2007-07-09 13:17:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/kernel/traps.c 2007-06-15 02:37:03 +0200
-@@ -878,8 +878,9 @@ void nonrecoverable_exception(struct pt_
+--- linux-2.6.25.4/arch/powerpc/kernel/traps.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/powerpc/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
+@@ -941,8 +941,9 @@ void nonrecoverable_exception(struct pt_
void trace_syscall(struct pt_regs *regs)
{
- printk("Task: %p(%d), PC: %08lX/%08lX, Syscall: %3ld, Result: %s%ld %s\n",
-- current, current->pid, regs->nip, regs->link, regs->gpr[0],
+- current, task_pid_nr(current), regs->nip, regs->link, regs->gpr[0],
+ printk("Task: %p(%d[#%u]), PC: %08lX/%08lX, Syscall: %3ld, Result: %s%ld %s\n",
-+ current, current->pid, current->xid,
++ current, task_pid_nr(current), current->xid,
+ regs->nip, regs->link, regs->gpr[0],
regs->ccr&0x10000000?"Error=":"", regs->gpr[3], print_tainted());
}
---- linux-2.6.22/arch/powerpc/kernel/vdso.c 2007-07-09 13:17:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/kernel/vdso.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/powerpc/kernel/vdso.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/powerpc/kernel/vdso.c 2008-04-19 15:14:52.000000000 -0400
@@ -21,6 +21,7 @@
#include <linux/elf.h>
#include <linux/security.h>
@@ -1247,8 +700,8 @@
#include <asm/pgtable.h>
#include <asm/system.h>
---- linux-2.6.22/arch/powerpc/mm/fault.c 2007-07-09 13:17:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/powerpc/mm/fault.c 2007-07-07 03:52:53 +0200
+--- linux-2.6.25.4/arch/powerpc/mm/fault.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/powerpc/mm/fault.c 2008-04-19 15:14:52.000000000 -0400
@@ -378,7 +378,8 @@ out_of_memory:
down_read(&mm->mmap_sem);
goto survive;
@@ -1257,11 +710,11 @@
+ printk("VM: killing process %s(%d:#%u)\n",
+ current->comm, current->pid, current->xid);
if (user_mode(regs))
- do_exit(SIGKILL);
+ do_group_exit(SIGKILL);
return SIGKILL;
---- linux-2.6.22/arch/ppc/Kconfig 2007-05-02 19:24:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ppc/Kconfig 2007-06-15 02:37:03 +0200
-@@ -1455,6 +1455,8 @@ source "arch/powerpc/oprofile/Kconfig"
+--- linux-2.6.25.4/arch/ppc/Kconfig 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ppc/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -1261,6 +1261,8 @@ source "lib/Kconfig"
source "arch/ppc/Kconfig.debug"
@@ -1270,30 +723,9 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/ppc/kernel/asm-offsets.c 2007-07-09 13:18:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ppc/kernel/asm-offsets.c 2007-06-15 02:37:03 +0200
-@@ -120,6 +120,7 @@ main(void)
- DEFINE(TRAP, STACK_FRAME_OVERHEAD+offsetof(struct pt_regs, trap));
- DEFINE(CLONE_VM, CLONE_VM);
- DEFINE(CLONE_UNTRACED, CLONE_UNTRACED);
-+ DEFINE(CLONE_KTHREAD, CLONE_KTHREAD);
- DEFINE(MM_PGD, offsetof(struct mm_struct, pgd));
-
- /* About the CPU features table */
---- linux-2.6.22/arch/ppc/kernel/misc.S 2006-11-30 21:18:32 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ppc/kernel/misc.S 2007-06-15 02:37:03 +0200
-@@ -848,7 +848,7 @@ _GLOBAL(kernel_thread)
- mr r30,r3 /* function */
- mr r31,r4 /* argument */
- ori r3,r5,CLONE_VM /* flags */
-- oris r3,r3,CLONE_UNTRACED>>16
-+ oris r3,r3,(CLONE_UNTRACED|CLONE_KTHREAD)>>16
- li r4,0 /* new sp (unused) */
- li r0,__NR_clone
- sc
---- linux-2.6.22/arch/ppc/kernel/traps.c 2007-07-09 13:18:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ppc/kernel/traps.c 2007-06-15 02:37:03 +0200
-@@ -696,8 +696,9 @@ void nonrecoverable_exception(struct pt_
+--- linux-2.6.25.4/arch/ppc/kernel/traps.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ppc/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
+@@ -669,8 +669,9 @@ void nonrecoverable_exception(struct pt_
void trace_syscall(struct pt_regs *regs)
{
@@ -1305,9 +737,9 @@
regs->ccr&0x10000000?"Error=":"", regs->gpr[3], print_tainted());
}
---- linux-2.6.22/arch/ppc/mm/fault.c 2006-11-30 21:18:32 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/ppc/mm/fault.c 2007-06-15 02:37:03 +0200
-@@ -296,7 +296,8 @@ out_of_memory:
+--- linux-2.6.25.4/arch/ppc/mm/fault.c 2008-04-17 11:31:25.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/ppc/mm/fault.c 2008-04-19 15:14:52.000000000 -0400
+@@ -295,7 +295,8 @@ out_of_memory:
down_read(&mm->mmap_sem);
goto survive;
}
@@ -1315,11 +747,11 @@
+ printk("VM: killing process %s(%d:#%u)\n",
+ current->comm, current->pid, current->xid);
if (user_mode(regs))
- do_exit(SIGKILL);
+ do_group_exit(SIGKILL);
return SIGKILL;
---- linux-2.6.22/arch/s390/Kconfig 2007-07-09 13:18:02 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/s390/Kconfig 2007-06-15 02:37:03 +0200
-@@ -551,6 +551,8 @@ endmenu
+--- linux-2.6.25.4/arch/s390/Kconfig 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/s390/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -544,6 +544,8 @@ source "fs/Kconfig"
source "arch/s390/Kconfig.debug"
@@ -1328,8 +760,8 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/s390/kernel/compat_linux.c 2007-07-09 13:18:02 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/s390/kernel/compat_linux.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/s390/kernel/compat_linux.c 2008-04-17 11:31:25.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/s390/kernel/compat_linux.c 2008-04-19 15:14:52.000000000 -0400
@@ -567,7 +567,7 @@ asmlinkage long sys32_gettimeofday(struc
{
if (tv) {
@@ -1339,32 +771,23 @@
if (put_tv32(tv, &ktv))
return -EFAULT;
}
---- linux-2.6.22/arch/s390/kernel/process.c 2007-07-09 13:18:02 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/s390/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -164,9 +164,9 @@ void show_regs(struct pt_regs *regs)
- struct task_struct *tsk = current;
-
- printk("CPU: %d %s\n", task_thread_info(tsk)->cpu, print_tainted());
-- printk("Process %s (pid: %d, task: %p, ksp: %p)\n",
-- current->comm, current->pid, (void *) tsk,
-- (void *) tsk->thread.ksp);
+--- linux-2.6.25.4/arch/s390/kernel/process.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/s390/kernel/process.c 2008-04-21 11:09:01.000000000 -0400
+@@ -194,9 +194,9 @@ void show_regs(struct pt_regs *regs)
+ init_utsname()->release,
+ (int)strcspn(init_utsname()->version, " "),
+ init_utsname()->version);
+- printk("Process %s (pid: %d, task: %p, ksp: %p)\n",
+- current->comm, current->pid, current,
+- (void *) current->thread.ksp);
+ printk("Process %s (pid: %d[#%u], task: %p, ksp: %p)\n",
+ current->comm, current->pid, current->xid,
-+ (void *) tsk, (void *) tsk->thread.ksp);
-
++ (void *) current, (void *) current->thread.ksp);
show_registers(regs);
/* Show stack backtrace if pt_regs is from kernel mode */
-@@ -197,7 +197,7 @@ int kernel_thread(int (*fn)(void *), voi
- regs.orig_gpr2 = -1;
-
- /* Ok, create the new process.. */
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED,
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
- 0, ®s, 0, NULL, NULL);
- }
-
---- linux-2.6.22/arch/s390/kernel/ptrace.c 2007-05-02 19:24:29 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/s390/kernel/ptrace.c 2007-06-15 02:37:03 +0200
+ if (!(regs->psw.mask & PSW_MASK_PSTATE))
+--- linux-2.6.25.4/arch/s390/kernel/ptrace.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/s390/kernel/ptrace.c 2008-04-19 15:14:52.000000000 -0400
@@ -33,6 +33,7 @@
#include <linux/security.h>
#include <linux/audit.h>
@@ -1373,7 +796,7 @@
#include <asm/segment.h>
#include <asm/page.h>
-@@ -725,7 +726,13 @@ sys_ptrace(long request, long pid, long
+@@ -710,7 +711,13 @@ sys_ptrace(long request, long pid, long
goto out;
}
@@ -1387,8 +810,8 @@
put_task_struct(child);
out:
unlock_kernel();
---- linux-2.6.22/arch/s390/kernel/syscalls.S 2007-07-09 13:18:02 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/s390/kernel/syscalls.S 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/s390/kernel/syscalls.S 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/s390/kernel/syscalls.S 2008-04-19 15:14:52.000000000 -0400
@@ -271,7 +271,7 @@ SYSCALL(sys_clock_settime,sys_clock_sett
SYSCALL(sys_clock_gettime,sys_clock_gettime,sys32_clock_gettime_wrapper) /* 260 */
SYSCALL(sys_clock_getres,sys_clock_getres,sys32_clock_getres_wrapper)
@@ -1398,9 +821,9 @@
SYSCALL(s390_fadvise64_64,sys_ni_syscall,sys32_fadvise64_64_wrapper)
SYSCALL(sys_statfs64,sys_statfs64,compat_sys_statfs64_wrapper)
SYSCALL(sys_fstatfs64,sys_fstatfs64,compat_sys_fstatfs64_wrapper)
---- linux-2.6.22/arch/s390/mm/fault.c 2007-07-09 13:18:02 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/s390/mm/fault.c 2007-06-15 02:37:03 +0200
-@@ -216,7 +216,8 @@ static int do_out_of_memory(struct pt_re
+--- linux-2.6.25.4/arch/s390/mm/fault.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/s390/mm/fault.c 2008-04-19 15:14:52.000000000 -0400
+@@ -217,7 +217,8 @@ static int do_out_of_memory(struct pt_re
down_read(&mm->mmap_sem);
return 1;
}
@@ -1408,11 +831,11 @@
+ printk("VM: killing process %s(%d:#%u)\n",
+ tsk->comm, tsk->pid, tsk->xid);
if (regs->psw.mask & PSW_MASK_PSTATE)
- do_exit(SIGKILL);
+ do_group_exit(SIGKILL);
do_no_context(regs, error_code, address);
---- linux-2.6.22/arch/sh/Kconfig 2007-07-09 13:18:02 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sh/Kconfig 2007-06-15 02:37:03 +0200
-@@ -723,6 +723,8 @@ source "arch/sh/oprofile/Kconfig"
+--- linux-2.6.25.4/arch/sh/Kconfig 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sh/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -913,6 +913,8 @@ source "fs/Kconfig"
source "arch/sh/Kconfig.debug"
@@ -1421,96 +844,29 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/sh/kernel/irq.c 2007-07-09 13:18:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sh/kernel/irq.c 2007-06-15 02:37:03 +0200
-@@ -12,6 +12,7 @@
+--- linux-2.6.25.4/arch/sh/kernel/irq.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sh/kernel/irq.c 2008-04-19 15:14:52.000000000 -0400
+@@ -11,6 +11,7 @@
+ #include <linux/module.h>
#include <linux/kernel_stat.h>
#include <linux/seq_file.h>
- #include <linux/irq.h>
+#include <linux/vs_context.h>
#include <asm/processor.h>
#include <asm/machvec.h>
#include <asm/uaccess.h>
---- linux-2.6.22/arch/sh/kernel/process.c 2007-07-09 13:18:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sh/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -118,7 +118,8 @@ void machine_power_off(void)
- void show_regs(struct pt_regs * regs)
- {
- printk("\n");
-- printk("Pid : %d, Comm: %20s\n", current->pid, current->comm);
-+ printk("Pid : %d:#%u, Comm: %20s\n",
-+ current->pid, current->xid, current->comm);
- print_symbol("PC is at %s\n", instruction_pointer(regs));
- printk("PC : %08lx SP : %08lx SR : %08lx ",
- regs->pc, regs->regs[15], regs->sr);
-@@ -179,7 +180,7 @@ int kernel_thread(int (*fn)(void *), voi
- regs.sr = (1 << 30);
-
- /* Ok, create the new process.. */
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0,
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD, 0,
- ®s, 0, NULL, NULL);
- }
-
---- linux-2.6.22/arch/sh/kernel/vsyscall/vsyscall.c 2007-07-09 13:18:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sh/kernel/vsyscall/vsyscall.c 2007-06-15 02:39:24 +0200
-@@ -18,6 +18,7 @@
- #include <linux/module.h>
+--- linux-2.6.25.4/arch/sh/kernel/vsyscall/vsyscall.c 2008-04-17 10:37:14.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sh/kernel/vsyscall/vsyscall.c 2008-04-19 15:14:52.000000000 -0400
+@@ -19,6 +19,7 @@
#include <linux/elf.h>
#include <linux/sched.h>
+ #include <linux/err.h>
+#include <linux/vs_memory.h>
/*
* Should the kernel map a VDSO page into processes and pass its
---- linux-2.6.22/arch/sh/mm/fault.c 2007-07-09 13:18:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sh/mm/fault.c 2007-06-15 02:37:03 +0200
-@@ -201,7 +201,8 @@ out_of_memory:
- down_read(&mm->mmap_sem);
- goto survive;
- }
-- printk("VM: killing process %s\n", tsk->comm);
-+ printk("VM: killing process %s(%d:#%u)\n",
-+ tsk->comm, tsk->pid, tsk->xid);
- if (user_mode(regs))
- do_exit(SIGKILL);
- goto no_context;
---- linux-2.6.22/arch/sh64/kernel/process.c 2007-07-09 13:18:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sh64/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -400,8 +400,8 @@ int kernel_thread(int (*fn)(void *), voi
- regs.pc = (unsigned long)kernel_thread_helper;
- regs.sr = (1 << 30);
-
-- return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0,
-- ®s, 0, NULL, NULL);
-+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD,
-+ 0, ®s, 0, NULL, NULL);
- }
-
- /*
---- linux-2.6.22/arch/sh64/mm/fault.c 2007-07-09 13:18:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sh64/mm/fault.c 2007-06-15 02:37:03 +0200
-@@ -81,7 +81,7 @@ static inline void print_vma(struct vm_a
-
- static inline void print_task(struct task_struct *tsk)
- {
-- printk("Task pid %d\n", tsk->pid);
-+ printk("Task pid %d:#%u\n", tsk->pid, tsk->xid);
- }
-
- static pte_t *lookup_pte(struct mm_struct *mm, unsigned long address)
-@@ -330,7 +330,8 @@ out_of_memory:
- down_read(&mm->mmap_sem);
- goto survive;
- }
-- printk("VM: killing process %s\n", tsk->comm);
-+ printk("VM: killing process %s(%d:#%u)\n",
-+ tsk->comm, tsk->pid, tsk->xid);
- if (user_mode(regs))
- do_exit(SIGKILL);
- goto no_context;
---- linux-2.6.22/arch/sparc/Kconfig 2007-07-09 13:18:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc/Kconfig 2007-06-15 02:37:03 +0200
-@@ -317,6 +317,8 @@ endmenu
+--- linux-2.6.25.4/arch/sparc/Kconfig 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -330,6 +330,8 @@ source "fs/Kconfig"
source "arch/sparc/Kconfig.debug"
@@ -1519,42 +875,30 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/sparc/kernel/process.c 2007-07-09 13:18:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -705,7 +705,8 @@ pid_t kernel_thread(int (*fn)(void *), v
- /* Notreached by child. */
- "1: mov %%o0, %0\n\t" :
- "=r" (retval) :
-- "i" (__NR_clone), "r" (flags | CLONE_VM | CLONE_UNTRACED),
-+ "i" (__NR_clone), "r" (flags |
-+ CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD),
- "i" (__NR_exit), "r" (fn), "r" (arg) :
- "g1", "g2", "g3", "o0", "o1", "memory", "cc");
- return retval;
---- linux-2.6.22/arch/sparc/kernel/ptrace.c 2007-02-06 03:00:18 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc/kernel/ptrace.c 2007-06-15 02:37:03 +0200
-@@ -19,6 +19,7 @@
- #include <linux/smp_lock.h>
- #include <linux/security.h>
+--- linux-2.6.25.4/arch/sparc/kernel/ptrace.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc/kernel/ptrace.c 2008-05-21 14:30:40.000000000 -0400
+@@ -21,6 +21,7 @@
#include <linux/signal.h>
+ #include <linux/regset.h>
+ #include <linux/elf.h>
+#include <linux/vs_base.h>
#include <asm/pgtable.h>
#include <asm/system.h>
-@@ -302,6 +303,10 @@ asmlinkage void do_ptrace(struct pt_regs
- pt_error_return(regs, -ret);
- goto out;
+@@ -270,6 +271,10 @@ static int fpregs32_set(struct task_stru
+ 33 * sizeof(u32),
+ 34 * sizeof(u32));
}
+ if (!vx_check(vx_task_xid(child), VS_WATCH_P | VS_IDENT)) {
+ pt_error_return(regs, ESRCH);
+ goto out_tsk;
+ }
- if ((current->personality == PER_SUNOS && request == PTRACE_SUNATTACH)
- || (current->personality != PER_SUNOS && request == PTRACE_ATTACH)) {
---- linux-2.6.22/arch/sparc/kernel/systbls.S 2007-07-09 13:18:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc/kernel/systbls.S 2007-06-15 02:37:03 +0200
-@@ -71,7 +71,7 @@ sys_call_table:
+ if (!ret)
+ ret = user_regset_copyin_ignore(&pos, &count, &kbuf, &ubuf,
+--- linux-2.6.25.4/arch/sparc/kernel/systbls.S 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc/kernel/systbls.S 2008-04-19 15:14:52.000000000 -0400
+@@ -70,7 +70,7 @@ sys_call_table:
/*250*/ .long sparc_mremap, sys_sysctl, sys_getsid, sys_fdatasync, sys_nfsservctl
/*255*/ .long sys_sync_file_range, sys_clock_settime, sys_clock_gettime, sys_clock_getres, sys_clock_nanosleep
/*260*/ .long sys_sched_getaffinity, sys_sched_setaffinity, sys_timer_settime, sys_timer_gettime, sys_timer_getoverrun
@@ -1563,20 +907,20 @@
/*270*/ .long sys_io_submit, sys_io_cancel, sys_io_getevents, sys_mq_open, sys_mq_unlink
/*275*/ .long sys_mq_timedsend, sys_mq_timedreceive, sys_mq_notify, sys_mq_getsetattr, sys_waitid
/*280*/ .long sys_tee, sys_add_key, sys_request_key, sys_keyctl, sys_openat
---- linux-2.6.22/arch/sparc/kernel/traps.c 2007-07-09 13:18:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc/kernel/traps.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/sparc/kernel/traps.c 2008-04-17 11:31:25.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
@@ -99,7 +99,8 @@ void die_if_kernel(char *str, struct pt_
" /_| \\__/ |_\\\n"
" \\__U_/\n");
-- printk("%s(%d): %s [#%d]\n", current->comm, current->pid, str, ++die_counter);
+- printk("%s(%d): %s [#%d]\n", current->comm, task_pid_nr(current), str, ++die_counter);
+ printk("%s(%d[#%u]): %s [#%d]\n", current->comm,
-+ current->pid, current->xid, str, ++die_counter);
++ task_pid_nr(current), current->xid, str, ++die_counter);
show_regs(regs);
+ add_taint(TAINT_DIE);
- __SAVE; __SAVE; __SAVE; __SAVE;
---- linux-2.6.22/arch/sparc/mm/fault.c 2007-07-09 13:18:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc/mm/fault.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/sparc/mm/fault.c 2008-04-17 11:31:25.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc/mm/fault.c 2008-04-19 15:14:52.000000000 -0400
@@ -367,7 +367,8 @@ no_context:
*/
out_of_memory:
@@ -1585,11 +929,11 @@
+ printk("VM: killing process %s(%d:#%u)\n",
+ tsk->comm, tsk->pid, tsk->xid);
if (from_user)
- do_exit(SIGKILL);
+ do_group_exit(SIGKILL);
goto no_context;
---- linux-2.6.22/arch/sparc64/Kconfig 2007-07-09 13:18:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/Kconfig 2007-06-17 05:54:16 +0200
-@@ -451,6 +451,8 @@ endmenu
+--- linux-2.6.25.4/arch/sparc64/Kconfig 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc64/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -471,6 +471,8 @@ source "fs/Kconfig"
source "arch/sparc64/Kconfig.debug"
@@ -1598,8 +942,8 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/sparc64/kernel/binfmt_aout32.c 2007-02-06 03:00:18 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/kernel/binfmt_aout32.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/sparc64/kernel/binfmt_aout32.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc64/kernel/binfmt_aout32.c 2008-04-19 15:14:52.000000000 -0400
@@ -27,6 +27,7 @@
#include <linux/binfmts.h>
#include <linux/personality.h>
@@ -1608,42 +952,30 @@
#include <asm/system.h>
#include <asm/uaccess.h>
---- linux-2.6.22/arch/sparc64/kernel/process.c 2007-07-09 13:18:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -707,7 +707,8 @@ pid_t kernel_thread(int (*fn)(void *), v
- /* Notreached by child. */
- "1:" :
- "=r" (retval) :
-- "i" (__NR_clone), "r" (flags | CLONE_VM | CLONE_UNTRACED),
-+ "i" (__NR_clone), "r" (flags |
-+ CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD),
- "i" (__NR_exit), "r" (fn), "r" (arg) :
- "g1", "g2", "g3", "o0", "o1", "memory", "cc");
- return retval;
---- linux-2.6.22/arch/sparc64/kernel/ptrace.c 2007-02-06 03:00:18 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/kernel/ptrace.c 2007-06-15 02:37:03 +0200
-@@ -22,6 +22,7 @@
- #include <linux/seccomp.h>
- #include <linux/audit.h>
- #include <linux/signal.h>
+--- linux-2.6.25.4/arch/sparc64/kernel/ptrace.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc64/kernel/ptrace.c 2008-05-21 14:30:40.000000000 -0400
+@@ -25,6 +25,7 @@
+ #include <linux/regset.h>
+ #include <linux/compat.h>
+ #include <linux/elf.h>
+#include <linux/vs_base.h>
#include <asm/asi.h>
#include <asm/pgtable.h>
-@@ -215,6 +216,10 @@ asmlinkage void do_ptrace(struct pt_regs
- pt_error_return(regs, -ret);
- goto out;
+@@ -222,6 +223,10 @@ static int genregs64_get(struct task_str
+ 16 * sizeof(u64),
+ 32 * sizeof(u64));
}
+ if (!vx_check(vx_task_xid(child), VS_WATCH_P | VS_IDENT)) {
+ pt_error_return(regs, ESRCH);
+ goto out_tsk;
+ }
- if ((current->personality == PER_SUNOS && request == PTRACE_SUNATTACH)
- || (current->personality != PER_SUNOS && request == PTRACE_ATTACH)) {
---- linux-2.6.22/arch/sparc64/kernel/sys_sparc32.c 2007-07-09 13:18:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/kernel/sys_sparc32.c 2007-06-15 02:37:03 +0200
-@@ -729,7 +729,7 @@ asmlinkage long sys32_gettimeofday(struc
+ if (!ret) {
+ /* TSTATE, TPC, TNPC */
+--- linux-2.6.25.4/arch/sparc64/kernel/sys_sparc32.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc64/kernel/sys_sparc32.c 2008-05-21 14:30:40.000000000 -0400
+@@ -722,7 +722,7 @@ asmlinkage long sys32_gettimeofday(struc
{
if (tv) {
struct timeval ktv;
@@ -1652,9 +984,9 @@
if (put_tv32(tv, &ktv))
return -EFAULT;
}
---- linux-2.6.22/arch/sparc64/kernel/systbls.S 2007-07-09 13:18:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/kernel/systbls.S 2007-06-15 02:37:03 +0200
-@@ -72,7 +72,7 @@ sys_call_table32:
+--- linux-2.6.25.4/arch/sparc64/kernel/systbls.S 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc64/kernel/systbls.S 2008-04-19 15:14:52.000000000 -0400
+@@ -71,7 +71,7 @@ sys_call_table32:
/*250*/ .word sys32_mremap, sys32_sysctl, sys32_getsid, sys_fdatasync, sys32_nfsservctl
.word sys32_sync_file_range, compat_sys_clock_settime, compat_sys_clock_gettime, compat_sys_clock_getres, sys32_clock_nanosleep
/*260*/ .word compat_sys_sched_getaffinity, compat_sys_sched_setaffinity, sys32_timer_settime, compat_sys_timer_gettime, sys_timer_getoverrun
@@ -1672,21 +1004,21 @@
/*270*/ .word sys_io_submit, sys_io_cancel, sys_io_getevents, sys_mq_open, sys_mq_unlink
.word sys_mq_timedsend, sys_mq_timedreceive, sys_mq_notify, sys_mq_getsetattr, sys_waitid
/*280*/ .word sys_tee, sys_add_key, sys_request_key, sys_keyctl, sys_openat
---- linux-2.6.22/arch/sparc64/kernel/traps.c 2007-07-09 13:18:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/kernel/traps.c 2007-06-15 02:37:03 +0200
-@@ -2221,7 +2221,8 @@ void die_if_kernel(char *str, struct pt_
+--- linux-2.6.25.4/arch/sparc64/kernel/traps.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc64/kernel/traps.c 2008-04-19 15:14:52.000000000 -0400
+@@ -2183,7 +2183,8 @@ void die_if_kernel(char *str, struct pt_
" /_| \\__/ |_\\\n"
" \\__U_/\n");
-- printk("%s(%d): %s [#%d]\n", current->comm, current->pid, str, ++die_counter);
+- printk("%s(%d): %s [#%d]\n", current->comm, task_pid_nr(current), str, ++die_counter);
+ printk("%s(%d[#%u]): %s [#%d]\n", current->comm,
-+ current->pid, current->xid, str, ++die_counter);
++ task_pid_nr(current), current->xid, str, ++die_counter);
notify_die(DIE_OOPS, str, regs, 0, 255, SIGSEGV);
__asm__ __volatile__("flushw");
__show_regs(regs);
---- linux-2.6.22/arch/sparc64/mm/fault.c 2007-07-09 13:18:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/mm/fault.c 2007-06-15 02:37:03 +0200
-@@ -466,7 +466,8 @@ handle_kernel_fault:
+--- linux-2.6.25.4/arch/sparc64/mm/fault.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc64/mm/fault.c 2008-04-19 15:14:52.000000000 -0400
+@@ -453,7 +453,8 @@ handle_kernel_fault:
out_of_memory:
insn = get_fault_insn(regs, insn);
up_read(&mm->mmap_sem);
@@ -1694,10 +1026,10 @@
+ printk("VM: killing process %s(%d:#%u)\n",
+ current->comm, current->pid, current->xid);
if (!(regs->tstate & TSTATE_PRIV))
- do_exit(SIGKILL);
+ do_group_exit(SIGKILL);
goto handle_kernel_fault;
---- linux-2.6.22/arch/sparc64/solaris/fs.c 2007-02-06 03:00:21 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/sparc64/solaris/fs.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/sparc64/solaris/fs.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/sparc64/solaris/fs.c 2008-04-19 15:14:52.000000000 -0400
@@ -368,7 +368,7 @@ static int report_statvfs(struct vfsmoun
int j = strlen (p);
@@ -1716,9 +1048,9 @@
if (mnt->mnt_flags & MNT_NOSUID) i |= 2;
if (!sysv_valid_dev(inode->i_sb->s_dev))
return -EOVERFLOW;
---- linux-2.6.22/arch/um/Kconfig 2007-07-09 13:18:07 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/um/Kconfig 2007-06-17 05:54:16 +0200
-@@ -316,6 +316,8 @@ source "drivers/connector/Kconfig"
+--- linux-2.6.25.4/arch/um/Kconfig 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/um/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -245,6 +245,8 @@ source "drivers/connector/Kconfig"
source "fs/Kconfig"
@@ -1727,21 +1059,21 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/um/kernel/trap.c 2007-07-09 13:18:07 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/um/kernel/trap.c 2007-06-15 02:37:03 +0200
-@@ -210,7 +210,8 @@ unsigned long segv(struct faultinfo fi,
+--- linux-2.6.25.4/arch/um/kernel/trap.c 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/um/kernel/trap.c 2008-04-19 15:14:52.000000000 -0400
+@@ -215,7 +215,8 @@ unsigned long segv(struct faultinfo fi,
current->thread.arch.faultinfo = fi;
force_sig_info(SIGBUS, &si, current);
} else if (err == -ENOMEM) {
-- printk("VM: killing process %s\n", current->comm);
-+ printk("VM: killing process %s(%d:#%u)\n",
-+ current->comm, current->pid, current->xid);
+- printk(KERN_INFO "VM: killing process %s\n", current->comm);
++ printk(KERN_INFO "VM: killing process %s(%d:#%u)\n",
++ current->comm, task_pid_nr(current), current->xid);
do_exit(SIGKILL);
} else {
BUG_ON(err != -EFAULT);
---- linux-2.6.22/arch/v850/Kconfig 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/v850/Kconfig 2007-06-15 02:37:03 +0200
-@@ -333,6 +333,8 @@ source "drivers/usb/Kconfig"
+--- linux-2.6.25.4/arch/v850/Kconfig 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/v850/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -344,6 +344,8 @@ source "drivers/usb/Kconfig"
source "arch/v850/Kconfig.debug"
@@ -1750,19 +1082,8 @@
source "security/Kconfig"
source "crypto/Kconfig"
---- linux-2.6.22/arch/v850/kernel/process.c 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/v850/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -82,7 +82,7 @@ int kernel_thread (int (*fn)(void *), vo
- /* Clone this thread. Note that we don't pass the clone syscall's
- second argument -- it's ignored for calls from kernel mode (the
- child's SP is always set to the top of the kernel stack). */
-- arg0 = flags | CLONE_VM;
-+ arg0 = flags | CLONE_VM | CLONE_KTHREAD;
- syscall = __NR_clone;
- asm volatile ("trap " SYSCALL_SHORT_TRAP
- : "=r" (ret), "=r" (syscall)
---- linux-2.6.22/arch/v850/kernel/ptrace.c 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/v850/kernel/ptrace.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/arch/v850/kernel/ptrace.c 2008-04-17 10:37:14.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/v850/kernel/ptrace.c 2008-04-19 15:14:52.000000000 -0400
@@ -23,6 +23,7 @@
#include <linux/sched.h>
#include <linux/ptrace.h>
@@ -1779,44 +1100,11 @@
+ goto out;
+
switch (request) {
- unsigned long val, copied;
-
---- linux-2.6.22/arch/x86_64/Kconfig 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/Kconfig 2007-06-15 02:37:03 +0200
-@@ -782,6 +782,8 @@ endmenu
-
- source "arch/x86_64/Kconfig.debug"
-
-+source "kernel/vserver/Kconfig"
-+
- source "security/Kconfig"
-
- source "crypto/Kconfig"
---- linux-2.6.22/arch/x86_64/ia32/ia32_aout.c 2007-02-06 03:00:21 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/ia32/ia32_aout.c 2007-06-15 02:37:03 +0200
-@@ -25,6 +25,7 @@
- #include <linux/binfmts.h>
- #include <linux/personality.h>
- #include <linux/init.h>
-+#include <linux/vs_memory.h>
+ unsigned long val;
- #include <asm/system.h>
- #include <asm/uaccess.h>
---- linux-2.6.22/arch/x86_64/ia32/ia32_binfmt.c 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/ia32/ia32_binfmt.c 2007-06-15 02:37:03 +0200
-@@ -324,7 +324,8 @@ int ia32_setup_arg_pages(struct linux_bi
- kmem_cache_free(vm_area_cachep, mpnt);
- return ret;
- }
-- mm->stack_vm = mm->total_vm = vma_pages(mpnt);
-+ vx_vmpages_sub(mm, mm->total_vm - vma_pages(mpnt));
-+ mm->stack_vm = mm->total_vm;
- }
-
- for (i = 0 ; i < MAX_ARG_PAGES ; i++) {
---- linux-2.6.22/arch/x86_64/ia32/ia32entry.S 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/ia32/ia32entry.S 2007-07-07 03:52:53 +0200
-@@ -668,7 +668,7 @@ ia32_sys_call_table:
+--- linux-2.6.25.4/arch/x86/ia32/ia32entry.S 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/x86/ia32/ia32entry.S 2008-04-22 20:07:49.000000000 -0400
+@@ -673,7 +673,7 @@ ia32_sys_call_table:
.quad sys_tgkill /* 270 */
.quad compat_sys_utimes
.quad sys32_fadvise64_64
@@ -1825,219 +1113,301 @@
.quad sys_mbind
.quad compat_sys_get_mempolicy /* 275 */
.quad sys_set_mempolicy
---- linux-2.6.22/arch/x86_64/ia32/sys_ia32.c 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/ia32/sys_ia32.c 2007-07-07 03:52:53 +0200
-@@ -454,7 +454,7 @@ sys32_gettimeofday(struct compat_timeval
- {
- if (tv) {
- struct timeval ktv;
-- do_gettimeofday(&ktv);
-+ vx_gettimeofday(&ktv);
- if (put_tv32(tv, &ktv))
- return -EFAULT;
- }
---- linux-2.6.22/arch/x86_64/ia32/syscall32.c 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/ia32/syscall32.c 2007-06-15 02:37:03 +0200
-@@ -10,6 +10,7 @@
- #include <linux/init.h>
- #include <linux/stringify.h>
- #include <linux/security.h>
-+#include <linux/vs_memory.h>
- #include <asm/proto.h>
- #include <asm/tlbflush.h>
- #include <asm/ia32_unistd.h>
---- linux-2.6.22/arch/x86_64/kernel/process.c 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/kernel/process.c 2007-06-15 02:37:03 +0200
-@@ -54,7 +54,8 @@
-
- asmlinkage extern void ret_from_fork(void);
-
--unsigned long kernel_thread_flags = CLONE_VM | CLONE_UNTRACED;
-+unsigned long kernel_thread_flags =
-+ CLONE_VM | CLONE_UNTRACED | CLONE_KTHREAD;
-
- unsigned long boot_option_idle_override = 0;
- EXPORT_SYMBOL(boot_option_idle_override);
-@@ -310,8 +311,8 @@ void __show_regs(struct pt_regs * regs)
-
- printk("\n");
- print_modules();
-- printk("Pid: %d, comm: %.20s %s %s %.*s\n",
-- current->pid, current->comm, print_tainted(),
-+ printk("Pid: %d:#%u, comm: %.20s %s %s %.*s\n",
-+ current->pid, current->xid, current->comm, print_tainted(),
- init_utsname()->release,
- (int)strcspn(init_utsname()->version, " "),
- init_utsname()->version);
---- linux-2.6.22/arch/x86_64/kernel/traps.c 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/kernel/traps.c 2007-06-17 05:54:16 +0200
-@@ -413,8 +413,9 @@ void show_registers(struct pt_regs *regs
- rsp = regs->rsp;
- printk("CPU %d ", cpu);
- __show_regs(regs);
-- printk("Process %s (pid: %d, threadinfo %p, task %p)\n",
-- cur->comm, cur->pid, task_thread_info(cur), cur);
-+ printk("Process %s (pid: %d[#%u], threadinfo %p, task %p)\n",
-+ cur->comm, cur->pid, cur->xid,
-+ task_thread_info(cur), cur);
-
- /*
- * When in-kernel, we also print out the stack and code at the
-@@ -580,8 +581,8 @@ static void __kprobes do_trap(int trapnr
+--- linux-2.6.25.4/arch/x86/Kconfig 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/x86/Kconfig 2008-05-21 14:30:40.000000000 -0400
+@@ -1623,6 +1623,8 @@ source "fs/Kconfig"
- if (exception_trace && unhandled_signal(tsk, signr))
- printk(KERN_INFO
-- "%s[%d] trap %s rip:%lx rsp:%lx error:%lx\n",
-- tsk->comm, tsk->pid, str,
-+ "%s[%d:#%u] trap %s rip:%lx rsp:%lx error:%lx\n",
-+ tsk->comm, tsk->pid, tsk->xid, str,
- regs->rip, regs->rsp, error_code);
-
- if (info)
-@@ -684,8 +685,8 @@ asmlinkage void __kprobes do_general_pro
-
- if (exception_trace && unhandled_signal(tsk, SIGSEGV))
- printk(KERN_INFO
-- "%s[%d] general protection rip:%lx rsp:%lx error:%lx\n",
-- tsk->comm, tsk->pid,
-+ "%s[%d:#%u] general protection rip:%lx rsp:%lx error:%lx\n",
-+ tsk->comm, tsk->pid, tsk->xid,
- regs->rip, regs->rsp, error_code);
-
- force_sig(SIGSEGV, tsk);
---- linux-2.6.22/arch/x86_64/mm/fault.c 2007-07-09 13:18:08 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/arch/x86_64/mm/fault.c 2007-06-17 05:54:16 +0200
-@@ -497,10 +497,10 @@ bad_area_nosemaphore:
-
- if (exception_trace && unhandled_signal(tsk, SIGSEGV)) {
- printk(
-- "%s%s[%d]: segfault at %016lx rip %016lx rsp %016lx error %lx\n",
-+ "%s%s[%d:#%u]: segfault at %016lx rip %016lx rsp %016lx error %lx\n",
- tsk->pid > 1 ? KERN_INFO : KERN_EMERG,
-- tsk->comm, tsk->pid, address, regs->rip,
-- regs->rsp, error_code);
-+ tsk->comm, tsk->pid, tsk->xid, address,
-+ regs->rip, regs->rsp, error_code);
- }
-
- tsk->thread.cr2 = address;
-@@ -567,7 +567,8 @@ out_of_memory:
- yield();
- goto again;
- }
-- printk("VM: killing process %s\n", tsk->comm);
-+ printk("VM: killing process %s(%d:#%u)\n",
-+ tsk->comm, tsk->pid, tsk->xid);
- if (error_code & 4)
- do_exit(SIGKILL);
- goto no_context;
---- linux-2.6.22/drivers/block/Kconfig 2007-07-09 13:18:16 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/block/Kconfig 2007-06-15 02:37:03 +0200
-@@ -311,6 +311,13 @@ config BLK_DEV_CRYPTOLOOP
- instead, which can be configured to be on-disk compatible with the
- cryptoloop device.
+ source "arch/x86/Kconfig.debug"
-+config BLK_DEV_VROOT
-+ tristate "Virtual Root device support"
-+ depends on QUOTACTL
-+ ---help---
-+ Saying Y here will allow you to use quota/fs ioctls on a shared
-+ partition within a virtual server without compromising security.
++source "kernel/vserver/Kconfig"
+
- config BLK_DEV_NBD
- tristate "Network block device support"
- depends on NET
---- linux-2.6.22/drivers/block/Makefile 2007-02-06 03:00:26 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/block/Makefile 2007-06-15 02:37:03 +0200
-@@ -28,4 +28,5 @@ obj-$(CONFIG_BLK_DEV_CRYPTOLOOP) += cryp
- obj-$(CONFIG_VIODASD) += viodasd.o
- obj-$(CONFIG_BLK_DEV_SX8) += sx8.o
- obj-$(CONFIG_BLK_DEV_UB) += ub.o
-+obj-$(CONFIG_BLK_DEV_VROOT) += vroot.o
-
---- linux-2.6.22/drivers/block/loop.c 2007-07-09 13:18:16 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/block/loop.c 2007-06-17 05:54:16 +0200
-@@ -74,6 +74,7 @@
- #include <linux/highmem.h>
- #include <linux/gfp.h>
- #include <linux/kthread.h>
-+#include <linux/vs_context.h>
-
- #include <asm/uaccess.h>
-
-@@ -790,6 +791,7 @@ static int loop_set_fd(struct loop_devic
- lo->lo_blocksize = lo_blocksize;
- lo->lo_device = bdev;
- lo->lo_flags = lo_flags;
-+ lo->lo_xid = vx_current_xid();
- lo->lo_backing_file = file;
- lo->transfer = transfer_none;
- lo->ioctl = NULL;
-@@ -930,7 +932,7 @@ loop_set_status(struct loop_device *lo,
- struct loop_func_table *xfer;
-
- if (lo->lo_encrypt_key_size && lo->lo_key_owner != current->uid &&
-- !capable(CAP_SYS_ADMIN))
-+ !vx_capable(CAP_SYS_ADMIN, VXC_ADMIN_CLOOP))
- return -EPERM;
- if (lo->lo_state != Lo_bound)
- return -ENXIO;
-@@ -1010,7 +1012,8 @@ loop_get_status(struct loop_device *lo,
- memcpy(info->lo_crypt_name, lo->lo_crypt_name, LO_NAME_SIZE);
- info->lo_encrypt_type =
- lo->lo_encryption ? lo->lo_encryption->number : 0;
-- if (lo->lo_encrypt_key_size && capable(CAP_SYS_ADMIN)) {
-+ if (lo->lo_encrypt_key_size &&
-+ vx_capable(CAP_SYS_ADMIN, VXC_ADMIN_CLOOP)) {
- info->lo_encrypt_key_size = lo->lo_encrypt_key_size;
- memcpy(info->lo_encrypt_key, lo->lo_encrypt_key,
- lo->lo_encrypt_key_size);
-@@ -1321,6 +1324,9 @@ static int lo_open(struct inode *inode,
- {
- struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
+ source "security/Kconfig"
-+ if (!vx_check(lo->lo_xid, VS_IDENT|VS_HOSTID))
-+ return -EACCES;
-+
- mutex_lock(&lo->lo_ctl_mutex);
- lo->lo_refcnt++;
- mutex_unlock(&lo->lo_ctl_mutex);
---- linux-2.6.22/drivers/block/vroot.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/block/vroot.c 2007-06-15 02:37:03 +0200
-@@ -0,0 +1,281 @@
-+/*
-+ * linux/drivers/block/vroot.c
-+ *
-+ * written by Herbert Pötzl, 9/11/2002
-+ * ported to 2.6.10 by Herbert Pötzl, 30/12/2004
-+ *
-+ * based on the loop.c code by Theodore Ts'o.
-+ *
-+ * Copyright (C) 2002-2007 by Herbert Pötzl.
-+ * Redistribution of this file is permitted under the
-+ * GNU General Public License.
-+ *
-+ */
-+
-+#include <linux/module.h>
-+#include <linux/moduleparam.h>
-+#include <linux/file.h>
-+#include <linux/major.h>
-+#include <linux/blkdev.h>
-+
-+#include <linux/vroot.h>
-+#include <linux/vs_context.h>
-+
+ source "crypto/Kconfig"
+--- linux-2.6.25.4/arch/x86/kernel/syscall_table_32.S 2008-04-17 12:05:30.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/arch/x86/kernel/syscall_table_32.S 2008-04-19 15:14:52.000000000 -0400
+@@ -272,7 +272,7 @@ ENTRY(sys_call_table)
+ .long sys_tgkill /* 270 */
+ .long sys_utimes
+ .long sys_fadvise64_64
+- .long sys_ni_syscall /* sys_vserver */
++ .long sys_vserver
+ .long sys_mbind
+ .long sys_get_mempolicy
+ .long sys_set_mempolicy
+--- linux-2.6.25.4/Documentation/vserver/debug.txt 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/Documentation/vserver/debug.txt 2008-04-19 15:14:51.000000000 -0400
+@@ -0,0 +1,154 @@
+
-+static int max_vroot = 8;
++debug_cvirt:
+
-+static struct vroot_device *vroot_dev;
-+static struct gendisk **disks;
++ 2 4 "vx_map_tgid: %p/%llx: %d -> %d"
++ "vx_rmap_tgid: %p/%llx: %d -> %d"
+
++debug_dlim:
+
-+static int vroot_set_dev(
-+ struct vroot_device *vr,
-+ struct file *vr_file,
++ 0 1 "ALLOC (%p,#%d)%c inode (%d)"
++ "FREE (%p,#%d)%c inode"
++ 1 2 "ALLOC (%p,#%d)%c %lld bytes (%d)"
++ "FREE (%p,#%d)%c %lld bytes"
++ 2 4 "ADJUST: %lld,%lld on %ld,%ld [mult=%d]"
++ 3 8 "ext3_has_free_blocks(%p): %lu<%lu+1, %c, %u!=%u r=%d"
++ "ext3_has_free_blocks(%p): free=%lu, root=%lu"
++ "rcu_free_dl_info(%p)"
++ 4 10 "alloc_dl_info(%p,%d) = %p"
++ "dealloc_dl_info(%p)"
++ "get_dl_info(%p[#%d.%d])"
++ "put_dl_info(%p[#%d.%d])"
++ 5 20 "alloc_dl_info(%p,%d)*"
++ 6 40 "__hash_dl_info: %p[#%d]"
++ "__unhash_dl_info: %p[#%d]"
++ 7 80 "locate_dl_info(%p,#%d) = %p"
++
++debug_misc:
++
++ 0 1 "destroy_dqhash: %p [#0x%08x] c=%d"
++ "new_dqhash: %p [#0x%08x]"
++ "vroot[%d]_clr_dev: dev=%p[%lu,%d:%d]"
++ "vroot[%d]_get_real_bdev: dev=%p[%lu,%d:%d]"
++ "vroot[%d]_set_dev: dev=%p[%lu,%d:%d]"
++ "vroot_get_real_bdev not set"
++ 1 2 "cow_break_link(»%s«)"
++ "temp copy »%s«"
++ 2 4 "dentry_open(new): %p"
++ "dentry_open(old): %p"
++ "lookup_create(new): %p"
++ "old path »%s«"
++ "path_lookup(old): %d"
++ "vfs_create(new): %d"
++ "vfs_rename: %d"
++ "vfs_sendfile: %d"
++ 3 8 "fput(new_file=%p[#%d])"
++ "fput(old_file=%p[#%d])"
++ 4 10 "vx_info_kill(%p[#%d],%d,%d) = %d"
++ "vx_info_kill(%p[#%d],%d,%d)*"
++ 5 20 "vs_reboot(%p[#%d],%d)"
++ 6 40 "dropping task %p[#%u,%u] for %p[#%u,%u]"
++
++debug_net:
++
++ 2 4 "nx_addr_conflict(%p,%p) %d.%d,%d.%d"
++ 3 8 "inet_bind(%p) %d.%d.%d.%d, %d.%d.%d.%d, %d.%d.%d.%d"
++ "inet_bind(%p)* %p,%p;%lx %d.%d.%d.%d"
++ 4 10 "ip_route_connect(%p) %p,%p;%lx"
++ 5 20 "__addr_in_socket(%p,%d.%d.%d.%d) %p:%d.%d.%d.%d %p;%lx"
++ 6 40 "sk,egf: %p [#%d] (from %d)"
++ "sk,egn: %p [#%d] (from %d)"
++ "sk,req: %p [#%d] (from %d)"
++ "sk: %p [#%d] (from %d)"
++ "tw: %p [#%d] (from %d)"
++ 7 80 "__sock_recvmsg: %p[%p,%p,%p;%d]:%d/%d"
++ "__sock_sendmsg: %p[%p,%p,%p;%d]:%d/%d"
++
++debug_nid:
++
++ 0 1 "__lookup_nx_info(#%u): %p[#%u]"
++ "alloc_nx_info(%d) = %p"
++ "create_nx_info(%d) (dynamic rejected)"
++ "create_nx_info(%d) = %p (already there)"
++ "create_nx_info(%d) = %p (new)"
++ "dealloc_nx_info(%p)"
++ 1 2 "alloc_nx_info(%d)*"
++ "create_nx_info(%d)*"
++ 2 4 "get_nx_info(%p[#%d.%d])"
++ "put_nx_info(%p[#%d.%d])"
++ 3 8 "claim_nx_info(%p[#%d.%d.%d]) %p"
++ "clr_nx_info(%p[#%d.%d])"
++ "init_nx_info(%p[#%d.%d])"
++ "release_nx_info(%p[#%d.%d.%d]) %p"
++ "set_nx_info(%p[#%d.%d])"
++ 4 10 "__hash_nx_info: %p[#%d]"
++ "__nx_dynamic_id: [#%d]"
++ "__unhash_nx_info: %p[#%d.%d.%d]"
++ 5 20 "moved task %p into nxi:%p[#%d]"
++ "nx_migrate_task(%p,%p[#%d.%d.%d])"
++ "task_get_nx_info(%p)"
++ 6 40 "nx_clear_persistent(%p[#%d])"
++
++debug_quota:
++
++ 0 1 "quota_sync_dqh(%p,%d) discard inode %p"
++ 1 2 "quota_sync_dqh(%p,%d)"
++ "sync_dquots(%p,%d)"
++ "sync_dquots_dqh(%p,%d)"
++ 3 8 "do_quotactl(%p,%d,cmd=%d,id=%d,%p)"
++
++debug_switch:
++
++ 0 1 "vc: VCMD_%02d_%d[%d], %d,%p [%d,%d,%x,%x]"
++ 1 2 "vc: VCMD_%02d_%d[%d] = %08lx(%ld) [%d,%d]"
++ 4 10 "%s: (%s %s) returned %s with %d"
++
++debug_tag:
++
++ 7 80 "dx_parse_tag(»%s«): %d:#%d"
++ "dx_propagate_tag(%p[#%lu.%d]): %d,%d"
++
++debug_xid:
++
++ 0 1 "__lookup_vx_info(#%u): %p[#%u]"
++ "alloc_vx_info(%d) = %p"
++ "alloc_vx_info(%d)*"
++ "create_vx_info(%d) (dynamic rejected)"
++ "create_vx_info(%d) = %p (already there)"
++ "create_vx_info(%d) = %p (new)"
++ "dealloc_vx_info(%p)"
++ "loc_vx_info(%d) = %p (found)"
++ "loc_vx_info(%d) = %p (new)"
++ "loc_vx_info(%d) = %p (not available)"
++ 1 2 "create_vx_info(%d)*"
++ "loc_vx_info(%d)*"
++ 2 4 "get_vx_info(%p[#%d.%d])"
++ "put_vx_info(%p[#%d.%d])"
++ 3 8 "claim_vx_info(%p[#%d.%d.%d]) %p"
++ "clr_vx_info(%p[#%d.%d])"
++ "init_vx_info(%p[#%d.%d])"
++ "release_vx_info(%p[#%d.%d.%d]) %p"
++ "set_vx_info(%p[#%d.%d])"
++ 4 10 "__hash_vx_info: %p[#%d]"
++ "__unhash_vx_info: %p[#%d.%d.%d]"
++ "__vx_dynamic_id: [#%d]"
++ 5 20 "enter_vx_info(%p[#%d],%p) %p[#%d,%p]"
++ "leave_vx_info(%p[#%d,%p]) %p[#%d,%p]"
++ "moved task %p into vxi:%p[#%d]"
++ "task_get_vx_info(%p)"
++ "vx_migrate_task(%p,%p[#%d.%d])"
++ 6 40 "vx_clear_persistent(%p[#%d])"
++ "vx_exit_init(%p[#%d],%p[#%d,%d,%d])"
++ "vx_set_init(%p[#%d],%p[#%d,%d,%d])"
++ "vx_set_persistent(%p[#%d])"
++ "vx_set_reaper(%p[#%d],%p[#%d,%d])"
++ 7 80 "vx_child_reaper(%p[#%u,%u]) = %p[#%u,%u]"
++
++
++debug_limit:
++
++ n 2^n "vx_acc_cres[%5d,%s,%2d]: %5d%s"
++ "vx_cres_avail[%5d,%s,%2d]: %5ld > %5d + %5d"
++
++ m 2^m "vx_acc_page[%5d,%s,%2d]: %5d%s"
++ "vx_acc_pages[%5d,%s,%2d]: %5d += %5d"
++ "vx_pages_avail[%5d,%s,%2d]: %5ld > %5d + %5d"
+--- linux-2.6.25.4/drivers/block/Kconfig 2008-04-17 12:05:31.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/block/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -263,6 +263,13 @@ config BLK_DEV_CRYPTOLOOP
+ instead, which can be configured to be on-disk compatible with the
+ cryptoloop device.
+
++config BLK_DEV_VROOT
++ tristate "Virtual Root device support"
++ depends on QUOTACTL
++ ---help---
++ Saying Y here will allow you to use quota/fs ioctls on a shared
++ partition within a virtual server without compromising security.
++
+ config BLK_DEV_NBD
+ tristate "Network block device support"
+ depends on NET
+--- linux-2.6.25.4/drivers/block/loop.c 2008-04-17 12:05:32.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/block/loop.c 2008-04-19 15:14:52.000000000 -0400
+@@ -76,6 +76,7 @@
+ #include <linux/gfp.h>
+ #include <linux/kthread.h>
+ #include <linux/splice.h>
++#include <linux/vs_context.h>
+
+ #include <asm/uaccess.h>
+
+@@ -789,6 +790,7 @@ static int loop_set_fd(struct loop_devic
+ lo->lo_blocksize = lo_blocksize;
+ lo->lo_device = bdev;
+ lo->lo_flags = lo_flags;
++ lo->lo_xid = vx_current_xid();
+ lo->lo_backing_file = file;
+ lo->transfer = transfer_none;
+ lo->ioctl = NULL;
+@@ -908,6 +910,7 @@ static int loop_clr_fd(struct loop_devic
+ lo->lo_encrypt_key_size = 0;
+ lo->lo_flags = 0;
+ lo->lo_thread = NULL;
++ lo->lo_xid = 0;
+ memset(lo->lo_encrypt_key, 0, LO_KEY_SIZE);
+ memset(lo->lo_crypt_name, 0, LO_NAME_SIZE);
+ memset(lo->lo_file_name, 0, LO_NAME_SIZE);
+@@ -929,7 +932,7 @@ loop_set_status(struct loop_device *lo,
+ struct loop_func_table *xfer;
+
+ if (lo->lo_encrypt_key_size && lo->lo_key_owner != current->uid &&
+- !capable(CAP_SYS_ADMIN))
++ !vx_capable(CAP_SYS_ADMIN, VXC_ADMIN_CLOOP))
+ return -EPERM;
+ if (lo->lo_state != Lo_bound)
+ return -ENXIO;
+@@ -1013,7 +1016,8 @@ loop_get_status(struct loop_device *lo,
+ memcpy(info->lo_crypt_name, lo->lo_crypt_name, LO_NAME_SIZE);
+ info->lo_encrypt_type =
+ lo->lo_encryption ? lo->lo_encryption->number : 0;
+- if (lo->lo_encrypt_key_size && capable(CAP_SYS_ADMIN)) {
++ if (lo->lo_encrypt_key_size &&
++ vx_capable(CAP_SYS_ADMIN, VXC_ADMIN_CLOOP)) {
+ info->lo_encrypt_key_size = lo->lo_encrypt_key_size;
+ memcpy(info->lo_encrypt_key, lo->lo_encrypt_key,
+ lo->lo_encrypt_key_size);
+@@ -1322,6 +1326,9 @@ static int lo_open(struct inode *inode,
+ {
+ struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
+
++ if (!vx_check(lo->lo_xid, VS_IDENT|VS_HOSTID|VS_ADMIN_P))
++ return -EACCES;
++
+ mutex_lock(&lo->lo_ctl_mutex);
+ lo->lo_refcnt++;
+ mutex_unlock(&lo->lo_ctl_mutex);
+--- linux-2.6.25.4/drivers/block/Makefile 2008-04-17 12:05:31.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/block/Makefile 2008-04-19 15:14:52.000000000 -0400
+@@ -29,5 +29,6 @@ obj-$(CONFIG_VIRTIO_BLK) += virtio_blk.o
+ obj-$(CONFIG_VIODASD) += viodasd.o
+ obj-$(CONFIG_BLK_DEV_SX8) += sx8.o
+ obj-$(CONFIG_BLK_DEV_UB) += ub.o
++obj-$(CONFIG_BLK_DEV_VROOT) += vroot.o
+
+ obj-$(CONFIG_XEN_BLKDEV_FRONTEND) += xen-blkfront.o
+--- linux-2.6.25.4/drivers/block/vroot.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/block/vroot.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,280 @@
++/*
++ * linux/drivers/block/vroot.c
++ *
++ * written by Herbert Pötzl, 9/11/2002
++ * ported to 2.6.10 by Herbert Pötzl, 30/12/2004
++ *
++ * based on the loop.c code by Theodore Ts'o.
++ *
++ * Copyright (C) 2002-2007 by Herbert Pötzl.
++ * Redistribution of this file is permitted under the
++ * GNU General Public License.
++ *
++ */
++
++#include <linux/module.h>
++#include <linux/moduleparam.h>
++#include <linux/file.h>
++#include <linux/major.h>
++#include <linux/blkdev.h>
++
++#include <linux/vroot.h>
++#include <linux/vs_context.h>
++
++
++static int max_vroot = 8;
++
++static struct vroot_device *vroot_dev;
++static struct gendisk **disks;
++
++
++static int vroot_set_dev(
++ struct vroot_device *vr,
++ struct file *vr_file,
+ struct block_device *bdev,
+ unsigned int arg)
+{
@@ -2264,8 +1634,7 @@
+ del_gendisk(disks[i]);
+ put_disk(disks[i]);
+ }
-+ if (unregister_blkdev(VROOT_MAJOR, "vroot"))
-+ printk(KERN_WARNING "vroot: cannot unregister blkdev\n");
++ unregister_blkdev(VROOT_MAJOR, "vroot");
+
+ kfree(disks);
+ kfree(vroot_dev);
@@ -2286,17 +1655,17 @@
+
+#endif
+
---- linux-2.6.22/drivers/char/sysrq.c 2007-07-09 13:18:20 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/char/sysrq.c 2007-06-15 02:37:03 +0200
-@@ -36,6 +36,7 @@
- #include <linux/kexec.h>
+--- linux-2.6.25.4/drivers/char/sysrq.c 2008-04-17 11:31:27.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/char/sysrq.c 2008-04-19 15:14:52.000000000 -0400
+@@ -37,6 +37,7 @@
#include <linux/irq.h>
#include <linux/hrtimer.h>
+ #include <linux/oom.h>
+#include <linux/vserver/debug.h>
#include <asm/ptrace.h>
#include <asm/irq_regs.h>
-@@ -309,6 +310,21 @@ static struct sysrq_key_op sysrq_unrt_op
+@@ -310,6 +311,21 @@ static struct sysrq_key_op sysrq_unrt_op
.enable_mask = SYSRQ_ENABLE_RTNICE,
};
@@ -2318,7 +1687,7 @@
/* Key Operations table and lock */
static DEFINE_SPINLOCK(sysrq_key_table_lock);
-@@ -357,7 +373,11 @@ static struct sysrq_key_op *sysrq_key_ta
+@@ -358,7 +374,11 @@ static struct sysrq_key_op *sysrq_key_ta
/* x: May be registered on ppc/powerpc for xmon */
NULL, /* x */
NULL, /* y */
@@ -2331,7 +1700,7 @@
};
/* key2index calculation, -1 on invalid index */
-@@ -369,6 +389,8 @@ static int sysrq_key_table_key2index(int
+@@ -370,6 +390,8 @@ static int sysrq_key_table_key2index(int
retval = key - '0';
else if ((key >= 'a') && (key <= 'z'))
retval = key + 10 - 'a';
@@ -2340,42 +1709,17 @@
else
retval = -1;
return retval;
---- linux-2.6.22/drivers/char/tty_io.c 2007-07-09 13:18:20 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/char/tty_io.c 2007-06-17 05:54:17 +0200
-@@ -103,6 +103,7 @@
- #include <linux/selection.h>
+--- linux-2.6.25.4/drivers/char/tty_io.c 2008-04-17 12:05:32.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/char/tty_io.c 2008-04-19 15:14:52.000000000 -0400
+@@ -105,6 +105,7 @@
#include <linux/kmod.h>
+ #include <linux/nsproxy.h>
+#include <linux/vs_pid.h>
#undef TTY_DEBUG_HANGUP
-@@ -191,6 +192,7 @@ static inline void free_tty_struct(struc
- {
- kfree(tty->write_buf);
- tty_buffer_free_all(tty);
-+ memset(tty, 0xDEADDEAD, sizeof(struct tty_struct));
- kfree(tty);
- }
-
-@@ -3049,13 +3051,15 @@ unlock:
-
- static int tiocgpgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t __user *p)
- {
-+ pid_t pgrp;
- /*
- * (tty == real_tty) is a cheap way of
- * testing if the tty is NOT a master pty.
- */
- if (tty == real_tty && current->signal->tty != real_tty)
- return -ENOTTY;
-- return put_user(pid_nr(real_tty->pgrp), p);
-+ pgrp = vx_map_pid(pid_nr(real_tty->pgrp));
-+ return put_user(pgrp, p);
- }
-
- /**
-@@ -3086,6 +3090,7 @@ static int tiocspgrp(struct tty_struct *
+@@ -3142,6 +3143,7 @@ static int tiocspgrp(struct tty_struct *
return -ENOTTY;
if (get_user(pgrp_nr, p))
return -EFAULT;
@@ -2383,18 +1727,8 @@
if (pgrp_nr < 0)
return -EINVAL;
rcu_read_lock();
-@@ -3910,6 +3915,9 @@ EXPORT_SYMBOL(tty_unregister_driver);
-
- dev_t tty_devnum(struct tty_struct *tty)
- {
-+ WARN_ON(!tty);
-+ WARN_ON(!tty->magic);
-+ WARN_ON(!tty->driver);
- return MKDEV(tty->driver->major, tty->driver->minor_start) + tty->index;
- }
- EXPORT_SYMBOL(tty_devnum);
---- linux-2.6.22/drivers/infiniband/hw/ipath/ipath_user_pages.c 2007-02-06 03:00:37 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/infiniband/hw/ipath/ipath_user_pages.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/drivers/infiniband/hw/ipath/ipath_user_pages.c 2008-04-17 10:37:17.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/infiniband/hw/ipath/ipath_user_pages.c 2008-04-19 15:14:52.000000000 -0400
@@ -33,6 +33,7 @@
#include <linux/mm.h>
@@ -2422,7 +1756,7 @@
ret = 0;
goto bail;
-@@ -203,7 +205,7 @@ void ipath_release_user_pages(struct pag
+@@ -177,7 +179,7 @@ void ipath_release_user_pages(struct pag
__ipath_release_user_pages(p, num_pages, 1);
@@ -2431,7 +1765,7 @@
up_write(¤t->mm->mmap_sem);
}
-@@ -220,7 +222,7 @@ static void user_pages_account(struct wo
+@@ -194,7 +196,7 @@ static void user_pages_account(struct wo
container_of(_work, struct ipath_user_pages_work, work);
down_write(&work->mm->mmap_sem);
@@ -2440,17 +1774,103 @@
up_write(&work->mm->mmap_sem);
mmput(work->mm);
kfree(work);
---- linux-2.6.22/drivers/md/dm-ioctl.c 2007-05-02 19:24:50 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/md/dm-ioctl.c 2007-06-15 02:37:03 +0200
-@@ -15,6 +15,7 @@
- #include <linux/slab.h>
- #include <linux/dm-ioctl.h>
+--- linux-2.6.25.4/drivers/md/dm.c 2008-04-17 12:05:33.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/md/dm.c 2008-04-19 15:14:52.000000000 -0400
+@@ -22,6 +22,7 @@
+ #include <linux/hdreg.h>
+ #include <linux/blktrace_api.h>
+ #include <linux/smp_lock.h>
++#include <linux/vs_base.h>
+
+ #define DM_MSG_PREFIX "core"
+
+@@ -91,6 +92,7 @@ struct mapped_device {
+ rwlock_t map_lock;
+ atomic_t holders;
+ atomic_t open_count;
++ xid_t xid;
+
+ unsigned long flags;
+
+@@ -250,6 +252,7 @@ static void __exit dm_exit(void)
+ static int dm_blk_open(struct inode *inode, struct file *file)
+ {
+ struct mapped_device *md;
++ int ret = -ENXIO;
+
+ spin_lock(&_minor_lock);
+
+@@ -258,18 +261,19 @@ static int dm_blk_open(struct inode *ino
+ goto out;
+
+ if (test_bit(DMF_FREEING, &md->flags) ||
+- test_bit(DMF_DELETING, &md->flags)) {
+- md = NULL;
++ test_bit(DMF_DELETING, &md->flags))
++ goto out;
++
++ ret = -EACCES;
++ if (!vx_check(md->xid, VS_IDENT|VS_HOSTID))
+ goto out;
+- }
+
+ dm_get(md);
+ atomic_inc(&md->open_count);
+-
++ ret = 0;
+ out:
+ spin_unlock(&_minor_lock);
+-
+- return md ? 0 : -ENXIO;
++ return ret;
+ }
+
+ static int dm_blk_close(struct inode *inode, struct file *file)
+@@ -465,6 +469,14 @@ int dm_set_geometry(struct mapped_device
+ return 0;
+ }
+
++/*
++ * Get the xid associated with a dm device
++ */
++xid_t dm_get_xid(struct mapped_device *md)
++{
++ return md->xid;
++}
++
+ /*-----------------------------------------------------------------
+ * CRUD START:
+ * A more elegant soln is in the works that uses the queue
+@@ -1022,6 +1034,7 @@ static struct mapped_device *alloc_dev(i
+ INIT_LIST_HEAD(&md->uevent_list);
+ spin_lock_init(&md->uevent_lock);
+
++ md->xid = vx_current_xid();
+ md->queue = blk_alloc_queue(GFP_KERNEL);
+ if (!md->queue)
+ goto bad_queue;
+--- linux-2.6.25.4/drivers/md/dm.h 2008-04-17 11:31:28.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/md/dm.h 2008-04-19 15:14:52.000000000 -0400
+@@ -127,6 +127,8 @@ void dm_put_target_type(struct target_ty
+ int dm_target_iterate(void (*iter_func)(struct target_type *tt,
+ void *param), void *param);
+
++xid_t dm_get_xid(struct mapped_device *md);
++
+ /*-----------------------------------------------------------------
+ * Useful inlines.
+ *---------------------------------------------------------------*/
+--- linux-2.6.25.4/drivers/md/dm-ioctl.c 2008-04-17 12:05:33.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/md/dm-ioctl.c 2008-04-21 10:45:53.000000000 -0400
+@@ -16,6 +16,7 @@
+ #include <linux/dm-ioctl.h>
#include <linux/hdreg.h>
+ #include <linux/compat.h>
+#include <linux/vs_context.h>
#include <asm/uaccess.h>
-@@ -100,7 +101,8 @@ static struct hash_cell *__get_name_cell
+@@ -101,7 +102,8 @@ static struct hash_cell *__get_name_cell
unsigned int h = hash_str(str);
list_for_each_entry (hc, _name_buckets + h, name_list)
@@ -2460,7 +1880,7 @@
dm_get(hc->md);
return hc;
}
-@@ -114,7 +116,8 @@ static struct hash_cell *__get_uuid_cell
+@@ -115,7 +117,8 @@ static struct hash_cell *__get_uuid_cell
unsigned int h = hash_str(str);
list_for_each_entry (hc, _uuid_buckets + h, uuid_list)
@@ -2470,7 +1890,7 @@
dm_get(hc->md);
return hc;
}
-@@ -349,6 +352,9 @@ typedef int (*ioctl_fn)(struct dm_ioctl
+@@ -352,6 +355,9 @@ typedef int (*ioctl_fn)(struct dm_ioctl
static int remove_all(struct dm_ioctl *param, size_t param_size)
{
@@ -2480,7 +1900,7 @@
dm_hash_remove_all(1);
param->data_size = 0;
return 0;
-@@ -396,6 +402,8 @@ static int list_devices(struct dm_ioctl
+@@ -399,6 +405,8 @@ static int list_devices(struct dm_ioctl
*/
for (i = 0; i < NUM_BUCKETS; i++) {
list_for_each_entry (hc, _name_buckets + i, name_list) {
@@ -2489,7 +1909,7 @@
needed += sizeof(struct dm_name_list);
needed += strlen(hc->name) + 1;
needed += ALIGN_MASK;
-@@ -419,6 +427,8 @@ static int list_devices(struct dm_ioctl
+@@ -422,6 +430,8 @@ static int list_devices(struct dm_ioctl
*/
for (i = 0; i < NUM_BUCKETS; i++) {
list_for_each_entry (hc, _name_buckets + i, name_list) {
@@ -2498,7 +1918,7 @@
if (old_nl)
old_nl->next = (uint32_t) ((void *) nl -
(void *) old_nl);
-@@ -609,10 +619,11 @@ static struct hash_cell *__find_device_h
+@@ -612,10 +622,11 @@ static struct hash_cell *__find_device_h
if (!md)
goto out;
@@ -2512,7 +1932,7 @@
out:
return mdptr;
}
-@@ -1409,8 +1420,8 @@ static int ctl_ioctl(struct inode *inode
+@@ -1406,8 +1417,8 @@ static int ctl_ioctl(uint command, struc
ioctl_fn fn = NULL;
size_t param_size;
@@ -2523,94 +1943,79 @@
return -EACCES;
if (_IOC_TYPE(command) != DM_IOCTL)
---- linux-2.6.22/drivers/md/dm.c 2007-07-09 13:18:44 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/md/dm.c 2007-06-15 02:37:03 +0200
-@@ -21,6 +21,7 @@
- #include <linux/hdreg.h>
- #include <linux/blktrace_api.h>
- #include <linux/smp_lock.h>
-+#include <linux/vs_base.h>
-
- #define DM_MSG_PREFIX "core"
+--- linux-2.6.25.4/drivers/net/tun.c 2008-04-17 12:05:36.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/drivers/net/tun.c 2008-04-19 15:39:04.000000000 -0400
+@@ -62,6 +62,7 @@
+ #include <linux/if_ether.h>
+ #include <linux/if_tun.h>
+ #include <linux/crc32.h>
++#include <linux/vs_network.h>
+ #include <net/net_namespace.h>
-@@ -77,6 +78,7 @@ struct mapped_device {
- rwlock_t map_lock;
- atomic_t holders;
- atomic_t open_count;
-+ xid_t xid;
+ #include <asm/system.h>
+@@ -86,6 +87,7 @@ struct tun_struct {
+ int attached;
+ uid_t owner;
+ gid_t group;
++ nid_t nid;
+
+ wait_queue_head_t read_wait;
+ struct sk_buff_head readq;
+@@ -465,6 +467,7 @@ static void tun_setup(struct net_device
+
+ tun->owner = -1;
+ tun->group = -1;
++ tun->nid = current->nid;
+
+ dev->open = tun_net_open;
+ dev->hard_start_xmit = tun_net_xmit;
+@@ -494,6 +497,9 @@ static int tun_set_iff(struct file *file
+
+ tun = tun_get_by_name(ifr->ifr_name);
+ if (tun) {
++ if (!nx_check(tun->nid, VS_IDENT | VS_HOSTID | VS_ADMIN_P))
++ return -EPERM;
++
+ if (tun->attached)
+ return -EBUSY;
- unsigned long flags;
+@@ -502,7 +508,7 @@ static int tun_set_iff(struct file *file
+ current->euid != tun->owner) ||
+ (tun->group != -1 &&
+ current->egid != tun->group)) &&
+- !capable(CAP_NET_ADMIN))
++ !cap_raised(current->cap_effective, CAP_NET_ADMIN))
+ return -EPERM;
+ }
+ else if (__dev_get_by_name(&init_net, ifr->ifr_name))
+@@ -513,7 +519,7 @@ static int tun_set_iff(struct file *file
-@@ -223,6 +225,7 @@ static void __exit dm_exit(void)
- static int dm_blk_open(struct inode *inode, struct file *file)
- {
- struct mapped_device *md;
-+ int ret = -ENXIO;
+ err = -EINVAL;
- spin_lock(&_minor_lock);
+- if (!capable(CAP_NET_ADMIN))
++ if (!nx_capable(CAP_NET_ADMIN, NXC_TUN_CREATE))
+ return -EPERM;
-@@ -231,18 +234,19 @@ static int dm_blk_open(struct inode *ino
- goto out;
+ /* Set dev type */
+@@ -656,6 +662,16 @@ static int tun_chr_ioctl(struct inode *i
+ DBG(KERN_INFO "%s: group set to %d\n", tun->dev->name, tun->group);
+ break;
- if (test_bit(DMF_FREEING, &md->flags) ||
-- test_bit(DMF_DELETING, &md->flags)) {
-- md = NULL;
-+ test_bit(DMF_DELETING, &md->flags))
-+ goto out;
++ case TUNSETNID:
++ if (!capable(CAP_CONTEXT))
++ return -EPERM;
+
-+ ret = -EACCES;
-+ if (!vx_check(md->xid, VS_IDENT|VS_HOSTID))
- goto out;
-- }
-
- dm_get(md);
- atomic_inc(&md->open_count);
--
-+ ret = 0;
- out:
- spin_unlock(&_minor_lock);
--
-- return md ? 0 : -ENXIO;
-+ return ret;
- }
-
- static int dm_blk_close(struct inode *inode, struct file *file)
-@@ -438,6 +442,14 @@ int dm_set_geometry(struct mapped_device
- return 0;
- }
-
-+/*
-+ * Get the xid associated with a dm device
-+ */
-+xid_t dm_get_xid(struct mapped_device *md)
-+{
-+ return md->xid;
-+}
++ /* Set nid owner of the device */
++ tun->nid = (nid_t) arg;
+
- /*-----------------------------------------------------------------
- * CRUD START:
- * A more elegant soln is in the works that uses the queue
-@@ -991,6 +1003,7 @@ static struct mapped_device *alloc_dev(i
- atomic_set(&md->holders, 1);
- atomic_set(&md->open_count, 0);
- atomic_set(&md->event_nr, 0);
-+ md->xid = vx_current_xid();
-
- md->queue = blk_alloc_queue(GFP_KERNEL);
- if (!md->queue)
---- linux-2.6.22/drivers/md/dm.h 2007-02-06 03:00:41 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/drivers/md/dm.h 2007-06-15 02:37:03 +0200
-@@ -91,6 +91,8 @@ void dm_put_target_type(struct target_ty
- int dm_target_iterate(void (*iter_func)(struct target_type *tt,
- void *param), void *param);
-
-+xid_t dm_get_xid(struct mapped_device *md);
++ DBG(KERN_INFO "%s: nid owner set to %u\n", tun->dev->name, tun->nid);
++ break;
+
- /*-----------------------------------------------------------------
- * Useful inlines.
- *---------------------------------------------------------------*/
---- linux-2.6.22/fs/attr.c 2007-07-09 13:19:22 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/attr.c 2007-06-15 02:37:03 +0200
+ case TUNSETLINK:
+ /* Only allow setting the type when the interface is down */
+ if (tun->dev->flags & IFF_UP) {
+--- linux-2.6.25.4/fs/attr.c 2008-04-17 11:31:35.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/attr.c 2008-04-19 15:14:52.000000000 -0400
@@ -14,6 +14,9 @@
#include <linux/fcntl.h>
#include <linux/quotaops.h>
@@ -2621,8 +2026,8 @@
/* Taken over from the old code... */
-@@ -55,6 +58,30 @@ int inode_change_ok(struct inode *inode,
- if (current->fsuid != inode->i_uid && !capable(CAP_FOWNER))
+@@ -55,6 +58,27 @@ int inode_change_ok(struct inode *inode,
+ if (!is_owner_or_cap(inode))
goto error;
}
+
@@ -2631,28 +2036,25 @@
+ goto fine;
+
+ if (IS_BARRIER(inode)) {
-+ vxwprintk(1, "xid=%d messing with the barrier.",
-+ vx_current_xid());
++ vxwprintk_task(1, "messing with the barrier.");
+ goto error;
+ }
+ switch (inode->i_sb->s_magic) {
+ case PROC_SUPER_MAGIC:
+ /* maybe allow that in the future? */
-+ vxwprintk(1, "xid=%d messing with the procfs.",
-+ vx_current_xid());
++ vxwprintk_task(1, "messing with the procfs.");
+ goto error;
+ case DEVPTS_SUPER_MAGIC:
+ /* devpts is xid tagged */
+ if (vx_check((xid_t)inode->i_tag, VS_IDENT))
+ goto fine;
-+ vxwprintk(1, "xid=%d messing with the devpts.",
-+ vx_current_xid());
++ vxwprintk_task(1, "messing with the devpts.");
+ goto error;
+ }
fine:
retval = 0;
error:
-@@ -78,6 +105,8 @@ int inode_setattr(struct inode * inode,
+@@ -78,6 +102,8 @@ int inode_setattr(struct inode * inode,
inode->i_uid = attr->ia_uid;
if (ia_valid & ATTR_GID)
inode->i_gid = attr->ia_gid;
@@ -2661,7 +2063,7 @@
if (ia_valid & ATTR_ATIME)
inode->i_atime = timespec_trunc(attr->ia_atime,
inode->i_sb->s_time_gran);
-@@ -152,7 +181,8 @@ int notify_change(struct dentry * dentry
+@@ -167,7 +193,8 @@ int notify_change(struct dentry * dentry
error = security_inode_setattr(dentry, attr);
if (!error) {
if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
@@ -2671,8 +2073,8 @@
error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0;
if (!error)
error = inode_setattr(inode, attr);
---- linux-2.6.22/fs/binfmt_aout.c 2007-02-06 03:01:16 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/binfmt_aout.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/binfmt_aout.c 2008-04-17 12:05:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/binfmt_aout.c 2008-04-19 15:14:52.000000000 -0400
@@ -24,6 +24,7 @@
#include <linux/binfmts.h>
#include <linux/personality.h>
@@ -2681,8 +2083,8 @@
#include <asm/system.h>
#include <asm/uaccess.h>
---- linux-2.6.22/fs/binfmt_elf.c 2007-07-09 13:19:22 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/binfmt_elf.c 2007-07-09 13:11:54 +0200
+--- linux-2.6.25.4/fs/binfmt_elf.c 2008-04-17 12:05:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/binfmt_elf.c 2008-04-19 15:14:52.000000000 -0400
@@ -39,6 +39,7 @@
#include <linux/random.h>
#include <linux/elf.h>
@@ -2691,9 +2093,9 @@
#include <asm/uaccess.h>
#include <asm/param.h>
#include <asm/page.h>
---- linux-2.6.22/fs/binfmt_flat.c 2007-07-09 13:19:22 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/binfmt_flat.c 2007-06-17 05:54:17 +0200
-@@ -36,6 +36,7 @@
+--- linux-2.6.25.4/fs/binfmt_flat.c 2008-04-17 12:05:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/binfmt_flat.c 2008-04-19 15:14:52.000000000 -0400
+@@ -35,6 +35,7 @@
#include <linux/init.h>
#include <linux/flat.h>
#include <linux/syscalls.h>
@@ -2701,18 +2103,91 @@
#include <asm/byteorder.h>
#include <asm/system.h>
---- linux-2.6.22/fs/binfmt_som.c 2006-11-30 21:19:19 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/binfmt_som.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/binfmt_som.c 2008-04-17 12:05:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/binfmt_som.c 2008-04-19 15:14:52.000000000 -0400
@@ -28,6 +28,7 @@
#include <linux/shm.h>
#include <linux/personality.h>
#include <linux/init.h>
+#include <linux/vs_memory.h>
- #include <asm/a.out.h>
#include <asm/uaccess.h>
---- linux-2.6.22/fs/dcache.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/dcache.c 2007-06-15 02:37:03 +0200
+ #include <asm/pgtable.h>
+--- linux-2.6.25.4/fs/block_dev.c 2008-04-17 12:05:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/block_dev.c 2008-04-19 15:14:52.000000000 -0400
+@@ -23,6 +23,7 @@
+ #include <linux/uio.h>
+ #include <linux/namei.h>
+ #include <linux/log2.h>
++#include <linux/vs_device.h>
+ #include <asm/uaccess.h>
+ #include "internal.h"
+
+@@ -388,6 +389,7 @@ struct block_device *bdget(dev_t dev)
+ bdev->bd_invalidated = 0;
+ inode->i_mode = S_IFBLK;
+ inode->i_rdev = dev;
++ inode->i_mdev = dev;
+ inode->i_bdev = bdev;
+ inode->i_data.a_ops = &def_blk_aops;
+ mapping_set_gfp_mask(&inode->i_data, GFP_USER);
+@@ -424,6 +426,11 @@ EXPORT_SYMBOL(bdput);
+ static struct block_device *bd_acquire(struct inode *inode)
+ {
+ struct block_device *bdev;
++ dev_t mdev;
++
++ if (!vs_map_blkdev(inode->i_rdev, &mdev, DATTR_OPEN))
++ return NULL;
++ inode->i_mdev = mdev;
+
+ spin_lock(&bdev_lock);
+ bdev = inode->i_bdev;
+@@ -434,7 +441,7 @@ static struct block_device *bd_acquire(s
+ }
+ spin_unlock(&bdev_lock);
+
+- bdev = bdget(inode->i_rdev);
++ bdev = bdget(mdev);
+ if (bdev) {
+ spin_lock(&bdev_lock);
+ if (!inode->i_bdev) {
+--- linux-2.6.25.4/fs/char_dev.c 2008-04-17 12:05:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/char_dev.c 2008-04-29 18:44:50.000000000 -0400
+@@ -21,6 +21,8 @@
+ #include <linux/cdev.h>
+ #include <linux/mutex.h>
+ #include <linux/backing-dev.h>
++#include <linux/vs_context.h>
++#include <linux/vs_device.h>
+
+ #ifdef CONFIG_KMOD
+ #include <linux/kmod.h>
+@@ -362,14 +364,21 @@ static int chrdev_open(struct inode *ino
+ struct cdev *p;
+ struct cdev *new = NULL;
+ int ret = 0;
++ dev_t mdev;
++
++ if (!vs_map_chrdev(inode->i_rdev, &mdev, DATTR_OPEN))
++ return -EPERM;
++ inode->i_mdev = mdev;
+
+ spin_lock(&cdev_lock);
+ p = inode->i_cdev;
+ if (!p) {
+ struct kobject *kobj;
+ int idx;
++
+ spin_unlock(&cdev_lock);
+- kobj = kobj_lookup(cdev_map, inode->i_rdev, &idx);
++
++ kobj = kobj_lookup(cdev_map, mdev, &idx);
+ if (!kobj)
+ return -ENXIO;
+ new = container_of(kobj, struct cdev, kobj);
+--- linux-2.6.25.4/fs/dcache.c 2008-04-17 12:05:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/dcache.c 2008-04-19 17:06:15.000000000 -0400
@@ -31,6 +31,7 @@
#include <linux/seqlock.h>
#include <linux/swap.h>
@@ -2721,7 +2196,7 @@
#include "internal.h"
-@@ -176,6 +177,7 @@ void dput(struct dentry *dentry)
+@@ -184,6 +185,7 @@ void dput(struct dentry *dentry)
if (!dentry)
return;
@@ -2729,7 +2204,7 @@
repeat:
if (atomic_read(&dentry->d_count) == 1)
might_sleep();
-@@ -189,6 +191,8 @@ repeat:
+@@ -197,6 +199,8 @@ repeat:
return;
}
@@ -2739,24 +2214,24 @@
* AV: ->d_delete() is _NOT_ allowed to block now.
*/
@@ -288,6 +292,7 @@ static inline struct dentry * __dget_loc
- if (!list_empty(&dentry->d_lru)) {
- dentry_stat.nr_unused--;
- list_del_init(&dentry->d_lru);
-+ vx_dentry_inc(dentry);
- }
+ {
+ atomic_inc(&dentry->d_count);
+ dentry_lru_remove(dentry);
++ vx_dentry_inc(dentry);
return dentry;
}
-@@ -898,6 +903,9 @@ struct dentry *d_alloc(struct dentry * p
+
+@@ -885,6 +890,9 @@ struct dentry *d_alloc(struct dentry * p
struct dentry *dentry;
char *dname;
+ if (!vx_dentry_avail(1))
+ return NULL;
+
- dentry = kmem_cache_alloc(dentry_cache, GFP_KERNEL);
+ dentry = kmem_cache_alloc(dentry_cache, GFP_KERNEL);
if (!dentry)
return NULL;
-@@ -946,6 +954,7 @@ struct dentry *d_alloc(struct dentry * p
+@@ -933,6 +941,7 @@ struct dentry *d_alloc(struct dentry * p
if (parent)
list_add(&dentry->d_u.d_child, &parent->d_subdirs);
dentry_stat.nr_dentry++;
@@ -2764,7 +2239,7 @@
spin_unlock(&dcache_lock);
return dentry;
-@@ -1295,6 +1304,7 @@ struct dentry * __d_lookup(struct dentry
+@@ -1282,6 +1291,7 @@ struct dentry * __d_lookup(struct dentry
if (!d_unhashed(dentry)) {
atomic_inc(&dentry->d_count);
@@ -2772,16 +2247,23 @@
found = dentry;
}
spin_unlock(&dentry->d_lock);
---- linux-2.6.22/fs/devpts/inode.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/devpts/inode.c 2007-06-15 02:41:30 +0200
-@@ -20,8 +20,22 @@
+--- linux-2.6.25.4/fs/devpts/inode.c 2008-04-17 12:05:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/devpts/inode.c 2008-04-21 09:23:34.000000000 -0400
+@@ -17,15 +17,30 @@
+ #include <linux/namei.h>
+ #include <linux/mount.h>
+ #include <linux/tty.h>
++#include <linux/magic.h>
#include <linux/devpts_fs.h>
#include <linux/parser.h>
#include <linux/fsnotify.h>
+ #include <linux/seq_file.h>
+#include <linux/vs_base.h>
-+
-#define DEVPTS_SUPER_MAGIC 0x1cd1
+
+ #define DEVPTS_DEFAULT_MODE 0600
+
+static int devpts_permission(struct inode *inode, int mask, struct nameidata *nd)
+{
+ int ret = -EACCES;
@@ -2795,10 +2277,11 @@
+static struct inode_operations devpts_file_inode_operations = {
+ .permission = devpts_permission,
+};
-
++
static struct vfsmount *devpts_mnt;
static struct dentry *devpts_root;
-@@ -92,6 +106,25 @@ static int devpts_remount(struct super_b
+
+@@ -106,6 +121,25 @@ static int devpts_show_options(struct se
return 0;
}
@@ -2824,7 +2307,7 @@
static const struct super_operations devpts_sops = {
.statfs = simple_statfs,
.remount_fs = devpts_remount,
-@@ -117,8 +150,10 @@ devpts_fill_super(struct super_block *s,
+@@ -132,8 +166,10 @@ devpts_fill_super(struct super_block *s,
inode->i_uid = inode->i_gid = 0;
inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR;
inode->i_op = &simple_dir_inode_operations;
@@ -2836,7 +2319,7 @@
devpts_root = s->s_root = d_alloc_root(inode);
if (s->s_root)
-@@ -176,6 +211,9 @@ int devpts_pty_new(struct tty_struct *tt
+@@ -191,6 +227,9 @@ int devpts_pty_new(struct tty_struct *tt
inode->i_gid = config.setgid ? config.gid : current->fsgid;
inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
init_special_inode(inode, S_IFCHR|config.mode, device);
@@ -2846,9 +2329,9 @@
inode->i_private = tty;
dentry = get_node(number);
---- linux-2.6.22/fs/ecryptfs/inode.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ecryptfs/inode.c 2007-07-07 03:52:53 +0200
-@@ -423,7 +423,7 @@ static int ecryptfs_link(struct dentry *
+--- linux-2.6.25.4/fs/ecryptfs/inode.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ecryptfs/inode.c 2008-04-19 15:14:52.000000000 -0400
+@@ -400,7 +400,7 @@ static int ecryptfs_link(struct dentry *
dget(lower_new_dentry);
lower_dir_dentry = lock_parent(lower_new_dentry);
rc = vfs_link(lower_old_dentry, lower_dir_dentry->d_inode,
@@ -2857,7 +2340,7 @@
if (rc || !lower_new_dentry->d_inode)
goto out_lock;
rc = ecryptfs_interpose(lower_new_dentry, new_dentry, dir->i_sb, 0);
-@@ -451,7 +451,7 @@ static int ecryptfs_unlink(struct inode
+@@ -428,7 +428,7 @@ static int ecryptfs_unlink(struct inode
struct inode *lower_dir_inode = ecryptfs_inode_to_lower(dir);
lock_parent(lower_dentry);
@@ -2866,7 +2349,7 @@
if (rc) {
printk(KERN_ERR "Error in vfs_unlink; rc = [%d]\n", rc);
goto out_unlock;
-@@ -488,7 +488,7 @@ static int ecryptfs_symlink(struct inode
+@@ -466,7 +466,7 @@ static int ecryptfs_symlink(struct inode
goto out_lock;
}
rc = vfs_symlink(lower_dir_dentry->d_inode, lower_dentry,
@@ -2875,7 +2358,7 @@
kfree(encoded_symname);
if (rc || !lower_dentry->d_inode)
goto out_lock;
-@@ -513,7 +513,7 @@ static int ecryptfs_mkdir(struct inode *
+@@ -491,7 +491,7 @@ static int ecryptfs_mkdir(struct inode *
lower_dentry = ecryptfs_dentry_to_lower(dentry);
lower_dir_dentry = lock_parent(lower_dentry);
@@ -2884,7 +2367,7 @@
if (rc || !lower_dentry->d_inode)
goto out;
rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0);
-@@ -539,7 +539,7 @@ static int ecryptfs_rmdir(struct inode *
+@@ -517,7 +517,7 @@ static int ecryptfs_rmdir(struct inode *
dget(dentry);
lower_dir_dentry = lock_parent(lower_dentry);
dget(lower_dentry);
@@ -2893,7 +2376,7 @@
dput(lower_dentry);
if (!rc)
d_delete(lower_dentry);
-@@ -561,7 +561,7 @@ ecryptfs_mknod(struct inode *dir, struct
+@@ -539,7 +539,7 @@ ecryptfs_mknod(struct inode *dir, struct
lower_dentry = ecryptfs_dentry_to_lower(dentry);
lower_dir_dentry = lock_parent(lower_dentry);
@@ -2902,27 +2385,20 @@
if (rc || !lower_dentry->d_inode)
goto out;
rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0);
---- linux-2.6.22/fs/exec.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/exec.c 2007-06-15 03:12:04 +0200
-@@ -51,6 +51,7 @@
- #include <linux/cn_proc.h>
- #include <linux/audit.h>
- #include <linux/signalfd.h>
-+#include <linux/vs_memory.h>
-
- #include <asm/uaccess.h>
- #include <asm/mmu_context.h>
-@@ -440,7 +441,8 @@ int setup_arg_pages(struct linux_binprm
- kmem_cache_free(vm_area_cachep, mpnt);
- return ret;
- }
-- mm->stack_vm = mm->total_vm = vma_pages(mpnt);
-+ vx_vmpages_sub(mm, mm->total_vm - vma_pages(mpnt));
-+ mm->stack_vm = mm->total_vm;
+--- linux-2.6.25.4/fs/exec.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/exec.c 2008-04-19 15:14:52.000000000 -0400
+@@ -249,7 +249,9 @@ static int __bprm_mm_init(struct linux_b
+ goto err;
}
- for (i = 0 ; i < MAX_ARG_PAGES ; i++) {
-@@ -1336,7 +1338,7 @@ static int format_corename(char *corenam
+- mm->stack_vm = mm->total_vm = 1;
++ mm->total_vm = 0;
++ vx_vmpages_inc(mm);
++ mm->stack_vm = 1;
+ up_write(&mm->mmap_sem);
+
+ bprm->p = vma->vm_end - sizeof(void *);
+@@ -1452,7 +1454,7 @@ static int format_corename(char *corenam
/* UNIX time of coredump */
case 't': {
struct timeval tv;
@@ -2931,8 +2407,8 @@
rc = snprintf(out_ptr, out_end - out_ptr,
"%lu", tv.tv_sec);
if (rc > out_end - out_ptr)
---- linux-2.6.22/fs/ext2/balloc.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/balloc.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext2/balloc.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/balloc.c 2008-04-19 15:14:52.000000000 -0400
@@ -16,6 +16,8 @@
#include <linux/sched.h>
#include <linux/buffer_head.h>
@@ -2942,23 +2418,7 @@
/*
* balloc.c contains the blocks allocation and deallocation routines
-@@ -102,12 +104,13 @@ static int reserve_blocks(struct super_b
- {
- struct ext2_sb_info *sbi = EXT2_SB(sb);
- struct ext2_super_block *es = sbi->s_es;
-- unsigned free_blocks;
-- unsigned root_blocks;
-+ unsigned long long free_blocks, root_blocks;
-
- free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter);
- root_blocks = le32_to_cpu(es->s_r_blocks_count);
-
-+ DLIMIT_ADJUST_BLOCK(sb, dx_current_tag(), &free_blocks, &root_blocks);
-+
- if (free_blocks < count)
- count = free_blocks;
-
-@@ -258,6 +261,7 @@ do_more:
+@@ -569,6 +571,7 @@ do_more:
}
error_return:
brelse(bitmap_bh);
@@ -2966,29 +2426,49 @@
release_blocks(sb, freed);
DQUOT_FREE_BLOCK(inode, freed);
}
-@@ -361,6 +365,10 @@ int ext2_new_block(struct inode *inode,
- *err = -ENOSPC;
- goto out_dquot;
+@@ -701,7 +704,6 @@ ext2_try_to_allocate(struct super_block
+ start = 0;
+ end = EXT2_BLOCKS_PER_GROUP(sb);
}
-+ if (DLIMIT_ALLOC_BLOCK(inode, es_alloc)) {
-+ *err = -ENOSPC;
-+ goto out_dlimit;
+-
+ BUG_ON(start > EXT2_BLOCKS_PER_GROUP(sb));
+
+ repeat:
+@@ -1251,6 +1253,11 @@ ext2_fsblk_t ext2_new_blocks(struct inod
+ *errp = -EDQUOT;
+ return 0;
+ }
++ if (DLIMIT_ALLOC_BLOCK(inode, num)) {
++ *errp = -ENOSPC;
++ DQUOT_FREE_BLOCK(inode, num);
++ return 0;
+ }
- ext2_debug ("goal=%lu.\n", goal);
+ sbi = EXT2_SB(sb);
+ es = EXT2_SB(sb)->s_es;
+@@ -1403,6 +1410,7 @@ allocated:
-@@ -508,6 +516,8 @@ got_block:
- *err = 0;
- out_release:
- group_release_blocks(sb, group_no, desc, gdp_bh, group_alloc);
-+ DLIMIT_FREE_BLOCK(inode, es_alloc);
-+out_dlimit:
- release_blocks(sb, es_alloc);
- out_dquot:
- DQUOT_FREE_BLOCK(inode, dq_alloc);
---- linux-2.6.22/fs/ext2/ext2.h 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/ext2.h 2007-06-15 02:37:03 +0200
-@@ -167,6 +167,7 @@ extern const struct file_operations ext2
+ *errp = 0;
+ brelse(bitmap_bh);
++ DLIMIT_FREE_BLOCK(inode, *count-num);
+ DQUOT_FREE_BLOCK(inode, *count-num);
+ *count = num;
+ return ret_block;
+@@ -1413,8 +1421,10 @@ out:
+ /*
+ * Undo the block allocation
+ */
+- if (!performed_allocation)
++ if (!performed_allocation) {
++ DLIMIT_FREE_BLOCK(inode, *count);
+ DQUOT_FREE_BLOCK(inode, *count);
++ }
+ brelse(bitmap_bh);
+ return 0;
+ }
+--- linux-2.6.25.4/fs/ext2/ext2.h 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/ext2.h 2008-04-19 15:14:52.000000000 -0400
+@@ -168,6 +168,7 @@ extern const struct file_operations ext2
extern const struct address_space_operations ext2_aops;
extern const struct address_space_operations ext2_aops_xip;
extern const struct address_space_operations ext2_nobh_aops;
@@ -2996,24 +2476,16 @@
/* namei.c */
extern const struct inode_operations ext2_dir_inode_operations;
---- linux-2.6.22/fs/ext2/file.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/file.c 2007-06-15 02:37:03 +0200
-@@ -54,6 +54,7 @@ const struct file_operations ext2_file_o
- .release = ext2_release_file,
- .fsync = ext2_sync_file,
- .sendfile = generic_file_sendfile,
-+ .sendpage = generic_file_sendpage,
- .splice_read = generic_file_splice_read,
- .splice_write = generic_file_splice_write,
- };
-@@ -85,4 +86,5 @@ const struct inode_operations ext2_file_
+--- linux-2.6.25.4/fs/ext2/file.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/file.c 2008-04-19 15:14:52.000000000 -0400
+@@ -86,4 +86,5 @@ const struct inode_operations ext2_file_
#endif
.setattr = ext2_setattr,
.permission = ext2_permission,
+ .sync_flags = ext2_sync_flags,
};
---- linux-2.6.22/fs/ext2/ialloc.c 2006-11-30 21:19:19 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/ialloc.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext2/ialloc.c 2008-04-17 11:31:35.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/ialloc.c 2008-04-19 15:14:52.000000000 -0400
@@ -17,6 +17,8 @@
#include <linux/backing-dev.h>
#include <linux/buffer_head.h>
@@ -3031,7 +2503,7 @@
}
es = EXT2_SB(sb)->s_es;
-@@ -464,6 +467,11 @@ struct inode *ext2_new_inode(struct inod
+@@ -456,6 +459,11 @@ struct inode *ext2_new_inode(struct inod
if (!inode)
return ERR_PTR(-ENOMEM);
@@ -3043,7 +2515,7 @@
ei = EXT2_I(inode);
sbi = EXT2_SB(sb);
es = sbi->s_es;
-@@ -577,7 +585,8 @@ got:
+@@ -569,7 +577,8 @@ got:
inode->i_blocks = 0;
inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
memset(ei->i_data, 0, sizeof(ei->i_data));
@@ -3053,7 +2525,7 @@
if (S_ISLNK(mode))
ei->i_flags &= ~(EXT2_IMMUTABLE_FL|EXT2_APPEND_FL);
/* dirsync is only applied to directories */
-@@ -625,12 +634,15 @@ fail_free_drop:
+@@ -614,12 +623,15 @@ fail_free_drop:
fail_drop:
DQUOT_DROP(inode);
@@ -3069,8 +2541,8 @@
make_bad_inode(inode);
iput(inode);
return ERR_PTR(err);
---- linux-2.6.22/fs/ext2/inode.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/inode.c 2007-06-15 02:57:56 +0200
+--- linux-2.6.25.4/fs/ext2/inode.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/inode.c 2008-04-21 10:14:57.000000000 -0400
@@ -31,6 +31,7 @@
#include <linux/writeback.h>
#include <linux/buffer_head.h>
@@ -3079,7 +2551,7 @@
#include "ext2.h"
#include "acl.h"
#include "xip.h"
-@@ -913,7 +914,7 @@ void ext2_truncate (struct inode * inode
+@@ -1011,7 +1012,7 @@ void ext2_truncate(struct inode *inode)
return;
if (ext2_inode_is_fast_symlink(inode))
return;
@@ -3087,8 +2559,8 @@
+ if (IS_APPEND(inode) || IS_IXORUNLINK(inode))
return;
- ext2_discard_prealloc(inode);
-@@ -1042,13 +1043,20 @@ void ext2_set_inode_flags(struct inode *
+ blocksize = inode->i_sb->s_blocksize;
+@@ -1149,13 +1150,20 @@ void ext2_set_inode_flags(struct inode *
{
unsigned int flags = EXT2_I(inode)->i_flags;
@@ -3112,7 +2584,7 @@
if (flags & EXT2_NOATIME_FL)
inode->i_flags |= S_NOATIME;
if (flags & EXT2_DIRSYNC_FL)
-@@ -1074,12 +1082,37 @@ void ext2_get_inode_flags(struct ext2_in
+@@ -1181,6 +1189,30 @@ void ext2_get_inode_flags(struct ext2_in
ei->i_flags |= EXT2_DIRSYNC_FL;
}
@@ -3139,19 +2611,21 @@
+ return 0;
+}
+
- void ext2_read_inode (struct inode * inode)
++
+ struct inode *ext2_iget (struct super_block *sb, unsigned long ino)
{
- struct ext2_inode_info *ei = EXT2_I(inode);
- ino_t ino = inode->i_ino;
- struct buffer_head * bh;
- struct ext2_inode * raw_inode = ext2_get_inode(inode->i_sb, ino, &bh);
+ struct ext2_inode_info *ei;
+@@ -1188,6 +1220,8 @@ struct inode *ext2_iget (struct super_bl
+ struct ext2_inode *raw_inode;
+ struct inode *inode;
+ long ret = -EIO;
+ uid_t uid;
+ gid_t gid;
int n;
- #ifdef CONFIG_EXT2_FS_POSIX_ACL
-@@ -1090,12 +1123,17 @@ void ext2_read_inode (struct inode * ino
- goto bad_inode;
+ inode = iget_locked(sb, ino);
+@@ -1210,12 +1244,17 @@ struct inode *ext2_iget (struct super_bl
+ }
inode->i_mode = le16_to_cpu(raw_inode->i_mode);
- inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low);
@@ -3172,7 +2646,7 @@
inode->i_nlink = le16_to_cpu(raw_inode->i_links_count);
inode->i_size = le32_to_cpu(raw_inode->i_size);
inode->i_atime.tv_sec = (signed)le32_to_cpu(raw_inode->i_atime);
-@@ -1192,8 +1230,8 @@ static int ext2_update_inode(struct inod
+@@ -1311,8 +1350,8 @@ static int ext2_update_inode(struct inod
struct ext2_inode_info *ei = EXT2_I(inode);
struct super_block *sb = inode->i_sb;
ino_t ino = inode->i_ino;
@@ -3183,7 +2657,7 @@
struct buffer_head * bh;
struct ext2_inode * raw_inode = ext2_get_inode(sb, ino, &bh);
int n;
-@@ -1229,6 +1267,9 @@ static int ext2_update_inode(struct inod
+@@ -1348,6 +1387,9 @@ static int ext2_update_inode(struct inod
raw_inode->i_uid_high = 0;
raw_inode->i_gid_high = 0;
}
@@ -3193,7 +2667,7 @@
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
raw_inode->i_size = cpu_to_le32(inode->i_size);
raw_inode->i_atime = cpu_to_le32(inode->i_atime.tv_sec);
-@@ -1315,7 +1356,8 @@ int ext2_setattr(struct dentry *dentry,
+@@ -1434,7 +1476,8 @@ int ext2_setattr(struct dentry *dentry,
if (error)
return error;
if ((iattr->ia_valid & ATTR_UID && iattr->ia_uid != inode->i_uid) ||
@@ -3203,8 +2677,8 @@
error = DQUOT_TRANSFER(inode, iattr) ? -EDQUOT : 0;
if (error)
return error;
---- linux-2.6.22/fs/ext2/ioctl.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/ioctl.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext2/ioctl.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/ioctl.c 2008-04-19 15:14:52.000000000 -0400
@@ -13,6 +13,7 @@
#include <linux/sched.h>
#include <linux/compat.h>
@@ -3213,7 +2687,7 @@
#include <asm/current.h>
#include <asm/uaccess.h>
-@@ -33,7 +34,8 @@ int ext2_ioctl (struct inode * inode, st
+@@ -34,7 +35,8 @@ long ext2_ioctl(struct file *filp, unsig
case EXT2_IOC_SETFLAGS: {
unsigned int oldflags;
@@ -3222,8 +2696,8 @@
+ (filp && MNT_IS_RDONLY(filp->f_vfsmnt)))
return -EROFS;
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
-@@ -54,7 +56,9 @@ int ext2_ioctl (struct inode * inode, st
+ if (!is_owner_or_cap(inode))
+@@ -60,7 +62,9 @@ long ext2_ioctl(struct file *filp, unsig
*
* This test looks nicer. Thanks to Pauline Middelink
*/
@@ -3234,9 +2708,9 @@
if (!capable(CAP_LINUX_IMMUTABLE)) {
mutex_unlock(&inode->i_mutex);
return -EPERM;
-@@ -76,7 +80,8 @@ int ext2_ioctl (struct inode * inode, st
+@@ -82,7 +86,8 @@ long ext2_ioctl(struct file *filp, unsig
case EXT2_IOC_SETVERSION:
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
+ if (!is_owner_or_cap(inode))
return -EPERM;
- if (IS_RDONLY(inode))
+ if (IS_RDONLY(inode) ||
@@ -3244,8 +2718,8 @@
return -EROFS;
if (get_user(inode->i_generation, (int __user *) arg))
return -EFAULT;
---- linux-2.6.22/fs/ext2/namei.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/namei.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext2/namei.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/namei.c 2008-04-21 10:12:49.000000000 -0400
@@ -31,6 +31,7 @@
*/
@@ -3255,9 +2729,9 @@
#include "xattr.h"
#include "acl.h"
@@ -66,6 +67,7 @@ static struct dentry *ext2_lookup(struct
- inode = iget(dir->i_sb, ino);
- if (!inode)
- return ERR_PTR(-EACCES);
+ inode = ext2_iget(dir->i_sb, ino);
+ if (IS_ERR(inode))
+ return ERR_CAST(inode);
+ dx_propagate_tag(nd, inode);
}
return d_splice_alias(inode, dentry);
@@ -3276,29 +2750,29 @@
.permission = ext2_permission,
+ .sync_flags = ext2_sync_flags,
};
---- linux-2.6.22/fs/ext2/super.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/super.c 2007-07-07 03:52:53 +0200
-@@ -321,7 +321,7 @@ enum {
+--- linux-2.6.25.4/fs/ext2/super.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/super.c 2008-04-19 15:14:52.000000000 -0400
+@@ -390,7 +390,8 @@ enum {
Opt_err_ro, Opt_nouid32, Opt_nocheck, Opt_debug,
Opt_oldalloc, Opt_orlov, Opt_nobh, Opt_user_xattr, Opt_nouser_xattr,
Opt_acl, Opt_noacl, Opt_xip, Opt_ignore, Opt_err, Opt_quota,
-- Opt_usrquota, Opt_grpquota
-+ Opt_usrquota, Opt_grpquota, Opt_tag, Opt_notag, Opt_tagid
+- Opt_usrquota, Opt_grpquota, Opt_reservation, Opt_noreservation
++ Opt_usrquota, Opt_grpquota, Opt_reservation, Opt_noreservation,
++ Opt_tag, Opt_notag, Opt_tagid
};
static match_table_t tokens = {
-@@ -349,6 +349,10 @@ static match_table_t tokens = {
+@@ -418,6 +419,9 @@ static match_table_t tokens = {
{Opt_acl, "acl"},
{Opt_noacl, "noacl"},
{Opt_xip, "xip"},
+ {Opt_tag, "tag"},
+ {Opt_notag, "notag"},
+ {Opt_tagid, "tagid=%u"},
-+ {Opt_tag, "tagxid"},
{Opt_grpquota, "grpquota"},
{Opt_ignore, "noquota"},
{Opt_quota, "quota"},
-@@ -417,6 +421,20 @@ static int parse_options (char * options
+@@ -488,6 +492,20 @@ static int parse_options (char * options
case Opt_nouid32:
set_opt (sbi->s_mount_opt, NO_UID32);
break;
@@ -3319,7 +2793,7 @@
case Opt_nocheck:
clear_opt (sbi->s_mount_opt, CHECK);
break;
-@@ -727,6 +745,8 @@ static int ext2_fill_super(struct super_
+@@ -831,6 +849,8 @@ static int ext2_fill_super(struct super_
if (!parse_options ((char *) data, sbi))
goto failed_mount;
@@ -3328,7 +2802,7 @@
sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
((EXT2_SB(sb)->s_mount_opt & EXT2_MOUNT_POSIX_ACL) ?
MS_POSIXACL : 0);
-@@ -1035,6 +1055,13 @@ static int ext2_remount (struct super_bl
+@@ -1164,6 +1184,13 @@ static int ext2_remount (struct super_bl
goto restore_opts;
}
@@ -3342,8 +2816,8 @@
sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
((sbi->s_mount_opt & EXT2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
---- linux-2.6.22/fs/ext2/symlink.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/symlink.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext2/symlink.c 2008-04-17 10:32:27.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/symlink.c 2008-04-19 15:14:52.000000000 -0400
@@ -38,6 +38,7 @@ const struct inode_operations ext2_symli
.listxattr = ext2_listxattr,
.removexattr = generic_removexattr,
@@ -3358,8 +2832,8 @@
#endif
+ .sync_flags = ext2_sync_flags,
};
---- linux-2.6.22/fs/ext2/xattr.c 2007-02-06 03:01:18 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext2/xattr.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext2/xattr.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext2/xattr.c 2008-04-19 15:14:52.000000000 -0400
@@ -60,6 +60,7 @@
#include <linux/mbcache.h>
#include <linux/quotaops.h>
@@ -3381,7 +2855,7 @@
unlock_buffer(new_bh);
goto cleanup;
}
-@@ -736,6 +741,7 @@ ext2_xattr_set2(struct inode *inode, str
+@@ -735,6 +740,7 @@ ext2_xattr_set2(struct inode *inode, str
le32_to_cpu(HDR(old_bh)->h_refcount) - 1);
if (ce)
mb_cache_entry_release(ce);
@@ -3389,7 +2863,7 @@
DQUOT_FREE_BLOCK(inode, 1);
mark_buffer_dirty(old_bh);
ea_bdebug(old_bh, "refcount now=%d",
-@@ -800,6 +806,7 @@ ext2_xattr_delete_inode(struct inode *in
+@@ -799,6 +805,7 @@ ext2_xattr_delete_inode(struct inode *in
mark_buffer_dirty(bh);
if (IS_SYNC(inode))
sync_dirty_buffer(bh);
@@ -3397,8 +2871,8 @@
DQUOT_FREE_BLOCK(inode, 1);
}
EXT2_I(inode)->i_file_acl = 0;
---- linux-2.6.22/fs/ext3/balloc.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/balloc.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext3/balloc.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/balloc.c 2008-04-19 15:14:52.000000000 -0400
@@ -19,6 +19,8 @@
#include <linux/ext3_jbd.h>
#include <linux/quotaops.h>
@@ -3408,7 +2882,7 @@
/*
* balloc.c contains the blocks allocation and deallocation routines
-@@ -613,8 +615,10 @@ void ext3_free_blocks(handle_t *handle,
+@@ -675,8 +677,10 @@ void ext3_free_blocks(handle_t *handle,
return;
}
ext3_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks);
@@ -3420,7 +2894,7 @@
return;
}
-@@ -1353,18 +1357,33 @@ out:
+@@ -1415,18 +1419,33 @@ out:
*
* Check if filesystem has at least 1 free block available for allocation.
*/
@@ -3461,7 +2935,7 @@
}
/**
-@@ -1381,7 +1400,7 @@ static int ext3_has_free_blocks(struct e
+@@ -1443,7 +1462,7 @@ static int ext3_has_free_blocks(struct e
*/
int ext3_should_retry_alloc(struct super_block *sb, int *retries)
{
@@ -3470,16 +2944,16 @@
return 0;
jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id);
-@@ -1444,6 +1463,8 @@ ext3_fsblk_t ext3_new_blocks(handle_t *h
+@@ -1506,6 +1525,8 @@ ext3_fsblk_t ext3_new_blocks(handle_t *h
*errp = -EDQUOT;
return 0;
}
-+ if (DLIMIT_ALLOC_BLOCK(inode, 1))
++ if (DLIMIT_ALLOC_BLOCK(inode, num))
+ goto out_dlimit;
sbi = EXT3_SB(sb);
es = EXT3_SB(sb)->s_es;
-@@ -1460,7 +1481,7 @@ ext3_fsblk_t ext3_new_blocks(handle_t *h
+@@ -1522,7 +1543,7 @@ ext3_fsblk_t ext3_new_blocks(handle_t *h
if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0))
my_rsv = &block_i->rsv_window_node;
@@ -3488,35 +2962,34 @@
*errp = -ENOSPC;
goto out;
}
-@@ -1653,6 +1674,9 @@ allocated:
+@@ -1710,12 +1731,16 @@ allocated:
+ *errp = 0;
+ brelse(bitmap_bh);
+ DQUOT_FREE_BLOCK(inode, *count-num);
++ DLIMIT_FREE_BLOCK(inode, *count-num);
+ *count = num;
+ return ret_block;
+
io_error:
*errp = -EIO;
out:
+ if (!performed_allocation)
-+ DLIMIT_FREE_BLOCK(inode, 1);
++ DLIMIT_FREE_BLOCK(inode, *count);
+out_dlimit:
if (fatal) {
*errp = fatal;
ext3_std_error(sb, fatal);
---- linux-2.6.22/fs/ext3/file.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/file.c 2007-06-15 02:37:03 +0200
-@@ -121,6 +121,7 @@ const struct file_operations ext3_file_o
- .release = ext3_release_file,
- .fsync = ext3_sync_file,
- .sendfile = generic_file_sendfile,
-+ .sendpage = generic_file_sendpage,
- .splice_read = generic_file_splice_read,
- .splice_write = generic_file_splice_write,
- };
-@@ -135,5 +136,6 @@ const struct inode_operations ext3_file_
+--- linux-2.6.25.4/fs/ext3/file.c 2008-04-17 10:37:23.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/file.c 2008-04-19 15:14:52.000000000 -0400
+@@ -134,5 +134,6 @@ const struct inode_operations ext3_file_
.removexattr = generic_removexattr,
#endif
.permission = ext3_permission,
+ .sync_flags = ext3_sync_flags,
};
---- linux-2.6.22/fs/ext3/ialloc.c 2006-11-30 21:19:19 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/ialloc.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext3/ialloc.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/ialloc.c 2008-04-19 15:14:52.000000000 -0400
@@ -23,6 +23,8 @@
#include <linux/buffer_head.h>
#include <linux/random.h>
@@ -3534,7 +3007,7 @@
is_directory = S_ISDIR(inode->i_mode);
-@@ -445,6 +448,12 @@ struct inode *ext3_new_inode(handle_t *h
+@@ -440,6 +443,12 @@ struct inode *ext3_new_inode(handle_t *h
inode = new_inode(sb);
if (!inode)
return ERR_PTR(-ENOMEM);
@@ -3547,7 +3020,7 @@
ei = EXT3_I(inode);
sbi = EXT3_SB(sb);
-@@ -566,7 +575,8 @@ got:
+@@ -559,7 +568,8 @@ got:
ei->i_dir_start_lookup = 0;
ei->i_disksize = 0;
@@ -3557,7 +3030,7 @@
if (S_ISLNK(mode))
ei->i_flags &= ~(EXT3_IMMUTABLE_FL|EXT3_APPEND_FL);
/* dirsync only applies to directories */
-@@ -621,6 +631,8 @@ got:
+@@ -614,6 +624,8 @@ got:
fail:
ext3_std_error(sb, err);
out:
@@ -3566,7 +3039,7 @@
iput(inode);
ret = ERR_PTR(err);
really_out:
-@@ -632,6 +644,7 @@ fail_free_drop:
+@@ -625,6 +637,7 @@ fail_free_drop:
fail_drop:
DQUOT_DROP(inode);
@@ -3574,8 +3047,8 @@
inode->i_flags |= S_NOQUOTA;
inode->i_nlink = 0;
iput(inode);
---- linux-2.6.22/fs/ext3/inode.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/inode.c 2007-07-07 03:52:53 +0200
+--- linux-2.6.25.4/fs/ext3/inode.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/inode.c 2008-04-20 13:25:49.000000000 -0400
@@ -36,6 +36,7 @@
#include <linux/mpage.h>
#include <linux/uio.h>
@@ -3584,7 +3057,7 @@
#include "xattr.h"
#include "acl.h"
-@@ -2237,7 +2238,7 @@ void ext3_truncate(struct inode *inode)
+@@ -2300,7 +2301,7 @@ void ext3_truncate(struct inode *inode)
return;
if (ext3_inode_is_fast_symlink(inode))
return;
@@ -3593,7 +3066,7 @@
return;
/*
-@@ -2559,13 +2560,20 @@ void ext3_set_inode_flags(struct inode *
+@@ -2622,13 +2623,20 @@ void ext3_set_inode_flags(struct inode *
{
unsigned int flags = EXT3_I(inode)->i_flags;
@@ -3617,7 +3090,7 @@
if (flags & EXT3_NOATIME_FL)
inode->i_flags |= S_NOATIME;
if (flags & EXT3_DIRSYNC_FL)
-@@ -2591,6 +2599,45 @@ void ext3_get_inode_flags(struct ext3_in
+@@ -2654,6 +2662,45 @@ void ext3_get_inode_flags(struct ext3_in
ei->i_flags |= EXT3_DIRSYNC_FL;
}
@@ -3660,19 +3133,19 @@
+ return err;
+}
+
- void ext3_read_inode(struct inode * inode)
+ struct inode *ext3_iget(struct super_block *sb, unsigned long ino)
{
struct ext3_iloc iloc;
-@@ -2598,6 +2645,8 @@ void ext3_read_inode(struct inode * inod
- struct ext3_inode_info *ei = EXT3_I(inode);
- struct buffer_head *bh;
+@@ -2663,6 +2710,8 @@ struct inode *ext3_iget(struct super_blo
+ struct inode *inode;
+ long ret;
int block;
+ uid_t uid;
+ gid_t gid;
- #ifdef CONFIG_EXT3_FS_POSIX_ACL
- ei->i_acl = EXT3_ACL_NOT_CACHED;
-@@ -2610,12 +2659,17 @@ void ext3_read_inode(struct inode * inod
+ inode = iget_locked(sb, ino);
+ if (!inode)
+@@ -2683,12 +2732,17 @@ struct inode *ext3_iget(struct super_blo
bh = iloc.bh;
raw_inode = ext3_raw_inode(&iloc);
inode->i_mode = le16_to_cpu(raw_inode->i_mode);
@@ -3694,7 +3167,7 @@
inode->i_nlink = le16_to_cpu(raw_inode->i_links_count);
inode->i_size = le32_to_cpu(raw_inode->i_size);
inode->i_atime.tv_sec = (signed)le32_to_cpu(raw_inode->i_atime);
-@@ -2741,6 +2795,8 @@ static int ext3_do_update_inode(handle_t
+@@ -2817,6 +2871,8 @@ static int ext3_do_update_inode(handle_t
struct ext3_inode *raw_inode = ext3_raw_inode(iloc);
struct ext3_inode_info *ei = EXT3_I(inode);
struct buffer_head *bh = iloc->bh;
@@ -3703,7 +3176,7 @@
int err = 0, rc, block;
/* For fields not not tracking in the in-memory inode,
-@@ -2751,29 +2807,32 @@ static int ext3_do_update_inode(handle_t
+@@ -2827,29 +2883,32 @@ static int ext3_do_update_inode(handle_t
ext3_get_inode_flags(ei);
raw_inode->i_mode = cpu_to_le16(inode->i_mode);
if(!(test_opt(inode->i_sb, NO_UID32))) {
@@ -3742,7 +3215,7 @@
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
raw_inode->i_size = cpu_to_le32(ei->i_disksize);
raw_inode->i_atime = cpu_to_le32(inode->i_atime.tv_sec);
-@@ -2926,7 +2985,8 @@ int ext3_setattr(struct dentry *dentry,
+@@ -3002,7 +3061,8 @@ int ext3_setattr(struct dentry *dentry,
return error;
if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
@@ -3752,7 +3225,7 @@
handle_t *handle;
/* (user+group)*(old+new) structure, inode write (sb,
-@@ -2948,6 +3008,8 @@ int ext3_setattr(struct dentry *dentry,
+@@ -3024,6 +3084,8 @@ int ext3_setattr(struct dentry *dentry,
inode->i_uid = attr->ia_uid;
if (attr->ia_valid & ATTR_GID)
inode->i_gid = attr->ia_gid;
@@ -3761,8 +3234,8 @@
error = ext3_mark_inode_dirty(handle, inode);
ext3_journal_stop(handle);
}
---- linux-2.6.22/fs/ext3/ioctl.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/ioctl.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext3/ioctl.c 2008-04-17 11:31:35.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/ioctl.c 2008-04-19 15:14:52.000000000 -0400
@@ -8,6 +8,7 @@
*/
@@ -3788,8 +3261,8 @@
+ (filp && MNT_IS_RDONLY(filp->f_vfsmnt)))
return -EROFS;
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
-@@ -62,7 +65,9 @@ int ext3_ioctl (struct inode * inode, st
+ if (!is_owner_or_cap(inode))
+@@ -67,7 +70,9 @@ int ext3_ioctl (struct inode * inode, st
*
* This test looks nicer. Thanks to Pauline Middelink
*/
@@ -3800,9 +3273,9 @@
if (!capable(CAP_LINUX_IMMUTABLE)) {
mutex_unlock(&inode->i_mutex);
return -EPERM;
-@@ -124,7 +129,8 @@ flags_err:
+@@ -129,7 +134,8 @@ flags_err:
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
+ if (!is_owner_or_cap(inode))
return -EPERM;
- if (IS_RDONLY(inode))
+ if (IS_RDONLY(inode) ||
@@ -3810,7 +3283,7 @@
return -EROFS;
if (get_user(generation, (int __user *) arg))
return -EFAULT;
-@@ -178,7 +184,8 @@ flags_err:
+@@ -183,7 +189,8 @@ flags_err:
if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode))
return -ENOTTY;
@@ -3819,8 +3292,8 @@
+ (filp && MNT_IS_RDONLY(filp->f_vfsmnt)))
return -EROFS;
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
-@@ -213,7 +220,8 @@ flags_err:
+ if (!is_owner_or_cap(inode))
+@@ -218,7 +225,8 @@ flags_err:
if (!capable(CAP_SYS_RESOURCE))
return -EPERM;
@@ -3830,7 +3303,7 @@
return -EROFS;
if (get_user(n_blocks_count, (__u32 __user *)arg))
-@@ -234,7 +242,8 @@ flags_err:
+@@ -239,7 +247,8 @@ flags_err:
if (!capable(CAP_SYS_RESOURCE))
return -EPERM;
@@ -3840,64 +3313,25 @@
return -EROFS;
if (copy_from_user(&input, (struct ext3_new_group_input __user *)arg,
-@@ -249,6 +258,38 @@ flags_err:
- return err;
- }
-
-+#if defined(CONFIG_VSERVER_LEGACY) && !defined(CONFIG_TAGGING_NONE)
-+ case EXT3_IOC_SETTAG: {
-+ handle_t *handle;
-+ struct ext3_iloc iloc;
-+ int tag;
-+ int err;
-+
-+ /* fixme: if stealth, return -ENOTTY */
-+ if (!capable(CAP_CONTEXT))
-+ return -EPERM;
-+ if (IS_RDONLY(inode))
-+ return -EROFS;
-+ if (!(inode->i_sb->s_flags & MS_TAGGED))
-+ return -ENOSYS;
-+ if (get_user(tag, (int __user *) arg))
-+ return -EFAULT;
-+
-+ handle = ext3_journal_start(inode, 1);
-+ if (IS_ERR(handle))
-+ return PTR_ERR(handle);
-+ err = ext3_reserve_inode_write(handle, inode, &iloc);
-+ if (err)
-+ return err;
-+
-+ inode->i_tag = (tag & 0xFFFF);
-+ inode->i_ctime = CURRENT_TIME;
-+
-+ err = ext3_mark_iloc_dirty(handle, inode, &iloc);
-+ ext3_journal_stop(handle);
-+ return err;
-+ }
-+#endif
-
- default:
- return -ENOTTY;
---- linux-2.6.22/fs/ext3/namei.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/namei.c 2007-06-15 03:00:11 +0200
-@@ -36,6 +36,7 @@
- #include <linux/quotaops.h>
- #include <linux/buffer_head.h>
- #include <linux/bio.h>
-+#include <linux/vs_tag.h>
+--- linux-2.6.25.4/fs/ext3/namei.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/namei.c 2008-04-19 15:14:52.000000000 -0400
+@@ -36,6 +36,7 @@
+ #include <linux/quotaops.h>
+ #include <linux/buffer_head.h>
+ #include <linux/bio.h>
++#include <linux/vs_tag.h>
#include "namei.h"
#include "xattr.h"
-@@ -1019,6 +1020,7 @@ static struct dentry *ext3_lookup(struct
-
- if (!inode)
- return ERR_PTR(-EACCES);
+@@ -907,6 +908,7 @@ restart:
+ if (bh)
+ ll_rw_block(READ_META, 1, &bh);
+ }
+ dx_propagate_tag(nd, inode);
- }
- return d_splice_alias(inode, dentry);
- }
-@@ -2386,6 +2388,7 @@ const struct inode_operations ext3_dir_i
+ }
+ if ((bh = bh_use[ra_ptr++]) == NULL)
+ goto next;
+@@ -2417,6 +2419,7 @@ const struct inode_operations ext3_dir_i
.removexattr = generic_removexattr,
#endif
.permission = ext3_permission,
@@ -3905,15 +3339,15 @@
};
const struct inode_operations ext3_special_inode_operations = {
-@@ -2397,4 +2400,5 @@ const struct inode_operations ext3_speci
+@@ -2428,4 +2431,5 @@ const struct inode_operations ext3_speci
.removexattr = generic_removexattr,
#endif
.permission = ext3_permission,
+ .sync_flags = ext3_sync_flags,
};
---- linux-2.6.22/fs/ext3/super.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/super.c 2007-06-15 02:37:03 +0200
-@@ -674,7 +674,7 @@ enum {
+--- linux-2.6.25.4/fs/ext3/super.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/super.c 2008-04-20 13:26:55.000000000 -0400
+@@ -756,7 +756,7 @@ enum {
Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota,
@@ -3922,18 +3356,17 @@
};
static match_table_t tokens = {
-@@ -724,6 +724,10 @@ static match_table_t tokens = {
- {Opt_quota, "quota"},
+@@ -807,6 +807,9 @@ static match_table_t tokens = {
{Opt_usrquota, "usrquota"},
{Opt_barrier, "barrier=%u"},
+ {Opt_resize, "resize"},
+ {Opt_tag, "tag"},
+ {Opt_notag, "notag"},
+ {Opt_tagid, "tagid=%u"},
-+ {Opt_tag, "tagxid"},
{Opt_err, NULL},
- {Opt_resize, "resize"},
};
-@@ -817,6 +821,20 @@ static int parse_options (char *options,
+
+@@ -899,6 +902,20 @@ static int parse_options (char *options,
case Opt_nouid32:
set_opt (sbi->s_mount_opt, NO_UID32);
break;
@@ -3954,7 +3387,7 @@
case Opt_nocheck:
clear_opt (sbi->s_mount_opt, CHECK);
break;
-@@ -1487,6 +1505,9 @@ static int ext3_fill_super (struct super
+@@ -1591,6 +1608,9 @@ static int ext3_fill_super (struct super
NULL, 0))
goto failed_mount;
@@ -3964,7 +3397,7 @@
sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
((sbi->s_mount_opt & EXT3_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
-@@ -2302,6 +2323,12 @@ static int ext3_remount (struct super_bl
+@@ -2428,6 +2448,12 @@ static int ext3_remount (struct super_bl
if (sbi->s_mount_opt & EXT3_MOUNT_ABORT)
ext3_abort(sb, __FUNCTION__, "Abort forced by user");
@@ -3977,8 +3410,8 @@
sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
((sbi->s_mount_opt & EXT3_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
---- linux-2.6.22/fs/ext3/symlink.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/symlink.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext3/symlink.c 2008-04-17 10:32:27.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/symlink.c 2008-04-19 15:14:52.000000000 -0400
@@ -40,6 +40,7 @@ const struct inode_operations ext3_symli
.listxattr = ext3_listxattr,
.removexattr = generic_removexattr,
@@ -3993,8 +3426,8 @@
#endif
+ .sync_flags = ext3_sync_flags,
};
---- linux-2.6.22/fs/ext3/xattr.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext3/xattr.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext3/xattr.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext3/xattr.c 2008-04-19 15:14:52.000000000 -0400
@@ -58,6 +58,7 @@
#include <linux/mbcache.h>
#include <linux/quotaops.h>
@@ -4003,7 +3436,7 @@
#include "xattr.h"
#include "acl.h"
-@@ -497,6 +498,7 @@ ext3_xattr_release_block(handle_t *handl
+@@ -496,6 +497,7 @@ ext3_xattr_release_block(handle_t *handl
error = ext3_journal_dirty_metadata(handle, bh);
if (IS_SYNC(inode))
handle->h_sync = 1;
@@ -4011,7 +3444,7 @@
DQUOT_FREE_BLOCK(inode, 1);
ea_bdebug(bh, "refcount now=%d; releasing",
le32_to_cpu(BHDR(bh)->h_refcount));
-@@ -771,11 +773,14 @@ inserted:
+@@ -769,11 +771,14 @@ inserted:
if (new_bh == bs->bh)
ea_bdebug(new_bh, "keeping");
else {
@@ -4027,7 +3460,7 @@
error = ext3_journal_get_write_access(handle,
new_bh);
if (error)
-@@ -852,6 +857,8 @@ cleanup:
+@@ -849,6 +854,8 @@ cleanup:
cleanup_dquot:
DQUOT_FREE_BLOCK(inode, 1);
@@ -4036,8 +3469,8 @@
goto cleanup;
bad_block:
---- linux-2.6.22/fs/ext4/balloc.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/balloc.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext4/balloc.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/balloc.c 2008-04-21 10:23:33.000000000 -0400
@@ -19,6 +19,8 @@
#include <linux/ext4_jbd2.h>
#include <linux/quotaops.h>
@@ -4045,12 +3478,12 @@
+#include <linux/vs_dlimit.h>
+#include <linux/vs_tag.h>
+ #include "group.h"
/*
- * balloc.c contains the blocks allocation and deallocation routines
-@@ -630,8 +632,10 @@ void ext4_free_blocks(handle_t *handle,
- return;
- }
- ext4_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks);
+@@ -810,8 +812,10 @@ void ext4_free_blocks(handle_t *handle,
+ else
+ ext4_mb_free_blocks(handle, inode, block, count,
+ metadata, &dquot_freed_blocks);
- if (dquot_freed_blocks)
+ if (dquot_freed_blocks) {
+ DLIMIT_FREE_BLOCK(inode, dquot_freed_blocks);
@@ -4059,7 +3492,7 @@
return;
}
-@@ -1370,18 +1374,33 @@ out:
+@@ -1551,18 +1555,33 @@ out:
*
* Check if filesystem has at least 1 free block available for allocation.
*/
@@ -4099,7 +3532,7 @@
}
/**
-@@ -1398,7 +1417,7 @@ static int ext4_has_free_blocks(struct e
+@@ -1579,7 +1598,7 @@ static int ext4_has_free_blocks(struct e
*/
int ext4_should_retry_alloc(struct super_block *sb, int *retries)
{
@@ -4108,16 +3541,16 @@
return 0;
jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id);
-@@ -1461,6 +1480,8 @@ ext4_fsblk_t ext4_new_blocks(handle_t *h
+@@ -1639,6 +1658,8 @@ ext4_fsblk_t ext4_new_blocks_old(handle_
*errp = -EDQUOT;
return 0;
}
-+ if (DLIMIT_ALLOC_BLOCK(inode, 1))
++ if (DLIMIT_ALLOC_BLOCK(inode, num))
+ goto out_dlimit;
sbi = EXT4_SB(sb);
es = EXT4_SB(sb)->s_es;
-@@ -1477,7 +1498,7 @@ ext4_fsblk_t ext4_new_blocks(handle_t *h
+@@ -1655,7 +1676,7 @@ ext4_fsblk_t ext4_new_blocks_old(handle_
if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0))
my_rsv = &block_i->rsv_window_node;
@@ -4126,35 +3559,34 @@
*errp = -ENOSPC;
goto out;
}
-@@ -1667,6 +1688,9 @@ allocated:
+@@ -1841,12 +1862,16 @@ allocated:
+ *errp = 0;
+ brelse(bitmap_bh);
+ DQUOT_FREE_BLOCK(inode, *count-num);
++ DLIMIT_FREE_BLOCK(inode, *count-num);
+ *count = num;
+ return ret_block;
+
io_error:
*errp = -EIO;
out:
+ if (!performed_allocation)
-+ DLIMIT_FREE_BLOCK(inode, 1);
++ DLIMIT_FREE_BLOCK(inode, *count);
+out_dlimit:
if (fatal) {
*errp = fatal;
ext4_std_error(sb, fatal);
---- linux-2.6.22/fs/ext4/file.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/file.c 2007-06-15 02:37:03 +0200
-@@ -121,6 +121,7 @@ const struct file_operations ext4_file_o
- .release = ext4_release_file,
- .fsync = ext4_sync_file,
- .sendfile = generic_file_sendfile,
-+ .sendpage = generic_file_sendpage,
- .splice_read = generic_file_splice_read,
- .splice_write = generic_file_splice_write,
- };
-@@ -135,5 +136,6 @@ const struct inode_operations ext4_file_
- .removexattr = generic_removexattr,
+--- linux-2.6.25.4/fs/ext4/file.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/file.c 2008-04-19 15:14:52.000000000 -0400
+@@ -152,5 +152,6 @@ const struct inode_operations ext4_file_
#endif
.permission = ext4_permission,
+ .fallocate = ext4_fallocate,
+ .sync_flags = ext4_sync_flags,
};
---- linux-2.6.22/fs/ext4/ialloc.c 2006-11-30 21:19:20 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/ialloc.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext4/ialloc.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/ialloc.c 2008-04-21 10:29:28.000000000 -0400
@@ -24,6 +24,8 @@
#include <linux/random.h>
#include <linux/bitops.h>
@@ -4164,7 +3596,7 @@
#include <asm/byteorder.h>
#include "xattr.h"
-@@ -127,6 +129,7 @@ void ext4_free_inode (handle_t *handle,
+@@ -186,6 +188,7 @@ void ext4_free_inode (handle_t *handle,
ext4_xattr_delete_inode(handle, inode);
DQUOT_FREE_INODE(inode);
DQUOT_DROP(inode);
@@ -4172,7 +3604,7 @@
is_directory = S_ISDIR(inode->i_mode);
-@@ -448,6 +451,12 @@ struct inode *ext4_new_inode(handle_t *h
+@@ -513,6 +516,12 @@ struct inode *ext4_new_inode(handle_t *h
inode = new_inode(sb);
if (!inode)
return ERR_PTR(-ENOMEM);
@@ -4185,17 +3617,17 @@
ei = EXT4_I(inode);
sbi = EXT4_SB(sb);
-@@ -569,7 +578,8 @@ got:
- ei->i_dir_start_lookup = 0;
- ei->i_disksize = 0;
-
-- ei->i_flags = EXT4_I(dir)->i_flags & ~EXT4_INDEX_FL;
+@@ -707,7 +716,8 @@ got:
+ * newly created directory and file only if -o extent mount option is
+ * specified
+ */
+- ei->i_flags = EXT4_I(dir)->i_flags & ~(EXT4_INDEX_FL|EXT4_EXTENTS_FL);
+ ei->i_flags = EXT4_I(dir)->i_flags &
-+ ~(EXT4_INDEX_FL|EXT4_IUNLINK_FL|EXT4_BARRIER_FL);
++ ~(EXT4_INDEX_FL|EXT4_EXTENTS_FL|EXT4_IUNLINK_FL|EXT4_BARRIER_FL);
if (S_ISLNK(mode))
ei->i_flags &= ~(EXT4_IMMUTABLE_FL|EXT4_APPEND_FL);
/* dirsync only applies to directories */
-@@ -635,6 +645,8 @@ got:
+@@ -766,6 +776,8 @@ got:
fail:
ext4_std_error(sb, err);
out:
@@ -4204,7 +3636,7 @@
iput(inode);
ret = ERR_PTR(err);
really_out:
-@@ -646,6 +658,7 @@ fail_free_drop:
+@@ -777,6 +789,7 @@ fail_free_drop:
fail_drop:
DQUOT_DROP(inode);
@@ -4212,8 +3644,8 @@
inode->i_flags |= S_NOQUOTA;
inode->i_nlink = 0;
iput(inode);
---- linux-2.6.22/fs/ext4/inode.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/inode.c 2007-07-07 03:52:53 +0200
+--- linux-2.6.25.4/fs/ext4/inode.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/inode.c 2008-04-21 10:20:53.000000000 -0400
@@ -36,6 +36,7 @@
#include <linux/mpage.h>
#include <linux/uio.h>
@@ -4222,7 +3654,7 @@
#include "xattr.h"
#include "acl.h"
-@@ -2244,7 +2245,7 @@ void ext4_truncate(struct inode *inode)
+@@ -2342,7 +2343,7 @@ void ext4_truncate(struct inode *inode)
return;
if (ext4_inode_is_fast_symlink(inode))
return;
@@ -4231,7 +3663,7 @@
return;
/*
-@@ -2570,19 +2571,65 @@ void ext4_set_inode_flags(struct inode *
+@@ -2671,13 +2672,20 @@ void ext4_set_inode_flags(struct inode *
{
unsigned int flags = EXT4_I(inode)->i_flags;
@@ -4255,9 +3687,11 @@
if (flags & EXT4_NOATIME_FL)
inode->i_flags |= S_NOATIME;
if (flags & EXT4_DIRSYNC_FL)
- inode->i_flags |= S_DIRSYNC;
+@@ -2702,6 +2710,46 @@ void ext4_get_inode_flags(struct ext4_in
+ if (flags & S_DIRSYNC)
+ ei->i_flags |= EXT4_DIRSYNC_FL;
}
-
++
+int ext4_sync_flags(struct inode *inode)
+{
+ unsigned int oldflags, newflags;
@@ -4297,19 +3731,19 @@
+ return err;
+}
+
- void ext4_read_inode(struct inode * inode)
+ static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode,
+ struct ext4_inode_info *ei)
{
- struct ext4_iloc iloc;
-@@ -2590,6 +2637,8 @@ void ext4_read_inode(struct inode * inod
- struct ext4_inode_info *ei = EXT4_I(inode);
- struct buffer_head *bh;
+@@ -2734,6 +2782,8 @@ struct inode *ext4_iget(struct super_blo
+ struct inode *inode;
+ long ret;
int block;
+ uid_t uid;
+ gid_t gid;
- #ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
- ei->i_acl = EXT4_ACL_NOT_CACHED;
-@@ -2602,12 +2651,17 @@ void ext4_read_inode(struct inode * inod
+ inode = iget_locked(sb, ino);
+ if (!inode)
+@@ -2754,12 +2804,17 @@ struct inode *ext4_iget(struct super_blo
bh = iloc.bh;
raw_inode = ext4_raw_inode(&iloc);
inode->i_mode = le16_to_cpu(raw_inode->i_mode);
@@ -4329,9 +3763,9 @@
+ le16_to_cpu(raw_inode->i_raw_tag));
+
inode->i_nlink = le16_to_cpu(raw_inode->i_links_count);
- inode->i_size = le32_to_cpu(raw_inode->i_size);
- inode->i_atime.tv_sec = (signed)le32_to_cpu(raw_inode->i_atime);
-@@ -2737,6 +2791,8 @@ static int ext4_do_update_inode(handle_t
+
+ ei->i_state = 0;
+@@ -2933,6 +2988,8 @@ static int ext4_do_update_inode(handle_t
struct ext4_inode *raw_inode = ext4_raw_inode(iloc);
struct ext4_inode_info *ei = EXT4_I(inode);
struct buffer_head *bh = iloc->bh;
@@ -4340,8 +3774,8 @@
int err = 0, rc, block;
/* For fields not not tracking in the in-memory inode,
-@@ -2746,29 +2802,32 @@ static int ext4_do_update_inode(handle_t
-
+@@ -2943,29 +3000,32 @@ static int ext4_do_update_inode(handle_t
+ ext4_get_inode_flags(ei);
raw_inode->i_mode = cpu_to_le16(inode->i_mode);
if(!(test_opt(inode->i_sb, NO_UID32))) {
- raw_inode->i_uid_low = cpu_to_le16(low_16_bits(inode->i_uid));
@@ -4377,9 +3811,9 @@
+ raw_inode->i_raw_tag = cpu_to_le16(inode->i_tag);
+#endif
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
- raw_inode->i_size = cpu_to_le32(ei->i_disksize);
- raw_inode->i_atime = cpu_to_le32(inode->i_atime.tv_sec);
-@@ -2925,7 +2984,8 @@ int ext4_setattr(struct dentry *dentry,
+
+ EXT4_INODE_SET_XTIME(i_ctime, inode, raw_inode);
+@@ -3121,7 +3181,8 @@ int ext4_setattr(struct dentry *dentry,
return error;
if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
@@ -4389,7 +3823,7 @@
handle_t *handle;
/* (user+group)*(old+new) structure, inode write (sb,
-@@ -2947,6 +3007,8 @@ int ext4_setattr(struct dentry *dentry,
+@@ -3143,6 +3204,8 @@ int ext4_setattr(struct dentry *dentry,
inode->i_uid = attr->ia_uid;
if (attr->ia_valid & ATTR_GID)
inode->i_gid = attr->ia_gid;
@@ -4398,8 +3832,8 @@
error = ext4_mark_inode_dirty(handle, inode);
ext4_journal_stop(handle);
}
---- linux-2.6.22/fs/ext4/ioctl.c 2007-02-06 03:01:18 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/ioctl.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext4/ioctl.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/ioctl.c 2008-04-19 15:14:52.000000000 -0400
@@ -8,6 +8,7 @@
*/
@@ -4416,7 +3850,7 @@
#include <asm/uaccess.h>
int ext4_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
-@@ -37,7 +39,8 @@ int ext4_ioctl (struct inode * inode, st
+@@ -38,7 +40,8 @@ int ext4_ioctl (struct inode * inode, st
unsigned int oldflags;
unsigned int jflag;
@@ -4425,8 +3859,8 @@
+ (filp && MNT_IS_RDONLY(filp->f_vfsmnt)))
return -EROFS;
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
-@@ -61,7 +64,9 @@ int ext4_ioctl (struct inode * inode, st
+ if (!is_owner_or_cap(inode))
+@@ -67,7 +70,9 @@ int ext4_ioctl (struct inode * inode, st
*
* This test looks nicer. Thanks to Pauline Middelink
*/
@@ -4437,9 +3871,9 @@
if (!capable(CAP_LINUX_IMMUTABLE)) {
mutex_unlock(&inode->i_mutex);
return -EPERM;
-@@ -123,7 +128,8 @@ flags_err:
+@@ -129,7 +134,8 @@ flags_err:
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
+ if (!is_owner_or_cap(inode))
return -EPERM;
- if (IS_RDONLY(inode))
+ if (IS_RDONLY(inode) ||
@@ -4447,7 +3881,7 @@
return -EROFS;
if (get_user(generation, (int __user *) arg))
return -EFAULT;
-@@ -177,7 +183,8 @@ flags_err:
+@@ -183,7 +189,8 @@ flags_err:
if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode))
return -ENOTTY;
@@ -4456,8 +3890,8 @@
+ (filp && MNT_IS_RDONLY(filp->f_vfsmnt)))
return -EROFS;
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
-@@ -212,7 +219,8 @@ flags_err:
+ if (!is_owner_or_cap(inode))
+@@ -218,7 +225,8 @@ flags_err:
if (!capable(CAP_SYS_RESOURCE))
return -EPERM;
@@ -4467,7 +3901,7 @@
return -EROFS;
if (get_user(n_blocks_count, (__u32 __user *)arg))
-@@ -233,7 +241,8 @@ flags_err:
+@@ -239,7 +247,8 @@ flags_err:
if (!capable(CAP_SYS_RESOURCE))
return -EPERM;
@@ -4477,48 +3911,8 @@
return -EROFS;
if (copy_from_user(&input, (struct ext4_new_group_input __user *)arg,
-@@ -248,6 +257,39 @@ flags_err:
- return err;
- }
-
-+#if defined(CONFIG_VSERVER_LEGACY) && !defined(CONFIG_TAGGING_NONE)
-+ case EXT4_IOC_SETTAG: {
-+ handle_t *handle;
-+ struct ext4_iloc iloc;
-+ int tag;
-+ int err;
-+
-+ /* fixme: if stealth, return -ENOTTY */
-+ if (!capable(CAP_CONTEXT))
-+ return -EPERM;
-+ if (IS_RDONLY(inode))
-+ return -EROFS;
-+ if (!(inode->i_sb->s_flags & MS_TAGGED))
-+ return -ENOSYS;
-+ if (get_user(tag, (int __user *) arg))
-+ return -EFAULT;
-+
-+ handle = ext4_journal_start(inode, 1);
-+ if (IS_ERR(handle))
-+ return PTR_ERR(handle);
-+ err = ext4_reserve_inode_write(handle, inode, &iloc);
-+ if (err)
-+ return err;
-+
-+ inode->i_tag = (tag & 0xFFFF);
-+ inode->i_ctime = CURRENT_TIME;
-+
-+ err = ext4_mark_iloc_dirty(handle, inode, &iloc);
-+ ext4_journal_stop(handle);
-+ return err;
-+ }
-+#endif
-+
- default:
- return -ENOTTY;
- }
---- linux-2.6.22/fs/ext4/namei.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/namei.c 2007-06-15 03:11:13 +0200
+--- linux-2.6.25.4/fs/ext4/namei.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/namei.c 2008-04-19 15:14:52.000000000 -0400
@@ -36,6 +36,7 @@
#include <linux/quotaops.h>
#include <linux/buffer_head.h>
@@ -4527,15 +3921,15 @@
#include "namei.h"
#include "xattr.h"
-@@ -1017,6 +1018,7 @@ static struct dentry *ext4_lookup(struct
-
- if (!inode)
- return ERR_PTR(-EACCES);
+@@ -908,6 +909,7 @@ restart:
+ if (bh)
+ ll_rw_block(READ_META, 1, &bh);
+ }
+ dx_propagate_tag(nd, inode);
- }
- return d_splice_alias(inode, dentry);
- }
-@@ -2384,6 +2386,7 @@ const struct inode_operations ext4_dir_i
+ }
+ if ((bh = bh_use[ra_ptr++]) == NULL)
+ goto next;
+@@ -2448,6 +2450,7 @@ const struct inode_operations ext4_dir_i
.removexattr = generic_removexattr,
#endif
.permission = ext4_permission,
@@ -4543,35 +3937,36 @@
};
const struct inode_operations ext4_special_inode_operations = {
-@@ -2395,4 +2398,5 @@ const struct inode_operations ext4_speci
+@@ -2459,4 +2462,5 @@ const struct inode_operations ext4_speci
.removexattr = generic_removexattr,
#endif
.permission = ext4_permission,
+ .sync_flags = ext4_sync_flags,
};
---- linux-2.6.22/fs/ext4/super.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/super.c 2007-06-15 02:37:03 +0200
-@@ -725,7 +725,7 @@ enum {
- Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
- Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
+--- linux-2.6.25.4/fs/ext4/super.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/super.c 2008-04-21 10:31:22.000000000 -0400
+@@ -887,6 +887,7 @@ enum {
Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota,
-- Opt_grpquota, Opt_extents,
-+ Opt_grpquota, Opt_extents, Opt_tag, Opt_notag, Opt_tagid
+ Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version,
+ Opt_mballoc, Opt_nomballoc, Opt_stripe,
++ Opt_tag, Opt_notag, Opt_tagid
};
static match_table_t tokens = {
-@@ -776,6 +776,10 @@ static match_table_t tokens = {
- {Opt_usrquota, "usrquota"},
- {Opt_barrier, "barrier=%u"},
- {Opt_extents, "extents"},
+@@ -944,8 +945,11 @@ static match_table_t tokens = {
+ {Opt_mballoc, "mballoc"},
+ {Opt_nomballoc, "nomballoc"},
+ {Opt_stripe, "stripe=%u"},
+- {Opt_err, NULL},
+ {Opt_resize, "resize"},
+ {Opt_tag, "tag"},
+ {Opt_notag, "notag"},
+ {Opt_tagid, "tagid=%u"},
-+ {Opt_tag, "tagxid"},
- {Opt_err, NULL},
- {Opt_resize, "resize"},
++ {Opt_err, NULL},
};
-@@ -869,6 +873,20 @@ static int parse_options (char *options,
+
+ static ext4_fsblk_t get_sb_block(void **data)
+@@ -1037,6 +1041,20 @@ static int parse_options (char *options,
case Opt_nouid32:
set_opt (sbi->s_mount_opt, NO_UID32);
break;
@@ -4592,7 +3987,7 @@
case Opt_nocheck:
clear_opt (sbi->s_mount_opt, CHECK);
break;
-@@ -1546,6 +1564,9 @@ static int ext4_fill_super (struct super
+@@ -1909,6 +1927,9 @@ static int ext4_fill_super (struct super
NULL, 0))
goto failed_mount;
@@ -4602,7 +3997,7 @@
sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
-@@ -2377,6 +2398,12 @@ static int ext4_remount (struct super_bl
+@@ -2828,6 +2849,12 @@ static int ext4_remount (struct super_bl
if (sbi->s_mount_opt & EXT4_MOUNT_ABORT)
ext4_abort(sb, __FUNCTION__, "Abort forced by user");
@@ -4615,8 +4010,8 @@
sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
---- linux-2.6.22/fs/ext4/symlink.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/symlink.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext4/symlink.c 2008-04-17 10:32:27.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/symlink.c 2008-04-19 15:14:52.000000000 -0400
@@ -40,6 +40,7 @@ const struct inode_operations ext4_symli
.listxattr = ext4_listxattr,
.removexattr = generic_removexattr,
@@ -4631,8 +4026,8 @@
#endif
+ .sync_flags = ext4_sync_flags,
};
---- linux-2.6.22/fs/ext4/xattr.c 2007-05-02 19:25:17 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ext4/xattr.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ext4/xattr.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ext4/xattr.c 2008-04-19 15:14:52.000000000 -0400
@@ -58,6 +58,7 @@
#include <linux/mbcache.h>
#include <linux/quotaops.h>
@@ -4641,7 +4036,7 @@
#include "xattr.h"
#include "acl.h"
-@@ -496,6 +497,7 @@ ext4_xattr_release_block(handle_t *handl
+@@ -489,6 +490,7 @@ ext4_xattr_release_block(handle_t *handl
error = ext4_journal_dirty_metadata(handle, bh);
if (IS_SYNC(inode))
handle->h_sync = 1;
@@ -4649,7 +4044,7 @@
DQUOT_FREE_BLOCK(inode, 1);
ea_bdebug(bh, "refcount now=%d; releasing",
le32_to_cpu(BHDR(bh)->h_refcount));
-@@ -769,11 +771,14 @@ inserted:
+@@ -779,11 +781,14 @@ inserted:
if (new_bh == bs->bh)
ea_bdebug(new_bh, "keeping");
else {
@@ -4665,7 +4060,7 @@
error = ext4_journal_get_write_access(handle,
new_bh);
if (error)
-@@ -850,6 +855,8 @@ cleanup:
+@@ -860,6 +865,8 @@ cleanup:
cleanup_dquot:
DQUOT_FREE_BLOCK(inode, 1);
@@ -4674,17 +4069,17 @@
goto cleanup;
bad_block:
---- linux-2.6.22/fs/fcntl.c 2007-02-06 03:01:18 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/fcntl.c 2007-06-15 02:37:03 +0200
-@@ -18,6 +18,7 @@
- #include <linux/ptrace.h>
+--- linux-2.6.25.4/fs/fcntl.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/fcntl.c 2008-04-19 15:14:52.000000000 -0400
+@@ -19,6 +19,7 @@
#include <linux/signal.h>
#include <linux/rcupdate.h>
+ #include <linux/pid_namespace.h>
+#include <linux/vs_limit.h>
#include <asm/poll.h>
#include <asm/siginfo.h>
-@@ -84,6 +85,8 @@ repeat:
+@@ -85,6 +86,8 @@ repeat:
error = -EMFILE;
if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
goto out;
@@ -4693,15 +4088,15 @@
error = expand_files(files, newfd);
if (error < 0)
-@@ -124,6 +127,7 @@ static int dupfd(struct file *file, unsi
- FD_SET(fd, fdt->open_fds);
- FD_CLR(fd, fdt->close_on_exec);
+@@ -128,6 +131,7 @@ static int dupfd(struct file *file, unsi
+ else
+ FD_CLR(fd, fdt->close_on_exec);
spin_unlock(&files->file_lock);
+ vx_openfd_inc(fd);
fd_install(fd, file);
} else {
spin_unlock(&files->file_lock);
-@@ -176,6 +180,9 @@ asmlinkage long sys_dup2(unsigned int ol
+@@ -180,6 +184,9 @@ asmlinkage long sys_dup2(unsigned int ol
if (tofree)
filp_close(tofree, files);
@@ -4711,8 +4106,8 @@
err = newfd;
out:
return err;
---- linux-2.6.22/fs/file_table.c 2007-07-09 13:19:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/file_table.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/file_table.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/file_table.c 2008-04-19 15:14:52.000000000 -0400
@@ -20,6 +20,8 @@
#include <linux/fsnotify.h>
#include <linux/sysctl.h>
@@ -4722,7 +4117,7 @@
#include <asm/atomic.h>
-@@ -119,6 +121,8 @@ struct file *get_empty_filp(void)
+@@ -124,6 +126,8 @@ struct file *get_empty_filp(void)
f->f_gid = tsk->fsgid;
eventpoll_init_file(f);
/* f->f_version: 0 */
@@ -4731,7 +4126,7 @@
return f;
over:
-@@ -174,6 +178,8 @@ void fastcall __fput(struct file *file)
+@@ -239,6 +243,8 @@ void __fput(struct file *file)
if (file->f_mode & FMODE_WRITE)
put_write_access(inode);
put_pid(file->f_owner.pid);
@@ -4740,7 +4135,7 @@
file_kill(file);
file->f_path.dentry = NULL;
file->f_path.mnt = NULL;
-@@ -239,6 +245,8 @@ void put_filp(struct file *file)
+@@ -304,6 +310,8 @@ void put_filp(struct file *file)
{
if (atomic_dec_and_test(&file->f_count)) {
security_file_free(file);
@@ -4749,8 +4144,8 @@
file_kill(file);
file_free(file);
}
---- linux-2.6.22/fs/hfsplus/ioctl.c 2006-11-30 21:19:25 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/hfsplus/ioctl.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/hfsplus/ioctl.c 2008-04-17 10:37:23.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/hfsplus/ioctl.c 2008-04-19 15:14:52.000000000 -0400
@@ -16,6 +16,7 @@
#include <linux/fs.h>
#include <linux/sched.h>
@@ -4768,10 +4163,10 @@
+ (filp && MNT_IS_RDONLY(filp->f_vfsmnt)))
return -EROFS;
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
---- linux-2.6.22/fs/inode.c 2007-07-09 13:19:24 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/inode.c 2007-06-15 02:37:03 +0200
-@@ -115,6 +115,9 @@ static struct inode *alloc_inode(struct
+ if (!is_owner_or_cap(inode))
+--- linux-2.6.25.4/fs/inode.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/inode.c 2008-04-19 15:14:52.000000000 -0400
+@@ -124,6 +124,9 @@ static struct inode *alloc_inode(struct
struct address_space * const mapping = &inode->i_data;
inode->i_sb = sb;
@@ -4781,7 +4176,15 @@
inode->i_blkbits = sb->s_blocksize_bits;
inode->i_flags = 0;
atomic_set(&inode->i_count, 1);
-@@ -231,6 +234,8 @@ void __iget(struct inode * inode)
+@@ -142,6 +145,7 @@ static struct inode *alloc_inode(struct
+ inode->i_bdev = NULL;
+ inode->i_cdev = NULL;
+ inode->i_rdev = 0;
++ inode->i_mdev = 0;
+ inode->dirtied_when = 0;
+ if (security_inode_alloc(inode)) {
+ if (inode->i_sb->s_op->destroy_inode)
+@@ -246,6 +250,8 @@ void __iget(struct inode * inode)
inodes_stat.nr_unused--;
}
@@ -4790,125 +4193,60 @@
/**
* clear_inode - clear an inode
* @inode: inode to clear
---- linux-2.6.22/fs/ioctl.c 2007-07-09 13:19:24 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ioctl.c 2007-06-15 03:12:40 +0200
-@@ -13,10 +13,19 @@
+@@ -1434,9 +1440,11 @@ void init_special_inode(struct inode *in
+ if (S_ISCHR(mode)) {
+ inode->i_fop = &def_chr_fops;
+ inode->i_rdev = rdev;
++ inode->i_mdev = rdev;
+ } else if (S_ISBLK(mode)) {
+ inode->i_fop = &def_blk_fops;
+ inode->i_rdev = rdev;
++ inode->i_mdev = rdev;
+ } else if (S_ISFIFO(mode))
+ inode->i_fop = &def_fifo_fops;
+ else if (S_ISSOCK(mode))
+--- linux-2.6.25.4/fs/ioctl.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ioctl.c 2008-04-21 09:25:46.000000000 -0400
+@@ -13,6 +13,9 @@
#include <linux/security.h>
#include <linux/module.h>
- #include <linux/kallsyms.h>
+ #include <linux/uaccess.h>
+#include <linux/proc_fs.h>
+#include <linux/vserver/inode.h>
+#include <linux/vs_tag.h>
- #include <asm/uaccess.h>
#include <asm/ioctls.h>
-+
-+#ifdef CONFIG_VSERVER_LEGACY
-+extern int vx_proc_ioctl(struct inode *, struct file *,
-+ unsigned int, unsigned long);
-+#endif
-+
- static long do_ioctl(struct file *filp, unsigned int cmd,
- unsigned long arg)
- {
-@@ -152,6 +161,48 @@ int vfs_ioctl(struct file *filp, unsigne
- else
- error = -ENOTTY;
- break;
-+#ifdef CONFIG_VSERVER_LEGACY
-+#ifndef CONFIG_TAGGING_NONE
-+ case FIOC_GETTAG: {
-+ struct inode *inode = filp->f_dentry->d_inode;
-+
-+ /* fixme: if stealth, return -ENOTTY */
-+ error = -EPERM;
-+ if (capable(CAP_CONTEXT))
-+ error = put_user(inode->i_tag, (int __user *) arg);
-+ break;
-+ }
-+ case FIOC_SETTAG: {
-+ struct inode *inode = filp->f_dentry->d_inode;
-+ int tag;
-+
-+ /* fixme: if stealth, return -ENOTTY */
-+ error = -EPERM;
-+ if (!capable(CAP_CONTEXT))
-+ break;
-+ error = -EROFS;
-+ if (IS_RDONLY(inode))
-+ break;
-+ error = -ENOSYS;
-+ if (!(inode->i_sb->s_flags & MS_TAGGED))
-+ break;
-+ error = -EFAULT;
-+ if (get_user(tag, (int __user *) arg))
-+ break;
-+ error = 0;
-+ inode->i_tag = (tag & 0xFFFF);
-+ inode->i_ctime = CURRENT_TIME;
-+ mark_inode_dirty(inode);
-+ break;
-+ }
-+#endif
-+ case FIOC_GETXFLG:
-+ case FIOC_SETXFLG:
-+ error = -ENOTTY;
-+ if (filp->f_dentry->d_inode->i_sb->s_magic == PROC_SUPER_MAGIC)
-+ error = vx_proc_ioctl(filp->f_dentry->d_inode, filp, cmd, arg);
-+ break;
-+#endif
- default:
- if (S_ISREG(filp->f_path.dentry->d_inode->i_mode))
- error = file_ioctl(filp, cmd, arg);
---- linux-2.6.22/fs/ioprio.c 2007-05-02 19:25:18 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ioprio.c 2007-06-15 02:37:03 +0200
-@@ -25,6 +25,7 @@
- #include <linux/capability.h>
+--- linux-2.6.25.4/fs/ioprio.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ioprio.c 2008-04-19 15:14:52.000000000 -0400
+@@ -26,6 +26,7 @@
#include <linux/syscalls.h>
#include <linux/security.h>
+ #include <linux/pid_namespace.h>
+#include <linux/vs_base.h>
static int set_task_ioprio(struct task_struct *task, int ioprio)
{
-@@ -103,6 +104,8 @@ asmlinkage long sys_ioprio_set(int which
+@@ -116,6 +117,8 @@ asmlinkage long sys_ioprio_set(int which
else
- pgrp = find_pid(who);
+ pgrp = find_vpid(who);
do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
+ if (!vx_check(p->xid, VS_ADMIN_P | VS_IDENT))
+ continue;
ret = set_task_ioprio(p, ioprio);
if (ret)
break;
-@@ -112,7 +115,7 @@ asmlinkage long sys_ioprio_set(int which
- if (!who)
- user = current->user;
+@@ -205,6 +208,8 @@ asmlinkage long sys_ioprio_get(int which
else
-- user = find_user(who);
-+ user = find_user(vx_current_xid(), who);
-
- if (!user)
- break;
-@@ -190,6 +193,8 @@ asmlinkage long sys_ioprio_get(int which
- else
- pgrp = find_pid(who);
+ pgrp = find_vpid(who);
do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
+ if (!vx_check(p->xid, VS_ADMIN_P | VS_IDENT))
+ continue;
tmpio = get_task_ioprio(p);
if (tmpio < 0)
continue;
-@@ -203,7 +208,7 @@ asmlinkage long sys_ioprio_get(int which
- if (!who)
- user = current->user;
- else
-- user = find_user(who);
-+ user = find_user(vx_current_xid(), who);
-
- if (!user)
- break;
---- linux-2.6.22/fs/jfs/acl.c 2006-11-30 21:19:25 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/acl.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/acl.c 2007-02-04 13:44:54.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/acl.c 2008-04-19 15:14:52.000000000 -0400
@@ -232,7 +232,8 @@ int jfs_setattr(struct dentry *dentry, s
return rc;
@@ -4919,8 +4257,8 @@
if (DQUOT_TRANSFER(inode, iattr))
return -EDQUOT;
}
---- linux-2.6.22/fs/jfs/file.c 2007-05-02 19:25:18 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/file.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/file.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/file.c 2008-04-19 15:14:52.000000000 -0400
@@ -98,6 +98,7 @@ const struct inode_operations jfs_file_i
.setattr = jfs_setattr,
.permission = jfs_permission,
@@ -4929,16 +4267,8 @@
};
const struct file_operations jfs_file_operations = {
-@@ -109,6 +110,7 @@ const struct file_operations jfs_file_op
- .aio_write = generic_file_aio_write,
- .mmap = generic_file_mmap,
- .sendfile = generic_file_sendfile,
-+ .sendpage = generic_file_sendpage,
- .splice_read = generic_file_splice_read,
- .splice_write = generic_file_splice_write,
- .fsync = jfs_fsync,
---- linux-2.6.22/fs/jfs/inode.c 2007-07-09 13:19:26 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/inode.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/inode.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/inode.c 2008-04-19 15:14:52.000000000 -0400
@@ -22,6 +22,7 @@
#include <linux/buffer_head.h>
#include <linux/pagemap.h>
@@ -4947,7 +4277,7 @@
#include "jfs_incore.h"
#include "jfs_inode.h"
#include "jfs_filsys.h"
-@@ -143,6 +144,7 @@ void jfs_delete_inode(struct inode *inod
+@@ -155,6 +156,7 @@ void jfs_delete_inode(struct inode *inod
DQUOT_INIT(inode);
DQUOT_FREE_INODE(inode);
DQUOT_DROP(inode);
@@ -4955,8 +4285,8 @@
}
clear_inode(inode);
---- linux-2.6.22/fs/jfs/ioctl.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/ioctl.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/ioctl.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/ioctl.c 2008-04-21 09:25:22.000000000 -0400
@@ -10,6 +10,7 @@
#include <linux/capability.h>
#include <linux/time.h>
@@ -4965,7 +4295,7 @@
#include <asm/current.h>
#include <asm/uaccess.h>
-@@ -66,7 +67,8 @@ int jfs_ioctl(struct inode * inode, stru
+@@ -66,7 +67,8 @@ long jfs_ioctl(struct file *filp, unsign
case JFS_IOC_SETFLAGS: {
unsigned int oldflags;
@@ -4974,8 +4304,8 @@
+ (filp && MNT_IS_RDONLY(filp->f_vfsmnt)))
return -EROFS;
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
-@@ -87,8 +89,8 @@ int jfs_ioctl(struct inode * inode, stru
+ if (!is_owner_or_cap(inode))
+@@ -94,8 +96,8 @@ long jfs_ioctl(struct file *filp, unsign
* the relevant capability.
*/
if ((oldflags & JFS_IMMUTABLE_FL) ||
@@ -4983,11 +4313,11 @@
- (JFS_APPEND_FL | JFS_IMMUTABLE_FL))) {
+ ((flags ^ oldflags) & (JFS_APPEND_FL |
+ JFS_IMMUTABLE_FL | JFS_IUNLINK_FL))) {
- if (!capable(CAP_LINUX_IMMUTABLE))
+ if (!capable(CAP_LINUX_IMMUTABLE)) {
+ mutex_unlock(&inode->i_mutex);
return -EPERM;
- }
---- linux-2.6.22/fs/jfs/jfs_dinode.h 2006-11-30 21:19:25 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/jfs_dinode.h 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/jfs_dinode.h 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/jfs_dinode.h 2008-04-19 15:14:52.000000000 -0400
@@ -162,9 +162,12 @@ struct dinode {
#define JFS_APPEND_FL 0x01000000 /* writes to file may only append */
#define JFS_IMMUTABLE_FL 0x02000000 /* Immutable file */
@@ -5003,8 +4333,8 @@
/* These are identical to EXT[23]_IOC_GETFLAGS/SETFLAGS */
#define JFS_IOC_GETFLAGS _IOR('f', 1, long)
---- linux-2.6.22/fs/jfs/jfs_dtree.c 2007-02-06 03:01:24 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/jfs_dtree.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/jfs_dtree.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/jfs_dtree.c 2008-04-19 15:14:52.000000000 -0400
@@ -102,6 +102,7 @@
#include <linux/fs.h>
@@ -5028,7 +4358,7 @@
/*
* Save the table, we're going to overwrite it with the
-@@ -479,6 +480,12 @@ static u32 add_index(tid_t tid, struct i
+@@ -480,6 +481,12 @@ static u32 add_index(tid_t tid, struct i
return index;
@@ -5041,7 +4371,7 @@
clean_up:
jfs_ip->next_index--;
-@@ -952,6 +959,7 @@ static int dtSplitUp(tid_t tid,
+@@ -951,6 +958,7 @@ static int dtSplitUp(tid_t tid,
struct tlock *tlck;
struct lv *lv;
int quota_allocation = 0;
@@ -5049,7 +4379,7 @@
/* get split page */
smp = split->mp;
-@@ -1036,6 +1044,12 @@ static int dtSplitUp(tid_t tid,
+@@ -1033,6 +1041,12 @@ static int dtSplitUp(tid_t tid,
}
quota_allocation += n;
@@ -5062,7 +4392,7 @@
if ((rc = dbReAlloc(sbi->ipbmap, xaddr, (s64) xlen,
(s64) n, &nxaddr)))
goto extendOut;
-@@ -1309,6 +1323,9 @@ static int dtSplitUp(tid_t tid,
+@@ -1306,6 +1320,9 @@ static int dtSplitUp(tid_t tid,
freeKeyName:
kfree(key.name);
@@ -5072,7 +4402,7 @@
/* Rollback quota allocation */
if (rc && quota_allocation)
DQUOT_FREE_BLOCK(ip, quota_allocation);
-@@ -1376,6 +1393,12 @@ static int dtSplitPage(tid_t tid, struct
+@@ -1373,6 +1390,12 @@ static int dtSplitPage(tid_t tid, struct
release_metapage(rmp);
return -EDQUOT;
}
@@ -5085,7 +4415,7 @@
jfs_info("dtSplitPage: ip:0x%p smp:0x%p rmp:0x%p", ip, smp, rmp);
-@@ -1926,6 +1949,12 @@ static int dtSplitRoot(tid_t tid,
+@@ -1920,6 +1943,12 @@ static int dtSplitRoot(tid_t tid,
release_metapage(rmp);
return -EDQUOT;
}
@@ -5098,7 +4428,7 @@
BT_MARK_DIRTY(rmp, ip);
/*
-@@ -2292,6 +2321,8 @@ static int dtDeleteUp(tid_t tid, struct
+@@ -2286,6 +2315,8 @@ static int dtDeleteUp(tid_t tid, struct
xlen = lengthPXD(&fp->header.self);
@@ -5107,7 +4437,7 @@
/* Free quota allocation. */
DQUOT_FREE_BLOCK(ip, xlen);
-@@ -2368,6 +2399,8 @@ static int dtDeleteUp(tid_t tid, struct
+@@ -2362,6 +2393,8 @@ static int dtDeleteUp(tid_t tid, struct
xlen = lengthPXD(&p->header.self);
@@ -5116,8 +4446,8 @@
/* Free quota allocation */
DQUOT_FREE_BLOCK(ip, xlen);
---- linux-2.6.22/fs/jfs/jfs_extent.c 2006-11-30 21:19:25 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/jfs_extent.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/jfs_extent.c 2008-04-17 10:37:23.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/jfs_extent.c 2008-04-19 15:14:52.000000000 -0400
@@ -18,6 +18,7 @@
#include <linux/fs.h>
@@ -5179,9 +4509,9 @@
DQUOT_FREE_BLOCK(ip, nxlen);
goto exit;
}
---- linux-2.6.22/fs/jfs/jfs_filsys.h 2007-02-06 03:01:24 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/jfs_filsys.h 2007-06-15 02:37:03 +0200
-@@ -264,6 +264,7 @@
+--- linux-2.6.25.4/fs/jfs/jfs_filsys.h 2008-04-17 10:37:23.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/jfs_filsys.h 2008-04-19 15:14:52.000000000 -0400
+@@ -263,6 +263,7 @@
#define JFS_NAME_MAX 255
#define JFS_PATH_MAX BPSIZE
@@ -5189,8 +4519,8 @@
/*
* file system state (superblock state)
---- linux-2.6.22/fs/jfs/jfs_imap.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/jfs_imap.c 2007-06-15 03:01:50 +0200
+--- linux-2.6.25.4/fs/jfs/jfs_imap.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/jfs_imap.c 2008-04-19 15:14:52.000000000 -0400
@@ -45,6 +45,7 @@
#include <linux/buffer_head.h>
#include <linux/pagemap.h>
@@ -5199,7 +4529,7 @@
#include "jfs_incore.h"
#include "jfs_inode.h"
-@@ -3075,6 +3076,8 @@ static int copy_from_dinode(struct dinod
+@@ -3061,6 +3062,8 @@ static int copy_from_dinode(struct dinod
{
struct jfs_inode_info *jfs_ip = JFS_IP(ip);
struct jfs_sb_info *sbi = JFS_SBI(ip->i_sb);
@@ -5208,7 +4538,7 @@
jfs_ip->fileset = le32_to_cpu(dip->di_fileset);
jfs_ip->mode2 = le32_to_cpu(dip->di_mode);
-@@ -3095,14 +3098,18 @@ static int copy_from_dinode(struct dinod
+@@ -3081,14 +3084,18 @@ static int copy_from_dinode(struct dinod
}
ip->i_nlink = le32_to_cpu(dip->di_nlink);
@@ -5229,7 +4559,7 @@
if (sbi->gid == -1)
ip->i_gid = jfs_ip->saved_gid;
else {
-@@ -3167,14 +3174,12 @@ static void copy_to_dinode(struct dinode
+@@ -3153,14 +3160,12 @@ static void copy_to_dinode(struct dinode
dip->di_size = cpu_to_le64(ip->i_size);
dip->di_nblocks = cpu_to_le64(PBLK2LBLK(ip->i_sb, ip->i_blocks));
dip->di_nlink = cpu_to_le32(ip->i_nlink);
@@ -5250,8 +4580,8 @@
jfs_get_inode_flags(jfs_ip);
/*
* mode2 is only needed for storing the higher order bits.
---- linux-2.6.22/fs/jfs/jfs_inode.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/jfs_inode.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/jfs_inode.c 2008-04-17 10:33:02.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/jfs_inode.c 2008-04-19 15:14:52.000000000 -0400
@@ -18,6 +18,8 @@
#include <linux/fs.h>
@@ -5320,7 +4650,7 @@
+ inode->i_tag = dx_current_fstag(sb);
+ if (DLIMIT_ALLOC_INODE(inode)) {
+ iput(inode);
-+ return NULL;
++ return ERR_PTR(-ENOSPC);
+ }
+
/*
@@ -5331,18 +4661,18 @@
DQUOT_DROP(inode);
inode->i_flags |= S_NOQUOTA;
inode->i_nlink = 0;
---- linux-2.6.22/fs/jfs/jfs_inode.h 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/jfs_inode.h 2007-06-15 03:02:28 +0200
-@@ -33,6 +33,7 @@ extern void jfs_free_zero_link(struct in
- extern struct dentry *jfs_get_parent(struct dentry *dentry);
- extern void jfs_get_inode_flags(struct jfs_inode_info *);
+--- linux-2.6.25.4/fs/jfs/jfs_inode.h 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/jfs_inode.h 2008-04-19 15:14:52.000000000 -0400
+@@ -39,6 +39,7 @@ extern struct dentry *jfs_fh_to_dentry(s
+ extern struct dentry *jfs_fh_to_parent(struct super_block *sb, struct fid *fid,
+ int fh_len, int fh_type);
extern void jfs_set_inode_flags(struct inode *);
+extern int jfs_sync_flags(struct inode *);
extern int jfs_get_block(struct inode *, sector_t, struct buffer_head *, int);
extern const struct address_space_operations jfs_aops;
---- linux-2.6.22/fs/jfs/jfs_xtree.c 2007-05-02 19:25:18 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/jfs_xtree.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/jfs_xtree.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/jfs_xtree.c 2008-04-19 15:14:52.000000000 -0400
@@ -21,6 +21,7 @@
#include <linux/fs.h>
@@ -5404,7 +4734,7 @@
/* Rollback quota allocation. */
if (quota_allocation)
DQUOT_FREE_BLOCK(ip, quota_allocation);
-@@ -1520,6 +1538,12 @@ xtSplitRoot(tid_t tid,
+@@ -1519,6 +1537,12 @@ xtSplitRoot(tid_t tid,
release_metapage(rmp);
return -EDQUOT;
}
@@ -5417,7 +4747,7 @@
jfs_info("xtSplitRoot: ip:0x%p rmp:0x%p", ip, rmp);
-@@ -3951,6 +3975,8 @@ s64 xtTruncate(tid_t tid, struct inode *
+@@ -3948,6 +3972,8 @@ s64 xtTruncate(tid_t tid, struct inode *
else
ip->i_size = newsize;
@@ -5426,25 +4756,25 @@
/* update quota allocation to reflect freed blocks */
DQUOT_FREE_BLOCK(ip, nfreed);
---- linux-2.6.22/fs/jfs/namei.c 2007-05-02 19:25:18 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/namei.c 2007-06-15 02:37:03 +0200
-@@ -20,6 +20,7 @@
- #include <linux/fs.h>
+--- linux-2.6.25.4/fs/jfs/namei.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/namei.c 2008-04-19 15:14:52.000000000 -0400
+@@ -21,6 +21,7 @@
#include <linux/ctype.h>
#include <linux/quotaops.h>
+ #include <linux/exportfs.h>
+#include <linux/vs_tag.h>
#include "jfs_incore.h"
#include "jfs_superblock.h"
#include "jfs_inode.h"
-@@ -1469,6 +1470,7 @@ static struct dentry *jfs_lookup(struct
- return ERR_PTR(-EACCES);
+@@ -1468,6 +1469,7 @@ static struct dentry *jfs_lookup(struct
+ return ERR_CAST(ip);
}
+ dx_propagate_tag(nd, ip);
dentry = d_splice_alias(ip, dentry);
if (dentry && (JFS_SBI(dip->i_sb)->mntflag & JFS_OS2))
-@@ -1521,6 +1523,7 @@ const struct inode_operations jfs_dir_in
+@@ -1550,6 +1552,7 @@ const struct inode_operations jfs_dir_in
.setattr = jfs_setattr,
.permission = jfs_permission,
#endif
@@ -5452,9 +4782,9 @@
};
const struct file_operations jfs_dir_operations = {
---- linux-2.6.22/fs/jfs/super.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/super.c 2007-06-15 02:37:03 +0200
-@@ -194,7 +194,8 @@ static void jfs_put_super(struct super_b
+--- linux-2.6.25.4/fs/jfs/super.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/super.c 2008-04-19 15:14:52.000000000 -0400
+@@ -195,7 +195,8 @@ static void jfs_put_super(struct super_b
enum {
Opt_integrity, Opt_nointegrity, Opt_iocharset, Opt_resize,
Opt_resize_nosize, Opt_errors, Opt_ignore, Opt_err, Opt_quota,
@@ -5464,7 +4794,7 @@
};
static match_table_t tokens = {
-@@ -204,6 +205,10 @@ static match_table_t tokens = {
+@@ -205,6 +206,10 @@ static match_table_t tokens = {
{Opt_resize, "resize=%u"},
{Opt_resize_nosize, "resize"},
{Opt_errors, "errors=%s"},
@@ -5475,7 +4805,7 @@
{Opt_ignore, "noquota"},
{Opt_ignore, "quota"},
{Opt_usrquota, "usrquota"},
-@@ -338,6 +343,20 @@ static int parse_options(char *options,
+@@ -339,6 +344,20 @@ static int parse_options(char *options,
}
break;
}
@@ -5496,7 +4826,7 @@
default:
printk("jfs: Unrecognized mount option \"%s\" "
" or missing value\n", p);
-@@ -368,6 +387,13 @@ static int jfs_remount(struct super_bloc
+@@ -369,6 +388,13 @@ static int jfs_remount(struct super_bloc
if (!parse_options(data, sb, &newLVSize, &flag)) {
return -EINVAL;
}
@@ -5510,7 +4840,7 @@
if (newLVSize) {
if (sb->s_flags & MS_RDONLY) {
printk(KERN_ERR
-@@ -439,6 +465,9 @@ static int jfs_fill_super(struct super_b
+@@ -440,6 +466,9 @@ static int jfs_fill_super(struct super_b
#ifdef CONFIG_JFS_POSIX_ACL
sb->s_flags |= MS_POSIXACL;
#endif
@@ -5520,8 +4850,8 @@
if (newLVSize) {
printk(KERN_ERR "resize option for remount only\n");
---- linux-2.6.22/fs/jfs/xattr.c 2006-11-30 21:19:26 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/jfs/xattr.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/jfs/xattr.c 2008-04-17 10:37:23.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/jfs/xattr.c 2008-04-19 15:14:52.000000000 -0400
@@ -23,6 +23,7 @@
#include <linux/posix_acl_xattr.h>
#include <linux/quotaops.h>
@@ -5577,7 +4907,7 @@
rc = dbAlloc(inode, INOHINT(inode), (s64) blocks_needed,
&blkno);
if (rc)
-@@ -599,6 +616,9 @@ static int ea_get(struct inode *inode, s
+@@ -600,6 +617,9 @@ static int ea_get(struct inode *inode, s
return ea_size;
clean_up:
@@ -5587,7 +4917,7 @@
/* Rollback quota allocation */
if (quota_allocation)
DQUOT_FREE_BLOCK(inode, quota_allocation);
-@@ -675,8 +695,10 @@ static int ea_put(tid_t tid, struct inod
+@@ -676,8 +696,10 @@ static int ea_put(tid_t tid, struct inod
}
/* If old blocks exist, they must be removed from quota allocation. */
@@ -5599,9 +4929,9 @@
inode->i_ctime = CURRENT_TIME;
---- linux-2.6.22/fs/libfs.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/libfs.c 2007-06-15 02:37:03 +0200
-@@ -124,7 +124,8 @@ static inline unsigned char dt_type(stru
+--- linux-2.6.25.4/fs/libfs.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/libfs.c 2008-04-19 15:14:52.000000000 -0400
+@@ -125,7 +125,8 @@ static inline unsigned char dt_type(stru
* both impossible due to the lock on directory.
*/
@@ -5611,7 +4941,7 @@
{
struct dentry *dentry = filp->f_path.dentry;
struct dentry *cursor = filp->private_data;
-@@ -157,6 +158,8 @@ int dcache_readdir(struct file * filp, v
+@@ -158,6 +159,8 @@ int dcache_readdir(struct file * filp, v
next = list_entry(p, struct dentry, d_u.d_child);
if (d_unhashed(next) || !next->d_inode)
continue;
@@ -5620,7 +4950,7 @@
spin_unlock(&dcache_lock);
if (filldir(dirent, next->d_name.name,
-@@ -175,6 +178,18 @@ int dcache_readdir(struct file * filp, v
+@@ -176,6 +179,18 @@ int dcache_readdir(struct file * filp, v
return 0;
}
@@ -5639,26 +4969,26 @@
ssize_t generic_read_dir(struct file *filp, char __user *buf, size_t siz, loff_t *ppos)
{
return -EISDIR;
-@@ -640,6 +655,7 @@ EXPORT_SYMBOL(dcache_dir_close);
+@@ -778,6 +793,7 @@ EXPORT_SYMBOL(dcache_dir_close);
EXPORT_SYMBOL(dcache_dir_lseek);
EXPORT_SYMBOL(dcache_dir_open);
EXPORT_SYMBOL(dcache_readdir);
+EXPORT_SYMBOL(dcache_readdir_filter);
EXPORT_SYMBOL(generic_read_dir);
EXPORT_SYMBOL(get_sb_pseudo);
- EXPORT_SYMBOL(simple_commit_write);
---- linux-2.6.22/fs/locks.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/locks.c 2007-06-15 03:16:26 +0200
-@@ -125,6 +125,8 @@
- #include <linux/syscalls.h>
+ EXPORT_SYMBOL(simple_write_begin);
+--- linux-2.6.25.4/fs/locks.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/locks.c 2008-05-21 14:30:41.000000000 -0400
+@@ -126,6 +126,8 @@
#include <linux/time.h>
#include <linux/rcupdate.h>
+ #include <linux/pid_namespace.h>
+#include <linux/vs_base.h>
+#include <linux/vs_limit.h>
#include <asm/semaphore.h>
#include <asm/uaccess.h>
-@@ -147,6 +149,8 @@ static struct kmem_cache *filelock_cache
+@@ -148,6 +150,8 @@ static struct kmem_cache *filelock_cache
/* Allocate an empty lock structure. */
static struct file_lock *locks_alloc_lock(void)
{
@@ -5667,7 +4997,7 @@
return kmem_cache_alloc(filelock_cache, GFP_KERNEL);
}
-@@ -172,6 +176,7 @@ static void locks_free_lock(struct file_
+@@ -173,6 +177,7 @@ static void locks_free_lock(struct file_
BUG_ON(!list_empty(&fl->fl_block));
BUG_ON(!list_empty(&fl->fl_link));
@@ -5675,7 +5005,7 @@
locks_release_private(fl);
kmem_cache_free(filelock_cache, fl);
}
-@@ -191,6 +196,7 @@ void locks_init_lock(struct file_lock *f
+@@ -193,6 +198,7 @@ void locks_init_lock(struct file_lock *f
fl->fl_start = fl->fl_end = 0;
fl->fl_ops = NULL;
fl->fl_lmops = NULL;
@@ -5683,7 +5013,7 @@
}
EXPORT_SYMBOL(locks_init_lock);
-@@ -244,6 +250,7 @@ void locks_copy_lock(struct file_lock *n
+@@ -246,6 +252,7 @@ void locks_copy_lock(struct file_lock *n
new->fl_file = fl->fl_file;
new->fl_ops = fl->fl_ops;
new->fl_lmops = fl->fl_lmops;
@@ -5691,7 +5021,7 @@
locks_copy_private(new, fl);
}
-@@ -282,6 +289,11 @@ static int flock_make_lock(struct file *
+@@ -284,6 +291,11 @@ static int flock_make_lock(struct file *
fl->fl_flags = FL_FLOCK;
fl->fl_type = type;
fl->fl_end = OFFSET_MAX;
@@ -5703,7 +5033,7 @@
*lock = fl;
return 0;
-@@ -447,6 +459,7 @@ static int lease_init(struct file *filp,
+@@ -449,6 +461,7 @@ static int lease_init(struct file *filp,
fl->fl_owner = current->files;
fl->fl_pid = current->tgid;
@@ -5711,9 +5041,9 @@
fl->fl_file = filp;
fl->fl_flags = FL_LEASE;
-@@ -466,6 +479,11 @@ static int lease_alloc(struct file *filp
+@@ -468,6 +481,11 @@ static struct file_lock *lease_alloc(str
if (fl == NULL)
- goto out;
+ return ERR_PTR(error);
+ fl->fl_xid = vx_current_xid();
+ if (filp)
@@ -5723,7 +5053,7 @@
error = lease_init(filp, type, fl);
if (error) {
locks_free_lock(fl);
-@@ -769,6 +787,7 @@ static int flock_lock_file(struct file *
+@@ -774,6 +792,7 @@ static int flock_lock_file(struct file *
if (found)
cond_resched();
@@ -5731,15 +5061,15 @@
find_conflict:
for_each_lock(inode, before) {
struct file_lock *fl = *before;
-@@ -787,6 +806,7 @@ find_conflict:
+@@ -792,6 +811,7 @@ find_conflict:
goto out;
locks_copy_lock(new_fl, request);
- locks_insert_lock(&inode->i_flock, new_fl);
+ locks_insert_lock(before, new_fl);
+ vx_locks_inc(new_fl);
new_fl = NULL;
error = 0;
-@@ -797,7 +817,8 @@ out:
+@@ -802,7 +822,8 @@ out:
return error;
}
@@ -5749,7 +5079,7 @@
{
struct file_lock *fl;
struct file_lock *new_fl = NULL;
-@@ -807,6 +828,8 @@ static int __posix_lock_file(struct inod
+@@ -812,6 +833,8 @@ static int __posix_lock_file(struct inod
struct file_lock **before;
int error, added = 0;
@@ -5758,7 +5088,7 @@
/*
* We may need two file_lock structures for this operation,
* so we get them in advance to avoid races.
-@@ -817,7 +840,11 @@ static int __posix_lock_file(struct inod
+@@ -822,7 +845,11 @@ static int __posix_lock_file(struct inod
(request->fl_type != F_UNLCK ||
request->fl_start != 0 || request->fl_end != OFFSET_MAX)) {
new_fl = locks_alloc_lock();
@@ -5770,7 +5100,7 @@
}
lock_kernel();
-@@ -1016,7 +1043,8 @@ static int __posix_lock_file(struct inod
+@@ -1021,7 +1048,8 @@ static int __posix_lock_file(struct inod
int posix_lock_file(struct file *filp, struct file_lock *fl,
struct file_lock *conflock)
{
@@ -5780,7 +5110,7 @@
}
EXPORT_SYMBOL(posix_lock_file);
-@@ -1106,7 +1134,7 @@ int locks_mandatory_area(int read_write,
+@@ -1111,7 +1139,7 @@ int locks_mandatory_area(int read_write,
fl.fl_end = offset + count - 1;
for (;;) {
@@ -5789,17 +5119,15 @@
if (error != -EAGAIN)
break;
if (!(fl.fl_flags & FL_SLEEP))
-@@ -1410,8 +1438,8 @@ static int __setlease(struct file *filp,
- goto out;
+@@ -1425,6 +1453,7 @@ int generic_setlease(struct file *filp,
- locks_copy_lock(fl, lease);
--
- locks_insert_lock(before, fl);
-+ vx_locks_inc(fl);
+ locks_copy_lock(new_fl, lease);
+ locks_insert_lock(before, new_fl);
++ vx_locks_inc(new_fl);
- *flp = fl;
- error = 0;
-@@ -1738,6 +1766,11 @@ int fcntl_setlk(unsigned int fd, struct
+ *flp = new_fl;
+ return 0;
+@@ -1756,6 +1785,11 @@ int fcntl_setlk(unsigned int fd, struct
if (file_lock == NULL)
return -ENOLCK;
@@ -5811,7 +5139,7 @@
/*
* This might block, so we do it before checking the inode.
*/
-@@ -1864,6 +1897,11 @@ int fcntl_setlk64(unsigned int fd, struc
+@@ -1893,6 +1927,11 @@ int fcntl_setlk64(unsigned int fd, struc
if (file_lock == NULL)
return -ENOLCK;
@@ -5823,40 +5151,43 @@
/*
* This might block, so we do it before checking the inode.
*/
-@@ -2168,6 +2206,10 @@ int get_locks_status(char *buffer, char
- list_for_each(tmp, &file_lock_list) {
- struct list_head *btmp;
- struct file_lock *fl = list_entry(tmp, struct file_lock, fl_link);
-+
+@@ -2176,8 +2215,11 @@ static int locks_show(struct seq_file *f
+
+ lock_get_status(f, fl, (long)f->private, "");
+
+- list_for_each_entry(bfl, &fl->fl_block, fl_block)
++ list_for_each_entry(bfl, &fl->fl_block, fl_block) {
+ if (!vx_check(fl->fl_xid, VS_WATCH_P | VS_IDENT))
+ continue;
-+
- lock_get_status(q, fl, ++i, "");
- move_lock_status(&q, &pos, offset);
+ lock_get_status(f, bfl, (long)f->private, " ->");
++ }
---- linux-2.6.22/fs/namei.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/namei.c 2007-06-15 02:37:03 +0200
-@@ -31,6 +31,11 @@
+ f->private++;
+ return 0;
+--- linux-2.6.25.4/fs/namei.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/namei.c 2008-04-22 19:23:22.000000000 -0400
+@@ -30,6 +30,13 @@
+ #include <linux/capability.h>
#include <linux/file.h>
#include <linux/fcntl.h>
- #include <linux/namei.h>
+#include <linux/proc_fs.h>
+#include <linux/vserver/inode.h>
+#include <linux/vs_base.h>
+#include <linux/vs_tag.h>
+#include <linux/vs_cowbl.h>
++#include <linux/vs_device.h>
++#include <linux/vs_context.h>
#include <asm/namei.h>
#include <asm/uaccess.h>
-@@ -224,6 +229,31 @@ int generic_permission(struct inode *ino
+@@ -225,6 +232,28 @@ int generic_permission(struct inode *ino
return -EACCES;
}
+static inline int dx_barrier(struct inode *inode)
+{
+ if (IS_BARRIER(inode) && !vx_check(0, VS_ADMIN)) {
-+ vxwprintk(1, "xid=%d did hit the barrier.",
-+ vx_current_xid());
++ vxwprintk_task(1, "did hit the barrier.");
+ return 1;
+ }
+ return 0;
@@ -5866,26 +5197,24 @@
+{
+ if (dx_barrier(inode))
+ return -EACCES;
-+ if (inode->i_tag == 0)
-+ return 0;
-+ if (dx_check(inode->i_tag, DX_ADMIN|DX_WATCH|DX_IDENT))
++ if (dx_notagcheck(nd) ||
++ dx_check(inode->i_tag, DX_HOSTID|DX_ADMIN|DX_WATCH|DX_IDENT))
+ return 0;
+
-+ vxwprintk(1, "xid=%d denied access to %p[#%d,%lu] »%s«.",
-+ vx_current_xid(), inode, inode->i_tag, inode->i_ino,
-+ vxd_cond_path(nd));
++ vxwprintk_task(1, "denied access to %p[#%d,%lu] »%s«.",
++ inode, inode->i_tag, inode->i_ino, vxd_cond_path(nd));
+ return -EACCES;
+}
+
int permission(struct inode *inode, int mask, struct nameidata *nd)
{
- umode_t mode = inode->i_mode;
-@@ -234,14 +264,14 @@ int permission(struct inode *inode, int
+ int retval, submask;
+@@ -239,14 +268,14 @@ int permission(struct inode *inode, int
/*
* Nobody gets write access to a read-only fs.
*/
- if (IS_RDONLY(inode) &&
-+ if ((IS_RDONLY(inode) || (nd && MNT_IS_RDONLY(nd->mnt))) &&
++ if ((IS_RDONLY(inode) || (nd && MNT_IS_RDONLY(nd->path.mnt))) &&
(S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)))
return -EROFS;
@@ -5897,16 +5226,19 @@
return -EACCES;
}
-@@ -257,6 +287,8 @@ int permission(struct inode *inode, int
+@@ -261,6 +290,11 @@ int permission(struct inode *inode, int
/* Ordinary permission routines do not understand MAY_APPEND. */
submask = mask & ~MAY_APPEND;
-+ if ((retval = dx_permission(inode, mask, nd)))
++ if ((inode->i_sb->s_magic != DEVPTS_SUPER_MAGIC) &&
++ (inode->i_sb->s_magic != PROC_SUPER_MAGIC) &&
++ (retval = dx_permission(inode, mask, nd)))
+ return retval;
- if (inode->i_op && inode->i_op->permission)
++
+ if (inode->i_op && inode->i_op->permission) {
retval = inode->i_op->permission(inode, submask, nd);
- else
-@@ -432,6 +464,8 @@ static int exec_permission_lite(struct i
+ if (!retval) {
+@@ -459,6 +493,8 @@ static int exec_permission_lite(struct i
{
umode_t mode = inode->i_mode;
@@ -5915,9 +5247,9 @@
if (inode->i_op && inode->i_op->permission)
return -EAGAIN;
-@@ -732,7 +766,8 @@ static __always_inline void follow_dotdo
- if (nd->dentry == fs->root &&
- nd->mnt == fs->rootmnt) {
+@@ -789,7 +825,8 @@ static __always_inline void follow_dotdo
+ if (nd->path.dentry == fs->root.dentry &&
+ nd->path.mnt == fs->root.mnt) {
read_unlock(&fs->lock);
- break;
+ /* for sane '/' avoid follow_mount() */
@@ -5925,10 +5257,10 @@
}
read_unlock(&fs->lock);
spin_lock(&dcache_lock);
-@@ -769,16 +804,34 @@ static int do_lookup(struct nameidata *n
+@@ -826,16 +863,39 @@ static int do_lookup(struct nameidata *n
{
- struct vfsmount *mnt = nd->mnt;
- struct dentry *dentry = __d_lookup(nd->dentry, name);
+ struct vfsmount *mnt = nd->path.mnt;
+ struct dentry *dentry = __d_lookup(nd->path.dentry, name);
+ struct inode *inode;
if (!dentry)
@@ -5938,29 +5270,34 @@
+ inode = dentry->d_inode;
+ if (!inode)
+ goto done;
++
+ if (inode->i_sb->s_magic == PROC_SUPER_MAGIC) {
+ struct proc_dir_entry *de = PDE(inode);
+
+ if (de && !vx_hide_check(0, de->vx_flags))
+ goto hidden;
++ } else if (inode->i_sb->s_magic == DEVPTS_SUPER_MAGIC) {
++ if (!vx_check((xid_t)inode->i_tag, VS_WATCH_P | VS_IDENT))
++ goto hidden;
++ } else {
++ if (!dx_notagcheck(nd) && !dx_check(inode->i_tag,
++ DX_WATCH | DX_ADMIN | DX_HOSTID | DX_IDENT))
++ goto hidden;
+ }
-+ if (!dx_check(inode->i_tag, DX_WATCH|DX_ADMIN|DX_HOSTID|DX_IDENT))
-+ goto hidden;
done:
path->mnt = mnt;
path->dentry = dentry;
__follow_mount(path);
return 0;
+hidden:
-+ vxwprintk(1, "xid=%d did lookup hidden %p[#%d,%lu] »%s«.",
-+ vx_current_xid(), inode, inode->i_tag, inode->i_ino,
-+ vxd_path(dentry, mnt));
++ vxwprintk_task(1, "did lookup hidden %p[#%d,%lu] »%s«.",
++ inode, inode->i_tag, inode->i_ino, vxd_path(path));
+ dput(dentry);
+ return -ENOENT;
need_lookup:
- dentry = real_lookup(nd->dentry, name, nd);
-@@ -1399,7 +1452,8 @@ static inline int check_sticky(struct in
+ dentry = real_lookup(nd->path.dentry, name, nd);
+@@ -1464,7 +1524,8 @@ static inline int check_sticky(struct in
* 10. We don't allow removal of NFS sillyrenamed files; it's handled by
* nfs_async_unlink().
*/
@@ -5970,9 +5307,9 @@
{
int error;
-@@ -1409,13 +1463,13 @@ static int may_delete(struct inode *dir,
+@@ -1474,13 +1535,13 @@ static int may_delete(struct inode *dir,
BUG_ON(victim->d_parent->d_inode != dir);
- audit_inode_child(victim->d_name.name, victim->d_inode, dir);
+ audit_inode_child(victim->d_name.name, victim, dir);
- error = permission(dir,MAY_WRITE | MAY_EXEC, NULL);
+ error = permission(dir,MAY_WRITE | MAY_EXEC, nd);
@@ -5986,9 +5323,9 @@
return -EPERM;
if (isdir) {
if (!S_ISDIR(victim->d_inode->i_mode))
-@@ -1546,6 +1600,14 @@ int may_open(struct nameidata *nd, int a
- if (S_ISDIR(inode->i_mode) && (flag & FMODE_WRITE))
- return -EISDIR;
+@@ -1626,6 +1687,14 @@ int may_open(struct nameidata *nd, int a
+ } else if (IS_RDONLY(inode) && (acc_mode & MAY_WRITE))
+ return -EROFS;
+#ifdef CONFIG_VSERVER_COWBL
+ if (IS_COW(inode) && (flag & FMODE_WRITE)) {
@@ -6001,17 +5338,7 @@
error = vfs_permission(nd, acc_mode);
if (error)
return error;
-@@ -1562,7 +1624,8 @@ int may_open(struct nameidata *nd, int a
- return -EACCES;
-
- flag &= ~O_TRUNC;
-- } else if (IS_RDONLY(inode) && (flag & FMODE_WRITE))
-+ } else if ((IS_RDONLY(inode) || MNT_IS_RDONLY(nd->mnt))
-+ && (flag & FMODE_WRITE))
- return -EROFS;
- /*
- * An append-only file must be opened in append mode for writing.
-@@ -1650,6 +1713,11 @@ int open_namei(int dfd, const char *path
+@@ -1717,6 +1786,11 @@ int open_namei(int dfd, const char *path
struct dentry *dir;
int count = 0;
@@ -6023,7 +5350,7 @@
acc_mode = ACC_MODE(flag);
/* O_TRUNC implies we need access checks for write permissions */
-@@ -1743,6 +1811,22 @@ do_last:
+@@ -1810,6 +1884,22 @@ do_last:
goto exit;
ok:
error = may_open(nd, acc_mode, flag);
@@ -6037,7 +5364,7 @@
+ }
+ dput(dentry);
+ release_open_intent(nd);
-+ path_release(nd);
++ path_put(&nd->path);
+ flag = rflag;
+ mode = rmode;
+ goto restart;
@@ -6046,7 +5373,7 @@
if (error)
goto exit;
return 0;
-@@ -1854,9 +1938,10 @@ fail:
+@@ -1921,16 +2011,25 @@ fail:
}
EXPORT_SYMBOL_GPL(lookup_create);
@@ -6059,23 +5386,39 @@
if (error)
return error;
-@@ -1906,11 +1991,12 @@ asmlinkage long sys_mknodat(int dfd, con
- error = vfs_create(nd.dentry->d_inode,dentry,mode,&nd);
+
+- if ((S_ISCHR(mode) || S_ISBLK(mode)) && !capable(CAP_MKNOD))
++ if (!(S_ISCHR(mode) || S_ISBLK(mode)))
++ goto okay;
++
++ if (!capable(CAP_MKNOD))
+ return -EPERM;
+
++ if (S_ISCHR(mode) && !vs_chrdev_perm(dev, DATTR_CREATE))
++ return -EPERM;
++ if (S_ISBLK(mode) && !vs_blkdev_perm(dev, DATTR_CREATE))
++ return -EPERM;
++okay:
+ if (!dir->i_op || !dir->i_op->mknod)
+ return -EPERM;
+
+@@ -1973,11 +2072,12 @@ asmlinkage long sys_mknodat(int dfd, con
+ error = vfs_create(nd.path.dentry->d_inode,dentry,mode,&nd);
break;
case S_IFCHR: case S_IFBLK:
-- error = vfs_mknod(nd.dentry->d_inode,dentry,mode,
+- error = vfs_mknod(nd.path.dentry->d_inode,dentry,mode,
- new_decode_dev(dev));
-+ error = vfs_mknod(nd.dentry->d_inode, dentry, mode,
++ error = vfs_mknod(nd.path.dentry->d_inode, dentry, mode,
+ new_decode_dev(dev), &nd);
break;
case S_IFIFO: case S_IFSOCK:
-- error = vfs_mknod(nd.dentry->d_inode,dentry,mode,0);
-+ error = vfs_mknod(nd.dentry->d_inode, dentry, mode,
+- error = vfs_mknod(nd.path.dentry->d_inode,dentry,mode,0);
++ error = vfs_mknod(nd.path.dentry->d_inode, dentry, mode,
+ 0, &nd);
break;
case S_IFDIR:
error = -EPERM;
-@@ -1933,9 +2019,10 @@ asmlinkage long sys_mknod(const char __u
+@@ -2000,9 +2100,10 @@ asmlinkage long sys_mknod(const char __u
return sys_mknodat(AT_FDCWD, filename, mode, dev);
}
@@ -6088,16 +5431,16 @@
if (error)
return error;
-@@ -1977,7 +2064,7 @@ asmlinkage long sys_mkdirat(int dfd, con
+@@ -2044,7 +2145,7 @@ asmlinkage long sys_mkdirat(int dfd, con
- if (!IS_POSIXACL(nd.dentry->d_inode))
+ if (!IS_POSIXACL(nd.path.dentry->d_inode))
mode &= ~current->fs->umask;
-- error = vfs_mkdir(nd.dentry->d_inode, dentry, mode);
-+ error = vfs_mkdir(nd.dentry->d_inode, dentry, mode, &nd);
+- error = vfs_mkdir(nd.path.dentry->d_inode, dentry, mode);
++ error = vfs_mkdir(nd.path.dentry->d_inode, dentry, mode, &nd);
dput(dentry);
out_unlock:
- mutex_unlock(&nd.dentry->d_inode->i_mutex);
-@@ -2020,9 +2107,10 @@ void dentry_unhash(struct dentry *dentry
+ mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+@@ -2087,9 +2188,10 @@ void dentry_unhash(struct dentry *dentry
spin_unlock(&dcache_lock);
}
@@ -6110,16 +5453,16 @@
if (error)
return error;
-@@ -2084,7 +2172,7 @@ static long do_rmdir(int dfd, const char
+@@ -2151,7 +2253,7 @@ static long do_rmdir(int dfd, const char
error = PTR_ERR(dentry);
if (IS_ERR(dentry))
goto exit2;
-- error = vfs_rmdir(nd.dentry->d_inode, dentry);
-+ error = vfs_rmdir(nd.dentry->d_inode, dentry, &nd);
+- error = vfs_rmdir(nd.path.dentry->d_inode, dentry);
++ error = vfs_rmdir(nd.path.dentry->d_inode, dentry, &nd);
dput(dentry);
exit2:
- mutex_unlock(&nd.dentry->d_inode->i_mutex);
-@@ -2100,9 +2188,10 @@ asmlinkage long sys_rmdir(const char __u
+ mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+@@ -2167,9 +2269,10 @@ asmlinkage long sys_rmdir(const char __u
return do_rmdir(AT_FDCWD, pathname);
}
@@ -6132,16 +5475,16 @@
if (error)
return error;
-@@ -2164,7 +2253,7 @@ static long do_unlinkat(int dfd, const c
+@@ -2232,7 +2335,7 @@ static long do_unlinkat(int dfd, const c
inode = dentry->d_inode;
if (inode)
atomic_inc(&inode->i_count);
-- error = vfs_unlink(nd.dentry->d_inode, dentry);
-+ error = vfs_unlink(nd.dentry->d_inode, dentry, &nd);
+- error = vfs_unlink(nd.path.dentry->d_inode, dentry);
++ error = vfs_unlink(nd.path.dentry->d_inode, dentry, &nd);
exit2:
dput(dentry);
}
-@@ -2199,9 +2288,10 @@ asmlinkage long sys_unlink(const char __
+@@ -2267,9 +2370,10 @@ asmlinkage long sys_unlink(const char __
return do_unlinkat(AT_FDCWD, pathname);
}
@@ -6154,16 +5497,17 @@
if (error)
return error;
-@@ -2245,7 +2335,7 @@ asmlinkage long sys_symlinkat(const char
+@@ -2313,7 +2417,8 @@ asmlinkage long sys_symlinkat(const char
if (IS_ERR(dentry))
goto out_unlock;
-- error = vfs_symlink(nd.dentry->d_inode, dentry, from, S_IALLUGO);
-+ error = vfs_symlink(nd.dentry->d_inode, dentry, from, S_IALLUGO, &nd);
+- error = vfs_symlink(nd.path.dentry->d_inode, dentry, from, S_IALLUGO);
++ error = vfs_symlink(nd.path.dentry->d_inode, dentry, from,
++ S_IALLUGO, &nd);
dput(dentry);
out_unlock:
- mutex_unlock(&nd.dentry->d_inode->i_mutex);
-@@ -2262,7 +2352,8 @@ asmlinkage long sys_symlink(const char _
+ mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+@@ -2330,7 +2435,8 @@ asmlinkage long sys_symlink(const char _
return sys_symlinkat(oldname, AT_FDCWD, newname);
}
@@ -6173,7 +5517,7 @@
{
struct inode *inode = old_dentry->d_inode;
int error;
-@@ -2270,7 +2361,7 @@ int vfs_link(struct dentry *old_dentry,
+@@ -2338,7 +2444,7 @@ int vfs_link(struct dentry *old_dentry,
if (!inode)
return -ENOENT;
@@ -6182,7 +5526,7 @@
if (error)
return error;
-@@ -2280,7 +2371,7 @@ int vfs_link(struct dentry *old_dentry,
+@@ -2348,7 +2454,7 @@ int vfs_link(struct dentry *old_dentry,
/*
* A link to an append-only or immutable file cannot be created.
*/
@@ -6191,16 +5535,17 @@
return -EPERM;
if (!dir->i_op || !dir->i_op->link)
return -EPERM;
-@@ -2340,7 +2431,7 @@ asmlinkage long sys_linkat(int olddfd, c
+@@ -2408,7 +2514,8 @@ asmlinkage long sys_linkat(int olddfd, c
error = PTR_ERR(new_dentry);
if (IS_ERR(new_dentry))
goto out_unlock;
-- error = vfs_link(old_nd.dentry, nd.dentry->d_inode, new_dentry);
-+ error = vfs_link(old_nd.dentry, nd.dentry->d_inode, new_dentry, &nd);
+- error = vfs_link(old_nd.path.dentry, nd.path.dentry->d_inode, new_dentry);
++ error = vfs_link(old_nd.path.dentry, nd.path.dentry->d_inode,
++ new_dentry, &nd);
dput(new_dentry);
out_unlock:
- mutex_unlock(&nd.dentry->d_inode->i_mutex);
-@@ -2472,14 +2563,14 @@ int vfs_rename(struct inode *old_dir, st
+ mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+@@ -2540,14 +2647,14 @@ int vfs_rename(struct inode *old_dir, st
if (old_dentry->d_inode == new_dentry->d_inode)
return 0;
@@ -6217,17 +5562,17 @@
if (error)
return error;
-@@ -2557,6 +2648,9 @@ static int do_rename(int olddfd, const c
+@@ -2625,6 +2732,9 @@ static int do_rename(int olddfd, const c
error = -EINVAL;
if (old_dentry == trap)
goto exit4;
+ error = -EROFS;
-+ if (MNT_IS_RDONLY(newnd.mnt))
++ if (MNT_IS_RDONLY(newnd.path.mnt))
+ goto exit4;
new_dentry = lookup_hash(&newnd);
error = PTR_ERR(new_dentry);
if (IS_ERR(new_dentry))
-@@ -2650,6 +2744,129 @@ int vfs_follow_link(struct nameidata *nd
+@@ -2718,6 +2828,214 @@ int vfs_follow_link(struct nameidata *nd
return __vfs_follow_link(nd, link);
}
@@ -6236,13 +5581,20 @@
+
+#include <linux/file.h>
+
++static inline
++long do_cow_splice(struct file *in, struct file *out, size_t len)
++{
++ loff_t ppos = 0;
++
++ return do_splice_direct(in, &ppos, out, len, 0);
++}
++
+struct dentry *cow_break_link(const char *pathname)
+{
-+ int ret, mode, pathlen;
++ int ret, mode, pathlen, redo = 0;
+ struct nameidata old_nd, dir_nd;
-+ struct dentry *old_dentry, *new_dentry;
-+ struct dentry *res = ERR_PTR(-EMLINK);
-+ struct vfsmount *old_mnt, *new_mnt;
++ struct path old_path, new_path;
++ struct dentry *dir, *res = NULL;
+ struct file *old_file;
+ struct file *new_file;
+ char *to, *path, pad='\251';
@@ -6250,85 +5602,136 @@
+
+ vxdprintk(VXD_CBIT(misc, 1), "cow_break_link(»%s«)", pathname);
+ path = kmalloc(PATH_MAX, GFP_KERNEL);
++ ret = -ENOMEM;
++ if (!path)
++ goto out;
+
++ /* old_nd will have refs to dentry and mnt */
+ ret = path_lookup(pathname, LOOKUP_FOLLOW, &old_nd);
+ vxdprintk(VXD_CBIT(misc, 2), "path_lookup(old): %d", ret);
-+ old_dentry = old_nd.dentry;
-+ old_mnt = old_nd.mnt;
-+ mode = old_dentry->d_inode->i_mode;
++ if (ret < 0)
++ goto out_free_path;
+
-+ to = d_path(old_dentry, old_mnt, path, PATH_MAX-2);
++ old_path = old_nd.path;
++ mode = old_path.dentry->d_inode->i_mode;
++
++ to = d_path(&old_path, path, PATH_MAX-2);
+ pathlen = strlen(to);
-+ vxdprintk(VXD_CBIT(misc, 2), "old path »%s«", to);
++ vxdprintk(VXD_CBIT(misc, 2), "old path »%s« [»%.*s«:%d]", to,
++ old_path.dentry->d_name.len, old_path.dentry->d_name.name,
++ old_path.dentry->d_name.len);
+
-+ to[pathlen+1] = 0;
++ to[pathlen + 1] = 0;
+retry:
+ to[pathlen] = pad--;
++ ret = -EMLINK;
+ if (pad <= '\240')
+ goto out_rel_old;
+
+ vxdprintk(VXD_CBIT(misc, 1), "temp copy »%s«", to);
++ /* dir_nd will have refs to dentry and mnt */
+ ret = path_lookup(to,
-+ LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE, &dir_nd);
++ LOOKUP_PARENT | LOOKUP_OPEN | LOOKUP_CREATE, &dir_nd);
++ vxdprintk(VXD_CBIT(misc, 2),
++ "path_lookup(new): %d", ret);
++ if (ret < 0)
++ goto retry;
+
-+ /* this puppy downs the inode sem */
-+ new_dentry = lookup_create(&dir_nd, 0);
++ /* this puppy downs the inode mutex */
++ new_path.dentry = lookup_create(&dir_nd, 0);
+ vxdprintk(VXD_CBIT(misc, 2),
-+ "lookup_create(new): %p", new_dentry);
-+ if (!new_dentry) {
-+ path_release(&dir_nd);
++ "lookup_create(new): %p [»%.*s«:%d]", new_path.dentry,
++ new_path.dentry->d_name.len, new_path.dentry->d_name.name,
++ new_path.dentry->d_name.len);
++ if (!new_path.dentry || IS_ERR(new_path.dentry)) {
++ path_put(&dir_nd.path);
+ goto retry;
+ }
++ dir = dir_nd.path.dentry;
+
-+ ret = vfs_create(dir_nd.dentry->d_inode, new_dentry, mode, &dir_nd);
++ ret = vfs_create(dir_nd.path.dentry->d_inode, new_path.dentry, mode, &dir_nd);
+ vxdprintk(VXD_CBIT(misc, 2),
+ "vfs_create(new): %d", ret);
+ if (ret == -EEXIST) {
-+ mutex_unlock(&dir_nd.dentry->d_inode->i_mutex);
-+ dput(new_dentry);
-+ path_release(&dir_nd);
++ mutex_unlock(&dir->d_inode->i_mutex);
++ dput(new_path.dentry);
++ path_put(&dir_nd.path);
+ goto retry;
+ }
-+ else if (ret < 0) {
-+ res = ERR_PTR(ret);
-+ goto out_rel_both;
-+ }
-+
-+ new_mnt = dir_nd.mnt;
++ else if (ret < 0)
++ goto out_unlock_new;
+
-+ dget(old_dentry);
-+ mntget(old_mnt);
-+ /* this one cleans up the dentry in case of failure */
-+ old_file = dentry_open(old_dentry, old_mnt, O_RDONLY);
++ /* drop out early, ret passes ENOENT */
++ ret = -ENOENT;
++ if ((redo = d_unhashed(old_path.dentry)))
++ goto out_unlock_new;
++
++ new_path.mnt = dir_nd.path.mnt;
++ dget(old_path.dentry);
++ mntget(old_path.mnt);
++ /* this one cleans up the dentry/mnt in case of failure */
++ old_file = dentry_open(old_path.dentry, old_path.mnt, O_RDONLY);
+ vxdprintk(VXD_CBIT(misc, 2),
+ "dentry_open(old): %p", old_file);
-+ if (!old_file)
-+ goto out_rel_both;
++ if (!old_file || IS_ERR(old_file)) {
++ res = IS_ERR(old_file) ? (void *) old_file : res;
++ goto out_unlock_new;
++ }
+
-+ dget(new_dentry);
-+ mntget(new_mnt);
-+ /* this one cleans up the dentry in case of failure */
-+ new_file = dentry_open(new_dentry, new_mnt, O_WRONLY);
++ dget(new_path.dentry);
++ mntget(new_path.mnt);
++ /* this one cleans up the dentry/mnt in case of failure */
++ new_file = dentry_open(new_path.dentry, new_path.mnt, O_WRONLY);
+ vxdprintk(VXD_CBIT(misc, 2),
+ "dentry_open(new): %p", new_file);
-+ if (!new_file)
++
++ ret = IS_ERR(new_file) ? PTR_ERR(new_file) : -ENOENT;
++ if (!new_file || IS_ERR(new_file))
+ goto out_fput_old;
+
+ size = i_size_read(old_file->f_dentry->d_inode);
-+ ret = vfs_sendfile(new_file, old_file, NULL, size, 0);
-+ vxdprintk(VXD_CBIT(misc, 2), "vfs_sendfile: %d", ret);
-+
-+ if (ret < 0)
++ ret = do_cow_splice(old_file, new_file, size);
++ vxdprintk(VXD_CBIT(misc, 2), "do_splice_direct: %d", ret);
++ if (ret < 0) {
++ goto out_fput_both;
++ } else if (ret < size) {
++ ret = -ENOSPC;
+ goto out_fput_both;
++ } else {
++ struct inode *old_inode = old_path.dentry->d_inode;
++ struct inode *new_inode = new_path.dentry->d_inode;
++ struct iattr attr = {
++ .ia_uid = old_inode->i_uid,
++ .ia_gid = old_inode->i_gid,
++ .ia_valid = ATTR_UID | ATTR_GID
++ };
+
-+ ret = vfs_rename(dir_nd.dentry->d_inode, new_dentry,
-+ old_nd.dentry->d_parent->d_inode, old_dentry);
-+ vxdprintk(VXD_CBIT(misc, 2), "vfs_rename: %d", ret);
-+ if (!ret) {
-+ res = new_dentry;
-+ dget(new_dentry);
++ ret = inode_setattr(new_inode, &attr);
++ if (ret)
++ goto out_fput_both;
+ }
+
++ mutex_lock(&old_path.dentry->d_inode->i_sb->s_vfs_rename_mutex);
++
++ /* drop out late */
++ ret = -ENOENT;
++ if ((redo = d_unhashed(old_path.dentry)))
++ goto out_unlock;
++
++ vxdprintk(VXD_CBIT(misc, 2),
++ "vfs_rename: [»%*s«:%d] -> [»%*s«:%d]",
++ new_path.dentry->d_name.len, new_path.dentry->d_name.name,
++ new_path.dentry->d_name.len,
++ old_path.dentry->d_name.len, old_path.dentry->d_name.name,
++ old_path.dentry->d_name.len);
++ ret = vfs_rename(dir_nd.path.dentry->d_inode, new_path.dentry,
++ old_nd.path.dentry->d_parent->d_inode, old_path.dentry);
++ vxdprintk(VXD_CBIT(misc, 2), "vfs_rename: %d", ret);
++ res = new_path.dentry;
++
++out_unlock:
++ mutex_unlock(&old_path.dentry->d_inode->i_sb->s_vfs_rename_mutex);
++
+out_fput_both:
+ vxdprintk(VXD_CBIT(misc, 3),
+ "fput(new_file=%p[#%d])", new_file,
@@ -6341,14 +5744,41 @@
+ atomic_read(&old_file->f_count));
+ fput(old_file);
+
-+out_rel_both:
-+ mutex_unlock(&dir_nd.dentry->d_inode->i_mutex);
-+ dput(new_dentry);
++out_unlock_new:
++ mutex_unlock(&dir->d_inode->i_mutex);
++ if (!ret)
++ goto out_redo;
+
-+ path_release(&dir_nd);
++ /* error path cleanup */
++ vfs_unlink(dir->d_inode, new_path.dentry, &dir_nd);
++ dput(new_path.dentry);
++
++out_redo:
++ if (!redo)
++ goto out_rel_both;
++ /* lookup dentry once again */
++ path_put(&old_nd.path);
++ ret = path_lookup(pathname, LOOKUP_FOLLOW, &old_nd);
++ if (ret)
++ goto out_rel_both;
++
++ new_path.dentry = old_nd.path.dentry;
++ vxdprintk(VXD_CBIT(misc, 2),
++ "path_lookup(redo): %p [»%.*s«:%d]", new_path.dentry,
++ new_path.dentry->d_name.len, new_path.dentry->d_name.name,
++ new_path.dentry->d_name.len);
++ dget(new_path.dentry);
++ res = new_path.dentry;
++
++out_rel_both:
++ path_put(&dir_nd.path);
+out_rel_old:
-+ path_release(&old_nd);
++ path_put(&old_nd.path);
++out_free_path:
+ kfree(path);
++out:
++ if (ret)
++ res = ERR_PTR(ret);
+ return res;
+}
+
@@ -6357,12 +5787,12 @@
/* get the link contents into pagecache */
static char *page_getlink(struct dentry * dentry, struct page **ppage)
{
---- linux-2.6.22/fs/namespace.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/namespace.c 2007-06-15 03:24:51 +0200
-@@ -25,6 +25,11 @@
- #include <linux/security.h>
+--- linux-2.6.25.4/fs/namespace.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/namespace.c 2008-04-21 17:20:53.000000000 -0400
+@@ -26,6 +26,11 @@
#include <linux/mount.h>
#include <linux/ramfs.h>
+ #include <linux/log2.h>
+#include <linux/vs_base.h>
+#include <linux/vs_context.h>
+#include <linux/vs_tag.h>
@@ -6371,7 +5801,7 @@
#include <asm/uaccess.h>
#include <asm/unistd.h>
#include "pnode.h"
-@@ -240,6 +245,7 @@ static struct vfsmount *clone_mnt(struct
+@@ -244,6 +249,7 @@ static struct vfsmount *clone_mnt(struct
mnt->mnt_root = dget(root);
mnt->mnt_mountpoint = mnt->mnt_root;
mnt->mnt_parent = mnt;
@@ -6379,39 +5809,40 @@
if (flag & CL_SLAVE) {
list_add(&mnt->mnt_slave, &old->mnt_slave_list);
-@@ -348,48 +354,91 @@ static inline void mangle(struct seq_fil
+@@ -323,6 +329,31 @@ static inline void mangle(struct seq_fil
seq_escape(m, s, " \t\n\\");
}
+static int mnt_is_reachable(struct vfsmount *mnt)
+{
-+ struct vfsmount *root_mnt;
-+ struct dentry *root, *point;
++ struct path root;
++ struct dentry *point;
+ int ret;
+
+ if (mnt == mnt->mnt_ns->root)
+ return 1;
+
+ spin_lock(&vfsmount_lock);
-+ root_mnt = current->fs->rootmnt;
+ root = current->fs->root;
-+ point = root;
++ point = root.dentry;
+
-+ while ((mnt != mnt->mnt_parent) && (mnt != root_mnt)) {
++ while ((mnt != mnt->mnt_parent) && (mnt != root.mnt)) {
+ point = mnt->mnt_mountpoint;
+ mnt = mnt->mnt_parent;
+ }
+
-+ ret = (mnt == root_mnt) && is_subdir(point, root);
++ ret = (mnt == root.mnt) && is_subdir(point, root.dentry);
+
+ spin_unlock(&vfsmount_lock);
+
+ return ret;
+}
+
- static int show_vfsmnt(struct seq_file *m, void *v)
- {
- struct vfsmount *mnt = v;
+ /*
+ * Simple .show_options callback for filesystems which don't want to
+ * implement more complex mount option showing.
+@@ -388,44 +419,61 @@ static int show_vfsmnt(struct seq_file *
+ struct vfsmount *mnt = list_entry(v, struct vfsmount, mnt_list);
int err = 0;
static struct proc_fs_info {
- int flag;
@@ -6448,13 +5879,14 @@
+ { 0, 0, NULL, NULL }
};
- struct proc_fs_info *fs_infop;
+- struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
+ struct proc_fs_info *p;
+ unsigned long s_flags = mnt->mnt_sb->s_flags;
+ int mnt_flags = mnt->mnt_flags;
- mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none");
- seq_putc(m, ' ');
-- seq_path(m, mnt, mnt->mnt_root, " \t\n\\");
+- seq_path(m, &mnt_path, " \t\n\\");
- seq_putc(m, ' ');
- mangle(m, mnt->mnt_sb->s_type->name);
- if (mnt->mnt_sb->s_subtype && mnt->mnt_sb->s_subtype[0]) {
@@ -6471,12 +5903,13 @@
+ return 0;
+
+ if (!vx_check(0, VS_ADMIN|VS_WATCH) &&
-+ mnt == current->fs->rootmnt) {
++ mnt == current->fs->root.mnt) {
+ seq_puts(m, "/dev/root / ");
+ } else {
++ struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
+ mangle(m, mnt->mnt_devname ? mnt->mnt_devname : "none");
+ seq_putc(m, ' ');
-+ seq_path(m, mnt, mnt->mnt_root, " \t\n\\");
++ seq_path(m, &mnt_path, " \t\n\\");
+ seq_putc(m, ' ');
+
+ if (mnt->mnt_sb->s_subtype && mnt->mnt_sb->s_subtype[0]) {
@@ -6503,8 +5936,8 @@
if (mnt->mnt_sb->s_op->show_options)
err = mnt->mnt_sb->s_op->show_options(m, mnt);
seq_puts(m, " 0 0\n");
-@@ -408,17 +457,27 @@ static int show_vfsstat(struct seq_file
- struct vfsmount *mnt = v;
+@@ -445,17 +493,27 @@ static int show_vfsstat(struct seq_file
+ struct path mnt_path = { .dentry = mnt->mnt_root, .mnt = mnt };
int err = 0;
- /* device */
@@ -6520,10 +5953,10 @@
- /* mount point */
- seq_puts(m, " mounted on ");
-- seq_path(m, mnt, mnt->mnt_root, " \t\n\\");
+- seq_path(m, &mnt_path, " \t\n\\");
- seq_putc(m, ' ');
+ if (!vx_check(0, VS_ADMIN|VS_WATCH) &&
-+ mnt == current->fs->rootmnt) {
++ mnt == current->fs->root.mnt) {
+ seq_puts(m, "device /dev/root mounted on / ");
+ } else {
+ /* device */
@@ -6535,13 +5968,13 @@
+
+ /* mount point */
+ seq_puts(m, " mounted on ");
-+ seq_path(m, mnt, mnt->mnt_root, " \t\n\\");
++ seq_path(m, &mnt_path, " \t\n\\");
+ seq_putc(m, ' ');
+ }
/* file system type */
seq_puts(m, "with fstype ");
-@@ -648,7 +707,7 @@ asmlinkage long sys_umount(char __user *
+@@ -693,7 +751,7 @@ asmlinkage long sys_umount(char __user *
goto dput_and_out;
retval = -EPERM;
@@ -6549,8 +5982,8 @@
+ if (!vx_capable(CAP_SYS_ADMIN, VXC_SECURE_MOUNT))
goto dput_and_out;
- retval = do_umount(nd.mnt, flags);
-@@ -672,7 +731,7 @@ asmlinkage long sys_oldumount(char __use
+ retval = do_umount(nd.path.mnt, flags);
+@@ -719,7 +777,7 @@ asmlinkage long sys_oldumount(char __use
static int mount_is_safe(struct nameidata *nd)
{
@@ -6559,22 +5992,22 @@
return 0;
return -EPERM;
#ifdef notyet
-@@ -904,11 +963,13 @@ static int do_change_type(struct nameida
- /*
- * do loopback mount.
+@@ -974,11 +1032,13 @@ static noinline int do_change_type(struc
+ * noinline this do_mount helper to save do_mount stack space.
*/
--static int do_loopback(struct nameidata *nd, char *old_name, int recurse)
-+static int do_loopback(struct nameidata *nd, char *old_name, tag_t tag,
-+ unsigned long flags, int mnt_flags)
+ static noinline int do_loopback(struct nameidata *nd, char *old_name,
+- int recurse)
++ tag_t tag, unsigned long flags, int mnt_flags)
{
struct nameidata old_nd;
struct vfsmount *mnt = NULL;
int err = mount_is_safe(nd);
+ int recurse = flags & MS_REC;
++
if (err)
return err;
if (!old_name || !*old_name)
-@@ -934,6 +995,12 @@ static int do_loopback(struct nameidata
+@@ -1004,6 +1064,12 @@ static noinline int do_loopback(struct n
if (!mnt)
goto out;
@@ -6587,7 +6020,7 @@
err = graft_tree(mnt, nd);
if (err) {
LIST_HEAD(umount_list);
-@@ -942,6 +1009,7 @@ static int do_loopback(struct nameidata
+@@ -1012,6 +1078,7 @@ static noinline int do_loopback(struct n
spin_unlock(&vfsmount_lock);
release_mounts(&umount_list);
}
@@ -6595,23 +6028,23 @@
out:
up_write(&namespace_sem);
-@@ -955,12 +1023,12 @@ out:
- * on it - tough luck.
+@@ -1026,12 +1093,12 @@ out:
+ * noinline this do_mount helper to save do_mount stack space.
*/
- static int do_remount(struct nameidata *nd, int flags, int mnt_flags,
+ static noinline int do_remount(struct nameidata *nd, int flags, int mnt_flags,
- void *data)
-+ void *data, xid_t xid)
++ void *data, xid_t xid)
{
int err;
- struct super_block *sb = nd->mnt->mnt_sb;
+ struct super_block *sb = nd->path.mnt->mnt_sb;
- if (!capable(CAP_SYS_ADMIN))
+ if (!vx_capable(CAP_SYS_ADMIN, VXC_SECURE_REMOUNT))
return -EPERM;
- if (!check_mnt(nd->mnt))
-@@ -994,7 +1062,7 @@ static int do_move_mount(struct nameidat
- struct nameidata old_nd, parent_nd;
+ if (!check_mnt(nd->path.mnt))
+@@ -1069,7 +1136,7 @@ static noinline int do_move_mount(struct
+ struct path parent_path;
struct vfsmount *p;
int err = 0;
- if (!capable(CAP_SYS_ADMIN))
@@ -6619,7 +6052,7 @@
return -EPERM;
if (!old_name || !*old_name)
return -EINVAL;
-@@ -1074,7 +1142,7 @@ static int do_new_mount(struct nameidata
+@@ -1152,7 +1219,7 @@ static noinline int do_new_mount(struct
return -EINVAL;
/* we need capabilities... */
@@ -6628,7 +6061,7 @@
return -EPERM;
mnt = do_kern_mount(type, flags, name, data);
-@@ -1386,6 +1454,7 @@ long do_mount(char *dev_name, char *dir_
+@@ -1397,6 +1464,7 @@ long do_mount(char *dev_name, char *dir_
struct nameidata nd;
int retval = 0;
int mnt_flags = 0;
@@ -6636,19 +6069,17 @@
/* Discard magic */
if ((flags & MS_MGC_MSK) == MS_MGC_VAL)
-@@ -1401,7 +1470,19 @@ long do_mount(char *dev_name, char *dir_
+@@ -1412,7 +1480,17 @@ long do_mount(char *dev_name, char *dir_
if (data_page)
((char *)data_page)[PAGE_SIZE - 1] = 0;
-+#ifdef CONFIG_PROPAGATE
+ retval = dx_parse_tag(data_page, &tag, 1);
+ if (retval) {
-+ mnt_flags |= MNT_TAGID;
-+ /* bind and re-mounts get the tag flag */
++ mnt_flags |= retval;
++ /* FIXME: bind and re-mounts get the tag flag? */
+ if (flags & (MS_BIND|MS_REMOUNT))
+ flags |= MS_TAGID;
+ }
-+#endif
+
/* Separate the per-mountpoint flags */
+ if (flags & MS_RDONLY)
@@ -6656,16 +6087,16 @@
if (flags & MS_NOSUID)
mnt_flags |= MNT_NOSUID;
if (flags & MS_NODEV)
-@@ -1415,6 +1496,8 @@ long do_mount(char *dev_name, char *dir_
+@@ -1426,6 +1504,8 @@ long do_mount(char *dev_name, char *dir_
if (flags & MS_RELATIME)
mnt_flags |= MNT_RELATIME;
+ if (!capable(CAP_SYS_ADMIN))
+ mnt_flags |= MNT_NODEV;
flags &= ~(MS_NOSUID | MS_NOEXEC | MS_NODEV | MS_ACTIVE |
- MS_NOATIME | MS_NODIRATIME | MS_RELATIME);
+ MS_NOATIME | MS_NODIRATIME | MS_RELATIME| MS_KERNMOUNT);
-@@ -1429,9 +1512,9 @@ long do_mount(char *dev_name, char *dir_
+@@ -1440,9 +1520,9 @@ long do_mount(char *dev_name, char *dir_
if (flags & MS_REMOUNT)
retval = do_remount(&nd, flags & ~MS_REMOUNT, mnt_flags,
@@ -6677,7 +6108,7 @@
else if (flags & (MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE))
retval = do_change_type(&nd, flags);
else if (flags & MS_MOVE)
-@@ -1504,6 +1587,7 @@ static struct mnt_namespace *dup_mnt_ns(
+@@ -1515,6 +1595,7 @@ static struct mnt_namespace *dup_mnt_ns(
q = next_mnt(q, new_ns->root);
}
up_write(&namespace_sem);
@@ -6685,18 +6116,18 @@
if (rootmnt)
mntput(rootmnt);
-@@ -1866,5 +1950,6 @@ void __put_mnt_ns(struct mnt_namespace *
+@@ -1850,5 +1931,6 @@ void __put_mnt_ns(struct mnt_namespace *
spin_unlock(&vfsmount_lock);
up_write(&namespace_sem);
release_mounts(&umount_list);
+ atomic_dec(&vs_global_mnt_ns);
kfree(ns);
}
---- linux-2.6.22/fs/nfs/client.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfs/client.c 2007-06-15 02:37:03 +0200
-@@ -521,6 +521,9 @@ static int nfs_init_server_rpcclient(str
- if (server->flags & NFS4_MOUNT_INTR)
- server->client->cl_intr = 1;
+--- linux-2.6.25.4/fs/nfs/client.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfs/client.c 2008-04-19 15:14:52.000000000 -0400
+@@ -589,6 +589,9 @@ static int nfs_init_server_rpcclient(str
+ if (server->flags & NFS_MOUNT_SOFT)
+ server->client->cl_softrtry = 1;
+ server->client->cl_tag = 0;
+ if (server->flags & NFS_MOUNT_TAGGED)
@@ -6704,7 +6135,7 @@
return 0;
}
-@@ -678,6 +681,10 @@ static void nfs_server_set_fsinfo(struct
+@@ -742,6 +745,10 @@ static void nfs_server_set_fsinfo(struct
server->acdirmin = server->acdirmax = 0;
}
@@ -6715,8 +6146,8 @@
server->maxfilesize = fsinfo->maxfilesize;
/* We're airborne Set socket buffersize */
---- linux-2.6.22/fs/nfs/dir.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfs/dir.c 2007-06-15 03:02:49 +0200
+--- linux-2.6.25.4/fs/nfs/dir.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfs/dir.c 2008-04-21 16:52:03.000000000 -0400
@@ -34,6 +34,7 @@
#include <linux/namei.h>
#include <linux/mount.h>
@@ -6725,26 +6156,26 @@
#include "nfs4_fs.h"
#include "delegation.h"
-@@ -956,6 +957,7 @@ static struct dentry *nfs_lookup(struct
+@@ -930,6 +931,7 @@ static struct dentry *nfs_lookup(struct
if (IS_ERR(res))
- goto out_unlock;
+ goto out_unblock_sillyrename;
+ dx_propagate_tag(nd, inode);
no_entry:
res = d_materialise_unique(dentry, inode);
if (res != NULL) {
-@@ -998,7 +1000,8 @@ static int is_atomic_open(struct inode *
+@@ -967,7 +969,8 @@ static int is_atomic_open(struct inode *
if (nd->flags & LOOKUP_DIRECTORY)
return 0;
/* Are we trying to write to a read only partition? */
- if (IS_RDONLY(dir) && (nd->intent.open.flags & (O_CREAT|O_TRUNC|FMODE_WRITE)))
-+ if ((IS_RDONLY(dir) || MNT_IS_RDONLY(nd->mnt)) &&
++ if ((IS_RDONLY(dir) || MNT_IS_RDONLY(nd->path.mnt)) &&
+ (nd->intent.open.flags & (O_CREAT|O_TRUNC|FMODE_WRITE)))
return 0;
return 1;
}
---- linux-2.6.22/fs/nfs/inode.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfs/inode.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/nfs/inode.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfs/inode.c 2008-04-19 15:14:52.000000000 -0400
@@ -37,6 +37,7 @@
#include <linux/vfs.h>
#include <linux/inet.h>
@@ -6753,7 +6184,7 @@
#include <asm/system.h>
#include <asm/uaccess.h>
-@@ -285,8 +286,10 @@ nfs_fhget(struct super_block *sb, struct
+@@ -316,8 +317,10 @@ nfs_fhget(struct super_block *sb, struct
nfsi->change_attr = fattr->change_attr;
inode->i_size = nfs_size_to_loff_t(fattr->size);
inode->i_nlink = fattr->nlink;
@@ -6766,7 +6197,7 @@
if (fattr->valid & (NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4)) {
/*
* report the blocks in 512byte units
-@@ -377,6 +380,8 @@ void nfs_setattr_update_inode(struct ino
+@@ -410,6 +413,8 @@ void nfs_setattr_update_inode(struct ino
inode->i_uid = attr->ia_uid;
if ((attr->ia_valid & ATTR_GID) != 0)
inode->i_gid = attr->ia_gid;
@@ -6775,19 +6206,19 @@
spin_lock(&inode->i_lock);
NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL;
spin_unlock(&inode->i_lock);
-@@ -825,6 +830,9 @@ static int nfs_check_inode_attributes(st
+@@ -841,6 +846,9 @@ static int nfs_check_inode_attributes(st
struct nfs_inode *nfsi = NFS_I(inode);
loff_t cur_size, new_isize;
- int data_unstable;
+ unsigned long invalid = 0;
+ uid_t uid;
+ gid_t gid;
+ tag_t tag;
/* Has the inode gone and changed behind our back? */
-@@ -852,10 +860,15 @@ static int nfs_check_inode_attributes(st
+@@ -865,10 +873,15 @@ static int nfs_check_inode_attributes(st
if (cur_size != new_isize && nfsi->npages == 0)
- nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE;
+ invalid |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE;
+ uid = INOTAG_UID(DX_TAG(inode), fattr->uid, fattr->gid);
+ gid = INOTAG_GID(DX_TAG(inode), fattr->uid, fattr->gid);
@@ -6800,22 +6231,22 @@
+ || inode->i_uid != uid
+ || inode->i_gid != gid
+ || inode->i_tag != tag)
- nfsi->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL;
+ invalid |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL;
/* Has the link count changed? */
-@@ -946,6 +959,9 @@ static int nfs_update_inode(struct inode
- unsigned int invalid = 0;
+@@ -989,6 +1002,9 @@ static int nfs_update_inode(struct inode
+ loff_t cur_isize, new_isize;
+ unsigned long invalid = 0;
unsigned long now = jiffies;
- int data_stable;
+ uid_t uid;
+ gid_t gid;
+ tag_t tag;
dfprintk(VFS, "NFS: %s(%s/%ld ct=%d info=0x%x)\n",
__FUNCTION__, inode->i_sb->s_id, inode->i_ino,
-@@ -1022,15 +1038,21 @@ static int nfs_update_inode(struct inode
- }
+@@ -1062,15 +1078,21 @@ static int nfs_update_inode(struct inode
memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime));
+ nfsi->change_attr = fattr->change_attr;
+ uid = INOTAG_UID(DX_TAG(inode), fattr->uid, fattr->gid);
+ gid = INOTAG_GID(DX_TAG(inode), fattr->uid, fattr->gid);
@@ -6839,8 +6270,8 @@
if (fattr->valid & (NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4)) {
/*
---- linux-2.6.22/fs/nfs/nfs3xdr.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfs/nfs3xdr.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/nfs/nfs3xdr.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfs/nfs3xdr.c 2008-04-19 15:14:52.000000000 -0400
@@ -22,6 +22,7 @@
#include <linux/nfs3.h>
#include <linux/nfs_fs.h>
@@ -6849,7 +6280,7 @@
#include "internal.h"
#define NFSDBG_FACILITY NFSDBG_XDR
-@@ -178,7 +179,7 @@ xdr_decode_fattr(__be32 *p, struct nfs_f
+@@ -180,7 +181,7 @@ xdr_decode_fattr(__be32 *p, struct nfs_f
}
static inline __be32 *
@@ -6858,7 +6289,7 @@
{
if (attr->ia_valid & ATTR_MODE) {
*p++ = xdr_one;
-@@ -186,15 +187,17 @@ xdr_encode_sattr(__be32 *p, struct iattr
+@@ -188,15 +189,17 @@ xdr_encode_sattr(__be32 *p, struct iattr
} else {
*p++ = xdr_zero;
}
@@ -6880,7 +6311,7 @@
} else {
*p++ = xdr_zero;
}
-@@ -279,7 +282,8 @@ static int
+@@ -281,7 +284,8 @@ static int
nfs3_xdr_sattrargs(struct rpc_rqst *req, __be32 *p, struct nfs3_sattrargs *args)
{
p = xdr_encode_fhandle(p, args->fh);
@@ -6890,7 +6321,7 @@
*p++ = htonl(args->guard);
if (args->guard)
p = xdr_encode_time3(p, &args->guardtime);
-@@ -370,7 +374,8 @@ nfs3_xdr_createargs(struct rpc_rqst *req
+@@ -386,7 +390,8 @@ nfs3_xdr_createargs(struct rpc_rqst *req
*p++ = args->verifier[0];
*p++ = args->verifier[1];
} else
@@ -6900,7 +6331,7 @@
req->rq_slen = xdr_adjust_iovec(req->rq_svec, p);
return 0;
-@@ -384,7 +389,8 @@ nfs3_xdr_mkdirargs(struct rpc_rqst *req,
+@@ -400,7 +405,8 @@ nfs3_xdr_mkdirargs(struct rpc_rqst *req,
{
p = xdr_encode_fhandle(p, args->fh);
p = xdr_encode_array(p, args->name, args->len);
@@ -6910,7 +6341,7 @@
req->rq_slen = xdr_adjust_iovec(req->rq_svec, p);
return 0;
}
-@@ -397,7 +403,8 @@ nfs3_xdr_symlinkargs(struct rpc_rqst *re
+@@ -413,7 +419,8 @@ nfs3_xdr_symlinkargs(struct rpc_rqst *re
{
p = xdr_encode_fhandle(p, args->fromfh);
p = xdr_encode_array(p, args->fromname, args->fromlen);
@@ -6920,7 +6351,7 @@
*p++ = htonl(args->pathlen);
req->rq_slen = xdr_adjust_iovec(req->rq_svec, p);
-@@ -415,7 +422,8 @@ nfs3_xdr_mknodargs(struct rpc_rqst *req,
+@@ -431,7 +438,8 @@ nfs3_xdr_mknodargs(struct rpc_rqst *req,
p = xdr_encode_fhandle(p, args->fh);
p = xdr_encode_array(p, args->name, args->len);
*p++ = htonl(args->type);
@@ -6930,9 +6361,9 @@
if (args->type == NF3CHR || args->type == NF3BLK) {
*p++ = htonl(MAJOR(args->rdev));
*p++ = htonl(MINOR(args->rdev));
---- linux-2.6.22/fs/nfs/nfsroot.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfs/nfsroot.c 2007-06-15 02:37:03 +0200
-@@ -118,12 +118,12 @@ static int mount_port __initdata = 0; /
+--- linux-2.6.25.4/fs/nfs/nfsroot.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfs/nfsroot.c 2008-04-19 15:14:52.000000000 -0400
+@@ -119,12 +119,12 @@ static int mount_port __initdata = 0; /
enum {
/* Options that take integer arguments */
Opt_port, Opt_rsize, Opt_wsize, Opt_timeo, Opt_retrans, Opt_acregmin,
@@ -6947,18 +6378,17 @@
/* Error token */
Opt_err
};
-@@ -160,6 +160,10 @@ static match_table_t __initdata tokens =
+@@ -161,6 +161,9 @@ static match_table_t __initdata tokens =
{Opt_tcp, "tcp"},
{Opt_acl, "acl"},
{Opt_noacl, "noacl"},
+ {Opt_tag, "tag"},
+ {Opt_notag, "notag"},
+ {Opt_tagid, "tagid=%u"},
-+ {Opt_tag, "tagxid"},
{Opt_err, NULL}
};
-@@ -274,6 +278,20 @@ static int __init root_nfs_parse(char *n
+@@ -272,6 +275,20 @@ static int __init root_nfs_parse(char *n
case Opt_noacl:
nfs_data.flags |= NFS_MOUNT_NOACL;
break;
@@ -6979,35 +6409,35 @@
default:
printk(KERN_WARNING "Root-NFS: unknown "
"option: %s\n", p);
---- linux-2.6.22/fs/nfs/super.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfs/super.c 2007-06-15 03:03:21 +0200
-@@ -45,6 +45,7 @@
- #include <linux/inet.h>
+--- linux-2.6.25.4/fs/nfs/super.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfs/super.c 2008-04-19 15:14:52.000000000 -0400
+@@ -50,6 +50,7 @@
#include <linux/nfs_xdr.h>
#include <linux/magic.h>
+ #include <linux/parser.h>
+#include <linux/vs_tag.h>
#include <asm/system.h>
#include <asm/uaccess.h>
-@@ -291,6 +292,7 @@ static void nfs_show_mount_options(struc
- { NFS_MOUNT_NONLM, ",nolock", "" },
+@@ -458,6 +459,7 @@ static void nfs_show_mount_options(struc
{ NFS_MOUNT_NOACL, ",noacl", "" },
{ NFS_MOUNT_NORDIRPLUS, ",nordirplus", "" },
+ { NFS_MOUNT_UNSHARED, ",nosharecache", ""},
+ { NFS_MOUNT_TAGGED, ",tag", "" },
{ 0, NULL, NULL }
};
const struct proc_nfs_info *nfs_infop;
---- linux-2.6.22/fs/nfsd/auth.c 2006-06-18 04:54:42 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfsd/auth.c 2007-06-15 02:37:03 +0200
-@@ -9,6 +9,7 @@
- #include <linux/sunrpc/svc.h>
+--- linux-2.6.25.4/fs/nfsd/auth.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfsd/auth.c 2008-04-19 17:07:41.000000000 -0400
+@@ -10,6 +10,7 @@
#include <linux/sunrpc/svcauth.h>
#include <linux/nfsd/nfsd.h>
+ #include <linux/nfsd/export.h>
+#include <linux/vs_tag.h>
- #define CAP_NFSD_MASK (CAP_FS_MASK|CAP_TO_MASK(CAP_SYS_RESOURCE))
-
-@@ -41,19 +42,22 @@ int nfsd_setuser(struct svc_rqst *rqstp,
+ int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp)
+ {
+@@ -54,19 +55,23 @@ int nfsd_setuser(struct svc_rqst *rqstp,
get_group_info(cred.cr_group_info);
if (cred.cr_uid != (uid_t) -1)
@@ -7029,21 +6459,22 @@
ret = set_current_groups(cred.cr_group_info);
put_group_info(cred.cr_group_info);
- if ((cred.cr_uid)) {
++
+ if (INOTAG_UID(DX_TAG_NFSD, cred.cr_uid, cred.cr_gid)) {
- cap_t(current->cap_effective) &= ~CAP_NFSD_MASK;
+ current->cap_effective =
+ cap_drop_nfsd_set(current->cap_effective);
} else {
- cap_t(current->cap_effective) |= (CAP_NFSD_MASK &
---- linux-2.6.22/fs/nfsd/nfs3xdr.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfsd/nfs3xdr.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/nfsd/nfs3xdr.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfsd/nfs3xdr.c 2008-04-19 17:08:09.000000000 -0400
@@ -21,6 +21,7 @@
#include <linux/sunrpc/svc.h>
#include <linux/nfsd/nfsd.h>
#include <linux/nfsd/xdr3.h>
+#include <linux/vs_tag.h>
+ #include "auth.h"
#define NFSDDBG_FACILITY NFSDDBG_XDR
-
-@@ -107,6 +108,8 @@ static __be32 *
+@@ -108,6 +109,8 @@ static __be32 *
decode_sattr3(__be32 *p, struct iattr *iap)
{
u32 tmp;
@@ -7052,7 +6483,7 @@
iap->ia_valid = 0;
-@@ -116,12 +119,15 @@ decode_sattr3(__be32 *p, struct iattr *i
+@@ -117,12 +120,15 @@ decode_sattr3(__be32 *p, struct iattr *i
}
if (*p++) {
iap->ia_valid |= ATTR_UID;
@@ -7070,31 +6501,33 @@
if (*p++) {
u64 newsize;
-@@ -180,8 +186,10 @@ encode_fattr3(struct svc_rqst *rqstp, __
+@@ -178,8 +184,12 @@ encode_fattr3(struct svc_rqst *rqstp, __
*p++ = htonl(nfs3_ftypes[(stat->mode & S_IFMT) >> 12]);
*p++ = htonl((u32) stat->mode);
*p++ = htonl((u32) stat->nlink);
- *p++ = htonl((u32) nfsd_ruid(rqstp, stat->uid));
- *p++ = htonl((u32) nfsd_rgid(rqstp, stat->gid));
+ *p++ = htonl((u32) nfsd_ruid(rqstp,
-+ TAGINO_UID(DX_TAG(dentry->d_inode), stat->uid, stat->tag)));
++ TAGINO_UID(0 /* FIXME: DX_TAG(dentry->d_inode) */,
++ stat->uid, stat->tag)));
+ *p++ = htonl((u32) nfsd_rgid(rqstp,
-+ TAGINO_GID(DX_TAG(dentry->d_inode), stat->gid, stat->tag)));
++ TAGINO_GID(0 /* FIXME: DX_TAG(dentry->d_inode) */,
++ stat->gid, stat->tag)));
if (S_ISLNK(stat->mode) && stat->size > NFS3_MAXPATHLEN) {
p = xdr_encode_hyper(p, (u64) NFS3_MAXPATHLEN);
} else {
---- linux-2.6.22/fs/nfsd/nfs4recover.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfsd/nfs4recover.c 2007-06-15 02:37:03 +0200
-@@ -156,7 +156,7 @@ nfsd4_create_clid_dir(struct nfs4_client
+--- linux-2.6.25.4/fs/nfsd/nfs4recover.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfsd/nfs4recover.c 2008-04-20 13:24:11.000000000 -0400
+@@ -154,7 +154,7 @@ nfsd4_create_clid_dir(struct nfs4_client
dprintk("NFSD: nfsd4_create_clid_dir: DIRECTORY EXISTS\n");
goto out_put;
}
-- status = vfs_mkdir(rec_dir.dentry->d_inode, dentry, S_IRWXU);
-+ status = vfs_mkdir(rec_dir.dentry->d_inode, dentry, S_IRWXU, NULL);
+- status = vfs_mkdir(rec_dir.path.dentry->d_inode, dentry, S_IRWXU);
++ status = vfs_mkdir(rec_dir.path.dentry->d_inode, dentry, S_IRWXU, NULL);
out_put:
dput(dentry);
out_unlock:
-@@ -260,7 +260,7 @@ nfsd4_remove_clid_file(struct dentry *di
+@@ -258,7 +258,7 @@ nfsd4_remove_clid_file(struct dentry *di
return -EINVAL;
}
mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT);
@@ -7103,7 +6536,7 @@
mutex_unlock(&dir->d_inode->i_mutex);
return status;
}
-@@ -275,7 +275,7 @@ nfsd4_clear_clid_dir(struct dentry *dir,
+@@ -273,7 +273,7 @@ nfsd4_clear_clid_dir(struct dentry *dir,
* a kernel from the future.... */
nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file);
mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT);
@@ -7112,17 +6545,17 @@
mutex_unlock(&dir->d_inode->i_mutex);
return status;
}
---- linux-2.6.22/fs/nfsd/nfs4xdr.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfsd/nfs4xdr.c 2007-06-15 02:37:03 +0200
-@@ -56,6 +56,7 @@
- #include <linux/nfsd_idmap.h>
- #include <linux/nfs4.h>
+--- linux-2.6.25.4/fs/nfsd/nfs4xdr.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfsd/nfs4xdr.c 2008-04-19 15:14:52.000000000 -0400
+@@ -58,6 +58,7 @@
#include <linux/nfs4_acl.h>
+ #include <linux/sunrpc/gss_api.h>
+ #include <linux/sunrpc/svcauth_gss.h>
+#include <linux/vs_tag.h>
#define NFSDDBG_FACILITY NFSDDBG_XDR
-@@ -1727,14 +1728,18 @@ out_acl:
+@@ -1759,14 +1760,18 @@ out_acl:
WRITE32(stat.nlink);
}
if (bmval1 & FATTR4_WORD1_OWNER) {
@@ -7143,17 +6576,17 @@
if (status == nfserr_resource)
goto out_resource;
if (status)
---- linux-2.6.22/fs/nfsd/nfsxdr.c 2007-07-09 13:19:27 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfsd/nfsxdr.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/nfsd/nfsxdr.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfsd/nfsxdr.c 2008-04-20 13:23:36.000000000 -0400
@@ -15,6 +15,7 @@
#include <linux/nfsd/nfsd.h>
#include <linux/nfsd/xdr.h>
#include <linux/mm.h>
+#include <linux/vs_tag.h>
+ #include "auth.h"
#define NFSDDBG_FACILITY NFSDDBG_XDR
-
-@@ -97,6 +98,8 @@ static __be32 *
+@@ -98,6 +99,8 @@ static __be32 *
decode_sattr(__be32 *p, struct iattr *iap)
{
u32 tmp, tmp1;
@@ -7162,7 +6595,7 @@
iap->ia_valid = 0;
-@@ -110,12 +113,15 @@ decode_sattr(__be32 *p, struct iattr *ia
+@@ -111,12 +114,15 @@ decode_sattr(__be32 *p, struct iattr *ia
}
if ((tmp = ntohl(*p++)) != (u32)-1) {
iap->ia_valid |= ATTR_UID;
@@ -7180,7 +6613,7 @@
if ((tmp = ntohl(*p++)) != (u32)-1) {
iap->ia_valid |= ATTR_SIZE;
iap->ia_size = tmp;
-@@ -160,8 +166,10 @@ encode_fattr(struct svc_rqst *rqstp, __b
+@@ -161,8 +167,10 @@ encode_fattr(struct svc_rqst *rqstp, __b
*p++ = htonl(nfs_ftypes[type >> 12]);
*p++ = htonl((u32) stat->mode);
*p++ = htonl((u32) stat->nlink);
@@ -7193,9 +6626,9 @@
if (S_ISLNK(type) && stat->size > NFS_MAXPATHLEN) {
*p++ = htonl(NFS_MAXPATHLEN);
---- linux-2.6.22/fs/nfsd/vfs.c 2007-05-02 19:25:19 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/nfsd/vfs.c 2007-06-15 02:37:03 +0200
-@@ -1186,13 +1186,13 @@ nfsd_create(struct svc_rqst *rqstp, stru
+--- linux-2.6.25.4/fs/nfsd/vfs.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/nfsd/vfs.c 2008-04-21 17:24:34.000000000 -0400
+@@ -1258,13 +1258,13 @@ nfsd_create(struct svc_rqst *rqstp, stru
host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL);
break;
case S_IFDIR:
@@ -7211,7 +6644,7 @@
break;
default:
printk("nfsd: bad file type %o in nfsd_create\n", type);
-@@ -1466,11 +1466,13 @@ nfsd_symlink(struct svc_rqst *rqstp, str
+@@ -1529,11 +1529,13 @@ nfsd_symlink(struct svc_rqst *rqstp, str
else {
strncpy(path_alloced, path, plen);
path_alloced[plen] = 0;
@@ -7227,7 +6660,7 @@
if (!host_err) {
if (EX_ISSYNC(fhp->fh_export))
-@@ -1529,7 +1531,7 @@ nfsd_link(struct svc_rqst *rqstp, struct
+@@ -1592,7 +1594,7 @@ nfsd_link(struct svc_rqst *rqstp, struct
dold = tfhp->fh_dentry;
dest = dold->d_inode;
@@ -7236,7 +6669,7 @@
if (!host_err) {
if (EX_ISSYNC(ffhp->fh_export)) {
err = nfserrno(nfsd_sync_dir(ddir));
-@@ -1694,9 +1696,9 @@ nfsd_unlink(struct svc_rqst *rqstp, stru
+@@ -1757,9 +1759,9 @@ nfsd_unlink(struct svc_rqst *rqstp, stru
host_err = -EPERM;
} else
#endif
@@ -7248,18 +6681,18 @@
}
dput(rdentry);
-@@ -1807,7 +1809,8 @@ nfsd_permission(struct svc_export *exp,
+@@ -1876,7 +1878,8 @@ nfsd_permission(struct svc_rqst *rqstp,
*/
if (!(acc & MAY_LOCAL_ACCESS))
if (acc & (MAY_WRITE | MAY_SATTR | MAY_TRUNC)) {
-- if (EX_RDONLY(exp) || IS_RDONLY(inode))
-+ if (EX_RDONLY(exp) || IS_RDONLY(inode)
-+ || MNT_IS_RDONLY(exp->ex_mnt))
+- if (exp_rdonly(rqstp, exp) || IS_RDONLY(inode))
++ if (exp_rdonly(rqstp, exp) || IS_RDONLY(inode)
++ || MNT_IS_RDONLY(exp->ex_path.mnt))
return nfserr_rofs;
if (/* (acc & MAY_WRITE) && */ IS_IMMUTABLE(inode))
return nfserr_perm;
---- linux-2.6.22/fs/ocfs2/dlm/dlmfs.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/dlm/dlmfs.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ocfs2/dlm/dlmfs.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/dlm/dlmfs.c 2008-04-19 15:14:52.000000000 -0400
@@ -43,6 +43,7 @@
#include <linux/init.h>
#include <linux/string.h>
@@ -7268,7 +6701,7 @@
#include <asm/uaccess.h>
-@@ -331,6 +332,7 @@ static struct inode *dlmfs_get_root_inod
+@@ -342,6 +343,7 @@ static struct inode *dlmfs_get_root_inod
inode->i_mode = mode;
inode->i_uid = current->fsuid;
inode->i_gid = current->fsgid;
@@ -7276,7 +6709,7 @@
inode->i_blocks = 0;
inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info;
inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-@@ -357,6 +359,7 @@ static struct inode *dlmfs_get_inode(str
+@@ -368,6 +370,7 @@ static struct inode *dlmfs_get_inode(str
inode->i_mode = mode;
inode->i_uid = current->fsuid;
inode->i_gid = current->fsgid;
@@ -7284,9 +6717,9 @@
inode->i_blocks = 0;
inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info;
inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
---- linux-2.6.22/fs/ocfs2/dlmglue.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/dlmglue.c 2007-06-15 02:37:03 +0200
-@@ -1475,6 +1475,7 @@ static void __ocfs2_stuff_meta_lvb(struc
+--- linux-2.6.25.4/fs/ocfs2/dlmglue.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/dlmglue.c 2008-04-19 15:14:52.000000000 -0400
+@@ -1665,6 +1665,7 @@ static void __ocfs2_stuff_meta_lvb(struc
lvb->lvb_iclusters = cpu_to_be32(oi->ip_clusters);
lvb->lvb_iuid = cpu_to_be32(inode->i_uid);
lvb->lvb_igid = cpu_to_be32(inode->i_gid);
@@ -7294,7 +6727,7 @@
lvb->lvb_imode = cpu_to_be16(inode->i_mode);
lvb->lvb_inlink = cpu_to_be16(inode->i_nlink);
lvb->lvb_iatime_packed =
-@@ -1527,6 +1528,7 @@ static void ocfs2_refresh_inode_from_lvb
+@@ -1719,6 +1720,7 @@ static void ocfs2_refresh_inode_from_lvb
inode->i_uid = be32_to_cpu(lvb->lvb_iuid);
inode->i_gid = be32_to_cpu(lvb->lvb_igid);
@@ -7302,20 +6735,21 @@
inode->i_mode = be16_to_cpu(lvb->lvb_imode);
inode->i_nlink = be16_to_cpu(lvb->lvb_inlink);
ocfs2_unpack_timespec(&inode->i_atime,
---- linux-2.6.22/fs/ocfs2/dlmglue.h 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/dlmglue.h 2007-06-15 02:37:03 +0200
-@@ -34,7 +34,7 @@
- struct ocfs2_meta_lvb {
- __u8 lvb_version;
- __u8 lvb_reserved0;
-- __be16 lvb_reserved1;
+--- linux-2.6.25.4/fs/ocfs2/dlmglue.h 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/dlmglue.h 2008-04-19 15:14:52.000000000 -0400
+@@ -46,7 +46,8 @@ struct ocfs2_meta_lvb {
+ __be16 lvb_inlink;
+ __be32 lvb_iattr;
+ __be32 lvb_igeneration;
+- __be32 lvb_reserved2;
+ __be16 lvb_itag;
- __be32 lvb_iclusters;
- __be32 lvb_iuid;
- __be32 lvb_igid;
---- linux-2.6.22/fs/ocfs2/file.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/file.c 2007-06-15 02:37:03 +0200
-@@ -943,13 +943,15 @@ int ocfs2_setattr(struct dentry *dentry,
++ __be16 lvb_reserved2;
+ };
+
+ /* ocfs2_inode_lock_full() 'arg_flags' flags */
+--- linux-2.6.25.4/fs/ocfs2/file.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/file.c 2008-04-19 15:14:52.000000000 -0400
+@@ -1054,13 +1054,15 @@ int ocfs2_setattr(struct dentry *dentry,
mlog(0, "uid change: %d\n", attr->ia_uid);
if (attr->ia_valid & ATTR_GID)
mlog(0, "gid change: %d\n", attr->ia_gid);
@@ -7332,16 +6766,16 @@
if (!(attr->ia_valid & OCFS2_VALID_ATTRS)) {
mlog(0, "can't handle attrs: 0x%x\n", attr->ia_valid);
return 0;
-@@ -1805,6 +1807,7 @@ bail:
- const struct inode_operations ocfs2_file_iops = {
+@@ -2229,6 +2231,7 @@ const struct inode_operations ocfs2_file
+ const struct inode_operations ocfs2_special_file_iops = {
.setattr = ocfs2_setattr,
.getattr = ocfs2_getattr,
+ .sync_flags = ocfs2_sync_flags,
.permission = ocfs2_permission,
};
---- linux-2.6.22/fs/ocfs2/inode.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/inode.c 2007-06-15 03:10:27 +0200
+--- linux-2.6.25.4/fs/ocfs2/inode.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/inode.c 2008-04-21 10:16:08.000000000 -0400
@@ -28,6 +28,7 @@
#include <linux/slab.h>
#include <linux/highmem.h>
@@ -7358,7 +6792,7 @@
#include "journal.h"
#include "namei.h"
#include "suballoc.h"
-@@ -77,6 +79,10 @@ void ocfs2_set_inode_flags(struct inode
+@@ -79,6 +81,10 @@ void ocfs2_set_inode_flags(struct inode
if (flags & OCFS2_IMMUTABLE_FL)
inode->i_flags |= S_IMMUTABLE;
@@ -7369,7 +6803,7 @@
if (flags & OCFS2_SYNC_FL)
inode->i_flags |= S_SYNC;
-@@ -107,6 +113,27 @@ void ocfs2_get_inode_flags(struct ocfs2_
+@@ -109,6 +115,27 @@ void ocfs2_get_inode_flags(struct ocfs2_
oi->ip_attr |= OCFS2_DIRSYNC_FL;
}
@@ -7394,10 +6828,10 @@
+ return 0;
+}
+
- struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, int flags)
+ struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, unsigned flags,
+ int sysfile_type)
{
- struct inode *inode = NULL;
-@@ -212,6 +239,8 @@ int ocfs2_populate_inode(struct inode *i
+@@ -219,6 +246,8 @@ int ocfs2_populate_inode(struct inode *i
struct super_block *sb;
struct ocfs2_super *osb;
int status = -EINVAL;
@@ -7406,7 +6840,7 @@
mlog_entry("(0x%p, size:%llu)\n", inode,
(unsigned long long)le64_to_cpu(fe->i_size));
-@@ -246,8 +275,12 @@ int ocfs2_populate_inode(struct inode *i
+@@ -254,8 +283,12 @@ int ocfs2_populate_inode(struct inode *i
inode->i_generation = le32_to_cpu(fe->i_generation);
inode->i_rdev = huge_decode_dev(le64_to_cpu(fe->id1.dev1.i_rdev));
inode->i_mode = le16_to_cpu(fe->i_mode);
@@ -7421,7 +6855,7 @@
/* Fast symlinks will have i_size but no allocated clusters. */
if (S_ISLNK(inode->i_mode) && !fe->i_clusters)
-@@ -1224,8 +1257,11 @@ int ocfs2_mark_inode_dirty(handle_t *han
+@@ -1230,8 +1263,11 @@ int ocfs2_mark_inode_dirty(handle_t *han
fe->i_size = cpu_to_le64(i_size_read(inode));
fe->i_links_count = cpu_to_le16(inode->i_nlink);
@@ -7435,7 +6869,7 @@
fe->i_mode = cpu_to_le16(inode->i_mode);
fe->i_atime = cpu_to_le64(inode->i_atime.tv_sec);
fe->i_atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec);
-@@ -1253,15 +1289,24 @@ leave:
+@@ -1259,16 +1295,25 @@ leave:
void ocfs2_refresh_inode(struct inode *inode,
struct ocfs2_dinode *fe)
{
@@ -7448,6 +6882,7 @@
OCFS2_I(inode)->ip_attr = le32_to_cpu(fe->i_attr);
+ /* OCFS2_I(inode)->ip_flags &= ~OCFS2_FL_MASK;
+ OCFS2_I(inode)->ip_flags |= le32_to_cpu(fe->i_flags) & OCFS2_FL_MASK; */
+ OCFS2_I(inode)->ip_dyn_features = le16_to_cpu(fe->i_dyn_features);
ocfs2_set_inode_flags(inode);
i_size_write(inode, le64_to_cpu(fe->i_size));
inode->i_nlink = le16_to_cpu(fe->i_links_count);
@@ -7462,9 +6897,9 @@
inode->i_mode = le16_to_cpu(fe->i_mode);
if (S_ISLNK(inode->i_mode) && le32_to_cpu(fe->i_clusters) == 0)
inode->i_blocks = 0;
---- linux-2.6.22/fs/ocfs2/inode.h 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/inode.h 2007-06-15 03:10:53 +0200
-@@ -142,6 +142,7 @@ int ocfs2_aio_write(struct file *file, s
+--- linux-2.6.25.4/fs/ocfs2/inode.h 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/inode.h 2008-04-19 15:14:52.000000000 -0400
+@@ -143,6 +143,7 @@ int ocfs2_aio_write(struct file *file, s
void ocfs2_set_inode_flags(struct inode *inode);
void ocfs2_get_inode_flags(struct ocfs2_inode_info *oi);
@@ -7472,9 +6907,9 @@
static inline blkcnt_t ocfs2_inode_sector_count(struct inode *inode)
{
---- linux-2.6.22/fs/ocfs2/ioctl.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/ioctl.c 2007-06-15 02:37:03 +0200
-@@ -39,7 +39,7 @@ static int ocfs2_get_inode_attr(struct i
+--- linux-2.6.25.4/fs/ocfs2/ioctl.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/ioctl.c 2008-04-19 15:14:52.000000000 -0400
+@@ -41,7 +41,7 @@ static int ocfs2_get_inode_attr(struct i
return status;
}
@@ -7483,8 +6918,8 @@
unsigned mask)
{
struct ocfs2_inode_info *ocfs2_inode = OCFS2_I(inode);
---- linux-2.6.22/fs/ocfs2/ioctl.h 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/ioctl.h 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ocfs2/ioctl.h 2008-04-17 10:33:02.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/ioctl.h 2008-04-19 15:14:52.000000000 -0400
@@ -10,6 +10,9 @@
#ifndef OCFS2_IOCTL_H
#define OCFS2_IOCTL_H
@@ -7495,8 +6930,8 @@
int ocfs2_ioctl(struct inode * inode, struct file * filp,
unsigned int cmd, unsigned long arg);
long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg);
---- linux-2.6.22/fs/ocfs2/namei.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/namei.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/ocfs2/namei.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/namei.c 2008-04-19 15:14:52.000000000 -0400
@@ -40,6 +40,7 @@
#include <linux/types.h>
#include <linux/slab.h>
@@ -7505,7 +6940,7 @@
#define MLOG_MASK_PREFIX ML_NAMEI
#include <cluster/masklog.h>
-@@ -483,6 +484,9 @@ static int ocfs2_mknod_locked(struct ocf
+@@ -366,6 +367,9 @@ static int ocfs2_mknod_locked(struct ocf
u64 fe_blkno = 0;
u16 suballoc_bit;
struct inode *inode = NULL;
@@ -7515,7 +6950,7 @@
mlog_entry("(0x%p, 0x%p, %d, %lu, '%.*s')\n", dir, dentry, mode,
(unsigned long)dev, dentry->d_name.len,
-@@ -542,13 +546,19 @@ static int ocfs2_mknod_locked(struct ocf
+@@ -425,13 +429,19 @@ static int ocfs2_mknod_locked(struct ocf
fe->i_blkno = cpu_to_le64(fe_blkno);
fe->i_suballoc_bit = cpu_to_le16(suballoc_bit);
fe->i_suballoc_slot = cpu_to_le16(osb->slot_num);
@@ -7538,26 +6973,16 @@
fe->i_mode = cpu_to_le16(mode);
if (S_ISCHR(mode) || S_ISBLK(mode))
fe->id1.dev1.i_rdev = cpu_to_le64(huge_encode_dev(dev));
-@@ -2316,5 +2326,6 @@ const struct inode_operations ocfs2_dir_
+@@ -1917,5 +1927,6 @@ const struct inode_operations ocfs2_dir_
.rename = ocfs2_rename,
.setattr = ocfs2_setattr,
.getattr = ocfs2_getattr,
+ .sync_flags = ocfs2_sync_flags,
.permission = ocfs2_permission,
};
---- linux-2.6.22/fs/ocfs2/ocfs2.h 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/ocfs2.h 2007-06-15 02:37:03 +0200
-@@ -170,6 +170,7 @@ enum ocfs2_mount_options
- OCFS2_MOUNT_NOINTR = 1 << 2, /* Don't catch signals */
- OCFS2_MOUNT_ERRORS_PANIC = 1 << 3, /* Panic on errors */
- OCFS2_MOUNT_DATA_WRITEBACK = 1 << 4, /* No data ordering */
-+ OCFS2_MOUNT_TAGGED = 1 << 8, /* use tagging */
- };
-
- #define OCFS2_OSB_SOFT_RO 0x0001
---- linux-2.6.22/fs/ocfs2/ocfs2_fs.h 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/ocfs2_fs.h 2007-06-15 02:37:03 +0200
-@@ -152,8 +152,12 @@
+--- linux-2.6.25.4/fs/ocfs2/ocfs2_fs.h 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/ocfs2_fs.h 2008-04-19 15:14:52.000000000 -0400
+@@ -188,8 +188,12 @@
#define OCFS2_NOATIME_FL (0x00000080) /* do not update atime */
#define OCFS2_DIRSYNC_FL (0x00010000) /* dirsync behaviour (directories only) */
@@ -7570,33 +6995,41 @@
/*
* Extent record flags (e_node.leaf.flags)
---- linux-2.6.22/fs/ocfs2/super.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/ocfs2/super.c 2007-06-15 02:37:03 +0200
-@@ -140,6 +140,7 @@ enum {
- Opt_data_ordered,
- Opt_data_writeback,
- Opt_atime_quantum,
+--- linux-2.6.25.4/fs/ocfs2/ocfs2.h 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/ocfs2.h 2008-04-21 10:15:31.000000000 -0400
+@@ -171,6 +171,7 @@ enum ocfs2_mount_options
+ OCFS2_MOUNT_ERRORS_PANIC = 1 << 3, /* Panic on errors */
+ OCFS2_MOUNT_DATA_WRITEBACK = 1 << 4, /* No data ordering */
+ OCFS2_MOUNT_LOCALFLOCKS = 1 << 5, /* No cluster aware user file locks */
++ OCFS2_MOUNT_TAGGED = 1 << 8, /* use tagging */
+ };
+
+ #define OCFS2_OSB_SOFT_RO 0x0001
+--- linux-2.6.25.4/fs/ocfs2/super.c 2008-04-17 12:05:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/ocfs2/super.c 2008-04-23 08:31:10.000000000 -0400
+@@ -154,6 +154,7 @@ enum {
+ Opt_commit,
+ Opt_localalloc,
+ Opt_localflocks,
+ Opt_tag, Opt_notag, Opt_tagid,
Opt_err,
};
-@@ -154,6 +155,10 @@ static match_table_t tokens = {
- {Opt_data_ordered, "data=ordered"},
- {Opt_data_writeback, "data=writeback"},
- {Opt_atime_quantum, "atime_quantum=%u"},
+@@ -172,6 +173,9 @@ static match_table_t tokens = {
+ {Opt_commit, "commit=%u"},
+ {Opt_localalloc, "localalloc=%d"},
+ {Opt_localflocks, "localflocks"},
+ {Opt_tag, "tag"},
-+ {Opt_tag, "tagxid"},
+ {Opt_notag, "notag"},
+ {Opt_tagid, "tagid=%u"},
{Opt_err, NULL}
};
-@@ -362,6 +367,14 @@ static int ocfs2_remount(struct super_bl
+@@ -391,6 +395,13 @@ static int ocfs2_remount(struct super_bl
goto out;
}
-+ printk("ocfs2_remount: %lx,%lx\n", osb->s_mount_opt, sb->s_flags);
-+ if ((parsed_options & OCFS2_MOUNT_TAGGED) &&
++ if ((parsed_options.mount_opt & OCFS2_MOUNT_TAGGED) &&
+ !(sb->s_flags & MS_TAGGED)) {
+ ret = -EINVAL;
+ mlog(ML_ERROR, "Cannot change tagging on remount\n");
@@ -7604,9 +7037,9 @@
+ }
+
if ((osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL) !=
- (parsed_options & OCFS2_MOUNT_HB_LOCAL)) {
+ (parsed_options.mount_opt & OCFS2_MOUNT_HB_LOCAL)) {
ret = -EINVAL;
-@@ -654,6 +667,9 @@ static int ocfs2_fill_super(struct super
+@@ -691,6 +702,9 @@ static int ocfs2_fill_super(struct super
ocfs2_complete_mount_recovery(osb);
@@ -7616,33 +7049,33 @@
if (ocfs2_mount_local(osb))
snprintf(nodestr, sizeof(nodestr), "local");
else
-@@ -782,6 +798,20 @@ static int ocfs2_parse_options(struct su
- else
- osb->s_atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM;
+@@ -864,6 +878,20 @@ static int ocfs2_parse_options(struct su
+ if (!is_remount)
+ mopt->mount_opt |= OCFS2_MOUNT_LOCALFLOCKS;
break;
+#ifndef CONFIG_TAGGING_NONE
+ case Opt_tag:
-+ *mount_opt |= OCFS2_MOUNT_TAGGED;
++ mopt->mount_opt |= OCFS2_MOUNT_TAGGED;
+ break;
+ case Opt_notag:
-+ *mount_opt &= ~OCFS2_MOUNT_TAGGED;
++ mopt->mount_opt &= ~OCFS2_MOUNT_TAGGED;
+ break;
+#endif
+#ifdef CONFIG_PROPAGATE
+ case Opt_tagid:
+ /* use args[0] */
-+ *mount_opt |= OCFS2_MOUNT_TAGGED;
++ mopt->mount_opt |= OCFS2_MOUNT_TAGGED;
+ break;
+#endif
default:
mlog(ML_ERROR,
"Unrecognized mount option \"%s\" "
---- linux-2.6.22/fs/open.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/open.c 2007-06-15 02:37:03 +0200
-@@ -26,22 +26,31 @@
- #include <linux/syscalls.h>
+--- linux-2.6.25.4/fs/open.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/open.c 2008-04-21 13:51:52.000000000 -0400
+@@ -27,22 +27,31 @@
#include <linux/rcupdate.h>
#include <linux/audit.h>
+ #include <linux/falloc.h>
+#include <linux/vs_base.h>
+#include <linux/vs_limit.h>
+#include <linux/vs_dlimit.h>
@@ -7673,26 +7106,26 @@
}
return retval;
}
-@@ -248,7 +257,7 @@ static long do_sys_truncate(const char _
+@@ -249,7 +258,7 @@ static long do_sys_truncate(const char _
goto dput_and_out;
error = -EROFS;
- if (IS_RDONLY(inode))
-+ if (IS_RDONLY(inode) || MNT_IS_RDONLY(nd.mnt))
++ if (IS_RDONLY(inode) || MNT_IS_RDONLY(nd.path.mnt))
goto dput_and_out;
error = -EPERM;
-@@ -397,7 +406,7 @@ asmlinkage long sys_faccessat(int dfd, c
- special_file(nd.dentry->d_inode->i_mode))
+@@ -458,7 +467,7 @@ asmlinkage long sys_faccessat(int dfd, c
+ special_file(nd.path.dentry->d_inode->i_mode))
goto out_path_release;
-- if(IS_RDONLY(nd.dentry->d_inode))
-+ if(IS_RDONLY(nd.dentry->d_inode) || MNT_IS_RDONLY(nd.mnt))
+- if(IS_RDONLY(nd.path.dentry->d_inode))
++ if(IS_RDONLY(nd.path.dentry->d_inode) || MNT_IS_RDONLY(nd.path.mnt))
res = -EROFS;
out_path_release:
-@@ -511,7 +520,7 @@ asmlinkage long sys_fchmod(unsigned int
- audit_inode(NULL, inode);
+@@ -568,7 +577,7 @@ asmlinkage long sys_fchmod(unsigned int
+ audit_inode(NULL, dentry);
err = -EROFS;
- if (IS_RDONLY(inode))
@@ -7700,22 +7133,23 @@
goto out_putf;
err = -EPERM;
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-@@ -541,11 +550,11 @@ asmlinkage long sys_fchmodat(int dfd, co
+@@ -598,11 +607,11 @@ asmlinkage long sys_fchmodat(int dfd, co
error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd);
if (error)
goto out;
-- inode = nd.dentry->d_inode;
-
+- inode = nd.path.dentry->d_inode;
+-
- error = -EROFS;
- if (IS_RDONLY(inode))
++
+ error = cow_check_and_break(&nd);
+ if (error)
goto dput_and_out;
-+ inode = nd.dentry->d_inode;
++ inode = nd.path.dentry->d_inode;
error = -EPERM;
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-@@ -570,7 +579,8 @@ asmlinkage long sys_chmod(const char __u
+@@ -627,7 +636,8 @@ asmlinkage long sys_chmod(const char __u
return sys_fchmodat(AT_FDCWD, filename, mode);
}
@@ -7725,7 +7159,7 @@
{
struct inode * inode;
int error;
-@@ -582,7 +592,7 @@ static int chown_common(struct dentry *
+@@ -639,7 +649,7 @@ static int chown_common(struct dentry *
goto out;
}
error = -EROFS;
@@ -7734,7 +7168,7 @@
goto out;
error = -EPERM;
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-@@ -590,11 +600,11 @@ static int chown_common(struct dentry *
+@@ -647,11 +657,11 @@ static int chown_common(struct dentry *
newattrs.ia_valid = ATTR_CTIME;
if (user != (uid_t) -1) {
newattrs.ia_valid |= ATTR_UID;
@@ -7747,83 +7181,83 @@
+ newattrs.ia_gid = dx_map_gid(group);
}
if (!S_ISDIR(inode->i_mode))
- newattrs.ia_valid |= ATTR_KILL_SUID|ATTR_KILL_SGID;
-@@ -613,7 +623,11 @@ asmlinkage long sys_chown(const char __u
+ newattrs.ia_valid |=
+@@ -671,7 +681,11 @@ asmlinkage long sys_chown(const char __u
error = user_path_walk(filename, &nd);
if (error)
goto out;
-- error = chown_common(nd.dentry, user, group);
+- error = chown_common(nd.path.dentry, user, group);
+#ifdef CONFIG_VSERVER_COWBL
+ error = cow_check_and_break(&nd);
+ if (!error)
+#endif
-+ error = chown_common(nd.dentry, nd.mnt, user, group);
- path_release(&nd);
++ error = chown_common(nd.path.dentry, nd.path.mnt, user, group);
+ path_put(&nd.path);
out:
return error;
-@@ -633,7 +647,11 @@ asmlinkage long sys_fchownat(int dfd, co
+@@ -691,7 +705,11 @@ asmlinkage long sys_fchownat(int dfd, co
error = __user_walk_fd(dfd, filename, follow, &nd);
if (error)
goto out;
-- error = chown_common(nd.dentry, user, group);
+- error = chown_common(nd.path.dentry, user, group);
+#ifdef CONFIG_VSERVER_COWBL
+ error = cow_check_and_break(&nd);
+ if (!error)
+#endif
-+ error = chown_common(nd.dentry, nd.mnt, user, group);
- path_release(&nd);
++ error = chown_common(nd.path.dentry, nd.path.mnt, user, group);
+ path_put(&nd.path);
out:
return error;
-@@ -647,7 +665,11 @@ asmlinkage long sys_lchown(const char __
+@@ -705,7 +723,11 @@ asmlinkage long sys_lchown(const char __
error = user_path_walk_link(filename, &nd);
if (error)
goto out;
-- error = chown_common(nd.dentry, user, group);
+- error = chown_common(nd.path.dentry, user, group);
+#ifdef CONFIG_VSERVER_COWBL
+ error = cow_check_and_break(&nd);
+ if (!error)
+#endif
-+ error = chown_common(nd.dentry, nd.mnt, user, group);
- path_release(&nd);
++ error = chown_common(nd.path.dentry, nd.path.mnt, user, group);
+ path_put(&nd.path);
out:
return error;
-@@ -666,7 +688,7 @@ asmlinkage long sys_fchown(unsigned int
+@@ -724,7 +746,7 @@ asmlinkage long sys_fchown(unsigned int
dentry = file->f_path.dentry;
- audit_inode(NULL, dentry->d_inode);
+ audit_inode(NULL, dentry);
- error = chown_common(dentry, user, group);
+ error = chown_common(dentry, file->f_vfsmnt, user, group);
fput(file);
out:
return error;
-@@ -893,6 +915,7 @@ repeat:
- FD_SET(fd, fdt->open_fds);
- FD_CLR(fd, fdt->close_on_exec);
+@@ -971,6 +993,7 @@ repeat:
+ else
+ FD_CLR(fd, fdt->close_on_exec);
files->next_fd = fd + 1;
+ vx_openfd_inc(fd);
#if 1
/* Sanity check */
if (fdt->fd[fd] != NULL) {
-@@ -915,6 +938,7 @@ static void __put_unused_fd(struct files
+@@ -998,6 +1021,7 @@ static void __put_unused_fd(struct files
__FD_CLR(fd, fdt->open_fds);
if (fd < files->next_fd)
files->next_fd = fd;
+ vx_openfd_dec(fd);
}
- void fastcall put_unused_fd(unsigned int fd)
---- linux-2.6.22/fs/proc/array.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/proc/array.c 2007-06-15 03:04:38 +0200
-@@ -75,6 +75,8 @@
- #include <linux/cpuset.h>
- #include <linux/rcupdate.h>
+ void put_unused_fd(unsigned int fd)
+--- linux-2.6.25.4/fs/proc/array.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/proc/array.c 2008-05-21 14:08:19.000000000 -0400
+@@ -79,6 +79,8 @@
#include <linux/delayacct.h>
+ #include <linux/seq_file.h>
+ #include <linux/pid_namespace.h>
+#include <linux/vs_context.h>
+#include <linux/vs_network.h>
- #include <asm/uaccess.h>
#include <asm/pgtable.h>
-@@ -134,8 +136,9 @@ static const char *task_state_array[] =
+ #include <asm/processor.h>
+@@ -140,8 +142,9 @@ static const char *task_state_array[] =
"D (disk sleep)", /* 2 */
"T (stopped)", /* 4 */
"T (tracing stop)", /* 8 */
@@ -7834,148 +7268,112 @@
+ "X (dead)", /* 64 */
};
- static inline const char * get_task_state(struct task_struct *tsk)
-@@ -144,7 +147,8 @@ static inline const char * get_task_stat
- TASK_INTERRUPTIBLE |
- TASK_UNINTERRUPTIBLE |
- TASK_STOPPED |
-- TASK_TRACED)) |
-+ TASK_TRACED |
-+ TASK_ONHOLD)) |
- (tsk->exit_state & (EXIT_ZOMBIE |
- EXIT_DEAD));
- const char **p = &task_state_array[0];
-@@ -161,8 +165,16 @@ static inline char * task_state(struct t
+ static inline const char *get_task_state(struct task_struct *tsk)
+@@ -162,6 +165,7 @@ static inline void task_state(struct seq
struct group_info *group_info;
int g;
struct fdtable *fdt = NULL;
-+ pid_t pid, ptgid, tppid, tgid;
++/* + pid_t pid, ptgid, tppid, tgid; */
+ pid_t ppid, tpid;
rcu_read_lock();
-+ tgid = vx_map_tgid(p->tgid);
+@@ -169,6 +173,12 @@ static inline void task_state(struct seq
+ task_tgid_nr_ns(rcu_dereference(p->real_parent), ns) : 0;
+ tpid = pid_alive(p) && p->ptrace ?
+ task_pid_nr_ns(rcu_dereference(p->parent), ns) : 0;
++/* tgid = vx_map_tgid(p->tgid);
+ pid = vx_map_pid(p->pid);
+ ptgid = vx_map_pid(pid_alive(p) ?
+ rcu_dereference(p->real_parent)->tgid : 0);
+ tppid = vx_map_pid(pid_alive(p) && p->ptrace ?
-+ rcu_dereference(p->parent)->pid : 0);
-+
- buffer += sprintf(buffer,
++ rcu_dereference(p->parent)->pid : 0); */
+ seq_printf(m,
"State:\t%s\n"
- "SleepAVG:\t%lu%%\n"
-@@ -174,9 +186,7 @@ static inline char * task_state(struct t
- "Gid:\t%d\t%d\t%d\t%d\n",
- get_task_state(p),
- (p->sleep_avg/1024)*100/(1020000000/1024),
-- p->tgid, p->pid,
-- pid_alive(p) ? rcu_dereference(p->real_parent)->tgid : 0,
-- pid_alive(p) && p->ptrace ? rcu_dereference(p->parent)->pid : 0,
-+ tgid, pid, (pid > 1) ? ptgid : 0, tppid,
- p->uid, p->euid, p->suid, p->fsuid,
- p->gid, p->egid, p->sgid, p->fsgid);
-
-@@ -283,17 +293,26 @@ static inline char * task_sig(struct tas
-
- static inline char *task_cap(struct task_struct *p, char *buffer)
- {
-- return buffer + sprintf(buffer, "CapInh:\t%016x\n"
-- "CapPrm:\t%016x\n"
-- "CapEff:\t%016x\n",
-- cap_t(p->cap_inheritable),
-- cap_t(p->cap_permitted),
-- cap_t(p->cap_effective));
-+ struct vx_info *vxi = p->vx_info;
-+
-+ return buffer + sprintf(buffer,
-+ "CapInh:\t%016x\n"
-+ "CapPrm:\t%016x\n"
-+ "CapEff:\t%016x\n",
-+ (unsigned)vx_info_mbcap(vxi, p->cap_inheritable),
-+ (unsigned)vx_info_mbcap(vxi, p->cap_permitted),
-+ (unsigned)vx_info_mbcap(vxi, p->cap_effective));
+ "Tgid:\t%d\n"
+@@ -280,23 +290,23 @@ static inline void task_sig(struct seq_f
}
- int proc_pid_status(struct task_struct *task, char * buffer)
- {
- char * orig = buffer;
-+#ifdef CONFIG_VSERVER_LEGACY
-+ struct vx_info *vxi;
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACYNET
-+ struct nx_info *nxi;
-+#endif
- struct mm_struct *mm = get_task_mm(task);
-
- buffer = task_name(task, buffer);
-@@ -306,6 +325,46 @@ int proc_pid_status(struct task_struct *
- buffer = task_sig(task, buffer);
- buffer = task_cap(task, buffer);
- buffer = cpuset_task_status_allowed(task, buffer);
+ static void render_cap_t(struct seq_file *m, const char *header,
+- kernel_cap_t *a)
++ struct vx_info *vxi, kernel_cap_t *a)
+ {
+ unsigned __capi;
+
+ seq_printf(m, "%s", header);
+ CAP_FOR_EACH_U32(__capi) {
+ seq_printf(m, "%08x",
+- a->cap[(_LINUX_CAPABILITY_U32S-1) - __capi]);
++ a->cap[(_LINUX_CAPABILITY_U32S-1) - __capi]);
+ }
+ seq_printf(m, "\n");
+ }
+
+ static inline void task_cap(struct seq_file *m, struct task_struct *p)
+ {
+- render_cap_t(m, "CapInh:\t", &p->cap_inheritable);
+- render_cap_t(m, "CapPrm:\t", &p->cap_permitted);
+- render_cap_t(m, "CapEff:\t", &p->cap_effective);
++ render_cap_t(m, "CapInh:\t", p->vx_info, &p->cap_inheritable);
++ render_cap_t(m, "CapPrm:\t", p->vx_info, &p->cap_permitted);
++ render_cap_t(m, "CapEff:\t", p->vx_info, &p->cap_effective);
+ }
+
+ static inline void task_context_switch_counts(struct seq_file *m,
+@@ -308,6 +318,45 @@ static inline void task_context_switch_c
+ p->nivcsw);
+ }
+
++int proc_pid_nsproxy(struct seq_file *m, struct pid_namespace *ns,
++ struct pid *pid, struct task_struct *task)
++{
++ seq_printf(m, "Proxy:\t%p(%c)\n"
++ "Count:\t%u\n"
++ "uts:\t%p(%c)\n"
++ "ipc:\t%p(%c)\n"
++ "mnt:\t%p(%c)\n"
++ "pid:\t%p(%c)\n"
++ "user:\t%p(%c)\n"
++ "net:\t%p(%c)\n",
++ task->nsproxy,
++ (task->nsproxy == init_task.nsproxy ? 'I' : '-'),
++ atomic_read(&task->nsproxy->count),
++ task->nsproxy->uts_ns,
++ (task->nsproxy->uts_ns == init_task.nsproxy->uts_ns ? 'I' : '-'),
++ task->nsproxy->ipc_ns,
++ (task->nsproxy->ipc_ns == init_task.nsproxy->ipc_ns ? 'I' : '-'),
++ task->nsproxy->mnt_ns,
++ (task->nsproxy->mnt_ns == init_task.nsproxy->mnt_ns ? 'I' : '-'),
++ task->nsproxy->pid_ns,
++ (task->nsproxy->pid_ns == init_task.nsproxy->pid_ns ? 'I' : '-'),
++ task->nsproxy->user_ns,
++ (task->nsproxy->user_ns == init_task.nsproxy->user_ns ? 'I' : '-'),
++ task->nsproxy->net_ns,
++ (task->nsproxy->net_ns == init_task.nsproxy->net_ns ? 'I' : '-'));
++ return 0;
++}
+
++void task_vs_id(struct seq_file *m, struct task_struct *task)
++{
+ if (task_vx_flags(task, VXF_HIDE_VINFO, 0))
-+ goto skip;
-+#ifdef CONFIG_VSERVER_LEGACY
-+ buffer += sprintf (buffer,"s_context: %d\n", vx_task_xid(task));
-+ vxi = task_get_vx_info(task);
-+ if (vxi) {
-+ buffer += sprintf (buffer,"ctxflags: %08llx\n"
-+ ,(unsigned long long)vxi->vx_flags);
-+ buffer += sprintf (buffer,"initpid: %d\n"
-+ ,vxi->vx_initpid);
-+ } else {
-+ buffer += sprintf (buffer,"ctxflags: none\n");
-+ buffer += sprintf (buffer,"initpid: none\n");
-+ }
-+ put_vx_info(vxi);
-+#else
-+ buffer += sprintf (buffer,"VxID: %d\n", vx_task_xid(task));
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACYNET
-+ nxi = task_get_nx_info(task);
-+ if (nxi) {
-+ int i;
++ return;
+
-+ buffer += sprintf (buffer,"ipv4root:");
-+ for (i=0; i<nxi->nbipv4; i++){
-+ buffer += sprintf (buffer," %08x/%08x"
-+ ,nxi->ipv4[i]
-+ ,nxi->mask[i]);
-+ }
-+ *buffer++ = '\n';
-+ buffer += sprintf (buffer,"ipv4root_bcast: %08x\n"
-+ ,nxi->v4_bcast);
-+ } else {
-+ buffer += sprintf (buffer,"ipv4root: 0\n");
-+ buffer += sprintf (buffer,"ipv4root_bcast: 0\n");
-+ }
-+ put_nx_info(nxi);
-+#endif
-+skip:
++ seq_printf(m, "VxID: %d\n", vx_task_xid(task));
++ seq_printf(m, "NxID: %d\n", nx_task_nid(task));
++}
++
++
+ int proc_pid_status(struct seq_file *m, struct pid_namespace *ns,
+ struct pid *pid, struct task_struct *task)
+ {
+@@ -323,6 +372,7 @@ int proc_pid_status(struct seq_file *m,
+ task_sig(m, task);
+ task_cap(m, task);
+ cpuset_task_status_allowed(m, task);
++ task_vs_id(m, task);
#if defined(CONFIG_S390)
- buffer = task_show_regs(task, buffer);
+ task_show_regs(m, task);
#endif
-@@ -320,7 +379,7 @@ static int do_task_stat(struct task_stru
- sigset_t sigign, sigcatch;
- char state;
- int res;
-- pid_t ppid = 0, pgid = -1, sid = -1;
-+ pid_t pid = 0, ppid = 0, pgid = -1, sid = -1;
- int num_threads = 0;
- struct mm_struct *mm;
- unsigned long long start_time;
-@@ -382,8 +441,10 @@ static int do_task_stat(struct task_stru
- }
-
- sid = signal_session(sig);
-- pgid = process_group(task);
-- ppid = rcu_dereference(task->real_parent)->tgid;
-+ pid = vx_info_map_pid(task->vx_info, task->pid);
-+ pgid = vx_info_map_pid(task->vx_info, process_group(task));
-+ ppid = (pid > 1) ? vx_info_map_tgid(task->vx_info,
-+ rcu_dereference(task->real_parent)->tgid) : 0;
-
- unlock_task_sighand(task, &flags);
- }
-@@ -410,10 +471,21 @@ static int do_task_stat(struct task_stru
+@@ -494,6 +544,17 @@ static int do_task_stat(struct seq_file
/* convert nsec -> ticks */
start_time = nsec_to_clock_t(start_time);
@@ -7990,27 +7388,21 @@
+ start_time = 0;
+ }
+
- res = sprintf(buffer,"%d (%s) %c %d %d %d %d %d %u %lu \
+ seq_printf(m, "%d (%s) %c %d %d %d %d %d %u %lu \
%lu %lu %lu %lu %lu %ld %ld %ld %ld %d 0 %llu %lu %ld %lu %lu %lu %lu %lu \
- %lu %lu %lu %lu %lu %lu %lu %lu %d %d %u %u %llu\n",
-- task->pid,
-+ pid,
- tcomm,
- state,
- ppid,
---- linux-2.6.22/fs/proc/base.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/proc/base.c 2007-06-15 03:06:14 +0200
-@@ -73,6 +73,9 @@
- #include <linux/poll.h>
- #include <linux/nsproxy.h>
+ %lu %lu %lu %lu %lu %lu %lu %lu %d %d %u %u %llu %lu %ld\n",
+--- linux-2.6.25.4/fs/proc/base.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/proc/base.c 2008-05-21 14:08:19.000000000 -0400
+@@ -76,6 +76,8 @@
#include <linux/oom.h>
+ #include <linux/elf.h>
+ #include <linux/pid_namespace.h>
+#include <linux/vs_context.h>
+#include <linux/vs_network.h>
-+
#include "internal.h"
/* NOTE:
-@@ -1049,6 +1052,8 @@ static struct inode *proc_pid_make_inode
+@@ -1290,6 +1292,8 @@ static struct inode *proc_pid_make_inode
inode->i_uid = task->euid;
inode->i_gid = task->egid;
}
@@ -8019,7 +7411,7 @@
security_task_to_inode(task, inode);
out:
-@@ -1595,6 +1600,13 @@ static struct dentry *proc_pident_lookup
+@@ -1833,6 +1837,13 @@ static struct dentry *proc_pident_lookup
if (!task)
goto out_no_task;
@@ -8033,24 +7425,7 @@
/*
* Yes, it does not scale. And it should not. Don't add
* new entries into /proc/<tgid>/ without very good reasons.
-@@ -1790,14 +1802,14 @@ static int proc_self_readlink(struct den
- int buflen)
- {
- char tmp[PROC_NUMBUF];
-- sprintf(tmp, "%d", current->tgid);
-+ sprintf(tmp, "%d", vx_map_tgid(current->tgid));
- return vfs_readlink(dentry,buffer,buflen,tmp);
- }
-
- static void *proc_self_follow_link(struct dentry *dentry, struct nameidata *nd)
- {
- char tmp[PROC_NUMBUF];
-- sprintf(tmp, "%d", current->tgid);
-+ sprintf(tmp, "%d", vx_map_tgid(current->tgid));
- return ERR_PTR(vfs_follow_link(nd,tmp));
- }
-
-@@ -1891,7 +1903,7 @@ out_iput:
+@@ -2220,7 +2231,7 @@ out_iput:
static struct dentry *proc_base_lookup(struct inode *dir, struct dentry *dentry)
{
struct dentry *error;
@@ -8059,7 +7434,7 @@
const struct pid_entry *p, *last;
error = ERR_PTR(-ENOENT);
-@@ -1956,6 +1968,9 @@ static int proc_pid_io_accounting(struct
+@@ -2285,6 +2296,9 @@ static int proc_pid_io_accounting(struct
static const struct file_operations proc_task_operations;
static const struct inode_operations proc_task_inode_operations;
@@ -8069,93 +7444,97 @@
static const struct pid_entry tgid_base_stuff[] = {
DIR("task", S_IRUGO|S_IXUGO, task),
DIR("fd", S_IRUSR|S_IXUSR, fd),
-@@ -1995,6 +2010,8 @@ static const struct pid_entry tgid_base_
- #ifdef CONFIG_CPUSETS
- REG("cpuset", S_IRUGO, cpuset),
+@@ -2335,6 +2349,8 @@ static const struct pid_entry tgid_base_
+ #ifdef CONFIG_CGROUPS
+ REG("cgroup", S_IRUGO, cgroup),
#endif
+ INF("vinfo", S_IRUGO, pid_vx_info),
+ INF("ninfo", S_IRUGO, pid_nx_info),
INF("oom_score", S_IRUGO, oom_score),
REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust),
#ifdef CONFIG_AUDITSYSCALL
-@@ -2143,7 +2160,7 @@ struct dentry *proc_pid_lookup(struct in
- goto out;
+@@ -2350,6 +2366,7 @@ static const struct pid_entry tgid_base_
+ #ifdef CONFIG_TASK_IO_ACCOUNTING
+ INF("io", S_IRUGO, pid_io_accounting),
+ #endif
++ ONE("nsproxy", S_IRUGO, pid_nsproxy),
+ };
- rcu_read_lock();
-- task = find_task_by_pid(tgid);
-+ task = vx_find_proc_task_by_pid(tgid);
- if (task)
- get_task_struct(task);
- rcu_read_unlock();
-@@ -2207,7 +2224,7 @@ static int proc_pid_fill_cache(struct fi
+ static int proc_tgid_base_readdir(struct file * filp,
+@@ -2547,7 +2564,7 @@ retry:
+ iter.task = NULL;
+ pid = find_ge_pid(iter.tgid, ns);
+ if (pid) {
+- iter.tgid = pid_nr_ns(pid, ns);
++ iter.tgid = pid_unmapped_nr_ns(pid, ns);
+ iter.task = pid_task(pid, PIDTYPE_PID);
+ /* What we to know is if the pid we have find is the
+ * pid of a thread_group_leader. Testing for task
+@@ -2577,7 +2594,7 @@ static int proc_pid_fill_cache(struct fi
+ struct tgid_iter iter)
+ {
+ char name[PROC_NUMBUF];
+- int len = snprintf(name, sizeof(name), "%d", iter.tgid);
++ int len = snprintf(name, sizeof(name), "%d", vx_map_tgid(iter.tgid));
+ return proc_fill_cache(filp, dirent, filldir, name, len,
+ proc_pid_instantiate, iter.task, NULL);
+ }
+@@ -2586,7 +2603,7 @@ static int proc_pid_fill_cache(struct fi
int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir)
{
unsigned int nr = filp->f_pos - FIRST_PROCESS_ENTRY;
- struct task_struct *reaper = get_proc_task(filp->f_path.dentry->d_inode);
+ struct task_struct *reaper = get_proc_task_real(filp->f_path.dentry->d_inode);
- struct task_struct *task;
- int tgid;
+ struct tgid_iter iter;
+ struct pid_namespace *ns;
-@@ -2226,6 +2243,8 @@ int proc_pid_readdir(struct file * filp,
- put_task_struct(task), task = next_tgid(tgid + 1)) {
- tgid = task->pid;
- filp->f_pos = tgid + TGID_OFFSET;
-+ if (!vx_proc_task_visible(task))
+@@ -2606,6 +2623,8 @@ int proc_pid_readdir(struct file * filp,
+ iter.task;
+ iter.tgid += 1, iter = next_tgid(ns, iter)) {
+ filp->f_pos = iter.tgid + TGID_OFFSET;
++ if (!vx_proc_task_visible(iter.task))
+ continue;
- if (proc_pid_fill_cache(filp, dirent, filldir, task, tgid) < 0) {
- put_task_struct(task);
+ if (proc_pid_fill_cache(filp, dirent, filldir, iter) < 0) {
+ put_task_struct(iter.task);
goto out;
-@@ -2352,9 +2371,11 @@ static struct dentry *proc_task_lookup(s
+@@ -2742,6 +2761,8 @@ static struct dentry *proc_task_lookup(s
tid = name_to_int(dentry);
if (tid == ~0U)
goto out;
+ if (vx_current_initpid(tid))
+ goto out;
+ ns = dentry->d_sb->s_fs_info;
rcu_read_lock();
-- task = find_task_by_pid(tid);
-+ task = vx_find_proc_task_by_pid(tid);
- if (task)
- get_task_struct(task);
- rcu_read_unlock();
-@@ -2499,7 +2520,7 @@ static int proc_task_readdir(struct file
- for (task = first_tid(leader, tid, pos - 2);
- task;
- task = next_tid(task), pos++) {
-- tid = task->pid;
-+ tid = vx_map_pid(task->pid);
- if (proc_task_fill_cache(filp, dirent, filldir, task, tid) < 0) {
- /* returning this tgid failed, save it as the first
- * pid for the next readir call */
---- linux-2.6.22/fs/proc/generic.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/proc/generic.c 2007-06-15 03:08:54 +0200
-@@ -20,6 +20,7 @@
- #include <linux/namei.h>
+--- linux-2.6.25.4/fs/proc/generic.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/proc/generic.c 2008-04-20 14:23:26.000000000 -0400
+@@ -21,6 +21,7 @@
#include <linux/bitops.h>
#include <linux/spinlock.h>
+ #include <linux/completion.h>
+#include <linux/vserver/inode.h>
#include <asm/uaccess.h>
#include "internal.h"
-@@ -395,6 +396,8 @@ struct dentry *proc_lookup(struct inode
+@@ -389,6 +390,8 @@ struct dentry *proc_lookup_de(struct pro
for (de = de->subdir; de ; de = de->next) {
if (de->namelen != dentry->d_name.len)
continue;
+ if (!vx_hide_check(0, de->vx_flags))
+ continue;
if (!memcmp(dentry->d_name.name, de->name, de->namelen)) {
- unsigned int ino = de->low_ino;
+ unsigned int ino;
-@@ -402,6 +405,8 @@ struct dentry *proc_lookup(struct inode
+@@ -397,6 +400,8 @@ struct dentry *proc_lookup_de(struct pro
spin_unlock(&proc_subdir_lock);
error = -EINVAL;
inode = proc_get_inode(dir->i_sb, ino, de);
+ /* generic proc entries belong to the host */
+ inode->i_tag = 0;
- spin_lock(&proc_subdir_lock);
- break;
+ goto out_unlock;
}
-@@ -482,6 +487,8 @@ int proc_readdir(struct file * filp,
+ }
+@@ -481,6 +486,8 @@ int proc_readdir_de(struct proc_dir_entr
/* filldir passes info to user space */
de_get(de);
@@ -8164,7 +7543,7 @@
spin_unlock(&proc_subdir_lock);
if (filldir(dirent, de->name, de->namelen, filp->f_pos,
de->low_ino, de->mode >> 12) < 0) {
-@@ -489,6 +496,7 @@ int proc_readdir(struct file * filp,
+@@ -488,6 +495,7 @@ int proc_readdir_de(struct proc_dir_entr
goto out;
}
spin_lock(&proc_subdir_lock);
@@ -8172,15 +7551,15 @@
filp->f_pos++;
next = de->next;
de_put(de);
-@@ -613,6 +621,7 @@ static struct proc_dir_entry *proc_creat
- ent->namelen = len;
- ent->mode = mode;
+@@ -602,6 +610,7 @@ static struct proc_dir_entry *__proc_cre
ent->nlink = nlink;
+ atomic_set(&ent->count, 1);
+ ent->pde_users = 0;
+ ent->vx_flags = IATTR_PROC_DEFAULT;
+ spin_lock_init(&ent->pde_unload_lock);
+ ent->pde_unload_completion = NULL;
out:
- return ent;
- }
-@@ -633,7 +642,8 @@ struct proc_dir_entry *proc_symlink(cons
+@@ -624,7 +633,8 @@ struct proc_dir_entry *proc_symlink(cons
kfree(ent->data);
kfree(ent);
ent = NULL;
@@ -8190,19 +7569,19 @@
} else {
kfree(ent);
ent = NULL;
---- linux-2.6.22/fs/proc/inode.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/proc/inode.c 2007-06-15 02:37:03 +0200
-@@ -160,6 +160,8 @@ struct inode *proc_get_inode(struct supe
- inode->i_uid = de->uid;
- inode->i_gid = de->gid;
- }
+--- linux-2.6.25.4/fs/proc/inode.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/proc/inode.c 2008-04-19 15:14:52.000000000 -0400
+@@ -408,6 +408,8 @@ struct inode *proc_get_inode(struct supe
+ inode->i_uid = de->uid;
+ inode->i_gid = de->gid;
+ }
+ if (de->vx_flags)
+ PROC_I(inode)->vx_flags = de->vx_flags;
- if (de->size)
- inode->i_size = de->size;
- if (de->nlink)
---- linux-2.6.22/fs/proc/internal.h 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/proc/internal.h 2007-06-15 02:37:03 +0200
+ if (de->size)
+ inode->i_size = de->size;
+ if (de->nlink)
+--- linux-2.6.25.4/fs/proc/internal.h 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/proc/internal.h 2008-05-21 14:08:19.000000000 -0400
@@ -10,6 +10,7 @@
*/
@@ -8211,7 +7590,17 @@
#ifdef CONFIG_PROC_SYSCTL
extern int proc_sys_init(void);
-@@ -64,11 +65,16 @@ static inline struct pid *proc_pid(struc
+@@ -57,6 +58,9 @@ extern int proc_pid_status(struct seq_fi
+ struct pid *pid, struct task_struct *task);
+ extern int proc_pid_statm(struct seq_file *m, struct pid_namespace *ns,
+ struct pid *pid, struct task_struct *task);
++extern int proc_pid_nsproxy(struct seq_file *m, struct pid_namespace *ns,
++ struct pid *pid, struct task_struct *task);
++
+ extern loff_t mem_lseek(struct file *file, loff_t offset, int orig);
+
+ extern const struct file_operations proc_maps_operations;
+@@ -76,11 +80,16 @@ static inline struct pid *proc_pid(struc
return PROC_I(inode)->pid;
}
@@ -8229,9 +7618,9 @@
static inline int proc_fd(struct inode *inode)
{
return PROC_I(inode)->fd;
---- linux-2.6.22/fs/proc/proc_misc.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/proc/proc_misc.c 2007-06-15 02:37:03 +0200
-@@ -53,6 +53,8 @@
+--- linux-2.6.25.4/fs/proc/proc_misc.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/proc/proc_misc.c 2008-04-23 11:56:24.000000000 -0400
+@@ -56,6 +56,8 @@
#include <asm/div64.h>
#include "internal.h"
@@ -8240,47 +7629,52 @@
#define LOAD_INT(x) ((x) >> FSHIFT)
#define LOAD_FRAC(x) LOAD_INT(((x) & (FIXED_1-1)) * 100)
/*
-@@ -82,17 +84,32 @@ static int proc_calc_metrics(char *page,
+@@ -83,22 +85,38 @@ static int proc_calc_metrics(char *page,
static int loadavg_read_proc(char *page, char **start, off_t off,
int count, int *eof, void *data)
{
-+ unsigned int running, threads;
++ unsigned long running;
++ unsigned int threads;
int a, b, c;
int len;
+ unsigned long seq;
-- a = avenrun[0] + (FIXED_1/200);
-- b = avenrun[1] + (FIXED_1/200);
-- c = avenrun[2] + (FIXED_1/200);
-- len = sprintf(page,"%d.%02d %d.%02d %d.%02d %ld/%d %d\n",
-+ if (vx_flags(VXF_VIRT_LOAD, 0)) {
-+ struct vx_info *vxi = current->vx_info;
-+
-+ a = vxi->cvirt.load[0] + (FIXED_1/200);
-+ b = vxi->cvirt.load[1] + (FIXED_1/200);
-+ c = vxi->cvirt.load[2] + (FIXED_1/200);
+ do {
+ seq = read_seqbegin(&xtime_lock);
+- a = avenrun[0] + (FIXED_1/200);
+- b = avenrun[1] + (FIXED_1/200);
+- c = avenrun[2] + (FIXED_1/200);
++ if (vx_flags(VXF_VIRT_LOAD, 0)) {
++ struct vx_info *vxi = current->vx_info;
++
++ a = vxi->cvirt.load[0] + (FIXED_1/200);
++ b = vxi->cvirt.load[1] + (FIXED_1/200);
++ c = vxi->cvirt.load[2] + (FIXED_1/200);
+
-+ running = atomic_read(&vxi->cvirt.nr_running);
-+ threads = atomic_read(&vxi->cvirt.nr_threads);
-+ } else {
-+ a = avenrun[0] + (FIXED_1/200);
-+ b = avenrun[1] + (FIXED_1/200);
-+ c = avenrun[2] + (FIXED_1/200);
++ running = atomic_read(&vxi->cvirt.nr_running);
++ threads = atomic_read(&vxi->cvirt.nr_threads);
++ } else {
++ a = avenrun[0] + (FIXED_1/200);
++ b = avenrun[1] + (FIXED_1/200);
++ c = avenrun[2] + (FIXED_1/200);
+
-+ running = nr_running();
-+ threads = nr_threads;
-+ }
-+ len = sprintf(page,"%d.%02d %d.%02d %d.%02d %d/%d %d\n",
++ running = nr_running();
++ threads = nr_threads;
++ }
+ } while (read_seqretry(&xtime_lock, seq));
+
+ len = sprintf(page,"%d.%02d %d.%02d %d.%02d %ld/%d %d\n",
LOAD_INT(a), LOAD_FRAC(a),
LOAD_INT(b), LOAD_FRAC(b),
LOAD_INT(c), LOAD_FRAC(c),
-- nr_running(), nr_threads, current->nsproxy->pid_ns->last_pid);
-+ running, threads, current->nsproxy->pid_ns->last_pid);
+- nr_running(), nr_threads,
++ running, threads,
+ task_active_pid_ns(current)->last_pid);
return proc_calc_metrics(page, start, off, count, eof, len);
}
-
-@@ -106,6 +123,9 @@ static int uptime_read_proc(char *page,
-
+@@ -114,6 +132,9 @@ static int uptime_read_proc(char *page,
do_posix_clock_monotonic_gettime(&uptime);
+ monotonic_to_bootbased(&uptime);
cputime_to_timespec(idletime, &idle);
+ if (vx_flags(VXF_VIRT_UPTIME, 0))
+ vx_vsi_uptime(&uptime, &idle);
@@ -8288,7 +7682,7 @@
len = sprintf(page,"%lu.%02lu %lu.%02lu\n",
(unsigned long) uptime.tv_sec,
(uptime.tv_nsec / (NSEC_PER_SEC / 100)),
-@@ -137,7 +157,7 @@ static int meminfo_read_proc(char *page,
+@@ -145,7 +166,7 @@ static int meminfo_read_proc(char *page,
cached = global_page_state(NR_FILE_PAGES) -
total_swapcache_pages - i.bufferram;
@@ -8297,19 +7691,19 @@
cached = 0;
get_vmalloc_info(&vmi);
---- linux-2.6.22/fs/proc/root.c 2007-05-02 19:25:21 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/proc/root.c 2007-06-15 02:37:03 +0200
-@@ -22,6 +22,9 @@
+--- linux-2.6.25.4/fs/proc/root.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/proc/root.c 2008-04-19 15:14:52.000000000 -0400
+@@ -23,6 +23,9 @@
#include "internal.h"
- struct proc_dir_entry *proc_net, *proc_net_stat, *proc_bus, *proc_root_fs, *proc_root_driver;
+ struct proc_dir_entry *proc_bus, *proc_root_fs, *proc_root_driver;
+struct proc_dir_entry *proc_virtual;
+
+extern void proc_vx_init(void);
- static int proc_get_sb(struct file_system_type *fs_type,
- int flags, const char *dev_name, void *data, struct vfsmount *mnt)
-@@ -79,6 +82,7 @@ void __init proc_root_init(void)
+ static int proc_test_super(struct super_block *sb, void *data)
+ {
+@@ -138,6 +141,7 @@ void __init proc_root_init(void)
proc_device_tree_init();
#endif
proc_bus = proc_mkdir("bus", NULL);
@@ -8317,17 +7711,17 @@
proc_sys_init();
}
---- linux-2.6.22/fs/quota.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/quota.c 2007-06-15 02:37:03 +0200
-@@ -16,6 +16,7 @@
- #include <linux/buffer_head.h>
+--- linux-2.6.25.4/fs/quota.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/quota.c 2008-04-21 09:21:04.000000000 -0400
+@@ -18,6 +18,7 @@
#include <linux/capability.h>
#include <linux/quotaops.h>
+ #include <linux/types.h>
+#include <linux/vs_context.h>
/* Check validity of generic quotactl commands */
static int generic_quotactl_valid(struct super_block *sb, int type, int cmd, qid_t id)
-@@ -80,11 +81,11 @@ static int generic_quotactl_valid(struct
+@@ -82,11 +83,11 @@ static int generic_quotactl_valid(struct
if (cmd == Q_GETQUOTA) {
if (((type == USRQUOTA && current->euid != id) ||
(type == GRPQUOTA && !in_egroup_p(id))) &&
@@ -8341,7 +7735,7 @@
return -EPERM;
return 0;
-@@ -131,10 +132,10 @@ static int xqm_quotactl_valid(struct sup
+@@ -133,10 +134,10 @@ static int xqm_quotactl_valid(struct sup
if (cmd == Q_XGETQUOTA) {
if (((type == XQM_USRQUOTA && current->euid != id) ||
(type == XQM_GRPQUOTA && !in_egroup_p(id))) &&
@@ -8354,7 +7748,7 @@
return -EPERM;
}
-@@ -327,6 +328,46 @@ static int do_quotactl(struct super_bloc
+@@ -329,6 +330,46 @@ static int do_quotactl(struct super_bloc
return 0;
}
@@ -8401,10 +7795,10 @@
/*
* look up a superblock on which quota ops will be performed
* - use the name of a block device to find the superblock thereon
-@@ -344,6 +385,22 @@ static inline struct super_block *quotac
+@@ -346,6 +387,22 @@ static inline struct super_block *quotac
putname(tmp);
if (IS_ERR(bdev))
- return ERR_PTR(PTR_ERR(bdev));
+ return ERR_CAST(bdev);
+#if defined(CONFIG_BLK_DEV_VROOT) || defined(CONFIG_BLK_DEV_VROOT_MODULE)
+ if (bdev && bdev->bd_inode &&
+ imajor(bdev->bd_inode) == VROOT_MAJOR) {
@@ -8424,160 +7818,8 @@
sb = get_super(bdev);
bdput(bdev);
if (!sb)
---- linux-2.6.22/fs/read_write.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/read_write.c 2007-06-15 02:37:03 +0200
-@@ -701,12 +701,77 @@ sys_writev(unsigned long fd, const struc
- return ret;
- }
-
-+ssize_t vfs_sendfile(struct file *out_file, struct file *in_file, loff_t *ppos,
-+ size_t count, loff_t max)
-+{
-+ struct inode * in_inode, * out_inode;
-+ loff_t pos;
-+ ssize_t ret;
-+
-+ /* verify in_file */
-+ in_inode = in_file->f_path.dentry->d_inode;
-+ if (!in_inode)
-+ return -EINVAL;
-+ if (!in_file->f_op || !in_file->f_op->sendfile)
-+ return -EINVAL;
-+
-+ if (!ppos)
-+ ppos = &in_file->f_pos;
-+ else
-+ if (!(in_file->f_mode & FMODE_PREAD))
-+ return -ESPIPE;
-+
-+ ret = rw_verify_area(READ, in_file, ppos, count);
-+ if (ret < 0)
-+ return ret;
-+ count = ret;
-+
-+ /* verify out_file */
-+ out_inode = out_file->f_path.dentry->d_inode;
-+ if (!out_inode)
-+ return -EINVAL;
-+ if (!out_file->f_op || !out_file->f_op->sendpage)
-+ return -EINVAL;
-+
-+ ret = rw_verify_area(WRITE, out_file, &out_file->f_pos, count);
-+ if (ret < 0)
-+ return ret;
-+ count = ret;
-+
-+ ret = security_file_permission (out_file, MAY_WRITE);
-+ if (ret)
-+ return ret;
-+
-+ if (!max)
-+ max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes);
-+
-+ pos = *ppos;
-+ if (unlikely(pos < 0))
-+ return -EINVAL;
-+ if (unlikely(pos + count > max)) {
-+ if (pos >= max)
-+ return -EOVERFLOW;
-+ count = max - pos;
-+ }
-+
-+ ret = in_file->f_op->sendfile(in_file, ppos, count, file_send_actor, out_file);
-+
-+ if (ret > 0) {
-+ add_rchar(current, ret);
-+ add_wchar(current, ret);
-+ }
-+
-+ if (*ppos > max)
-+ return -EOVERFLOW;
-+ return ret;
-+}
-+
-+EXPORT_SYMBOL(vfs_sendfile);
-+
- static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
- size_t count, loff_t max)
- {
- struct file * in_file, * out_file;
-- struct inode * in_inode, * out_inode;
-- loff_t pos;
- ssize_t retval;
- int fput_needed_in, fput_needed_out;
-
-@@ -719,22 +784,6 @@ static ssize_t do_sendfile(int out_fd, i
- goto out;
- if (!(in_file->f_mode & FMODE_READ))
- goto fput_in;
-- retval = -EINVAL;
-- in_inode = in_file->f_path.dentry->d_inode;
-- if (!in_inode)
-- goto fput_in;
-- if (!in_file->f_op || !in_file->f_op->sendfile)
-- goto fput_in;
-- retval = -ESPIPE;
-- if (!ppos)
-- ppos = &in_file->f_pos;
-- else
-- if (!(in_file->f_mode & FMODE_PREAD))
-- goto fput_in;
-- retval = rw_verify_area(READ, in_file, ppos, count);
-- if (retval < 0)
-- goto fput_in;
-- count = retval;
-
- retval = security_file_permission (in_file, MAY_READ);
- if (retval)
-@@ -749,45 +798,11 @@ static ssize_t do_sendfile(int out_fd, i
- goto fput_in;
- if (!(out_file->f_mode & FMODE_WRITE))
- goto fput_out;
-- retval = -EINVAL;
-- if (!out_file->f_op || !out_file->f_op->sendpage)
-- goto fput_out;
-- out_inode = out_file->f_path.dentry->d_inode;
-- retval = rw_verify_area(WRITE, out_file, &out_file->f_pos, count);
-- if (retval < 0)
-- goto fput_out;
-- count = retval;
--
-- retval = security_file_permission (out_file, MAY_WRITE);
-- if (retval)
-- goto fput_out;
--
-- if (!max)
-- max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes);
--
-- pos = *ppos;
-- retval = -EINVAL;
-- if (unlikely(pos < 0))
-- goto fput_out;
-- if (unlikely(pos + count > max)) {
-- retval = -EOVERFLOW;
-- if (pos >= max)
-- goto fput_out;
-- count = max - pos;
-- }
--
-- retval = in_file->f_op->sendfile(in_file, ppos, count, file_send_actor, out_file);
-
-- if (retval > 0) {
-- add_rchar(current, retval);
-- add_wchar(current, retval);
-- }
-+ retval = vfs_sendfile(out_file, in_file, ppos, count, max);
-
- inc_syscr(current);
- inc_syscw(current);
-- if (*ppos > max)
-- retval = -EOVERFLOW;
--
- fput_out:
- fput_light(out_file, fput_needed_out);
- fput_in:
---- linux-2.6.22/fs/reiserfs/bitmap.c 2007-02-06 03:01:29 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/reiserfs/bitmap.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/reiserfs/bitmap.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/reiserfs/bitmap.c 2008-04-19 15:14:52.000000000 -0400
@@ -13,6 +13,7 @@
#include <linux/reiserfs_fs_sb.h>
#include <linux/reiserfs_fs_i.h>
@@ -8586,7 +7828,7 @@
#define PREALLOCATION_SIZE 9
-@@ -425,8 +426,10 @@ static void _reiserfs_free_block(struct
+@@ -429,8 +430,10 @@ static void _reiserfs_free_block(struct
set_sb_free_blocks(rs, sb_free_blocks(rs) + 1);
journal_mark_dirty(th, s, sbh);
@@ -8598,7 +7840,7 @@
}
void reiserfs_free_block(struct reiserfs_transaction_handle *th,
-@@ -1034,6 +1037,7 @@ static inline int blocknrs_and_prealloc_
+@@ -1045,6 +1048,7 @@ static inline int blocknrs_and_prealloc_
b_blocknr_t finish = SB_BLOCK_COUNT(s) - 1;
int passno = 0;
int nr_allocated = 0;
@@ -8606,7 +7848,7 @@
determine_prealloc_size(hint);
if (!hint->formatted_node) {
-@@ -1043,19 +1047,30 @@ static inline int blocknrs_and_prealloc_
+@@ -1054,19 +1058,30 @@ static inline int blocknrs_and_prealloc_
"reiserquota: allocating %d blocks id=%u",
amount_needed, hint->inode->i_uid);
#endif
@@ -8643,7 +7885,7 @@
if (quota_ret)
hint->preallocate = hint->prealloc_size = 0;
}
-@@ -1087,7 +1102,10 @@ static inline int blocknrs_and_prealloc_
+@@ -1098,7 +1113,10 @@ static inline int blocknrs_and_prealloc_
nr_allocated,
hint->inode->i_uid);
#endif
@@ -8655,7 +7897,7 @@
}
while (nr_allocated--)
reiserfs_free_block(hint->th, hint->inode,
-@@ -1118,10 +1136,10 @@ static inline int blocknrs_and_prealloc_
+@@ -1129,10 +1147,10 @@ static inline int blocknrs_and_prealloc_
REISERFS_I(hint->inode)->i_prealloc_count,
hint->inode->i_uid);
#endif
@@ -8670,34 +7912,26 @@
}
return CARRY_ON;
---- linux-2.6.22/fs/reiserfs/file.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/reiserfs/file.c 2007-06-15 02:37:03 +0200
-@@ -1532,6 +1532,7 @@ const struct file_operations reiserfs_fi
- .release = reiserfs_file_release,
- .fsync = reiserfs_sync_file,
- .sendfile = generic_file_sendfile,
-+ .sendpage = generic_file_sendpage,
- .aio_read = generic_file_aio_read,
- .aio_write = generic_file_aio_write,
- .splice_read = generic_file_splice_read,
-@@ -1546,4 +1547,5 @@ const struct inode_operations reiserfs_f
+--- linux-2.6.25.4/fs/reiserfs/file.c 2008-04-17 11:31:38.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/reiserfs/file.c 2008-04-19 15:14:52.000000000 -0400
+@@ -306,4 +306,5 @@ const struct inode_operations reiserfs_f
.listxattr = reiserfs_listxattr,
.removexattr = reiserfs_removexattr,
.permission = reiserfs_permission,
+ .sync_flags = reiserfs_sync_flags,
};
---- linux-2.6.22/fs/reiserfs/inode.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/reiserfs/inode.c 2007-06-15 02:37:03 +0200
-@@ -16,6 +16,8 @@
- #include <linux/mpage.h>
+--- linux-2.6.25.4/fs/reiserfs/inode.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/reiserfs/inode.c 2008-04-19 15:14:52.000000000 -0400
+@@ -18,6 +18,8 @@
#include <linux/writeback.h>
#include <linux/quotaops.h>
+ #include <linux/swap.h>
+#include <linux/vs_dlimit.h>
+#include <linux/vs_tag.h>
- static int reiserfs_commit_write(struct file *f, struct page *page,
- unsigned from, unsigned to);
-@@ -50,6 +52,7 @@ void reiserfs_delete_inode(struct inode
+ int reiserfs_commit_write(struct file *f, struct page *page,
+ unsigned from, unsigned to);
+@@ -52,6 +54,7 @@ void reiserfs_delete_inode(struct inode
* stat data deletion */
if (!err)
DQUOT_FREE_INODE(inode);
@@ -8705,7 +7939,7 @@
if (journal_end(&th, inode->i_sb, jbegin_count))
goto out;
-@@ -1112,6 +1115,8 @@ static void init_inode(struct inode *ino
+@@ -1114,6 +1117,8 @@ static void init_inode(struct inode *ino
struct buffer_head *bh;
struct item_head *ih;
__u32 rdev;
@@ -8714,7 +7948,7 @@
//int version = ITEM_VERSION_1;
bh = PATH_PLAST_BUFFER(path);
-@@ -1135,12 +1140,13 @@ static void init_inode(struct inode *ino
+@@ -1137,12 +1142,13 @@ static void init_inode(struct inode *ino
(struct stat_data_v1 *)B_I_PITEM(bh, ih);
unsigned long blocks;
@@ -8730,7 +7964,7 @@
inode->i_size = sd_v1_size(sd);
inode->i_atime.tv_sec = sd_v1_atime(sd);
inode->i_mtime.tv_sec = sd_v1_mtime(sd);
-@@ -1182,11 +1188,12 @@ static void init_inode(struct inode *ino
+@@ -1184,11 +1190,12 @@ static void init_inode(struct inode *ino
// (directories and symlinks)
struct stat_data *sd = (struct stat_data *)B_I_PITEM(bh, ih);
@@ -8745,7 +7979,7 @@
inode->i_mtime.tv_sec = sd_v2_mtime(sd);
inode->i_atime.tv_sec = sd_v2_atime(sd);
inode->i_ctime.tv_sec = sd_v2_ctime(sd);
-@@ -1216,6 +1223,10 @@ static void init_inode(struct inode *ino
+@@ -1218,6 +1225,10 @@ static void init_inode(struct inode *ino
sd_attrs_to_i_attrs(sd_v2_attrs(sd), inode);
}
@@ -8756,7 +7990,7 @@
pathrelse(path);
if (S_ISREG(inode->i_mode)) {
inode->i_op = &reiserfs_file_inode_operations;
-@@ -1238,13 +1249,15 @@ static void init_inode(struct inode *ino
+@@ -1240,13 +1251,15 @@ static void init_inode(struct inode *ino
static void inode2sd(void *sd, struct inode *inode, loff_t size)
{
struct stat_data *sd_v2 = (struct stat_data *)sd;
@@ -8774,7 +8008,7 @@
set_sd_v2_mtime(sd_v2, inode->i_mtime.tv_sec);
set_sd_v2_atime(sd_v2, inode->i_atime.tv_sec);
set_sd_v2_ctime(sd_v2, inode->i_ctime.tv_sec);
-@@ -1775,6 +1788,10 @@ int reiserfs_new_inode(struct reiserfs_t
+@@ -1769,6 +1782,10 @@ int reiserfs_new_inode(struct reiserfs_t
BUG_ON(!th->t_trans_id);
@@ -8785,7 +8019,7 @@
if (DQUOT_ALLOC_INODE(inode)) {
err = -EDQUOT;
goto out_end_trans;
-@@ -1960,6 +1977,9 @@ int reiserfs_new_inode(struct reiserfs_t
+@@ -1954,6 +1971,9 @@ int reiserfs_new_inode(struct reiserfs_t
DQUOT_FREE_INODE(inode);
out_end_trans:
@@ -8795,7 +8029,7 @@
journal_end(th, th->t_super, th->t_blocks_allocated);
/* Drop can be outside and it needs more credits so it's better to have it outside */
DQUOT_DROP(inode);
-@@ -2690,6 +2710,14 @@ void sd_attrs_to_i_attrs(__u16 sd_attrs,
+@@ -2848,6 +2868,14 @@ void sd_attrs_to_i_attrs(__u16 sd_attrs,
inode->i_flags |= S_IMMUTABLE;
else
inode->i_flags &= ~S_IMMUTABLE;
@@ -8810,7 +8044,7 @@
if (sd_attrs & REISERFS_APPEND_FL)
inode->i_flags |= S_APPEND;
else
-@@ -2712,6 +2740,14 @@ void i_attrs_to_sd_attrs(struct inode *i
+@@ -2870,6 +2898,14 @@ void i_attrs_to_sd_attrs(struct inode *i
*sd_attrs |= REISERFS_IMMUTABLE_FL;
else
*sd_attrs &= ~REISERFS_IMMUTABLE_FL;
@@ -8825,7 +8059,7 @@
if (inode->i_flags & S_SYNC)
*sd_attrs |= REISERFS_SYNC_FL;
else
-@@ -2891,6 +2927,22 @@ static ssize_t reiserfs_direct_IO(int rw
+@@ -3049,6 +3085,22 @@ static ssize_t reiserfs_direct_IO(int rw
reiserfs_get_blocks_direct_io, NULL);
}
@@ -8848,7 +8082,7 @@
int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
{
struct inode *inode = dentry->d_inode;
-@@ -2940,9 +2992,11 @@ int reiserfs_setattr(struct dentry *dent
+@@ -3102,9 +3154,11 @@ int reiserfs_setattr(struct dentry *dent
}
error = inode_change_ok(inode, attr);
@@ -8861,7 +8095,7 @@
error = reiserfs_chown_xattrs(inode, attr);
if (!error) {
-@@ -2972,6 +3026,9 @@ int reiserfs_setattr(struct dentry *dent
+@@ -3134,6 +3188,9 @@ int reiserfs_setattr(struct dentry *dent
inode->i_uid = attr->ia_uid;
if (attr->ia_valid & ATTR_GID)
inode->i_gid = attr->ia_gid;
@@ -8871,17 +8105,17 @@
mark_inode_dirty(inode);
error =
journal_end(&th, inode->i_sb, jbegin_count);
---- linux-2.6.22/fs/reiserfs/ioctl.c 2007-02-06 03:01:29 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/reiserfs/ioctl.c 2007-06-15 02:37:03 +0200
-@@ -4,6 +4,7 @@
-
- #include <linux/capability.h>
+--- linux-2.6.25.4/fs/reiserfs/ioctl.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/reiserfs/ioctl.c 2008-05-21 14:30:41.000000000 -0400
+@@ -6,6 +6,7 @@
#include <linux/fs.h>
-+#include <linux/mount.h>
#include <linux/reiserfs_fs.h>
#include <linux/time.h>
++#include <linux/mount.h>
#include <asm/uaccess.h>
-@@ -24,7 +25,7 @@ static int reiserfs_unpack(struct inode
+ #include <linux/pagemap.h>
+ #include <linux/smp_lock.h>
+@@ -22,7 +23,7 @@
int reiserfs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
unsigned long arg)
{
@@ -8890,7 +8124,7 @@
switch (cmd) {
case REISERFS_IOC_UNPACK:
-@@ -43,12 +44,14 @@ int reiserfs_ioctl(struct inode *inode,
+@@ -41,12 +42,14 @@ int reiserfs_ioctl(struct inode *inode,
flags = REISERFS_I(inode)->i_attrs;
i_attrs_to_sd_attrs(inode, (__u16 *) & flags);
@@ -8905,37 +8139,21 @@
+ (filp && MNT_IS_RDONLY(filp->f_vfsmnt)))
return -EROFS;
- if ((current->fsuid != inode->i_uid)
-@@ -58,10 +61,12 @@ int reiserfs_ioctl(struct inode *inode,
- if (get_user(flags, (int __user *)arg))
- return -EFAULT;
-
-- if (((flags ^ REISERFS_I(inode)->
-- i_attrs) & (REISERFS_IMMUTABLE_FL |
-- REISERFS_APPEND_FL))
-- && !capable(CAP_LINUX_IMMUTABLE))
-+ oldflags = REISERFS_I(inode) -> i_attrs;
-+ if (((oldflags & REISERFS_IMMUTABLE_FL) ||
-+ ((flags ^ oldflags) &
-+ (REISERFS_IMMUTABLE_FL | REISERFS_IUNLINK_FL |
-+ REISERFS_APPEND_FL))) &&
-+ !capable(CAP_LINUX_IMMUTABLE))
- return -EPERM;
-
- if ((flags & REISERFS_NOTAIL_FL) &&
-@@ -72,6 +77,9 @@ int reiserfs_ioctl(struct inode *inode,
+ if (!is_owner_or_cap(inode))
+@@ -72,6 +75,10 @@ int reiserfs_ioctl(struct inode *inode,
if (result)
return result;
}
+
++ oldflags = REISERFS_I(inode)->i_attrs;
+ flags = flags & REISERFS_FL_USER_MODIFIABLE;
+ flags |= oldflags & ~REISERFS_FL_USER_MODIFIABLE;
sd_attrs_to_i_attrs(flags, inode);
REISERFS_I(inode)->i_attrs = flags;
inode->i_ctime = CURRENT_TIME_SEC;
-@@ -83,7 +91,8 @@ int reiserfs_ioctl(struct inode *inode,
+@@ -83,7 +90,8 @@ int reiserfs_ioctl(struct inode *inode,
case REISERFS_IOC_SETVERSION:
- if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER))
+ if (!is_owner_or_cap(inode))
return -EPERM;
- if (IS_RDONLY(inode))
+ if (IS_RDONLY(inode) ||
@@ -8943,8 +8161,8 @@
return -EROFS;
if (get_user(inode->i_generation, (int __user *)arg))
return -EFAULT;
---- linux-2.6.22/fs/reiserfs/namei.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/reiserfs/namei.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/reiserfs/namei.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/reiserfs/namei.c 2008-04-19 15:14:52.000000000 -0400
@@ -17,6 +17,7 @@
#include <linux/reiserfs_acl.h>
#include <linux/reiserfs_xattr.h>
@@ -8992,8 +8210,8 @@
+ .sync_flags = reiserfs_sync_flags,
};
---- linux-2.6.22/fs/reiserfs/stree.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/reiserfs/stree.c 2007-06-15 02:37:03 +0200
+--- linux-2.6.25.4/fs/reiserfs/stree.c 2008-04-17 11:31:38.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/reiserfs/stree.c 2008-04-19 15:14:52.000000000 -0400
@@ -55,6 +55,7 @@
#include <linux/reiserfs_fs.h>
#include <linux/buffer_head.h>
@@ -9002,7 +8220,7 @@
/* Does the buffer contain a disk block which is in the tree. */
inline int B_IS_IN_TREE(const struct buffer_head *p_s_bh)
-@@ -1296,6 +1297,7 @@ int reiserfs_delete_item(struct reiserfs
+@@ -1297,6 +1298,7 @@ int reiserfs_delete_item(struct reiserfs
"reiserquota delete_item(): freeing %u, id=%u type=%c",
quota_cut_bytes, p_s_inode->i_uid, head2type(&s_ih));
#endif
@@ -9010,7 +8228,7 @@
DQUOT_FREE_SPACE_NODIRTY(p_s_inode, quota_cut_bytes);
/* Return deleted body length */
-@@ -1384,6 +1386,7 @@ void reiserfs_delete_solid_item(struct r
+@@ -1385,6 +1387,7 @@ void reiserfs_delete_solid_item(struct r
#endif
DQUOT_FREE_SPACE_NODIRTY(inode,
quota_cut_bytes);
@@ -9018,7 +8236,7 @@
}
break;
}
-@@ -1737,6 +1740,7 @@ int reiserfs_cut_from_item(struct reiser
+@@ -1735,6 +1738,7 @@ int reiserfs_cut_from_item(struct reiser
"reiserquota cut_from_item(): freeing %u id=%u type=%c",
quota_cut_bytes, p_s_inode->i_uid, '?');
#endif
@@ -9026,7 +8244,7 @@
DQUOT_FREE_SPACE_NODIRTY(p_s_inode, quota_cut_bytes);
return n_ret_value;
}
-@@ -1978,6 +1982,11 @@ int reiserfs_paste_into_item(struct reis
+@@ -1976,6 +1980,11 @@ int reiserfs_paste_into_item(struct reis
pathrelse(p_s_search_path);
return -EDQUOT;
}
@@ -9038,7 +8256,7 @@
init_tb_struct(th, &s_paste_balance, th->t_super, p_s_search_path,
n_pasted_size);
#ifdef DISPLACE_NEW_PACKING_LOCALITIES
-@@ -2030,6 +2039,7 @@ int reiserfs_paste_into_item(struct reis
+@@ -2028,6 +2037,7 @@ int reiserfs_paste_into_item(struct reis
n_pasted_size, inode->i_uid,
key2type(&(p_s_key->on_disk_key)));
#endif
@@ -9046,7 +8264,7 @@
DQUOT_FREE_SPACE_NODIRTY(inode, n_pasted_size);
return retval;
}
-@@ -2067,6 +2077,11 @@ int reiserfs_insert_item(struct reiserfs
+@@ -2065,6 +2075,11 @@ int reiserfs_insert_item(struct reiserfs
pathrelse(p_s_path);
return -EDQUOT;
}
@@ -9058,7 +8276,7 @@
}
init_tb_struct(th, &s_ins_balance, th->t_super, p_s_path,
IH_SIZE + ih_item_len(p_s_ih));
-@@ -2114,7 +2129,9 @@ int reiserfs_insert_item(struct reiserfs
+@@ -2112,7 +2127,9 @@ int reiserfs_insert_item(struct reiserfs
"reiserquota insert_item(): freeing %u id=%u type=%c",
quota_bytes, inode->i_uid, head2type(p_s_ih));
#endif
@@ -9069,9 +8287,9 @@
+ }
return retval;
}
---- linux-2.6.22/fs/reiserfs/super.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/reiserfs/super.c 2007-06-15 02:37:03 +0200
-@@ -882,6 +882,14 @@ static int reiserfs_parse_options(struct
+--- linux-2.6.25.4/fs/reiserfs/super.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/reiserfs/super.c 2008-05-21 14:30:41.000000000 -0400
+@@ -896,6 +896,14 @@ static int reiserfs_parse_options(struct
{"user_xattr",.setmask = 1 << REISERFS_UNSUPPORTED_OPT},
{"nouser_xattr",.clrmask = 1 << REISERFS_UNSUPPORTED_OPT},
#endif
@@ -9086,8 +8304,8 @@
#ifdef CONFIG_REISERFS_FS_POSIX_ACL
{"acl",.setmask = 1 << REISERFS_POSIXACL},
{"noacl",.clrmask = 1 << REISERFS_POSIXACL},
-@@ -1143,6 +1151,12 @@ static int reiserfs_remount(struct super
- return -EINVAL;
+@@ -1159,6 +1167,12 @@ static int reiserfs_remount(struct super
+ goto out_err;
}
+ if ((mount_options & (1 << REISERFS_TAGGED)) &&
@@ -9099,7 +8317,7 @@
handle_attrs(s);
/* Add options that are safe here */
-@@ -1591,6 +1605,10 @@ static int reiserfs_fill_super(struct su
+@@ -1618,6 +1632,10 @@ static int reiserfs_fill_super(struct su
goto error;
}
@@ -9110,8 +8328,8 @@
rs = SB_DISK_SUPER_BLOCK(s);
/* Let's do basic sanity check to verify that underlying device is not
smaller than the filesystem. If the check fails then abort and scream,
---- linux-2.6.22/fs/reiserfs/xattr.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/reiserfs/xattr.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/fs/reiserfs/xattr.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/reiserfs/xattr.c 2008-04-19 15:14:52.000000000 -0400
@@ -35,6 +35,7 @@
#include <linux/namei.h>
#include <linux/errno.h>
@@ -9120,7 +8338,7 @@
#include <linux/file.h>
#include <linux/pagemap.h>
#include <linux/xattr.h>
-@@ -775,7 +776,7 @@ int reiserfs_delete_xattrs(struct inode
+@@ -747,7 +748,7 @@ int reiserfs_delete_xattrs(struct inode
if (dir->d_inode->i_nlink <= 2) {
root = get_xa_root(inode->i_sb, XATTR_REPLACE);
reiserfs_write_lock_xattrs(inode->i_sb);
@@ -9129,8 +8347,8 @@
reiserfs_write_unlock_xattrs(inode->i_sb);
dput(root);
} else {
---- linux-2.6.22/fs/stat.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/stat.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/fs/stat.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/stat.c 2008-04-19 15:14:52.000000000 -0400
@@ -26,6 +26,7 @@ void generic_fillattr(struct inode *inod
stat->nlink = inode->i_nlink;
stat->uid = inode->i_uid;
@@ -9139,8 +8357,8 @@
stat->rdev = inode->i_rdev;
stat->atime = inode->i_atime;
stat->mtime = inode->i_mtime;
---- linux-2.6.22/fs/super.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/super.c 2007-06-15 03:26:55 +0200
+--- linux-2.6.25.4/fs/super.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/super.c 2008-04-19 15:14:52.000000000 -0400
@@ -37,6 +37,9 @@
#include <linux/idr.h>
#include <linux/kobject.h>
@@ -9151,7 +8369,7 @@
#include <asm/uaccess.h>
-@@ -860,12 +863,18 @@ struct vfsmount *
+@@ -859,12 +862,18 @@ struct vfsmount *
vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void *data)
{
struct vfsmount *mnt;
@@ -9171,8 +8389,8 @@
mnt = alloc_vfsmnt(name);
if (!mnt)
@@ -885,7 +894,14 @@ vfs_kern_mount(struct file_system_type *
- if (error < 0)
goto out_free_secdata;
+ BUG_ON(!mnt->mnt_sb);
- error = security_sb_kern_mount(mnt->mnt_sb, secdata);
+ sb = mnt->mnt_sb;
@@ -9186,18 +8404,18 @@
if (error)
goto out_sb;
---- linux-2.6.22/fs/sysfs/mount.c 2007-07-09 13:19:28 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/sysfs/mount.c 2007-06-17 05:54:17 +0200
-@@ -12,8 +12,6 @@
+--- linux-2.6.25.4/fs/sysfs/mount.c 2008-04-17 11:31:38.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/sysfs/mount.c 2008-04-19 15:14:52.000000000 -0400
+@@ -19,8 +19,6 @@
#include "sysfs.h"
-/* Random magic number */
-#define SYSFS_MAGIC 0x62656572
- struct vfsmount *sysfs_mount;
+ static struct vfsmount *sysfs_mount;
struct super_block * sysfs_sb = NULL;
-@@ -48,7 +46,7 @@ static int sysfs_fill_super(struct super
+@@ -46,7 +44,7 @@ static int sysfs_fill_super(struct super
sb->s_blocksize = PAGE_CACHE_SIZE;
sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
@@ -9206,38 +8424,29 @@
sb->s_op = &sysfs_ops;
sb->s_time_gran = 1;
sysfs_sb = sb;
---- linux-2.6.22/fs/utimes.c 2007-07-09 13:19:29 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/utimes.c 2007-07-09 13:11:54 +0200
-@@ -6,6 +6,8 @@
- #include <linux/sched.h>
+--- linux-2.6.25.4/fs/utimes.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/utimes.c 2008-05-21 14:30:41.000000000 -0400
+@@ -7,6 +7,8 @@
#include <linux/stat.h>
#include <linux/utime.h>
+ #include <linux/syscalls.h>
+#include <linux/mount.h>
+#include <linux/vs_cowbl.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
-@@ -70,14 +72,16 @@ long do_utimes(int dfd, char __user *fil
+@@ -89,6 +91,9 @@ long do_utimes(int dfd, char __user *fil
if (error)
goto out;
+ error = cow_check_and_break(&nd);
+ if (error)
+ goto dput_and_out;
- dentry = nd.dentry;
+ dentry = nd.path.dentry;
}
- inode = dentry->d_inode;
--
- error = -EROFS;
- if (IS_RDONLY(inode))
-- goto dput_and_out;
-+ goto dput_and_out;
-
- /* Don't worry, the checks are done in inode_change_ok() */
- newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME;
---- linux-2.6.22/fs/xattr.c 2007-07-09 13:19:29 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xattr.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/fs/xattr.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xattr.c 2008-04-21 13:52:50.000000000 -0400
@@ -17,6 +17,7 @@
#include <linux/module.h>
#include <linux/fsnotify.h>
@@ -9246,7 +8455,7 @@
#include <asm/uaccess.h>
-@@ -194,7 +195,7 @@ EXPORT_SYMBOL_GPL(vfs_removexattr);
+@@ -220,7 +221,7 @@ EXPORT_SYMBOL_GPL(vfs_removexattr);
*/
static long
setxattr(struct dentry *d, char __user *name, void __user *value,
@@ -9255,7 +8464,7 @@
{
int error;
void *kvalue = NULL;
-@@ -221,6 +222,9 @@ setxattr(struct dentry *d, char __user *
+@@ -247,6 +248,9 @@ setxattr(struct dentry *d, char __user *
}
}
@@ -9265,34 +8474,34 @@
error = vfs_setxattr(d, kname, kvalue, size, flags);
kfree(kvalue);
return error;
-@@ -236,7 +240,7 @@ sys_setxattr(char __user *path, char __u
+@@ -262,7 +266,7 @@ sys_setxattr(char __user *path, char __u
error = user_path_walk(path, &nd);
if (error)
return error;
-- error = setxattr(nd.dentry, name, value, size, flags);
-+ error = setxattr(nd.dentry, name, value, size, flags, nd.mnt);
- path_release(&nd);
+- error = setxattr(nd.path.dentry, name, value, size, flags);
++ error = setxattr(nd.path.dentry, name, value, size, flags, nd.path.mnt);
+ path_put(&nd.path);
return error;
}
-@@ -251,7 +255,7 @@ sys_lsetxattr(char __user *path, char __
+@@ -277,7 +281,7 @@ sys_lsetxattr(char __user *path, char __
error = user_path_walk_link(path, &nd);
if (error)
return error;
-- error = setxattr(nd.dentry, name, value, size, flags);
-+ error = setxattr(nd.dentry, name, value, size, flags, nd.mnt);
- path_release(&nd);
+- error = setxattr(nd.path.dentry, name, value, size, flags);
++ error = setxattr(nd.path.dentry, name, value, size, flags, nd.path.mnt);
+ path_put(&nd.path);
return error;
}
-@@ -269,7 +273,7 @@ sys_fsetxattr(int fd, char __user *name,
+@@ -295,7 +299,7 @@ sys_fsetxattr(int fd, char __user *name,
return error;
dentry = f->f_path.dentry;
- audit_inode(NULL, dentry->d_inode);
+ audit_inode(NULL, dentry);
- error = setxattr(dentry, name, value, size, flags);
+ error = setxattr(dentry, name, value, size, flags, f->f_vfsmnt);
fput(f);
return error;
}
-@@ -433,7 +437,7 @@ sys_flistxattr(int fd, char __user *list
+@@ -459,7 +463,7 @@ sys_flistxattr(int fd, char __user *list
* Extended attribute REMOVE operations
*/
static long
@@ -9301,7 +8510,7 @@
{
int error;
char kname[XATTR_NAME_MAX + 1];
-@@ -444,6 +448,9 @@ removexattr(struct dentry *d, char __use
+@@ -470,6 +474,9 @@ removexattr(struct dentry *d, char __use
if (error < 0)
return error;
@@ -9311,75 +8520,77 @@
return vfs_removexattr(d, kname);
}
-@@ -456,7 +463,7 @@ sys_removexattr(char __user *path, char
+@@ -482,7 +489,7 @@ sys_removexattr(char __user *path, char
error = user_path_walk(path, &nd);
if (error)
return error;
-- error = removexattr(nd.dentry, name);
-+ error = removexattr(nd.dentry, name, nd.mnt);
- path_release(&nd);
+- error = removexattr(nd.path.dentry, name);
++ error = removexattr(nd.path.dentry, name, nd.path.mnt);
+ path_put(&nd.path);
return error;
}
-@@ -470,7 +477,7 @@ sys_lremovexattr(char __user *path, char
+@@ -496,7 +503,7 @@ sys_lremovexattr(char __user *path, char
error = user_path_walk_link(path, &nd);
if (error)
return error;
-- error = removexattr(nd.dentry, name);
-+ error = removexattr(nd.dentry, name, nd.mnt);
- path_release(&nd);
+- error = removexattr(nd.path.dentry, name);
++ error = removexattr(nd.path.dentry, name, nd.path.mnt);
+ path_put(&nd.path);
return error;
}
-@@ -487,7 +494,7 @@ sys_fremovexattr(int fd, char __user *na
+@@ -513,7 +520,7 @@ sys_fremovexattr(int fd, char __user *na
return error;
dentry = f->f_path.dentry;
- audit_inode(NULL, dentry->d_inode);
+ audit_inode(NULL, dentry);
- error = removexattr(dentry, name);
+ error = removexattr(dentry, name, f->f_vfsmnt);
fput(f);
return error;
}
---- linux-2.6.22/fs/xfs/linux-2.6/xfs_file.c 2007-05-02 19:25:22 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/linux-2.6/xfs_file.c 2007-06-15 02:37:04 +0200
-@@ -453,6 +453,7 @@ const struct file_operations xfs_file_op
- .aio_read = xfs_file_aio_read,
- .aio_write = xfs_file_aio_write,
- .sendfile = xfs_file_sendfile,
-+ .sendpage = generic_file_sendpage,
- .splice_read = xfs_file_splice_read,
- .splice_write = xfs_file_splice_write,
- .unlocked_ioctl = xfs_file_ioctl,
-@@ -476,6 +477,7 @@ const struct file_operations xfs_invis_f
- .aio_read = xfs_file_aio_read_invis,
- .aio_write = xfs_file_aio_write_invis,
- .sendfile = xfs_file_sendfile_invis,
-+ .sendpage = generic_file_sendpage,
- .splice_read = xfs_file_splice_read_invis,
- .splice_write = xfs_file_splice_write_invis,
- .unlocked_ioctl = xfs_file_ioctl_invis,
---- linux-2.6.22/fs/xfs/linux-2.6/xfs_ioctl.c 2007-05-02 19:25:22 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/linux-2.6/xfs_ioctl.c 2007-06-15 02:37:04 +0200
-@@ -1128,6 +1128,10 @@ xfs_di2lxflags(
-
- if (di_flags & XFS_DIFLAG_IMMUTABLE)
- flags |= FS_IMMUTABLE_FL;
-+ if (di_flags & XFS_DIFLAG_IUNLINK)
+--- linux-2.6.25.4/fs/xfs/linux-2.6/xfs_ioctl.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/linux-2.6/xfs_ioctl.c 2008-04-19 15:14:52.000000000 -0400
+@@ -1125,7 +1125,8 @@ xfs_merge_ioc_xflags(
+
+ STATIC unsigned int
+ xfs_di2lxflags(
+- __uint16_t di_flags)
++ __uint16_t di_flags,
++ __uint16_t di_vflags)
+ {
+ unsigned int flags = 0;
+
+@@ -1139,6 +1140,11 @@ xfs_di2lxflags(
+ flags |= FS_NOATIME_FL;
+ if (di_flags & XFS_DIFLAG_NODUMP)
+ flags |= FS_NODUMP_FL;
++
++ if (di_vflags & XFS_DIVFLAG_IUNLINK)
+ flags |= FS_IUNLINK_FL;
-+ if (di_flags & XFS_DIFLAG_BARRIER)
++ if (di_vflags & XFS_DIVFLAG_BARRIER)
+ flags |= FS_BARRIER_FL;
- if (di_flags & XFS_DIFLAG_APPEND)
- flags |= FS_APPEND_FL;
- if (di_flags & XFS_DIFLAG_SYNC)
---- linux-2.6.22/fs/xfs/linux-2.6/xfs_iops.c 2007-05-02 19:25:22 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/linux-2.6/xfs_iops.c 2007-06-15 02:37:04 +0200
-@@ -51,6 +51,7 @@
- #include <linux/xattr.h>
+ return flags;
+ }
+
+@@ -1219,7 +1225,7 @@ xfs_ioc_xattr(
+ }
+
+ case XFS_IOC_GETXFLAGS: {
+- flags = xfs_di2lxflags(ip->i_d.di_flags);
++ flags = xfs_di2lxflags(ip->i_d.di_flags, ip->i_d.di_vflags);
+ if (copy_to_user(arg, &flags, sizeof(flags)))
+ error = -EFAULT;
+ break;
+--- linux-2.6.25.4/fs/xfs/linux-2.6/xfs_iops.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/linux-2.6/xfs_iops.c 2008-04-19 17:01:15.000000000 -0400
+@@ -53,6 +53,7 @@
#include <linux/namei.h>
#include <linux/security.h>
+ #include <linux/falloc.h>
+#include <linux/vs_tag.h>
/*
- * Get a XFS inode from a given vnode.
-@@ -400,6 +401,7 @@ xfs_vn_lookup(
+ * Bring the atime in the XFS inode uptodate.
+@@ -391,6 +392,7 @@ xfs_vn_lookup(
d_add(dentry, NULL);
return NULL;
}
@@ -9387,7 +8598,7 @@
return d_splice_alias(vn_to_inode(cvp), dentry);
}
-@@ -657,6 +659,10 @@ xfs_vn_setattr(
+@@ -655,6 +657,10 @@ xfs_vn_setattr(
int flags = 0;
int error;
@@ -9398,7 +8609,7 @@
if (ia_valid & ATTR_UID) {
vattr.va_mask |= XFS_AT_UID;
vattr.va_uid = attr->ia_uid;
-@@ -665,6 +671,10 @@ xfs_vn_setattr(
+@@ -663,6 +669,10 @@ xfs_vn_setattr(
vattr.va_mask |= XFS_AT_GID;
vattr.va_gid = attr->ia_gid;
}
@@ -9409,7 +8620,7 @@
if (ia_valid & ATTR_SIZE) {
vattr.va_mask |= XFS_AT_SIZE;
vattr.va_size = attr->ia_size;
-@@ -710,6 +720,42 @@ xfs_vn_truncate(
+@@ -708,6 +718,42 @@ xfs_vn_truncate(
}
STATIC int
@@ -9424,7 +8635,7 @@
+ memset(&vattr, 0, sizeof vattr);
+
+ vattr.va_mask = XFS_AT_XFLAGS;
-+ error = bhv_vop_getattr(vp, &vattr, 0, NULL);
++ error = xfs_getattr(XFS_I(inode), &vattr, 0);
+
+ if (error)
+ return error;
@@ -9442,7 +8653,7 @@
+ if (oldflags ^ newflags) {
+ vattr.va_xflags = newflags;
+ vattr.va_mask |= XFS_AT_XFLAGS;
-+ error = bhv_vop_setattr(vp, &vattr, flags, NULL);
++ error = xfs_setattr(XFS_I(inode), &vattr, flags, NULL);
+ }
+ vn_revalidate(vp);
+ return error;
@@ -9452,31 +8663,24 @@
xfs_vn_setxattr(
struct dentry *dentry,
const char *name,
-@@ -822,6 +868,7 @@ const struct inode_operations xfs_inode_
+@@ -881,6 +927,8 @@ const struct inode_operations xfs_dir_in
.getxattr = xfs_vn_getxattr,
.listxattr = xfs_vn_listxattr,
.removexattr = xfs_vn_removexattr,
+ .sync_flags = xfs_vn_sync_flags,
- };
-
- const struct inode_operations xfs_dir_inode_operations = {
-@@ -841,6 +888,7 @@ const struct inode_operations xfs_dir_in
- .getxattr = xfs_vn_getxattr,
- .listxattr = xfs_vn_listxattr,
- .removexattr = xfs_vn_removexattr,
+ .sync_flags = xfs_vn_sync_flags,
};
const struct inode_operations xfs_symlink_inode_operations = {
-@@ -854,4 +902,5 @@ const struct inode_operations xfs_symlin
+@@ -894,4 +942,5 @@ const struct inode_operations xfs_symlin
.getxattr = xfs_vn_getxattr,
.listxattr = xfs_vn_listxattr,
.removexattr = xfs_vn_removexattr,
+ .sync_flags = xfs_vn_sync_flags,
};
---- linux-2.6.22/fs/xfs/linux-2.6/xfs_linux.h 2007-05-02 19:25:22 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/linux-2.6/xfs_linux.h 2007-06-15 02:37:04 +0200
-@@ -129,6 +129,7 @@
+--- linux-2.6.25.4/fs/xfs/linux-2.6/xfs_linux.h 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/linux-2.6/xfs_linux.h 2008-04-19 15:14:52.000000000 -0400
+@@ -128,6 +128,7 @@
#define current_pid() (current->pid)
#define current_fsuid(cred) (current->fsuid)
#define current_fsgid(cred) (current->fsgid)
@@ -9484,9 +8688,39 @@
#define current_test_flags(f) (current->flags & (f))
#define current_set_flags_nested(sp, f) \
(*(sp) = current->flags, current->flags |= (f))
---- linux-2.6.22/fs/xfs/linux-2.6/xfs_super.c 2007-07-09 13:19:29 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/linux-2.6/xfs_super.c 2007-06-15 02:37:04 +0200
-@@ -157,6 +157,7 @@ xfs_revalidate_inode(
+--- linux-2.6.25.4/fs/xfs/linux-2.6/xfs_super.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/linux-2.6/xfs_super.c 2008-04-26 09:51:47.000000000 -0400
+@@ -137,6 +137,9 @@ xfs_args_allocate(
+ #define MNTOPT_DMAPI "dmapi" /* DMI enabled (DMAPI / XDSM) */
+ #define MNTOPT_XDSM "xdsm" /* DMI enabled (DMAPI / XDSM) */
+ #define MNTOPT_DMI "dmi" /* DMI enabled (DMAPI / XDSM) */
++#define MNTOPT_TAGXID "tagxid" /* context tagging for inodes */
++#define MNTOPT_TAGGED "tag" /* context tagging for inodes */
++#define MNTOPT_NOTAGTAG "notag" /* do not use context tagging */
+
+ STATIC unsigned long
+ suffix_strtoul(char *s, char **endp, unsigned int base)
+@@ -355,6 +358,19 @@ xfs_parseargs(
+ } else if (!strcmp(this_char, "irixsgid")) {
+ cmn_err(CE_WARN,
+ "XFS: irixsgid is now a sysctl(2) variable, option is deprecated.");
++#ifndef CONFIG_TAGGING_NONE
++ } else if (!strcmp(this_char, MNTOPT_TAGGED)) {
++ args->flags2 |= XFSMNT2_TAGGED;
++ } else if (!strcmp(this_char, MNTOPT_NOTAGTAG)) {
++ args->flags2 &= ~XFSMNT2_TAGGED;
++ } else if (!strcmp(this_char, MNTOPT_TAGXID)) {
++ args->flags2 |= XFSMNT2_TAGGED;
++#endif
++#ifdef CONFIG_PROPAGATE
++ } else if (!strcmp(this_char, MNTOPT_TAGGED)) {
++ /* use value */
++ args->flags2 |= XFSMNT2_TAGGED;
++#endif
+ } else {
+ cmn_err(CE_WARN,
+ "XFS: unknown mount option [%s].", this_char);
+@@ -591,6 +607,7 @@ xfs_revalidate_inode(
inode->i_nlink = ip->i_d.di_nlink;
inode->i_uid = ip->i_d.di_uid;
inode->i_gid = ip->i_d.di_gid;
@@ -9494,25 +8728,34 @@
switch (inode->i_mode & S_IFMT) {
case S_IFBLK:
-@@ -184,6 +185,14 @@ xfs_revalidate_inode(
+@@ -612,6 +629,7 @@ xfs_revalidate_inode(
+ inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;
+ inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec;
+ inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec;
++
+ if (ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE)
inode->i_flags |= S_IMMUTABLE;
else
- inode->i_flags &= ~S_IMMUTABLE;
-+ if (ip->i_d.di_flags & XFS_DIFLAG_IUNLINK)
+@@ -628,6 +646,15 @@ xfs_revalidate_inode(
+ inode->i_flags |= S_NOATIME;
+ else
+ inode->i_flags &= ~S_NOATIME;
++
++ if (ip->i_d.di_vflags & XFS_DIVFLAG_IUNLINK)
+ inode->i_flags |= S_IUNLINK;
+ else
+ inode->i_flags &= ~S_IUNLINK;
-+ if (ip->i_d.di_flags & XFS_DIFLAG_BARRIER)
++ if (ip->i_d.di_vflags & XFS_DIVFLAG_BARRIER)
+ inode->i_flags |= S_BARRIER;
+ else
+ inode->i_flags &= ~S_BARRIER;
- if (ip->i_d.di_flags & XFS_DIFLAG_APPEND)
- inode->i_flags |= S_APPEND;
- else
-@@ -712,6 +721,12 @@ xfs_fs_remount(
+ xfs_iflags_clear(ip, XFS_IMODIFIED);
+ }
+
+@@ -1220,6 +1247,12 @@ xfs_fs_remount(
int error;
- error = bhv_vfs_parseargs(vfsp, options, args, 1);
+ error = xfs_parseargs(mp, options, args, 1);
+ if ((args->flags2 & XFSMNT2_TAGGED) &&
+ !(sb->s_flags & MS_TAGGED)) {
+ printk("XFS: %s: tagging not permitted on remount.\n",
@@ -9520,36 +8763,46 @@
+ error = EINVAL;
+ }
if (!error)
- error = bhv_vfs_mntupdate(vfsp, flags, args);
+ error = xfs_mntupdate(mp, flags, args);
kmem_free(args, sizeof(*args));
---- linux-2.6.22/fs/xfs/linux-2.6/xfs_vnode.c 2007-05-02 19:25:22 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/linux-2.6/xfs_vnode.c 2007-06-15 02:37:04 +0200
-@@ -119,6 +119,7 @@ vn_revalidate_core(
- inode->i_nlink = vap->va_nlink;
- inode->i_uid = vap->va_uid;
- inode->i_gid = vap->va_gid;
-+ inode->i_tag = vap->va_tag;
- inode->i_blocks = vap->va_nblocks;
- inode->i_mtime = vap->va_mtime;
- inode->i_ctime = vap->va_ctime;
-@@ -126,6 +127,14 @@ vn_revalidate_core(
+@@ -1336,6 +1369,9 @@ xfs_fs_fill_super(
+ if (error)
+ goto fail_vfsop;
+
++ if (mp->m_flags & XFS_MOUNT_TAGGED)
++ sb->s_flags |= MS_TAGGED;
++
+ sb->s_dirt = 1;
+ sb->s_magic = XFS_SB_MAGIC;
+ sb->s_blocksize = mp->m_sb.sb_blocksize;
+--- linux-2.6.25.4/fs/xfs/linux-2.6/xfs_vnode.c 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/linux-2.6/xfs_vnode.c 2008-04-19 17:04:58.000000000 -0400
+@@ -105,6 +105,7 @@ vn_revalidate(
+ inode->i_mode = ip->i_d.di_mode;
+ inode->i_uid = ip->i_d.di_uid;
+ inode->i_gid = ip->i_d.di_gid;
++ inode->i_tag = ip->i_d.di_tag;
+ inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec;
+ inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;
+ inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec;
+@@ -115,6 +116,14 @@ vn_revalidate(
inode->i_flags |= S_IMMUTABLE;
else
inode->i_flags &= ~S_IMMUTABLE;
-+ if (vap->va_xflags & XFS_XFLAG_IUNLINK)
++ if (xflags & XFS_XFLAG_IUNLINK)
+ inode->i_flags |= S_IUNLINK;
+ else
+ inode->i_flags &= ~S_IUNLINK;
-+ if (vap->va_xflags & XFS_XFLAG_BARRIER)
++ if (xflags & XFS_XFLAG_BARRIER)
+ inode->i_flags |= S_BARRIER;
+ else
+ inode->i_flags &= ~S_BARRIER;
- if (vap->va_xflags & XFS_XFLAG_APPEND)
+ if (xflags & XFS_XFLAG_APPEND)
inode->i_flags |= S_APPEND;
else
---- linux-2.6.22/fs/xfs/linux-2.6/xfs_vnode.h 2007-07-09 13:19:29 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/linux-2.6/xfs_vnode.h 2007-06-15 02:37:04 +0200
-@@ -350,6 +350,7 @@ typedef struct bhv_vattr {
+--- linux-2.6.25.4/fs/xfs/linux-2.6/xfs_vnode.h 2008-04-17 12:05:41.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/linux-2.6/xfs_vnode.h 2008-04-19 15:14:52.000000000 -0400
+@@ -99,6 +99,7 @@ typedef struct bhv_vattr {
xfs_nlink_t va_nlink; /* number of references to file */
uid_t va_uid; /* owner user id */
gid_t va_gid; /* owner group id */
@@ -9557,7 +8810,7 @@
xfs_ino_t va_nodeid; /* file id */
xfs_off_t va_size; /* file size in bytes */
u_long va_blocksize; /* blocksize preferred for i/o */
-@@ -398,13 +399,15 @@ typedef struct bhv_vattr {
+@@ -147,13 +148,15 @@ typedef struct bhv_vattr {
#define XFS_AT_PROJID 0x04000000
#define XFS_AT_SIZE_NOPERM 0x08000000
#define XFS_AT_GENCOUNT 0x10000000
@@ -9574,8 +8827,8 @@
#define XFS_AT_STAT (XFS_AT_TYPE|XFS_AT_MODE|XFS_AT_UID|XFS_AT_GID|\
XFS_AT_FSID|XFS_AT_NODEID|XFS_AT_NLINK|XFS_AT_SIZE|\
---- linux-2.6.22/fs/xfs/quota/xfs_qm_syscalls.c 2007-07-09 13:19:29 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/quota/xfs_qm_syscalls.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/fs/xfs/quota/xfs_qm_syscalls.c 2008-04-17 12:05:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/quota/xfs_qm_syscalls.c 2008-04-19 15:14:52.000000000 -0400
@@ -17,6 +17,7 @@
*/
@@ -9584,7 +8837,7 @@
#include "xfs.h"
#include "xfs_fs.h"
-@@ -211,7 +212,7 @@ xfs_qm_scall_quotaoff(
+@@ -205,7 +206,7 @@ xfs_qm_scall_quotaoff(
xfs_qoff_logitem_t *qoffstart;
int nculprits;
@@ -9593,7 +8846,7 @@
return XFS_ERROR(EPERM);
/*
* No file system can have quotas enabled on disk but not in core.
-@@ -380,7 +381,7 @@ xfs_qm_scall_trunc_qfiles(
+@@ -374,7 +375,7 @@ xfs_qm_scall_trunc_qfiles(
int error;
xfs_inode_t *qip;
@@ -9601,8 +8854,8 @@
+ if (!vx_capable(CAP_SYS_ADMIN, VXC_QUOTA_CTL))
return XFS_ERROR(EPERM);
error = 0;
- if (!XFS_SB_VERSION_HASQUOTA(&mp->m_sb) || flags == 0) {
-@@ -425,7 +426,7 @@ xfs_qm_scall_quotaon(
+ if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) {
+@@ -418,7 +419,7 @@ xfs_qm_scall_quotaon(
uint accflags;
__int64_t sbflags;
@@ -9611,7 +8864,7 @@
return XFS_ERROR(EPERM);
flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD);
-@@ -594,7 +595,7 @@ xfs_qm_scall_setqlim(
+@@ -587,7 +588,7 @@ xfs_qm_scall_setqlim(
int error;
xfs_qcnt_t hard, soft;
@@ -9620,67 +8873,67 @@
return XFS_ERROR(EPERM);
if ((newlim->d_fieldmask &
---- linux-2.6.22/fs/xfs/xfs_clnt.h 2006-06-18 04:54:50 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/xfs_clnt.h 2007-06-15 02:37:04 +0200
-@@ -99,5 +99,7 @@ struct xfs_mount_args {
- */
- #define XFSMNT2_COMPAT_IOSIZE 0x00000001 /* don't report large preferred
+--- linux-2.6.25.4/fs/xfs/xfs_clnt.h 2008-04-17 12:05:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_clnt.h 2008-04-19 15:14:52.000000000 -0400
+@@ -100,5 +100,6 @@ struct xfs_mount_args {
* I/O size in stat(2) */
+ #define XFSMNT2_FILESTREAMS 0x00000002 /* enable the filestreams
+ * allocator */
+#define XFSMNT2_TAGGED 0x80000000 /* context tagging */
-+
#endif /* __XFS_CLNT_H__ */
---- linux-2.6.22/fs/xfs/xfs_dinode.h 2006-09-20 16:58:40 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/xfs_dinode.h 2007-06-15 02:37:04 +0200
-@@ -53,7 +53,8 @@ typedef struct xfs_dinode_core
- __uint32_t di_gid; /* owner's group id */
- __uint32_t di_nlink; /* number of links to file */
- __uint16_t di_projid; /* owner's project id */
-- __uint8_t di_pad[8]; /* unused, zeroed space */
-+ __uint16_t di_tag; /* context tagging */
-+ __uint8_t di_pad[6]; /* unused, zeroed space */
- __uint16_t di_flushiter; /* incremented on flush */
+--- linux-2.6.25.4/fs/xfs/xfs_dinode.h 2008-04-17 12:05:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_dinode.h 2008-04-27 10:33:37.000000000 -0400
+@@ -53,7 +53,9 @@ typedef struct xfs_dinode_core {
+ __be32 di_gid; /* owner's group id */
+ __be32 di_nlink; /* number of links to file */
+ __be16 di_projid; /* owner's project id */
+- __u8 di_pad[8]; /* unused, zeroed space */
++ __be16 di_tag; /* context tagging */
++ __be16 di_vflags; /* vserver specific flags */
++ __u8 di_pad[4]; /* unused, zeroed space */
+ __be16 di_flushiter; /* incremented on flush */
xfs_timestamp_t di_atime; /* time last accessed */
xfs_timestamp_t di_mtime; /* time last modified */
-@@ -257,6 +258,9 @@ typedef enum xfs_dinode_fmt
- #define XFS_DIFLAG_EXTSIZE_BIT 11 /* inode extent size allocator hint */
+@@ -136,7 +138,9 @@ typedef struct xfs_dinode
+ #define XFS_DI_NEXT_UNLINKED 0x1000000
+ #define XFS_DI_U 0x2000000
+ #define XFS_DI_A 0x4000000
+-#define XFS_DI_NUM_BITS 27
++#define XFS_DI_VFLAGS 0x8000000
++#define XFS_DI_TAG 0x10000000
++#define XFS_DI_NUM_BITS 29
+ #define XFS_DI_ALL_BITS ((1 << XFS_DI_NUM_BITS) - 1)
+ #define XFS_DI_CORE_BITS (XFS_DI_ALL_BITS & ~(XFS_DI_U|XFS_DI_A))
+
+@@ -223,6 +227,7 @@ typedef enum xfs_dinode_fmt
#define XFS_DIFLAG_EXTSZINHERIT_BIT 12 /* inherit inode extent size */
#define XFS_DIFLAG_NODEFRAG_BIT 13 /* do not reorganize/defragment */
-+#define XFS_DIFLAG_BARRIER_BIT 14 /* chroot() barrier */
-+#define XFS_DIFLAG_IUNLINK_BIT 15 /* immutable unlink */
+ #define XFS_DIFLAG_FILESTREAM_BIT 14 /* use filestream allocator */
+
#define XFS_DIFLAG_REALTIME (1 << XFS_DIFLAG_REALTIME_BIT)
#define XFS_DIFLAG_PREALLOC (1 << XFS_DIFLAG_PREALLOC_BIT)
#define XFS_DIFLAG_NEWRTBM (1 << XFS_DIFLAG_NEWRTBM_BIT)
-@@ -271,12 +275,15 @@ typedef enum xfs_dinode_fmt
- #define XFS_DIFLAG_EXTSIZE (1 << XFS_DIFLAG_EXTSIZE_BIT)
- #define XFS_DIFLAG_EXTSZINHERIT (1 << XFS_DIFLAG_EXTSZINHERIT_BIT)
- #define XFS_DIFLAG_NODEFRAG (1 << XFS_DIFLAG_NODEFRAG_BIT)
-+#define XFS_DIFLAG_BARRIER (1 << XFS_DIFLAG_BARRIER_BIT)
-+#define XFS_DIFLAG_IUNLINK (1 << XFS_DIFLAG_IUNLINK_BIT)
-
- #define XFS_DIFLAG_ANY \
- (XFS_DIFLAG_REALTIME | XFS_DIFLAG_PREALLOC | XFS_DIFLAG_NEWRTBM | \
- XFS_DIFLAG_IMMUTABLE | XFS_DIFLAG_APPEND | XFS_DIFLAG_SYNC | \
- XFS_DIFLAG_NOATIME | XFS_DIFLAG_NODUMP | XFS_DIFLAG_RTINHERIT | \
+@@ -252,4 +257,7 @@ typedef enum xfs_dinode_fmt
XFS_DIFLAG_PROJINHERIT | XFS_DIFLAG_NOSYMLINKS | XFS_DIFLAG_EXTSIZE | \
-- XFS_DIFLAG_EXTSZINHERIT | XFS_DIFLAG_NODEFRAG)
-+ XFS_DIFLAG_EXTSZINHERIT | XFS_DIFLAG_NODEFRAG | XFS_DIFLAG_BARRIER | \
-+ XFS_DIFLAG_IUNLINK)
+ XFS_DIFLAG_EXTSZINHERIT | XFS_DIFLAG_NODEFRAG | XFS_DIFLAG_FILESTREAM)
++#define XFS_DIVFLAG_BARRIER 0x01
++#define XFS_DIVFLAG_IUNLINK 0x02
++
#endif /* __XFS_DINODE_H__ */
---- linux-2.6.22/fs/xfs/xfs_fs.h 2006-11-30 21:19:29 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/xfs_fs.h 2007-06-15 02:37:04 +0200
-@@ -66,6 +66,8 @@ struct fsxattr {
- #define XFS_XFLAG_EXTSIZE 0x00000800 /* extent size allocator hint */
+--- linux-2.6.25.4/fs/xfs/xfs_fs.h 2008-04-17 12:05:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_fs.h 2008-04-19 15:14:52.000000000 -0400
+@@ -67,6 +67,8 @@ struct fsxattr {
#define XFS_XFLAG_EXTSZINHERIT 0x00001000 /* inherit inode extent size */
#define XFS_XFLAG_NODEFRAG 0x00002000 /* do not defragment */
-+#define XFS_XFLAG_BARRIER 0x00004000 /* chroot() barrier */
-+#define XFS_XFLAG_IUNLINK 0x00008000 /* immutable unlink */
+ #define XFS_XFLAG_FILESTREAM 0x00004000 /* use filestream allocator */
++#define XFS_XFLAG_BARRIER 0x10000000 /* chroot() barrier */
++#define XFS_XFLAG_IUNLINK 0x20000000 /* immutable unlink */
#define XFS_XFLAG_HASATTR 0x80000000 /* no DIFLAG for this */
/*
-@@ -294,7 +296,8 @@ typedef struct xfs_bstat {
+@@ -296,7 +298,8 @@ typedef struct xfs_bstat {
__s32 bs_extents; /* number of extents */
__u32 bs_gen; /* generation count */
__u16 bs_projid; /* project id */
@@ -9690,66 +8943,105 @@
__u32 bs_dmevmask; /* DMIG event mask */
__u16 bs_dmstate; /* DMIG state info */
__u16 bs_aextents; /* attribute number of extents */
---- linux-2.6.22/fs/xfs/xfs_inode.c 2007-07-09 13:19:34 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/xfs_inode.c 2007-06-15 02:37:04 +0200
-@@ -49,6 +49,7 @@
- #include "xfs_quota.h"
- #include "xfs_acl.h"
+--- linux-2.6.25.4/fs/xfs/xfs_ialloc.c 2008-04-17 12:05:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_ialloc.c 2008-04-27 10:30:39.000000000 -0400
+@@ -66,6 +66,8 @@ xfs_ialloc_log_di(
+ offsetof(xfs_dinode_core_t, di_gid),
+ offsetof(xfs_dinode_core_t, di_nlink),
+ offsetof(xfs_dinode_core_t, di_projid),
++ offsetof(xfs_dinode_core_t, di_tag),
++ offsetof(xfs_dinode_core_t, di_vflags),
+ offsetof(xfs_dinode_core_t, di_pad),
+ offsetof(xfs_dinode_core_t, di_atime),
+ offsetof(xfs_dinode_core_t, di_mtime),
+--- linux-2.6.25.4/fs/xfs/xfs_inode.c 2008-04-17 12:05:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_inode.c 2008-04-26 11:06:34.000000000 -0400
+@@ -219,6 +219,7 @@ xfs_inotobp(
+ return 0;
+ }
+#include <linux/vs_tag.h>
- kmem_zone_t *xfs_ifork_zone;
- kmem_zone_t *xfs_inode_zone;
-@@ -736,20 +737,35 @@ xfs_xlate_dinode_core(
- xfs_dinode_core_t *buf_core = (xfs_dinode_core_t *)buf;
- xfs_dinode_core_t *mem_core = (xfs_dinode_core_t *)dip;
- xfs_arch_t arch = ARCH_CONVERT;
-+ uint32_t uid = 0, gid = 0;
-+ uint16_t tag = 0;
-
- ASSERT(dir);
-
-+ if (dir < 0) {
-+ tag = mem_core->di_tag;
-+ /* FIXME: supposed to use superblock flag */
-+ uid = TAGINO_UID(1, mem_core->di_uid, tag);
-+ gid = TAGINO_GID(1, mem_core->di_gid, tag);
-+ tag = TAGINO_TAG(1, tag);
-+ }
-+
- INT_XLATE(buf_core->di_magic, mem_core->di_magic, dir, arch);
- INT_XLATE(buf_core->di_mode, mem_core->di_mode, dir, arch);
- INT_XLATE(buf_core->di_version, mem_core->di_version, dir, arch);
- INT_XLATE(buf_core->di_format, mem_core->di_format, dir, arch);
- INT_XLATE(buf_core->di_onlink, mem_core->di_onlink, dir, arch);
-- INT_XLATE(buf_core->di_uid, mem_core->di_uid, dir, arch);
-- INT_XLATE(buf_core->di_gid, mem_core->di_gid, dir, arch);
-+ INT_XLATE(buf_core->di_uid, uid, dir, arch);
-+ INT_XLATE(buf_core->di_gid, gid, dir, arch);
-+ INT_XLATE(buf_core->di_tag, tag, dir, arch);
- INT_XLATE(buf_core->di_nlink, mem_core->di_nlink, dir, arch);
- INT_XLATE(buf_core->di_projid, mem_core->di_projid, dir, arch);
-
- if (dir > 0) {
-+ /* FIXME: supposed to use superblock flag */
-+ mem_core->di_uid = INOTAG_UID(1, uid, gid);
-+ mem_core->di_gid = INOTAG_GID(1, uid, gid);
-+ mem_core->di_tag = INOTAG_TAG(1, uid, gid, tag);
- memcpy(mem_core->di_pad, buf_core->di_pad,
- sizeof(buf_core->di_pad));
- } else {
-@@ -797,6 +813,10 @@ _xfs_dic2xflags(
- flags |= XFS_XFLAG_PREALLOC;
- if (di_flags & XFS_DIFLAG_IMMUTABLE)
- flags |= XFS_XFLAG_IMMUTABLE;
-+ if (di_flags & XFS_DIFLAG_IUNLINK)
-+ flags |= XFS_XFLAG_IUNLINK;
-+ if (di_flags & XFS_DIFLAG_BARRIER)
-+ flags |= XFS_XFLAG_BARRIER;
- if (di_flags & XFS_DIFLAG_APPEND)
- flags |= XFS_XFLAG_APPEND;
- if (di_flags & XFS_DIFLAG_SYNC)
-@@ -1129,6 +1149,7 @@ xfs_ialloc(
+ /*
+ * This routine is called to map an inode to the buffer containing
+@@ -716,13 +717,21 @@ xfs_dinode_from_disk(
+ xfs_icdinode_t *to,
+ xfs_dinode_core_t *from)
+ {
++ uint32_t uid, gid;
++
+ to->di_magic = be16_to_cpu(from->di_magic);
+ to->di_mode = be16_to_cpu(from->di_mode);
+ to->di_version = from ->di_version;
+ to->di_format = from->di_format;
+ to->di_onlink = be16_to_cpu(from->di_onlink);
+- to->di_uid = be32_to_cpu(from->di_uid);
+- to->di_gid = be32_to_cpu(from->di_gid);
++
++ uid = be32_to_cpu(from->di_uid);
++ gid = be32_to_cpu(from->di_gid);
++
++ to->di_uid = INOTAG_UID(1, uid, gid);
++ to->di_gid = INOTAG_GID(1, uid, gid);
++ to->di_tag = INOTAG_TAG(1, uid, gid, 0);
++
+ to->di_nlink = be32_to_cpu(from->di_nlink);
+ to->di_projid = be16_to_cpu(from->di_projid);
+ memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad));
+@@ -756,8 +765,10 @@ xfs_dinode_to_disk(
+ to->di_version = from ->di_version;
+ to->di_format = from->di_format;
+ to->di_onlink = cpu_to_be16(from->di_onlink);
+- to->di_uid = cpu_to_be32(from->di_uid);
+- to->di_gid = cpu_to_be32(from->di_gid);
++
++ to->di_uid = cpu_to_be32(TAGINO_UID(1, from->di_uid, from->di_tag));
++ to->di_gid = cpu_to_be32(TAGINO_GID(1, from->di_gid, from->di_tag));
++
+ to->di_nlink = cpu_to_be32(from->di_nlink);
+ to->di_projid = cpu_to_be16(from->di_projid);
+ memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad));
+@@ -783,7 +794,8 @@ xfs_dinode_to_disk(
+
+ STATIC uint
+ _xfs_dic2xflags(
+- __uint16_t di_flags)
++ __uint16_t di_flags,
++ __uint16_t di_vflags)
+ {
+ uint flags = 0;
+
+@@ -817,7 +829,10 @@ _xfs_dic2xflags(
+ if (di_flags & XFS_DIFLAG_FILESTREAM)
+ flags |= XFS_XFLAG_FILESTREAM;
+ }
+-
++ if (di_vflags & XFS_DIVFLAG_IUNLINK)
++ flags |= XFS_XFLAG_IUNLINK;
++ if (di_vflags & XFS_DIVFLAG_BARRIER)
++ flags |= XFS_XFLAG_BARRIER;
+ return flags;
+ }
+
+@@ -827,7 +842,7 @@ xfs_ip2xflags(
+ {
+ xfs_icdinode_t *dic = &ip->i_d;
+
+- return _xfs_dic2xflags(dic->di_flags) |
++ return _xfs_dic2xflags(dic->di_flags, dic->di_vflags) |
+ (XFS_IFORK_Q(ip) ? XFS_XFLAG_HASATTR : 0);
+ }
+
+@@ -837,7 +852,7 @@ xfs_dic2xflags(
+ {
+ xfs_dinode_core_t *dic = &dip->di_core;
+
+- return _xfs_dic2xflags(be16_to_cpu(dic->di_flags)) |
++ return _xfs_dic2xflags(be16_to_cpu(dic->di_flags), be16_to_cpu(dic->di_vflags)) |
+ (XFS_DFORK_Q(dip) ? XFS_XFLAG_HASATTR : 0);
+ }
+
+@@ -1138,6 +1153,7 @@ xfs_ialloc(
ASSERT(ip->i_d.di_nlink == nlink);
ip->i_d.di_uid = current_fsuid(cr);
ip->i_d.di_gid = current_fsgid(cr);
@@ -9757,8 +9049,21 @@
ip->i_d.di_projid = prid;
memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
---- linux-2.6.22/fs/xfs/xfs_itable.c 2007-07-09 13:19:34 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/xfs_itable.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/fs/xfs/xfs_inode.h 2008-04-17 12:05:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_inode.h 2008-04-19 15:14:52.000000000 -0400
+@@ -187,7 +187,9 @@ typedef struct xfs_icdinode {
+ __uint32_t di_gid; /* owner's group id */
+ __uint32_t di_nlink; /* number of links to file */
+ __uint16_t di_projid; /* owner's project id */
+- __uint8_t di_pad[8]; /* unused, zeroed space */
++ __uint16_t di_tag; /* context tagging */
++ __uint16_t di_vflags; /* vserver specific flags */
++ __uint8_t di_pad[4]; /* unused, zeroed space */
+ __uint16_t di_flushiter; /* incremented on flush */
+ xfs_ictimestamp_t di_atime; /* time last accessed */
+ xfs_ictimestamp_t di_mtime; /* time last modified */
+--- linux-2.6.25.4/fs/xfs/xfs_itable.c 2008-04-17 12:05:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_itable.c 2008-04-19 15:14:52.000000000 -0400
@@ -89,6 +89,7 @@ xfs_bulkstat_one_iget(
buf->bs_mode = dic->di_mode;
buf->bs_uid = dic->di_uid;
@@ -9767,27 +9072,19 @@
buf->bs_size = dic->di_size;
vn_atime_to_bstime(vp, &buf->bs_atime);
buf->bs_mtime.tv_sec = dic->di_mtime.t_sec;
-@@ -163,6 +164,7 @@ xfs_bulkstat_one_dinode(
- buf->bs_mode = INT_GET(dic->di_mode, ARCH_CONVERT);
- buf->bs_uid = INT_GET(dic->di_uid, ARCH_CONVERT);
- buf->bs_gid = INT_GET(dic->di_gid, ARCH_CONVERT);
-+ buf->bs_tag = INT_GET(dic->di_tag, ARCH_CONVERT);
- buf->bs_size = INT_GET(dic->di_size, ARCH_CONVERT);
- buf->bs_atime.tv_sec = INT_GET(dic->di_atime.t_sec, ARCH_CONVERT);
- buf->bs_atime.tv_nsec = INT_GET(dic->di_atime.t_nsec, ARCH_CONVERT);
---- linux-2.6.22/fs/xfs/xfs_mount.h 2007-05-02 19:25:23 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/xfs_mount.h 2007-06-15 02:37:04 +0200
-@@ -464,6 +464,7 @@ typedef struct xfs_mount {
- #define XFS_MOUNT_NO_PERCPU_SB (1ULL << 23) /* don't use per-cpu superblock
- counters */
+--- linux-2.6.25.4/fs/xfs/xfs_mount.h 2008-04-17 12:05:43.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_mount.h 2008-04-19 15:14:52.000000000 -0400
+@@ -378,6 +378,7 @@ typedef struct xfs_mount {
+ #define XFS_MOUNT_FILESTREAMS (1ULL << 24) /* enable the filestreams
+ allocator */
+#define XFS_MOUNT_TAGGED (1ULL << 31) /* context tagging */
/*
* Default minimum read and write sizes.
---- linux-2.6.22/fs/xfs/xfs_vfsops.c 2007-07-09 13:19:34 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/xfs_vfsops.c 2007-06-15 02:37:04 +0200
-@@ -300,6 +300,8 @@ xfs_start_flags(
+--- linux-2.6.25.4/fs/xfs/xfs_vfsops.c 2008-04-17 12:05:43.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_vfsops.c 2008-04-26 09:52:49.000000000 -0400
+@@ -290,6 +290,8 @@ xfs_start_flags(
if (ap->flags2 & XFSMNT2_COMPAT_IOSIZE)
mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
@@ -9796,48 +9093,17 @@
/*
* no recovery flag requires a read-only mount
-@@ -394,6 +396,8 @@ xfs_finish_flags(
- return XFS_ERROR(EINVAL);
+@@ -402,7 +404,6 @@ xfs_finish_flags(
+ if (ap->flags & XFSMNT_PQUOTAENF)
+ mp->m_qflags |= XFS_OQUOTA_ENFD;
}
-
-+ if (ap->flags2 & XFSMNT2_TAGGED)
-+ vfs->vfs_super->s_flags |= MS_TAGGED;
+-
return 0;
}
-@@ -1645,6 +1649,9 @@ xfs_vget(
- * in stat(). */
- #define MNTOPT_ATTR2 "attr2" /* do use attr2 attribute format */
- #define MNTOPT_NOATTR2 "noattr2" /* do not use attr2 attribute format */
-+#define MNTOPT_TAGXID "tagxid" /* context tagging for inodes */
-+#define MNTOPT_TAGGED "tag" /* context tagging for inodes */
-+#define MNTOPT_NOTAGTAG "notag" /* do not use context tagging */
-
- STATIC unsigned long
- suffix_strtoul(char *s, char **endp, unsigned int base)
-@@ -1831,6 +1838,19 @@ xfs_parseargs(
- args->flags |= XFSMNT_ATTR2;
- } else if (!strcmp(this_char, MNTOPT_NOATTR2)) {
- args->flags &= ~XFSMNT_ATTR2;
-+#ifndef CONFIG_TAGGING_NONE
-+ } else if (!strcmp(this_char, MNTOPT_TAGGED)) {
-+ args->flags2 |= XFSMNT2_TAGGED;
-+ } else if (!strcmp(this_char, MNTOPT_NOTAGTAG)) {
-+ args->flags2 &= ~XFSMNT2_TAGGED;
-+ } else if (!strcmp(this_char, MNTOPT_TAGXID)) {
-+ args->flags2 |= XFSMNT2_TAGGED;
-+#endif
-+#ifdef CONFIG_PROPAGATE
-+ } else if (!strcmp(this_char, MNTOPT_TAGGED)) {
-+ /* use value */
-+ args->flags2 |= XFSMNT2_TAGGED;
-+#endif
- } else if (!strcmp(this_char, "osyncisdsync")) {
- /* no-op, this is now the default */
- cmn_err(CE_WARN,
---- linux-2.6.22/fs/xfs/xfs_vnodeops.c 2007-07-09 13:19:34 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/fs/xfs/xfs_vnodeops.c 2007-06-15 02:37:04 +0200
-@@ -159,6 +159,7 @@ xfs_getattr(
+--- linux-2.6.25.4/fs/xfs/xfs_vnodeops.c 2008-04-17 12:05:43.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/fs/xfs/xfs_vnodeops.c 2008-04-19 15:14:52.000000000 -0400
+@@ -122,6 +122,7 @@ xfs_getattr(
vap->va_mode = ip->i_d.di_mode;
vap->va_uid = ip->i_d.di_uid;
vap->va_gid = ip->i_d.di_gid;
@@ -9845,15 +9111,15 @@
vap->va_projid = ip->i_d.di_projid;
/*
-@@ -259,6 +260,7 @@ xfs_setattr(
+@@ -221,6 +222,7 @@ xfs_setattr(
uint commit_flags=0;
uid_t uid=0, iuid=0;
gid_t gid=0, igid=0;
+ tag_t tag=0, itag=0;
int timeflags = 0;
- bhv_vnode_t *vp;
xfs_prid_t projid=0, iprojid=0;
-@@ -315,6 +317,7 @@ xfs_setattr(
+ int mandlock_before, mandlock_after;
+@@ -272,6 +274,7 @@ xfs_setattr(
(mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID))) {
uint qflags = 0;
@@ -9861,7 +9127,7 @@
if ((mask & XFS_AT_UID) && XFS_IS_UQUOTA_ON(mp)) {
uid = vap->va_uid;
qflags |= XFS_QMOPT_UQUOTA;
-@@ -394,6 +397,8 @@ xfs_setattr(
+@@ -351,6 +354,8 @@ xfs_setattr(
if (mask &
(XFS_AT_MODE|XFS_AT_XFLAGS|XFS_AT_EXTSIZE|XFS_AT_UID|
XFS_AT_GID|XFS_AT_PROJID)) {
@@ -9870,7 +9136,7 @@
/*
* CAP_FOWNER overrides the following restrictions:
*
-@@ -442,7 +447,7 @@ xfs_setattr(
+@@ -399,7 +404,7 @@ xfs_setattr(
* and can change the group id only to a group of which he
* or she is a member.
*/
@@ -9879,7 +9145,7 @@
/*
* These IDs could have changed since we last looked at them.
* But, we're assured that if the ownership did change
-@@ -450,10 +455,12 @@ xfs_setattr(
+@@ -407,10 +412,12 @@ xfs_setattr(
* would have changed also.
*/
iuid = ip->i_d.di_uid;
@@ -9894,7 +9160,7 @@
projid = (mask & XFS_AT_PROJID) ? (xfs_prid_t)vap->va_projid :
iprojid;
-@@ -481,6 +488,7 @@ xfs_setattr(
+@@ -438,6 +445,7 @@ xfs_setattr(
if ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) ||
(XFS_IS_PQUOTA_ON(mp) && iprojid != projid) ||
(XFS_IS_GQUOTA_ON(mp) && igid != gid)) {
@@ -9902,7 +9168,7 @@
ASSERT(tp);
code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp,
capable(CAP_FOWNER) ?
-@@ -706,7 +714,7 @@ xfs_setattr(
+@@ -686,7 +694,7 @@ xfs_setattr(
* and can change the group id only to a group of which he
* or she is a member.
*/
@@ -9911,7 +9177,7 @@
/*
* CAP_FSETID overrides the following restrictions:
*
-@@ -722,6 +730,9 @@ xfs_setattr(
+@@ -702,6 +710,9 @@ xfs_setattr(
* Change the ownerships and register quota modifications
* in the transaction.
*/
@@ -9921,19 +9187,28 @@
if (iuid != uid) {
if (XFS_IS_UQUOTA_ON(mp)) {
ASSERT(mask & XFS_AT_UID);
-@@ -802,6 +813,10 @@ xfs_setattr(
+@@ -777,6 +788,7 @@ xfs_setattr(
+ }
+ if (mask & XFS_AT_XFLAGS) {
+ uint di_flags;
++ uint di_vflags = 0;
+
+ /* can't set PREALLOC this way, just preserve it */
di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC);
- if (vap->va_xflags & XFS_XFLAG_IMMUTABLE)
- di_flags |= XFS_DIFLAG_IMMUTABLE;
+@@ -810,6 +822,11 @@ xfs_setattr(
+ di_flags |= XFS_DIFLAG_EXTSIZE;
+ }
+ ip->i_d.di_flags = di_flags;
+ if (vap->va_xflags & XFS_XFLAG_IUNLINK)
-+ di_flags |= XFS_DIFLAG_IUNLINK;
++ di_vflags |= XFS_DIVFLAG_IUNLINK;
+ if (vap->va_xflags & XFS_XFLAG_BARRIER)
-+ di_flags |= XFS_DIFLAG_BARRIER;
- if (vap->va_xflags & XFS_XFLAG_APPEND)
- di_flags |= XFS_DIFLAG_APPEND;
- if (vap->va_xflags & XFS_XFLAG_SYNC)
---- linux-2.6.22/include/asm-arm/tlb.h 2006-06-18 04:54:58 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-arm/tlb.h 2007-06-15 02:37:04 +0200
++ di_vflags |= XFS_DIVFLAG_BARRIER;
++ ip->i_d.di_vflags = di_vflags;
+ }
+ xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ timeflags |= XFS_ICHGTIME_CHG;
+--- linux-2.6.25.4/include/asm-arm/tlb.h 2008-04-17 12:05:43.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-arm/tlb.h 2008-04-19 15:14:52.000000000 -0400
@@ -28,6 +28,7 @@
#else /* !CONFIG_MMU */
@@ -9942,29 +9217,19 @@
/*
* TLB handling. This allows us to remove pages from the page
---- linux-2.6.22/include/asm-arm26/tlb.h 2006-01-03 17:30:02 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-arm26/tlb.h 2007-06-15 02:37:04 +0200
-@@ -3,6 +3,7 @@
-
- #include <asm/pgalloc.h>
- #include <asm/tlbflush.h>
-+#include <linux/vs_memory.h>
-
- /*
- * TLB handling. This allows us to remove pages from the page
---- linux-2.6.22/include/asm-arm26/unistd.h 2007-02-06 03:01:35 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-arm26/unistd.h 2007-06-15 02:37:04 +0200
-@@ -302,6 +302,8 @@
- #define __NR_mq_getsetattr (__NR_SYSCALL_BASE+279)
- #define __NR_waitid (__NR_SYSCALL_BASE+280)
-
-+#define __NR_vserver (__NR_SYSCALL_BASE+313)
-+
- /*
- * The following SWIs are ARM private. FIXME - make appropriate for arm26
- */
---- linux-2.6.22/include/asm-generic/tlb.h 2006-11-30 21:19:31 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-generic/tlb.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/include/asm-blackfin/unistd.h 2008-04-17 12:05:43.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-blackfin/unistd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -278,7 +278,7 @@
+ #define __NR_tgkill 271
+ #define __NR_utimes 272
+ #define __NR_fadvise64_64 273
+- /* 274 __NR_vserver */
++#define __NR_vserver 274
+ /* 275 __NR_mbind */
+ /* 276 __NR_get_mempolicy */
+ /* 277 __NR_set_mempolicy */
+--- linux-2.6.25.4/include/asm-generic/tlb.h 2008-04-17 12:05:43.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-generic/tlb.h 2008-04-19 15:50:15.000000000 -0400
@@ -14,6 +14,7 @@
#define _ASM_GENERIC__TLB_H
@@ -9973,8 +9238,8 @@
#include <asm/pgalloc.h>
#include <asm/tlbflush.h>
---- linux-2.6.22/include/asm-ia64/tlb.h 2006-09-20 16:58:40 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-ia64/tlb.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/include/asm-ia64/tlb.h 2007-02-04 13:44:54.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-ia64/tlb.h 2008-04-19 15:14:52.000000000 -0400
@@ -40,6 +40,7 @@
#include <linux/mm.h>
#include <linux/pagemap.h>
@@ -9983,8 +9248,8 @@
#include <asm/pgalloc.h>
#include <asm/processor.h>
---- linux-2.6.22/include/asm-powerpc/systbl.h 2007-07-09 13:19:44 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-powerpc/systbl.h 2007-07-07 03:52:53 +0200
+--- linux-2.6.25.4/include/asm-powerpc/systbl.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-powerpc/systbl.h 2008-04-19 15:14:52.000000000 -0400
@@ -260,7 +260,7 @@ COMPAT_SYS_SPU(fstatfs64)
SYSX(sys_ni_syscall, ppc_fadvise64_64, ppc_fadvise64_64)
PPC_SYS_SPU(rtas)
@@ -9994,8 +9259,8 @@
COMPAT_SYS(migrate_pages)
COMPAT_SYS(mbind)
COMPAT_SYS(get_mempolicy)
---- linux-2.6.22/include/asm-powerpc/unistd.h 2007-07-09 13:19:45 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-powerpc/unistd.h 2007-07-07 03:52:53 +0200
+--- linux-2.6.25.4/include/asm-powerpc/unistd.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-powerpc/unistd.h 2008-04-19 15:14:52.000000000 -0400
@@ -275,7 +275,7 @@
#endif
#define __NR_rtas 255
@@ -10005,8 +9270,8 @@
#define __NR_migrate_pages 258
#define __NR_mbind 259
#define __NR_get_mempolicy 260
---- linux-2.6.22/include/asm-s390/unistd.h 2007-07-09 13:19:45 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-s390/unistd.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/include/asm-s390/unistd.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-s390/unistd.h 2008-04-19 15:14:52.000000000 -0400
@@ -202,7 +202,7 @@
#define __NR_clock_gettime (__NR_timer_create+6)
#define __NR_clock_getres (__NR_timer_create+7)
@@ -10016,9 +9281,9 @@
#define __NR_statfs64 265
#define __NR_fstatfs64 266
#define __NR_remap_file_pages 267
---- linux-2.6.22/include/asm-sparc/unistd.h 2007-07-09 13:19:54 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-sparc/unistd.h 2007-06-15 02:37:04 +0200
-@@ -283,7 +283,7 @@
+--- linux-2.6.25.4/include/asm-sparc/unistd.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-sparc/unistd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -282,7 +282,7 @@
#define __NR_timer_getoverrun 264
#define __NR_timer_delete 265
#define __NR_timer_create 266
@@ -10027,8 +9292,8 @@
#define __NR_io_setup 268
#define __NR_io_destroy 269
#define __NR_io_submit 270
---- linux-2.6.22/include/asm-sparc64/tlb.h 2007-07-09 13:19:54 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-sparc64/tlb.h 2007-07-07 03:54:19 +0200
+--- linux-2.6.25.4/include/asm-sparc64/tlb.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-sparc64/tlb.h 2008-04-19 15:14:52.000000000 -0400
@@ -3,6 +3,7 @@
#include <linux/swap.h>
@@ -10037,9 +9302,9 @@
#include <asm/pgalloc.h>
#include <asm/tlbflush.h>
#include <asm/mmu_context.h>
---- linux-2.6.22/include/asm-sparc64/unistd.h 2007-07-09 13:19:54 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-sparc64/unistd.h 2007-06-15 02:37:04 +0200
-@@ -285,7 +285,7 @@
+--- linux-2.6.25.4/include/asm-sparc64/unistd.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-sparc64/unistd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -284,7 +284,7 @@
#define __NR_timer_getoverrun 264
#define __NR_timer_delete 265
#define __NR_timer_create 266
@@ -10048,29 +9313,20 @@
#define __NR_io_setup 268
#define __NR_io_destroy 269
#define __NR_io_submit 270
---- linux-2.6.22/include/asm-x86_64/unistd.h 2007-07-09 13:19:55 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/asm-x86_64/unistd.h 2007-07-07 03:52:53 +0200
-@@ -532,7 +532,7 @@ __SYSCALL(__NR_tgkill, sys_tgkill)
- #define __NR_utimes 235
+--- linux-2.6.25.4/include/asm-x86/unistd_64.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/asm-x86/unistd_64.h 2008-04-19 15:14:52.000000000 -0400
+@@ -535,7 +535,7 @@ __SYSCALL(__NR_tgkill, sys_tgkill)
+ #define __NR_utimes 235
__SYSCALL(__NR_utimes, sys_utimes)
- #define __NR_vserver 236
+ #define __NR_vserver 236
-__SYSCALL(__NR_vserver, sys_ni_syscall)
+__SYSCALL(__NR_vserver, sys_vserver)
- #define __NR_mbind 237
+ #define __NR_mbind 237
__SYSCALL(__NR_mbind, sys_mbind)
- #define __NR_set_mempolicy 238
---- linux-2.6.22/include/linux/Kbuild 2007-07-09 13:19:55 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/Kbuild 2007-06-15 02:37:04 +0200
-@@ -349,3 +349,6 @@ unifdef-y += xattr.h
- unifdef-y += xfrm.h
-
- objhdr-y += version.h
-+
-+header-y += vserver/
-+
---- linux-2.6.22/include/linux/capability.h 2007-07-09 13:19:55 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/capability.h 2007-06-15 02:37:04 +0200
-@@ -237,6 +237,7 @@ typedef __u32 kernel_cap_t;
+ #define __NR_set_mempolicy 238
+--- linux-2.6.25.4/include/linux/capability.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/capability.h 2008-04-22 11:01:12.000000000 -0400
+@@ -260,6 +260,7 @@ typedef struct kernel_cap_struct {
arbitrary SCSI commands */
/* Allow setting encryption key on loopback filesystem */
/* Allow setting zone reclaim policy */
@@ -10078,31 +9334,32 @@
#define CAP_SYS_ADMIN 21
-@@ -290,6 +291,11 @@ typedef __u32 kernel_cap_t;
+@@ -332,7 +333,13 @@ typedef struct kernel_cap_struct {
- #define CAP_AUDIT_CONTROL 30
+ #define CAP_MAC_ADMIN 33
+-#define CAP_LAST_CAP CAP_MAC_ADMIN
+/* Allow context manipulations */
+/* Allow changing context info on files */
+
-+#define CAP_CONTEXT 31
++#define CAP_CONTEXT 34
++
+
- #ifdef __KERNEL__
- /*
- * Bounding set
---- linux-2.6.22/include/linux/devpts_fs.h 2004-08-14 12:55:59 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/devpts_fs.h 2007-06-15 02:37:04 +0200
-@@ -30,5 +30,7 @@ static inline void devpts_pty_kill(int n
++#define CAP_LAST_CAP CAP_CONTEXT
- #endif
+ #define cap_valid(x) ((x) >= 0 && (x) <= CAP_LAST_CAP)
-+#define DEVPTS_SUPER_MAGIC 0x00001cd1
-+
+--- linux-2.6.25.4/include/linux/devpts_fs.h 2007-02-04 13:44:54.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/devpts_fs.h 2008-04-19 15:14:52.000000000 -0400
+@@ -30,5 +30,4 @@ static inline void devpts_pty_kill(int n
+
+ #endif
+-
#endif /* _LINUX_DEVPTS_FS_H */
---- linux-2.6.22/include/linux/ext2_fs.h 2006-11-30 21:19:37 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/ext2_fs.h 2007-06-15 02:37:04 +0200
-@@ -188,6 +188,8 @@ struct ext2_group_desc
+--- linux-2.6.25.4/include/linux/ext2_fs.h 2008-04-17 11:31:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/ext2_fs.h 2008-04-19 15:14:52.000000000 -0400
+@@ -189,6 +189,8 @@ struct ext2_group_desc
#define EXT2_NOTAIL_FL FS_NOTAIL_FL /* file tail should not be merged */
#define EXT2_DIRSYNC_FL FS_DIRSYNC_FL /* dirsync behaviour (directories only) */
#define EXT2_TOPDIR_FL FS_TOPDIR_FL /* Top of directory hierarchies*/
@@ -10111,7 +9368,7 @@
#define EXT2_RESERVED_FL FS_RESERVED_FL /* reserved for ext2 lib */
#define EXT2_FL_USER_VISIBLE FS_FL_USER_VISIBLE /* User visible flags */
-@@ -244,7 +246,7 @@ struct ext2_inode {
+@@ -247,7 +249,7 @@ struct ext2_inode {
struct {
__u8 l_i_frag; /* Fragment number */
__u8 l_i_fsize; /* Fragment size */
@@ -10120,7 +9377,7 @@
__le16 l_i_uid_high; /* these 2 fields */
__le16 l_i_gid_high; /* were reserved2[0] */
__u32 l_i_reserved2;
-@@ -276,6 +278,7 @@ struct ext2_inode {
+@@ -279,6 +281,7 @@ struct ext2_inode {
#define i_gid_low i_gid
#define i_uid_high osd2.linux2.l_i_uid_high
#define i_gid_high osd2.linux2.l_i_gid_high
@@ -10128,21 +9385,17 @@
#define i_reserved2 osd2.linux2.l_i_reserved2
#endif
-@@ -317,8 +320,9 @@ struct ext2_inode {
- #define EXT2_MOUNT_XATTR_USER 0x004000 /* Extended user attributes */
- #define EXT2_MOUNT_POSIX_ACL 0x008000 /* POSIX Access Control Lists */
- #define EXT2_MOUNT_XIP 0x010000 /* Execute in place */
--#define EXT2_MOUNT_USRQUOTA 0x020000 /* user quota */
--#define EXT2_MOUNT_GRPQUOTA 0x040000 /* group quota */
-+#define EXT2_MOUNT_USRQUOTA 0x020000 /* user quota */
-+#define EXT2_MOUNT_GRPQUOTA 0x040000 /* group quota */
+@@ -323,6 +326,7 @@ struct ext2_inode {
+ #define EXT2_MOUNT_USRQUOTA 0x020000 /* user quota */
+ #define EXT2_MOUNT_GRPQUOTA 0x040000 /* group quota */
+ #define EXT2_MOUNT_RESERVATION 0x080000 /* Preallocation */
+#define EXT2_MOUNT_TAGGED (1<<24) /* Enable Context Tags */
#define clear_opt(o, opt) o &= ~EXT2_MOUNT_##opt
---- linux-2.6.22/include/linux/ext3_fs.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/ext3_fs.h 2007-06-15 02:37:04 +0200
-@@ -177,10 +177,20 @@ struct ext3_group_desc
+--- linux-2.6.25.4/include/linux/ext3_fs.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/ext3_fs.h 2008-04-19 15:14:52.000000000 -0400
+@@ -173,6 +173,8 @@ struct ext3_group_desc
#define EXT3_NOTAIL_FL 0x00008000 /* file tail should not be merged */
#define EXT3_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */
#define EXT3_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/
@@ -10150,20 +9403,8 @@
+#define EXT3_IUNLINK_FL 0x08000000 /* Immutable unlink */
#define EXT3_RESERVED_FL 0x80000000 /* reserved for ext3 lib */
-+#ifdef CONFIG_VSERVER_LEGACY
-+#define EXT3_FL_USER_VISIBLE 0x0803DFFF /* User visible flags */
-+#define EXT3_FL_USER_MODIFIABLE 0x080380FF /* User modifiable flags */
-+#else
#define EXT3_FL_USER_VISIBLE 0x0003DFFF /* User visible flags */
- #define EXT3_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACY
-+#define EXT3_IOC_SETTAG FIOC_SETTAGJ
-+#endif
-
- /*
- * Inode dynamic state flags
-@@ -296,7 +306,7 @@ struct ext3_inode {
+@@ -292,7 +294,7 @@ struct ext3_inode {
struct {
__u8 l_i_frag; /* Fragment number */
__u8 l_i_fsize; /* Fragment size */
@@ -10172,7 +9413,7 @@
__le16 l_i_uid_high; /* these 2 fields */
__le16 l_i_gid_high; /* were reserved2[0] */
__u32 l_i_reserved2;
-@@ -330,6 +340,7 @@ struct ext3_inode {
+@@ -326,6 +328,7 @@ struct ext3_inode {
#define i_gid_low i_gid
#define i_uid_high osd2.linux2.l_i_uid_high
#define i_gid_high osd2.linux2.l_i_gid_high
@@ -10180,7 +9421,7 @@
#define i_reserved2 osd2.linux2.l_i_reserved2
#elif defined(__GNU__)
-@@ -384,6 +395,7 @@ struct ext3_inode {
+@@ -380,6 +383,7 @@ struct ext3_inode {
#define EXT3_MOUNT_QUOTA 0x80000 /* Some quota option set */
#define EXT3_MOUNT_USRQUOTA 0x100000 /* "old" user quota */
#define EXT3_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */
@@ -10188,40 +9429,26 @@
/* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */
#ifndef _LINUX_EXT2_FS_H
-@@ -812,6 +824,7 @@ struct buffer_head * ext3_bread (handle_
+@@ -822,6 +826,7 @@ struct buffer_head * ext3_bread (handle_
int ext3_get_blocks_handle(handle_t *handle, struct inode *inode,
sector_t iblock, unsigned long maxblocks, struct buffer_head *bh_result,
int create, int extend_disksize);
+extern int ext3_sync_flags(struct inode *inode);
- extern void ext3_read_inode (struct inode *);
+ extern struct inode *ext3_iget(struct super_block *, unsigned long);
extern int ext3_write_inode (struct inode *, int);
---- linux-2.6.22/include/linux/ext4_fs.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/ext4_fs.h 2007-06-15 02:37:04 +0200
-@@ -189,11 +189,21 @@ struct ext4_group_desc
- #define EXT4_NOTAIL_FL 0x00008000 /* file tail should not be merged */
- #define EXT4_DIRSYNC_FL 0x00010000 /* dirsync behaviour (directories only) */
+--- linux-2.6.25.4/include/linux/ext4_fs.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/ext4_fs.h 2008-04-19 15:43:04.000000000 -0400
+@@ -231,6 +231,8 @@ struct ext4_group_desc
#define EXT4_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/
--#define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */
+ #define EXT4_HUGE_FILE_FL 0x00040000 /* Set to each huge file */
#define EXT4_EXTENTS_FL 0x00080000 /* Inode uses extents */
+#define EXT4_BARRIER_FL 0x04000000 /* Barrier for chroot() */
+#define EXT4_IUNLINK_FL 0x08000000 /* Immutable unlink */
-+#define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */
+ #define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */
-+#ifdef CONFIG_VSERVER_LEGACY
-+#define EXT4_FL_USER_VISIBLE 0x080BDFFF /* User visible flags */
-+#define EXT4_FL_USER_MODIFIABLE 0x080380FF /* User modifiable flags */
-+#else
#define EXT4_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */
- #define EXT4_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACY
-+#define EXT4_IOC_SETTAG FIOC_SETTAGJ
-+#endif
-
- /*
- * Inode dynamic state flags
-@@ -312,7 +322,8 @@ struct ext4_inode {
+@@ -359,7 +361,8 @@ struct ext4_inode {
__le16 l_i_file_acl_high;
__le16 l_i_uid_high; /* these 2 fields */
__le16 l_i_gid_high; /* were reserved2[0] */
@@ -10230,8 +9457,8 @@
+ __u16 l_i_reserved2;
} linux2;
struct {
- __u8 h_i_frag; /* Fragment number */
-@@ -344,6 +355,7 @@ struct ext4_inode {
+ __le16 h_i_reserved1; /* Obsoleted fragment number/size which are removed in ext4 */
+@@ -465,6 +468,7 @@ do { \
#define i_gid_low i_gid
#define i_uid_high osd2.linux2.l_i_uid_high
#define i_gid_high osd2.linux2.l_i_gid_high
@@ -10239,34 +9466,34 @@
#define i_reserved2 osd2.linux2.l_i_reserved2
#elif defined(__GNU__)
-@@ -400,6 +412,7 @@ struct ext4_inode {
- #define EXT4_MOUNT_USRQUOTA 0x100000 /* "old" user quota */
- #define EXT4_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */
- #define EXT4_MOUNT_EXTENTS 0x400000 /* Extents support */
-+#define EXT4_MOUNT_TAGGED (1<<24) /* Enable Context Tags */
-
+@@ -528,6 +532,7 @@ do { \
+ #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */
+ #define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */
+ #define EXT4_MOUNT_MBALLOC 0x4000000 /* Buddy allocation support */
++#define EXT4_MOUNT_TAGGED 0x8000000 /* Enable Context Tags */
/* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */
#ifndef _LINUX_EXT2_FS_H
-@@ -850,6 +863,7 @@ struct buffer_head * ext4_bread (handle_
- int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
- sector_t iblock, unsigned long maxblocks, struct buffer_head *bh_result,
- int create, int extend_disksize);
+ #define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt
+@@ -1030,6 +1035,7 @@ int ext4_get_blocks_handle(handle_t *han
+ ext4_lblk_t iblock, unsigned long maxblocks,
+ struct buffer_head *bh_result,
+ int create, int extend_disksize);
+extern int ext4_sync_flags(struct inode *inode);
- extern void ext4_read_inode (struct inode *);
+ extern struct inode *ext4_iget(struct super_block *, unsigned long);
extern int ext4_write_inode (struct inode *, int);
---- linux-2.6.22/include/linux/fs.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/fs.h 2007-07-07 03:52:54 +0200
-@@ -123,6 +123,8 @@ extern int dir_notify_enable;
- #define MS_SLAVE (1<<19) /* change to slave */
- #define MS_SHARED (1<<20) /* change to shared */
+--- linux-2.6.25.4/include/linux/fs.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/fs.h 2008-04-19 15:37:28.000000000 -0400
+@@ -125,6 +125,8 @@ extern int dir_notify_enable;
#define MS_RELATIME (1<<21) /* Update atime relative to mtime/ctime. */
+ #define MS_KERNMOUNT (1<<22) /* this is a kern_mount call */
+ #define MS_I_VERSION (1<<23) /* Update inode I_version field */
+#define MS_TAGGED (1<<24) /* use generic inode tagging */
+#define MS_TAGID (1<<25) /* use specific tag for this mount */
#define MS_ACTIVE (1<<30)
#define MS_NOUSER (1<<31)
-@@ -149,6 +151,8 @@ extern int dir_notify_enable;
+@@ -151,6 +153,8 @@ extern int dir_notify_enable;
#define S_NOCMTIME 128 /* Do not update file c/mtime */
#define S_SWAPFILE 256 /* Do not truncate: swapon got its bmaps */
#define S_PRIVATE 512 /* Inode is fs-internal */
@@ -10275,7 +9502,7 @@
/*
* Note that nosuid etc flags are inode-specific: setting some file-system
-@@ -165,24 +169,36 @@ extern int dir_notify_enable;
+@@ -167,25 +171,37 @@ extern int dir_notify_enable;
*/
#define __IS_FLG(inode,flg) ((inode)->i_sb->s_flags & (flg))
@@ -10287,7 +9514,9 @@
((inode)->i_flags & (S_SYNC|S_DIRSYNC)))
#define IS_MANDLOCK(inode) __IS_FLG(inode, MS_MANDLOCK)
-#define IS_NOATIME(inode) __IS_FLG(inode, MS_RDONLY|MS_NOATIME)
-+#define IS_NOATIME(inode) __IS_FLG(inode, MS_RDONLY|MS_NOATIME)
+-#define IS_I_VERSION(inode) __IS_FLG(inode, MS_I_VERSION)
++#define IS_NOATIME(inode) __IS_FLG(inode, MS_RDONLY|MS_NOATIME)
++#define IS_I_VERSION(inode) __IS_FLG(inode, MS_I_VERSION)
+#define IS_TAGGED(inode) __IS_FLG(inode, MS_TAGGED)
#define IS_NOQUOTA(inode) ((inode)->i_flags & S_NOQUOTA)
@@ -10314,7 +9543,7 @@
/* the read-only stuff doesn't really belong here, but any other place is
probably as bad and I don't want to create yet another include file. */
-@@ -256,11 +272,17 @@ extern int dir_notify_enable;
+@@ -259,12 +275,13 @@ extern int dir_notify_enable;
#define FS_TOPDIR_FL 0x00020000 /* Top of directory hierarchies*/
#define FS_EXTENT_FL 0x00080000 /* Extents */
#define FS_DIRECTIO_FL 0x00100000 /* Use direct i/o */
@@ -10322,26 +9551,22 @@
+#define FS_IUNLINK_FL 0x08000000 /* Immutable unlink */
#define FS_RESERVED_FL 0x80000000 /* reserved for ext2 lib */
-+#ifdef CONFIG_VSERVER_LEGACY
-+#define FS_FL_USER_VISIBLE 0x0803DFFF /* User visible flags */
-+#define FS_FL_USER_MODIFIABLE 0x080380FF /* User modifiable flags */
-+#else
#define FS_FL_USER_VISIBLE 0x0003DFFF /* User visible flags */
#define FS_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */
--
-+#endif
+-
#define SYNC_FILE_RANGE_WAIT_BEFORE 1
#define SYNC_FILE_RANGE_WRITE 2
-@@ -327,6 +349,7 @@ typedef void (dio_iodone_t)(struct kiocb
- #define ATTR_KILL_SUID 2048
- #define ATTR_KILL_SGID 4096
+ #define SYNC_FILE_RANGE_WAIT_AFTER 4
+@@ -334,6 +351,7 @@ typedef void (dio_iodone_t)(struct kiocb
#define ATTR_FILE 8192
-+#define ATTR_TAG 16384
+ #define ATTR_KILL_PRIV 16384
+ #define ATTR_OPEN 32768 /* Truncating from open(O_TRUNC) */
++#define ATTR_TAG 65536
/*
* This is the Inode Attributes structure, used for notify_change(). It
-@@ -342,6 +365,7 @@ struct iattr {
+@@ -349,6 +367,7 @@ struct iattr {
umode_t ia_mode;
uid_t ia_uid;
gid_t ia_gid;
@@ -10349,7 +9574,7 @@
loff_t ia_size;
struct timespec ia_atime;
struct timespec ia_mtime;
-@@ -355,6 +379,9 @@ struct iattr {
+@@ -362,6 +381,9 @@ struct iattr {
struct file *ia_file;
};
@@ -10359,23 +9584,40 @@
/*
* Includes for diskquotas.
*/
-@@ -537,6 +564,7 @@ struct inode {
+@@ -600,7 +622,9 @@ struct inode {
unsigned int i_nlink;
uid_t i_uid;
gid_t i_gid;
+ tag_t i_tag;
dev_t i_rdev;
- unsigned long i_version;
++ dev_t i_mdev;
+ u64 i_version;
loff_t i_size;
-@@ -728,6 +756,7 @@ struct file {
+ #ifdef __NEED_I_SIZE_ORDERED
+@@ -735,12 +759,12 @@ static inline void i_size_write(struct i
+
+ static inline unsigned iminor(const struct inode *inode)
+ {
+- return MINOR(inode->i_rdev);
++ return MINOR(inode->i_mdev);
+ }
+
+ static inline unsigned imajor(const struct inode *inode)
+ {
+- return MAJOR(inode->i_rdev);
++ return MAJOR(inode->i_mdev);
+ }
+
+ extern struct block_device *I_BDEV(struct inode *inode);
+@@ -795,6 +819,7 @@ struct file {
loff_t f_pos;
struct fown_struct f_owner;
unsigned int f_uid, f_gid;
+ xid_t f_xid;
struct file_ra_state f_ra;
- unsigned long f_version;
-@@ -811,6 +840,7 @@ struct file_lock {
+ u64 f_version;
+@@ -879,6 +904,7 @@ struct file_lock {
unsigned char fl_type;
loff_t fl_start;
loff_t fl_end;
@@ -10383,7 +9625,7 @@
struct fasync_struct * fl_fasync; /* for lease break notifications */
unsigned long fl_break_time; /* for nonblocking lease breaks */
-@@ -993,12 +1023,12 @@ extern void unlock_super(struct super_bl
+@@ -1076,12 +1102,12 @@ extern void unlock_super(struct super_bl
*/
extern int vfs_permission(struct nameidata *, int);
extern int vfs_create(struct inode *, struct dentry *, int, struct nameidata *);
@@ -10402,15 +9644,15 @@
extern int vfs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *);
/*
-@@ -1138,6 +1168,7 @@ struct inode_operations {
- ssize_t (*listxattr) (struct dentry *, char *, size_t);
- int (*removexattr) (struct dentry *, const char *);
+@@ -1223,6 +1249,7 @@ struct inode_operations {
void (*truncate_range)(struct inode *, loff_t, loff_t);
+ long (*fallocate)(struct inode *inode, int mode, loff_t offset,
+ loff_t len);
+ int (*sync_flags) (struct inode *);
};
struct seq_file;
-@@ -1153,6 +1184,7 @@ extern ssize_t vfs_readv(struct file *,
+@@ -1238,6 +1265,7 @@ extern ssize_t vfs_readv(struct file *,
unsigned long, loff_t *);
extern ssize_t vfs_writev(struct file *, const struct iovec __user *,
unsigned long, loff_t *);
@@ -10418,15 +9660,7 @@
/*
* NOTE: write_inode, delete_inode, clear_inode, put_inode can be called
-@@ -1763,6 +1795,7 @@ extern ssize_t generic_file_buffered_wri
- extern ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos);
- extern ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos);
- extern ssize_t generic_file_sendfile(struct file *, loff_t *, size_t, read_actor_t, void *);
-+extern ssize_t generic_file_sendpage(struct file *, struct page *, int, size_t, loff_t *, int);
- extern void do_generic_mapping_read(struct address_space *mapping,
- struct file_ra_state *, struct file *,
- loff_t *, read_descriptor_t *, read_actor_t);
-@@ -1898,6 +1931,7 @@ extern int dcache_dir_open(struct inode
+@@ -1934,6 +1962,7 @@ extern int dcache_dir_open(struct inode
extern int dcache_dir_close(struct inode *, struct file *);
extern loff_t dcache_dir_lseek(struct file *, loff_t, int);
extern int dcache_readdir(struct file *, void *, filldir_t);
@@ -10434,10 +9668,20 @@
extern int simple_getattr(struct vfsmount *, struct dentry *, struct kstat *);
extern int simple_statfs(struct dentry *, struct kstatfs *);
extern int simple_link(struct dentry *, struct inode *, struct dentry *);
---- linux-2.6.22/include/linux/init_task.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/init_task.h 2007-06-15 02:37:04 +0200
-@@ -169,6 +169,10 @@ extern struct group_info init_groups;
- }, \
+--- linux-2.6.25.4/include/linux/if_tun.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/if_tun.h 2008-04-19 15:14:52.000000000 -0400
+@@ -42,6 +42,7 @@
+ #define TUNSETOWNER _IOW('T', 204, int)
+ #define TUNSETLINK _IOW('T', 205, int)
+ #define TUNSETGROUP _IOW('T', 206, int)
++#define TUNSETNID _IOW('T', 215, int)
+
+ /* TUNSETIFF ifr flags */
+ #define IFF_TUN 0x0001
+--- linux-2.6.25.4/include/linux/init_task.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/init_task.h 2008-04-19 15:14:52.000000000 -0400
+@@ -196,6 +196,10 @@ extern struct group_info init_groups;
+ INIT_IDS \
INIT_TRACE_IRQFLAGS \
INIT_LOCKDEP \
+ .xid = 0, \
@@ -10447,9 +9691,21 @@
}
---- linux-2.6.22/include/linux/ipc.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/ipc.h 2007-06-15 02:37:04 +0200
-@@ -63,6 +63,7 @@ struct kern_ipc_perm
+--- linux-2.6.25.4/include/linux/interrupt.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/interrupt.h 2008-04-19 15:14:52.000000000 -0400
+@@ -8,8 +8,8 @@
+ #include <linux/preempt.h>
+ #include <linux/cpumask.h>
+ #include <linux/irqreturn.h>
+-#include <linux/hardirq.h>
+ #include <linux/sched.h>
++#include <linux/hardirq.h>
+ #include <linux/irqflags.h>
+ #include <asm/atomic.h>
+ #include <asm/ptrace.h>
+--- linux-2.6.25.4/include/linux/ipc.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/ipc.h 2008-04-19 15:14:52.000000000 -0400
+@@ -93,6 +93,7 @@ struct kern_ipc_perm
key_t key;
uid_t uid;
gid_t gid;
@@ -10457,8 +9713,17 @@
uid_t cuid;
gid_t cgid;
mode_t mode;
---- linux-2.6.22/include/linux/loop.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/loop.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/include/linux/Kbuild 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/Kbuild 2008-04-19 15:14:52.000000000 -0400
+@@ -354,3 +354,6 @@ unifdef-y += xattr.h
+ unifdef-y += xfrm.h
+
+ objhdr-y += version.h
++
++header-y += vserver/
++
+--- linux-2.6.25.4/include/linux/loop.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/loop.h 2008-04-19 15:14:52.000000000 -0400
@@ -45,6 +45,7 @@ struct loop_device {
struct loop_func_table *lo_encryption;
__u32 lo_init[2];
@@ -10467,8 +9732,27 @@
int (*ioctl)(struct loop_device *, int cmd,
unsigned long arg);
---- linux-2.6.22/include/linux/major.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/major.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/include/linux/magic.h 2008-04-17 11:31:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/magic.h 2008-04-19 15:14:52.000000000 -0400
+@@ -3,7 +3,7 @@
+
+ #define ADFS_SUPER_MAGIC 0xadf5
+ #define AFFS_SUPER_MAGIC 0xadff
+-#define AFS_SUPER_MAGIC 0x5346414F
++#define AFS_SUPER_MAGIC 0x5346414F
+ #define AUTOFS_SUPER_MAGIC 0x0187
+ #define CODA_SUPER_MAGIC 0x73757245
+ #define EFS_SUPER_MAGIC 0x414A53
+@@ -26,6 +26,7 @@
+ #define NFS_SUPER_MAGIC 0x6969
+ #define OPENPROM_SUPER_MAGIC 0x9fa1
+ #define PROC_SUPER_MAGIC 0x9fa0
++#define DEVPTS_SUPER_MAGIC 0x1cd1
+ #define QNX4_SUPER_MAGIC 0x002f /* qnx4 fs detection */
+
+ #define REISERFS_SUPER_MAGIC 0x52654973 /* used by gcc */
+--- linux-2.6.25.4/include/linux/major.h 2008-04-17 10:37:24.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/major.h 2008-04-19 15:14:52.000000000 -0400
@@ -15,6 +15,7 @@
#define HD_MAJOR IDE0_MAJOR
#define PTY_SLAVE_MAJOR 3
@@ -10477,9 +9761,19 @@
#define TTYAUX_MAJOR 5
#define LP_MAJOR 6
#define VCS_MAJOR 7
---- linux-2.6.22/include/linux/mount.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/mount.h 2007-06-15 03:54:15 +0200
-@@ -28,12 +28,16 @@ struct mnt_namespace;
+--- linux-2.6.25.4/include/linux/mm_types.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/mm_types.h 2008-04-19 15:14:52.000000000 -0400
+@@ -201,6 +201,7 @@ struct mm_struct {
+
+ /* Architecture-specific MM context */
+ mm_context_t context;
++ struct vx_info *mm_vx_info;
+
+ /* Swap token stuff */
+ /*
+--- linux-2.6.25.4/include/linux/mount.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/mount.h 2008-04-19 15:45:19.000000000 -0400
+@@ -28,6 +28,9 @@ struct mnt_namespace;
#define MNT_NOATIME 0x08
#define MNT_NODIRATIME 0x10
#define MNT_RELATIME 0x20
@@ -10489,24 +9783,28 @@
#define MNT_SHRINKABLE 0x100
- #define MNT_SHARED 0x1000 /* if the vfsmount is a shared mount */
+@@ -35,6 +38,10 @@ struct mnt_namespace;
#define MNT_UNBINDABLE 0x2000 /* if the vfsmount is a unbindable mount */
#define MNT_PNODE_MASK 0x3000 /* propagation flag mask */
-+#define MNT_TAGID 0x8000
++#define MNT_TAGID 0x10000
++#define MNT_NOTAG 0x20000
++#define MNT_NOTAGCHECK 0x40000
++
struct vfsmount {
struct list_head mnt_hash;
-@@ -61,6 +65,7 @@ struct vfsmount {
- atomic_t mnt_count;
+ struct vfsmount *mnt_parent; /* fs we are mounted on */
+@@ -62,6 +69,7 @@ struct vfsmount {
int mnt_expiry_mark; /* true if marked for expiry */
int mnt_pinned;
+ int mnt_ghosts;
+ tag_t mnt_tag; /* tagging used for vfsmount */
};
static inline struct vfsmount *mntget(struct vfsmount *mnt)
---- linux-2.6.22/include/linux/net.h 2007-07-09 13:19:56 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/net.h 2007-06-15 02:37:04 +0200
-@@ -63,6 +63,7 @@ typedef enum {
+--- linux-2.6.25.4/include/linux/net.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/net.h 2008-04-19 15:14:52.000000000 -0400
+@@ -65,6 +65,7 @@ typedef enum {
#define SOCK_NOSPACE 2
#define SOCK_PASSCRED 3
#define SOCK_PASSSEC 4
@@ -10514,18 +9812,19 @@
#ifndef ARCH_HAS_SOCKET_TYPES
/**
---- linux-2.6.22/include/linux/nfs_mount.h 2007-07-09 13:20:00 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/nfs_mount.h 2007-06-15 03:54:39 +0200
-@@ -62,6 +62,7 @@ struct nfs_mount_data {
- #define NFS_MOUNT_STRICTLOCK 0x1000 /* reserved for NFSv4 */
+--- linux-2.6.25.4/include/linux/nfs_mount.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/nfs_mount.h 2008-04-19 15:14:52.000000000 -0400
+@@ -63,6 +63,7 @@ struct nfs_mount_data {
#define NFS_MOUNT_SECFLAVOUR 0x2000 /* 5 */
#define NFS_MOUNT_NORDIRPLUS 0x4000 /* 5 */
-+#define NFS_MOUNT_TAGGED 0x8000 /* context tagging */
- #define NFS_MOUNT_FLAGMASK 0xFFFF
+ #define NFS_MOUNT_UNSHARED 0x8000 /* 5 */
+-#define NFS_MOUNT_FLAGMASK 0xFFFF
++#define NFS_MOUNT_TAGGED 0x10000 /* context tagging */
++#define NFS_MOUNT_FLAGMASK 0x1FFFF
#endif
---- linux-2.6.22/include/linux/nsproxy.h 2007-07-09 13:20:00 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/nsproxy.h 2007-06-16 05:37:45 +0200
+--- linux-2.6.25.4/include/linux/nsproxy.h 2008-04-17 11:31:39.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/nsproxy.h 2008-04-19 15:14:52.000000000 -0400
@@ -3,6 +3,7 @@
#include <linux/spinlock.h>
@@ -10534,12 +9833,13 @@
struct mnt_namespace;
struct uts_namespace;
-@@ -32,26 +33,46 @@ struct nsproxy {
- extern struct nsproxy init_nsproxy;
+@@ -63,22 +64,33 @@ static inline struct nsproxy *task_nspro
+ }
- int copy_namespaces(int flags, struct task_struct *tsk);
+ int copy_namespaces(unsigned long flags, struct task_struct *tsk);
+struct nsproxy *copy_nsproxy(struct nsproxy *orig);
- void get_task_namespaces(struct task_struct *tsk);
+ void exit_task_namespaces(struct task_struct *tsk);
+ void switch_task_namespaces(struct task_struct *tsk, struct nsproxy *new);
void free_nsproxy(struct nsproxy *ns);
int unshare_nsproxy_namespaces(unsigned long, struct nsproxy **,
struct fs_struct *);
@@ -10550,42 +9850,31 @@
+static inline void __get_nsproxy(struct nsproxy *ns,
+ const char *_file, int _line)
{
+- if (atomic_dec_and_test(&ns->count)) {
+- free_nsproxy(ns);
+- }
+ vxlprintk(VXD_CBIT(space, 0), "get_nsproxy(%p[%u])",
+ ns, atomic_read(&ns->count), _file, _line);
+ atomic_inc(&ns->count);
-+}
-+
+ }
+
+-static inline void get_nsproxy(struct nsproxy *ns)
+#define put_nsproxy(n) __put_nsproxy(n, __FILE__, __LINE__)
+
+static inline void __put_nsproxy(struct nsproxy *ns,
+ const char *_file, int _line)
-+{
+ {
+- atomic_inc(&ns->count);
+ vxlprintk(VXD_CBIT(space, 0), "put_nsproxy(%p[%u])",
+ ns, atomic_read(&ns->count), _file, _line);
- if (atomic_dec_and_test(&ns->count)) {
- free_nsproxy(ns);
- }
++ if (atomic_dec_and_test(&ns->count)) {
++ free_nsproxy(ns);
++ }
}
--static inline void exit_task_namespaces(struct task_struct *p)
-+#define exit_task_namespaces(p) __exit_task_namespaces(p, __FILE__, __LINE__)
-+
-+static inline void __exit_task_namespaces(struct task_struct *p,
-+ const char *_file, int _line)
- {
- struct nsproxy *ns = p->nsproxy;
- if (ns) {
- task_lock(p);
- p->nsproxy = NULL;
- task_unlock(p);
-- put_nsproxy(ns);
-+ __put_nsproxy(ns, _file, _line);
- }
- }
-+
- #endif
---- linux-2.6.22/include/linux/pid.h 2007-07-09 13:20:00 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/pid.h 2007-06-15 02:37:04 +0200
+ #ifdef CONFIG_CGROUP_NS
+--- linux-2.6.25.4/include/linux/pid.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/pid.h 2008-04-29 17:56:00.000000000 -0400
@@ -8,7 +8,8 @@ enum pid_type
PIDTYPE_PID,
PIDTYPE_PGID,
@@ -10596,17 +9885,25 @@
};
/*
---- linux-2.6.22/include/linux/proc_fs.h 2007-07-09 13:20:00 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/proc_fs.h 2007-06-15 02:37:04 +0200
-@@ -54,6 +54,7 @@ struct proc_dir_entry {
+@@ -142,6 +143,7 @@ static inline pid_t pid_nr(struct pid *p
+ }
+
+ pid_t pid_nr_ns(struct pid *pid, struct pid_namespace *ns);
++pid_t pid_unmapped_nr_ns(struct pid *pid, struct pid_namespace *ns);
+ pid_t pid_vnr(struct pid *pid);
+
+ #define do_each_pid_task(pid, type, task) \
+--- linux-2.6.25.4/include/linux/proc_fs.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/proc_fs.h 2008-04-19 15:44:54.000000000 -0400
+@@ -59,6 +59,7 @@ struct proc_dir_entry {
nlink_t nlink;
uid_t uid;
gid_t gid;
+ int vx_flags;
loff_t size;
const struct inode_operations *proc_iops;
- const struct file_operations *proc_fops;
-@@ -246,13 +247,20 @@ static inline void kclist_add(struct kco
+ /*
+@@ -265,16 +266,23 @@ static inline void kclist_add(struct kco
extern void kclist_add(struct kcore_list *, void *, size_t);
#endif
@@ -10614,8 +9911,11 @@
+struct nx_info;
+
union proc_op {
- int (*proc_get_link)(struct inode *, struct dentry **, struct vfsmount **);
+ int (*proc_get_link)(struct inode *, struct path *);
int (*proc_read)(struct task_struct *task, char *page);
+ int (*proc_show)(struct seq_file *m,
+ struct pid_namespace *ns, struct pid *pid,
+ struct task_struct *task);
+ int (*proc_vs_read)(char *page);
+ int (*proc_vxi_read)(struct vx_info *vxi, char *page);
+ int (*proc_nxi_read)(struct nx_info *nxi, char *page);
@@ -10627,9 +9927,9 @@
int fd;
union proc_op op;
struct proc_dir_entry *pde;
---- linux-2.6.22/include/linux/reiserfs_fs.h 2007-05-02 19:25:34 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/reiserfs_fs.h 2007-06-15 02:37:04 +0200
-@@ -821,6 +821,10 @@ struct stat_data_v1 {
+--- linux-2.6.25.4/include/linux/reiserfs_fs.h 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/reiserfs_fs.h 2008-05-21 14:30:41.000000000 -0400
+@@ -837,6 +837,10 @@ struct stat_data_v1 {
#define REISERFS_COMPR_FL FS_COMPR_FL
#define REISERFS_NOTAIL_FL FS_NOTAIL_FL
@@ -10640,22 +9940,17 @@
/* persistent flags that file inherits from the parent directory */
#define REISERFS_INHERIT_MASK ( REISERFS_IMMUTABLE_FL | \
REISERFS_SYNC_FL | \
-@@ -830,6 +834,14 @@ struct stat_data_v1 {
+@@ -846,6 +850,9 @@ struct stat_data_v1 {
REISERFS_COMPR_FL | \
REISERFS_NOTAIL_FL )
-+#ifdef CONFIG_VSERVER_LEGACY
-+#define REISERFS_FL_USER_VISIBLE (REISERFS_IUNLINK_FL|0x80FF)
-+#define REISERFS_FL_USER_MODIFIABLE (REISERFS_IUNLINK_FL|0x80FF)
-+#else
+#define REISERFS_FL_USER_VISIBLE 0x80FF
+#define REISERFS_FL_USER_MODIFIABLE 0x80FF
-+#endif
+
/* Stat Data on disk (reiserfs version of UFS disk inode minus the
address blocks) */
struct stat_data {
-@@ -1901,6 +1913,7 @@ static inline void reiserfs_update_sd(st
+@@ -1911,6 +1918,7 @@ static inline void reiserfs_update_sd(st
void sd_attrs_to_i_attrs(__u16 sd_attrs, struct inode *inode);
void i_attrs_to_sd_attrs(struct inode *inode, __u16 * sd_attrs);
int reiserfs_setattr(struct dentry *dentry, struct iattr *attr);
@@ -10663,9 +9958,9 @@
/* namei.c */
void set_de_name_and_namelen(struct reiserfs_dir_entry *de);
---- linux-2.6.22/include/linux/reiserfs_fs_sb.h 2007-07-09 13:20:00 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/reiserfs_fs_sb.h 2007-06-15 02:37:04 +0200
-@@ -458,6 +458,7 @@ enum reiserfs_mount_options {
+--- linux-2.6.25.4/include/linux/reiserfs_fs_sb.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/reiserfs_fs_sb.h 2008-04-19 15:14:52.000000000 -0400
+@@ -456,6 +456,7 @@ enum reiserfs_mount_options {
REISERFS_POSIXACL,
REISERFS_BARRIER_NONE,
REISERFS_BARRIER_FLUSH,
@@ -10673,44 +9968,25 @@
/* Actions on error */
REISERFS_ERROR_PANIC,
---- linux-2.6.22/include/linux/sched.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/sched.h 2007-06-17 05:54:19 +0200
-@@ -26,6 +26,7 @@
- #define CLONE_STOPPED 0x02000000 /* Start in stopped state */
- #define CLONE_NEWUTS 0x04000000 /* New utsname group? */
- #define CLONE_NEWIPC 0x08000000 /* New ipcs */
-+#define CLONE_KTHREAD 0x10000000 /* clone a kernel thread */
+--- linux-2.6.25.4/include/linux/sched.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/sched.h 2008-04-21 11:09:01.000000000 -0400
+@@ -73,7 +73,6 @@ struct sched_param {
+ #include <linux/fs_struct.h>
+ #include <linux/compiler.h>
+ #include <linux/completion.h>
+-#include <linux/pid.h>
+ #include <linux/percpu.h>
+ #include <linux/topology.h>
+ #include <linux/proportions.h>
+@@ -89,6 +88,7 @@ struct sched_param {
+ #include <linux/task_io_accounting.h>
+ #include <linux/kobject.h>
+ #include <linux/latencytop.h>
++#include <linux/pid.h>
- /*
- * Scheduling policies
-@@ -94,7 +95,7 @@ struct bio;
- * List of flags we want to share for kernel threads,
- * if only because they are not used by them anyway.
- */
--#define CLONE_KERNEL (CLONE_FS | CLONE_FILES | CLONE_SIGHAND)
-+#define CLONE_KERNEL (CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_KTHREAD)
+ #include <asm/processor.h>
- /*
- * These are the constant used to fake the fixed-point load-average
-@@ -146,12 +147,13 @@ extern unsigned long weighted_cpuload(co
- #define TASK_UNINTERRUPTIBLE 2
- #define TASK_STOPPED 4
- #define TASK_TRACED 8
-+#define TASK_ONHOLD 16
- /* in tsk->exit_state */
--#define EXIT_ZOMBIE 16
--#define EXIT_DEAD 32
-+#define EXIT_ZOMBIE 32
-+#define EXIT_DEAD 64
- /* in tsk->state again */
--#define TASK_NONINTERACTIVE 64
--#define TASK_DEAD 128
-+#define TASK_NONINTERACTIVE 128
-+#define TASK_DEAD 256
-
- #define __set_task_state(tsk, state_value) \
- do { (tsk)->state = (state_value); } while (0)
-@@ -287,27 +289,30 @@ extern void arch_unmap_area_topdown(stru
+@@ -355,25 +355,27 @@ extern void arch_unmap_area_topdown(stru
* The mm counters are not protected by its page_table_lock,
* so must be incremented atomically.
*/
@@ -10719,25 +9995,22 @@
-#define add_mm_counter(mm, member, value) atomic_long_add(value, &(mm)->_##member)
-#define inc_mm_counter(mm, member) atomic_long_inc(&(mm)->_##member)
-#define dec_mm_counter(mm, member) atomic_long_dec(&(mm)->_##member)
- typedef atomic_long_t mm_counter_t;
+-
+#define __set_mm_counter(mm, member, value) \
+ atomic_long_set(&(mm)->_##member, value)
+#define get_mm_counter(mm, member) \
+ ((unsigned long)atomic_long_read(&(mm)->_##member))
-
#else /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
/*
* The mm counters are protected by its page_table_lock,
* so can be incremented directly.
*/
-#define set_mm_counter(mm, member, value) (mm)->_##member = (value)
--#define get_mm_counter(mm, member) ((mm)->_##member)
++#define __set_mm_counter(mm, member, value) (mm)->_##member = (value)
+ #define get_mm_counter(mm, member) ((mm)->_##member)
-#define add_mm_counter(mm, member, value) (mm)->_##member += (value)
-#define inc_mm_counter(mm, member) (mm)->_##member++
-#define dec_mm_counter(mm, member) (mm)->_##member--
- typedef unsigned long mm_counter_t;
-+#define __set_mm_counter(mm, member, value) (mm)->_##member = (value)
-+#define get_mm_counter(mm, member) ((mm)->_##member)
#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
@@ -10751,62 +10024,53 @@
#define get_mm_rss(mm) \
(get_mm_counter(mm, file_rss) + get_mm_counter(mm, anon_rss))
#define update_hiwater_rss(mm) do { \
-@@ -365,6 +370,7 @@ struct mm_struct {
-
- /* Architecture-specific MM context */
- mm_context_t context;
-+ struct vx_info *mm_vx_info;
-
- /* Swap token stuff */
- /*
-@@ -570,9 +576,10 @@ struct user_struct {
- /* Hash table maintenance information */
- struct list_head uidhash_list;
- uid_t uid;
-+ xid_t xid;
- };
-
--extern struct user_struct *find_user(uid_t);
-+extern struct user_struct *find_user(xid_t, uid_t);
+@@ -1162,6 +1164,14 @@ struct task_struct {
+ #endif
+ seccomp_t seccomp;
- extern struct user_struct root_user;
- #define INIT_USER (&root_user)
-@@ -969,6 +976,14 @@ struct task_struct {
-
- void *security;
- struct audit_context *audit_context;
-+
+/* vserver context data */
+ struct vx_info *vx_info;
+ struct nx_info *nx_info;
+
+ xid_t xid;
+ nid_t nid;
++ tag_t tag;
+
- seccomp_t seccomp;
-
/* Thread group tracking */
-@@ -1290,12 +1305,16 @@ extern struct task_struct init_task;
+ u32 parent_exec_id;
+ u32 self_exec_id;
+@@ -1350,6 +1360,11 @@ struct pid_namespace;
+ * see also pid_nr() etc in include/linux/pid.h
+ */
+
++#include <linux/vserver/base.h>
++#include <linux/vserver/context.h>
++#include <linux/vserver/debug.h>
++#include <linux/vserver/pid.h>
++
+ static inline pid_t task_pid_nr(struct task_struct *tsk)
+ {
+ return tsk->pid;
+@@ -1359,7 +1374,7 @@ pid_t task_pid_nr_ns(struct task_struct
+
+ static inline pid_t task_pid_vnr(struct task_struct *tsk)
+ {
+- return pid_vnr(task_pid(tsk));
++ return vx_map_pid(pid_vnr(task_pid(tsk)));
+ }
+
+
+@@ -1372,7 +1387,7 @@ pid_t task_tgid_nr_ns(struct task_struct
+
+ static inline pid_t task_tgid_vnr(struct task_struct *tsk)
+ {
+- return pid_vnr(task_tgid(tsk));
++ return vx_map_tgid(pid_vnr(task_tgid(tsk)));
+ }
- extern struct mm_struct init_mm;
--#define find_task_by_pid(nr) find_task_by_pid_type(PIDTYPE_PID, nr)
-+#define find_task_by_real_pid(nr) \
-+ find_task_by_pid_type(PIDTYPE_REALPID, nr)
-+#define find_task_by_pid(nr) \
-+ find_task_by_pid_type(PIDTYPE_PID, nr)
-+
- extern struct task_struct *find_task_by_pid_type(int type, int pid);
- extern void __set_special_pids(pid_t session, pid_t pgrp);
-
- /* per-UID process charging. */
--extern struct user_struct * alloc_uid(uid_t);
-+extern struct user_struct * alloc_uid(xid_t, uid_t);
- static inline struct user_struct *get_uid(struct user_struct *u)
- {
- atomic_inc(&u->__count);
---- linux-2.6.22/include/linux/shmem_fs.h 2006-11-30 21:19:39 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/shmem_fs.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/include/linux/shmem_fs.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/shmem_fs.h 2008-04-19 15:14:52.000000000 -0400
@@ -8,6 +8,9 @@
#define SHMEM_NR_DIRECT 16
@@ -10817,8 +10081,8 @@
struct shmem_inode_info {
spinlock_t lock;
unsigned long flags;
---- linux-2.6.22/include/linux/stat.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/stat.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/include/linux/stat.h 2008-04-17 10:33:07.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/stat.h 2008-04-19 15:14:52.000000000 -0400
@@ -66,6 +66,7 @@ struct kstat {
unsigned int nlink;
uid_t uid;
@@ -10827,9 +10091,9 @@
dev_t rdev;
loff_t size;
struct timespec atime;
---- linux-2.6.22/include/linux/sunrpc/auth.h 2006-11-30 21:19:40 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/sunrpc/auth.h 2007-06-15 02:37:04 +0200
-@@ -24,6 +24,7 @@
+--- linux-2.6.25.4/include/linux/sunrpc/auth.h 2008-04-17 10:37:24.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/sunrpc/auth.h 2008-04-19 15:14:52.000000000 -0400
+@@ -25,6 +25,7 @@
struct auth_cred {
uid_t uid;
gid_t gid;
@@ -10837,20 +10101,20 @@
struct group_info *group_info;
};
---- linux-2.6.22/include/linux/sunrpc/clnt.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/sunrpc/clnt.h 2007-06-15 02:37:04 +0200
-@@ -43,7 +43,8 @@ struct rpc_clnt {
+--- linux-2.6.25.4/include/linux/sunrpc/clnt.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/sunrpc/clnt.h 2008-04-19 15:14:52.000000000 -0400
+@@ -42,7 +42,8 @@ struct rpc_clnt {
+
+ unsigned int cl_softrtry : 1,/* soft timeouts */
cl_discrtry : 1,/* disconnect before retry */
- cl_autobind : 1,/* use getport() */
- cl_oneshot : 1,/* dispose after use */
-- cl_dead : 1;/* abandoned */
-+ cl_dead : 1,/* abandoned */
+- cl_autobind : 1;/* use getport() */
++ cl_autobind : 1,/* use getport() */
+ cl_tag : 1;/* context tagging */
struct rpc_rtt * cl_rtt; /* RTO estimator data */
-
---- linux-2.6.22/include/linux/syscalls.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/syscalls.h 2007-07-07 03:52:54 +0200
+ const struct rpc_timeout *cl_timeout; /* Timeout strategy */
+--- linux-2.6.25.4/include/linux/syscalls.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/syscalls.h 2008-04-19 15:14:52.000000000 -0400
@@ -294,6 +294,8 @@ asmlinkage long sys_symlink(const char _
asmlinkage long sys_unlink(const char __user *pathname);
asmlinkage long sys_rename(const char __user *oldname,
@@ -10860,19 +10124,27 @@
asmlinkage long sys_chmod(const char __user *filename, mode_t mode);
asmlinkage long sys_fchmod(unsigned int fd, mode_t mode);
---- linux-2.6.22/include/linux/sysctl.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/sysctl.h 2007-06-15 02:37:04 +0200
-@@ -106,6 +106,7 @@ enum
- KERN_CAP_BSET=14, /* int: capability bounding set */
+--- linux-2.6.25.4/include/linux/sysctl.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/sysctl.h 2008-04-19 15:14:52.000000000 -0400
+@@ -70,6 +70,7 @@ enum
+ CTL_ABI=9, /* Binary emulation */
+ CTL_CPU=10, /* CPU stuff (speed scaling, etc) */
+ CTL_ARLAN=254, /* arlan wireless driver */
++ CTL_VSERVER=4242, /* Linux-VServer debug */
+ CTL_S390DBF=5677, /* s390 debug */
+ CTL_SUNRPC=7249, /* sunrpc debug */
+ CTL_PM=9899, /* frv power management */
+@@ -104,6 +105,7 @@ enum
+
KERN_PANIC=15, /* int: panic timeout */
KERN_REALROOTDEV=16, /* real root device to mount after initrd */
+ KERN_VSHELPER=17, /* string: path to vshelper policy agent */
KERN_SPARC_REBOOT=21, /* reboot command on Sparc */
KERN_CTLALTDEL=22, /* int: allow ctl-alt-del to reboot */
---- linux-2.6.22/include/linux/sysfs.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/sysfs.h 2007-06-15 02:37:04 +0200
-@@ -15,6 +15,8 @@
+--- linux-2.6.25.4/include/linux/sysfs.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/sysfs.h 2008-04-19 15:14:52.000000000 -0400
+@@ -17,6 +17,8 @@
#include <linux/list.h>
#include <asm/atomic.h>
@@ -10880,10 +10152,10 @@
+
struct kobject;
struct module;
- struct nameidata;
---- linux-2.6.22/include/linux/time.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/time.h 2007-06-15 02:37:04 +0200
-@@ -176,6 +176,9 @@ static inline void timespec_add_ns(struc
+
+--- linux-2.6.25.4/include/linux/time.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/time.h 2008-04-19 15:14:52.000000000 -0400
+@@ -183,6 +183,9 @@ static inline void timespec_add_ns(struc
}
a->tv_nsec = ns;
}
@@ -10893,9 +10165,9 @@
#endif /* __KERNEL__ */
#define NFDBITS __NFDBITS
---- linux-2.6.22/include/linux/types.h 2007-02-06 03:01:52 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/types.h 2007-06-15 02:37:04 +0200
-@@ -39,6 +39,9 @@ typedef __kernel_uid32_t uid_t;
+--- linux-2.6.25.4/include/linux/types.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/types.h 2008-04-19 15:14:52.000000000 -0400
+@@ -36,6 +36,9 @@ typedef __kernel_uid32_t uid_t;
typedef __kernel_gid32_t gid_t;
typedef __kernel_uid16_t uid16_t;
typedef __kernel_gid16_t gid16_t;
@@ -10903,10 +10175,10 @@
+typedef unsigned int nid_t;
+typedef unsigned int tag_t;
- #ifdef CONFIG_UID16
- /* This is defined by include/asm-{arch}/posix_types.h */
---- linux-2.6.22/include/linux/vroot.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vroot.h 2007-06-15 02:37:04 +0200
+ typedef unsigned long uintptr_t;
+
+--- linux-2.6.25.4/include/linux/vroot.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vroot.h 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,51 @@
+
+/*
@@ -10959,29 +10231,33 @@
+#define VROOT_CLR_DEV 0x5601
+
+#endif /* _LINUX_VROOT_H */
---- linux-2.6.22/include/linux/vs_base.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_base.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,9 @@
+--- linux-2.6.25.4/include/linux/vs_base.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_base.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,10 @@
+#ifndef _VS_BASE_H
+#define _VS_BASE_H
+
+#include "vserver/base.h"
++#include "vserver/check.h"
+#include "vserver/debug.h"
+
+#else
+#warning duplicate inclusion
+#endif
---- linux-2.6.22/include/linux/vs_context.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_context.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,224 @@
+--- linux-2.6.25.4/include/linux/vs_context.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_context.h 2008-04-29 18:42:09.000000000 -0400
+@@ -0,0 +1,227 @@
+#ifndef _VS_CONTEXT_H
+#define _VS_CONTEXT_H
+
+#include "vserver/base.h"
++#include "vserver/check.h"
+#include "vserver/context.h"
+#include "vserver/history.h"
+#include "vserver/debug.h"
+
++#include <linux/sched.h>
++
+
+#define get_vx_info(i) __get_vx_info(i, __FILE__, __LINE__, __HERE__)
+
@@ -11198,8 +10474,8 @@
+#else
+#warning duplicate inclusion
+#endif
---- linux-2.6.22/include/linux/vs_cowbl.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_cowbl.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/include/linux/vs_cowbl.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_cowbl.h 2008-04-21 13:49:51.000000000 -0400
@@ -0,0 +1,44 @@
+#ifndef _VS_COWBL_H
+#define _VS_COWBL_H
@@ -11212,24 +10488,24 @@
+
+static inline int cow_check_and_break(struct nameidata *nd)
+{
-+ struct inode *inode = nd->dentry->d_inode;
++ struct inode *inode = nd->path.dentry->d_inode;
+ int error = 0;
-+ if (IS_RDONLY(inode) || MNT_IS_RDONLY(nd->mnt))
++ if (IS_RDONLY(inode) || MNT_IS_RDONLY(nd->path.mnt))
+ return -EROFS;
+ if (IS_COW(inode)) {
+ if (IS_COW_LINK(inode)) {
-+ struct dentry *new_dentry, *old_dentry = nd->dentry;
++ struct dentry *new_dentry, *old_dentry = nd->path.dentry;
+ char *path, *buf;
+
+ buf = kmalloc(PATH_MAX, GFP_KERNEL);
+ if (!buf) {
+ return -ENOMEM;
+ }
-+ path = d_path(nd->dentry, nd->mnt, buf, PATH_MAX);
++ path = d_path(&nd->path, buf, PATH_MAX);
+ new_dentry = cow_break_link(path);
+ kfree(buf);
+ if (!IS_ERR(new_dentry)) {
-+ nd->dentry = new_dentry;
++ nd->path.dentry = new_dentry;
+ dput(old_dentry);
+ } else
+ error = PTR_ERR(new_dentry);
@@ -11245,15 +10521,16 @@
+#else
+#warning duplicate inclusion
+#endif
---- linux-2.6.22/include/linux/vs_cvirt.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_cvirt.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,49 @@
+--- linux-2.6.25.4/include/linux/vs_cvirt.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_cvirt.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,50 @@
+#ifndef _VS_CVIRT_H
+#define _VS_CVIRT_H
+
+#include "vserver/cvirt.h"
+#include "vserver/context.h"
+#include "vserver/base.h"
++#include "vserver/check.h"
+#include "vserver/debug.h"
+
+
@@ -11297,12 +10574,62 @@
+#else
+#warning duplicate inclusion
+#endif
---- linux-2.6.22/include/linux/vs_dlimit.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_dlimit.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,209 @@
+--- linux-2.6.25.4/include/linux/vs_device.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_device.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,45 @@
++#ifndef _VS_DEVICE_H
++#define _VS_DEVICE_H
++
++#include "vserver/base.h"
++#include "vserver/device.h"
++#include "vserver/debug.h"
++
++
++#ifdef CONFIG_VSERVER_DEVICE
++
++int vs_map_device(struct vx_info *, dev_t, dev_t *, umode_t);
++
++#define vs_device_perm(v, d, m, p) \
++ ((vs_map_device(current_vx_info(), d, NULL, m) & (p)) == (p))
++
++#else
++
++static inline
++int vs_map_device(struct vx_info *vxi,
++ dev_t device, dev_t *target, umode_t mode)
++{
++ if (target)
++ *target = device;
++ return ~0;
++}
++
++#define vs_device_perm(v, d, m, p) ((p) == (p))
++
++#endif
++
++
++#define vs_map_chrdev(d, t, p) \
++ ((vs_map_device(current_vx_info(), d, t, S_IFCHR) & (p)) == (p))
++#define vs_map_blkdev(d, t, p) \
++ ((vs_map_device(current_vx_info(), d, t, S_IFBLK) & (p)) == (p))
++
++#define vs_chrdev_perm(d, p) \
++ vs_device_perm(current_vx_info(), d, S_IFCHR, p)
++#define vs_blkdev_perm(d, p) \
++ vs_device_perm(current_vx_info(), d, S_IFBLK, p)
++
++
++#else
++#warning duplicate inclusion
++#endif
+--- linux-2.6.25.4/include/linux/vs_dlimit.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_dlimit.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,211 @@
+#ifndef _VS_DLIMIT_H
+#define _VS_DLIMIT_H
+
++#include <linux/fs.h>
++
+#include "vserver/dlimit.h"
+#include "vserver/base.h"
+#include "vserver/debug.h"
@@ -11509,6032 +10836,4917 @@
+#else
+#warning duplicate inclusion
+#endif
---- linux-2.6.22/include/linux/vs_limit.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_limit.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,140 @@
-+#ifndef _VS_LIMIT_H
-+#define _VS_LIMIT_H
+--- linux-2.6.25.4/include/linux/vserver/base.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/base.h 2008-04-23 20:52:31.000000000 -0400
+@@ -0,0 +1,157 @@
++#ifndef _VX_BASE_H
++#define _VX_BASE_H
+
-+#include "vserver/limit.h"
-+#include "vserver/base.h"
-+#include "vserver/context.h"
-+#include "vserver/debug.h"
-+#include "vserver/context.h"
-+#include "vserver/limit_int.h"
+
++/* context state changes */
+
-+#define vx_acc_cres(v, d, p, r) \
-+ __vx_acc_cres(v, r, d, p, __FILE__, __LINE__)
++enum {
++ VSC_STARTUP = 1,
++ VSC_SHUTDOWN,
+
-+#define vx_acc_cres_cond(x, d, p, r) \
-+ __vx_acc_cres(((x) == vx_current_xid()) ? current->vx_info : 0, \
-+ r, d, p, __FILE__, __LINE__)
++ VSC_NETUP,
++ VSC_NETDOWN,
++};
+
+
-+#define vx_add_cres(v, a, p, r) \
-+ __vx_add_cres(v, r, a, p, __FILE__, __LINE__)
-+#define vx_sub_cres(v, a, p, r) vx_add_cres(v, -(a), p, r)
+
-+#define vx_add_cres_cond(x, a, p, r) \
-+ __vx_add_cres(((x) == vx_current_xid()) ? current->vx_info : 0, \
-+ r, a, p, __FILE__, __LINE__)
-+#define vx_sub_cres_cond(x, a, p, r) vx_add_cres_cond(x, -(a), p, r)
++#define vx_task_xid(t) ((t)->xid)
+
++#define vx_current_xid() vx_task_xid(current)
+
-+/* process and file limits */
++#define current_vx_info() (current->vx_info)
+
-+#define vx_nproc_inc(p) \
-+ vx_acc_cres((p)->vx_info, 1, p, RLIMIT_NPROC)
+
-+#define vx_nproc_dec(p) \
-+ vx_acc_cres((p)->vx_info,-1, p, RLIMIT_NPROC)
++#define nx_task_nid(t) ((t)->nid)
+
-+#define vx_files_inc(f) \
-+ vx_acc_cres_cond((f)->f_xid, 1, f, RLIMIT_NOFILE)
++#define nx_current_nid() nx_task_nid(current)
+
-+#define vx_files_dec(f) \
-+ vx_acc_cres_cond((f)->f_xid,-1, f, RLIMIT_NOFILE)
++#define current_nx_info() (current->nx_info)
+
-+#define vx_locks_inc(l) \
-+ vx_acc_cres_cond((l)->fl_xid, 1, l, RLIMIT_LOCKS)
+
-+#define vx_locks_dec(l) \
-+ vx_acc_cres_cond((l)->fl_xid,-1, l, RLIMIT_LOCKS)
++/* generic flag merging */
+
-+#define vx_openfd_inc(f) \
-+ vx_acc_cres(current->vx_info, 1, (void *)(long)(f), VLIMIT_OPENFD)
++#define vs_check_flags(v, m, f) (((v) & (m)) ^ (f))
+
-+#define vx_openfd_dec(f) \
-+ vx_acc_cres(current->vx_info,-1, (void *)(long)(f), VLIMIT_OPENFD)
++#define vs_mask_flags(v, f, m) (((v) & ~(m)) | ((f) & (m)))
+
++#define vs_mask_mask(v, f, m) (((v) & ~(m)) | ((v) & (f) & (m)))
+
-+#define vx_cres_avail(v, n, r) \
-+ __vx_cres_avail(v, r, n, __FILE__, __LINE__)
++#define vs_check_bit(v, n) ((v) & (1LL << (n)))
+
+
-+#define vx_nproc_avail(n) \
-+ vx_cres_avail(current->vx_info, n, RLIMIT_NPROC)
++/* context flags */
+
-+#define vx_files_avail(n) \
-+ vx_cres_avail(current->vx_info, n, RLIMIT_NOFILE)
++#define __vx_flags(v) ((v) ? (v)->vx_flags : 0)
+
-+#define vx_locks_avail(n) \
-+ vx_cres_avail(current->vx_info, n, RLIMIT_LOCKS)
++#define vx_current_flags() __vx_flags(current->vx_info)
+
-+#define vx_openfd_avail(n) \
-+ vx_cres_avail(current->vx_info, n, VLIMIT_OPENFD)
++#define vx_info_flags(v, m, f) \
++ vs_check_flags(__vx_flags(v), m, f)
+
++#define task_vx_flags(t, m, f) \
++ ((t) && vx_info_flags((t)->vx_info, m, f))
+
-+/* dentry limits */
++#define vx_flags(m, f) vx_info_flags(current->vx_info, m, f)
+
-+#define vx_dentry_inc(d) do { \
-+ if (atomic_read(&d->d_count) == 1) \
-+ vx_acc_cres(current->vx_info, 1, d, VLIMIT_DENTRY); \
-+ } while (0)
+
-+#define vx_dentry_dec(d) do { \
-+ if (atomic_read(&d->d_count) == 0) \
-+ vx_acc_cres(current->vx_info,-1, d, VLIMIT_DENTRY); \
-+ } while (0)
++/* context caps */
+
-+#define vx_dentry_avail(n) \
-+ vx_cres_avail(current->vx_info, n, VLIMIT_DENTRY)
++#define __vx_ccaps(v) ((v) ? (v)->vx_ccaps : 0)
+
++#define vx_current_ccaps() __vx_ccaps(current->vx_info)
+
-+/* socket limits */
++#define vx_info_ccaps(v, c) (__vx_ccaps(v) & (c))
+
-+#define vx_sock_inc(s) \
-+ vx_acc_cres((s)->sk_vx_info, 1, s, VLIMIT_NSOCK)
++#define vx_ccaps(c) vx_info_ccaps(current->vx_info, (c))
+
-+#define vx_sock_dec(s) \
-+ vx_acc_cres((s)->sk_vx_info,-1, s, VLIMIT_NSOCK)
+
-+#define vx_sock_avail(n) \
-+ vx_cres_avail(current->vx_info, n, VLIMIT_NSOCK)
+
++/* network flags */
+
-+/* ipc resource limits */
++#define __nx_flags(n) ((n) ? (n)->nx_flags : 0)
+
-+#define vx_ipcmsg_add(v, u, a) \
-+ vx_add_cres(v, a, u, RLIMIT_MSGQUEUE)
++#define nx_current_flags() __nx_flags(current->nx_info)
+
-+#define vx_ipcmsg_sub(v, u, a) \
-+ vx_sub_cres(v, a, u, RLIMIT_MSGQUEUE)
++#define nx_info_flags(n, m, f) \
++ vs_check_flags(__nx_flags(n), m, f)
+
-+#define vx_ipcmsg_avail(v, a) \
-+ vx_cres_avail(v, a, RLIMIT_MSGQUEUE)
++#define task_nx_flags(t, m, f) \
++ ((t) && nx_info_flags((t)->nx_info, m, f))
+
++#define nx_flags(m, f) nx_info_flags(current->nx_info, m, f)
+
-+#define vx_ipcshm_add(v, k, a) \
-+ vx_add_cres(v, a, (void *)(long)(k), VLIMIT_SHMEM)
+
-+#define vx_ipcshm_sub(v, k, a) \
-+ vx_sub_cres(v, a, (void *)(long)(k), VLIMIT_SHMEM)
++/* network caps */
+
-+#define vx_ipcshm_avail(v, a) \
-+ vx_cres_avail(v, a, VLIMIT_SHMEM)
++#define __nx_ncaps(n) ((n) ? (n)->nx_ncaps : 0)
+
++#define nx_current_ncaps() __nx_ncaps(current->nx_info)
+
-+#define vx_semary_inc(a) \
-+ vx_acc_cres(current->vx_info, 1, a, VLIMIT_SEMARY)
++#define nx_info_ncaps(n, c) (__nx_ncaps(n) & (c))
+
-+#define vx_semary_dec(a) \
-+ vx_acc_cres(current->vx_info, -1, a, VLIMIT_SEMARY)
++#define nx_ncaps(c) nx_info_ncaps(current->nx_info, c)
+
+
-+#define vx_nsems_add(a,n) \
-+ vx_add_cres(current->vx_info, n, a, VLIMIT_NSEMS)
++/* context mask capabilities */
+
-+#define vx_nsems_sub(a,n) \
-+ vx_sub_cres(current->vx_info, n, a, VLIMIT_NSEMS)
++#define __vx_mcaps(v) ((v) ? (v)->vx_ccaps >> 32UL : ~0 )
+
++#define vx_info_mcaps(v, c) (__vx_mcaps(v) & (c))
+
-+#else
-+#warning duplicate inclusion
-+#endif
---- linux-2.6.22/include/linux/vs_memory.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_memory.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,159 @@
-+#ifndef _VS_MEMORY_H
-+#define _VS_MEMORY_H
++#define vx_mcaps(c) vx_info_mcaps(current->vx_info, c)
+
-+#include "vserver/limit.h"
-+#include "vserver/base.h"
-+#include "vserver/context.h"
-+#include "vserver/debug.h"
-+#include "vserver/context.h"
-+#include "vserver/limit_int.h"
+
++/* context bcap mask */
+
-+#define __acc_add_long(a, v) (*(v) += (a))
-+#define __acc_inc_long(v) (++*(v))
-+#define __acc_dec_long(v) (--*(v))
++#define __vx_bcaps(v) ((v)->vx_bcaps)
+
-+#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
-+#define __acc_add_atomic(a, v) atomic_long_add(a, v)
-+#define __acc_inc_atomic(v) atomic_long_inc(v)
-+#define __acc_dec_atomic(v) atomic_long_dec(v)
-+#else /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
-+#define __acc_add_atomic(a, v) __acc_add_long(a, v)
-+#define __acc_inc_atomic(v) __acc_inc_long(v)
-+#define __acc_dec_atomic(v) __acc_dec_long(v)
-+#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
++#define vx_current_bcaps() __vx_bcaps(current->vx_info)
+
+
-+#define vx_acc_page(m, d, v, r) do { \
-+ if ((d) > 0) \
-+ __acc_inc_long(&(m)->v); \
-+ else \
-+ __acc_dec_long(&(m)->v); \
-+ __vx_acc_cres(m->mm_vx_info, r, d, m, __FILE__, __LINE__); \
-+} while (0)
++/* mask given bcaps */
+
-+#define vx_acc_page_atomic(m, d, v, r) do { \
-+ if ((d) > 0) \
-+ __acc_inc_atomic(&(m)->v); \
-+ else \
-+ __acc_dec_atomic(&(m)->v); \
-+ __vx_acc_cres(m->mm_vx_info, r, d, m, __FILE__, __LINE__); \
-+} while (0)
++#define vx_info_mbcaps(v, c) ((v) ? cap_intersect(__vx_bcaps(v), c) : c)
+
++#define vx_mbcaps(c) vx_info_mbcaps(current->vx_info, c)
+
-+#define vx_acc_pages(m, p, v, r) do { \
-+ unsigned long __p = (p); \
-+ __acc_add_long(__p, &(m)->v); \
-+ __vx_add_cres(m->mm_vx_info, r, __p, m, __FILE__, __LINE__); \
-+} while (0)
+
-+#define vx_acc_pages_atomic(m, p, v, r) do { \
-+ unsigned long __p = (p); \
-+ __acc_add_atomic(__p, &(m)->v); \
-+ __vx_add_cres(m->mm_vx_info, r, __p, m, __FILE__, __LINE__); \
-+} while (0)
++/* masked cap_bset */
+
++#define vx_info_cap_bset(v) vx_info_mbcaps(v, current->cap_bset)
+
++#define vx_current_cap_bset() vx_info_cap_bset(current->vx_info)
+
-+#define vx_acc_vmpage(m, d) \
-+ vx_acc_page(m, d, total_vm, RLIMIT_AS)
-+#define vx_acc_vmlpage(m, d) \
-+ vx_acc_page(m, d, locked_vm, RLIMIT_MEMLOCK)
-+#define vx_acc_file_rsspage(m, d) \
-+ vx_acc_page_atomic(m, d, _file_rss, VLIMIT_MAPPED)
-+#define vx_acc_anon_rsspage(m, d) \
-+ vx_acc_page_atomic(m, d, _anon_rss, VLIMIT_ANON)
++#if 0
++#define vx_info_mbcap(v, b) \
++ (!vx_info_flags(v, VXF_STATE_SETUP, 0) ? \
++ vx_info_bcaps(v, b) : (b))
+
-+#define vx_acc_vmpages(m, p) \
-+ vx_acc_pages(m, p, total_vm, RLIMIT_AS)
-+#define vx_acc_vmlpages(m, p) \
-+ vx_acc_pages(m, p, locked_vm, RLIMIT_MEMLOCK)
-+#define vx_acc_file_rsspages(m, p) \
-+ vx_acc_pages_atomic(m, p, _file_rss, VLIMIT_MAPPED)
-+#define vx_acc_anon_rsspages(m, p) \
-+ vx_acc_pages_atomic(m, p, _anon_rss, VLIMIT_ANON)
++#define task_vx_mbcap(t, b) \
++ vx_info_mbcap((t)->vx_info, (t)->b)
+
-+#define vx_pages_add(s, r, p) __vx_add_cres(s, r, p, 0, __FILE__, __LINE__)
-+#define vx_pages_sub(s, r, p) vx_pages_add(s, r, -(p))
++#define vx_mbcap(b) task_vx_mbcap(current, b)
++#endif
+
-+#define vx_vmpages_inc(m) vx_acc_vmpage(m, 1)
-+#define vx_vmpages_dec(m) vx_acc_vmpage(m, -1)
-+#define vx_vmpages_add(m, p) vx_acc_vmpages(m, p)
-+#define vx_vmpages_sub(m, p) vx_acc_vmpages(m, -(p))
++#define vx_cap_raised(v, c, f) cap_raised(vx_info_mbcaps(v, c), f)
+
-+#define vx_vmlocked_inc(m) vx_acc_vmlpage(m, 1)
-+#define vx_vmlocked_dec(m) vx_acc_vmlpage(m, -1)
-+#define vx_vmlocked_add(m, p) vx_acc_vmlpages(m, p)
-+#define vx_vmlocked_sub(m, p) vx_acc_vmlpages(m, -(p))
++#define vx_capable(b, c) (capable(b) || \
++ (cap_raised(current->cap_effective, b) && vx_ccaps(c)))
+
-+#define vx_file_rsspages_inc(m) vx_acc_file_rsspage(m, 1)
-+#define vx_file_rsspages_dec(m) vx_acc_file_rsspage(m, -1)
-+#define vx_file_rsspages_add(m, p) vx_acc_file_rsspages(m, p)
-+#define vx_file_rsspages_sub(m, p) vx_acc_file_rsspages(m, -(p))
++#define nx_capable(b, c) (capable(b) || \
++ (cap_raised(current->cap_effective, b) && nx_ncaps(c)))
+
-+#define vx_anon_rsspages_inc(m) vx_acc_anon_rsspage(m, 1)
-+#define vx_anon_rsspages_dec(m) vx_acc_anon_rsspage(m, -1)
-+#define vx_anon_rsspages_add(m, p) vx_acc_anon_rsspages(m, p)
-+#define vx_anon_rsspages_sub(m, p) vx_acc_anon_rsspages(m, -(p))
++#define vx_current_initpid(n) \
++ (current->vx_info && \
++ (current->vx_info->vx_initpid == (n)))
+
+
-+#define vx_pages_avail(m, p, r) \
-+ __vx_cres_avail((m)->mm_vx_info, r, p, __FILE__, __LINE__)
++#define __vx_state(v) ((v) ? ((v)->vx_state) : 0)
+
-+#define vx_vmpages_avail(m, p) vx_pages_avail(m, p, RLIMIT_AS)
-+#define vx_vmlocked_avail(m, p) vx_pages_avail(m, p, RLIMIT_MEMLOCK)
-+#define vx_anon_avail(m, p) vx_pages_avail(m, p, VLIMIT_ANON)
-+#define vx_mapped_avail(m, p) vx_pages_avail(m, p, VLIMIT_MAPPED)
++#define vx_info_state(v, m) (__vx_state(v) & (m))
+
-+#define vx_rss_avail(m, p) \
-+ __vx_cres_array_avail((m)->mm_vx_info, VLA_RSS, p, __FILE__, __LINE__)
+
++#define __nx_state(n) ((n) ? ((n)->nx_state) : 0)
+
-+enum {
-+ VXPT_UNKNOWN = 0,
-+ VXPT_ANON,
-+ VXPT_NONE,
-+ VXPT_FILE,
-+ VXPT_SWAP,
-+ VXPT_WRITE
-+};
++#define nx_info_state(n, m) (__nx_state(n) & (m))
+
-+#if 0
-+#define vx_page_fault(mm, vma, type, ret)
-+#else
++#endif
+--- linux-2.6.25.4/include/linux/vserver/cacct_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/cacct_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,23 @@
++#ifndef _VX_CACCT_CMD_H
++#define _VX_CACCT_CMD_H
+
-+static inline
-+void __vx_page_fault(struct mm_struct *mm,
-+ struct vm_area_struct *vma, int type, int ret)
-+{
-+ struct vx_info *vxi = mm->mm_vx_info;
-+ int what;
-+/*
-+ static char *page_type[6] =
-+ { "UNKNOWN", "ANON", "NONE", "FILE", "SWAP", "WRITE" };
-+ static char *page_what[4] =
-+ { "FAULT_OOM", "FAULT_SIGBUS", "FAULT_MINOR", "FAULT_MAJOR" };
-+*/
+
-+ if (!vxi)
-+ return;
++/* virtual host info name commands */
+
-+ what = (ret & 0x3);
++#define VCMD_sock_stat VC_CMD(VSTAT, 5, 0)
+
-+/* printk("[%d] page[%d][%d] %2x %s %s\n", vxi->vx_id,
-+ type, what, ret, page_type[type], page_what[what]);
-+*/
-+ if (ret & VM_FAULT_WRITE)
-+ what |= 0x4;
-+ atomic_inc(&vxi->cacct.page[type][what]);
-+}
++struct vcmd_sock_stat_v0 {
++ uint32_t field;
++ uint32_t count[3];
++ uint64_t total[3];
++};
+
-+#define vx_page_fault(mm, vma, type, ret) __vx_page_fault(mm, vma, type, ret)
-+#endif
+
++#ifdef __KERNEL__
+
-+extern unsigned long vx_badness(struct task_struct *task, struct mm_struct *mm);
++#include <linux/compiler.h>
+
-+#else
-+#warning duplicate inclusion
-+#endif
---- linux-2.6.22/include/linux/vs_network.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_network.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,185 @@
-+#ifndef _NX_VS_NETWORK_H
-+#define _NX_VS_NETWORK_H
++extern int vc_sock_stat(struct vx_info *, void __user *);
+
-+#include "vserver/context.h"
-+#include "vserver/network.h"
-+#include "vserver/base.h"
-+#include "vserver/debug.h"
++#endif /* __KERNEL__ */
++#endif /* _VX_CACCT_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/cacct_def.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/cacct_def.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,43 @@
++#ifndef _VX_CACCT_DEF_H
++#define _VX_CACCT_DEF_H
+
++#include <asm/atomic.h>
++#include <linux/vserver/cacct.h>
+
-+#define get_nx_info(i) __get_nx_info(i, __FILE__, __LINE__)
-+
-+static inline struct nx_info *__get_nx_info(struct nx_info *nxi,
-+ const char *_file, int _line)
-+{
-+ if (!nxi)
-+ return NULL;
-+
-+ vxlprintk(VXD_CBIT(nid, 2), "get_nx_info(%p[#%d.%d])",
-+ nxi, nxi ? nxi->nx_id : 0,
-+ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
-+ _file, _line);
+
-+ atomic_inc(&nxi->nx_usecnt);
-+ return nxi;
-+}
++struct _vx_sock_acc {
++ atomic_long_t count;
++ atomic_long_t total;
++};
+
++/* context sub struct */
+
-+extern void free_nx_info(struct nx_info *);
++struct _vx_cacct {
++ struct _vx_sock_acc sock[VXA_SOCK_SIZE][3];
++ atomic_t slab[8];
++ atomic_t page[6][8];
++};
+
-+#define put_nx_info(i) __put_nx_info(i, __FILE__, __LINE__)
++#ifdef CONFIG_VSERVER_DEBUG
+
-+static inline void __put_nx_info(struct nx_info *nxi, const char *_file, int _line)
++static inline void __dump_vx_cacct(struct _vx_cacct *cacct)
+{
-+ if (!nxi)
-+ return;
-+
-+ vxlprintk(VXD_CBIT(nid, 2), "put_nx_info(%p[#%d.%d])",
-+ nxi, nxi ? nxi->nx_id : 0,
-+ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
-+ _file, _line);
-+
-+ if (atomic_dec_and_test(&nxi->nx_usecnt))
-+ free_nx_info(nxi);
-+}
-+
-+
-+#define init_nx_info(p, i) __init_nx_info(p, i, __FILE__, __LINE__)
++ int i, j;
+
-+static inline void __init_nx_info(struct nx_info **nxp, struct nx_info *nxi,
-+ const char *_file, int _line)
-+{
-+ if (nxi) {
-+ vxlprintk(VXD_CBIT(nid, 3),
-+ "init_nx_info(%p[#%d.%d])",
-+ nxi, nxi ? nxi->nx_id : 0,
-+ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
-+ _file, _line);
++ printk("\t_vx_cacct:");
++ for (i = 0; i < 6; i++) {
++ struct _vx_sock_acc *ptr = cacct->sock[i];
+
-+ atomic_inc(&nxi->nx_usecnt);
++ printk("\t [%d] =", i);
++ for (j = 0; j < 3; j++) {
++ printk(" [%d] = %8lu, %8lu", j,
++ atomic_long_read(&ptr[j].count),
++ atomic_long_read(&ptr[j].total));
++ }
++ printk("\n");
+ }
-+ *nxp = nxi;
+}
+
++#endif
+
-+#define set_nx_info(p, i) __set_nx_info(p, i, __FILE__, __LINE__)
++#endif /* _VX_CACCT_DEF_H */
+--- linux-2.6.25.4/include/linux/vserver/cacct.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/cacct.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,15 @@
++#ifndef _VX_CACCT_H
++#define _VX_CACCT_H
+
-+static inline void __set_nx_info(struct nx_info **nxp, struct nx_info *nxi,
-+ const char *_file, int _line)
-+{
-+ struct nx_info *nxo;
+
-+ if (!nxi)
-+ return;
++enum sock_acc_field {
++ VXA_SOCK_UNSPEC = 0,
++ VXA_SOCK_UNIX,
++ VXA_SOCK_INET,
++ VXA_SOCK_INET6,
++ VXA_SOCK_PACKET,
++ VXA_SOCK_OTHER,
++ VXA_SOCK_SIZE /* array size */
++};
+
-+ vxlprintk(VXD_CBIT(nid, 3), "set_nx_info(%p[#%d.%d])",
-+ nxi, nxi ? nxi->nx_id : 0,
-+ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
-+ _file, _line);
++#endif /* _VX_CACCT_H */
+--- linux-2.6.25.4/include/linux/vserver/cacct_int.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/cacct_int.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,21 @@
++#ifndef _VX_CACCT_INT_H
++#define _VX_CACCT_INT_H
+
-+ atomic_inc(&nxi->nx_usecnt);
-+ nxo = xchg(nxp, nxi);
-+ BUG_ON(nxo);
-+}
+
-+#define clr_nx_info(p) __clr_nx_info(p, __FILE__, __LINE__)
++#ifdef __KERNEL__
+
-+static inline void __clr_nx_info(struct nx_info **nxp,
-+ const char *_file, int _line)
++static inline
++unsigned long vx_sock_count(struct _vx_cacct *cacct, int type, int pos)
+{
-+ struct nx_info *nxo;
-+
-+ nxo = xchg(nxp, NULL);
-+ if (!nxo)
-+ return;
-+
-+ vxlprintk(VXD_CBIT(nid, 3), "clr_nx_info(%p[#%d.%d])",
-+ nxo, nxo ? nxo->nx_id : 0,
-+ nxo ? atomic_read(&nxo->nx_usecnt) : 0,
-+ _file, _line);
-+
-+ if (atomic_dec_and_test(&nxo->nx_usecnt))
-+ free_nx_info(nxo);
++ return atomic_long_read(&cacct->sock[type][pos].count);
+}
+
+
-+#define claim_nx_info(v, p) __claim_nx_info(v, p, __FILE__, __LINE__)
-+
-+static inline void __claim_nx_info(struct nx_info *nxi,
-+ struct task_struct *task, const char *_file, int _line)
++static inline
++unsigned long vx_sock_total(struct _vx_cacct *cacct, int type, int pos)
+{
-+ vxlprintk(VXD_CBIT(nid, 3), "claim_nx_info(%p[#%d.%d.%d]) %p",
-+ nxi, nxi ? nxi->nx_id : 0,
-+ nxi?atomic_read(&nxi->nx_usecnt):0,
-+ nxi?atomic_read(&nxi->nx_tasks):0,
-+ task, _file, _line);
-+
-+ atomic_inc(&nxi->nx_tasks);
++ return atomic_long_read(&cacct->sock[type][pos].total);
+}
+
++#endif /* __KERNEL__ */
++#endif /* _VX_CACCT_INT_H */
+--- linux-2.6.25.4/include/linux/vserver/check.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/check.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,89 @@
++#ifndef _VS_CHECK_H
++#define _VS_CHECK_H
+
-+extern void unhash_nx_info(struct nx_info *);
-+
-+#define release_nx_info(v, p) __release_nx_info(v, p, __FILE__, __LINE__)
+
-+static inline void __release_nx_info(struct nx_info *nxi,
-+ struct task_struct *task, const char *_file, int _line)
-+{
-+ vxlprintk(VXD_CBIT(nid, 3), "release_nx_info(%p[#%d.%d.%d]) %p",
-+ nxi, nxi ? nxi->nx_id : 0,
-+ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
-+ nxi ? atomic_read(&nxi->nx_tasks) : 0,
-+ task, _file, _line);
++#define MAX_S_CONTEXT 65535 /* Arbitrary limit */
+
-+ might_sleep();
++#ifdef CONFIG_VSERVER_DYNAMIC_IDS
++#define MIN_D_CONTEXT 49152 /* dynamic contexts start here */
++#else
++#define MIN_D_CONTEXT 65536
++#endif
+
-+ if (atomic_dec_and_test(&nxi->nx_tasks))
-+ unhash_nx_info(nxi);
-+}
++/* check conditions */
+
++#define VS_ADMIN 0x0001
++#define VS_WATCH 0x0002
++#define VS_HIDE 0x0004
++#define VS_HOSTID 0x0008
+
-+#define task_get_nx_info(i) __task_get_nx_info(i, __FILE__, __LINE__)
++#define VS_IDENT 0x0010
++#define VS_EQUIV 0x0020
++#define VS_PARENT 0x0040
++#define VS_CHILD 0x0080
+
-+static __inline__ struct nx_info *__task_get_nx_info(struct task_struct *p,
-+ const char *_file, int _line)
-+{
-+ struct nx_info *nxi;
++#define VS_ARG_MASK 0x00F0
+
-+ task_lock(p);
-+ vxlprintk(VXD_CBIT(nid, 5), "task_get_nx_info(%p)",
-+ p, _file, _line);
-+ nxi = __get_nx_info(p->nx_info, _file, _line);
-+ task_unlock(p);
-+ return nxi;
-+}
++#define VS_DYNAMIC 0x0100
++#define VS_STATIC 0x0200
+
++#define VS_ATR_MASK 0x0F00
+
++#ifdef CONFIG_VSERVER_PRIVACY
++#define VS_ADMIN_P (0)
++#define VS_WATCH_P (0)
++#else
++#define VS_ADMIN_P VS_ADMIN
++#define VS_WATCH_P VS_WATCH
++#endif
+
++#define VS_HARDIRQ 0x1000
++#define VS_SOFTIRQ 0x2000
++#define VS_IRQ 0x4000
+
-+static inline int addr_in_nx_info(struct nx_info *nxi, uint32_t addr)
-+{
-+ int n, i;
++#define VS_IRQ_MASK 0xF000
+
-+ if (!nxi)
-+ return 1;
++#include <linux/hardirq.h>
+
-+ n = nxi->nbipv4;
-+ if (n && (nxi->ipv4[0] == 0))
-+ return 1;
-+ for (i = 0; i < n; i++) {
-+ if (nxi->ipv4[i] == addr)
++/*
++ * check current context for ADMIN/WATCH and
++ * optionally against supplied argument
++ */
++static inline int __vs_check(int cid, int id, unsigned int mode)
++{
++ if (mode & VS_ARG_MASK) {
++ if ((mode & VS_IDENT) && (id == cid))
+ return 1;
+ }
-+ return 0;
++ if (mode & VS_ATR_MASK) {
++ if ((mode & VS_DYNAMIC) &&
++ (id >= MIN_D_CONTEXT) &&
++ (id <= MAX_S_CONTEXT))
++ return 1;
++ if ((mode & VS_STATIC) &&
++ (id > 1) && (id < MIN_D_CONTEXT))
++ return 1;
++ }
++ if (mode & VS_IRQ_MASK) {
++ if ((mode & VS_IRQ) && unlikely(in_interrupt()))
++ return 1;
++ if ((mode & VS_HARDIRQ) && unlikely(in_irq()))
++ return 1;
++ if ((mode & VS_SOFTIRQ) && unlikely(in_softirq()))
++ return 1;
++ }
++ return (((mode & VS_ADMIN) && (cid == 0)) ||
++ ((mode & VS_WATCH) && (cid == 1)) ||
++ ((mode & VS_HOSTID) && (id == 0)));
+}
+
-+static inline void exit_nx_info(struct task_struct *p)
-+{
-+ if (p->nx_info)
-+ release_nx_info(p->nx_info, p);
-+}
++#define vx_check(c, m) __vs_check(vx_current_xid(), c, (m) | VS_IRQ)
+
++#define vx_weak_check(c, m) ((m) ? vx_check(c, m) : 1)
+
-+#else
-+#warning duplicate inclusion
-+#endif
---- linux-2.6.22/include/linux/vs_pid.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_pid.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,133 @@
-+#ifndef _VS_PID_H
-+#define _VS_PID_H
+
-+#include "vserver/base.h"
-+#include "vserver/context.h"
-+#include "vserver/debug.h"
-+#include <linux/pid_namespace.h>
++#define nx_check(c, m) __vs_check(nx_current_nid(), c, m)
+
++#define nx_weak_check(c, m) ((m) ? nx_check(c, m) : 1)
+
-+/* pid faking stuff */
++#endif
+--- linux-2.6.25.4/include/linux/vserver/context_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/context_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,128 @@
++#ifndef _VX_CONTEXT_CMD_H
++#define _VX_CONTEXT_CMD_H
+
+
-+#define vx_info_map_pid(v, p) \
-+ __vx_info_map_pid((v), (p), __FUNC__, __FILE__, __LINE__)
-+#define vx_info_map_tgid(v,p) vx_info_map_pid(v,p)
-+#define vx_map_pid(p) vx_info_map_pid(current->vx_info, p)
-+#define vx_map_tgid(p) vx_map_pid(p)
++/* vinfo commands */
+
-+static inline int __vx_info_map_pid(struct vx_info *vxi, int pid,
-+ const char *func, const char *file, int line)
-+{
-+ if (vx_info_flags(vxi, VXF_INFO_INIT, 0)) {
-+ vxfprintk(VXD_CBIT(cvirt, 2),
-+ "vx_map_tgid: %p/%llx: %d -> %d",
-+ vxi, (long long)vxi->vx_flags, pid,
-+ (pid && pid == vxi->vx_initpid) ? 1 : pid,
-+ func, file, line);
-+ if (pid == 0)
-+ return 0;
-+ if (pid == vxi->vx_initpid)
-+ return 1;
-+ }
-+ return pid;
-+}
++#define VCMD_task_xid VC_CMD(VINFO, 1, 0)
+
-+#define vx_info_rmap_pid(v, p) \
-+ __vx_info_rmap_pid((v), (p), __FUNC__, __FILE__, __LINE__)
-+#define vx_rmap_pid(p) vx_info_rmap_pid(current->vx_info, p)
-+#define vx_rmap_tgid(p) vx_rmap_pid(p)
++#ifdef __KERNEL__
++extern int vc_task_xid(uint32_t);
+
-+static inline int __vx_info_rmap_pid(struct vx_info *vxi, int pid,
-+ const char *func, const char *file, int line)
-+{
-+ if (vx_info_flags(vxi, VXF_INFO_INIT, 0)) {
-+ vxfprintk(VXD_CBIT(cvirt, 2),
-+ "vx_rmap_tgid: %p/%llx: %d -> %d",
-+ vxi, (long long)vxi->vx_flags, pid,
-+ (pid == 1) ? vxi->vx_initpid : pid,
-+ func, file, line);
-+ if ((pid == 1) && vxi->vx_initpid)
-+ return vxi->vx_initpid;
-+ if (pid == vxi->vx_initpid)
-+ return ~0U;
-+ }
-+ return pid;
-+}
++#endif /* __KERNEL__ */
+
++#define VCMD_vx_info VC_CMD(VINFO, 5, 0)
+
-+#define VXF_FAKE_INIT (VXF_INFO_INIT | VXF_STATE_INIT)
++struct vcmd_vx_info_v0 {
++ uint32_t xid;
++ uint32_t initpid;
++ /* more to come */
++};
+
-+static inline
-+int vx_proc_task_visible(struct task_struct *task)
-+{
-+ if ((task->pid == 1) &&
-+ !vx_flags(VXF_FAKE_INIT, VXF_FAKE_INIT))
-+ /* show a blend through init */
-+ goto visible;
-+ if (vx_check(vx_task_xid(task), VS_WATCH | VS_IDENT))
-+ goto visible;
-+ return 0;
-+visible:
-+ return 1;
-+}
++#ifdef __KERNEL__
++extern int vc_vx_info(struct vx_info *, void __user *);
+
-+static inline
-+struct task_struct *vx_find_proc_task_by_pid(int pid)
-+{
-+ struct task_struct *task = find_task_by_real_pid(pid);
++#endif /* __KERNEL__ */
+
-+ if (task && !vx_proc_task_visible(task)) {
-+ vxdprintk(VXD_CBIT(misc, 6),
-+ "dropping task (find) %p[#%u,%u] for %p[#%u,%u]",
-+ task, task->xid, task->pid,
-+ current, current->xid, current->pid);
-+ task = NULL;
-+ }
-+ return task;
-+}
++#define VCMD_ctx_stat VC_CMD(VSTAT, 0, 0)
+
-+static inline
-+struct task_struct *vx_get_proc_task(struct inode *inode, struct pid *pid)
-+{
-+ struct task_struct *task = get_pid_task(pid, PIDTYPE_PID);
++struct vcmd_ctx_stat_v0 {
++ uint32_t usecnt;
++ uint32_t tasks;
++ /* more to come */
++};
+
-+ if (task && !vx_proc_task_visible(task)) {
-+ vxdprintk(VXD_CBIT(misc, 6),
-+ "dropping task (get) %p[#%u,%u] for %p[#%u,%u]",
-+ task, task->xid, task->pid,
-+ current, current->xid, current->pid);
-+ put_task_struct(task);
-+ task = NULL;
-+ }
-+ return task;
-+}
++#ifdef __KERNEL__
++extern int vc_ctx_stat(struct vx_info *, void __user *);
+
++#endif /* __KERNEL__ */
+
-+static inline
-+struct task_struct *vx_child_reaper(struct task_struct *p)
-+{
-+ struct vx_info *vxi = p->vx_info;
-+ struct task_struct *reaper = child_reaper(p);
++/* context commands */
+
-+ if (!vxi)
-+ goto out;
++#define VCMD_ctx_create_v0 VC_CMD(VPROC, 1, 0)
++#define VCMD_ctx_create VC_CMD(VPROC, 1, 1)
+
-+ BUG_ON(!p->vx_info->vx_reaper);
++struct vcmd_ctx_create {
++ uint64_t flagword;
++};
+
-+ /* child reaper for the guest reaper */
-+ if (vxi->vx_reaper == p)
-+ goto out;
++#define VCMD_ctx_migrate_v0 VC_CMD(PROCMIG, 1, 0)
++#define VCMD_ctx_migrate VC_CMD(PROCMIG, 1, 1)
+
-+ reaper = vxi->vx_reaper;
-+out:
-+ vxdprintk(VXD_CBIT(xid, 7),
-+ "vx_child_reaper(%p[#%u,%u]) = %p[#%u,%u]",
-+ p, p->xid, p->pid, reaper, reaper->xid, reaper->pid);
-+ return reaper;
-+}
++struct vcmd_ctx_migrate {
++ uint64_t flagword;
++};
+
++#ifdef __KERNEL__
++extern int vc_ctx_create(uint32_t, void __user *);
++extern int vc_ctx_migrate(struct vx_info *, void __user *);
+
-+#else
-+#warning duplicate inclusion
-+#endif
---- linux-2.6.22/include/linux/vs_sched.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_sched.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,110 @@
-+#ifndef _VS_SCHED_H
-+#define _VS_SCHED_H
++#endif /* __KERNEL__ */
+
-+#include "vserver/base.h"
-+#include "vserver/context.h"
-+#include "vserver/sched.h"
+
++/* flag commands */
+
-+#define VAVAVOOM_RATIO 50
++#define VCMD_get_cflags VC_CMD(FLAGS, 1, 0)
++#define VCMD_set_cflags VC_CMD(FLAGS, 2, 0)
+
-+#define MAX_PRIO_BIAS 20
-+#define MIN_PRIO_BIAS -20
++struct vcmd_ctx_flags_v0 {
++ uint64_t flagword;
++ uint64_t mask;
++};
+
++#ifdef __KERNEL__
++extern int vc_get_cflags(struct vx_info *, void __user *);
++extern int vc_set_cflags(struct vx_info *, void __user *);
+
-+#ifdef CONFIG_VSERVER_HARDCPU
++#endif /* __KERNEL__ */
+
-+/*
-+ * effective_prio - return the priority that is based on the static
-+ * priority but is modified by bonuses/penalties.
-+ *
-+ * We scale the actual sleep average [0 .... MAX_SLEEP_AVG]
-+ * into a -4 ... 0 ... +4 bonus/penalty range.
-+ *
-+ * Additionally, we scale another amount based on the number of
-+ * CPU tokens currently held by the context, if the process is
-+ * part of a context (and the appropriate SCHED flag is set).
-+ * This ranges from -5 ... 0 ... +15, quadratically.
-+ *
-+ * So, the total bonus is -9 .. 0 .. +19
-+ * We use ~50% of the full 0...39 priority range so that:
-+ *
-+ * 1) nice +19 interactive tasks do not preempt nice 0 CPU hogs.
-+ * 2) nice -20 CPU hogs do not get preempted by nice 0 tasks.
-+ * unless that context is far exceeding its CPU allocation.
-+ *
-+ * Both properties are important to certain workloads.
-+ */
-+static inline
-+int vx_effective_vavavoom(struct _vx_sched_pc *sched_pc, int max_prio)
-+{
-+ int vavavoom, max;
+
-+ /* lots of tokens = lots of vavavoom
-+ * no tokens = no vavavoom */
-+ if ((vavavoom = sched_pc->tokens) >= 0) {
-+ max = sched_pc->tokens_max;
-+ vavavoom = max - vavavoom;
-+ max = max * max;
-+ vavavoom = max_prio * VAVAVOOM_RATIO / 100
-+ * (vavavoom*vavavoom - (max >> 2)) / max;
-+ return vavavoom;
-+ }
-+ return 0;
-+}
++/* context caps commands */
+
++#define VCMD_get_ccaps VC_CMD(FLAGS, 3, 1)
++#define VCMD_set_ccaps VC_CMD(FLAGS, 4, 1)
+
-+static inline
-+int vx_adjust_prio(struct task_struct *p, int prio, int max_user)
-+{
-+ struct vx_info *vxi = p->vx_info;
-+ struct _vx_sched_pc *sched_pc;
++struct vcmd_ctx_caps_v1 {
++ uint64_t ccaps;
++ uint64_t cmask;
++};
+
-+ if (!vxi)
-+ return prio;
++#ifdef __KERNEL__
++extern int vc_get_ccaps(struct vx_info *, void __user *);
++extern int vc_set_ccaps(struct vx_info *, void __user *);
+
-+ sched_pc = &vx_cpu(vxi, sched_pc);
-+ if (vx_info_flags(vxi, VXF_SCHED_PRIO, 0)) {
-+ int vavavoom = vx_effective_vavavoom(sched_pc, max_user);
++#endif /* __KERNEL__ */
+
-+ sched_pc->vavavoom = vavavoom;
-+ prio += vavavoom;
-+ }
-+ prio += sched_pc->prio_bias;
-+ return prio;
-+}
+
-+#else /* !CONFIG_VSERVER_HARDCPU */
++/* bcaps commands */
+
-+static inline
-+int vx_adjust_prio(struct task_struct *p, int prio, int max_user)
-+{
-+ struct vx_info *vxi = p->vx_info;
++#define VCMD_get_bcaps VC_CMD(FLAGS, 9, 0)
++#define VCMD_set_bcaps VC_CMD(FLAGS, 10, 0)
+
-+ if (vxi)
-+ prio += vx_cpu(vxi, sched_pc).prio_bias;
-+ return prio;
-+}
++struct vcmd_bcaps {
++ uint64_t bcaps;
++ uint64_t bmask;
++};
+
-+#endif /* CONFIG_VSERVER_HARDCPU */
++#ifdef __KERNEL__
++extern int vc_get_bcaps(struct vx_info *, void __user *);
++extern int vc_set_bcaps(struct vx_info *, void __user *);
+
++#endif /* __KERNEL__ */
+
-+static inline void vx_account_user(struct vx_info *vxi,
-+ cputime_t cputime, int nice)
-+{
-+ if (!vxi)
-+ return;
-+ vx_cpu(vxi, sched_pc).user_ticks += cputime;
-+}
+
-+static inline void vx_account_system(struct vx_info *vxi,
-+ cputime_t cputime, int idle)
-+{
-+ if (!vxi)
-+ return;
-+ vx_cpu(vxi, sched_pc).sys_ticks += cputime;
-+}
++/* OOM badness */
+
-+#else
-+#warning duplicate inclusion
-+#endif
---- linux-2.6.22/include/linux/vs_socket.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_socket.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,67 @@
-+#ifndef _VS_SOCKET_H
-+#define _VS_SOCKET_H
++#define VCMD_get_badness VC_CMD(MEMCTRL, 5, 0)
++#define VCMD_set_badness VC_CMD(MEMCTRL, 6, 0)
+
-+#include "vserver/debug.h"
-+#include "vserver/base.h"
-+#include "vserver/cacct.h"
-+#include "vserver/context.h"
++struct vcmd_badness_v0 {
++ int64_t bias;
++};
+
++#ifdef __KERNEL__
++extern int vc_get_badness(struct vx_info *, void __user *);
++extern int vc_set_badness(struct vx_info *, void __user *);
+
-+/* socket accounting */
++#endif /* __KERNEL__ */
++#endif /* _VX_CONTEXT_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/context.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/context.h 2008-04-22 15:14:28.000000000 -0400
+@@ -0,0 +1,176 @@
++#ifndef _VX_CONTEXT_H
++#define _VX_CONTEXT_H
+
-+#include <linux/socket.h>
++#include <linux/types.h>
++#include <linux/capability.h>
+
-+static inline int vx_sock_type(int family)
-+{
-+ switch (family) {
-+ case PF_UNSPEC:
-+ return VXA_SOCK_UNSPEC;
-+ case PF_UNIX:
-+ return VXA_SOCK_UNIX;
-+ case PF_INET:
-+ return VXA_SOCK_INET;
-+ case PF_INET6:
-+ return VXA_SOCK_INET6;
-+ case PF_PACKET:
-+ return VXA_SOCK_PACKET;
-+ default:
-+ return VXA_SOCK_OTHER;
-+ }
-+}
+
-+#define vx_acc_sock(v, f, p, s) \
-+ __vx_acc_sock(v, f, p, s, __FILE__, __LINE__)
++/* context flags */
+
-+static inline void __vx_acc_sock(struct vx_info *vxi,
-+ int family, int pos, int size, char *file, int line)
-+{
-+ if (vxi) {
-+ int type = vx_sock_type(family);
++#define VXF_INFO_SCHED 0x00000002
++#define VXF_INFO_NPROC 0x00000004
++#define VXF_INFO_PRIVATE 0x00000008
+
-+ atomic_long_inc(&vxi->cacct.sock[type][pos].count);
-+ atomic_long_add(size, &vxi->cacct.sock[type][pos].total);
-+ }
-+}
++#define VXF_INFO_INIT 0x00000010
++#define VXF_INFO_HIDE 0x00000020
++#define VXF_INFO_ULIMIT 0x00000040
++#define VXF_INFO_NSPACE 0x00000080
+
-+#define vx_sock_recv(sk, s) \
-+ vx_acc_sock((sk)->sk_vx_info, (sk)->sk_family, 0, s)
-+#define vx_sock_send(sk, s) \
-+ vx_acc_sock((sk)->sk_vx_info, (sk)->sk_family, 1, s)
-+#define vx_sock_fail(sk, s) \
-+ vx_acc_sock((sk)->sk_vx_info, (sk)->sk_family, 2, s)
++#define VXF_SCHED_HARD 0x00000100
++#define VXF_SCHED_PRIO 0x00000200
++#define VXF_SCHED_PAUSE 0x00000400
+
++#define VXF_VIRT_MEM 0x00010000
++#define VXF_VIRT_UPTIME 0x00020000
++#define VXF_VIRT_CPU 0x00040000
++#define VXF_VIRT_LOAD 0x00080000
++#define VXF_VIRT_TIME 0x00100000
+
-+#define sock_vx_init(s) do { \
-+ (s)->sk_xid = 0; \
-+ (s)->sk_vx_info = NULL; \
-+ } while (0)
++#define VXF_HIDE_MOUNT 0x01000000
++/* was VXF_HIDE_NETIF 0x02000000 */
++#define VXF_HIDE_VINFO 0x04000000
+
-+#define sock_nx_init(s) do { \
-+ (s)->sk_nid = 0; \
-+ (s)->sk_nx_info = NULL; \
-+ } while (0)
++#define VXF_STATE_SETUP (1ULL << 32)
++#define VXF_STATE_INIT (1ULL << 33)
++#define VXF_STATE_ADMIN (1ULL << 34)
+
++#define VXF_SC_HELPER (1ULL << 36)
++#define VXF_REBOOT_KILL (1ULL << 37)
++#define VXF_PERSISTENT (1ULL << 38)
+
-+#else
-+#warning duplicate inclusion
-+#endif
---- linux-2.6.22/include/linux/vs_tag.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_tag.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,43 @@
-+#ifndef _VS_TAG_H
-+#define _VS_TAG_H
++#define VXF_FORK_RSS (1ULL << 48)
++#define VXF_PROLIFIC (1ULL << 49)
+
-+#include <linux/vserver/tag.h>
++#define VXF_IGNEG_NICE (1ULL << 52)
+
-+/* check conditions */
++#define VXF_ONE_TIME (0x0007ULL << 32)
+
-+#define DX_ADMIN 0x0001
-+#define DX_WATCH 0x0002
-+#define DX_HOSTID 0x0008
++#define VXF_INIT_SET (VXF_STATE_SETUP | VXF_STATE_INIT | VXF_STATE_ADMIN)
+
-+#define DX_IDENT 0x0010
+
-+#define DX_ARG_MASK 0x0010
++/* context migration */
+
++#define VXM_SET_INIT 0x00000001
++#define VXM_SET_REAPER 0x00000002
+
-+#define dx_task_tag(t) ((t)->xid)
++/* context caps */
+
-+#define dx_current_tag() dx_task_tag(current)
++#define VXC_CAP_MASK 0x00000000
+
-+#define dx_check(c, m) __dx_check(dx_current_tag(), c, m)
++#define VXC_SET_UTSNAME 0x00000001
++#define VXC_SET_RLIMIT 0x00000002
+
-+#define dx_weak_check(c, m) ((m) ? dx_check(c, m) : 1)
++/* was VXC_RAW_ICMP 0x00000100 */
++#define VXC_SYSLOG 0x00001000
+
++#define VXC_SECURE_MOUNT 0x00010000
++#define VXC_SECURE_REMOUNT 0x00020000
++#define VXC_BINARY_MOUNT 0x00040000
+
-+/*
-+ * check current context for ADMIN/WATCH and
-+ * optionally against supplied argument
-+ */
-+static inline int __dx_check(tag_t cid, tag_t id, unsigned int mode)
-+{
-+ if (mode & DX_ARG_MASK) {
-+ if ((mode & DX_IDENT) && (id == cid))
-+ return 1;
-+ }
-+ return (((mode & DX_ADMIN) && (cid == 0)) ||
-+ ((mode & DX_WATCH) && (cid == 1)) ||
-+ ((mode & DX_HOSTID) && (id == 0)));
-+}
++#define VXC_QUOTA_CTL 0x00100000
++#define VXC_ADMIN_MAPPER 0x00200000
++#define VXC_ADMIN_CLOOP 0x00400000
+
-+#else
-+#warning duplicate inclusion
-+#endif
---- linux-2.6.22/include/linux/vs_time.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vs_time.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,19 @@
-+#ifndef _VS_TIME_H
-+#define _VS_TIME_H
++#define VXC_KTHREAD 0x01000000
+
+
-+/* time faking stuff */
++#ifdef __KERNEL__
+
-+#ifdef CONFIG_VSERVER_VTIME
++#include <linux/list.h>
++#include <linux/spinlock.h>
++#include <linux/rcupdate.h>
+
-+extern void vx_gettimeofday(struct timeval *tv);
-+extern int vx_settimeofday(struct timespec *ts);
++#include "limit_def.h"
++#include "sched_def.h"
++#include "cvirt_def.h"
++#include "cacct_def.h"
++#include "device_def.h"
+
-+#else
-+#define vx_gettimeofday(t) do_gettimeofday(t)
-+#define vx_settimeofday(t) do_settimeofday(t)
-+#endif
++struct _vx_info_pc {
++ struct _vx_sched_pc sched_pc;
++ struct _vx_cvirt_pc cvirt_pc;
++};
+
-+#else
-+#warning duplicate inclusion
-+#endif
---- linux-2.6.22/include/linux/vserver/Kbuild 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/Kbuild 2007-06-16 19:48:30 +0200
-@@ -0,0 +1,9 @@
++struct vx_info {
++ struct hlist_node vx_hlist; /* linked list of contexts */
++ xid_t vx_id; /* context id */
++ atomic_t vx_usecnt; /* usage count */
++ atomic_t vx_tasks; /* tasks count */
++ struct vx_info *vx_parent; /* parent context */
++ int vx_state; /* context state */
+
-+unifdef-y += context_cmd.h network_cmd.h space_cmd.h \
-+ cacct_cmd.h cvirt_cmd.h limit_cmd.h dlimit_cmd.h \
-+ inode_cmd.h sched_cmd.h signal_cmd.h debug_cmd.h
++ unsigned long vx_nsmask; /* assignment mask */
++ struct nsproxy *vx_nsproxy; /* private namespace */
++ struct fs_struct *vx_fs; /* private namespace fs */
+
-+unifdef-y += switch.h network.h monitor.h inode.h
++ uint64_t vx_flags; /* context flags */
++ uint64_t vx_ccaps; /* context caps (vserver) */
++ kernel_cap_t vx_bcaps; /* bounding caps (system) */
++ // kernel_cap_t vx_cap_bset; /* the guest's bset */
+
-+unifdef-y += legacy.h
++ struct task_struct *vx_reaper; /* guest reaper process */
++ pid_t vx_initpid; /* PID of guest init */
++ int64_t vx_badness_bias; /* OOM points bias */
+
---- linux-2.6.22/include/linux/vserver/base.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/base.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,236 @@
-+#ifndef _VX_BASE_H
-+#define _VX_BASE_H
++ struct _vx_limit limit; /* vserver limits */
++ struct _vx_sched sched; /* vserver scheduler */
++ struct _vx_cvirt cvirt; /* virtual/bias stuff */
++ struct _vx_cacct cacct; /* context accounting */
+
++ struct _vx_device dmap; /* default device map targets */
+
-+/* context state changes */
++#ifndef CONFIG_SMP
++ struct _vx_info_pc info_pc; /* per cpu data */
++#else
++ struct _vx_info_pc *ptr_pc; /* per cpu array */
++#endif
+
-+enum {
-+ VSC_STARTUP = 1,
-+ VSC_SHUTDOWN,
++ wait_queue_head_t vx_wait; /* context exit waitqueue */
++ int reboot_cmd; /* last sys_reboot() cmd */
++ int exit_code; /* last process exit code */
+
-+ VSC_NETUP,
-+ VSC_NETDOWN,
++ char vx_name[65]; /* vserver name */
+};
+
-+
-+#define MAX_S_CONTEXT 65535 /* Arbitrary limit */
-+
-+#ifdef CONFIG_VSERVER_DYNAMIC_IDS
-+#define MIN_D_CONTEXT 49152 /* dynamic contexts start here */
++#ifndef CONFIG_SMP
++#define vx_ptr_pc(vxi) (&(vxi)->info_pc)
++#define vx_per_cpu(vxi, v, id) vx_ptr_pc(vxi)->v
+#else
-+#define MIN_D_CONTEXT 65536
++#define vx_ptr_pc(vxi) ((vxi)->ptr_pc)
++#define vx_per_cpu(vxi, v, id) per_cpu_ptr(vx_ptr_pc(vxi), id)->v
+#endif
+
-+/* check conditions */
++#define vx_cpu(vxi, v) vx_per_cpu(vxi, v, smp_processor_id())
+
-+#define VS_ADMIN 0x0001
-+#define VS_WATCH 0x0002
-+#define VS_HIDE 0x0004
-+#define VS_HOSTID 0x0008
+
-+#define VS_IDENT 0x0010
-+#define VS_EQUIV 0x0020
-+#define VS_PARENT 0x0040
-+#define VS_CHILD 0x0080
++struct vx_info_save {
++ struct vx_info *vxi;
++ xid_t xid;
++};
+
-+#define VS_ARG_MASK 0x00F0
+
-+#define VS_DYNAMIC 0x0100
-+#define VS_STATIC 0x0200
++/* status flags */
+
-+#define VS_ATR_MASK 0x0F00
++#define VXS_HASHED 0x0001
++#define VXS_PAUSED 0x0010
++#define VXS_SHUTDOWN 0x0100
++#define VXS_HELPER 0x1000
++#define VXS_RELEASED 0x8000
+
-+#ifdef CONFIG_VSERVER_PRIVACY
-+#define VS_ADMIN_P (0)
-+#define VS_WATCH_P (0)
-+#else
-+#define VS_ADMIN_P VS_ADMIN
-+#define VS_WATCH_P VS_WATCH
-+#endif
+
-+#define VS_HARDIRQ 0x1000
-+#define VS_SOFTIRQ 0x2000
-+#define VS_IRQ 0x4000
++extern void claim_vx_info(struct vx_info *, struct task_struct *);
++extern void release_vx_info(struct vx_info *, struct task_struct *);
+
-+#define VS_IRQ_MASK 0xF000
++extern struct vx_info *lookup_vx_info(int);
++extern struct vx_info *lookup_or_create_vx_info(int);
+
-+#include <linux/hardirq.h>
++extern int get_xid_list(int, unsigned int *, int);
++extern int xid_is_hashed(xid_t);
+
-+/*
-+ * check current context for ADMIN/WATCH and
-+ * optionally against supplied argument
-+ */
-+static inline int __vs_check(int cid, int id, unsigned int mode)
-+{
-+ if (mode & VS_ARG_MASK) {
-+ if ((mode & VS_IDENT) && (id == cid))
-+ return 1;
-+ }
-+ if (mode & VS_ATR_MASK) {
-+ if ((mode & VS_DYNAMIC) &&
-+ (id >= MIN_D_CONTEXT) &&
-+ (id <= MAX_S_CONTEXT))
-+ return 1;
-+ if ((mode & VS_STATIC) &&
-+ (id > 1) && (id < MIN_D_CONTEXT))
-+ return 1;
-+ }
-+ if (mode & VS_IRQ_MASK) {
-+ if ((mode & VS_IRQ) && unlikely(in_interrupt()))
-+ return 1;
-+ if ((mode & VS_HARDIRQ) && unlikely(in_irq()))
-+ return 1;
-+ if ((mode & VS_SOFTIRQ) && unlikely(in_softirq()))
-+ return 1;
-+ }
-+ return (((mode & VS_ADMIN) && (cid == 0)) ||
-+ ((mode & VS_WATCH) && (cid == 1)) ||
-+ ((mode & VS_HOSTID) && (id == 0)));
-+}
-+
-+#define vx_task_xid(t) ((t)->xid)
++extern int vx_migrate_task(struct task_struct *, struct vx_info *, int);
+
-+#define vx_current_xid() vx_task_xid(current)
++extern long vs_state_change(struct vx_info *, unsigned int);
+
-+#define current_vx_info() (current->vx_info)
+
++#endif /* __KERNEL__ */
++#endif /* _VX_CONTEXT_H */
+--- linux-2.6.25.4/include/linux/vserver/cvirt_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/cvirt_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,53 @@
++#ifndef _VX_CVIRT_CMD_H
++#define _VX_CVIRT_CMD_H
+
-+#define vx_check(c, m) __vs_check(vx_current_xid(), c, (m) | VS_IRQ)
+
-+#define vx_weak_check(c, m) ((m) ? vx_check(c, m) : 1)
++/* virtual host info name commands */
+
++#define VCMD_set_vhi_name VC_CMD(VHOST, 1, 0)
++#define VCMD_get_vhi_name VC_CMD(VHOST, 2, 0)
+
-+#define nx_task_nid(t) ((t)->nid)
++struct vcmd_vhi_name_v0 {
++ uint32_t field;
++ char name[65];
++};
+
-+#define nx_current_nid() nx_task_nid(current)
+
-+#define current_nx_info() (current->nx_info)
++enum vhi_name_field {
++ VHIN_CONTEXT = 0,
++ VHIN_SYSNAME,
++ VHIN_NODENAME,
++ VHIN_RELEASE,
++ VHIN_VERSION,
++ VHIN_MACHINE,
++ VHIN_DOMAINNAME,
++};
+
+
-+#define nx_check(c, m) __vs_check(nx_current_nid(), c, m)
++#ifdef __KERNEL__
+
-+#define nx_weak_check(c, m) ((m) ? nx_check(c, m) : 1)
++#include <linux/compiler.h>
+
++extern int vc_set_vhi_name(struct vx_info *, void __user *);
++extern int vc_get_vhi_name(struct vx_info *, void __user *);
+
++#endif /* __KERNEL__ */
+
-+/* generic flag merging */
++#define VCMD_virt_stat VC_CMD(VSTAT, 3, 0)
+
-+#define vs_check_flags(v, m, f) (((v) & (m)) ^ (f))
++struct vcmd_virt_stat_v0 {
++ uint64_t offset;
++ uint64_t uptime;
++ uint32_t nr_threads;
++ uint32_t nr_running;
++ uint32_t nr_uninterruptible;
++ uint32_t nr_onhold;
++ uint32_t nr_forks;
++ uint32_t load[3];
++};
+
-+#define vs_mask_flags(v, f, m) (((v) & ~(m)) | ((f) & (m)))
++#ifdef __KERNEL__
++extern int vc_virt_stat(struct vx_info *, void __user *);
+
-+#define vs_mask_mask(v, f, m) (((v) & ~(m)) | ((v) & (f) & (m)))
++#endif /* __KERNEL__ */
++#endif /* _VX_CVIRT_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/cvirt_def.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/cvirt_def.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,80 @@
++#ifndef _VX_CVIRT_DEF_H
++#define _VX_CVIRT_DEF_H
+
-+#define vs_check_bit(v, n) ((v) & (1LL << (n)))
++#include <linux/jiffies.h>
++#include <linux/spinlock.h>
++#include <linux/wait.h>
++#include <linux/time.h>
++#include <asm/atomic.h>
+
+
-+/* context flags */
++struct _vx_usage_stat {
++ uint64_t user;
++ uint64_t nice;
++ uint64_t system;
++ uint64_t softirq;
++ uint64_t irq;
++ uint64_t idle;
++ uint64_t iowait;
++};
+
-+#define __vx_flags(v) ((v) ? (v)->vx_flags : 0)
++struct _vx_syslog {
++ wait_queue_head_t log_wait;
++ spinlock_t logbuf_lock; /* lock for the log buffer */
+
-+#define vx_current_flags() __vx_flags(current->vx_info)
++ unsigned long log_start; /* next char to be read by syslog() */
++ unsigned long con_start; /* next char to be sent to consoles */
++ unsigned long log_end; /* most-recently-written-char + 1 */
++ unsigned long logged_chars; /* #chars since last read+clear operation */
+
-+#define vx_info_flags(v, m, f) \
-+ vs_check_flags(__vx_flags(v), m, f)
++ char log_buf[1024];
++};
+
-+#define task_vx_flags(t, m, f) \
-+ ((t) && vx_info_flags((t)->vx_info, m, f))
+
-+#define vx_flags(m, f) vx_info_flags(current->vx_info, m, f)
++/* context sub struct */
+
++struct _vx_cvirt {
++ atomic_t nr_threads; /* number of current threads */
++ atomic_t nr_running; /* number of running threads */
++ atomic_t nr_uninterruptible; /* number of uninterruptible threads */
+
-+/* context caps */
++ atomic_t nr_onhold; /* processes on hold */
++ uint32_t onhold_last; /* jiffies when put on hold */
+
-+#define __vx_ccaps(v) ((v) ? (v)->vx_ccaps : 0)
++ struct timeval bias_tv; /* time offset to the host */
++ struct timespec bias_idle;
++ struct timespec bias_uptime; /* context creation point */
++ uint64_t bias_clock; /* offset in clock_t */
+
-+#define vx_current_ccaps() __vx_ccaps(current->vx_info)
++ spinlock_t load_lock; /* lock for the load averages */
++ atomic_t load_updates; /* nr of load updates done so far */
++ uint32_t load_last; /* last time load was calculated */
++ uint32_t load[3]; /* load averages 1,5,15 */
+
-+#define vx_info_ccaps(v, c) (__vx_ccaps(v) & (c))
++ atomic_t total_forks; /* number of forks so far */
+
-+#define vx_ccaps(c) vx_info_ccaps(current->vx_info, (c))
++ struct _vx_syslog syslog;
++};
+
++struct _vx_cvirt_pc {
++ struct _vx_usage_stat cpustat;
++};
+
+
-+/* network flags */
++#ifdef CONFIG_VSERVER_DEBUG
+
-+#define __nx_flags(v) ((v) ? (v)->nx_flags : 0)
++static inline void __dump_vx_cvirt(struct _vx_cvirt *cvirt)
++{
++ printk("\t_vx_cvirt:\n");
++ printk("\t threads: %4d, %4d, %4d, %4d\n",
++ atomic_read(&cvirt->nr_threads),
++ atomic_read(&cvirt->nr_running),
++ atomic_read(&cvirt->nr_uninterruptible),
++ atomic_read(&cvirt->nr_onhold));
++ /* add rest here */
++ printk("\t total_forks = %d\n", atomic_read(&cvirt->total_forks));
++}
+
-+#define nx_current_flags() __nx_flags(current->nx_info)
++#endif
+
-+#define nx_info_flags(v, m, f) \
-+ vs_check_flags(__nx_flags(v), m, f)
++#endif /* _VX_CVIRT_DEF_H */
+--- linux-2.6.25.4/include/linux/vserver/cvirt.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/cvirt.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,20 @@
++#ifndef _VX_CVIRT_H
++#define _VX_CVIRT_H
+
-+#define task_nx_flags(t, m, f) \
-+ ((t) && nx_info_flags((t)->nx_info, m, f))
+
-+#define nx_flags(m, f) nx_info_flags(current->nx_info, m, f)
++#ifdef __KERNEL__
+
++struct timespec;
+
-+/* network caps */
++void vx_vsi_uptime(struct timespec *, struct timespec *);
+
-+#define __nx_ncaps(v) ((v) ? (v)->nx_ncaps : 0)
+
-+#define nx_current_ncaps() __nx_ncaps(current->nx_info)
++struct vx_info;
+
-+#define nx_info_ncaps(v, c) (__nx_ncaps(v) & (c))
++void vx_update_load(struct vx_info *);
+
-+#define nx_ncaps(c) nx_info_ncaps(current->nx_info, c)
+
++int vx_do_syslog(int, char __user *, int);
+
-+/* context mask capabilities */
++#endif /* __KERNEL__ */
++#endif /* _VX_CVIRT_H */
+--- linux-2.6.25.4/include/linux/vserver/debug_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/debug_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,58 @@
++#ifndef _VX_DEBUG_CMD_H
++#define _VX_DEBUG_CMD_H
+
-+#define __vx_mcaps(v) ((v) ? (v)->vx_ccaps >> 32UL : ~0 )
+
-+#define vx_info_mcaps(v, c) (__vx_mcaps(v) & (c))
++/* debug commands */
+
-+#define vx_mcaps(c) vx_info_mcaps(current->vx_info, c)
++#define VCMD_dump_history VC_CMD(DEBUG, 1, 0)
+
++#define VCMD_read_history VC_CMD(DEBUG, 5, 0)
++#define VCMD_read_monitor VC_CMD(DEBUG, 6, 0)
+
-+/* context bcap mask */
++struct vcmd_read_history_v0 {
++ uint32_t index;
++ uint32_t count;
++ char __user *data;
++};
+
-+#define __vx_bcaps(v) ((v) ? (v)->vx_bcaps : ~0 )
++struct vcmd_read_monitor_v0 {
++ uint32_t index;
++ uint32_t count;
++ char __user *data;
++};
+
-+#define vx_current_bcaps() __vx_bcaps(current->vx_info)
+
-+#define vx_info_bcaps(v, c) (__vx_bcaps(v) & (c))
++#ifdef __KERNEL__
+
-+#define vx_bcaps(c) vx_info_bcaps(current->vx_info, c)
++#ifdef CONFIG_COMPAT
+
++#include <asm/compat.h>
+
-+#define vx_info_cap_bset(v) ((v) ? (v)->vx_cap_bset : cap_bset)
++struct vcmd_read_history_v0_x32 {
++ uint32_t index;
++ uint32_t count;
++ compat_uptr_t data_ptr;
++};
+
-+#define vx_current_cap_bset() vx_info_cap_bset(current->vx_info)
++struct vcmd_read_monitor_v0_x32 {
++ uint32_t index;
++ uint32_t count;
++ compat_uptr_t data_ptr;
++};
+
++#endif /* CONFIG_COMPAT */
+
-+#define __vx_info_mbcap(v, b) \
-+ (!vx_info_flags(v, VXF_STATE_SETUP, 0) ? \
-+ vx_info_bcaps(v, b) : (b))
++extern int vc_dump_history(uint32_t);
+
-+#define vx_info_mbcap(v, b) __vx_info_mbcap(v, cap_t(b))
++extern int vc_read_history(uint32_t, void __user *);
++extern int vc_read_monitor(uint32_t, void __user *);
+
-+#define task_vx_mbcap(t, b) \
-+ vx_info_mbcap((t)->vx_info, (t)->b)
++#ifdef CONFIG_COMPAT
+
-+#define vx_mbcap(b) task_vx_mbcap(current, b)
++extern int vc_read_history_x32(uint32_t, void __user *);
++extern int vc_read_monitor_x32(uint32_t, void __user *);
+
-+#define vx_cap_raised(v, c, f) (vx_info_mbcap(v, c) & CAP_TO_MASK(f))
++#endif /* CONFIG_COMPAT */
+
-+#define vx_capable(b, c) (capable(b) || \
-+ (cap_raised(current->cap_effective, b) && vx_ccaps(c)))
++#endif /* __KERNEL__ */
++#endif /* _VX_DEBUG_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/debug.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/debug.h 2008-04-21 13:55:14.000000000 -0400
+@@ -0,0 +1,130 @@
++#ifndef _VX_DEBUG_H
++#define _VX_DEBUG_H
+
+
-+#define vx_current_initpid(n) \
-+ (current->vx_info && \
-+ (current->vx_info->vx_initpid == (n)))
++#define VXD_CBIT(n, m) (vx_debug_ ## n & (1 << (m)))
++#define VXD_CMIN(n, m) (vx_debug_ ## n > (m))
++#define VXD_MASK(n, m) (vx_debug_ ## n & (m))
+
++#define VXD_DEV(d) (d), (d)->bd_inode->i_ino, \
++ imajor((d)->bd_inode), iminor((d)->bd_inode)
++#define VXF_DEV "%p[%lu,%d:%d]"
+
-+#define __vx_state(v) ((v) ? ((v)->vx_state) : 0)
+
-+#define vx_info_state(v, m) (__vx_state(v) & (m))
++#define __FUNC__ __func__
+
+
-+#define __nx_state(v) ((v) ? ((v)->nx_state) : 0)
++#define vxd_path(p) \
++ ({ static char _buffer[PATH_MAX]; \
++ d_path(p, _buffer, sizeof(_buffer)); })
+
-+#define nx_info_state(v, m) (__nx_state(v) & (m))
++#define vxd_cond_path(n) \
++ ((n) ? vxd_path(&(n)->path) : "<null>" )
+
-+#endif
---- linux-2.6.22/include/linux/vserver/cacct.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/cacct.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,15 @@
-+#ifndef _VX_CACCT_H
-+#define _VX_CACCT_H
+
++#ifdef CONFIG_VSERVER_DEBUG
+
-+enum sock_acc_field {
-+ VXA_SOCK_UNSPEC = 0,
-+ VXA_SOCK_UNIX,
-+ VXA_SOCK_INET,
-+ VXA_SOCK_INET6,
-+ VXA_SOCK_PACKET,
-+ VXA_SOCK_OTHER,
-+ VXA_SOCK_SIZE /* array size */
-+};
-+
-+#endif /* _VX_CACCT_H */
---- linux-2.6.22/include/linux/vserver/cacct_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/cacct_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,23 @@
-+#ifndef _VX_CACCT_CMD_H
-+#define _VX_CACCT_CMD_H
++extern unsigned int vx_debug_switch;
++extern unsigned int vx_debug_xid;
++extern unsigned int vx_debug_nid;
++extern unsigned int vx_debug_tag;
++extern unsigned int vx_debug_net;
++extern unsigned int vx_debug_limit;
++extern unsigned int vx_debug_cres;
++extern unsigned int vx_debug_dlim;
++extern unsigned int vx_debug_quota;
++extern unsigned int vx_debug_cvirt;
++extern unsigned int vx_debug_space;
++extern unsigned int vx_debug_misc;
+
+
-+/* virtual host info name commands */
++#define VX_LOGLEVEL "vxD: "
++#define VX_PROC_FMT "%p: "
++#define VX_PROCESS current
+
-+#define VCMD_sock_stat VC_CMD(VSTAT, 5, 0)
++#define vxdprintk(c, f, x...) \
++ do { \
++ if (c) \
++ printk(VX_LOGLEVEL VX_PROC_FMT f "\n", \
++ VX_PROCESS , ##x); \
++ } while (0)
+
-+struct vcmd_sock_stat_v0 {
-+ uint32_t field;
-+ uint32_t count[3];
-+ uint64_t total[3];
-+};
++#define vxlprintk(c, f, x...) \
++ do { \
++ if (c) \
++ printk(VX_LOGLEVEL f " @%s:%d\n", x); \
++ } while (0)
+
++#define vxfprintk(c, f, x...) \
++ do { \
++ if (c) \
++ printk(VX_LOGLEVEL f " %s@%s:%d\n", x); \
++ } while (0)
+
-+#ifdef __KERNEL__
+
-+#include <linux/compiler.h>
++struct vx_info;
+
-+extern int vc_sock_stat(struct vx_info *, void __user *);
++void dump_vx_info(struct vx_info *, int);
++void dump_vx_info_inactive(int);
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_CACCT_CMD_H */
---- linux-2.6.22/include/linux/vserver/cacct_def.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/cacct_def.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,43 @@
-+#ifndef _VX_CACCT_DEF_H
-+#define _VX_CACCT_DEF_H
++#else /* CONFIG_VSERVER_DEBUG */
+
-+#include <asm/atomic.h>
-+#include <linux/vserver/cacct.h>
++#define vx_debug_switch 0
++#define vx_debug_xid 0
++#define vx_debug_nid 0
++#define vx_debug_tag 0
++#define vx_debug_net 0
++#define vx_debug_limit 0
++#define vx_debug_cres 0
++#define vx_debug_dlim 0
++#define vx_debug_cvirt 0
+
++#define vxdprintk(x...) do { } while (0)
++#define vxlprintk(x...) do { } while (0)
++#define vxfprintk(x...) do { } while (0)
+
-+struct _vx_sock_acc {
-+ atomic_long_t count;
-+ atomic_long_t total;
-+};
++#endif /* CONFIG_VSERVER_DEBUG */
+
-+/* context sub struct */
+
-+struct _vx_cacct {
-+ struct _vx_sock_acc sock[VXA_SOCK_SIZE][3];
-+ atomic_t slab[8];
-+ atomic_t page[6][8];
-+};
++#ifdef CONFIG_VSERVER_WARN
+
-+#ifdef CONFIG_VSERVER_DEBUG
++#define VX_WARNLEVEL KERN_WARNING "vxW: "
++#define VX_WARN_TASK "[»%s«,%u:#%u|%u|%u] "
++#define VX_WARN_XID "[xid #%u] "
++#define VX_WARN_NID "[nid #%u] "
++#define VX_WARN_TAG "[tag #%u] "
+
-+static inline void __dump_vx_cacct(struct _vx_cacct *cacct)
-+{
-+ int i, j;
++#define vxwprintk(c, f, x...) \
++ do { \
++ if (c) \
++ printk(VX_WARNLEVEL f "\n", ##x); \
++ } while (0)
+
-+ printk("\t_vx_cacct:");
-+ for (i = 0; i < 6; i++) {
-+ struct _vx_sock_acc *ptr = cacct->sock[i];
++#else /* CONFIG_VSERVER_WARN */
+
-+ printk("\t [%d] =", i);
-+ for (j = 0; j < 3; j++) {
-+ printk(" [%d] = %8lu, %8lu", j,
-+ atomic_long_read(&ptr[j].count),
-+ atomic_long_read(&ptr[j].total));
-+ }
-+ printk("\n");
-+ }
-+}
++#define vxwprintk(x...) do { } while (0)
+
-+#endif
++#endif /* CONFIG_VSERVER_WARN */
+
-+#endif /* _VX_CACCT_DEF_H */
---- linux-2.6.22/include/linux/vserver/cacct_int.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/cacct_int.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,21 @@
-+#ifndef _VX_CACCT_INT_H
-+#define _VX_CACCT_INT_H
++#define vxwprintk_task(c, f, x...) \
++ vxwprintk(c, VX_WARN_TASK f, \
++ current->comm, current->pid, \
++ current->xid, current->nid, current->tag, ##x)
++#define vxwprintk_xid(c, f, x...) \
++ vxwprintk(c, VX_WARN_XID f, current->xid, x)
++#define vxwprintk_nid(c, f, x...) \
++ vxwprintk(c, VX_WARN_NID f, current->nid, x)
++#define vxwprintk_tag(c, f, x...) \
++ vxwprintk(c, VX_WARN_TAG f, current->tag, x)
+
++#ifdef CONFIG_VSERVER_DEBUG
++#define vxd_assert_lock(l) assert_spin_locked(l)
++#define vxd_assert(c, f, x...) vxlprintk(!(c), \
++ "assertion [" f "] failed.", ##x, __FILE__, __LINE__)
++#else
++#define vxd_assert_lock(l) do { } while (0)
++#define vxd_assert(c, f, x...) do { } while (0)
++#endif
+
-+#ifdef __KERNEL__
+
-+static inline
-+unsigned long vx_sock_count(struct _vx_cacct *cacct, int type, int pos)
-+{
-+ return atomic_long_read(&cacct->sock[type][pos].count);
-+}
++#endif /* _VX_DEBUG_H */
+--- linux-2.6.25.4/include/linux/vserver/device_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/device_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,44 @@
++#ifndef _VX_DEVICE_CMD_H
++#define _VX_DEVICE_CMD_H
+
+
-+static inline
-+unsigned long vx_sock_total(struct _vx_cacct *cacct, int type, int pos)
-+{
-+ return atomic_long_read(&cacct->sock[type][pos].total);
-+}
++/* device vserver commands */
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_CACCT_INT_H */
---- linux-2.6.22/include/linux/vserver/context.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/context.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,172 @@
-+#ifndef _VX_CONTEXT_H
-+#define _VX_CONTEXT_H
++#define VCMD_set_mapping VC_CMD(DEVICE, 1, 0)
++#define VCMD_unset_mapping VC_CMD(DEVICE, 2, 0)
+
-+#include <linux/types.h>
-+#include <linux/capability.h>
++struct vcmd_set_mapping_v0 {
++ const char __user *device;
++ const char __user *target;
++ uint32_t flags;
++};
+
+
-+#define VX_DYNAMIC_ID ((uint32_t)-1) /* id for dynamic context */
++#ifdef __KERNEL__
+
-+/* context flags */
++#ifdef CONFIG_COMPAT
+
-+#define VXF_INFO_SCHED 0x00000002
-+#define VXF_INFO_NPROC 0x00000004
-+#define VXF_INFO_PRIVATE 0x00000008
++#include <asm/compat.h>
+
-+#define VXF_INFO_INIT 0x00000010
-+#define VXF_INFO_HIDE 0x00000020
-+#define VXF_INFO_ULIMIT 0x00000040
-+#define VXF_INFO_NSPACE 0x00000080
++struct vcmd_set_mapping_v0_x32 {
++ compat_uptr_t device_ptr;
++ compat_uptr_t target_ptr;
++ uint32_t flags;
++};
+
-+#define VXF_SCHED_HARD 0x00000100
-+#define VXF_SCHED_PRIO 0x00000200
-+#define VXF_SCHED_PAUSE 0x00000400
++#endif /* CONFIG_COMPAT */
+
-+#define VXF_VIRT_MEM 0x00010000
-+#define VXF_VIRT_UPTIME 0x00020000
-+#define VXF_VIRT_CPU 0x00040000
-+#define VXF_VIRT_LOAD 0x00080000
-+#define VXF_VIRT_TIME 0x00100000
++#include <linux/compiler.h>
+
-+#define VXF_HIDE_MOUNT 0x01000000
-+#define VXF_HIDE_NETIF 0x02000000
-+#define VXF_HIDE_VINFO 0x04000000
++extern int vc_set_mapping(struct vx_info *, void __user *);
++extern int vc_unset_mapping(struct vx_info *, void __user *);
+
-+#define VXF_STATE_SETUP (1ULL << 32)
-+#define VXF_STATE_INIT (1ULL << 33)
-+#define VXF_STATE_ADMIN (1ULL << 34)
++#ifdef CONFIG_COMPAT
+
-+#define VXF_SC_HELPER (1ULL << 36)
-+#define VXF_REBOOT_KILL (1ULL << 37)
-+#define VXF_PERSISTENT (1ULL << 38)
++extern int vc_set_mapping_x32(struct vx_info *, void __user *);
++extern int vc_unset_mapping_x32(struct vx_info *, void __user *);
+
-+#define VXF_FORK_RSS (1ULL << 48)
-+#define VXF_PROLIFIC (1ULL << 49)
++#endif /* CONFIG_COMPAT */
+
-+#define VXF_IGNEG_NICE (1ULL << 52)
++#endif /* __KERNEL__ */
++#endif /* _VX_DEVICE_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/device_def.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/device_def.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,17 @@
++#ifndef _VX_DEVICE_DEF_H
++#define _VX_DEVICE_DEF_H
+
-+#define VXF_ONE_TIME (0x0007ULL << 32)
++#include <linux/types.h>
+
-+#define VXF_INIT_SET (VXF_STATE_SETUP | VXF_STATE_INIT | VXF_STATE_ADMIN)
++struct vx_dmap_target {
++ dev_t target;
++ uint32_t flags;
++};
+
++struct _vx_device {
++#ifdef CONFIG_VSERVER_DEVICE
++ struct vx_dmap_target targets[2];
++#endif
++};
+
-+/* context migration */
++#endif /* _VX_DEVICE_DEF_H */
+--- linux-2.6.25.4/include/linux/vserver/device.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/device.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,15 @@
++#ifndef _VX_DEVICE_H
++#define _VX_DEVICE_H
+
-+#define VXM_SET_INIT 0x00000001
-+#define VXM_SET_REAPER 0x00000002
+
-+/* context caps */
++#define DATTR_CREATE 0x00000001
++#define DATTR_OPEN 0x00000002
+
-+#define VXC_CAP_MASK 0x00000000
++#define DATTR_REMAP 0x00000010
+
-+#define VXC_SET_UTSNAME 0x00000001
-+#define VXC_SET_RLIMIT 0x00000002
++#define DATTR_MASK 0x00000013
+
-+#define VXC_RAW_ICMP 0x00000100
-+#define VXC_SYSLOG 0x00001000
+
-+#define VXC_SECURE_MOUNT 0x00010000
-+#define VXC_SECURE_REMOUNT 0x00020000
-+#define VXC_BINARY_MOUNT 0x00040000
++#else /* _VX_DEVICE_H */
++#warning duplicate inclusion
++#endif /* _VX_DEVICE_H */
+--- linux-2.6.25.4/include/linux/vserver/dlimit_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/dlimit_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,74 @@
++#ifndef _VX_DLIMIT_CMD_H
++#define _VX_DLIMIT_CMD_H
+
-+#define VXC_QUOTA_CTL 0x00100000
-+#define VXC_ADMIN_MAPPER 0x00200000
-+#define VXC_ADMIN_CLOOP 0x00400000
+
++/* dlimit vserver commands */
+
-+#ifdef __KERNEL__
++#define VCMD_add_dlimit VC_CMD(DLIMIT, 1, 0)
++#define VCMD_rem_dlimit VC_CMD(DLIMIT, 2, 0)
+
-+#include <linux/list.h>
-+#include <linux/spinlock.h>
-+#include <linux/rcupdate.h>
++#define VCMD_set_dlimit VC_CMD(DLIMIT, 5, 0)
++#define VCMD_get_dlimit VC_CMD(DLIMIT, 6, 0)
+
-+#include "limit_def.h"
-+#include "sched_def.h"
-+#include "cvirt_def.h"
-+#include "cacct_def.h"
++struct vcmd_ctx_dlimit_base_v0 {
++ const char __user *name;
++ uint32_t flags;
++};
+
-+struct _vx_info_pc {
-+ struct _vx_sched_pc sched_pc;
-+ struct _vx_cvirt_pc cvirt_pc;
++struct vcmd_ctx_dlimit_v0 {
++ const char __user *name;
++ uint32_t space_used; /* used space in kbytes */
++ uint32_t space_total; /* maximum space in kbytes */
++ uint32_t inodes_used; /* used inodes */
++ uint32_t inodes_total; /* maximum inodes */
++ uint32_t reserved; /* reserved for root in % */
++ uint32_t flags;
+};
+
-+struct vx_info {
-+ struct hlist_node vx_hlist; /* linked list of contexts */
-+ xid_t vx_id; /* context id */
-+ atomic_t vx_usecnt; /* usage count */
-+ atomic_t vx_tasks; /* tasks count */
-+ struct vx_info *vx_parent; /* parent context */
-+ int vx_state; /* context state */
++#define CDLIM_UNSET ((uint32_t)0UL)
++#define CDLIM_INFINITY ((uint32_t)~0UL)
++#define CDLIM_KEEP ((uint32_t)~1UL)
+
-+ unsigned long vx_nsmask; /* assignment mask */
-+ struct nsproxy *vx_nsproxy; /* private namespace */
-+ struct fs_struct *vx_fs; /* private namespace fs */
++#ifdef __KERNEL__
+
-+ uint64_t vx_flags; /* context flags */
-+ uint64_t vx_bcaps; /* bounding caps (system) */
-+ uint64_t vx_ccaps; /* context caps (vserver) */
-+ kernel_cap_t vx_cap_bset; /* the guest's bset */
++#ifdef CONFIG_COMPAT
+
-+ struct task_struct *vx_reaper; /* guest reaper process */
-+ pid_t vx_initpid; /* PID of guest init */
++#include <asm/compat.h>
+
-+ struct _vx_limit limit; /* vserver limits */
-+ struct _vx_sched sched; /* vserver scheduler */
-+ struct _vx_cvirt cvirt; /* virtual/bias stuff */
-+ struct _vx_cacct cacct; /* context accounting */
-+
-+#ifndef CONFIG_SMP
-+ struct _vx_info_pc info_pc; /* per cpu data */
-+#else
-+ struct _vx_info_pc *ptr_pc; /* per cpu array */
-+#endif
-+
-+ wait_queue_head_t vx_wait; /* context exit waitqueue */
-+ int reboot_cmd; /* last sys_reboot() cmd */
-+ int exit_code; /* last process exit code */
-+
-+ char vx_name[65]; /* vserver name */
++struct vcmd_ctx_dlimit_base_v0_x32 {
++ compat_uptr_t name_ptr;
++ uint32_t flags;
+};
+
-+#ifndef CONFIG_SMP
-+#define vx_ptr_pc(vxi) (&(vxi)->info_pc)
-+#define vx_per_cpu(vxi, v, id) vx_ptr_pc(vxi)->v
-+#else
-+#define vx_ptr_pc(vxi) ((vxi)->ptr_pc)
-+#define vx_per_cpu(vxi, v, id) per_cpu_ptr(vx_ptr_pc(vxi), id)->v
-+#endif
-+
-+#define vx_cpu(vxi, v) vx_per_cpu(vxi, v, smp_processor_id())
-+
-+
-+struct vx_info_save {
-+ struct vx_info *vxi;
-+ xid_t xid;
++struct vcmd_ctx_dlimit_v0_x32 {
++ compat_uptr_t name_ptr;
++ uint32_t space_used; /* used space in kbytes */
++ uint32_t space_total; /* maximum space in kbytes */
++ uint32_t inodes_used; /* used inodes */
++ uint32_t inodes_total; /* maximum inodes */
++ uint32_t reserved; /* reserved for root in % */
++ uint32_t flags;
+};
+
++#endif /* CONFIG_COMPAT */
+
-+/* status flags */
-+
-+#define VXS_HASHED 0x0001
-+#define VXS_PAUSED 0x0010
-+#define VXS_SHUTDOWN 0x0100
-+#define VXS_HELPER 0x1000
-+#define VXS_RELEASED 0x8000
-+
++#include <linux/compiler.h>
+
-+extern void claim_vx_info(struct vx_info *, struct task_struct *);
-+extern void release_vx_info(struct vx_info *, struct task_struct *);
++extern int vc_add_dlimit(uint32_t, void __user *);
++extern int vc_rem_dlimit(uint32_t, void __user *);
+
-+extern struct vx_info *lookup_vx_info(int);
-+extern struct vx_info *lookup_or_create_vx_info(int);
++extern int vc_set_dlimit(uint32_t, void __user *);
++extern int vc_get_dlimit(uint32_t, void __user *);
+
-+extern int get_xid_list(int, unsigned int *, int);
-+extern int xid_is_hashed(xid_t);
++#ifdef CONFIG_COMPAT
+
-+extern int vx_migrate_task(struct task_struct *, struct vx_info *, int);
++extern int vc_add_dlimit_x32(uint32_t, void __user *);
++extern int vc_rem_dlimit_x32(uint32_t, void __user *);
+
-+extern long vs_state_change(struct vx_info *, unsigned int);
++extern int vc_set_dlimit_x32(uint32_t, void __user *);
++extern int vc_get_dlimit_x32(uint32_t, void __user *);
+
++#endif /* CONFIG_COMPAT */
+
+#endif /* __KERNEL__ */
-+#endif /* _VX_CONTEXT_H */
---- linux-2.6.22/include/linux/vserver/context_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/context_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,123 @@
-+#ifndef _VX_CONTEXT_CMD_H
-+#define _VX_CONTEXT_CMD_H
-+
++#endif /* _VX_DLIMIT_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/dlimit.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/dlimit.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,54 @@
++#ifndef _VX_DLIMIT_H
++#define _VX_DLIMIT_H
+
-+/* vinfo commands */
++#include "switch.h"
+
-+#define VCMD_task_xid VC_CMD(VINFO, 1, 0)
+
+#ifdef __KERNEL__
-+extern int vc_task_xid(uint32_t, void __user *);
+
-+#endif /* __KERNEL__ */
++/* keep in sync with CDLIM_INFINITY */
+
-+#define VCMD_vx_info VC_CMD(VINFO, 5, 0)
++#define DLIM_INFINITY (~0ULL)
+
-+struct vcmd_vx_info_v0 {
-+ uint32_t xid;
-+ uint32_t initpid;
-+ /* more to come */
-+};
++#include <linux/spinlock.h>
++#include <linux/rcupdate.h>
+
-+#ifdef __KERNEL__
-+extern int vc_vx_info(struct vx_info *, void __user *);
++struct super_block;
+
-+#endif /* __KERNEL__ */
++struct dl_info {
++ struct hlist_node dl_hlist; /* linked list of contexts */
++ struct rcu_head dl_rcu; /* the rcu head */
++ tag_t dl_tag; /* context tag */
++ atomic_t dl_usecnt; /* usage count */
++ atomic_t dl_refcnt; /* reference count */
+
-+#define VCMD_ctx_stat VC_CMD(VSTAT, 0, 0)
++ struct super_block *dl_sb; /* associated superblock */
+
-+struct vcmd_ctx_stat_v0 {
-+ uint32_t usecnt;
-+ uint32_t tasks;
-+ /* more to come */
-+};
++ spinlock_t dl_lock; /* protect the values */
+
-+#ifdef __KERNEL__
-+extern int vc_ctx_stat(struct vx_info *, void __user *);
++ unsigned long long dl_space_used; /* used space in bytes */
++ unsigned long long dl_space_total; /* maximum space in bytes */
++ unsigned long dl_inodes_used; /* used inodes */
++ unsigned long dl_inodes_total; /* maximum inodes */
+
-+#endif /* __KERNEL__ */
++ unsigned int dl_nrlmult; /* non root limit mult */
++};
+
-+/* context commands */
++struct rcu_head;
+
-+#define VCMD_ctx_create_v0 VC_CMD(VPROC, 1, 0)
-+#define VCMD_ctx_create VC_CMD(VPROC, 1, 1)
++extern void rcu_free_dl_info(struct rcu_head *);
++extern void unhash_dl_info(struct dl_info *);
+
-+struct vcmd_ctx_create {
-+ uint64_t flagword;
-+};
++extern struct dl_info *locate_dl_info(struct super_block *, tag_t);
+
-+#define VCMD_ctx_migrate_v0 VC_CMD(PROCMIG, 1, 0)
-+#define VCMD_ctx_migrate VC_CMD(PROCMIG, 1, 1)
+
-+struct vcmd_ctx_migrate {
-+ uint64_t flagword;
-+};
++struct kstatfs;
+
-+#ifdef __KERNEL__
-+extern int vc_ctx_create(uint32_t, void __user *);
-+extern int vc_ctx_migrate(struct vx_info *, void __user *);
++extern void vx_vsi_statfs(struct super_block *, struct kstatfs *);
++
++typedef uint64_t dlsize_t;
+
+#endif /* __KERNEL__ */
++#else /* _VX_DLIMIT_H */
++#warning duplicate inclusion
++#endif /* _VX_DLIMIT_H */
+--- linux-2.6.25.4/include/linux/vserver/global.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/global.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,20 @@
++#ifndef _VX_GLOBAL_H
++#define _VX_GLOBAL_H
+
+
-+/* flag commands */
++extern atomic_t vx_global_ctotal;
++extern atomic_t vx_global_cactive;
+
-+#define VCMD_get_cflags VC_CMD(FLAGS, 1, 0)
-+#define VCMD_set_cflags VC_CMD(FLAGS, 2, 0)
++extern atomic_t nx_global_ctotal;
++extern atomic_t nx_global_cactive;
+
-+struct vcmd_ctx_flags_v0 {
-+ uint64_t flagword;
-+ uint64_t mask;
-+};
++extern atomic_t vs_global_nsproxy;
++extern atomic_t vs_global_fs;
++extern atomic_t vs_global_mnt_ns;
++extern atomic_t vs_global_uts_ns;
++extern atomic_t vs_global_ipc_ns;
++extern atomic_t vs_global_user_ns;
++extern atomic_t vs_global_pid_ns;
+
-+#ifdef __KERNEL__
-+extern int vc_get_cflags(struct vx_info *, void __user *);
-+extern int vc_set_cflags(struct vx_info *, void __user *);
+
-+#endif /* __KERNEL__ */
++#endif /* _VX_GLOBAL_H */
+--- linux-2.6.25.4/include/linux/vserver/history.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/history.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,197 @@
++#ifndef _VX_HISTORY_H
++#define _VX_HISTORY_H
+
+
-+/* context caps commands */
++enum {
++ VXH_UNUSED = 0,
++ VXH_THROW_OOPS = 1,
+
-+#define VCMD_get_ccaps_v0 VC_CMD(FLAGS, 3, 0)
-+#define VCMD_set_ccaps_v0 VC_CMD(FLAGS, 4, 0)
++ VXH_GET_VX_INFO,
++ VXH_PUT_VX_INFO,
++ VXH_INIT_VX_INFO,
++ VXH_SET_VX_INFO,
++ VXH_CLR_VX_INFO,
++ VXH_CLAIM_VX_INFO,
++ VXH_RELEASE_VX_INFO,
++ VXH_ALLOC_VX_INFO,
++ VXH_DEALLOC_VX_INFO,
++ VXH_HASH_VX_INFO,
++ VXH_UNHASH_VX_INFO,
++ VXH_LOC_VX_INFO,
++ VXH_LOOKUP_VX_INFO,
++ VXH_CREATE_VX_INFO,
++};
+
-+struct vcmd_ctx_caps_v0 {
-+ uint64_t bcaps;
-+ uint64_t ccaps;
-+ uint64_t cmask;
++struct _vxhe_vxi {
++ struct vx_info *ptr;
++ unsigned xid;
++ unsigned usecnt;
++ unsigned tasks;
+};
+
-+#define VCMD_get_ccaps VC_CMD(FLAGS, 3, 1)
-+#define VCMD_set_ccaps VC_CMD(FLAGS, 4, 1)
++struct _vxhe_set_clr {
++ void *data;
++};
+
-+struct vcmd_ctx_caps_v1 {
-+ uint64_t ccaps;
-+ uint64_t cmask;
++struct _vxhe_loc_lookup {
++ unsigned arg;
+};
+
-+#ifdef __KERNEL__
-+extern int vc_get_ccaps_v0(struct vx_info *, void __user *);
-+extern int vc_set_ccaps_v0(struct vx_info *, void __user *);
-+extern int vc_get_ccaps(struct vx_info *, void __user *);
-+extern int vc_set_ccaps(struct vx_info *, void __user *);
++struct _vx_hist_entry {
++ void *loc;
++ unsigned short seq;
++ unsigned short type;
++ struct _vxhe_vxi vxi;
++ union {
++ struct _vxhe_set_clr sc;
++ struct _vxhe_loc_lookup ll;
++ };
++};
+
-+#endif /* __KERNEL__ */
++#ifdef CONFIG_VSERVER_HISTORY
+
++extern unsigned volatile int vxh_active;
+
-+/* bcaps commands */
++struct _vx_hist_entry *vxh_advance(void *loc);
+
-+#define VCMD_get_bcaps VC_CMD(FLAGS, 9, 0)
-+#define VCMD_set_bcaps VC_CMD(FLAGS, 10, 0)
+
-+struct vcmd_bcaps {
-+ uint64_t bcaps;
-+ uint64_t bmask;
-+};
++static inline
++void __vxh_copy_vxi(struct _vx_hist_entry *entry, struct vx_info *vxi)
++{
++ entry->vxi.ptr = vxi;
++ if (vxi) {
++ entry->vxi.usecnt = atomic_read(&vxi->vx_usecnt);
++ entry->vxi.tasks = atomic_read(&vxi->vx_tasks);
++ entry->vxi.xid = vxi->vx_id;
++ }
++}
+
-+#ifdef __KERNEL__
-+extern int vc_get_bcaps(struct vx_info *, void __user *);
-+extern int vc_set_bcaps(struct vx_info *, void __user *);
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_CONTEXT_CMD_H */
---- linux-2.6.22/include/linux/vserver/cvirt.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/cvirt.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,20 @@
-+#ifndef _VX_CVIRT_H
-+#define _VX_CVIRT_H
++#define __HERE__ current_text_addr()
+
++#define __VXH_BODY(__type, __data, __here) \
++ struct _vx_hist_entry *entry; \
++ \
++ preempt_disable(); \
++ entry = vxh_advance(__here); \
++ __data; \
++ entry->type = __type; \
++ preempt_enable();
+
-+#ifdef __KERNEL__
+
-+struct timespec;
++ /* pass vxi only */
+
-+void vx_vsi_uptime(struct timespec *, struct timespec *);
++#define __VXH_SMPL \
++ __vxh_copy_vxi(entry, vxi)
+
++static inline
++void __vxh_smpl(struct vx_info *vxi, int __type, void *__here)
++{
++ __VXH_BODY(__type, __VXH_SMPL, __here)
++}
+
-+struct vx_info;
++ /* pass vxi and data (void *) */
+
-+void vx_update_load(struct vx_info *);
++#define __VXH_DATA \
++ __vxh_copy_vxi(entry, vxi); \
++ entry->sc.data = data
+
++static inline
++void __vxh_data(struct vx_info *vxi, void *data,
++ int __type, void *__here)
++{
++ __VXH_BODY(__type, __VXH_DATA, __here)
++}
+
-+int vx_do_syslog(int, char __user *, int);
++ /* pass vxi and arg (long) */
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_CVIRT_H */
---- linux-2.6.22/include/linux/vserver/cvirt_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/cvirt_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,53 @@
-+#ifndef _VX_CVIRT_CMD_H
-+#define _VX_CVIRT_CMD_H
++#define __VXH_LONG \
++ __vxh_copy_vxi(entry, vxi); \
++ entry->ll.arg = arg
+
++static inline
++void __vxh_long(struct vx_info *vxi, long arg,
++ int __type, void *__here)
++{
++ __VXH_BODY(__type, __VXH_LONG, __here)
++}
+
-+/* virtual host info name commands */
+
-+#define VCMD_set_vhi_name VC_CMD(VHOST, 1, 0)
-+#define VCMD_get_vhi_name VC_CMD(VHOST, 2, 0)
++static inline
++void __vxh_throw_oops(void *__here)
++{
++ __VXH_BODY(VXH_THROW_OOPS, {}, __here);
++ /* prevent further acquisition */
++ vxh_active = 0;
++}
+
-+struct vcmd_vhi_name_v0 {
-+ uint32_t field;
-+ char name[65];
-+};
+
++#define vxh_throw_oops() __vxh_throw_oops(__HERE__);
+
-+enum vhi_name_field {
-+ VHIN_CONTEXT = 0,
-+ VHIN_SYSNAME,
-+ VHIN_NODENAME,
-+ VHIN_RELEASE,
-+ VHIN_VERSION,
-+ VHIN_MACHINE,
-+ VHIN_DOMAINNAME,
-+};
++#define __vxh_get_vx_info(v, h) __vxh_smpl(v, VXH_GET_VX_INFO, h);
++#define __vxh_put_vx_info(v, h) __vxh_smpl(v, VXH_PUT_VX_INFO, h);
+
++#define __vxh_init_vx_info(v, d, h) \
++ __vxh_data(v, d, VXH_INIT_VX_INFO, h);
++#define __vxh_set_vx_info(v, d, h) \
++ __vxh_data(v, d, VXH_SET_VX_INFO, h);
++#define __vxh_clr_vx_info(v, d, h) \
++ __vxh_data(v, d, VXH_CLR_VX_INFO, h);
+
-+#ifdef __KERNEL__
++#define __vxh_claim_vx_info(v, d, h) \
++ __vxh_data(v, d, VXH_CLAIM_VX_INFO, h);
++#define __vxh_release_vx_info(v, d, h) \
++ __vxh_data(v, d, VXH_RELEASE_VX_INFO, h);
+
-+#include <linux/compiler.h>
++#define vxh_alloc_vx_info(v) \
++ __vxh_smpl(v, VXH_ALLOC_VX_INFO, __HERE__);
++#define vxh_dealloc_vx_info(v) \
++ __vxh_smpl(v, VXH_DEALLOC_VX_INFO, __HERE__);
+
-+extern int vc_set_vhi_name(struct vx_info *, void __user *);
-+extern int vc_get_vhi_name(struct vx_info *, void __user *);
++#define vxh_hash_vx_info(v) \
++ __vxh_smpl(v, VXH_HASH_VX_INFO, __HERE__);
++#define vxh_unhash_vx_info(v) \
++ __vxh_smpl(v, VXH_UNHASH_VX_INFO, __HERE__);
+
-+#endif /* __KERNEL__ */
++#define vxh_loc_vx_info(v, l) \
++ __vxh_long(v, l, VXH_LOC_VX_INFO, __HERE__);
++#define vxh_lookup_vx_info(v, l) \
++ __vxh_long(v, l, VXH_LOOKUP_VX_INFO, __HERE__);
++#define vxh_create_vx_info(v, l) \
++ __vxh_long(v, l, VXH_CREATE_VX_INFO, __HERE__);
+
-+#define VCMD_virt_stat VC_CMD(VSTAT, 3, 0)
++extern void vxh_dump_history(void);
+
-+struct vcmd_virt_stat_v0 {
-+ uint64_t offset;
-+ uint64_t uptime;
-+ uint32_t nr_threads;
-+ uint32_t nr_running;
-+ uint32_t nr_uninterruptible;
-+ uint32_t nr_onhold;
-+ uint32_t nr_forks;
-+ uint32_t load[3];
-+};
+
-+#ifdef __KERNEL__
-+extern int vc_virt_stat(struct vx_info *, void __user *);
++#else /* CONFIG_VSERVER_HISTORY */
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_CVIRT_CMD_H */
---- linux-2.6.22/include/linux/vserver/cvirt_def.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/cvirt_def.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,81 @@
-+#ifndef _VX_CVIRT_DEF_H
-+#define _VX_CVIRT_DEF_H
++#define __HERE__ 0
+
-+#include <linux/jiffies.h>
-+#include <linux/spinlock.h>
-+#include <linux/wait.h>
-+#include <linux/time.h>
-+#include <linux/utsname.h>
-+#include <asm/atomic.h>
++#define vxh_throw_oops() do { } while (0)
+
++#define __vxh_get_vx_info(v, h) do { } while (0)
++#define __vxh_put_vx_info(v, h) do { } while (0)
+
-+struct _vx_usage_stat {
-+ uint64_t user;
-+ uint64_t nice;
-+ uint64_t system;
-+ uint64_t softirq;
-+ uint64_t irq;
-+ uint64_t idle;
-+ uint64_t iowait;
-+};
++#define __vxh_init_vx_info(v, d, h) do { } while (0)
++#define __vxh_set_vx_info(v, d, h) do { } while (0)
++#define __vxh_clr_vx_info(v, d, h) do { } while (0)
+
-+struct _vx_syslog {
-+ wait_queue_head_t log_wait;
-+ spinlock_t logbuf_lock; /* lock for the log buffer */
++#define __vxh_claim_vx_info(v, d, h) do { } while (0)
++#define __vxh_release_vx_info(v, d, h) do { } while (0)
+
-+ unsigned long log_start; /* next char to be read by syslog() */
-+ unsigned long con_start; /* next char to be sent to consoles */
-+ unsigned long log_end; /* most-recently-written-char + 1 */
-+ unsigned long logged_chars; /* #chars since last read+clear operation */
++#define vxh_alloc_vx_info(v) do { } while (0)
++#define vxh_dealloc_vx_info(v) do { } while (0)
+
-+ char log_buf[1024];
-+};
++#define vxh_hash_vx_info(v) do { } while (0)
++#define vxh_unhash_vx_info(v) do { } while (0)
+
++#define vxh_loc_vx_info(v, l) do { } while (0)
++#define vxh_lookup_vx_info(v, l) do { } while (0)
++#define vxh_create_vx_info(v, l) do { } while (0)
+
-+/* context sub struct */
++#define vxh_dump_history() do { } while (0)
+
-+struct _vx_cvirt {
-+ atomic_t nr_threads; /* number of current threads */
-+ atomic_t nr_running; /* number of running threads */
-+ atomic_t nr_uninterruptible; /* number of uninterruptible threads */
+
-+ atomic_t nr_onhold; /* processes on hold */
-+ uint32_t onhold_last; /* jiffies when put on hold */
++#endif /* CONFIG_VSERVER_HISTORY */
+
-+ struct timeval bias_tv; /* time offset to the host */
-+ struct timespec bias_idle;
-+ struct timespec bias_uptime; /* context creation point */
-+ uint64_t bias_clock; /* offset in clock_t */
++#endif /* _VX_HISTORY_H */
+--- linux-2.6.25.4/include/linux/vserver/inode_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/inode_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,59 @@
++#ifndef _VX_INODE_CMD_H
++#define _VX_INODE_CMD_H
+
-+ spinlock_t load_lock; /* lock for the load averages */
-+ atomic_t load_updates; /* nr of load updates done so far */
-+ uint32_t load_last; /* last time load was calculated */
-+ uint32_t load[3]; /* load averages 1,5,15 */
+
-+ atomic_t total_forks; /* number of forks so far */
++/* inode vserver commands */
+
-+ struct _vx_syslog syslog;
++#define VCMD_get_iattr VC_CMD(INODE, 1, 1)
++#define VCMD_set_iattr VC_CMD(INODE, 2, 1)
++
++#define VCMD_fget_iattr VC_CMD(INODE, 3, 0)
++#define VCMD_fset_iattr VC_CMD(INODE, 4, 0)
++
++struct vcmd_ctx_iattr_v1 {
++ const char __user *name;
++ uint32_t tag;
++ uint32_t flags;
++ uint32_t mask;
+};
+
-+struct _vx_cvirt_pc {
-+ struct _vx_usage_stat cpustat;
++struct vcmd_ctx_fiattr_v0 {
++ uint32_t tag;
++ uint32_t flags;
++ uint32_t mask;
+};
+
+
-+#ifdef CONFIG_VSERVER_DEBUG
++#ifdef __KERNEL__
+
-+static inline void __dump_vx_cvirt(struct _vx_cvirt *cvirt)
-+{
-+ printk("\t_vx_cvirt:\n");
-+ printk("\t threads: %4d, %4d, %4d, %4d\n",
-+ atomic_read(&cvirt->nr_threads),
-+ atomic_read(&cvirt->nr_running),
-+ atomic_read(&cvirt->nr_uninterruptible),
-+ atomic_read(&cvirt->nr_onhold));
-+ /* add rest here */
-+ printk("\t total_forks = %d\n", atomic_read(&cvirt->total_forks));
-+}
+
-+#endif
++#ifdef CONFIG_COMPAT
+
-+#endif /* _VX_CVIRT_DEF_H */
---- linux-2.6.22/include/linux/vserver/debug.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/debug.h 2007-06-16 05:47:03 +0200
-@@ -0,0 +1,113 @@
-+#ifndef _VX_DEBUG_H
-+#define _VX_DEBUG_H
++#include <asm/compat.h>
+
++struct vcmd_ctx_iattr_v1_x32 {
++ compat_uptr_t name_ptr;
++ uint32_t tag;
++ uint32_t flags;
++ uint32_t mask;
++};
+
-+#define VXD_CBIT(n, m) (vx_debug_ ## n & (1 << (m)))
-+#define VXD_CMIN(n, m) (vx_debug_ ## n > (m))
-+#define VXD_MASK(n, m) (vx_debug_ ## n & (m))
++#endif /* CONFIG_COMPAT */
+
-+#define VXD_DEV(d) (d), (d)->bd_inode->i_ino, \
-+ imajor((d)->bd_inode), iminor((d)->bd_inode)
-+#define VXF_DEV "%p[%lu,%d:%d]"
++#include <linux/compiler.h>
+
++extern int vc_get_iattr(void __user *);
++extern int vc_set_iattr(void __user *);
+
-+#define __FUNC__ __func__
++extern int vc_fget_iattr(uint32_t, void __user *);
++extern int vc_fset_iattr(uint32_t, void __user *);
+
++#ifdef CONFIG_COMPAT
+
-+#define vxd_path(d, m) \
-+ ({ static char _buffer[PATH_MAX]; \
-+ d_path(d, m, _buffer, sizeof(_buffer)); })
++extern int vc_get_iattr_x32(void __user *);
++extern int vc_set_iattr_x32(void __user *);
+
-+#define vxd_cond_path(n) \
-+ ((n) ? vxd_path((n)->dentry, (n)->mnt) : "<null>" )
++#endif /* CONFIG_COMPAT */
+
++#endif /* __KERNEL__ */
++#endif /* _VX_INODE_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/inode.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/inode.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,38 @@
++#ifndef _VX_INODE_H
++#define _VX_INODE_H
+
-+#ifdef CONFIG_VSERVER_DEBUG
+
-+extern unsigned int vx_debug_switch;
-+extern unsigned int vx_debug_xid;
-+extern unsigned int vx_debug_nid;
-+extern unsigned int vx_debug_tag;
-+extern unsigned int vx_debug_net;
-+extern unsigned int vx_debug_limit;
-+extern unsigned int vx_debug_cres;
-+extern unsigned int vx_debug_dlim;
-+extern unsigned int vx_debug_quota;
-+extern unsigned int vx_debug_cvirt;
-+extern unsigned int vx_debug_space;
-+extern unsigned int vx_debug_misc;
++#define IATTR_TAG 0x01000000
+
++#define IATTR_ADMIN 0x00000001
++#define IATTR_WATCH 0x00000002
++#define IATTR_HIDE 0x00000004
++#define IATTR_FLAGS 0x00000007
+
-+#define VX_LOGLEVEL "vxD: "
++#define IATTR_BARRIER 0x00010000
++#define IATTR_IUNLINK 0x00020000
++#define IATTR_IMMUTABLE 0x00040000
+
-+#define vxdprintk(c, f, x...) \
-+ do { \
-+ if (c) \
-+ printk(VX_LOGLEVEL f "\n" , ##x); \
-+ } while (0)
++#ifdef __KERNEL__
+
-+#define vxlprintk(c, f, x...) \
-+ do { \
-+ if (c) \
-+ printk(VX_LOGLEVEL f " @%s:%d\n", x); \
-+ } while (0)
-+
-+#define vxfprintk(c, f, x...) \
-+ do { \
-+ if (c) \
-+ printk(VX_LOGLEVEL f " %s@%s:%d\n", x); \
-+ } while (0)
+
++#ifdef CONFIG_VSERVER_PROC_SECURE
++#define IATTR_PROC_DEFAULT ( IATTR_ADMIN | IATTR_HIDE )
++#define IATTR_PROC_SYMLINK ( IATTR_ADMIN )
++#else
++#define IATTR_PROC_DEFAULT ( IATTR_ADMIN )
++#define IATTR_PROC_SYMLINK ( IATTR_ADMIN )
++#endif
+
-+struct vx_info;
++#define vx_hide_check(c, m) (((m) & IATTR_HIDE) ? vx_check(c, m) : 1)
+
-+void dump_vx_info(struct vx_info *, int);
-+void dump_vx_info_inactive(int);
++#endif /* __KERNEL__ */
+
-+#else /* CONFIG_VSERVER_DEBUG */
++/* inode ioctls */
+
-+#define vx_debug_switch 0
-+#define vx_debug_xid 0
-+#define vx_debug_nid 0
-+#define vx_debug_tag 0
-+#define vx_debug_net 0
-+#define vx_debug_limit 0
-+#define vx_debug_cres 0
-+#define vx_debug_dlim 0
-+#define vx_debug_cvirt 0
++#define FIOC_GETXFLG _IOR('x', 5, long)
++#define FIOC_SETXFLG _IOW('x', 6, long)
+
-+#define vxdprintk(x...) do { } while (0)
-+#define vxlprintk(x...) do { } while (0)
-+#define vxfprintk(x...) do { } while (0)
++#else /* _VX_INODE_H */
++#warning duplicate inclusion
++#endif /* _VX_INODE_H */
+--- linux-2.6.25.4/include/linux/vserver/Kbuild 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/Kbuild 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,8 @@
+
-+#endif /* CONFIG_VSERVER_DEBUG */
++unifdef-y += context_cmd.h network_cmd.h space_cmd.h \
++ cacct_cmd.h cvirt_cmd.h limit_cmd.h dlimit_cmd.h \
++ inode_cmd.h tag_cmd.h sched_cmd.h signal_cmd.h \
++ debug_cmd.h device_cmd.h
+
++unifdef-y += switch.h network.h monitor.h inode.h device.h
+
-+#ifdef CONFIG_VSERVER_WARN
+--- linux-2.6.25.4/include/linux/vserver/limit_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/limit_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,69 @@
++#ifndef _VX_LIMIT_CMD_H
++#define _VX_LIMIT_CMD_H
+
-+#define VX_WARNLEVEL KERN_WARNING "vxW: "
+
-+#define vxwprintk(c, f, x...) \
-+ do { \
-+ if (c) \
-+ printk(VX_WARNLEVEL f "\n" , ##x); \
-+ } while (0)
++/* rlimit vserver commands */
+
-+#else /* CONFIG_VSERVER_WARN */
++#define VCMD_get_rlimit VC_CMD(RLIMIT, 1, 0)
++#define VCMD_set_rlimit VC_CMD(RLIMIT, 2, 0)
++#define VCMD_get_rlimit_mask VC_CMD(RLIMIT, 3, 0)
++#define VCMD_reset_minmax VC_CMD(RLIMIT, 9, 0)
+
-+#define vxwprintk(x...) do { } while (0)
++struct vcmd_ctx_rlimit_v0 {
++ uint32_t id;
++ uint64_t minimum;
++ uint64_t softlimit;
++ uint64_t maximum;
++};
+
-+#endif /* CONFIG_VSERVER_WARN */
++struct vcmd_ctx_rlimit_mask_v0 {
++ uint32_t minimum;
++ uint32_t softlimit;
++ uint32_t maximum;
++};
+
++#define VCMD_rlimit_stat VC_CMD(VSTAT, 1, 0)
+
-+#ifdef CONFIG_VSERVER_DEBUG
-+#define vxd_assert_lock(l) assert_spin_locked(l)
-+#define vxd_assert(c, f, x...) vxlprintk(!(c), \
-+ "assertion [" f "] failed.", ##x, __FILE__, __LINE__)
-+#else
-+#define vxd_assert_lock(l) do { } while (0)
-+#define vxd_assert(c, f, x...) do { } while (0)
-+#endif
++struct vcmd_rlimit_stat_v0 {
++ uint32_t id;
++ uint32_t hits;
++ uint64_t value;
++ uint64_t minimum;
++ uint64_t maximum;
++};
+
++#define CRLIM_UNSET (0ULL)
++#define CRLIM_INFINITY (~0ULL)
++#define CRLIM_KEEP (~1ULL)
+
-+#endif /* _VX_DEBUG_H */
---- linux-2.6.22/include/linux/vserver/debug_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/debug_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,58 @@
-+#ifndef _VX_DEBUG_CMD_H
-+#define _VX_DEBUG_CMD_H
++#ifdef __KERNEL__
+
++#ifdef CONFIG_IA32_EMULATION
+
-+/* debug commands */
++struct vcmd_ctx_rlimit_v0_x32 {
++ uint32_t id;
++ uint64_t minimum;
++ uint64_t softlimit;
++ uint64_t maximum;
++} __attribute__ ((aligned (4)));
+
-+#define VCMD_dump_history VC_CMD(DEBUG, 1, 0)
++#endif /* CONFIG_IA32_EMULATION */
+
-+#define VCMD_read_history VC_CMD(DEBUG, 5, 0)
-+#define VCMD_read_monitor VC_CMD(DEBUG, 6, 0)
++#include <linux/compiler.h>
+
-+struct vcmd_read_history_v0 {
-+ uint32_t index;
-+ uint32_t count;
-+ char __user *data;
-+};
++extern int vc_get_rlimit_mask(uint32_t, void __user *);
++extern int vc_get_rlimit(struct vx_info *, void __user *);
++extern int vc_set_rlimit(struct vx_info *, void __user *);
++extern int vc_reset_minmax(struct vx_info *, void __user *);
+
-+struct vcmd_read_monitor_v0 {
-+ uint32_t index;
-+ uint32_t count;
-+ char __user *data;
-+};
++extern int vc_rlimit_stat(struct vx_info *, void __user *);
+
++#ifdef CONFIG_IA32_EMULATION
+
-+#ifdef __KERNEL__
++extern int vc_get_rlimit_x32(struct vx_info *, void __user *);
++extern int vc_set_rlimit_x32(struct vx_info *, void __user *);
+
-+#ifdef CONFIG_COMPAT
++#endif /* CONFIG_IA32_EMULATION */
+
-+#include <asm/compat.h>
++#endif /* __KERNEL__ */
++#endif /* _VX_LIMIT_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/limit_def.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/limit_def.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,47 @@
++#ifndef _VX_LIMIT_DEF_H
++#define _VX_LIMIT_DEF_H
+
-+struct vcmd_read_history_v0_x32 {
-+ uint32_t index;
-+ uint32_t count;
-+ compat_uptr_t data_ptr;
-+};
++#include <asm/atomic.h>
++#include <asm/resource.h>
+
-+struct vcmd_read_monitor_v0_x32 {
-+ uint32_t index;
-+ uint32_t count;
-+ compat_uptr_t data_ptr;
-+};
++#include "limit.h"
+
-+#endif /* CONFIG_COMPAT */
+
-+extern int vc_dump_history(uint32_t);
++struct _vx_res_limit {
++ rlim_t soft; /* Context soft limit */
++ rlim_t hard; /* Context hard limit */
+
-+extern int vc_read_history(uint32_t, void __user *);
-+extern int vc_read_monitor(uint32_t, void __user *);
++ rlim_atomic_t rcur; /* Current value */
++ rlim_t rmin; /* Context minimum */
++ rlim_t rmax; /* Context maximum */
+
-+#ifdef CONFIG_COMPAT
++ atomic_t lhit; /* Limit hits */
++};
+
-+extern int vc_read_history_x32(uint32_t, void __user *);
-+extern int vc_read_monitor_x32(uint32_t, void __user *);
++/* context sub struct */
+
-+#endif /* CONFIG_COMPAT */
++struct _vx_limit {
++ struct _vx_res_limit res[NUM_LIMITS];
++};
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_DEBUG_CMD_H */
---- linux-2.6.22/include/linux/vserver/dlimit.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/dlimit.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,53 @@
-+#ifndef _VX_DLIMIT_H
-+#define _VX_DLIMIT_H
++#ifdef CONFIG_VSERVER_DEBUG
+
-+#include "switch.h"
++static inline void __dump_vx_limit(struct _vx_limit *limit)
++{
++ int i;
+
++ printk("\t_vx_limit:");
++ for (i = 0; i < NUM_LIMITS; i++) {
++ printk("\t [%2d] = %8lu %8lu/%8lu, %8ld/%8ld, %8d\n",
++ i, (unsigned long)__rlim_get(limit, i),
++ (unsigned long)__rlim_rmin(limit, i),
++ (unsigned long)__rlim_rmax(limit, i),
++ (long)__rlim_soft(limit, i),
++ (long)__rlim_hard(limit, i),
++ atomic_read(&__rlim_lhit(limit, i)));
++ }
++}
+
-+#ifdef __KERNEL__
++#endif
+
-+/* keep in sync with CDLIM_INFINITY */
++#endif /* _VX_LIMIT_DEF_H */
+--- linux-2.6.25.4/include/linux/vserver/limit.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/limit.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,70 @@
++#ifndef _VX_LIMIT_H
++#define _VX_LIMIT_H
+
-+#define DLIM_INFINITY (~0ULL)
++#define VLIMIT_NSOCK 16
++#define VLIMIT_OPENFD 17
++#define VLIMIT_ANON 18
++#define VLIMIT_SHMEM 19
++#define VLIMIT_SEMARY 20
++#define VLIMIT_NSEMS 21
++#define VLIMIT_DENTRY 22
++#define VLIMIT_MAPPED 23
+
-+#include <linux/spinlock.h>
+
-+struct super_block;
++#ifdef __KERNEL__
+
-+struct dl_info {
-+ struct hlist_node dl_hlist; /* linked list of contexts */
-+ struct rcu_head dl_rcu; /* the rcu head */
-+ tag_t dl_tag; /* context tag */
-+ atomic_t dl_usecnt; /* usage count */
-+ atomic_t dl_refcnt; /* reference count */
++#define VLIM_NOCHECK ((1L << VLIMIT_DENTRY) | (1L << RLIMIT_RSS))
+
-+ struct super_block *dl_sb; /* associated superblock */
++/* keep in sync with CRLIM_INFINITY */
+
-+ spinlock_t dl_lock; /* protect the values */
++#define VLIM_INFINITY (~0ULL)
+
-+ unsigned long long dl_space_used; /* used space in bytes */
-+ unsigned long long dl_space_total; /* maximum space in bytes */
-+ unsigned long dl_inodes_used; /* used inodes */
-+ unsigned long dl_inodes_total; /* maximum inodes */
++#include <asm/atomic.h>
++#include <asm/resource.h>
+
-+ unsigned int dl_nrlmult; /* non root limit mult */
-+};
++#ifndef RLIM_INFINITY
++#warning RLIM_INFINITY is undefined
++#endif
+
-+struct rcu_head;
++#define __rlim_val(l, r, v) ((l)->res[r].v)
+
-+extern void rcu_free_dl_info(struct rcu_head *);
-+extern void unhash_dl_info(struct dl_info *);
++#define __rlim_soft(l, r) __rlim_val(l, r, soft)
++#define __rlim_hard(l, r) __rlim_val(l, r, hard)
+
-+extern struct dl_info *locate_dl_info(struct super_block *, tag_t);
++#define __rlim_rcur(l, r) __rlim_val(l, r, rcur)
++#define __rlim_rmin(l, r) __rlim_val(l, r, rmin)
++#define __rlim_rmax(l, r) __rlim_val(l, r, rmax)
+
++#define __rlim_lhit(l, r) __rlim_val(l, r, lhit)
++#define __rlim_hit(l, r) atomic_inc(&__rlim_lhit(l, r))
+
-+struct kstatfs;
++typedef atomic_long_t rlim_atomic_t;
++typedef unsigned long rlim_t;
+
-+extern void vx_vsi_statfs(struct super_block *, struct kstatfs *);
++#define __rlim_get(l, r) atomic_long_read(&__rlim_rcur(l, r))
++#define __rlim_set(l, r, v) atomic_long_set(&__rlim_rcur(l, r), v)
++#define __rlim_inc(l, r) atomic_long_inc(&__rlim_rcur(l, r))
++#define __rlim_dec(l, r) atomic_long_dec(&__rlim_rcur(l, r))
++#define __rlim_add(l, r, v) atomic_long_add(v, &__rlim_rcur(l, r))
++#define __rlim_sub(l, r, v) atomic_long_sub(v, &__rlim_rcur(l, r))
+
-+typedef uint64_t dlsize_t;
+
-+#endif /* __KERNEL__ */
-+#else /* _VX_DLIMIT_H */
-+#warning duplicate inclusion
-+#endif /* _VX_DLIMIT_H */
---- linux-2.6.22/include/linux/vserver/dlimit_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/dlimit_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,74 @@
-+#ifndef _VX_DLIMIT_CMD_H
-+#define _VX_DLIMIT_CMD_H
-+
-+
-+/* dlimit vserver commands */
++#if (RLIM_INFINITY == VLIM_INFINITY)
++#define VX_VLIM(r) ((long long)(long)(r))
++#define VX_RLIM(v) ((rlim_t)(v))
++#else
++#define VX_VLIM(r) (((r) == RLIM_INFINITY) \
++ ? VLIM_INFINITY : (long long)(r))
++#define VX_RLIM(v) (((v) == VLIM_INFINITY) \
++ ? RLIM_INFINITY : (rlim_t)(v))
++#endif
+
-+#define VCMD_add_dlimit VC_CMD(DLIMIT, 1, 0)
-+#define VCMD_rem_dlimit VC_CMD(DLIMIT, 2, 0)
++struct sysinfo;
+
-+#define VCMD_set_dlimit VC_CMD(DLIMIT, 5, 0)
-+#define VCMD_get_dlimit VC_CMD(DLIMIT, 6, 0)
++void vx_vsi_meminfo(struct sysinfo *);
++void vx_vsi_swapinfo(struct sysinfo *);
+
-+struct vcmd_ctx_dlimit_base_v0 {
-+ const char __user *name;
-+ uint32_t flags;
-+};
++#define NUM_LIMITS 24
+
-+struct vcmd_ctx_dlimit_v0 {
-+ const char __user *name;
-+ uint32_t space_used; /* used space in kbytes */
-+ uint32_t space_total; /* maximum space in kbytes */
-+ uint32_t inodes_used; /* used inodes */
-+ uint32_t inodes_total; /* maximum inodes */
-+ uint32_t reserved; /* reserved for root in % */
-+ uint32_t flags;
-+};
++#endif /* __KERNEL__ */
++#endif /* _VX_LIMIT_H */
+--- linux-2.6.25.4/include/linux/vserver/limit_int.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/limit_int.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,198 @@
++#ifndef _VX_LIMIT_INT_H
++#define _VX_LIMIT_INT_H
+
-+#define CDLIM_UNSET ((uint32_t)0UL)
-+#define CDLIM_INFINITY ((uint32_t)~0UL)
-+#define CDLIM_KEEP ((uint32_t)~1UL)
++#include "context.h"
+
+#ifdef __KERNEL__
+
-+#ifdef CONFIG_COMPAT
-+
-+#include <asm/compat.h>
-+
-+struct vcmd_ctx_dlimit_base_v0_x32 {
-+ compat_uptr_t name_ptr;
-+ uint32_t flags;
-+};
-+
-+struct vcmd_ctx_dlimit_v0_x32 {
-+ compat_uptr_t name_ptr;
-+ uint32_t space_used; /* used space in kbytes */
-+ uint32_t space_total; /* maximum space in kbytes */
-+ uint32_t inodes_used; /* used inodes */
-+ uint32_t inodes_total; /* maximum inodes */
-+ uint32_t reserved; /* reserved for root in % */
-+ uint32_t flags;
-+};
-+
-+#endif /* CONFIG_COMPAT */
-+
-+#include <linux/compiler.h>
-+
-+extern int vc_add_dlimit(uint32_t, void __user *);
-+extern int vc_rem_dlimit(uint32_t, void __user *);
-+
-+extern int vc_set_dlimit(uint32_t, void __user *);
-+extern int vc_get_dlimit(uint32_t, void __user *);
-+
-+#ifdef CONFIG_COMPAT
++#define VXD_RCRES_COND(r) VXD_CBIT(cres, r)
++#define VXD_RLIMIT_COND(r) VXD_CBIT(limit, r)
+
-+extern int vc_add_dlimit_x32(uint32_t, void __user *);
-+extern int vc_rem_dlimit_x32(uint32_t, void __user *);
++extern const char *vlimit_name[NUM_LIMITS];
+
-+extern int vc_set_dlimit_x32(uint32_t, void __user *);
-+extern int vc_get_dlimit_x32(uint32_t, void __user *);
++static inline void __vx_acc_cres(struct vx_info *vxi,
++ int res, int dir, void *_data, char *_file, int _line)
++{
++ if (VXD_RCRES_COND(res))
++ vxlprintk(1, "vx_acc_cres[%5d,%s,%2d]: %5ld%s (%p)",
++ (vxi ? vxi->vx_id : -1), vlimit_name[res], res,
++ (vxi ? (long)__rlim_get(&vxi->limit, res) : 0),
++ (dir > 0) ? "++" : "--", _data, _file, _line);
++ if (!vxi)
++ return;
+
-+#endif /* CONFIG_COMPAT */
++ if (dir > 0)
++ __rlim_inc(&vxi->limit, res);
++ else
++ __rlim_dec(&vxi->limit, res);
++}
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_DLIMIT_CMD_H */
---- linux-2.6.22/include/linux/vserver/global.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/global.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,18 @@
-+#ifndef _VX_GLOBAL_H
-+#define _VX_GLOBAL_H
++static inline void __vx_add_cres(struct vx_info *vxi,
++ int res, int amount, void *_data, char *_file, int _line)
++{
++ if (VXD_RCRES_COND(res))
++ vxlprintk(1, "vx_add_cres[%5d,%s,%2d]: %5ld += %5d (%p)",
++ (vxi ? vxi->vx_id : -1), vlimit_name[res], res,
++ (vxi ? (long)__rlim_get(&vxi->limit, res) : 0),
++ amount, _data, _file, _line);
++ if (amount == 0)
++ return;
++ if (!vxi)
++ return;
++ __rlim_add(&vxi->limit, res, amount);
++}
+
++static inline
++int __vx_cres_adjust_max(struct _vx_limit *limit, int res, rlim_t value)
++{
++ int cond = (value > __rlim_rmax(limit, res));
+
-+extern atomic_t vx_global_ctotal;
-+extern atomic_t vx_global_cactive;
++ if (cond)
++ __rlim_rmax(limit, res) = value;
++ return cond;
++}
+
-+extern atomic_t nx_global_ctotal;
-+extern atomic_t nx_global_cactive;
++static inline
++int __vx_cres_adjust_min(struct _vx_limit *limit, int res, rlim_t value)
++{
++ int cond = (value < __rlim_rmin(limit, res));
+
-+extern atomic_t vs_global_nsproxy;
-+extern atomic_t vs_global_fs;
-+extern atomic_t vs_global_mnt_ns;
-+extern atomic_t vs_global_uts_ns;
-+extern atomic_t vs_global_ipc_ns;
++ if (cond)
++ __rlim_rmin(limit, res) = value;
++ return cond;
++}
+
++static inline
++void __vx_cres_fixup(struct _vx_limit *limit, int res, rlim_t value)
++{
++ if (!__vx_cres_adjust_max(limit, res, value))
++ __vx_cres_adjust_min(limit, res, value);
++}
+
-+#endif /* _VX_GLOBAL_H */
---- linux-2.6.22/include/linux/vserver/history.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/history.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,197 @@
-+#ifndef _VX_HISTORY_H
-+#define _VX_HISTORY_H
+
++/* return values:
++ +1 ... no limit hit
++ -1 ... over soft limit
++ 0 ... over hard limit */
+
-+enum {
-+ VXH_UNUSED = 0,
-+ VXH_THROW_OOPS = 1,
++static inline int __vx_cres_avail(struct vx_info *vxi,
++ int res, int num, char *_file, int _line)
++{
++ struct _vx_limit *limit;
++ rlim_t value;
+
-+ VXH_GET_VX_INFO,
-+ VXH_PUT_VX_INFO,
-+ VXH_INIT_VX_INFO,
-+ VXH_SET_VX_INFO,
-+ VXH_CLR_VX_INFO,
-+ VXH_CLAIM_VX_INFO,
-+ VXH_RELEASE_VX_INFO,
-+ VXH_ALLOC_VX_INFO,
-+ VXH_DEALLOC_VX_INFO,
-+ VXH_HASH_VX_INFO,
-+ VXH_UNHASH_VX_INFO,
-+ VXH_LOC_VX_INFO,
-+ VXH_LOOKUP_VX_INFO,
-+ VXH_CREATE_VX_INFO,
-+};
++ if (VXD_RLIMIT_COND(res))
++ vxlprintk(1, "vx_cres_avail[%5d,%s,%2d]: %5ld/%5ld > %5ld + %5d",
++ (vxi ? vxi->vx_id : -1), vlimit_name[res], res,
++ (vxi ? (long)__rlim_soft(&vxi->limit, res) : -1),
++ (vxi ? (long)__rlim_hard(&vxi->limit, res) : -1),
++ (vxi ? (long)__rlim_get(&vxi->limit, res) : 0),
++ num, _file, _line);
++ if (!vxi)
++ return 1;
+
-+struct _vxhe_vxi {
-+ struct vx_info *ptr;
-+ unsigned xid;
-+ unsigned usecnt;
-+ unsigned tasks;
-+};
++ limit = &vxi->limit;
++ value = __rlim_get(limit, res);
+
-+struct _vxhe_set_clr {
-+ void *data;
-+};
++ if (!__vx_cres_adjust_max(limit, res, value))
++ __vx_cres_adjust_min(limit, res, value);
+
-+struct _vxhe_loc_lookup {
-+ unsigned arg;
-+};
++ if (num == 0)
++ return 1;
+
-+struct _vx_hist_entry {
-+ void *loc;
-+ unsigned short seq;
-+ unsigned short type;
-+ struct _vxhe_vxi vxi;
-+ union {
-+ struct _vxhe_set_clr sc;
-+ struct _vxhe_loc_lookup ll;
-+ };
-+};
++ if (__rlim_soft(limit, res) == RLIM_INFINITY)
++ return -1;
++ if (value + num <= __rlim_soft(limit, res))
++ return -1;
+
-+#ifdef CONFIG_VSERVER_HISTORY
++ if (__rlim_hard(limit, res) == RLIM_INFINITY)
++ return 1;
++ if (value + num <= __rlim_hard(limit, res))
++ return 1;
+
-+extern unsigned volatile int vxh_active;
++ __rlim_hit(limit, res);
++ return 0;
++}
+
-+struct _vx_hist_entry *vxh_advance(void *loc);
+
++static const int VLA_RSS[] = { RLIMIT_RSS, VLIMIT_ANON, VLIMIT_MAPPED, 0 };
+
+static inline
-+void __vxh_copy_vxi(struct _vx_hist_entry *entry, struct vx_info *vxi)
++rlim_t __vx_cres_array_sum(struct _vx_limit *limit, const int *array)
+{
-+ entry->vxi.ptr = vxi;
-+ if (vxi) {
-+ entry->vxi.usecnt = atomic_read(&vxi->vx_usecnt);
-+ entry->vxi.tasks = atomic_read(&vxi->vx_tasks);
-+ entry->vxi.xid = vxi->vx_id;
++ rlim_t value, sum = 0;
++ int res;
++
++ while ((res = *array++)) {
++ value = __rlim_get(limit, res);
++ __vx_cres_fixup(limit, res, value);
++ sum += value;
+ }
++ return sum;
+}
+
++static inline
++rlim_t __vx_cres_array_fixup(struct _vx_limit *limit, const int *array)
++{
++ rlim_t value = __vx_cres_array_sum(limit, array + 1);
++ int res = *array;
+
-+#define __HERE__ current_text_addr()
++ if (value == __rlim_get(limit, res))
++ return value;
+
-+#define __VXH_BODY(__type, __data, __here) \
-+ struct _vx_hist_entry *entry; \
-+ \
-+ preempt_disable(); \
-+ entry = vxh_advance(__here); \
-+ __data; \
-+ entry->type = __type; \
-+ preempt_enable();
-+
-+
-+ /* pass vxi only */
-+
-+#define __VXH_SMPL \
-+ __vxh_copy_vxi(entry, vxi)
++ __rlim_set(limit, res, value);
++ /* now adjust min/max */
++ if (!__vx_cres_adjust_max(limit, res, value))
++ __vx_cres_adjust_min(limit, res, value);
+
-+static inline
-+void __vxh_smpl(struct vx_info *vxi, int __type, void *__here)
-+{
-+ __VXH_BODY(__type, __VXH_SMPL, __here)
++ return value;
+}
+
-+ /* pass vxi and data (void *) */
-+
-+#define __VXH_DATA \
-+ __vxh_copy_vxi(entry, vxi); \
-+ entry->sc.data = data
-+
-+static inline
-+void __vxh_data(struct vx_info *vxi, void *data,
-+ int __type, void *__here)
++static inline int __vx_cres_array_avail(struct vx_info *vxi,
++ const int *array, int num, char *_file, int _line)
+{
-+ __VXH_BODY(__type, __VXH_DATA, __here)
-+}
++ struct _vx_limit *limit;
++ rlim_t value = 0;
++ int res;
+
-+ /* pass vxi and arg (long) */
++ if (num == 0)
++ return 1;
++ if (!vxi)
++ return 1;
+
-+#define __VXH_LONG \
-+ __vxh_copy_vxi(entry, vxi); \
-+ entry->ll.arg = arg
++ limit = &vxi->limit;
++ res = *array;
++ value = __vx_cres_array_sum(limit, array + 1);
+
-+static inline
-+void __vxh_long(struct vx_info *vxi, long arg,
-+ int __type, void *__here)
-+{
-+ __VXH_BODY(__type, __VXH_LONG, __here)
++ __rlim_set(limit, res, value);
++ __vx_cres_fixup(limit, res, value);
++
++ return __vx_cres_avail(vxi, res, num, _file, _line);
+}
+
+
-+static inline
-+void __vxh_throw_oops(void *__here)
++static inline void vx_limit_fixup(struct _vx_limit *limit, int id)
+{
-+ __VXH_BODY(VXH_THROW_OOPS, {}, __here);
-+ /* prevent further acquisition */
-+ vxh_active = 0;
-+}
++ rlim_t value;
++ int res;
+
++ /* complex resources first */
++ if ((id < 0) || (id == RLIMIT_RSS))
++ __vx_cres_array_fixup(limit, VLA_RSS);
+
-+#define vxh_throw_oops() __vxh_throw_oops(__HERE__);
++ for (res = 0; res < NUM_LIMITS; res++) {
++ if ((id > 0) && (res != id))
++ continue;
+
-+#define __vxh_get_vx_info(v, h) __vxh_smpl(v, VXH_GET_VX_INFO, h);
-+#define __vxh_put_vx_info(v, h) __vxh_smpl(v, VXH_PUT_VX_INFO, h);
++ value = __rlim_get(limit, res);
++ __vx_cres_fixup(limit, res, value);
+
-+#define __vxh_init_vx_info(v, d, h) \
-+ __vxh_data(v, d, VXH_INIT_VX_INFO, h);
-+#define __vxh_set_vx_info(v, d, h) \
-+ __vxh_data(v, d, VXH_SET_VX_INFO, h);
-+#define __vxh_clr_vx_info(v, d, h) \
-+ __vxh_data(v, d, VXH_CLR_VX_INFO, h);
++ /* not supposed to happen, maybe warn? */
++ if (__rlim_rmax(limit, res) > __rlim_hard(limit, res))
++ __rlim_rmax(limit, res) = __rlim_hard(limit, res);
++ }
++}
+
-+#define __vxh_claim_vx_info(v, d, h) \
-+ __vxh_data(v, d, VXH_CLAIM_VX_INFO, h);
-+#define __vxh_release_vx_info(v, d, h) \
-+ __vxh_data(v, d, VXH_RELEASE_VX_INFO, h);
+
-+#define vxh_alloc_vx_info(v) \
-+ __vxh_smpl(v, VXH_ALLOC_VX_INFO, __HERE__);
-+#define vxh_dealloc_vx_info(v) \
-+ __vxh_smpl(v, VXH_DEALLOC_VX_INFO, __HERE__);
++#endif /* __KERNEL__ */
++#endif /* _VX_LIMIT_INT_H */
+--- linux-2.6.25.4/include/linux/vserver/monitor.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/monitor.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,96 @@
++#ifndef _VX_MONITOR_H
++#define _VX_MONITOR_H
+
-+#define vxh_hash_vx_info(v) \
-+ __vxh_smpl(v, VXH_HASH_VX_INFO, __HERE__);
-+#define vxh_unhash_vx_info(v) \
-+ __vxh_smpl(v, VXH_UNHASH_VX_INFO, __HERE__);
++#include <linux/types.h>
+
-+#define vxh_loc_vx_info(v, l) \
-+ __vxh_long(v, l, VXH_LOC_VX_INFO, __HERE__);
-+#define vxh_lookup_vx_info(v, l) \
-+ __vxh_long(v, l, VXH_LOOKUP_VX_INFO, __HERE__);
-+#define vxh_create_vx_info(v, l) \
-+ __vxh_long(v, l, VXH_CREATE_VX_INFO, __HERE__);
++enum {
++ VXM_UNUSED = 0,
+
-+extern void vxh_dump_history(void);
++ VXM_SYNC = 0x10,
+
++ VXM_UPDATE = 0x20,
++ VXM_UPDATE_1,
++ VXM_UPDATE_2,
+
-+#else /* CONFIG_VSERVER_HISTORY */
++ VXM_RQINFO_1 = 0x24,
++ VXM_RQINFO_2,
+
-+#define __HERE__ 0
++ VXM_ACTIVATE = 0x40,
++ VXM_DEACTIVATE,
++ VXM_IDLE,
+
-+#define vxh_throw_oops() do { } while (0)
++ VXM_HOLD = 0x44,
++ VXM_UNHOLD,
+
-+#define __vxh_get_vx_info(v, h) do { } while (0)
-+#define __vxh_put_vx_info(v, h) do { } while (0)
++ VXM_MIGRATE = 0x48,
++ VXM_RESCHED,
+
-+#define __vxh_init_vx_info(v, d, h) do { } while (0)
-+#define __vxh_set_vx_info(v, d, h) do { } while (0)
-+#define __vxh_clr_vx_info(v, d, h) do { } while (0)
++ /* all other bits are flags */
++ VXM_SCHED = 0x80,
++};
+
-+#define __vxh_claim_vx_info(v, d, h) do { } while (0)
-+#define __vxh_release_vx_info(v, d, h) do { } while (0)
++struct _vxm_update_1 {
++ uint32_t tokens_max;
++ uint32_t fill_rate;
++ uint32_t interval;
++};
+
-+#define vxh_alloc_vx_info(v) do { } while (0)
-+#define vxh_dealloc_vx_info(v) do { } while (0)
++struct _vxm_update_2 {
++ uint32_t tokens_min;
++ uint32_t fill_rate;
++ uint32_t interval;
++};
+
-+#define vxh_hash_vx_info(v) do { } while (0)
-+#define vxh_unhash_vx_info(v) do { } while (0)
++struct _vxm_rqinfo_1 {
++ uint16_t running;
++ uint16_t onhold;
++ uint16_t iowait;
++ uint16_t uintr;
++ uint32_t idle_tokens;
++};
+
-+#define vxh_loc_vx_info(v, l) do { } while (0)
-+#define vxh_lookup_vx_info(v, l) do { } while (0)
-+#define vxh_create_vx_info(v, l) do { } while (0)
++struct _vxm_rqinfo_2 {
++ uint32_t norm_time;
++ uint32_t idle_time;
++ uint32_t idle_skip;
++};
+
-+#define vxh_dump_history() do { } while (0)
++struct _vxm_sched {
++ uint32_t tokens;
++ uint32_t norm_time;
++ uint32_t idle_time;
++};
+
++struct _vxm_task {
++ uint16_t pid;
++ uint16_t state;
++};
+
-+#endif /* CONFIG_VSERVER_HISTORY */
++struct _vxm_event {
++ uint32_t jif;
++ union {
++ uint32_t seq;
++ uint32_t sec;
++ };
++ union {
++ uint32_t tokens;
++ uint32_t nsec;
++ struct _vxm_task tsk;
++ };
++};
+
-+#endif /* _VX_HISTORY_H */
---- linux-2.6.22/include/linux/vserver/inode.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/inode.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,38 @@
-+#ifndef _VX_INODE_H
-+#define _VX_INODE_H
++struct _vx_mon_entry {
++ uint16_t type;
++ uint16_t xid;
++ union {
++ struct _vxm_event ev;
++ struct _vxm_sched sd;
++ struct _vxm_update_1 u1;
++ struct _vxm_update_2 u2;
++ struct _vxm_rqinfo_1 q1;
++ struct _vxm_rqinfo_2 q2;
++ };
++};
+
+
-+#define IATTR_TAG 0x01000000
++#endif /* _VX_MONITOR_H */
+--- linux-2.6.25.4/include/linux/vserver/network_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/network_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,150 @@
++#ifndef _VX_NETWORK_CMD_H
++#define _VX_NETWORK_CMD_H
+
-+#define IATTR_ADMIN 0x00000001
-+#define IATTR_WATCH 0x00000002
-+#define IATTR_HIDE 0x00000004
-+#define IATTR_FLAGS 0x00000007
+
-+#define IATTR_BARRIER 0x00010000
-+#define IATTR_IUNLINK 0x00020000
-+#define IATTR_IMMUTABLE 0x00040000
++/* vinfo commands */
++
++#define VCMD_task_nid VC_CMD(VINFO, 2, 0)
+
+#ifdef __KERNEL__
++extern int vc_task_nid(uint32_t);
++
++#endif /* __KERNEL__ */
+
++#define VCMD_nx_info VC_CMD(VINFO, 6, 0)
+
-+#ifdef CONFIG_VSERVER_PROC_SECURE
-+#define IATTR_PROC_DEFAULT ( IATTR_ADMIN | IATTR_HIDE )
-+#define IATTR_PROC_SYMLINK ( IATTR_ADMIN )
-+#else
-+#define IATTR_PROC_DEFAULT ( IATTR_ADMIN )
-+#define IATTR_PROC_SYMLINK ( IATTR_ADMIN )
-+#endif
++struct vcmd_nx_info_v0 {
++ uint32_t nid;
++ /* more to come */
++};
+
-+#define vx_hide_check(c, m) (((m) & IATTR_HIDE) ? vx_check(c, m) : 1)
++#ifdef __KERNEL__
++extern int vc_nx_info(struct nx_info *, void __user *);
+
+#endif /* __KERNEL__ */
+
-+/* inode ioctls */
++#include <linux/in.h>
++#include <linux/in6.h>
+
-+#define FIOC_GETXFLG _IOR('x', 5, long)
-+#define FIOC_SETXFLG _IOW('x', 6, long)
++#define VCMD_net_create_v0 VC_CMD(VNET, 1, 0)
++#define VCMD_net_create VC_CMD(VNET, 1, 1)
+
-+#else /* _VX_INODE_H */
-+#warning duplicate inclusion
-+#endif /* _VX_INODE_H */
---- linux-2.6.22/include/linux/vserver/inode_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/inode_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,61 @@
-+#ifndef _VX_INODE_CMD_H
-+#define _VX_INODE_CMD_H
++struct vcmd_net_create {
++ uint64_t flagword;
++};
+
++#define VCMD_net_migrate VC_CMD(NETMIG, 1, 0)
+
-+/* inode vserver commands */
-+
-+#define VCMD_get_iattr_v0 VC_CMD(INODE, 1, 0)
-+#define VCMD_set_iattr_v0 VC_CMD(INODE, 2, 0)
-+
-+#define VCMD_get_iattr VC_CMD(INODE, 1, 1)
-+#define VCMD_set_iattr VC_CMD(INODE, 2, 1)
++#define VCMD_net_add VC_CMD(NETALT, 1, 0)
++#define VCMD_net_remove VC_CMD(NETALT, 2, 0)
+
-+struct vcmd_ctx_iattr_v0 {
-+ /* device handle in id */
-+ uint64_t ino;
-+ uint32_t xid;
-+ uint32_t flags;
-+ uint32_t mask;
++struct vcmd_net_addr_v0 {
++ uint16_t type;
++ uint16_t count;
++ struct in_addr ip[4];
++ struct in_addr mask[4];
+};
+
-+struct vcmd_ctx_iattr_v1 {
-+ const char __user *name;
-+ uint32_t xid;
-+ uint32_t flags;
-+ uint32_t mask;
++#define VCMD_net_add_ipv4 VC_CMD(NETALT, 1, 1)
++#define VCMD_net_remove_ipv4 VC_CMD(NETALT, 2, 1)
++
++struct vcmd_net_addr_ipv4_v1 {
++ uint16_t type;
++ uint16_t flags;
++ struct in_addr ip;
++ struct in_addr mask;
+};
+
++#define VCMD_net_add_ipv6 VC_CMD(NETALT, 3, 1)
++#define VCMD_net_remove_ipv6 VC_CMD(NETALT, 4, 1)
+
-+#ifdef __KERNEL__
++struct vcmd_net_addr_ipv6_v1 {
++ uint16_t type;
++ uint16_t flags;
++ uint32_t prefix;
++ struct in6_addr ip;
++ struct in6_addr mask;
++};
+
++#define VCMD_add_match_ipv4 VC_CMD(NETALT, 5, 0)
++#define VCMD_get_match_ipv4 VC_CMD(NETALT, 6, 0)
+
-+#ifdef CONFIG_COMPAT
++struct vcmd_match_ipv4_v0 {
++ uint16_t type;
++ uint16_t flags;
++ uint16_t parent;
++ uint16_t prefix;
++ struct in_addr ip;
++ struct in_addr ip2;
++ struct in_addr mask;
++};
+
-+#include <asm/compat.h>
++#define VCMD_add_match_ipv6 VC_CMD(NETALT, 7, 0)
++#define VCMD_get_match_ipv6 VC_CMD(NETALT, 8, 0)
+
-+struct vcmd_ctx_iattr_v1_x32 {
-+ compat_uptr_t name_ptr;
-+ uint32_t xid;
-+ uint32_t flags;
-+ uint32_t mask;
++struct vcmd_match_ipv6_v0 {
++ uint16_t type;
++ uint16_t flags;
++ uint16_t parent;
++ uint16_t prefix;
++ struct in6_addr ip;
++ struct in6_addr ip2;
++ struct in6_addr mask;
+};
+
-+#endif /* CONFIG_COMPAT */
+
-+#include <linux/compiler.h>
++#ifdef __KERNEL__
++extern int vc_net_create(uint32_t, void __user *);
++extern int vc_net_migrate(struct nx_info *, void __user *);
+
-+extern int vc_get_iattr_v0(uint32_t, void __user *);
-+extern int vc_set_iattr_v0(uint32_t, void __user *);
++extern int vc_net_add(struct nx_info *, void __user *);
++extern int vc_net_remove(struct nx_info *, void __user *);
+
-+extern int vc_get_iattr(uint32_t, void __user *);
-+extern int vc_set_iattr(uint32_t, void __user *);
++extern int vc_net_add_ipv4(struct nx_info *, void __user *);
++extern int vc_net_remove_ipv4(struct nx_info *, void __user *);
+
-+#ifdef CONFIG_COMPAT
++extern int vc_net_add_ipv6(struct nx_info *, void __user *);
++extern int vc_net_remove_ipv6(struct nx_info *, void __user *);
+
-+extern int vc_get_iattr_x32(uint32_t, void __user *);
-+extern int vc_set_iattr_x32(uint32_t, void __user *);
++extern int vc_add_match_ipv4(struct nx_info *, void __user *);
++extern int vc_get_match_ipv4(struct nx_info *, void __user *);
+
-+#endif /* CONFIG_COMPAT */
++extern int vc_add_match_ipv6(struct nx_info *, void __user *);
++extern int vc_get_match_ipv6(struct nx_info *, void __user *);
+
+#endif /* __KERNEL__ */
-+#endif /* _VX_INODE_CMD_H */
---- linux-2.6.22/include/linux/vserver/legacy.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/legacy.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,49 @@
-+#ifndef _VX_LEGACY_H
-+#define _VX_LEGACY_H
-+
-+#include "switch.h"
+
+
-+/* compatibiliy vserver commands */
++/* flag commands */
+
-+#define VCMD_new_s_context VC_CMD(COMPAT, 1, 1)
-+#define VCMD_set_ipv4root VC_CMD(COMPAT, 2, 3)
++#define VCMD_get_nflags VC_CMD(FLAGS, 5, 0)
++#define VCMD_set_nflags VC_CMD(FLAGS, 6, 0)
+
-+#define VCMD_create_context VC_CMD(VSETUP, 1, 0)
++struct vcmd_net_flags_v0 {
++ uint64_t flagword;
++ uint64_t mask;
++};
+
-+/* compatibiliy vserver arguments */
++#ifdef __KERNEL__
++extern int vc_get_nflags(struct nx_info *, void __user *);
++extern int vc_set_nflags(struct nx_info *, void __user *);
+
-+struct vcmd_new_s_context_v1 {
-+ uint32_t remove_cap;
-+ uint32_t flags;
-+};
++#endif /* __KERNEL__ */
+
-+struct vcmd_set_ipv4root_v3 {
-+ /* number of pairs in id */
-+ uint32_t broadcast;
-+ struct {
-+ uint32_t ip;
-+ uint32_t mask;
-+ } nx_mask_pair[NB_IPV4ROOT];
-+};
+
++/* network caps commands */
+
-+#define VX_INFO_LOCK 1 /* Can't request a new vx_id */
-+#define VX_INFO_NPROC 4 /* Limit number of processes in a context */
-+#define VX_INFO_PRIVATE 8 /* Noone can join this security context */
-+#define VX_INFO_INIT 16 /* This process wants to become the */
-+ /* logical process 1 of the security */
-+ /* context */
-+#define VX_INFO_HIDEINFO 32 /* Hide some information in /proc */
-+#define VX_INFO_ULIMIT 64 /* Use ulimit of the current process */
-+ /* to become the global limits */
-+ /* of the context */
-+#define VX_INFO_NAMESPACE 128 /* save private namespace */
++#define VCMD_get_ncaps VC_CMD(FLAGS, 7, 0)
++#define VCMD_set_ncaps VC_CMD(FLAGS, 8, 0)
+
++struct vcmd_net_caps_v0 {
++ uint64_t ncaps;
++ uint64_t cmask;
++};
+
+#ifdef __KERNEL__
-+extern int vc_new_s_context(uint32_t, void __user *);
-+extern int vc_set_ipv4root(uint32_t, void __user *);
++extern int vc_get_ncaps(struct nx_info *, void __user *);
++extern int vc_set_ncaps(struct nx_info *, void __user *);
+
+#endif /* __KERNEL__ */
-+#endif /* _VX_LEGACY_H */
---- linux-2.6.22/include/linux/vserver/limit.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/limit.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,68 @@
-+#ifndef _VX_LIMIT_H
-+#define _VX_LIMIT_H
-+
-+
-+#define VLIMIT_NSOCK 16
-+#define VLIMIT_OPENFD 17
-+#define VLIMIT_ANON 18
-+#define VLIMIT_SHMEM 19
-+#define VLIMIT_SEMARY 20
-+#define VLIMIT_NSEMS 21
-+#define VLIMIT_DENTRY 22
-+#define VLIMIT_MAPPED 23
++#endif /* _VX_CONTEXT_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/network.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/network.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,146 @@
++#ifndef _VX_NETWORK_H
++#define _VX_NETWORK_H
+
++#include <linux/types.h>
+
-+#ifdef __KERNEL__
+
-+#define VLIM_NOCHECK ((1L << VLIMIT_DENTRY) | (1L << RLIMIT_RSS))
++#define MAX_N_CONTEXT 65535 /* Arbitrary limit */
+
-+/* keep in sync with CRLIM_INFINITY */
+
-+#define VLIM_INFINITY (~0ULL)
++/* network flags */
+
-+#ifndef RLIM_INFINITY
-+#warning RLIM_INFINITY is undefined
-+#endif
++#define NXF_INFO_PRIVATE 0x00000008
+
-+#define __rlim_val(l, r, v) ((l)->res[r].v)
++#define NXF_SINGLE_IP 0x00000100
++#define NXF_LBACK_REMAP 0x00000200
++#define NXF_LBACK_ALLOW 0x00000400
+
-+#define __rlim_soft(l, r) __rlim_val(l, r, soft)
-+#define __rlim_hard(l, r) __rlim_val(l, r, hard)
++#define NXF_HIDE_NETIF 0x02000000
++#define NXF_HIDE_LBACK 0x04000000
+
-+#define __rlim_rcur(l, r) __rlim_val(l, r, rcur)
-+#define __rlim_rmin(l, r) __rlim_val(l, r, rmin)
-+#define __rlim_rmax(l, r) __rlim_val(l, r, rmax)
++#define NXF_STATE_SETUP (1ULL << 32)
++#define NXF_STATE_ADMIN (1ULL << 34)
+
-+#define __rlim_lhit(l, r) __rlim_val(l, r, lhit)
-+#define __rlim_hit(l, r) atomic_inc(&__rlim_lhit(l, r))
++#define NXF_SC_HELPER (1ULL << 36)
++#define NXF_PERSISTENT (1ULL << 38)
+
-+typedef atomic_long_t rlim_atomic_t;
-+typedef unsigned long rlim_t;
++#define NXF_ONE_TIME (0x0005ULL << 32)
+
-+#define __rlim_get(l, r) atomic_long_read(&__rlim_rcur(l, r))
-+#define __rlim_set(l, r, v) atomic_long_set(&__rlim_rcur(l, r), v)
-+#define __rlim_inc(l, r) atomic_long_inc(&__rlim_rcur(l, r))
-+#define __rlim_dec(l, r) atomic_long_dec(&__rlim_rcur(l, r))
-+#define __rlim_add(l, r, v) atomic_long_add(v, &__rlim_rcur(l, r))
-+#define __rlim_sub(l, r, v) atomic_long_sub(v, &__rlim_rcur(l, r))
+
++#define NXF_INIT_SET (__nxf_init_set())
+
-+#if (RLIM_INFINITY == VLIM_INFINITY)
-+#define VX_VLIM(r) ((long long)(long)(r))
-+#define VX_RLIM(v) ((rlim_t)(v))
-+#else
-+#define VX_VLIM(r) (((r) == RLIM_INFINITY) \
-+ ? VLIM_INFINITY : (long long)(r))
-+#define VX_RLIM(v) (((v) == VLIM_INFINITY) \
-+ ? RLIM_INFINITY : (rlim_t)(v))
++static inline uint64_t __nxf_init_set(void) {
++ return NXF_STATE_ADMIN
++#ifdef CONFIG_VSERVER_AUTO_LBACK
++ | NXF_LBACK_REMAP
++ | NXF_HIDE_LBACK
++#endif
++#ifdef CONFIG_VSERVER_AUTO_SINGLE
++ | NXF_SINGLE_IP
+#endif
++ | NXF_HIDE_NETIF;
++}
+
-+struct sysinfo;
+
-+void vx_vsi_meminfo(struct sysinfo *);
-+void vx_vsi_swapinfo(struct sysinfo *);
++/* network caps */
+
-+#define NUM_LIMITS 24
++#define NXC_TUN_CREATE 0x00000001
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_LIMIT_H */
---- linux-2.6.22/include/linux/vserver/limit_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/limit_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,69 @@
-+#ifndef _VX_LIMIT_CMD_H
-+#define _VX_LIMIT_CMD_H
++#define NXC_RAW_ICMP 0x00000100
+
+
-+/* rlimit vserver commands */
++/* address types */
+
-+#define VCMD_get_rlimit VC_CMD(RLIMIT, 1, 0)
-+#define VCMD_set_rlimit VC_CMD(RLIMIT, 2, 0)
-+#define VCMD_get_rlimit_mask VC_CMD(RLIMIT, 3, 0)
-+#define VCMD_reset_minmax VC_CMD(RLIMIT, 9, 0)
++#define NXA_TYPE_IPV4 0x0001
++#define NXA_TYPE_IPV6 0x0002
+
-+struct vcmd_ctx_rlimit_v0 {
-+ uint32_t id;
-+ uint64_t minimum;
-+ uint64_t softlimit;
-+ uint64_t maximum;
-+};
++#define NXA_TYPE_NONE 0x0000
++#define NXA_TYPE_ANY 0x00FF
+
-+struct vcmd_ctx_rlimit_mask_v0 {
-+ uint32_t minimum;
-+ uint32_t softlimit;
-+ uint32_t maximum;
-+};
++#define NXA_TYPE_ADDR 0x0010
++#define NXA_TYPE_MASK 0x0020
++#define NXA_TYPE_RANGE 0x0040
+
-+#define VCMD_rlimit_stat VC_CMD(VSTAT, 1, 0)
++#define NXA_MASK_ALL (NXA_TYPE_ADDR | NXA_TYPE_MASK | NXA_TYPE_RANGE)
+
-+struct vcmd_rlimit_stat_v0 {
-+ uint32_t id;
-+ uint32_t hits;
-+ uint64_t value;
-+ uint64_t minimum;
-+ uint64_t maximum;
-+};
++#define NXA_MOD_BCAST 0x0100
++#define NXA_MOD_LBACK 0x0200
+
-+#define CRLIM_UNSET (0ULL)
-+#define CRLIM_INFINITY (~0ULL)
-+#define CRLIM_KEEP (~1ULL)
++#define NXA_LOOPBACK 0x1000
++
++#define NXA_MASK_BIND (NXA_MASK_ALL | NXA_MOD_BCAST | NXA_MOD_LBACK)
++#define NXA_MASK_SHOW (NXA_MASK_ALL | NXA_LOOPBACK)
+
+#ifdef __KERNEL__
+
-+#ifdef CONFIG_IA32_EMULATION
++#include <linux/list.h>
++#include <linux/spinlock.h>
++#include <linux/rcupdate.h>
++#include <linux/in.h>
++#include <linux/in6.h>
++#include <asm/atomic.h>
+
-+struct vcmd_ctx_rlimit_v0_x32 {
-+ uint32_t id;
-+ uint64_t minimum;
-+ uint64_t softlimit;
-+ uint64_t maximum;
-+} __attribute__ ((aligned (4)));
++struct nx_addr_v4 {
++ struct nx_addr_v4 *next;
++ struct in_addr ip[2];
++ struct in_addr mask;
++ uint16_t type;
++ uint16_t flags;
++};
+
-+#endif /* CONFIG_IA32_EMULATION */
++struct nx_addr_v6 {
++ struct nx_addr_v6 *next;
++ struct in6_addr ip;
++ struct in6_addr mask;
++ uint32_t prefix;
++ uint16_t type;
++ uint16_t flags;
++};
+
-+#include <linux/compiler.h>
++struct nx_info {
++ struct hlist_node nx_hlist; /* linked list of nxinfos */
++ nid_t nx_id; /* vnet id */
++ atomic_t nx_usecnt; /* usage count */
++ atomic_t nx_tasks; /* tasks count */
++ int nx_state; /* context state */
+
-+extern int vc_get_rlimit_mask(uint32_t, void __user *);
-+extern int vc_get_rlimit(struct vx_info *, void __user *);
-+extern int vc_set_rlimit(struct vx_info *, void __user *);
-+extern int vc_reset_minmax(struct vx_info *, void __user *);
++ uint64_t nx_flags; /* network flag word */
++ uint64_t nx_ncaps; /* network capabilities */
+
-+extern int vc_rlimit_stat(struct vx_info *, void __user *);
++ struct in_addr v4_lback; /* Loopback address */
++ struct in_addr v4_bcast; /* Broadcast address */
++ struct nx_addr_v4 v4; /* First/Single ipv4 address */
++#ifdef CONFIG_IPV6
++ struct nx_addr_v6 v6; /* First/Single ipv6 address */
++#endif
++ char nx_name[65]; /* network context name */
++};
+
-+#ifdef CONFIG_IA32_EMULATION
+
-+extern int vc_get_rlimit_x32(struct vx_info *, void __user *);
-+extern int vc_set_rlimit_x32(struct vx_info *, void __user *);
++/* status flags */
+
-+#endif /* CONFIG_IA32_EMULATION */
++#define NXS_HASHED 0x0001
++#define NXS_SHUTDOWN 0x0100
++#define NXS_RELEASED 0x8000
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_LIMIT_CMD_H */
---- linux-2.6.22/include/linux/vserver/limit_def.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/limit_def.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,47 @@
-+#ifndef _VX_LIMIT_DEF_H
-+#define _VX_LIMIT_DEF_H
++extern struct nx_info *lookup_nx_info(int);
+
-+#include <asm/atomic.h>
-+#include <asm/resource.h>
++extern int get_nid_list(int, unsigned int *, int);
++extern int nid_is_hashed(nid_t);
+
-+#include "limit.h"
++extern int nx_migrate_task(struct task_struct *, struct nx_info *);
+
++extern long vs_net_change(struct nx_info *, unsigned int);
+
-+struct _vx_res_limit {
-+ rlim_t soft; /* Context soft limit */
-+ rlim_t hard; /* Context hard limit */
++struct sock;
+
-+ rlim_atomic_t rcur; /* Current value */
-+ rlim_t rmin; /* Context minimum */
-+ rlim_t rmax; /* Context maximum */
+
-+ atomic_t lhit; /* Limit hits */
-+};
++#define NX_IPV4(n) ((n)->v4.type != NXA_TYPE_NONE)
++#ifdef CONFIG_IPV6
++#define NX_IPV6(n) ((n)->v6.type != NXA_TYPE_NONE)
++#else
++#define NX_IPV6(n) (0)
++#endif
+
-+/* context sub struct */
++#endif /* __KERNEL__ */
++#endif /* _VX_NETWORK_H */
+--- linux-2.6.25.4/include/linux/vserver/percpu.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/percpu.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,14 @@
++#ifndef _VX_PERCPU_H
++#define _VX_PERCPU_H
+
-+struct _vx_limit {
-+ struct _vx_res_limit res[NUM_LIMITS];
++#include "cvirt_def.h"
++#include "sched_def.h"
++
++struct _vx_percpu {
++ struct _vx_cvirt_pc cvirt;
++ struct _vx_sched_pc sched;
+};
+
-+#ifdef CONFIG_VSERVER_DEBUG
++#define PERCPU_PERCTX (sizeof(struct _vx_percpu))
+
-+static inline void __dump_vx_limit(struct _vx_limit *limit)
-+{
-+ int i;
++#endif /* _VX_PERCPU_H */
+--- linux-2.6.25.4/include/linux/vserver/pid.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/pid.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,51 @@
++#ifndef _VSERVER_PID_H
++#define _VSERVER_PID_H
+
-+ printk("\t_vx_limit:");
-+ for (i = 0; i < NUM_LIMITS; i++) {
-+ printk("\t [%2d] = %8lu %8lu/%8lu, %8ld/%8ld, %8d\n",
-+ i, (unsigned long)__rlim_get(limit, i),
-+ (unsigned long)__rlim_rmin(limit, i),
-+ (unsigned long)__rlim_rmax(limit, i),
-+ (long)__rlim_soft(limit, i),
-+ (long)__rlim_hard(limit, i),
-+ atomic_read(&__rlim_lhit(limit, i)));
++/* pid faking stuff */
++
++#define vx_info_map_pid(v, p) \
++ __vx_info_map_pid((v), (p), __FUNC__, __FILE__, __LINE__)
++#define vx_info_map_tgid(v,p) vx_info_map_pid(v,p)
++#define vx_map_pid(p) vx_info_map_pid(current->vx_info, p)
++#define vx_map_tgid(p) vx_map_pid(p)
++
++static inline int __vx_info_map_pid(struct vx_info *vxi, int pid,
++ const char *func, const char *file, int line)
++{
++ if (vx_info_flags(vxi, VXF_INFO_INIT, 0)) {
++ vxfprintk(VXD_CBIT(cvirt, 2),
++ "vx_map_tgid: %p/%llx: %d -> %d",
++ vxi, (long long)vxi->vx_flags, pid,
++ (pid && pid == vxi->vx_initpid) ? 1 : pid,
++ func, file, line);
++ if (pid == 0)
++ return 0;
++ if (pid == vxi->vx_initpid)
++ return 1;
+ }
++ return pid;
+}
+
-+#endif
++#define vx_info_rmap_pid(v, p) \
++ __vx_info_rmap_pid((v), (p), __FUNC__, __FILE__, __LINE__)
++#define vx_rmap_pid(p) vx_info_rmap_pid(current->vx_info, p)
++#define vx_rmap_tgid(p) vx_rmap_pid(p)
+
-+#endif /* _VX_LIMIT_DEF_H */
---- linux-2.6.22/include/linux/vserver/limit_int.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/limit_int.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,198 @@
-+#ifndef _VX_LIMIT_INT_H
-+#define _VX_LIMIT_INT_H
++static inline int __vx_info_rmap_pid(struct vx_info *vxi, int pid,
++ const char *func, const char *file, int line)
++{
++ if (vx_info_flags(vxi, VXF_INFO_INIT, 0)) {
++ vxfprintk(VXD_CBIT(cvirt, 2),
++ "vx_rmap_tgid: %p/%llx: %d -> %d",
++ vxi, (long long)vxi->vx_flags, pid,
++ (pid == 1) ? vxi->vx_initpid : pid,
++ func, file, line);
++ if ((pid == 1) && vxi->vx_initpid)
++ return vxi->vx_initpid;
++ if (pid == vxi->vx_initpid)
++ return ~0U;
++ }
++ return pid;
++}
+
-+#include "context.h"
++#endif
+--- linux-2.6.25.4/include/linux/vserver/sched_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/sched_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,108 @@
++#ifndef _VX_SCHED_CMD_H
++#define _VX_SCHED_CMD_H
+
-+#ifdef __KERNEL__
+
-+#define VXD_RCRES_COND(r) VXD_CBIT(cres, r)
-+#define VXD_RLIMIT_COND(r) VXD_CBIT(limit, r)
++/* sched vserver commands */
+
-+extern const char *vlimit_name[NUM_LIMITS];
++#define VCMD_set_sched_v2 VC_CMD(SCHED, 1, 2)
++#define VCMD_set_sched_v3 VC_CMD(SCHED, 1, 3)
++#define VCMD_set_sched_v4 VC_CMD(SCHED, 1, 4)
+
-+static inline void __vx_acc_cres(struct vx_info *vxi,
-+ int res, int dir, void *_data, char *_file, int _line)
-+{
-+ if (VXD_RCRES_COND(res))
-+ vxlprintk(1, "vx_acc_cres[%5d,%s,%2d]: %5ld%s (%p)",
-+ (vxi ? vxi->vx_id : -1), vlimit_name[res], res,
-+ (vxi ? (long)__rlim_get(&vxi->limit, res) : 0),
-+ (dir > 0) ? "++" : "--", _data, _file, _line);
-+ if (!vxi)
-+ return;
++struct vcmd_set_sched_v2 {
++ int32_t fill_rate;
++ int32_t interval;
++ int32_t tokens;
++ int32_t tokens_min;
++ int32_t tokens_max;
++ uint64_t cpu_mask;
++};
+
-+ if (dir > 0)
-+ __rlim_inc(&vxi->limit, res);
-+ else
-+ __rlim_dec(&vxi->limit, res);
-+}
++struct vcmd_set_sched_v3 {
++ uint32_t set_mask;
++ int32_t fill_rate;
++ int32_t interval;
++ int32_t tokens;
++ int32_t tokens_min;
++ int32_t tokens_max;
++ int32_t priority_bias;
++};
+
-+static inline void __vx_add_cres(struct vx_info *vxi,
-+ int res, int amount, void *_data, char *_file, int _line)
-+{
-+ if (VXD_RCRES_COND(res))
-+ vxlprintk(1, "vx_add_cres[%5d,%s,%2d]: %5ld += %5d (%p)",
-+ (vxi ? vxi->vx_id : -1), vlimit_name[res], res,
-+ (vxi ? (long)__rlim_get(&vxi->limit, res) : 0),
-+ amount, _data, _file, _line);
-+ if (amount == 0)
-+ return;
-+ if (!vxi)
-+ return;
-+ __rlim_add(&vxi->limit, res, amount);
-+}
++struct vcmd_set_sched_v4 {
++ uint32_t set_mask;
++ int32_t fill_rate;
++ int32_t interval;
++ int32_t tokens;
++ int32_t tokens_min;
++ int32_t tokens_max;
++ int32_t prio_bias;
++ int32_t cpu_id;
++ int32_t bucket_id;
++};
+
-+static inline
-+int __vx_cres_adjust_max(struct _vx_limit *limit, int res, rlim_t value)
-+{
-+ int cond = (value > __rlim_rmax(limit, res));
++#define VCMD_set_sched VC_CMD(SCHED, 1, 5)
++#define VCMD_get_sched VC_CMD(SCHED, 2, 5)
+
-+ if (cond)
-+ __rlim_rmax(limit, res) = value;
-+ return cond;
-+}
++struct vcmd_sched_v5 {
++ uint32_t mask;
++ int32_t cpu_id;
++ int32_t bucket_id;
++ int32_t fill_rate[2];
++ int32_t interval[2];
++ int32_t tokens;
++ int32_t tokens_min;
++ int32_t tokens_max;
++ int32_t prio_bias;
++};
+
-+static inline
-+int __vx_cres_adjust_min(struct _vx_limit *limit, int res, rlim_t value)
-+{
-+ int cond = (value < __rlim_rmin(limit, res));
++#define VXSM_FILL_RATE 0x0001
++#define VXSM_INTERVAL 0x0002
++#define VXSM_FILL_RATE2 0x0004
++#define VXSM_INTERVAL2 0x0008
++#define VXSM_TOKENS 0x0010
++#define VXSM_TOKENS_MIN 0x0020
++#define VXSM_TOKENS_MAX 0x0040
++#define VXSM_PRIO_BIAS 0x0100
+
-+ if (cond)
-+ __rlim_rmin(limit, res) = value;
-+ return cond;
-+}
++#define VXSM_IDLE_TIME 0x0200
++#define VXSM_FORCE 0x0400
+
-+static inline
-+void __vx_cres_fixup(struct _vx_limit *limit, int res, rlim_t value)
-+{
-+ if (!__vx_cres_adjust_max(limit, res, value))
-+ __vx_cres_adjust_min(limit, res, value);
-+}
++#define VXSM_V3_MASK 0x0173
++#define VXSM_SET_MASK 0x01FF
+
++#define VXSM_CPU_ID 0x1000
++#define VXSM_BUCKET_ID 0x2000
+
-+/* return values:
-+ +1 ... no limit hit
-+ -1 ... over soft limit
-+ 0 ... over hard limit */
++#define VXSM_MSEC 0x4000
+
-+static inline int __vx_cres_avail(struct vx_info *vxi,
-+ int res, int num, char *_file, int _line)
-+{
-+ struct _vx_limit *limit;
-+ rlim_t value;
++#define SCHED_KEEP (-2) /* only for v2 */
+
-+ if (VXD_RLIMIT_COND(res))
-+ vxlprintk(1, "vx_cres_avail[%5d,%s,%2d]: %5ld/%5ld > %5ld + %5d",
-+ (vxi ? vxi->vx_id : -1), vlimit_name[res], res,
-+ (vxi ? (long)__rlim_soft(&vxi->limit, res) : -1),
-+ (vxi ? (long)__rlim_hard(&vxi->limit, res) : -1),
-+ (vxi ? (long)__rlim_get(&vxi->limit, res) : 0),
-+ num, _file, _line);
-+ if (!vxi)
-+ return 1;
++#ifdef __KERNEL__
+
-+ limit = &vxi->limit;
-+ value = __rlim_get(limit, res);
++#include <linux/compiler.h>
+
-+ if (!__vx_cres_adjust_max(limit, res, value))
-+ __vx_cres_adjust_min(limit, res, value);
++extern int vc_set_sched_v2(struct vx_info *, void __user *);
++extern int vc_set_sched_v3(struct vx_info *, void __user *);
++extern int vc_set_sched_v4(struct vx_info *, void __user *);
++extern int vc_set_sched(struct vx_info *, void __user *);
++extern int vc_get_sched(struct vx_info *, void __user *);
+
-+ if (num == 0)
-+ return 1;
++#endif /* __KERNEL__ */
+
-+ if (__rlim_soft(limit, res) == RLIM_INFINITY)
-+ return -1;
-+ if (value + num <= __rlim_soft(limit, res))
-+ return -1;
++#define VCMD_sched_info VC_CMD(SCHED, 3, 0)
+
-+ if (__rlim_hard(limit, res) == RLIM_INFINITY)
-+ return 1;
-+ if (value + num <= __rlim_hard(limit, res))
-+ return 1;
++struct vcmd_sched_info {
++ int32_t cpu_id;
++ int32_t bucket_id;
++ uint64_t user_msec;
++ uint64_t sys_msec;
++ uint64_t hold_msec;
++ uint32_t token_usec;
++ int32_t vavavoom;
++};
+
-+ __rlim_hit(limit, res);
-+ return 0;
-+}
++#ifdef __KERNEL__
+
++extern int vc_sched_info(struct vx_info *, void __user *);
+
-+static const int VLA_RSS[] = { RLIMIT_RSS, VLIMIT_ANON, VLIMIT_MAPPED, 0 };
++#endif /* __KERNEL__ */
++#endif /* _VX_SCHED_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/sched_def.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/sched_def.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,68 @@
++#ifndef _VX_SCHED_DEF_H
++#define _VX_SCHED_DEF_H
+
-+static inline
-+rlim_t __vx_cres_array_sum(struct _vx_limit *limit, const int *array)
-+{
-+ rlim_t value, sum = 0;
-+ int res;
++#include <linux/spinlock.h>
++#include <linux/jiffies.h>
++#include <linux/cpumask.h>
++#include <asm/atomic.h>
++#include <asm/param.h>
+
-+ while ((res = *array++)) {
-+ value = __rlim_get(limit, res);
-+ __vx_cres_fixup(limit, res, value);
-+ sum += value;
-+ }
-+ return sum;
-+}
+
-+static inline
-+rlim_t __vx_cres_array_fixup(struct _vx_limit *limit, const int *array)
-+{
-+ rlim_t value = __vx_cres_array_sum(limit, array + 1);
-+ int res = *array;
++/* context sub struct */
+
-+ if (value == __rlim_get(limit, res))
-+ return value;
-+
-+ __rlim_set(limit, res, value);
-+ /* now adjust min/max */
-+ if (!__vx_cres_adjust_max(limit, res, value))
-+ __vx_cres_adjust_min(limit, res, value);
++struct _vx_sched {
++ spinlock_t tokens_lock; /* lock for token bucket */
+
-+ return value;
-+}
++ int tokens; /* number of CPU tokens */
++ int fill_rate[2]; /* Fill rate: add X tokens... */
++ int interval[2]; /* Divisor: per Y jiffies */
++ int tokens_min; /* Limit: minimum for unhold */
++ int tokens_max; /* Limit: no more than N tokens */
+
-+static inline int __vx_cres_array_avail(struct vx_info *vxi,
-+ const int *array, int num, char *_file, int _line)
-+{
-+ struct _vx_limit *limit;
-+ rlim_t value = 0;
-+ int res;
++ int prio_bias; /* bias offset for priority */
+
-+ if (num == 0)
-+ return 1;
-+ if (!vxi)
-+ return 1;
++ unsigned update_mask; /* which features should be updated */
++ cpumask_t update; /* CPUs which should update */
++};
+
-+ limit = &vxi->limit;
-+ res = *array;
-+ value = __vx_cres_array_sum(limit, array + 1);
++struct _vx_sched_pc {
++ int tokens; /* number of CPU tokens */
++ int flags; /* bucket flags */
+
-+ __rlim_set(limit, res, value);
-+ __vx_cres_fixup(limit, res, value);
++ int fill_rate[2]; /* Fill rate: add X tokens... */
++ int interval[2]; /* Divisor: per Y jiffies */
++ int tokens_min; /* Limit: minimum for unhold */
++ int tokens_max; /* Limit: no more than N tokens */
+
-+ return __vx_cres_avail(vxi, res, num, _file, _line);
-+}
++ int prio_bias; /* bias offset for priority */
++ int vavavoom; /* last calculated vavavoom */
+
++ unsigned long norm_time; /* last time accounted */
++ unsigned long idle_time; /* non linear time for fair sched */
++ unsigned long token_time; /* token time for accounting */
++ unsigned long onhold; /* jiffies when put on hold */
+
-+static inline void vx_limit_fixup(struct _vx_limit *limit, int id)
-+{
-+ rlim_t value;
-+ int res;
++ uint64_t user_ticks; /* token tick events */
++ uint64_t sys_ticks; /* token tick events */
++ uint64_t hold_ticks; /* token ticks paused */
++};
+
-+ /* complex resources first */
-+ if ((id < 0) || (id == RLIMIT_RSS))
-+ __vx_cres_array_fixup(limit, VLA_RSS);
+
-+ for (res = 0; res < NUM_LIMITS; res++) {
-+ if ((id > 0) && (res != id))
-+ continue;
++#define VXSF_ONHOLD 0x0001
++#define VXSF_IDLE_TIME 0x0100
+
-+ value = __rlim_get(limit, res);
-+ __vx_cres_fixup(limit, res, value);
++#ifdef CONFIG_VSERVER_DEBUG
+
-+ /* not supposed to happen, maybe warn? */
-+ if (__rlim_rmax(limit, res) > __rlim_hard(limit, res))
-+ __rlim_rmax(limit, res) = __rlim_hard(limit, res);
-+ }
++static inline void __dump_vx_sched(struct _vx_sched *sched)
++{
++ printk("\t_vx_sched:\n");
++ printk("\t tokens: %4d/%4d, %4d/%4d, %4d, %4d\n",
++ sched->fill_rate[0], sched->interval[0],
++ sched->fill_rate[1], sched->interval[1],
++ sched->tokens_min, sched->tokens_max);
++ printk("\t priority = %4d\n", sched->prio_bias);
+}
+
++#endif
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_LIMIT_INT_H */
---- linux-2.6.22/include/linux/vserver/monitor.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/monitor.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,95 @@
-+#ifndef _VX_MONITOR_H
-+#define _VX_MONITOR_H
++#endif /* _VX_SCHED_DEF_H */
+--- linux-2.6.25.4/include/linux/vserver/sched.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/sched.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,26 @@
++#ifndef _VX_SCHED_H
++#define _VX_SCHED_H
+
+
-+enum {
-+ VXM_UNUSED = 0,
++#ifdef __KERNEL__
+
-+ VXM_SYNC = 0x10,
++struct timespec;
+
-+ VXM_UPDATE = 0x20,
-+ VXM_UPDATE_1,
-+ VXM_UPDATE_2,
++void vx_vsi_uptime(struct timespec *, struct timespec *);
+
-+ VXM_RQINFO_1 = 0x24,
-+ VXM_RQINFO_2,
+
-+ VXM_ACTIVATE = 0x40,
-+ VXM_DEACTIVATE,
-+ VXM_IDLE,
++struct vx_info;
+
-+ VXM_HOLD = 0x44,
-+ VXM_UNHOLD,
++void vx_update_load(struct vx_info *);
+
-+ VXM_MIGRATE = 0x48,
-+ VXM_RESCHED,
+
-+ /* all other bits are flags */
-+ VXM_SCHED = 0x80,
-+};
++int vx_tokens_recalc(struct _vx_sched_pc *,
++ unsigned long *, unsigned long *, int [2]);
+
-+struct _vxm_update_1 {
-+ uint32_t tokens_max;
-+ uint32_t fill_rate;
-+ uint32_t interval;
-+};
++void vx_update_sched_param(struct _vx_sched *sched,
++ struct _vx_sched_pc *sched_pc);
+
-+struct _vxm_update_2 {
-+ uint32_t tokens_min;
-+ uint32_t fill_rate;
-+ uint32_t interval;
-+};
++#endif /* __KERNEL__ */
++#else /* _VX_SCHED_H */
++#warning duplicate inclusion
++#endif /* _VX_SCHED_H */
+--- linux-2.6.25.4/include/linux/vserver/signal_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/signal_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,43 @@
++#ifndef _VX_SIGNAL_CMD_H
++#define _VX_SIGNAL_CMD_H
+
-+struct _vxm_rqinfo_1 {
-+ uint16_t running;
-+ uint16_t onhold;
-+ uint16_t iowait;
-+ uint16_t uintr;
-+ uint32_t idle_tokens;
-+};
+
-+struct _vxm_rqinfo_2 {
-+ uint32_t norm_time;
-+ uint32_t idle_time;
-+ uint32_t idle_skip;
-+};
++/* signalling vserver commands */
+
-+struct _vxm_sched {
-+ uint32_t tokens;
-+ uint32_t norm_time;
-+ uint32_t idle_time;
-+};
++#define VCMD_ctx_kill VC_CMD(PROCTRL, 1, 0)
++#define VCMD_wait_exit VC_CMD(EVENT, 99, 0)
+
-+struct _vxm_task {
-+ uint16_t pid;
-+ uint16_t state;
++struct vcmd_ctx_kill_v0 {
++ int32_t pid;
++ int32_t sig;
+};
+
-+struct _vxm_event {
-+ uint32_t jif;
-+ union {
-+ uint32_t seq;
-+ uint32_t sec;
-+ };
-+ union {
-+ uint32_t tokens;
-+ uint32_t nsec;
-+ struct _vxm_task tsk;
-+ };
++struct vcmd_wait_exit_v0 {
++ int32_t reboot_cmd;
++ int32_t exit_code;
+};
+
-+struct _vx_mon_entry {
-+ uint16_t type;
-+ uint16_t xid;
-+ union {
-+ struct _vxm_event ev;
-+ struct _vxm_sched sd;
-+ struct _vxm_update_1 u1;
-+ struct _vxm_update_2 u2;
-+ struct _vxm_rqinfo_1 q1;
-+ struct _vxm_rqinfo_2 q2;
-+ };
-+};
++#ifdef __KERNEL__
+
++extern int vc_ctx_kill(struct vx_info *, void __user *);
++extern int vc_wait_exit(struct vx_info *, void __user *);
+
-+#endif /* _VX_MONITOR_H */
---- linux-2.6.22/include/linux/vserver/network.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/network.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,142 @@
-+#ifndef _VX_NETWORK_H
-+#define _VX_NETWORK_H
++#endif /* __KERNEL__ */
+
-+#include <linux/types.h>
++/* process alteration commands */
+
++#define VCMD_get_pflags VC_CMD(PROCALT, 5, 0)
++#define VCMD_set_pflags VC_CMD(PROCALT, 6, 0)
+
-+#define MAX_N_CONTEXT 65535 /* Arbitrary limit */
++struct vcmd_pflags_v0 {
++ uint32_t flagword;
++ uint32_t mask;
++};
+
-+#define NX_DYNAMIC_ID ((uint32_t)-1) /* id for dynamic context */
++#ifdef __KERNEL__
+
-+#define NB_IPV4ROOT 16
++extern int vc_get_pflags(uint32_t pid, void __user *);
++extern int vc_set_pflags(uint32_t pid, void __user *);
+
++#endif /* __KERNEL__ */
++#endif /* _VX_SIGNAL_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/signal.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/signal.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,14 @@
++#ifndef _VX_SIGNAL_H
++#define _VX_SIGNAL_H
+
-+/* network flags */
+
-+#define NXF_INFO_PRIVATE 0x00000008
++#ifdef __KERNEL__
+
-+#define NXF_STATE_SETUP (1ULL << 32)
-+#define NXF_STATE_ADMIN (1ULL << 34)
++struct vx_info;
+
-+#define NXF_SC_HELPER (1ULL << 36)
-+#define NXF_PERSISTENT (1ULL << 38)
++int vx_info_kill(struct vx_info *, int, int);
+
-+#define NXF_ONE_TIME (0x0005ULL << 32)
++#endif /* __KERNEL__ */
++#else /* _VX_SIGNAL_H */
++#warning duplicate inclusion
++#endif /* _VX_SIGNAL_H */
+--- linux-2.6.25.4/include/linux/vserver/space_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/space_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,26 @@
++#ifndef _VX_SPACE_CMD_H
++#define _VX_SPACE_CMD_H
+
-+#define NXF_INIT_SET (NXF_STATE_ADMIN)
+
++#define VCMD_enter_space_v0 VC_CMD(PROCALT, 1, 0)
++#define VCMD_enter_space VC_CMD(PROCALT, 1, 1)
+
-+/* address types */
++#define VCMD_set_space_v0 VC_CMD(PROCALT, 3, 0)
++#define VCMD_set_space VC_CMD(PROCALT, 3, 1)
+
-+#define NXA_TYPE_IPV4 1
-+#define NXA_TYPE_IPV6 2
++#define VCMD_get_space_mask VC_CMD(PROCALT, 4, 0)
+
-+#define NXA_MOD_BCAST (1 << 8)
+
-+#define NXA_TYPE_ANY ((uint16_t)-1)
++struct vcmd_space_mask {
++ uint64_t mask;
++};
+
+
+#ifdef __KERNEL__
+
-+#include <linux/list.h>
-+#include <linux/spinlock.h>
-+#include <linux/rcupdate.h>
-+#include <asm/atomic.h>
++extern int vc_enter_space(struct vx_info *, void __user *);
++extern int vc_set_space(struct vx_info *, void __user *);
++extern int vc_get_space_mask(struct vx_info *, void __user *);
+
++#endif /* __KERNEL__ */
++#endif /* _VX_SPACE_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/space.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/space.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,13 @@
++#ifndef _VX_SPACE_H
++#define _VX_SPACE_H
+
-+struct nx_info {
-+ struct hlist_node nx_hlist; /* linked list of nxinfos */
-+ nid_t nx_id; /* vnet id */
-+ atomic_t nx_usecnt; /* usage count */
-+ atomic_t nx_tasks; /* tasks count */
-+ int nx_state; /* context state */
+
-+ uint64_t nx_flags; /* network flag word */
-+ uint64_t nx_ncaps; /* network capabilities */
++#include <linux/types.h>
+
-+ int nbipv4;
-+ __u32 ipv4[NB_IPV4ROOT]; /* Process can only bind to these IPs */
-+ /* The first one is used to connect */
-+ /* and for bind any service */
-+ /* The other must be used explicity */
-+ __u32 mask[NB_IPV4ROOT]; /* Netmask for each ipv4 */
-+ /* Used to select the proper source */
-+ /* address for sockets */
-+ __u32 v4_bcast; /* Broadcast address to receive UDP */
++struct vx_info;
+
-+ char nx_name[65]; /* network context name */
-+};
++int vx_set_space(struct vx_info *vxi, unsigned long mask);
+
++#else /* _VX_SPACE_H */
++#warning duplicate inclusion
++#endif /* _VX_SPACE_H */
+--- linux-2.6.25.4/include/linux/vserver/switch.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/switch.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,99 @@
++#ifndef _VX_SWITCH_H
++#define _VX_SWITCH_H
+
-+/* status flags */
++#include <linux/types.h>
+
-+#define NXS_HASHED 0x0001
-+#define NXS_SHUTDOWN 0x0100
-+#define NXS_RELEASED 0x8000
+
-+/* check conditions */
++#define VC_CATEGORY(c) (((c) >> 24) & 0x3F)
++#define VC_COMMAND(c) (((c) >> 16) & 0xFF)
++#define VC_VERSION(c) ((c) & 0xFFF)
+
-+#define NX_ADMIN 0x0001
-+#define NX_WATCH 0x0002
-+#define NX_BLEND 0x0004
-+#define NX_HOSTID 0x0008
-+
-+#define NX_IDENT 0x0010
-+#define NX_EQUIV 0x0020
-+#define NX_PARENT 0x0040
-+#define NX_CHILD 0x0080
++#define VC_CMD(c, i, v) ((((VC_CAT_ ## c) & 0x3F) << 24) \
++ | (((i) & 0xFF) << 16) | ((v) & 0xFFF))
+
-+#define NX_ARG_MASK 0x00F0
++/*
+
-+#define NX_DYNAMIC 0x0100
-+#define NX_STATIC 0x0200
++ Syscall Matrix V2.8
+
-+#define NX_ATR_MASK 0x0F00
++ |VERSION|CREATE |MODIFY |MIGRATE|CONTROL|EXPERIM| |SPECIAL|SPECIAL|
++ |STATS |DESTROY|ALTER |CHANGE |LIMIT |TEST | | | |
++ |INFO |SETUP | |MOVE | | | | | |
++ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
++ SYSTEM |VERSION|VSETUP |VHOST | | | | |DEVICE | |
++ HOST | 00| 01| 02| 03| 04| 05| | 06| 07|
++ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
++ CPU | |VPROC |PROCALT|PROCMIG|PROCTRL| | |SCHED. | |
++ PROCESS| 08| 09| 10| 11| 12| 13| | 14| 15|
++ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
++ MEMORY | | | | |MEMCTRL| | |SWAP | |
++ | 16| 17| 18| 19| 20| 21| | 22| 23|
++ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
++ NETWORK| |VNET |NETALT |NETMIG |NETCTL | | |SERIAL | |
++ | 24| 25| 26| 27| 28| 29| | 30| 31|
++ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
++ DISK | | | |TAGMIG |DLIMIT | | |INODE | |
++ VFS | 32| 33| 34| 35| 36| 37| | 38| 39|
++ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
++ OTHER |VSTAT | | | | | | |VINFO | |
++ | 40| 41| 42| 43| 44| 45| | 46| 47|
++ =======+=======+=======+=======+=======+=======+=======+ +=======+=======+
++ SPECIAL|EVENT | | | |FLAGS | | | | |
++ | 48| 49| 50| 51| 52| 53| | 54| 55|
++ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
++ SPECIAL|DEBUG | | | |RLIMIT |SYSCALL| | |COMPAT |
++ | 56| 57| 58| 59| 60|TEST 61| | 62| 63|
++ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
+
++*/
+
-+extern struct nx_info *lookup_nx_info(int);
++#define VC_CAT_VERSION 0
+
-+extern int get_nid_list(int, unsigned int *, int);
-+extern int nid_is_hashed(nid_t);
++#define VC_CAT_VSETUP 1
++#define VC_CAT_VHOST 2
+
-+extern int nx_migrate_task(struct task_struct *, struct nx_info *);
++#define VC_CAT_DEVICE 6
+
-+extern long vs_net_change(struct nx_info *, unsigned int);
++#define VC_CAT_VPROC 9
++#define VC_CAT_PROCALT 10
++#define VC_CAT_PROCMIG 11
++#define VC_CAT_PROCTRL 12
+
-+struct in_ifaddr;
-+struct net_device;
++#define VC_CAT_SCHED 14
++#define VC_CAT_MEMCTRL 20
+
-+#ifdef CONFIG_INET
-+int ifa_in_nx_info(struct in_ifaddr *, struct nx_info *);
-+int dev_in_nx_info(struct net_device *, struct nx_info *);
++#define VC_CAT_VNET 25
++#define VC_CAT_NETALT 26
++#define VC_CAT_NETMIG 27
++#define VC_CAT_NETCTRL 28
+
-+#else /* CONFIG_INET */
-+static inline
-+int ifa_in_nx_info(struct in_ifaddr *a, struct nx_info *n)
-+{
-+ return 1;
-+}
++#define VC_CAT_TAGMIG 35
++#define VC_CAT_DLIMIT 36
++#define VC_CAT_INODE 38
+
-+static inline
-+int dev_in_nx_info(struct net_device *d, struct nx_info *n)
-+{
-+ return 1;
-+}
-+#endif /* CONFIG_INET */
++#define VC_CAT_VSTAT 40
++#define VC_CAT_VINFO 46
++#define VC_CAT_EVENT 48
+
-+struct sock;
++#define VC_CAT_FLAGS 52
++#define VC_CAT_DEBUG 56
++#define VC_CAT_RLIMIT 60
+
-+#ifdef CONFIG_INET
-+int nx_addr_conflict(struct nx_info *, uint32_t, const struct sock *);
-+#else /* CONFIG_INET */
-+static inline
-+int nx_addr_conflict(struct nx_info *n, uint32_t a, const struct sock *s)
-+{
-+ return 1;
-+}
-+#endif /* CONFIG_INET */
++#define VC_CAT_SYSTEST 61
++#define VC_CAT_COMPAT 63
++
++/* query version */
++
++#define VCMD_get_version VC_CMD(VERSION, 0, 0)
++#define VCMD_get_vci VC_CMD(VERSION, 1, 0)
+
++
++#ifdef __KERNEL__
++
++#include <linux/errno.h>
++
++
++#else /* __KERNEL__ */
++#define __user
+#endif /* __KERNEL__ */
-+#else /* _VX_NETWORK_H */
-+#warning duplicate inclusion
-+#endif /* _VX_NETWORK_H */
---- linux-2.6.22/include/linux/vserver/network_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/network_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,89 @@
-+#ifndef _VX_NETWORK_CMD_H
-+#define _VX_NETWORK_CMD_H
++
++#endif /* _VX_SWITCH_H */
+--- linux-2.6.25.4/include/linux/vserver/tag_cmd.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/tag_cmd.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,22 @@
++#ifndef _VX_TAG_CMD_H
++#define _VX_TAG_CMD_H
+
+
+/* vinfo commands */
+
-+#define VCMD_task_nid VC_CMD(VINFO, 2, 0)
++#define VCMD_task_tag VC_CMD(VINFO, 3, 0)
+
+#ifdef __KERNEL__
-+extern int vc_task_nid(uint32_t, void __user *);
++extern int vc_task_tag(uint32_t);
+
+#endif /* __KERNEL__ */
+
-+#define VCMD_nx_info VC_CMD(VINFO, 6, 0)
++/* context commands */
+
-+struct vcmd_nx_info_v0 {
-+ uint32_t nid;
-+ /* more to come */
-+};
++#define VCMD_tag_migrate VC_CMD(TAGMIG, 1, 0)
+
+#ifdef __KERNEL__
-+extern int vc_nx_info(struct nx_info *, void __user *);
++extern int vc_tag_migrate(uint32_t);
+
+#endif /* __KERNEL__ */
++#endif /* _VX_TAG_CMD_H */
+--- linux-2.6.25.4/include/linux/vserver/tag.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vserver/tag.h 2008-04-21 13:53:47.000000000 -0400
+@@ -0,0 +1,143 @@
++#ifndef _DX_TAG_H
++#define _DX_TAG_H
+
-+#define VCMD_net_create_v0 VC_CMD(VNET, 1, 0)
-+#define VCMD_net_create VC_CMD(VNET, 1, 1)
-+
-+struct vcmd_net_create {
-+ uint64_t flagword;
-+};
-+
-+#define VCMD_net_migrate VC_CMD(NETMIG, 1, 0)
++#include <linux/types.h>
+
-+#define VCMD_net_add VC_CMD(NETALT, 1, 0)
-+#define VCMD_net_remove VC_CMD(NETALT, 2, 0)
+
-+struct vcmd_net_addr_v0 {
-+ uint16_t type;
-+ uint16_t count;
-+ uint32_t ip[4];
-+ uint32_t mask[4];
-+ /* more to come */
-+};
++#define DX_TAG(in) (IS_TAGGED(in))
+
+
-+#ifdef __KERNEL__
-+extern int vc_net_create(uint32_t, void __user *);
-+extern int vc_net_migrate(struct nx_info *, void __user *);
++#ifdef CONFIG_DX_TAG_NFSD
++#define DX_TAG_NFSD 1
++#else
++#define DX_TAG_NFSD 0
++#endif
+
-+extern int vc_net_add(struct nx_info *, void __user *);
-+extern int vc_net_remove(struct nx_info *, void __user *);
+
-+#endif /* __KERNEL__ */
++#ifdef CONFIG_TAGGING_NONE
+
++#define MAX_UID 0xFFFFFFFF
++#define MAX_GID 0xFFFFFFFF
+
-+/* flag commands */
++#define INOTAG_TAG(cond, uid, gid, tag) (0)
+
-+#define VCMD_get_nflags VC_CMD(FLAGS, 5, 0)
-+#define VCMD_set_nflags VC_CMD(FLAGS, 6, 0)
++#define TAGINO_UID(cond, uid, tag) (uid)
++#define TAGINO_GID(cond, gid, tag) (gid)
+
-+struct vcmd_net_flags_v0 {
-+ uint64_t flagword;
-+ uint64_t mask;
-+};
++#endif
+
-+#ifdef __KERNEL__
-+extern int vc_get_nflags(struct nx_info *, void __user *);
-+extern int vc_set_nflags(struct nx_info *, void __user *);
+
-+#endif /* __KERNEL__ */
++#ifdef CONFIG_TAGGING_GID16
+
++#define MAX_UID 0xFFFFFFFF
++#define MAX_GID 0x0000FFFF
+
-+/* network caps commands */
++#define INOTAG_TAG(cond, uid, gid, tag) \
++ ((cond) ? (((gid) >> 16) & 0xFFFF) : 0)
+
-+#define VCMD_get_ncaps VC_CMD(FLAGS, 7, 0)
-+#define VCMD_set_ncaps VC_CMD(FLAGS, 8, 0)
++#define TAGINO_UID(cond, uid, tag) (uid)
++#define TAGINO_GID(cond, gid, tag) \
++ ((cond) ? (((gid) & 0xFFFF) | ((tag) << 16)) : (gid))
+
-+struct vcmd_net_caps_v0 {
-+ uint64_t ncaps;
-+ uint64_t cmask;
-+};
++#endif
+
-+#ifdef __KERNEL__
-+extern int vc_get_ncaps(struct nx_info *, void __user *);
-+extern int vc_set_ncaps(struct nx_info *, void __user *);
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_CONTEXT_CMD_H */
---- linux-2.6.22/include/linux/vserver/percpu.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/percpu.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,14 @@
-+#ifndef _VX_PERCPU_H
-+#define _VX_PERCPU_H
++#ifdef CONFIG_TAGGING_ID24
+
-+#include "cvirt_def.h"
-+#include "sched_def.h"
++#define MAX_UID 0x00FFFFFF
++#define MAX_GID 0x00FFFFFF
+
-+struct _vx_percpu {
-+ struct _vx_cvirt_pc cvirt;
-+ struct _vx_sched_pc sched;
-+};
++#define INOTAG_TAG(cond, uid, gid, tag) \
++ ((cond) ? ((((uid) >> 16) & 0xFF00) | (((gid) >> 24) & 0xFF)) : 0)
+
-+#define PERCPU_PERCTX (sizeof(struct _vx_percpu))
++#define TAGINO_UID(cond, uid, tag) \
++ ((cond) ? (((uid) & 0xFFFFFF) | (((tag) & 0xFF00) << 16)) : (uid))
++#define TAGINO_GID(cond, gid, tag) \
++ ((cond) ? (((gid) & 0xFFFFFF) | (((tag) & 0x00FF) << 24)) : (gid))
+
-+#endif /* _VX_PERCPU_H */
---- linux-2.6.22/include/linux/vserver/sched.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/sched.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,26 @@
-+#ifndef _VX_SCHED_H
-+#define _VX_SCHED_H
++#endif
+
+
-+#ifdef __KERNEL__
++#ifdef CONFIG_TAGGING_UID16
+
-+struct timespec;
++#define MAX_UID 0x0000FFFF
++#define MAX_GID 0xFFFFFFFF
+
-+void vx_vsi_uptime(struct timespec *, struct timespec *);
++#define INOTAG_TAG(cond, uid, gid, tag) \
++ ((cond) ? (((uid) >> 16) & 0xFFFF) : 0)
+
++#define TAGINO_UID(cond, uid, tag) \
++ ((cond) ? (((uid) & 0xFFFF) | ((tag) << 16)) : (uid))
++#define TAGINO_GID(cond, gid, tag) (gid)
+
-+struct vx_info;
++#endif
+
-+void vx_update_load(struct vx_info *);
+
++#ifdef CONFIG_TAGGING_INTERN
+
-+int vx_tokens_recalc(struct _vx_sched_pc *,
-+ unsigned long *, unsigned long *, int [2]);
++#define MAX_UID 0xFFFFFFFF
++#define MAX_GID 0xFFFFFFFF
+
-+void vx_update_sched_param(struct _vx_sched *sched,
-+ struct _vx_sched_pc *sched_pc);
++#define INOTAG_TAG(cond, uid, gid, tag) \
++ ((cond) ? (tag) : 0)
+
-+#endif /* __KERNEL__ */
-+#else /* _VX_SCHED_H */
-+#warning duplicate inclusion
-+#endif /* _VX_SCHED_H */
---- linux-2.6.22/include/linux/vserver/sched_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/sched_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,108 @@
-+#ifndef _VX_SCHED_CMD_H
-+#define _VX_SCHED_CMD_H
++#define TAGINO_UID(cond, uid, tag) (uid)
++#define TAGINO_GID(cond, gid, tag) (gid)
+
++#endif
+
-+/* sched vserver commands */
+
-+#define VCMD_set_sched_v2 VC_CMD(SCHED, 1, 2)
-+#define VCMD_set_sched_v3 VC_CMD(SCHED, 1, 3)
-+#define VCMD_set_sched_v4 VC_CMD(SCHED, 1, 4)
++#ifndef CONFIG_TAGGING_NONE
++#define dx_current_fstag(sb) \
++ ((sb)->s_flags & MS_TAGGED ? dx_current_tag() : 0)
++#else
++#define dx_current_fstag(sb) (0)
++#endif
+
-+struct vcmd_set_sched_v2 {
-+ int32_t fill_rate;
-+ int32_t interval;
-+ int32_t tokens;
-+ int32_t tokens_min;
-+ int32_t tokens_max;
-+ uint64_t cpu_mask;
-+};
++#ifndef CONFIG_TAGGING_INTERN
++#define TAGINO_TAG(cond, tag) (0)
++#else
++#define TAGINO_TAG(cond, tag) ((cond) ? (tag) : 0)
++#endif
+
-+struct vcmd_set_sched_v3 {
-+ uint32_t set_mask;
-+ int32_t fill_rate;
-+ int32_t interval;
-+ int32_t tokens;
-+ int32_t tokens_min;
-+ int32_t tokens_max;
-+ int32_t priority_bias;
-+};
++#define INOTAG_UID(cond, uid, gid) \
++ ((cond) ? ((uid) & MAX_UID) : (uid))
++#define INOTAG_GID(cond, uid, gid) \
++ ((cond) ? ((gid) & MAX_GID) : (gid))
+
-+struct vcmd_set_sched_v4 {
-+ uint32_t set_mask;
-+ int32_t fill_rate;
-+ int32_t interval;
-+ int32_t tokens;
-+ int32_t tokens_min;
-+ int32_t tokens_max;
-+ int32_t prio_bias;
-+ int32_t cpu_id;
-+ int32_t bucket_id;
-+};
+
-+#define VCMD_set_sched VC_CMD(SCHED, 1, 5)
-+#define VCMD_get_sched VC_CMD(SCHED, 2, 5)
++static inline uid_t dx_map_uid(uid_t uid)
++{
++ if ((uid > MAX_UID) && (uid != -1))
++ uid = -2;
++ return (uid & MAX_UID);
++}
+
-+struct vcmd_sched_v5 {
-+ uint32_t mask;
-+ int32_t cpu_id;
-+ int32_t bucket_id;
-+ int32_t fill_rate[2];
-+ int32_t interval[2];
-+ int32_t tokens;
-+ int32_t tokens_min;
-+ int32_t tokens_max;
-+ int32_t prio_bias;
++static inline gid_t dx_map_gid(gid_t gid)
++{
++ if ((gid > MAX_GID) && (gid != -1))
++ gid = -2;
++ return (gid & MAX_GID);
++}
++
++struct peer_tag {
++ int32_t xid;
++ int32_t nid;
+};
+
-+#define VXSM_FILL_RATE 0x0001
-+#define VXSM_INTERVAL 0x0002
-+#define VXSM_FILL_RATE2 0x0004
-+#define VXSM_INTERVAL2 0x0008
-+#define VXSM_TOKENS 0x0010
-+#define VXSM_TOKENS_MIN 0x0020
-+#define VXSM_TOKENS_MAX 0x0040
-+#define VXSM_PRIO_BIAS 0x0100
++#define dx_notagcheck(nd) ((nd) && (nd)->path.mnt && \
++ ((nd)->path.mnt->mnt_flags & MNT_NOTAGCHECK))
+
-+#define VXSM_IDLE_TIME 0x0200
-+#define VXSM_FORCE 0x0400
++int dx_parse_tag(char *string, tag_t *tag, int remove);
+
-+#define VXSM_V3_MASK 0x0173
-+#define VXSM_SET_MASK 0x01FF
++#ifdef CONFIG_PROPAGATE
+
-+#define VXSM_CPU_ID 0x1000
-+#define VXSM_BUCKET_ID 0x2000
++void __dx_propagate_tag(struct nameidata *nd, struct inode *inode);
+
-+#define VXSM_MSEC 0x4000
++#define dx_propagate_tag(n, i) __dx_propagate_tag(n, i)
+
-+#define SCHED_KEEP (-2) /* only for v2 */
++#else
++#define dx_propagate_tag(n, i) do { } while (0)
++#endif
+
-+#ifdef __KERNEL__
++#endif /* _DX_TAG_H */
+--- linux-2.6.25.4/include/linux/vs_inet6.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_inet6.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,229 @@
++#ifndef _VS_INET6_H
++#define _VS_INET6_H
+
-+#include <linux/compiler.h>
++#include "vserver/base.h"
++#include "vserver/network.h"
++#include "vserver/debug.h"
+
-+extern int vc_set_sched_v2(struct vx_info *, void __user *);
-+extern int vc_set_sched_v3(struct vx_info *, void __user *);
-+extern int vc_set_sched_v4(struct vx_info *, void __user *);
-+extern int vc_set_sched(struct vx_info *, void __user *);
-+extern int vc_get_sched(struct vx_info *, void __user *);
++#include <net/ipv6.h>
+
-+#endif /* __KERNEL__ */
++#define NXAV6(a) NIP6((a)->ip), NIP6((a)->mask), (a)->prefix, (a)->type
++#define NXAV6_FMT "[" NIP6_FMT "/" NIP6_FMT "/%d:%04x]"
+
-+#define VCMD_sched_info VC_CMD(SCHED, 3, 0)
+
-+struct vcmd_sched_info {
-+ int32_t cpu_id;
-+ int32_t bucket_id;
-+ uint64_t user_msec;
-+ uint64_t sys_msec;
-+ uint64_t hold_msec;
-+ uint32_t token_usec;
-+ int32_t vavavoom;
-+};
++#ifdef CONFIG_IPV6
+
-+#ifdef __KERNEL__
++static inline
++int v6_addr_match(struct nx_addr_v6 *nxa,
++ const struct in6_addr *addr, uint16_t mask)
++{
++ switch (nxa->type & mask) {
++ case NXA_TYPE_MASK:
++ return ipv6_masked_addr_cmp(&nxa->ip, &nxa->mask, addr);
++ case NXA_TYPE_ADDR:
++ return ipv6_addr_equal(&nxa->ip, addr);
++ case NXA_TYPE_ANY:
++ return 1;
++ default:
++ return 0;
++ }
++}
+
-+extern int vc_sched_info(struct vx_info *, void __user *);
++static inline
++int v6_addr_in_nx_info(struct nx_info *nxi,
++ const struct in6_addr *addr, uint16_t mask)
++{
++ struct nx_addr_v6 *nxa;
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_SCHED_CMD_H */
---- linux-2.6.22/include/linux/vserver/sched_def.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/sched_def.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,68 @@
-+#ifndef _VX_SCHED_DEF_H
-+#define _VX_SCHED_DEF_H
++ if (!nxi)
++ return 1;
++ for (nxa = &nxi->v6; nxa; nxa = nxa->next)
++ if (v6_addr_match(nxa, addr, mask))
++ return 1;
++ return 0;
++}
+
-+#include <linux/spinlock.h>
-+#include <linux/jiffies.h>
-+#include <linux/cpumask.h>
-+#include <asm/atomic.h>
-+#include <asm/param.h>
++static inline
++int v6_nx_addr_match(struct nx_addr_v6 *nxa, struct nx_addr_v6 *addr, uint16_t mask)
++{
++ /* FIXME: needs full range checks */
++ return v6_addr_match(nxa, &addr->ip, mask);
++}
+
++static inline
++int v6_nx_addr_in_nx_info(struct nx_info *nxi, struct nx_addr_v6 *nxa, uint16_t mask)
++{
++ struct nx_addr_v6 *ptr;
+
-+/* context sub struct */
++ for (ptr = &nxi->v6; ptr; ptr = ptr->next)
++ if (v6_nx_addr_match(ptr, nxa, mask))
++ return 1;
++ return 0;
++}
+
-+struct _vx_sched {
-+ spinlock_t tokens_lock; /* lock for token bucket */
+
-+ int tokens; /* number of CPU tokens */
-+ int fill_rate[2]; /* Fill rate: add X tokens... */
-+ int interval[2]; /* Divisor: per Y jiffies */
-+ int tokens_min; /* Limit: minimum for unhold */
-+ int tokens_max; /* Limit: no more than N tokens */
++/*
++ * Check if a given address matches for a socket
++ *
++ * nxi: the socket's nx_info if any
++ * addr: to be verified address
++ */
++static inline
++int v6_sock_addr_match (
++ struct nx_info *nxi,
++ struct inet_sock *inet,
++ struct in6_addr *addr)
++{
++ struct sock *sk = &inet->sk;
++ struct in6_addr *saddr = inet6_rcv_saddr(sk);
+
-+ int prio_bias; /* bias offset for priority */
++ if (!ipv6_addr_any(addr) &&
++ ipv6_addr_equal(saddr, addr))
++ return 1;
++ if (ipv6_addr_any(saddr))
++ return v6_addr_in_nx_info(nxi, addr, -1);
++ return 0;
++}
+
-+ unsigned update_mask; /* which features should be updated */
-+ cpumask_t update; /* CPUs which should update */
-+};
++/*
++ * check if address is covered by socket
++ *
++ * sk: the socket to check against
++ * addr: the address in question (must be != 0)
++ */
+
-+struct _vx_sched_pc {
-+ int tokens; /* number of CPU tokens */
-+ int flags; /* bucket flags */
++static inline
++int __v6_addr_match_socket(const struct sock *sk, struct nx_addr_v6 *nxa)
++{
++ struct nx_info *nxi = sk->sk_nx_info;
++ struct in6_addr *saddr = inet6_rcv_saddr(sk);
+
-+ int fill_rate[2]; /* Fill rate: add X tokens... */
-+ int interval[2]; /* Divisor: per Y jiffies */
-+ int tokens_min; /* Limit: minimum for unhold */
-+ int tokens_max; /* Limit: no more than N tokens */
++ vxdprintk(VXD_CBIT(net, 5),
++ "__v6_addr_in_socket(%p," NXAV6_FMT ") %p:" NIP6_FMT " %p;%lx",
++ sk, NXAV6(nxa), nxi, NIP6(*saddr), sk->sk_socket,
++ (sk->sk_socket?sk->sk_socket->flags:0));
+
-+ int prio_bias; /* bias offset for priority */
-+ int vavavoom; /* last calculated vavavoom */
++ if (!ipv6_addr_any(saddr)) { /* direct address match */
++ return v6_addr_match(nxa, saddr, -1);
++ } else if (nxi) { /* match against nx_info */
++ return v6_nx_addr_in_nx_info(nxi, nxa, -1);
++ } else { /* unrestricted any socket */
++ return 1;
++ }
++}
+
-+ unsigned long norm_time; /* last time accounted */
-+ unsigned long idle_time; /* non linear time for fair sched */
-+ unsigned long token_time; /* token time for accounting */
-+ unsigned long onhold; /* jiffies when put on hold */
+
-+ uint64_t user_ticks; /* token tick events */
-+ uint64_t sys_ticks; /* token tick events */
-+ uint64_t hold_ticks; /* token ticks paused */
-+};
++/* inet related checks and helpers */
+
+
-+#define VXSF_ONHOLD 0x0001
-+#define VXSF_IDLE_TIME 0x0100
++struct in_ifaddr;
++struct net_device;
++struct sock;
+
-+#ifdef CONFIG_VSERVER_DEBUG
+
-+static inline void __dump_vx_sched(struct _vx_sched *sched)
-+{
-+ printk("\t_vx_sched:\n");
-+ printk("\t tokens: %4d/%4d, %4d/%4d, %4d, %4d\n",
-+ sched->fill_rate[0], sched->interval[0],
-+ sched->fill_rate[1], sched->interval[1],
-+ sched->tokens_min, sched->tokens_max);
-+ printk("\t priority = %4d\n", sched->prio_bias);
-+}
++#include <linux/netdevice.h>
++#include <linux/inetdevice.h>
++#include <net/inet_timewait_sock.h>
+
-+#endif
+
-+#endif /* _VX_SCHED_DEF_H */
---- linux-2.6.22/include/linux/vserver/signal.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/signal.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,14 @@
-+#ifndef _VX_SIGNAL_H
-+#define _VX_SIGNAL_H
++int dev_in_nx_info(struct net_device *, struct nx_info *);
++int v6_dev_in_nx_info(struct net_device *, struct nx_info *);
++int nx_v6_addr_conflict(struct nx_info *, struct nx_info *);
+
+
-+#ifdef __KERNEL__
+
-+struct vx_info;
++static inline
++int v6_ifa_in_nx_info(struct inet6_ifaddr *ifa, struct nx_info *nxi)
++{
++ if (!nxi)
++ return 1;
++ if (!ifa)
++ return 0;
++ return v6_addr_in_nx_info(nxi, &ifa->addr, -1);
++}
+
-+int vx_info_kill(struct vx_info *, int, int);
++static inline
++int nx_v6_ifa_visible(struct nx_info *nxi, struct inet6_ifaddr *ifa)
++{
++ if (!nx_info_flags(nxi, NXF_HIDE_NETIF, 0))
++ return 1;
++ if (v6_ifa_in_nx_info(ifa, nxi))
++ return 1;
++ return 0;
++}
+
-+#endif /* __KERNEL__ */
-+#else /* _VX_SIGNAL_H */
-+#warning duplicate inclusion
-+#endif /* _VX_SIGNAL_H */
---- linux-2.6.22/include/linux/vserver/signal_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/signal_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,43 @@
-+#ifndef _VX_SIGNAL_CMD_H
-+#define _VX_SIGNAL_CMD_H
+
++struct nx_v6_sock_addr {
++ struct in6_addr saddr; /* Address used for validation */
++ struct in6_addr baddr; /* Address used for socket bind */
++};
+
-+/* signalling vserver commands */
++static inline
++int v6_map_sock_addr(struct inet_sock *inet, struct sockaddr_in6 *addr,
++ struct nx_v6_sock_addr *nsa)
++{
++ // struct sock *sk = &inet->sk;
++ // struct nx_info *nxi = sk->sk_nx_info;
++ struct in6_addr saddr = addr->sin6_addr;
++ struct in6_addr baddr = saddr;
+
-+#define VCMD_ctx_kill VC_CMD(PROCTRL, 1, 0)
-+#define VCMD_wait_exit VC_CMD(EVENT, 99, 0)
-+
-+struct vcmd_ctx_kill_v0 {
-+ int32_t pid;
-+ int32_t sig;
-+};
-+
-+struct vcmd_wait_exit_v0 {
-+ int32_t reboot_cmd;
-+ int32_t exit_code;
-+};
-+
-+#ifdef __KERNEL__
-+
-+extern int vc_ctx_kill(struct vx_info *, void __user *);
-+extern int vc_wait_exit(struct vx_info *, void __user *);
++ nsa->saddr = saddr;
++ nsa->baddr = baddr;
++ return 0;
++}
+
-+#endif /* __KERNEL__ */
++static inline
++void v6_set_sock_addr(struct inet_sock *inet, struct nx_v6_sock_addr *nsa)
++{
++ // struct sock *sk = &inet->sk;
++ // struct in6_addr *saddr = inet6_rcv_saddr(sk);
+
-+/* process alteration commands */
++ // *saddr = nsa->baddr;
++ // inet->saddr = nsa->baddr;
++}
+
-+#define VCMD_get_pflags VC_CMD(PROCALT, 5, 0)
-+#define VCMD_set_pflags VC_CMD(PROCALT, 6, 0)
++static inline
++int nx_info_has_v6(struct nx_info *nxi)
++{
++ if (!nxi)
++ return 1;
++ if (NX_IPV6(nxi))
++ return 1;
++ return 0;
++}
+
-+struct vcmd_pflags_v0 {
-+ uint32_t flagword;
-+ uint32_t mask;
-+};
++#else /* CONFIG_IPV6 */
+
-+#ifdef __KERNEL__
++static inline
++int nx_v6_dev_visible(struct nx_info *n, struct net_device *d)
++{
++ return 1;
++}
+
-+extern int vc_get_pflags(uint32_t pid, void __user *);
-+extern int vc_set_pflags(uint32_t pid, void __user *);
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_SIGNAL_CMD_H */
---- linux-2.6.22/include/linux/vserver/space.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/space.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,13 @@
-+#ifndef _VX_SPACE_H
-+#define _VX_SPACE_H
++static inline
++int nx_v6_addr_conflict(struct nx_info *n, uint32_t a, const struct sock *s)
++{
++ return 1;
++}
+
++static inline
++int v6_ifa_in_nx_info(struct in_ifaddr *a, struct nx_info *n)
++{
++ return 1;
++}
+
-+#include <linux/types.h>
++static inline
++int nx_info_has_v6(struct nx_info *nxi)
++{
++ return 0;
++}
+
-+struct vx_info;
++#endif /* CONFIG_IPV6 */
+
-+int vx_set_space(struct vx_info *vxi, unsigned long mask);
++#define current_nx_info_has_v6() \
++ nx_info_has_v6(current_nx_info())
+
-+#else /* _VX_SPACE_H */
++#else
+#warning duplicate inclusion
-+#endif /* _VX_SPACE_H */
---- linux-2.6.22/include/linux/vserver/space_cmd.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/space_cmd.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,26 @@
-+#ifndef _VX_SPACE_CMD_H
-+#define _VX_SPACE_CMD_H
++#endif
+--- linux-2.6.25.4/include/linux/vs_inet.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_inet.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,342 @@
++#ifndef _VS_INET_H
++#define _VS_INET_H
+
++#include "vserver/base.h"
++#include "vserver/network.h"
++#include "vserver/debug.h"
+
-+#define VCMD_enter_space_v0 VC_CMD(PROCALT, 1, 0)
-+#define VCMD_enter_space VC_CMD(PROCALT, 1, 1)
++#define IPI_LOOPBACK htonl(INADDR_LOOPBACK)
+
-+#define VCMD_set_space_v0 VC_CMD(PROCALT, 3, 0)
-+#define VCMD_set_space VC_CMD(PROCALT, 3, 1)
++#define NXAV4(a) NIPQUAD((a)->ip[0]), NIPQUAD((a)->ip[1]), \
++ NIPQUAD((a)->mask), (a)->type
++#define NXAV4_FMT "[" NIPQUAD_FMT "-" NIPQUAD_FMT "/" NIPQUAD_FMT ":%04x]"
+
-+#define VCMD_get_space_mask VC_CMD(PROCALT, 4, 0)
+
++static inline
++int v4_addr_match(struct nx_addr_v4 *nxa, __be32 addr, uint16_t tmask)
++{
++ __be32 ip = nxa->ip[0].s_addr;
++ __be32 mask = nxa->mask.s_addr;
++ __be32 bcast = ip | ~mask;
++ int ret = 0;
+
-+struct vcmd_space_mask {
-+ uint64_t mask;
-+};
++ switch (nxa->type & tmask) {
++ case NXA_TYPE_MASK:
++ ret = (ip == (addr & mask));
++ break;
++ case NXA_TYPE_ADDR:
++ ret = 3;
++ if (addr == ip)
++ break;
++ /* fall through to broadcast */
++ case NXA_MOD_BCAST:
++ ret = ((tmask & NXA_MOD_BCAST) && (addr == bcast));
++ break;
++ case NXA_TYPE_RANGE:
++ ret = ((nxa->ip[0].s_addr <= addr) &&
++ (nxa->ip[1].s_addr > addr));
++ break;
++ case NXA_TYPE_ANY:
++ ret = 2;
++ break;
++ }
+
++ vxdprintk(VXD_CBIT(net, 0),
++ "v4_addr_match(%p" NXAV4_FMT "," NIPQUAD_FMT ",%04x) = %d",
++ nxa, NXAV4(nxa), NIPQUAD(addr), tmask, ret);
++ return ret;
++}
+
-+#ifdef __KERNEL__
++static inline
++int v4_addr_in_nx_info(struct nx_info *nxi, __be32 addr, uint16_t tmask)
++{
++ struct nx_addr_v4 *nxa;
++ int ret = 1;
+
-+extern int vc_enter_space(struct vx_info *, void __user *);
-+extern int vc_set_space(struct vx_info *, void __user *);
-+extern int vc_get_space_mask(struct vx_info *, void __user *);
++ if (!nxi)
++ goto out;
+
-+#endif /* __KERNEL__ */
-+#endif /* _VX_SPACE_CMD_H */
---- linux-2.6.22/include/linux/vserver/switch.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/switch.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,95 @@
-+#ifndef _VX_SWITCH_H
-+#define _VX_SWITCH_H
++ ret = 2;
++ /* allow 127.0.0.1 when remapping lback */
++ if ((tmask & NXA_LOOPBACK) &&
++ (addr == IPI_LOOPBACK) &&
++ nx_info_flags(nxi, NXF_LBACK_REMAP, 0))
++ goto out;
++ ret = 3;
++ /* check for lback address */
++ if ((tmask & NXA_MOD_LBACK) &&
++ (nxi->v4_lback.s_addr == addr))
++ goto out;
++ ret = 4;
++ /* check for broadcast address */
++ if ((tmask & NXA_MOD_BCAST) &&
++ (nxi->v4_bcast.s_addr == addr))
++ goto out;
++ ret = 5;
++ /* check for v4 addresses */
++ for (nxa = &nxi->v4; nxa; nxa = nxa->next)
++ if (v4_addr_match(nxa, addr, tmask))
++ goto out;
++ ret = 0;
++out:
++ vxdprintk(VXD_CBIT(net, 0),
++ "v4_addr_in_nx_info(%p[#%u]," NIPQUAD_FMT ",%04x) = %d",
++ nxi, nxi ? nxi->nx_id : 0, NIPQUAD(addr), tmask, ret);
++ return ret;
++}
+
-+#include <linux/types.h>
++static inline
++int v4_nx_addr_match(struct nx_addr_v4 *nxa, struct nx_addr_v4 *addr, uint16_t mask)
++{
++ /* FIXME: needs full range checks */
++ return v4_addr_match(nxa, addr->ip[0].s_addr, mask);
++}
+
++static inline
++int v4_nx_addr_in_nx_info(struct nx_info *nxi, struct nx_addr_v4 *nxa, uint16_t mask)
++{
++ struct nx_addr_v4 *ptr;
+
-+#define VC_CATEGORY(c) (((c) >> 24) & 0x3F)
-+#define VC_COMMAND(c) (((c) >> 16) & 0xFF)
-+#define VC_VERSION(c) ((c) & 0xFFF)
++ for (ptr = &nxi->v4; ptr; ptr = ptr->next)
++ if (v4_nx_addr_match(ptr, nxa, mask))
++ return 1;
++ return 0;
++}
+
-+#define VC_CMD(c, i, v) ((((VC_CAT_ ## c) & 0x3F) << 24) \
-+ | (((i) & 0xFF) << 16) | ((v) & 0xFFF))
++#include <net/inet_sock.h>
+
+/*
++ * Check if a given address matches for a socket
++ *
++ * nxi: the socket's nx_info if any
++ * addr: to be verified address
++ */
++static inline
++int v4_sock_addr_match (
++ struct nx_info *nxi,
++ struct inet_sock *inet,
++ __be32 addr)
++{
++ __be32 saddr = inet->rcv_saddr;
++ __be32 bcast = nxi ? nxi->v4_bcast.s_addr : INADDR_BROADCAST;
+
-+ Syscall Matrix V2.8
-+
-+ |VERSION|CREATE |MODIFY |MIGRATE|CONTROL|EXPERIM| |SPECIAL|SPECIAL|
-+ |STATS |DESTROY|ALTER |CHANGE |LIMIT |TEST | | | |
-+ |INFO |SETUP | |MOVE | | | | | |
-+ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
-+ SYSTEM |VERSION|VSETUP |VHOST | | | | |DEVICES| |
-+ HOST | 00| 01| 02| 03| 04| 05| | 06| 07|
-+ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
-+ CPU | |VPROC |PROCALT|PROCMIG|PROCTRL| | |SCHED. | |
-+ PROCESS| 08| 09| 10| 11| 12| 13| | 14| 15|
-+ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
-+ MEMORY | | | | | | | |SWAP | |
-+ | 16| 17| 18| 19| 20| 21| | 22| 23|
-+ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
-+ NETWORK| |VNET |NETALT |NETMIG |NETCTL | | |SERIAL | |
-+ | 24| 25| 26| 27| 28| 29| | 30| 31|
-+ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
-+ DISK | | | | |DLIMIT | | |INODE | |
-+ VFS | 32| 33| 34| 35| 36| 37| | 38| 39|
-+ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
-+ OTHER |VSTAT | | | | | | |VINFO | |
-+ | 40| 41| 42| 43| 44| 45| | 46| 47|
-+ =======+=======+=======+=======+=======+=======+=======+ +=======+=======+
-+ SPECIAL|EVENT | | | |FLAGS | | | | |
-+ | 48| 49| 50| 51| 52| 53| | 54| 55|
-+ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
-+ SPECIAL|DEBUG | | | |RLIMIT |SYSCALL| | |COMPAT |
-+ | 56| 57| 58| 59| 60|TEST 61| | 62| 63|
-+ -------+-------+-------+-------+-------+-------+-------+ +-------+-------+
++ if (addr && (saddr == addr || bcast == addr))
++ return 1;
++ if (!saddr)
++ return v4_addr_in_nx_info(nxi, addr, NXA_MASK_BIND);
++ return 0;
++}
+
-+*/
+
-+#define VC_CAT_VERSION 0
++/* inet related checks and helpers */
+
-+#define VC_CAT_VSETUP 1
-+#define VC_CAT_VHOST 2
+
-+#define VC_CAT_VPROC 9
-+#define VC_CAT_PROCALT 10
-+#define VC_CAT_PROCMIG 11
-+#define VC_CAT_PROCTRL 12
++struct in_ifaddr;
++struct net_device;
++struct sock;
+
-+#define VC_CAT_SCHED 14
++#ifdef CONFIG_INET
+
-+#define VC_CAT_VNET 25
-+#define VC_CAT_NETALT 26
-+#define VC_CAT_NETMIG 27
-+#define VC_CAT_NETCTRL 28
++#include <linux/netdevice.h>
++#include <linux/inetdevice.h>
++#include <net/inet_sock.h>
++#include <net/inet_timewait_sock.h>
+
-+#define VC_CAT_DLIMIT 36
-+#define VC_CAT_INODE 38
+
-+#define VC_CAT_VSTAT 40
-+#define VC_CAT_VINFO 46
-+#define VC_CAT_EVENT 48
++int dev_in_nx_info(struct net_device *, struct nx_info *);
++int v4_dev_in_nx_info(struct net_device *, struct nx_info *);
++int nx_v4_addr_conflict(struct nx_info *, struct nx_info *);
+
-+#define VC_CAT_FLAGS 52
-+#define VC_CAT_DEBUG 56
-+#define VC_CAT_RLIMIT 60
+
-+#define VC_CAT_SYSTEST 61
-+#define VC_CAT_COMPAT 63
++/*
++ * check if address is covered by socket
++ *
++ * sk: the socket to check against
++ * addr: the address in question (must be != 0)
++ */
+
-+/* query version */
++static inline
++int __v4_addr_match_socket(const struct sock *sk, struct nx_addr_v4 *nxa)
++{
++ struct nx_info *nxi = sk->sk_nx_info;
++ __be32 saddr = inet_rcv_saddr(sk);
+
-+#define VCMD_get_version VC_CMD(VERSION, 0, 0)
-+#define VCMD_get_vci VC_CMD(VERSION, 1, 0)
++ vxdprintk(VXD_CBIT(net, 5),
++ "__v4_addr_in_socket(%p," NXAV4_FMT ") %p:" NIPQUAD_FMT " %p;%lx",
++ sk, NXAV4(nxa), nxi, NIPQUAD(saddr), sk->sk_socket,
++ (sk->sk_socket?sk->sk_socket->flags:0));
+
++ if (saddr) { /* direct address match */
++ return v4_addr_match(nxa, saddr, -1);
++ } else if (nxi) { /* match against nx_info */
++ return v4_nx_addr_in_nx_info(nxi, nxa, -1);
++ } else { /* unrestricted any socket */
++ return 1;
++ }
++}
+
-+#ifdef __KERNEL__
+
-+#include <linux/errno.h>
+
++static inline
++int nx_dev_visible(struct nx_info *nxi, struct net_device *dev)
++{
++ vxdprintk(VXD_CBIT(net, 1), "nx_dev_visible(%p[#%u],%p »%s«) %d",
++ nxi, nxi ? nxi->nx_id : 0, dev, dev->name,
++ nxi ? dev_in_nx_info(dev, nxi) : 0);
+
-+#else /* __KERNEL__ */
-+#define __user
-+#endif /* __KERNEL__ */
++ if (!nx_info_flags(nxi, NXF_HIDE_NETIF, 0))
++ return 1;
++ if (dev_in_nx_info(dev, nxi))
++ return 1;
++ return 0;
++}
+
-+#endif /* _VX_SWITCH_H */
---- linux-2.6.22/include/linux/vserver/tag.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/include/linux/vserver/tag.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,140 @@
-+#ifndef _DX_TAG_H
-+#define _DX_TAG_H
+
++static inline
++int v4_ifa_in_nx_info(struct in_ifaddr *ifa, struct nx_info *nxi)
++{
++ if (!nxi)
++ return 1;
++ if (!ifa)
++ return 0;
++ return v4_addr_in_nx_info(nxi, ifa->ifa_local, NXA_MASK_SHOW);
++}
+
-+#define DX_TAG(in) (IS_TAGGED(in))
++static inline
++int nx_v4_ifa_visible(struct nx_info *nxi, struct in_ifaddr *ifa)
++{
++ vxdprintk(VXD_CBIT(net, 1), "nx_v4_ifa_visible(%p[#%u],%p) %d",
++ nxi, nxi ? nxi->nx_id : 0, ifa,
++ nxi ? v4_ifa_in_nx_info(ifa, nxi) : 0);
+
++ if (!nx_info_flags(nxi, NXF_HIDE_NETIF, 0))
++ return 1;
++ if (v4_ifa_in_nx_info(ifa, nxi))
++ return 1;
++ return 0;
++}
+
-+#ifdef CONFIG_DX_TAG_NFSD
-+#define DX_TAG_NFSD 1
-+#else
-+#define DX_TAG_NFSD 0
-+#endif
+
++struct nx_v4_sock_addr {
++ __be32 saddr; /* Address used for validation */
++ __be32 baddr; /* Address used for socket bind */
++};
+
-+#ifdef CONFIG_TAGGING_NONE
++static inline
++int v4_map_sock_addr(struct inet_sock *inet, struct sockaddr_in *addr,
++ struct nx_v4_sock_addr *nsa)
++{
++ struct sock *sk = &inet->sk;
++ struct nx_info *nxi = sk->sk_nx_info;
++ __be32 saddr = addr->sin_addr.s_addr;
++ __be32 baddr = saddr;
+
-+#define MAX_UID 0xFFFFFFFF
-+#define MAX_GID 0xFFFFFFFF
++ vxdprintk(VXD_CBIT(net, 3),
++ "inet_bind(%p)* %p,%p;%lx " NIPQUAD_FMT,
++ sk, sk->sk_nx_info, sk->sk_socket,
++ (sk->sk_socket ? sk->sk_socket->flags : 0),
++ NIPQUAD(saddr));
+
-+#define INOTAG_TAG(cond, uid, gid, tag) (0)
++ if (nxi) {
++ if (saddr == INADDR_ANY) {
++ if (nx_info_flags(nxi, NXF_SINGLE_IP, 0))
++ baddr = nxi->v4.ip[0].s_addr;
++ } else if (saddr == IPI_LOOPBACK) {
++ if (nx_info_flags(nxi, NXF_LBACK_REMAP, 0))
++ baddr = nxi->v4_lback.s_addr;
++ } else { /* normal address bind */
++ if (!v4_addr_in_nx_info(nxi, saddr, NXA_MASK_BIND))
++ return -EADDRNOTAVAIL;
++ }
++ }
+
-+#define TAGINO_UID(cond, uid, tag) (uid)
-+#define TAGINO_GID(cond, gid, tag) (gid)
++ vxdprintk(VXD_CBIT(net, 3),
++ "inet_bind(%p) " NIPQUAD_FMT ", " NIPQUAD_FMT,
++ sk, NIPQUAD(saddr), NIPQUAD(baddr));
+
-+#endif
++ nsa->saddr = saddr;
++ nsa->baddr = baddr;
++ return 0;
++}
+
++static inline
++void v4_set_sock_addr(struct inet_sock *inet, struct nx_v4_sock_addr *nsa)
++{
++ inet->saddr = nsa->baddr;
++ inet->rcv_saddr = nsa->baddr;
++}
+
-+#ifdef CONFIG_TAGGING_GID16
+
-+#define MAX_UID 0xFFFFFFFF
-+#define MAX_GID 0x0000FFFF
++/*
++ * helper to simplify inet_lookup_listener
++ *
++ * nxi: the socket's nx_info if any
++ * addr: to be verified address
++ * saddr: socket address
++ */
++static inline int v4_inet_addr_match (
++ struct nx_info *nxi,
++ __be32 addr,
++ __be32 saddr)
++{
++ if (addr && (saddr == addr))
++ return 1;
++ if (!saddr)
++ return nxi ? v4_addr_in_nx_info(nxi, addr, NXA_MASK_BIND) : 1;
++ return 0;
++}
+
-+#define INOTAG_TAG(cond, uid, gid, tag) \
-+ ((cond) ? (((gid) >> 16) & 0xFFFF) : 0)
++static inline __be32 nx_map_sock_lback(struct nx_info *nxi, __be32 addr)
++{
++ if (nx_info_flags(nxi, NXF_HIDE_LBACK, 0) &&
++ (addr == nxi->v4_lback.s_addr))
++ return IPI_LOOPBACK;
++ return addr;
++}
+
-+#define TAGINO_UID(cond, uid, tag) (uid)
-+#define TAGINO_GID(cond, gid, tag) \
-+ ((cond) ? (((gid) & 0xFFFF) | ((tag) << 16)) : (gid))
++static inline
++int nx_info_has_v4(struct nx_info *nxi)
++{
++ if (!nxi)
++ return 1;
++ if (NX_IPV4(nxi))
++ return 1;
++ if (nx_info_flags(nxi, NXF_LBACK_REMAP, 0))
++ return 1;
++ return 0;
++}
+
-+#endif
++#else /* CONFIG_INET */
+
++static inline
++int nx_dev_visible(struct nx_info *n, struct net_device *d)
++{
++ return 1;
++}
+
-+#ifdef CONFIG_TAGGING_ID24
++static inline
++int nx_v4_addr_conflict(struct nx_info *n, uint32_t a, const struct sock *s)
++{
++ return 1;
++}
+
-+#define MAX_UID 0x00FFFFFF
-+#define MAX_GID 0x00FFFFFF
++static inline
++int v4_ifa_in_nx_info(struct in_ifaddr *a, struct nx_info *n)
++{
++ return 1;
++}
+
-+#define INOTAG_TAG(cond, uid, gid, tag) \
-+ ((cond) ? ((((uid) >> 16) & 0xFF00) | (((gid) >> 24) & 0xFF)) : 0)
++static inline
++int nx_info_has_v4(struct nx_info *nxi)
++{
++ return 0;
++}
+
-+#define TAGINO_UID(cond, uid, tag) \
-+ ((cond) ? (((uid) & 0xFFFFFF) | (((tag) & 0xFF00) << 16)) : (uid))
-+#define TAGINO_GID(cond, gid, tag) \
-+ ((cond) ? (((gid) & 0xFFFFFF) | (((tag) & 0x00FF) << 24)) : (gid))
++#endif /* CONFIG_INET */
++
++#define current_nx_info_has_v4() \
++ nx_info_has_v4(current_nx_info())
+
++#else
++// #warning duplicate inclusion
+#endif
+--- linux-2.6.25.4/include/linux/vs_limit.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_limit.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,140 @@
++#ifndef _VS_LIMIT_H
++#define _VS_LIMIT_H
+
++#include "vserver/limit.h"
++#include "vserver/base.h"
++#include "vserver/context.h"
++#include "vserver/debug.h"
++#include "vserver/context.h"
++#include "vserver/limit_int.h"
+
-+#ifdef CONFIG_TAGGING_UID16
+
-+#define MAX_UID 0x0000FFFF
-+#define MAX_GID 0xFFFFFFFF
++#define vx_acc_cres(v, d, p, r) \
++ __vx_acc_cres(v, r, d, p, __FILE__, __LINE__)
+
-+#define INOTAG_TAG(cond, uid, gid, tag) \
-+ ((cond) ? (((uid) >> 16) & 0xFFFF) : 0)
++#define vx_acc_cres_cond(x, d, p, r) \
++ __vx_acc_cres(((x) == vx_current_xid()) ? current->vx_info : 0, \
++ r, d, p, __FILE__, __LINE__)
+
-+#define TAGINO_UID(cond, uid, tag) \
-+ ((cond) ? (((uid) & 0xFFFF) | ((tag) << 16)) : (uid))
-+#define TAGINO_GID(cond, gid, tag) (gid)
+
-+#endif
++#define vx_add_cres(v, a, p, r) \
++ __vx_add_cres(v, r, a, p, __FILE__, __LINE__)
++#define vx_sub_cres(v, a, p, r) vx_add_cres(v, -(a), p, r)
+
++#define vx_add_cres_cond(x, a, p, r) \
++ __vx_add_cres(((x) == vx_current_xid()) ? current->vx_info : 0, \
++ r, a, p, __FILE__, __LINE__)
++#define vx_sub_cres_cond(x, a, p, r) vx_add_cres_cond(x, -(a), p, r)
+
-+#ifdef CONFIG_TAGGING_INTERN
+
-+#define MAX_UID 0xFFFFFFFF
-+#define MAX_GID 0xFFFFFFFF
++/* process and file limits */
+
-+#define INOTAG_TAG(cond, uid, gid, tag) \
-+ ((cond) ? (tag) : 0)
++#define vx_nproc_inc(p) \
++ vx_acc_cres((p)->vx_info, 1, p, RLIMIT_NPROC)
+
-+#define TAGINO_UID(cond, uid, tag) (uid)
-+#define TAGINO_GID(cond, gid, tag) (gid)
++#define vx_nproc_dec(p) \
++ vx_acc_cres((p)->vx_info,-1, p, RLIMIT_NPROC)
+
-+#endif
++#define vx_files_inc(f) \
++ vx_acc_cres_cond((f)->f_xid, 1, f, RLIMIT_NOFILE)
+
++#define vx_files_dec(f) \
++ vx_acc_cres_cond((f)->f_xid,-1, f, RLIMIT_NOFILE)
+
-+#ifndef CONFIG_TAGGING_NONE
-+#define dx_current_fstag(sb) \
-+ ((sb)->s_flags & MS_TAGGED ? dx_current_tag() : 0)
-+#else
-+#define dx_current_fstag(sb) (0)
-+#endif
++#define vx_locks_inc(l) \
++ vx_acc_cres_cond((l)->fl_xid, 1, l, RLIMIT_LOCKS)
+
-+#ifndef CONFIG_TAGGING_INTERN
-+#define TAGINO_TAG(cond, tag) (0)
-+#else
-+#define TAGINO_TAG(cond, tag) ((cond) ? (tag) : 0)
-+#endif
++#define vx_locks_dec(l) \
++ vx_acc_cres_cond((l)->fl_xid,-1, l, RLIMIT_LOCKS)
+
-+#define INOTAG_UID(cond, uid, gid) \
-+ ((cond) ? ((uid) & MAX_UID) : (uid))
-+#define INOTAG_GID(cond, uid, gid) \
-+ ((cond) ? ((gid) & MAX_GID) : (gid))
++#define vx_openfd_inc(f) \
++ vx_acc_cres(current->vx_info, 1, (void *)(long)(f), VLIMIT_OPENFD)
+
++#define vx_openfd_dec(f) \
++ vx_acc_cres(current->vx_info,-1, (void *)(long)(f), VLIMIT_OPENFD)
+
-+static inline uid_t dx_map_uid(uid_t uid)
-+{
-+ if ((uid > MAX_UID) && (uid != -1))
-+ uid = -2;
-+ return (uid & MAX_UID);
-+}
+
-+static inline gid_t dx_map_gid(gid_t gid)
-+{
-+ if ((gid > MAX_GID) && (gid != -1))
-+ gid = -2;
-+ return (gid & MAX_GID);
-+}
++#define vx_cres_avail(v, n, r) \
++ __vx_cres_avail(v, r, n, __FILE__, __LINE__)
+
+
-+#ifdef CONFIG_VSERVER_LEGACY
-+#define FIOC_GETTAG _IOR('x', 1, long)
-+#define FIOC_SETTAG _IOW('x', 2, long)
-+#define FIOC_SETTAGJ _IOW('x', 3, long)
-+#endif
++#define vx_nproc_avail(n) \
++ vx_cres_avail(current->vx_info, n, RLIMIT_NPROC)
+
-+#ifdef CONFIG_PROPAGATE
++#define vx_files_avail(n) \
++ vx_cres_avail(current->vx_info, n, RLIMIT_NOFILE)
+
-+int dx_parse_tag(char *string, tag_t *tag, int remove);
++#define vx_locks_avail(n) \
++ vx_cres_avail(current->vx_info, n, RLIMIT_LOCKS)
+
-+void __dx_propagate_tag(struct nameidata *nd, struct inode *inode);
++#define vx_openfd_avail(n) \
++ vx_cres_avail(current->vx_info, n, VLIMIT_OPENFD)
+
-+#define dx_propagate_tag(n, i) __dx_propagate_tag(n, i)
+
-+#else
-+#define dx_propagate_tag(n, i) do { } while (0)
-+#endif
++/* dentry limits */
+
-+#endif /* _DX_TAG_H */
---- linux-2.6.22/include/net/af_unix.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/net/af_unix.h 2007-06-15 02:37:04 +0200
-@@ -4,6 +4,7 @@
- #include <linux/socket.h>
- #include <linux/un.h>
- #include <linux/mutex.h>
-+#include <linux/vs_base.h>
- #include <net/sock.h>
-
- extern void unix_inflight(struct file *fp);
-@@ -17,9 +18,9 @@ extern spinlock_t unix_table_lock;
-
- extern atomic_t unix_tot_inflight;
-
--static inline struct sock *first_unix_socket(int *i)
-+static inline struct sock *next_unix_socket_table(int *i)
- {
-- for (*i = 0; *i <= UNIX_HASH_SIZE; (*i)++) {
-+ for ((*i)++; *i <= UNIX_HASH_SIZE; (*i)++) {
- if (!hlist_empty(&unix_socket_table[*i]))
- return __sk_head(&unix_socket_table[*i]);
- }
-@@ -28,16 +29,19 @@ static inline struct sock *first_unix_so
-
- static inline struct sock *next_unix_socket(int *i, struct sock *s)
- {
-- struct sock *next = sk_next(s);
-- /* More in this chain? */
-- if (next)
-- return next;
-- /* Look for next non-empty chain. */
-- for ((*i)++; *i <= UNIX_HASH_SIZE; (*i)++) {
-- if (!hlist_empty(&unix_socket_table[*i]))
-- return __sk_head(&unix_socket_table[*i]);
-- }
-- return NULL;
-+ do {
-+ if (s)
-+ s = sk_next(s);
-+ if (!s)
-+ s = next_unix_socket_table(i);
-+ } while (s && !nx_check(s->sk_nid, VS_WATCH_P | VS_IDENT));
-+ return s;
-+}
-+
-+static inline struct sock *first_unix_socket(int *i)
-+{
-+ *i = 0;
-+ return next_unix_socket(i, NULL);
- }
-
- #define forall_unix_sockets(i, s) \
---- linux-2.6.22/include/net/inet_hashtables.h 2007-05-02 19:25:35 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/net/inet_hashtables.h 2007-06-15 02:37:04 +0200
-@@ -271,6 +271,26 @@ static inline int inet_iif(const struct
- return ((struct rtable *)skb->dst)->rt_iif;
- }
-
-+/*
-+ * Check if a given address matches for an inet socket
-+ *
-+ * nxi: the socket's nx_info if any
-+ * addr: to be verified address
-+ * saddr: socket addresses
-+ */
-+static inline int inet_addr_match (
-+ struct nx_info *nxi,
-+ uint32_t addr,
-+ uint32_t saddr)
-+{
-+ if (addr && (saddr == addr))
-+ return 1;
-+ if (!saddr)
-+ return addr_in_nx_info(nxi, addr);
-+ return 0;
-+}
++#define vx_dentry_inc(d) do { \
++ if (atomic_read(&d->d_count) == 1) \
++ vx_acc_cres(current->vx_info, 1, d, VLIMIT_DENTRY); \
++ } while (0)
+
++#define vx_dentry_dec(d) do { \
++ if (atomic_read(&d->d_count) == 0) \
++ vx_acc_cres(current->vx_info,-1, d, VLIMIT_DENTRY); \
++ } while (0)
+
- extern struct sock *__inet_lookup_listener(struct inet_hashinfo *hashinfo,
- const __be32 daddr,
- const unsigned short hnum,
---- linux-2.6.22/include/net/inet_sock.h 2007-07-09 13:20:01 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/net/inet_sock.h 2007-06-15 02:37:04 +0200
-@@ -113,6 +113,7 @@ struct inet_sock {
- /* Socket demultiplex comparisons on incoming packets. */
- __be32 daddr;
- __be32 rcv_saddr;
-+ __be32 rcv_saddr2; /* Second bound ipv4 addr, for ipv4root */
- __be16 dport;
- __u16 num;
- __be32 saddr;
---- linux-2.6.22/include/net/inet_timewait_sock.h 2007-05-02 19:25:35 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/net/inet_timewait_sock.h 2007-06-15 02:37:04 +0200
-@@ -115,6 +115,10 @@ struct inet_timewait_sock {
- #define tw_refcnt __tw_common.skc_refcnt
- #define tw_hash __tw_common.skc_hash
- #define tw_prot __tw_common.skc_prot
-+#define tw_xid __tw_common.skc_xid
-+#define tw_vx_info __tw_common.skc_vx_info
-+#define tw_nid __tw_common.skc_nid
-+#define tw_nx_info __tw_common.skc_nx_info
- volatile unsigned char tw_substate;
- /* 3 bits hole, try to pack */
- unsigned char tw_rcv_wscale;
---- linux-2.6.22/include/net/route.h 2007-05-02 19:25:35 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/net/route.h 2007-06-15 02:37:04 +0200
-@@ -27,12 +27,16 @@
- #include <net/dst.h>
- #include <net/inetpeer.h>
- #include <net/flow.h>
-+#include <net/inet_sock.h>
- #include <linux/in_route.h>
- #include <linux/rtnetlink.h>
- #include <linux/route.h>
- #include <linux/ip.h>
- #include <linux/cache.h>
- #include <linux/security.h>
-+#include <linux/vs_base.h>
-+#include <linux/vs_network.h>
-+#include <linux/in.h>
-
- #ifndef __KERNEL__
- #warning This file is not supposed to be used outside of kernel.
-@@ -143,6 +147,59 @@ static inline char rt_tos2priority(u8 to
- return ip_tos2prio[IPTOS_TOS(tos)>>1];
- }
-
-+#define IPI_LOOPBACK htonl(INADDR_LOOPBACK)
++#define vx_dentry_avail(n) \
++ vx_cres_avail(current->vx_info, n, VLIMIT_DENTRY)
+
-+static inline int ip_find_src(struct nx_info *nxi, struct rtable **rp, struct flowi *fl)
-+{
-+ int err;
-+ int i, n = nxi->nbipv4;
-+ u32 ipv4root = nxi->ipv4[0];
+
-+ if (ipv4root == 0)
-+ return 0;
++/* socket limits */
+
-+ if (fl->fl4_src == 0) {
-+ if (n > 1) {
-+ u32 foundsrc;
-+
-+ err = __ip_route_output_key(rp, fl);
-+ if (err) {
-+ fl->fl4_src = ipv4root;
-+ err = __ip_route_output_key(rp, fl);
-+ }
-+ if (err)
-+ return err;
++#define vx_sock_inc(s) \
++ vx_acc_cres((s)->sk_vx_info, 1, s, VLIMIT_NSOCK)
+
-+ foundsrc = (*rp)->rt_src;
-+ ip_rt_put(*rp);
++#define vx_sock_dec(s) \
++ vx_acc_cres((s)->sk_vx_info,-1, s, VLIMIT_NSOCK)
+
-+ for (i=0; i<n; i++){
-+ u32 mask = nxi->mask[i];
-+ u32 ipv4 = nxi->ipv4[i];
-+ u32 net4 = ipv4 & mask;
-+
-+ if (foundsrc == ipv4) {
-+ fl->fl4_src = ipv4;
-+ break;
-+ }
-+ if (!fl->fl4_src && (foundsrc & mask) == net4)
-+ fl->fl4_src = ipv4;
-+ }
-+ }
-+ if (fl->fl4_src == 0)
-+ fl->fl4_src = (fl->fl4_dst == IPI_LOOPBACK)
-+ ? IPI_LOOPBACK : ipv4root;
-+ } else {
-+ for (i=0; i<n; i++) {
-+ if (nxi->ipv4[i] == fl->fl4_src)
-+ break;
-+ }
-+ if (i == n)
-+ return -EPERM;
-+ }
-+ return 0;
-+}
++#define vx_sock_avail(n) \
++ vx_cres_avail(current->vx_info, n, VLIMIT_NSOCK)
+
- static inline int ip_route_connect(struct rtable **rp, __be32 dst,
- __be32 src, u32 tos, int oif, u8 protocol,
- __be16 sport, __be16 dport, struct sock *sk,
-@@ -158,7 +215,27 @@ static inline int ip_route_connect(struc
- .dport = dport } } };
-
- int err;
-- if (!dst || !src) {
-+ struct nx_info *nx_info = current->nx_info;
+
-+ if (sk)
-+ nx_info = sk->sk_nx_info;
-+ vxdprintk(VXD_CBIT(net, 4),
-+ "ip_route_connect(%p) %p,%p;%lx",
-+ sk, nx_info, sk->sk_socket,
-+ (sk->sk_socket?sk->sk_socket->flags:0));
++/* ipc resource limits */
+
-+ if (nx_info) {
-+ err = ip_find_src(nx_info, rp, &fl);
-+ if (err)
-+ return err;
-+ if (fl.fl4_dst == IPI_LOOPBACK && !nx_check(0, VS_ADMIN))
-+ fl.fl4_dst = nx_info->ipv4[0];
-+#ifdef CONFIG_VSERVER_REMAP_SADDR
-+ if (fl.fl4_src == IPI_LOOPBACK && !nx_check(0, VS_ADMIN))
-+ fl.fl4_src = nx_info->ipv4[0];
-+#endif
-+ }
-+ if (!fl.fl4_dst || !fl.fl4_src) {
- err = __ip_route_output_key(rp, &fl);
- if (err)
- return err;
---- linux-2.6.22/include/net/sock.h 2007-07-09 13:20:02 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/include/net/sock.h 2007-06-15 02:37:04 +0200
-@@ -119,6 +119,10 @@ struct sock_common {
- atomic_t skc_refcnt;
- unsigned int skc_hash;
- struct proto *skc_prot;
-+ xid_t skc_xid;
-+ struct vx_info *skc_vx_info;
-+ nid_t skc_nid;
-+ struct nx_info *skc_nx_info;
- };
-
- /**
-@@ -195,6 +199,10 @@ struct sock {
- #define sk_refcnt __sk_common.skc_refcnt
- #define sk_hash __sk_common.skc_hash
- #define sk_prot __sk_common.skc_prot
-+#define sk_xid __sk_common.skc_xid
-+#define sk_vx_info __sk_common.skc_vx_info
-+#define sk_nid __sk_common.skc_nid
-+#define sk_nx_info __sk_common.skc_nx_info
- unsigned char sk_shutdown : 2,
- sk_no_check : 2,
- sk_userlocks : 4;
---- linux-2.6.22/init/main.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/init/main.c 2007-06-15 04:03:00 +0200
-@@ -55,6 +55,7 @@
- #include <linux/pid_namespace.h>
- #include <linux/device.h>
- #include <linux/kthread.h>
-+#include <linux/vserver/percpu.h>
-
- #include <asm/io.h>
- #include <asm/bugs.h>
-@@ -364,12 +365,14 @@ EXPORT_SYMBOL(__per_cpu_offset);
-
- static void __init setup_per_cpu_areas(void)
- {
-- unsigned long size, i;
-+ unsigned long size, vspc, i;
- char *ptr;
- unsigned long nr_possible_cpus = num_possible_cpus();
-
-+ vspc = PERCPU_PERCTX * CONFIG_VSERVER_CONTEXTS;
++#define vx_ipcmsg_add(v, u, a) \
++ vx_add_cres(v, a, u, RLIMIT_MSGQUEUE)
+
- /* Copy section for each CPU (we discard the original) */
-- size = ALIGN(PERCPU_ENOUGH_ROOM, PAGE_SIZE);
-+ size = ALIGN(PERCPU_ENOUGH_ROOM + vspc, PAGE_SIZE);
- ptr = alloc_bootmem_pages(size * nr_possible_cpus);
-
- for_each_possible_cpu(i) {
---- linux-2.6.22/ipc/mqueue.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/ipc/mqueue.c 2007-06-15 02:37:04 +0200
-@@ -29,6 +29,8 @@
- #include <linux/audit.h>
- #include <linux/signal.h>
- #include <linux/mutex.h>
-+#include <linux/vs_context.h>
-+#include <linux/vs_limit.h>
-
- #include <net/sock.h>
- #include "util.h"
-@@ -151,17 +153,20 @@ static struct inode *mqueue_get_inode(st
- spin_lock(&mq_lock);
- if (u->mq_bytes + mq_bytes < u->mq_bytes ||
- u->mq_bytes + mq_bytes >
-- p->signal->rlim[RLIMIT_MSGQUEUE].rlim_cur) {
-+ p->signal->rlim[RLIMIT_MSGQUEUE].rlim_cur ||
-+ !vx_ipcmsg_avail(p->vx_info, mq_bytes)) {
- spin_unlock(&mq_lock);
- goto out_inode;
- }
- u->mq_bytes += mq_bytes;
-+ vx_ipcmsg_add(p->vx_info, u, mq_bytes);
- spin_unlock(&mq_lock);
-
- info->messages = kmalloc(mq_msg_tblsz, GFP_KERNEL);
- if (!info->messages) {
- spin_lock(&mq_lock);
- u->mq_bytes -= mq_bytes;
-+ vx_ipcmsg_sub(p->vx_info, u, mq_bytes);
- spin_unlock(&mq_lock);
- goto out_inode;
- }
-@@ -257,10 +262,14 @@ static void mqueue_delete_inode(struct i
- (info->attr.mq_maxmsg * info->attr.mq_msgsize));
- user = info->user;
- if (user) {
-+ struct vx_info *vxi = lookup_vx_info(user->xid);
++#define vx_ipcmsg_sub(v, u, a) \
++ vx_sub_cres(v, a, u, RLIMIT_MSGQUEUE)
+
- spin_lock(&mq_lock);
- user->mq_bytes -= mq_bytes;
-+ vx_ipcmsg_sub(vxi, user, mq_bytes);
- queues_count--;
- spin_unlock(&mq_lock);
-+ put_vx_info(vxi);
- free_uid(user);
- }
- }
-@@ -748,7 +757,7 @@ asmlinkage long sys_mq_unlink(const char
- if (inode)
- atomic_inc(&inode->i_count);
-
-- err = vfs_unlink(dentry->d_parent->d_inode, dentry);
-+ err = vfs_unlink(dentry->d_parent->d_inode, dentry, NULL);
- out_err:
- dput(dentry);
-
---- linux-2.6.22/ipc/msg.c 2007-02-06 03:01:55 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/ipc/msg.c 2007-06-15 02:37:04 +0200
-@@ -36,6 +36,7 @@
- #include <linux/seq_file.h>
- #include <linux/mutex.h>
- #include <linux/nsproxy.h>
-+#include <linux/vs_base.h>
-
- #include <asm/current.h>
- #include <asm/uaccess.h>
-@@ -149,6 +150,7 @@ static int newque (struct ipc_namespace
-
- msq->q_perm.mode = msgflg & S_IRWXUGO;
- msq->q_perm.key = key;
-+ msq->q_perm.xid = vx_current_xid();
-
- msq->q_perm.security = NULL;
- retval = security_msg_queue_alloc(msq);
---- linux-2.6.22/ipc/sem.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/ipc/sem.c 2007-06-15 02:37:04 +0200
-@@ -82,6 +82,8 @@
- #include <linux/seq_file.h>
- #include <linux/mutex.h>
- #include <linux/nsproxy.h>
-+#include <linux/vs_base.h>
-+#include <linux/vs_limit.h>
-
- #include <asm/uaccess.h>
- #include "util.h"
-@@ -229,6 +231,7 @@ static int newary (struct ipc_namespace
-
- sma->sem_perm.mode = (semflg & S_IRWXUGO);
- sma->sem_perm.key = key;
-+ sma->sem_perm.xid = vx_current_xid();
-
- sma->sem_perm.security = NULL;
- retval = security_sem_alloc(sma);
-@@ -244,6 +247,9 @@ static int newary (struct ipc_namespace
- return -ENOSPC;
- }
- ns->used_sems += nsems;
-+ /* FIXME: obsoleted? */
-+ vx_semary_inc(sma);
-+ vx_nsems_add(sma, nsems);
-
- sma->sem_id = sem_buildid(ns, id, sma->sem_perm.seq);
- sma->sem_base = (struct sem *) &sma[1];
-@@ -525,6 +531,9 @@ static void freeary (struct ipc_namespac
- sem_unlock(sma);
-
- ns->used_sems -= sma->sem_nsems;
-+ /* FIXME: obsoleted? */
-+ vx_nsems_sub(sma, sma->sem_nsems);
-+ vx_semary_dec(sma);
- size = sizeof (*sma) + sma->sem_nsems * sizeof (struct sem);
- security_sem_free(sma);
- ipc_rcu_putref(sma);
---- linux-2.6.22/ipc/shm.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/ipc/shm.c 2007-06-17 05:54:20 +0200
-@@ -38,6 +38,8 @@
- #include <linux/mutex.h>
- #include <linux/nsproxy.h>
- #include <linux/mount.h>
-+#include <linux/vs_context.h>
-+#include <linux/vs_limit.h>
-
- #include <asm/uaccess.h>
-
-@@ -185,7 +187,12 @@ static void shm_open(struct vm_area_stru
- */
- static void shm_destroy(struct ipc_namespace *ns, struct shmid_kernel *shp)
- {
-- ns->shm_tot -= (shp->shm_segsz + PAGE_SIZE - 1) >> PAGE_SHIFT;
-+ struct vx_info *vxi = lookup_vx_info(shp->shm_perm.xid);
-+ int numpages = (shp->shm_segsz + PAGE_SIZE - 1) >> PAGE_SHIFT;
++#define vx_ipcmsg_avail(v, a) \
++ vx_cres_avail(v, a, RLIMIT_MSGQUEUE)
+
-+ vx_ipcshm_sub(vxi, shp, numpages);
-+ ns->shm_tot -= numpages;
+
- shm_rmid(ns, shp->id);
- shm_unlock(shp);
- if (!is_file_hugepages(shp->shm_file))
-@@ -195,6 +202,7 @@ static void shm_destroy(struct ipc_names
- shp->mlock_user);
- fput (shp->shm_file);
- security_shm_free(shp);
-+ put_vx_info(vxi);
- ipc_rcu_putref(shp);
- }
-
-@@ -351,11 +359,15 @@ static int newseg (struct ipc_namespace
- if (ns->shm_tot + numpages > ns->shm_ctlall)
- return -ENOSPC;
-
-+ if (!vx_ipcshm_avail(current->vx_info, numpages))
-+ return -ENOSPC;
++#define vx_ipcshm_add(v, k, a) \
++ vx_add_cres(v, a, (void *)(long)(k), VLIMIT_SHMEM)
+
- shp = ipc_rcu_alloc(sizeof(*shp));
- if (!shp)
- return -ENOMEM;
-
- shp->shm_perm.key = key;
-+ shp->shm_perm.xid = vx_current_xid();
- shp->shm_perm.mode = (shmflg & S_IRWXUGO);
- shp->mlock_user = NULL;
-
-@@ -406,6 +418,7 @@ static int newseg (struct ipc_namespace
- file->f_dentry->d_inode->i_ino = shp->id;
-
- ns->shm_tot += numpages;
-+ vx_ipcshm_add(current->vx_info, key, numpages);
- shm_unlock(shp);
- return shp->id;
-
---- linux-2.6.22/ipc/util.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/ipc/util.c 2007-06-15 02:37:04 +0200
-@@ -32,6 +32,8 @@
- #include <linux/proc_fs.h>
- #include <linux/audit.h>
- #include <linux/nsproxy.h>
-+#include <linux/vs_base.h>
-+#include <linux/vserver/global.h>
-
- #include <asm/unistd.h>
-
-@@ -72,6 +74,7 @@ static struct ipc_namespace *clone_ipc_n
- goto err_shm;
-
- kref_init(&ns->kref);
-+ atomic_inc(&vs_global_ipc_ns);
- return ns;
-
- err_shm:
-@@ -108,6 +111,7 @@ void free_ipc_ns(struct kref *kref)
- sem_exit_ns(ns);
- msg_exit_ns(ns);
- shm_exit_ns(ns);
-+ atomic_dec(&vs_global_ipc_ns);
- kfree(ns);
- }
- #else
---- linux-2.6.22/kernel/Makefile 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/Makefile 2007-06-15 02:37:04 +0200
-@@ -10,6 +10,8 @@ obj-y = sched.o fork.o exec_domain.o
- kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \
- hrtimer.o rwsem.o latency.o nsproxy.o srcu.o die_notifier.o
-
-+obj-y += vserver/
++#define vx_ipcshm_sub(v, k, a) \
++ vx_sub_cres(v, a, (void *)(long)(k), VLIMIT_SHMEM)
+
- obj-$(CONFIG_STACKTRACE) += stacktrace.o
- obj-y += time/
- obj-$(CONFIG_DEBUG_MUTEXES) += mutex-debug.o
---- linux-2.6.22/kernel/capability.c 2007-05-02 19:25:37 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/capability.c 2007-06-15 02:37:04 +0200
-@@ -12,6 +12,7 @@
- #include <linux/module.h>
- #include <linux/security.h>
- #include <linux/syscalls.h>
-+#include <linux/vs_context.h>
- #include <asm/uaccess.h>
-
- unsigned securebits = SECUREBITS_DEFAULT; /* systemwide security settings */
-@@ -103,6 +104,8 @@ static inline int cap_set_pg(int pgrp_nr
-
- pgrp = find_pid(pgrp_nr);
- do_each_pid_task(pgrp, PIDTYPE_PGID, g) {
-+ if (!vx_check(g->xid, VS_ADMIN_P | VS_IDENT))
-+ continue;
- target = g;
- while_each_thread(g, target) {
- if (!security_capset_check(target, effective,
-@@ -246,8 +249,12 @@ int __capable(struct task_struct *t, int
- }
- EXPORT_SYMBOL(__capable);
-
-+#include <linux/vserver/base.h>
- int capable(int cap)
- {
-+ /* here for now so we don't require task locking */
-+ if (vs_check_bit(VXC_CAP_MASK, cap) && !vx_mcaps(1L << cap))
-+ return 0;
- return __capable(current, cap);
- }
- EXPORT_SYMBOL(capable);
---- linux-2.6.22/kernel/compat.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/compat.c 2007-06-15 02:37:04 +0200
-@@ -846,7 +846,7 @@ asmlinkage long compat_sys_time(compat_t
- compat_time_t i;
- struct timeval tv;
-
-- do_gettimeofday(&tv);
-+ vx_gettimeofday(&tv);
- i = tv.tv_sec;
-
- if (tloc) {
-@@ -870,7 +870,7 @@ asmlinkage long compat_sys_stime(compat_
- if (err)
- return err;
-
-- do_settimeofday(&tv);
-+ vx_settimeofday(&tv);
- return 0;
- }
-
---- linux-2.6.22/kernel/exit.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/exit.c 2007-06-17 05:54:20 +0200
-@@ -44,6 +44,11 @@
- #include <linux/resource.h>
- #include <linux/blkdev.h>
- #include <linux/task_io_accounting_ops.h>
-+#include <linux/vs_limit.h>
-+#include <linux/vs_context.h>
-+#include <linux/vs_network.h>
-+#include <linux/vs_pid.h>
-+#include <linux/vserver/global.h>
-
- #include <asm/uaccess.h>
- #include <asm/unistd.h>
-@@ -443,9 +448,11 @@ static void close_files(struct files_str
- filp_close(file, files);
- cond_resched();
- }
-+ vx_openfd_dec(i);
- }
- i++;
- set >>= 1;
-+ cond_resched();
- }
- }
- }
-@@ -525,6 +532,7 @@ static inline void __put_fs_struct(struc
- dput(fs->altroot);
- mntput(fs->altrootmnt);
- }
-+ atomic_dec(&vs_global_fs);
- kmem_cache_free(fs_cachep, fs);
- }
- }
-@@ -596,6 +604,14 @@ static void exit_mm(struct task_struct *
- static inline void
- choose_new_parent(struct task_struct *p, struct task_struct *reaper)
- {
-+ /* check for reaper context */
-+ vxwprintk((p->xid != reaper->xid) && (reaper != child_reaper(p)),
-+ "rogue reaper: %p[%d,#%u] <> %p[%d,#%u]",
-+ p, p->pid, p->xid, reaper, reaper->pid, reaper->xid);
-+
-+ if (p == reaper)
-+ reaper = vx_child_reaper(p);
++#define vx_ipcshm_avail(v, a) \
++ vx_cres_avail(v, a, VLIMIT_SHMEM)
+
- /*
- * Make sure we're not reparenting to ourselves and that
- * the parent is not a zombie.
-@@ -687,7 +703,8 @@ forget_original_parent(struct task_struc
- do {
- reaper = next_thread(reaper);
- if (reaper == father) {
-- reaper = child_reaper(father);
-+ // reaper = child_reaper(father);
-+ reaper = vx_child_reaper(father);
- break;
- }
- } while (reaper->exit_state);
-@@ -964,7 +981,11 @@ fastcall NORET_TYPE void do_exit(long co
- tsk->exit_code = code;
- proc_exit_connector(tsk);
- exit_task_namespaces(tsk);
-+ // ns = exit_task_namespaces_early(tsk);
-+ /* needs to stay before exit_notify() */
-+ exit_vx_info_early(tsk, code);
- exit_notify(tsk);
-+ // exit_task_namespaces(tsk, ns);
- #ifdef CONFIG_NUMA
- mpol_free(tsk->mempolicy);
- tsk->mempolicy = NULL;
-@@ -994,6 +1015,10 @@ fastcall NORET_TYPE void do_exit(long co
- if (tsk->splice_pipe)
- __free_pipe_info(tsk->splice_pipe);
-
-+ /* needs to stay after exit_notify() */
-+ exit_vx_info(tsk, code);
-+ exit_nx_info(tsk);
+
- preempt_disable();
- /* causes final put_task_struct in finish_task_switch(). */
- tsk->state = TASK_DEAD;
---- linux-2.6.22/kernel/fork.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/fork.c 2007-06-15 02:37:04 +0200
-@@ -49,6 +49,11 @@
- #include <linux/delayacct.h>
- #include <linux/taskstats_kern.h>
- #include <linux/random.h>
-+#include <linux/vs_context.h>
-+#include <linux/vs_network.h>
-+#include <linux/vs_limit.h>
-+#include <linux/vs_memory.h>
-+#include <linux/vserver/global.h>
-
- #include <asm/pgtable.h>
- #include <asm/pgalloc.h>
-@@ -108,6 +113,8 @@ void free_task(struct task_struct *tsk)
- {
- free_thread_info(tsk->stack);
- rt_mutex_debug_task_free(tsk);
-+ clr_vx_info(&tsk->vx_info);
-+ clr_nx_info(&tsk->nx_info);
- free_task_struct(tsk);
- }
- EXPORT_SYMBOL(free_task);
-@@ -215,6 +222,8 @@ static inline int dup_mmap(struct mm_str
- mm->free_area_cache = oldmm->mmap_base;
- mm->cached_hole_size = ~0UL;
- mm->map_count = 0;
-+ __set_mm_counter(mm, file_rss, 0);
-+ __set_mm_counter(mm, anon_rss, 0);
- cpus_clear(mm->cpu_vm_mask);
- mm->mm_rb = RB_ROOT;
- rb_link = &mm->mm_rb.rb_node;
-@@ -226,7 +235,7 @@ static inline int dup_mmap(struct mm_str
-
- if (mpnt->vm_flags & VM_DONTCOPY) {
- long pages = vma_pages(mpnt);
-- mm->total_vm -= pages;
-+ vx_vmpages_sub(mm, pages);
- vm_stat_account(mm, mpnt->vm_flags, mpnt->vm_file,
- -pages);
- continue;
-@@ -335,8 +344,8 @@ static struct mm_struct * mm_init(struct
- INIT_LIST_HEAD(&mm->mmlist);
- mm->core_waiters = 0;
- mm->nr_ptes = 0;
-- set_mm_counter(mm, file_rss, 0);
-- set_mm_counter(mm, anon_rss, 0);
-+ __set_mm_counter(mm, file_rss, 0);
-+ __set_mm_counter(mm, anon_rss, 0);
- spin_lock_init(&mm->page_table_lock);
- rwlock_init(&mm->ioctx_list_lock);
- mm->ioctx_list = NULL;
-@@ -345,6 +354,7 @@ static struct mm_struct * mm_init(struct
-
- if (likely(!mm_alloc_pgd(mm))) {
- mm->def_flags = 0;
-+ set_vx_info(&mm->mm_vx_info, current->vx_info);
- return mm;
- }
- free_mm(mm);
-@@ -376,6 +386,7 @@ void fastcall __mmdrop(struct mm_struct
- BUG_ON(mm == &init_mm);
- mm_free_pgd(mm);
- destroy_context(mm);
-+ clr_vx_info(&mm->mm_vx_info);
- free_mm(mm);
- }
-
-@@ -490,6 +501,7 @@ static struct mm_struct *dup_mm(struct t
- goto fail_nomem;
-
- memcpy(mm, oldmm, sizeof(*mm));
-+ mm->mm_vx_info = NULL;
-
- /* Initializing for Swap token stuff */
- mm->token_priority = 0;
-@@ -521,6 +533,7 @@ fail_nocontext:
- * If init_new_context() failed, we cannot use mmput() to free the mm
- * because it calls destroy_context()
- */
-+ clr_vx_info(&mm->mm_vx_info);
- mm_free_pgd(mm);
- free_mm(mm);
- return NULL;
-@@ -591,6 +604,7 @@ static inline struct fs_struct *__copy_f
- fs->altroot = NULL;
- }
- read_unlock(&old->lock);
-+ atomic_inc(&vs_global_fs);
- }
- return fs;
- }
-@@ -709,6 +723,8 @@ static struct files_struct *dup_fd(struc
- struct file *f = *old_fds++;
- if (f) {
- get_file(f);
-+ /* TODO: sum it first for check and performance */
-+ vx_openfd_inc(open_files - i);
- } else {
- /*
- * The fd may be claimed in the fd bitmap but not yet
-@@ -961,6 +977,8 @@ static struct task_struct *copy_process(
- {
- int retval;
- struct task_struct *p = NULL;
-+ struct vx_info *vxi;
-+ struct nx_info *nxi;
-
- if ((clone_flags & (CLONE_NEWNS|CLONE_FS)) == (CLONE_NEWNS|CLONE_FS))
- return ERR_PTR(-EINVAL);
-@@ -995,12 +1013,30 @@ static struct task_struct *copy_process(
- DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled);
- DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled);
- #endif
-+ init_vx_info(&p->vx_info, current->vx_info);
-+ init_nx_info(&p->nx_info, current->nx_info);
++#define vx_semary_inc(a) \
++ vx_acc_cres(current->vx_info, 1, a, VLIMIT_SEMARY)
+
-+ /* check vserver memory */
-+ if (p->mm && !(clone_flags & CLONE_VM)) {
-+ if (vx_vmpages_avail(p->mm, p->mm->total_vm))
-+ vx_pages_add(p->vx_info, RLIMIT_AS, p->mm->total_vm);
-+ else
-+ goto bad_fork_free;
-+ }
-+ if (p->mm && vx_flags(VXF_FORK_RSS, 0)) {
-+ if (!vx_rss_avail(p->mm, get_mm_counter(p->mm, file_rss)))
-+ goto bad_fork_cleanup_vm;
-+ }
++#define vx_semary_dec(a) \
++ vx_acc_cres(current->vx_info, -1, a, VLIMIT_SEMARY)
+
- retval = -EAGAIN;
-+ if (!vx_nproc_avail(1))
-+ goto bad_fork_cleanup_vm;
+
- if (atomic_read(&p->user->processes) >=
- p->signal->rlim[RLIMIT_NPROC].rlim_cur) {
- if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE) &&
- p->user != &root_user)
-- goto bad_fork_free;
-+ goto bad_fork_cleanup_vm;
- }
-
- atomic_inc(&p->user->__count);
-@@ -1265,6 +1301,18 @@ static struct task_struct *copy_process(
-
- total_forks++;
- spin_unlock(¤t->sighand->siglock);
++#define vx_nsems_add(a,n) \
++ vx_add_cres(current->vx_info, n, a, VLIMIT_NSEMS)
+
-+ /* p is copy of current */
-+ vxi = p->vx_info;
-+ if (vxi) {
-+ claim_vx_info(vxi, p);
-+ atomic_inc(&vxi->cvirt.nr_threads);
-+ atomic_inc(&vxi->cvirt.total_forks);
-+ vx_nproc_inc(p);
-+ }
-+ nxi = p->nx_info;
-+ if (nxi)
-+ claim_nx_info(nxi, p);
- write_unlock_irq(&tasklist_lock);
- proc_fork_connector(p);
- return p;
-@@ -1306,6 +1354,9 @@ bad_fork_cleanup_count:
- put_group_info(p->group_info);
- atomic_dec(&p->user->processes);
- free_uid(p->user);
-+bad_fork_cleanup_vm:
-+ if (p->mm && !(clone_flags & CLONE_VM))
-+ vx_pages_sub(p->vx_info, RLIMIT_AS, p->mm->total_vm);
- bad_fork_free:
- free_task(p);
- fork_out:
-@@ -1367,6 +1418,15 @@ long do_fork(unsigned long clone_flags,
-
- if (!pid)
- return -EAGAIN;
++#define vx_nsems_sub(a,n) \
++ vx_sub_cres(current->vx_info, n, a, VLIMIT_NSEMS)
+
-+ /* kernel threads are host only */
-+ if ((clone_flags & CLONE_KTHREAD) && !vx_check(0, VS_ADMIN)) {
-+ vxwprintk(1, "xid=%d tried to spawn a kernel thread.",
-+ vx_current_xid());
-+ free_pid(pid);
-+ return -EPERM;
-+ }
+
- nr = pid->nr;
- if (unlikely(current->ptrace)) {
- trace = fork_traceflag (clone_flags);
---- linux-2.6.22/kernel/kthread.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/kthread.c 2007-06-15 02:37:04 +0200
-@@ -96,7 +96,7 @@ static void create_kthread(struct kthrea
- } else {
- wait_for_completion(&create->started);
- read_lock(&tasklist_lock);
-- create->result = find_task_by_pid(pid);
-+ create->result = find_task_by_real_pid(pid);
- read_unlock(&tasklist_lock);
- }
- complete(&create->done);
---- linux-2.6.22/kernel/nsproxy.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/nsproxy.c 2007-07-07 04:12:31 +0200
-@@ -20,14 +20,11 @@
- #include <linux/mnt_namespace.h>
- #include <linux/utsname.h>
- #include <linux/pid_namespace.h>
-+#include <linux/vserver/global.h>
-+#include <linux/vserver/debug.h>
-
- struct nsproxy init_nsproxy = INIT_NSPROXY(init_nsproxy);
-
--static inline void get_nsproxy(struct nsproxy *ns)
--{
-- atomic_inc(&ns->count);
--}
--
- void get_task_namespaces(struct task_struct *tsk)
- {
- struct nsproxy *ns = tsk->nsproxy;
-@@ -46,6 +43,9 @@ static inline struct nsproxy *clone_nspr
- ns = kmemdup(orig, sizeof(struct nsproxy), GFP_KERNEL);
- if (ns)
- atomic_set(&ns->count, 1);
-+ vxdprintk(VXD_CBIT(space, 2), "clone_nsproxy(%p[%u] = %p[1]",
-+ orig, atomic_read(&orig->count), ns);
-+ atomic_inc(&vs_global_nsproxy);
- return ns;
- }
-
-@@ -54,28 +54,32 @@ static inline struct nsproxy *clone_nspr
- * Return the newly created nsproxy. Do not attach this to the task,
- * leave it to the caller to do proper locking and attach it to task.
- */
--static struct nsproxy *create_new_namespaces(int flags, struct task_struct *tsk,
-+static struct nsproxy *unshare_namespaces(int flags, struct nsproxy *orig,
- struct fs_struct *new_fs)
- {
- struct nsproxy *new_nsp;
-
-- new_nsp = clone_nsproxy(tsk->nsproxy);
-+ vxdprintk(VXD_CBIT(space, 4),
-+ "unshare_namespaces(0x%08x,%p,%p)",
-+ flags, orig, new_fs);
++#else
++#warning duplicate inclusion
++#endif
+--- linux-2.6.25.4/include/linux/vs_memory.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_memory.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,159 @@
++#ifndef _VS_MEMORY_H
++#define _VS_MEMORY_H
+
-+ new_nsp = clone_nsproxy(orig);
- if (!new_nsp)
- return ERR_PTR(-ENOMEM);
-
-- new_nsp->mnt_ns = copy_mnt_ns(flags, tsk->nsproxy->mnt_ns, new_fs);
-+ new_nsp->mnt_ns = copy_mnt_ns(flags, orig->mnt_ns, new_fs);
- if (IS_ERR(new_nsp->mnt_ns))
- goto out_ns;
-
-- new_nsp->uts_ns = copy_utsname(flags, tsk->nsproxy->uts_ns);
-+ new_nsp->uts_ns = copy_utsname(flags, orig->uts_ns);
- if (IS_ERR(new_nsp->uts_ns))
- goto out_uts;
-
-- new_nsp->ipc_ns = copy_ipcs(flags, tsk->nsproxy->ipc_ns);
-+ new_nsp->ipc_ns = copy_ipcs(flags, orig->ipc_ns);
- if (IS_ERR(new_nsp->ipc_ns))
- goto out_ipc;
-
-- new_nsp->pid_ns = copy_pid_ns(flags, tsk->nsproxy->pid_ns);
-+ new_nsp->pid_ns = copy_pid_ns(flags, orig->pid_ns);
- if (IS_ERR(new_nsp->pid_ns))
- goto out_pid;
-
-@@ -95,6 +99,33 @@ out_ns:
- return ERR_PTR(-ENOMEM);
- }
-
-+static struct nsproxy *create_new_namespaces(int flags, struct task_struct *tsk,
-+ struct fs_struct *new_fs)
-+{
-+ return unshare_namespaces(flags, tsk->nsproxy, new_fs);
-+}
++#include "vserver/limit.h"
++#include "vserver/base.h"
++#include "vserver/context.h"
++#include "vserver/debug.h"
++#include "vserver/context.h"
++#include "vserver/limit_int.h"
+
-+/*
-+ * copies the nsproxy, setting refcount to 1, and grabbing a
-+ * reference to all contained namespaces.
-+ */
-+struct nsproxy *copy_nsproxy(struct nsproxy *orig)
++
++#define __acc_add_long(a, v) (*(v) += (a))
++#define __acc_inc_long(v) (++*(v))
++#define __acc_dec_long(v) (--*(v))
++
++#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
++#define __acc_add_atomic(a, v) atomic_long_add(a, v)
++#define __acc_inc_atomic(v) atomic_long_inc(v)
++#define __acc_dec_atomic(v) atomic_long_dec(v)
++#else /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
++#define __acc_add_atomic(a, v) __acc_add_long(a, v)
++#define __acc_inc_atomic(v) __acc_inc_long(v)
++#define __acc_dec_atomic(v) __acc_dec_long(v)
++#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
++
++
++#define vx_acc_page(m, d, v, r) do { \
++ if ((d) > 0) \
++ __acc_inc_long(&(m)->v); \
++ else \
++ __acc_dec_long(&(m)->v); \
++ __vx_acc_cres(m->mm_vx_info, r, d, m, __FILE__, __LINE__); \
++} while (0)
++
++#define vx_acc_page_atomic(m, d, v, r) do { \
++ if ((d) > 0) \
++ __acc_inc_atomic(&(m)->v); \
++ else \
++ __acc_dec_atomic(&(m)->v); \
++ __vx_acc_cres(m->mm_vx_info, r, d, m, __FILE__, __LINE__); \
++} while (0)
++
++
++#define vx_acc_pages(m, p, v, r) do { \
++ unsigned long __p = (p); \
++ __acc_add_long(__p, &(m)->v); \
++ __vx_add_cres(m->mm_vx_info, r, __p, m, __FILE__, __LINE__); \
++} while (0)
++
++#define vx_acc_pages_atomic(m, p, v, r) do { \
++ unsigned long __p = (p); \
++ __acc_add_atomic(__p, &(m)->v); \
++ __vx_add_cres(m->mm_vx_info, r, __p, m, __FILE__, __LINE__); \
++} while (0)
++
++
++
++#define vx_acc_vmpage(m, d) \
++ vx_acc_page(m, d, total_vm, RLIMIT_AS)
++#define vx_acc_vmlpage(m, d) \
++ vx_acc_page(m, d, locked_vm, RLIMIT_MEMLOCK)
++#define vx_acc_file_rsspage(m, d) \
++ vx_acc_page_atomic(m, d, _file_rss, VLIMIT_MAPPED)
++#define vx_acc_anon_rsspage(m, d) \
++ vx_acc_page_atomic(m, d, _anon_rss, VLIMIT_ANON)
++
++#define vx_acc_vmpages(m, p) \
++ vx_acc_pages(m, p, total_vm, RLIMIT_AS)
++#define vx_acc_vmlpages(m, p) \
++ vx_acc_pages(m, p, locked_vm, RLIMIT_MEMLOCK)
++#define vx_acc_file_rsspages(m, p) \
++ vx_acc_pages_atomic(m, p, _file_rss, VLIMIT_MAPPED)
++#define vx_acc_anon_rsspages(m, p) \
++ vx_acc_pages_atomic(m, p, _anon_rss, VLIMIT_ANON)
++
++#define vx_pages_add(s, r, p) __vx_add_cres(s, r, p, 0, __FILE__, __LINE__)
++#define vx_pages_sub(s, r, p) vx_pages_add(s, r, -(p))
++
++#define vx_vmpages_inc(m) vx_acc_vmpage(m, 1)
++#define vx_vmpages_dec(m) vx_acc_vmpage(m, -1)
++#define vx_vmpages_add(m, p) vx_acc_vmpages(m, p)
++#define vx_vmpages_sub(m, p) vx_acc_vmpages(m, -(p))
++
++#define vx_vmlocked_inc(m) vx_acc_vmlpage(m, 1)
++#define vx_vmlocked_dec(m) vx_acc_vmlpage(m, -1)
++#define vx_vmlocked_add(m, p) vx_acc_vmlpages(m, p)
++#define vx_vmlocked_sub(m, p) vx_acc_vmlpages(m, -(p))
++
++#define vx_file_rsspages_inc(m) vx_acc_file_rsspage(m, 1)
++#define vx_file_rsspages_dec(m) vx_acc_file_rsspage(m, -1)
++#define vx_file_rsspages_add(m, p) vx_acc_file_rsspages(m, p)
++#define vx_file_rsspages_sub(m, p) vx_acc_file_rsspages(m, -(p))
++
++#define vx_anon_rsspages_inc(m) vx_acc_anon_rsspage(m, 1)
++#define vx_anon_rsspages_dec(m) vx_acc_anon_rsspage(m, -1)
++#define vx_anon_rsspages_add(m, p) vx_acc_anon_rsspages(m, p)
++#define vx_anon_rsspages_sub(m, p) vx_acc_anon_rsspages(m, -(p))
++
++
++#define vx_pages_avail(m, p, r) \
++ __vx_cres_avail((m)->mm_vx_info, r, p, __FILE__, __LINE__)
++
++#define vx_vmpages_avail(m, p) vx_pages_avail(m, p, RLIMIT_AS)
++#define vx_vmlocked_avail(m, p) vx_pages_avail(m, p, RLIMIT_MEMLOCK)
++#define vx_anon_avail(m, p) vx_pages_avail(m, p, VLIMIT_ANON)
++#define vx_mapped_avail(m, p) vx_pages_avail(m, p, VLIMIT_MAPPED)
++
++#define vx_rss_avail(m, p) \
++ __vx_cres_array_avail((m)->mm_vx_info, VLA_RSS, p, __FILE__, __LINE__)
++
++
++enum {
++ VXPT_UNKNOWN = 0,
++ VXPT_ANON,
++ VXPT_NONE,
++ VXPT_FILE,
++ VXPT_SWAP,
++ VXPT_WRITE
++};
++
++#if 0
++#define vx_page_fault(mm, vma, type, ret)
++#else
++
++static inline
++void __vx_page_fault(struct mm_struct *mm,
++ struct vm_area_struct *vma, int type, int ret)
+{
-+ struct nsproxy *ns = clone_nsproxy(orig);
++ struct vx_info *vxi = mm->mm_vx_info;
++ int what;
++/*
++ static char *page_type[6] =
++ { "UNKNOWN", "ANON", "NONE", "FILE", "SWAP", "WRITE" };
++ static char *page_what[4] =
++ { "FAULT_OOM", "FAULT_SIGBUS", "FAULT_MINOR", "FAULT_MAJOR" };
++*/
+
-+ if (ns) {
-+ if (ns->mnt_ns)
-+ get_mnt_ns(ns->mnt_ns);
-+ if (ns->uts_ns)
-+ get_uts_ns(ns->uts_ns);
-+ if (ns->ipc_ns)
-+ get_ipc_ns(ns->ipc_ns);
-+ if (ns->pid_ns)
-+ get_pid_ns(ns->pid_ns);
-+ }
-+ return ns;
++ if (!vxi)
++ return;
++
++ what = (ret & 0x3);
++
++/* printk("[%d] page[%d][%d] %2x %s %s\n", vxi->vx_id,
++ type, what, ret, page_type[type], page_what[what]);
++*/
++ if (ret & VM_FAULT_WRITE)
++ what |= 0x4;
++ atomic_inc(&vxi->cacct.page[type][what]);
+}
+
- /*
- * called from clone. This now handles copy for nsproxy and all
- * namespaces therein.
-@@ -102,9 +133,12 @@ out_ns:
- int copy_namespaces(int flags, struct task_struct *tsk)
- {
- struct nsproxy *old_ns = tsk->nsproxy;
-- struct nsproxy *new_ns;
-+ struct nsproxy *new_ns = NULL;
- int err = 0;
-
-+ vxdprintk(VXD_CBIT(space, 7), "copy_namespaces(0x%08x,%p[%p])",
-+ flags, tsk, old_ns);
++#define vx_page_fault(mm, vma, type, ret) __vx_page_fault(mm, vma, type, ret)
++#endif
+
- if (!old_ns)
- return 0;
-
-@@ -127,6 +161,9 @@ int copy_namespaces(int flags, struct ta
- tsk->nsproxy = new_ns;
- out:
- put_nsproxy(old_ns);
-+ vxdprintk(VXD_CBIT(space, 3),
-+ "copy_namespaces(0x%08x,%p[%p]) = %d [%p]",
-+ flags, tsk, old_ns, err, new_ns);
- return err;
- }
-
-@@ -140,6 +177,7 @@ void free_nsproxy(struct nsproxy *ns)
- put_ipc_ns(ns->ipc_ns);
- if (ns->pid_ns)
- put_pid_ns(ns->pid_ns);
-+ atomic_dec(&vs_global_nsproxy);
- kfree(ns);
- }
-
-@@ -152,6 +190,10 @@ int unshare_nsproxy_namespaces(unsigned
- {
- int err = 0;
-
-+ vxdprintk(VXD_CBIT(space, 4),
-+ "unshare_nsproxy_namespaces(0x%08lx,[%p])",
-+ unshare_flags, current->nsproxy);
+
- if (!(unshare_flags & (CLONE_NEWNS | CLONE_NEWUTS | CLONE_NEWIPC)))
- return 0;
-
---- linux-2.6.22/kernel/pid.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/pid.c 2007-06-15 04:04:09 +0200
-@@ -28,6 +28,7 @@
- #include <linux/hash.h>
- #include <linux/pid_namespace.h>
- #include <linux/init_task.h>
-+#include <linux/vs_pid.h>
-
- #define pid_hashfn(nr) hash_long((unsigned long)nr, pidhash_shift)
- static struct hlist_head *pid_hash;
-@@ -295,6 +296,9 @@ void fastcall transfer_pid(struct task_s
- struct task_struct * fastcall pid_task(struct pid *pid, enum pid_type type)
- {
- struct task_struct *result = NULL;
++extern unsigned long vx_badness(struct task_struct *task, struct mm_struct *mm);
++
++#else
++#warning duplicate inclusion
++#endif
+--- linux-2.6.25.4/include/linux/vs_network.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_network.h 2008-04-29 18:42:49.000000000 -0400
+@@ -0,0 +1,169 @@
++#ifndef _NX_VS_NETWORK_H
++#define _NX_VS_NETWORK_H
++
++#include "vserver/context.h"
++#include "vserver/network.h"
++#include "vserver/base.h"
++#include "vserver/check.h"
++#include "vserver/debug.h"
++
++#include <linux/sched.h>
+
-+ if (type == PIDTYPE_REALPID)
-+ type = PIDTYPE_PID;
- if (pid) {
- struct hlist_node *first;
- first = rcu_dereference(pid->tasks[type].first);
-@@ -309,7 +313,17 @@ struct task_struct * fastcall pid_task(s
- */
- struct task_struct *find_task_by_pid_type(int type, int nr)
- {
-- return pid_task(find_pid(nr), type);
-+ struct task_struct *task;
+
-+ if (type == PIDTYPE_PID)
-+ nr = vx_rmap_pid(nr);
++#define get_nx_info(i) __get_nx_info(i, __FILE__, __LINE__)
+
-+ task = pid_task(find_pid(nr), type);
-+ if (task && (type != PIDTYPE_REALPID) &&
-+ /* maybe VS_WATCH_P in the future? */
-+ !vx_check(task->xid, VS_WATCH|VS_IDENT))
++static inline struct nx_info *__get_nx_info(struct nx_info *nxi,
++ const char *_file, int _line)
++{
++ if (!nxi)
+ return NULL;
-+ return task;
- }
-
- EXPORT_SYMBOL(find_task_by_pid_type);
---- linux-2.6.22/kernel/posix-timers.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/posix-timers.c 2007-07-07 03:52:54 +0200
-@@ -47,6 +47,7 @@
- #include <linux/wait.h>
- #include <linux/workqueue.h>
- #include <linux/module.h>
-+#include <linux/vs_context.h>
-
- /*
- * Management arrays for POSIX timers. Timers are kept in slab memory
-@@ -297,6 +298,10 @@ void do_schedule_next_timer(struct sigin
-
- int posix_timer_event(struct k_itimer *timr,int si_private)
- {
-+ struct vx_info_save vxis;
-+ int ret;
-+
-+ enter_vx_info(task_get_vx_info(timr->it_process), &vxis);
- memset(&timr->sigq->info, 0, sizeof(siginfo_t));
- timr->sigq->info.si_sys_private = si_private;
- /* Send signal to the process that owns this timer.*/
-@@ -309,11 +314,11 @@ int posix_timer_event(struct k_itimer *t
-
- if (timr->it_sigev_notify & SIGEV_THREAD_ID) {
- struct task_struct *leader;
-- int ret = send_sigqueue(timr->it_sigev_signo, timr->sigq,
-- timr->it_process);
-
-+ ret = send_sigqueue(timr->it_sigev_signo, timr->sigq,
-+ timr->it_process);
- if (likely(ret >= 0))
-- return ret;
-+ goto out;
-
- timr->it_sigev_notify = SIGEV_SIGNAL;
- leader = timr->it_process->group_leader;
-@@ -321,8 +326,12 @@ int posix_timer_event(struct k_itimer *t
- timr->it_process = leader;
- }
-
-- return send_group_sigqueue(timr->it_sigev_signo, timr->sigq,
-+ ret = send_group_sigqueue(timr->it_sigev_signo, timr->sigq,
- timr->it_process);
-+out:
-+ leave_vx_info(&vxis);
-+ put_vx_info(vxis.vxi);
-+ return ret;
- }
- EXPORT_SYMBOL_GPL(posix_timer_event);
-
-@@ -402,7 +411,7 @@ static struct task_struct * good_sigeven
- struct task_struct *rtn = current->group_leader;
-
- if ((event->sigev_notify & SIGEV_THREAD_ID ) &&
-- (!(rtn = find_task_by_pid(event->sigev_notify_thread_id)) ||
-+ (!(rtn = find_task_by_real_pid(event->sigev_notify_thread_id)) ||
- rtn->tgid != current->tgid ||
- (event->sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_SIGNAL))
- return NULL;
---- linux-2.6.22/kernel/printk.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/printk.c 2007-06-15 02:37:04 +0200
-@@ -31,6 +31,7 @@
- #include <linux/bootmem.h>
- #include <linux/syscalls.h>
- #include <linux/jiffies.h>
-+#include <linux/vs_cvirt.h>
-
- #include <asm/uaccess.h>
-
-@@ -182,18 +183,13 @@ int do_syslog(int type, char __user *buf
- unsigned long i, j, limit, count;
- int do_clear = 0;
- char c;
-- int error = 0;
-+ int error;
-
- error = security_syslog(type);
- if (error)
- return error;
-
-- switch (type) {
-- case 0: /* Close log */
-- break;
-- case 1: /* Open log */
-- break;
-- case 2: /* Read from log */
-+ if ((type >= 2) && (type <= 4)) {
- error = -EINVAL;
- if (!buf || len < 0)
- goto out;
-@@ -204,6 +200,16 @@ int do_syslog(int type, char __user *buf
- error = -EFAULT;
- goto out;
- }
-+ }
-+ if (!vx_check(0, VS_ADMIN|VS_WATCH))
-+ return vx_do_syslog(type, buf, len);
+
-+ switch (type) {
-+ case 0: /* Close log */
-+ break;
-+ case 1: /* Open log */
-+ break;
-+ case 2: /* Read from log */
- error = wait_event_interruptible(log_wait,
- (log_start - log_end));
- if (error)
-@@ -228,16 +234,6 @@ int do_syslog(int type, char __user *buf
- do_clear = 1;
- /* FALL THRU */
- case 3: /* Read last kernel messages */
-- error = -EINVAL;
-- if (!buf || len < 0)
-- goto out;
-- error = 0;
-- if (!len)
-- goto out;
-- if (!access_ok(VERIFY_WRITE, buf, len)) {
-- error = -EFAULT;
-- goto out;
-- }
- count = len;
- if (count > log_buf_len)
- count = log_buf_len;
---- linux-2.6.22/kernel/ptrace.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/ptrace.c 2007-06-15 04:28:02 +0200
-@@ -19,6 +19,7 @@
- #include <linux/security.h>
- #include <linux/signal.h>
- #include <linux/audit.h>
-+#include <linux/vs_context.h>
-
- #include <asm/pgtable.h>
- #include <asm/uaccess.h>
-@@ -145,6 +146,11 @@ static int may_attach(struct task_struct
- dumpable = task->mm->dumpable;
- if (!dumpable && !capable(CAP_SYS_PTRACE))
- return -EPERM;
-+ if (!vx_check(task->xid, VS_ADMIN_P|VS_IDENT))
-+ return -EPERM;
-+ if (!vx_check(task->xid, VS_IDENT) &&
-+ !task_vx_flags(task, VXF_STATE_ADMIN, 0))
-+ return -EACCES;
-
- return security_ptrace(current, task);
- }
-@@ -471,6 +477,10 @@ asmlinkage long sys_ptrace(long request,
- goto out;
- }
-
-+ ret = -EPERM;
-+ if (!vx_check(vx_task_xid(child), VS_WATCH_P | VS_IDENT))
-+ goto out_put_task_struct;
++ vxlprintk(VXD_CBIT(nid, 2), "get_nx_info(%p[#%d.%d])",
++ nxi, nxi ? nxi->nx_id : 0,
++ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
++ _file, _line);
+
- if (request == PTRACE_ATTACH) {
- ret = ptrace_attach(child);
- goto out_put_task_struct;
---- linux-2.6.22/kernel/sched.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/sched.c 2007-07-07 03:52:54 +0200
-@@ -56,6 +56,8 @@
-
- #include <asm/tlb.h>
- #include <asm/unistd.h>
-+#include <linux/vs_sched.h>
-+#include <linux/vs_cvirt.h>
-
- /*
- * Scheduler clock - returns current time in nanosec units.
-@@ -281,6 +283,16 @@ struct rq {
- struct task_struct *migration_thread;
- struct list_head migration_queue;
- #endif
-+ unsigned long norm_time;
-+ unsigned long idle_time;
-+#ifdef CONFIG_VSERVER_IDLETIME
-+ int idle_skip;
-+#endif
-+#ifdef CONFIG_VSERVER_HARDCPU
-+ struct list_head hold_queue;
-+ unsigned long nr_onhold;
-+ int idle_tokens;
-+#endif
-
- #ifdef CONFIG_SCHEDSTATS
- /* latency stats */
-@@ -714,6 +726,7 @@ sched_info_switch(struct task_struct *pr
- */
- static void dequeue_task(struct task_struct *p, struct prio_array *array)
- {
-+ BUG_ON(p->state & TASK_ONHOLD);
- array->nr_active--;
- list_del(&p->run_list);
- if (list_empty(array->queue + p->prio))
-@@ -722,6 +735,7 @@ static void dequeue_task(struct task_str
-
- static void enqueue_task(struct task_struct *p, struct prio_array *array)
- {
-+ BUG_ON(p->state & TASK_ONHOLD);
- sched_info_queued(p);
- list_add_tail(&p->run_list, array->queue + p->prio);
- __set_bit(p->prio, array->bitmap);
-@@ -735,12 +749,14 @@ static void enqueue_task(struct task_str
- */
- static void requeue_task(struct task_struct *p, struct prio_array *array)
- {
-+ BUG_ON(p->state & TASK_ONHOLD);
- list_move_tail(&p->run_list, array->queue + p->prio);
- }
-
- static inline void
- enqueue_task_head(struct task_struct *p, struct prio_array *array)
- {
-+ BUG_ON(p->state & TASK_ONHOLD);
- list_add(&p->run_list, array->queue + p->prio);
- __set_bit(p->prio, array->bitmap);
- array->nr_active++;
-@@ -769,6 +785,10 @@ static inline int __normal_prio(struct t
- bonus = CURRENT_BONUS(p) - MAX_BONUS / 2;
-
- prio = p->static_prio - bonus;
++ atomic_inc(&nxi->nx_usecnt);
++ return nxi;
++}
+
-+ /* adjust effective priority */
-+ prio = vx_adjust_prio(p, prio, MAX_USER_PRIO);
+
- if (prio < MAX_RT_PRIO)
- prio = MAX_RT_PRIO;
- if (prio > MAX_PRIO-1)
-@@ -878,6 +898,9 @@ static int effective_prio(struct task_st
- return p->prio;
- }
-
-+#include "sched_mon.h"
++extern void free_nx_info(struct nx_info *);
+
++#define put_nx_info(i) __put_nx_info(i, __FILE__, __LINE__)
+
- /*
- * __activate_task - move a task to the runqueue.
- */
-@@ -887,6 +910,7 @@ static void __activate_task(struct task_
-
- if (batch_task(p))
- target = rq->expired;
-+ vxm_activate_task(p, rq);
- enqueue_task(p, target);
- inc_nr_running(p, rq);
- }
-@@ -896,6 +920,7 @@ static void __activate_task(struct task_
- */
- static inline void __activate_idle_task(struct task_struct *p, struct rq *rq)
- {
-+ vxm_activate_idle(p, rq);
- enqueue_task_head(p, rq->active);
- inc_nr_running(p, rq);
- }
-@@ -1030,19 +1055,30 @@ static void activate_task(struct task_st
- }
- p->timestamp = now;
- out:
-+ vx_activate_task(p);
- __activate_task(p, rq);
- }
-
- /*
-- * deactivate_task - remove a task from the runqueue.
-+ * __deactivate_task - remove a task from the runqueue.
- */
--static void deactivate_task(struct task_struct *p, struct rq *rq)
-+static void __deactivate_task(struct task_struct *p, struct rq *rq)
- {
- dec_nr_running(p, rq);
- dequeue_task(p, p->array);
-+ vxm_deactivate_task(p, rq);
- p->array = NULL;
- }
-
-+static inline
-+void deactivate_task(struct task_struct *p, struct rq *rq)
++static inline void __put_nx_info(struct nx_info *nxi, const char *_file, int _line)
+{
-+ vx_deactivate_task(p);
-+ __deactivate_task(p, rq);
++ if (!nxi)
++ return;
++
++ vxlprintk(VXD_CBIT(nid, 2), "put_nx_info(%p[#%d.%d])",
++ nxi, nxi ? nxi->nx_id : 0,
++ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
++ _file, _line);
++
++ if (atomic_dec_and_test(&nxi->nx_usecnt))
++ free_nx_info(nxi);
+}
+
-+#include "sched_hard.h"
+
- /*
- * resched_task - mark a task 'to be rescheduled now'.
- *
-@@ -1129,6 +1165,7 @@ migrate_task(struct task_struct *p, int
- {
- struct rq *rq = task_rq(p);
-
-+ vxm_migrate_task(p, rq, dest_cpu);
- /*
- * If the task is not on a runqueue (and not running), then
- * it is sufficient to simply update the task's cpu field.
-@@ -1518,6 +1555,12 @@ static int try_to_wake_up(struct task_st
-
- rq = task_rq_lock(p, &flags);
- old_state = p->state;
++#define init_nx_info(p, i) __init_nx_info(p, i, __FILE__, __LINE__)
+
-+ /* we need to unhold suspended tasks */
-+ if (old_state & TASK_ONHOLD) {
-+ vx_unhold_task(p, rq);
-+ old_state = p->state;
++static inline void __init_nx_info(struct nx_info **nxp, struct nx_info *nxi,
++ const char *_file, int _line)
++{
++ if (nxi) {
++ vxlprintk(VXD_CBIT(nid, 3),
++ "init_nx_info(%p[#%d.%d])",
++ nxi, nxi ? nxi->nx_id : 0,
++ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
++ _file, _line);
++
++ atomic_inc(&nxi->nx_usecnt);
+ }
- if (!(old_state & state))
- goto out;
-
-@@ -1625,6 +1668,7 @@ out_activate:
- #endif /* CONFIG_SMP */
- if (old_state == TASK_UNINTERRUPTIBLE) {
- rq->nr_uninterruptible--;
-+ vx_uninterruptible_dec(p);
- /*
- * Tasks on involuntary sleep don't earn
- * sleep_avg beyond just interactive state.
-@@ -1676,7 +1720,7 @@ int fastcall wake_up_state(struct task_s
- return try_to_wake_up(p, state, 0);
- }
-
--static void task_running_tick(struct rq *rq, struct task_struct *p);
-+static void task_running_tick(struct rq *rq, struct task_struct *p, int cpu);
- /*
- * Perform scheduler related setup for a newly forked process p.
- * p is forked by current.
-@@ -1737,7 +1781,7 @@ void fastcall sched_fork(struct task_str
- * runqueue lock is not a problem.
- */
- current->time_slice = 1;
-- task_running_tick(cpu_rq(cpu), current);
-+ task_running_tick(cpu_rq(cpu), current, cpu);
- }
- local_irq_enable();
- put_cpu();
-@@ -1772,6 +1816,7 @@ void fastcall wake_up_new_task(struct ta
-
- p->prio = effective_prio(p);
-
-+ vx_activate_task(p);
- if (likely(cpu == this_cpu)) {
- if (!(clone_flags & CLONE_VM)) {
- /*
-@@ -1783,6 +1828,7 @@ void fastcall wake_up_new_task(struct ta
- __activate_task(p, rq);
- else {
- p->prio = current->prio;
-+ BUG_ON(p->state & TASK_ONHOLD);
- p->normal_prio = current->normal_prio;
- list_add_tail(&p->run_list, ¤t->run_list);
- p->array = current->array;
-@@ -3351,13 +3397,16 @@ static inline int expired_starving(struc
- void account_user_time(struct task_struct *p, cputime_t cputime)
- {
- struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
-+ struct vx_info *vxi = p->vx_info; /* p is _always_ current */
- cputime64_t tmp;
-+ int nice = (TASK_NICE(p) > 0);
-
- p->utime = cputime_add(p->utime, cputime);
-+ vx_account_user(vxi, cputime, nice);
-
- /* Add user time to cpustat. */
- tmp = cputime_to_cputime64(cputime);
-- if (TASK_NICE(p) > 0)
-+ if (nice)
- cpustat->nice = cputime64_add(cpustat->nice, tmp);
- else
- cpustat->user = cputime64_add(cpustat->user, tmp);
-@@ -3373,10 +3422,12 @@ void account_system_time(struct task_str
- cputime_t cputime)
- {
- struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
-+ struct vx_info *vxi = p->vx_info; /* p is _always_ current */
- struct rq *rq = this_rq();
- cputime64_t tmp;
-
- p->stime = cputime_add(p->stime, cputime);
-+ vx_account_system(vxi, cputime, (p == rq->idle));
-
- /* Add system time to cpustat. */
- tmp = cputime_to_cputime64(cputime);
-@@ -3415,7 +3466,7 @@ void account_steal_time(struct task_stru
- cpustat->steal = cputime64_add(cpustat->steal, tmp);
- }
-
--static void task_running_tick(struct rq *rq, struct task_struct *p)
-+static void task_running_tick(struct rq *rq, struct task_struct *p, int cpu)
- {
- if (p->array != rq->active) {
- /* Task has expired but was not scheduled yet */
-@@ -3445,7 +3496,7 @@ static void task_running_tick(struct rq
- }
- goto out_unlock;
- }
-- if (!--p->time_slice) {
-+ if (vx_need_resched(p, --p->time_slice, cpu)) {
- dequeue_task(p, rq->active);
- set_tsk_need_resched(p);
- p->prio = effective_prio(p);
-@@ -3506,9 +3557,12 @@ void scheduler_tick(void)
- struct rq *rq = cpu_rq(cpu);
-
- update_cpu_clock(p, rq, now);
--
-- if (!idle_at_tick)
-- task_running_tick(rq, p);
-+ vxm_sync(now, cpu);
++ *nxp = nxi;
++}
+
-+ if (idle_at_tick)
-+ vx_idle_resched(rq);
-+ else
-+ task_running_tick(rq, p, cpu);
- #ifdef CONFIG_SMP
- update_load(rq);
- rq->idle_at_tick = idle_at_tick;
-@@ -3630,14 +3684,25 @@ need_resched_nonpreemptible:
- unlikely(signal_pending(prev))))
- prev->state = TASK_RUNNING;
- else {
-- if (prev->state == TASK_UNINTERRUPTIBLE)
-+ if (prev->state == TASK_UNINTERRUPTIBLE) {
- rq->nr_uninterruptible++;
-+ vx_uninterruptible_inc(prev);
-+ }
- deactivate_task(prev, rq);
- }
- }
-
- cpu = smp_processor_id();
-+ vx_set_rq_time(rq, jiffies);
-+try_unhold:
-+ vx_try_unhold(rq, cpu);
-+pick_next:
-+
- if (unlikely(!rq->nr_running)) {
-+ /* can we skip idle time? */
-+ if (vx_try_skip(rq, cpu))
-+ goto try_unhold;
-+
- idle_balance(cpu, rq);
- if (!rq->nr_running) {
- next = rq->idle;
-@@ -3663,6 +3728,10 @@ need_resched_nonpreemptible:
- queue = array->queue + idx;
- next = list_entry(queue->next, struct task_struct, run_list);
-
-+ /* check before we schedule this context */
-+ if (!vx_schedule(next, rq, cpu))
-+ goto pick_next;
-+
- if (!rt_task(next) && interactive_sleep(next->sleep_type)) {
- unsigned long long delta = now - next->timestamp;
- if (unlikely((long long)(now - next->timestamp) < 0))
-@@ -4263,7 +4332,7 @@ asmlinkage long sys_nice(int increment)
- nice = 19;
-
- if (increment < 0 && !can_nice(current, nice))
-- return -EPERM;
-+ return vx_flags(VXF_IGNEG_NICE, 0) ? 0 : -EPERM;
-
- retval = security_task_setnice(current, nice);
- if (retval)
-@@ -4435,6 +4504,7 @@ recheck:
- oldprio = p->prio;
- __setscheduler(p, policy, param->sched_priority);
- if (array) {
-+ vx_activate_task(p);
- __activate_task(p, rq);
- /*
- * Reschedule if we are currently running on this runqueue and
-@@ -5188,6 +5258,7 @@ static int __migrate_task(struct task_st
- p->timestamp = p->timestamp - rq_src->most_recent_timestamp
- + rq_dest->most_recent_timestamp;
- deactivate_task(p, rq_src);
-+ vx_activate_task(p);
- __activate_task(p, rq_dest);
- if (TASK_PREEMPTS_CURR(p, rq_dest))
- resched_task(rq_dest->curr);
-@@ -7058,7 +7129,10 @@ void __init sched_init(void)
- INIT_LIST_HEAD(&rq->migration_queue);
- #endif
- atomic_set(&rq->nr_iowait, 0);
--
-+#ifdef CONFIG_VSERVER_HARDCPU
-+ INIT_LIST_HEAD(&rq->hold_queue);
-+ rq->nr_onhold = 0;
-+#endif
- for (j = 0; j < 2; j++) {
- array = rq->arrays + j;
- for (k = 0; k < MAX_PRIO; k++) {
-@@ -7144,6 +7218,7 @@ void normalize_rt_tasks(void)
- deactivate_task(p, task_rq(p));
- __setscheduler(p, SCHED_NORMAL, 0);
- if (array) {
-+ vx_activate_task(p);
- __activate_task(p, task_rq(p));
- resched_task(rq->curr);
- }
---- linux-2.6.22/kernel/sched_hard.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/sched_hard.h 2007-06-15 19:20:10 +0200
-@@ -0,0 +1,324 @@
+
-+#ifdef CONFIG_VSERVER_IDLELIMIT
++#define set_nx_info(p, i) __set_nx_info(p, i, __FILE__, __LINE__)
+
-+/*
-+ * vx_idle_resched - reschedule after maxidle
-+ */
-+static inline
-+void vx_idle_resched(struct rq *rq)
++static inline void __set_nx_info(struct nx_info **nxp, struct nx_info *nxi,
++ const char *_file, int _line)
+{
-+ /* maybe have a better criterion for paused */
-+ if (!--rq->idle_tokens && !list_empty(&rq->hold_queue))
-+ set_need_resched();
-+}
-+
-+#else /* !CONFIG_VSERVER_IDLELIMIT */
++ struct nx_info *nxo;
+
-+#define vx_idle_resched(rq)
++ if (!nxi)
++ return;
+
-+#endif /* CONFIG_VSERVER_IDLELIMIT */
++ vxlprintk(VXD_CBIT(nid, 3), "set_nx_info(%p[#%d.%d])",
++ nxi, nxi ? nxi->nx_id : 0,
++ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
++ _file, _line);
+
++ atomic_inc(&nxi->nx_usecnt);
++ nxo = xchg(nxp, nxi);
++ BUG_ON(nxo);
++}
+
++#define clr_nx_info(p) __clr_nx_info(p, __FILE__, __LINE__)
+
-+#ifdef CONFIG_VSERVER_IDLETIME
++static inline void __clr_nx_info(struct nx_info **nxp,
++ const char *_file, int _line)
++{
++ struct nx_info *nxo;
+
-+#define vx_set_rq_min_skip(rq, min) \
-+ (rq)->idle_skip = (min)
++ nxo = xchg(nxp, NULL);
++ if (!nxo)
++ return;
+
-+#define vx_save_min_skip(ret, min, val) \
-+ __vx_save_min_skip(ret, min, val)
++ vxlprintk(VXD_CBIT(nid, 3), "clr_nx_info(%p[#%d.%d])",
++ nxo, nxo ? nxo->nx_id : 0,
++ nxo ? atomic_read(&nxo->nx_usecnt) : 0,
++ _file, _line);
+
-+static inline
-+void __vx_save_min_skip(int ret, int *min, int val)
-+{
-+ if (ret > -2)
-+ return;
-+ if ((*min > val) || !*min)
-+ *min = val;
++ if (atomic_dec_and_test(&nxo->nx_usecnt))
++ free_nx_info(nxo);
+}
+
-+static inline
-+int vx_try_skip(struct rq *rq, int cpu)
++
++#define claim_nx_info(v, p) __claim_nx_info(v, p, __FILE__, __LINE__)
++
++static inline void __claim_nx_info(struct nx_info *nxi,
++ struct task_struct *task, const char *_file, int _line)
+{
-+ /* artificially advance time */
-+ if (rq->idle_skip > 0) {
-+ vxdprintk(list_empty(&rq->hold_queue),
-+ "hold queue empty on cpu %d", cpu);
-+ rq->idle_time += rq->idle_skip;
-+ vxm_idle_skip(rq, cpu);
-+ return 1;
-+ }
-+ return 0;
++ vxlprintk(VXD_CBIT(nid, 3), "claim_nx_info(%p[#%d.%d.%d]) %p",
++ nxi, nxi ? nxi->nx_id : 0,
++ nxi?atomic_read(&nxi->nx_usecnt):0,
++ nxi?atomic_read(&nxi->nx_tasks):0,
++ task, _file, _line);
++
++ atomic_inc(&nxi->nx_tasks);
+}
+
-+#else /* !CONFIG_VSERVER_IDLETIME */
+
-+#define vx_set_rq_min_skip(rq, min) \
-+ ({ int dummy = (min); dummy; })
++extern void unhash_nx_info(struct nx_info *);
+
-+#define vx_save_min_skip(ret, min, val)
++#define release_nx_info(v, p) __release_nx_info(v, p, __FILE__, __LINE__)
+
-+static inline
-+int vx_try_skip(struct rq *rq, int cpu)
++static inline void __release_nx_info(struct nx_info *nxi,
++ struct task_struct *task, const char *_file, int _line)
+{
-+ return 0;
-+}
-+
-+#endif /* CONFIG_VSERVER_IDLETIME */
-+
++ vxlprintk(VXD_CBIT(nid, 3), "release_nx_info(%p[#%d.%d.%d]) %p",
++ nxi, nxi ? nxi->nx_id : 0,
++ nxi ? atomic_read(&nxi->nx_usecnt) : 0,
++ nxi ? atomic_read(&nxi->nx_tasks) : 0,
++ task, _file, _line);
+
++ might_sleep();
+
-+#ifdef CONFIG_VSERVER_HARDCPU
++ if (atomic_dec_and_test(&nxi->nx_tasks))
++ unhash_nx_info(nxi);
++}
+
-+#define vx_set_rq_max_idle(rq, max) \
-+ (rq)->idle_tokens = (max)
+
-+#define vx_save_max_idle(ret, min, val) \
-+ __vx_save_max_idle(ret, min, val)
++#define task_get_nx_info(i) __task_get_nx_info(i, __FILE__, __LINE__)
+
-+static inline
-+void __vx_save_max_idle(int ret, int *min, int val)
++static __inline__ struct nx_info *__task_get_nx_info(struct task_struct *p,
++ const char *_file, int _line)
+{
-+ if (*min > val)
-+ *min = val;
++ struct nx_info *nxi;
++
++ task_lock(p);
++ vxlprintk(VXD_CBIT(nid, 5), "task_get_nx_info(%p)",
++ p, _file, _line);
++ nxi = __get_nx_info(p->nx_info, _file, _line);
++ task_unlock(p);
++ return nxi;
+}
+
+
-+/*
-+ * vx_hold_task - put a task on the hold queue
-+ */
-+static inline
-+void vx_hold_task(struct task_struct *p, struct rq *rq)
++static inline void exit_nx_info(struct task_struct *p)
+{
-+ __deactivate_task(p, rq);
-+ p->state |= TASK_ONHOLD;
-+ /* a new one on hold */
-+ rq->nr_onhold++;
-+ vxm_hold_task(p, rq);
-+ list_add_tail(&p->run_list, &rq->hold_queue);
++ if (p->nx_info)
++ release_nx_info(p->nx_info, p);
+}
+
-+/*
-+ * vx_unhold_task - put a task back to the runqueue
-+ */
-+static inline
-+void vx_unhold_task(struct task_struct *p, struct rq *rq)
-+{
-+ list_del(&p->run_list);
-+ /* one less waiting */
-+ rq->nr_onhold--;
-+ p->state &= ~TASK_ONHOLD;
-+ enqueue_task(p, rq->expired);
-+ inc_nr_running(p, rq);
-+ vxm_unhold_task(p, rq);
+
-+ if (p->static_prio < rq->best_expired_prio)
-+ rq->best_expired_prio = p->static_prio;
-+}
++#else
++#warning duplicate inclusion
++#endif
+--- linux-2.6.25.4/include/linux/vs_pid.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_pid.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,95 @@
++#ifndef _VS_PID_H
++#define _VS_PID_H
+
-+unsigned long nr_onhold(void)
-+{
-+ unsigned long i, sum = 0;
++#include "vserver/base.h"
++#include "vserver/check.h"
++#include "vserver/context.h"
++#include "vserver/debug.h"
++#include "vserver/pid.h"
++#include <linux/pid_namespace.h>
+
-+ for_each_online_cpu(i)
-+ sum += cpu_rq(i)->nr_onhold;
+
-+ return sum;
++#define VXF_FAKE_INIT (VXF_INFO_INIT | VXF_STATE_INIT)
++
++static inline
++int vx_proc_task_visible(struct task_struct *task)
++{
++ if ((task->pid == 1) &&
++ !vx_flags(VXF_FAKE_INIT, VXF_FAKE_INIT))
++ /* show a blend through init */
++ goto visible;
++ if (vx_check(vx_task_xid(task), VS_WATCH | VS_IDENT))
++ goto visible;
++ return 0;
++visible:
++ return 1;
+}
+
++#define find_task_by_real_pid find_task_by_pid
+
++#if 0
+
+static inline
-+int __vx_tokens_avail(struct _vx_sched_pc *sched_pc)
++struct task_struct *vx_find_proc_task_by_pid(int pid)
+{
-+ return sched_pc->tokens;
++ struct task_struct *task = find_task_by_real_pid(pid);
++
++ if (task && !vx_proc_task_visible(task)) {
++ vxdprintk(VXD_CBIT(misc, 6),
++ "dropping task (find) %p[#%u,%u] for %p[#%u,%u]",
++ task, task->xid, task->pid,
++ current, current->xid, current->pid);
++ task = NULL;
++ }
++ return task;
+}
+
++#endif
++
+static inline
-+void __vx_consume_token(struct _vx_sched_pc *sched_pc)
++struct task_struct *vx_get_proc_task(struct inode *inode, struct pid *pid)
+{
-+ sched_pc->tokens--;
++ struct task_struct *task = get_pid_task(pid, PIDTYPE_PID);
++
++ if (task && !vx_proc_task_visible(task)) {
++ vxdprintk(VXD_CBIT(misc, 6),
++ "dropping task (get) %p[#%u,%u] for %p[#%u,%u]",
++ task, task->xid, task->pid,
++ current, current->xid, current->pid);
++ put_task_struct(task);
++ task = NULL;
++ }
++ return task;
+}
+
++#if 0
++
+static inline
-+int vx_need_resched(struct task_struct *p, int slice, int cpu)
++struct task_struct *vx_child_reaper(struct task_struct *p)
+{
+ struct vx_info *vxi = p->vx_info;
++ struct task_struct *reaper = child_reaper(p);
+
-+ if (vx_info_flags(vxi, VXF_SCHED_HARD|VXF_SCHED_PRIO, 0)) {
-+ struct _vx_sched_pc *sched_pc =
-+ &vx_per_cpu(vxi, sched_pc, cpu);
-+ int tokens;
++ if (!vxi)
++ goto out;
+
-+ /* maybe we can simplify that to decrement
-+ the token counter unconditional? */
++ BUG_ON(!p->vx_info->vx_reaper);
+
-+ if ((tokens = __vx_tokens_avail(sched_pc)) > 0)
-+ __vx_consume_token(sched_pc);
++ /* child reaper for the guest reaper */
++ if (vxi->vx_reaper == p)
++ goto out;
+
-+ /* for tokens > 0, one token was consumed */
-+ if (tokens < 2)
-+ slice = 0;
-+ }
-+ vxm_need_resched(p, slice, cpu);
-+ return (slice == 0);
++ reaper = vxi->vx_reaper;
++out:
++ vxdprintk(VXD_CBIT(xid, 7),
++ "vx_child_reaper(%p[#%u,%u]) = %p[#%u,%u]",
++ p, p->xid, p->pid, reaper, reaper->xid, reaper->pid);
++ return reaper;
+}
+
++#endif
+
-+#define vx_set_rq_time(rq, time) do { \
-+ rq->norm_time = time; \
-+} while (0)
+
++#else
++#warning duplicate inclusion
++#endif
+--- linux-2.6.25.4/include/linux/vs_sched.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_sched.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,110 @@
++#ifndef _VS_SCHED_H
++#define _VS_SCHED_H
+
-+static inline
-+void vx_try_unhold(struct rq *rq, int cpu)
-+{
-+ struct vx_info *vxi = NULL;
-+ struct list_head *l, *n;
-+ int maxidle = HZ;
-+ int minskip = 0;
++#include "vserver/base.h"
++#include "vserver/context.h"
++#include "vserver/sched.h"
+
-+ /* nothing to do? what about pause? */
-+ if (list_empty(&rq->hold_queue))
-+ return;
+
-+ list_for_each_safe(l, n, &rq->hold_queue) {
-+ int ret, delta_min[2];
-+ struct _vx_sched_pc *sched_pc;
-+ struct task_struct *p;
++#define VAVAVOOM_RATIO 50
+
-+ p = list_entry(l, struct task_struct, run_list);
-+ /* don't bother with same context */
-+ if (vxi == p->vx_info)
-+ continue;
++#define MAX_PRIO_BIAS 20
++#define MIN_PRIO_BIAS -20
+
-+ vxi = p->vx_info;
-+ /* ignore paused contexts */
-+ if (vx_info_flags(vxi, VXF_SCHED_PAUSE, 0))
-+ continue;
+
-+ sched_pc = &vx_per_cpu(vxi, sched_pc, cpu);
++#ifdef CONFIG_VSERVER_HARDCPU
+
-+ /* recalc tokens */
-+ vxm_sched_info(sched_pc, vxi, cpu);
-+ ret = vx_tokens_recalc(sched_pc,
-+ &rq->norm_time, &rq->idle_time, delta_min);
-+ vxm_tokens_recalc(sched_pc, rq, vxi, cpu);
++/*
++ * effective_prio - return the priority that is based on the static
++ * priority but is modified by bonuses/penalties.
++ *
++ * We scale the actual sleep average [0 .... MAX_SLEEP_AVG]
++ * into a -4 ... 0 ... +4 bonus/penalty range.
++ *
++ * Additionally, we scale another amount based on the number of
++ * CPU tokens currently held by the context, if the process is
++ * part of a context (and the appropriate SCHED flag is set).
++ * This ranges from -5 ... 0 ... +15, quadratically.
++ *
++ * So, the total bonus is -9 .. 0 .. +19
++ * We use ~50% of the full 0...39 priority range so that:
++ *
++ * 1) nice +19 interactive tasks do not preempt nice 0 CPU hogs.
++ * 2) nice -20 CPU hogs do not get preempted by nice 0 tasks.
++ * unless that context is far exceeding its CPU allocation.
++ *
++ * Both properties are important to certain workloads.
++ */
++static inline
++int vx_effective_vavavoom(struct _vx_sched_pc *sched_pc, int max_prio)
++{
++ int vavavoom, max;
+
-+ if (ret > 0) {
-+ /* we found a runable context */
-+ vx_unhold_task(p, rq);
-+ break;
-+ }
-+ vx_save_max_idle(ret, &maxidle, delta_min[0]);
-+ vx_save_min_skip(ret, &minskip, delta_min[1]);
++ /* lots of tokens = lots of vavavoom
++ * no tokens = no vavavoom */
++ if ((vavavoom = sched_pc->tokens) >= 0) {
++ max = sched_pc->tokens_max;
++ vavavoom = max - vavavoom;
++ max = max * max;
++ vavavoom = max_prio * VAVAVOOM_RATIO / 100
++ * (vavavoom*vavavoom - (max >> 2)) / max;
++ return vavavoom;
+ }
-+ vx_set_rq_max_idle(rq, maxidle);
-+ vx_set_rq_min_skip(rq, minskip);
-+ vxm_rq_max_min(rq, cpu);
++ return 0;
+}
+
+
+static inline
-+int vx_schedule(struct task_struct *next, struct rq *rq, int cpu)
++int vx_adjust_prio(struct task_struct *p, int prio, int max_user)
+{
-+ struct vx_info *vxi = next->vx_info;
++ struct vx_info *vxi = p->vx_info;
+ struct _vx_sched_pc *sched_pc;
-+ int delta_min[2];
-+ int flags, ret;
+
+ if (!vxi)
-+ return 1;
-+
-+ flags = vxi->vx_flags;
-+
-+ if (unlikely(vs_check_flags(flags, VXF_SCHED_PAUSE, 0)))
-+ goto put_on_hold;
-+ if (!vs_check_flags(flags, VXF_SCHED_HARD | VXF_SCHED_PRIO, 0))
-+ return 1;
-+
-+ sched_pc = &vx_per_cpu(vxi, sched_pc, cpu);
-+#ifdef CONFIG_SMP
-+ /* update scheduler params */
-+ if (cpu_isset(cpu, vxi->sched.update)) {
-+ vx_update_sched_param(&vxi->sched, sched_pc);
-+ vxm_update_sched(sched_pc, vxi, cpu);
-+ cpu_clear(cpu, vxi->sched.update);
-+ }
-+#endif
-+ vxm_sched_info(sched_pc, vxi, cpu);
-+ ret = vx_tokens_recalc(sched_pc,
-+ &rq->norm_time, &rq->idle_time, delta_min);
-+ vxm_tokens_recalc(sched_pc, rq, vxi, cpu);
++ return prio;
+
-+ if (!vs_check_flags(flags, VXF_SCHED_HARD, 0))
-+ return 1;
++ sched_pc = &vx_cpu(vxi, sched_pc);
++ if (vx_info_flags(vxi, VXF_SCHED_PRIO, 0)) {
++ int vavavoom = vx_effective_vavavoom(sched_pc, max_user);
+
-+ if (unlikely(ret < 0)) {
-+ vx_save_max_idle(ret, &rq->idle_tokens, delta_min[0]);
-+ vx_save_min_skip(ret, &rq->idle_skip, delta_min[1]);
-+ vxm_rq_max_min(rq, cpu);
-+ put_on_hold:
-+ vx_hold_task(next, rq);
-+ return 0;
++ sched_pc->vavavoom = vavavoom;
++ prio += vavavoom;
+ }
-+ return 1;
++ prio += sched_pc->prio_bias;
++ return prio;
+}
+
-+
-+#else /* CONFIG_VSERVER_HARDCPU */
-+
-+static inline
-+void vx_hold_task(struct task_struct *p, struct rq *rq)
-+{
-+ return;
-+}
++#else /* !CONFIG_VSERVER_HARDCPU */
+
+static inline
-+void vx_unhold_task(struct task_struct *p, struct rq *rq)
-+{
-+ return;
-+}
-+
-+unsigned long nr_onhold(void)
++int vx_adjust_prio(struct task_struct *p, int prio, int max_user)
+{
-+ return 0;
-+}
-+
++ struct vx_info *vxi = p->vx_info;
+
-+static inline
-+int vx_need_resched(struct task_struct *p, int slice, int cpu)
-+{
-+ return (slice == 0);
++ if (vxi)
++ prio += vx_cpu(vxi, sched_pc).prio_bias;
++ return prio;
+}
+
++#endif /* CONFIG_VSERVER_HARDCPU */
+
-+#define vx_set_rq_time(rq, time)
+
-+static inline
-+void vx_try_unhold(struct rq *rq, int cpu)
++static inline void vx_account_user(struct vx_info *vxi,
++ cputime_t cputime, int nice)
+{
-+ return;
++ if (!vxi)
++ return;
++ vx_cpu(vxi, sched_pc).user_ticks += cputime;
+}
+
-+static inline
-+int vx_schedule(struct task_struct *next, struct rq *rq, int cpu)
++static inline void vx_account_system(struct vx_info *vxi,
++ cputime_t cputime, int idle)
+{
-+ struct vx_info *vxi = next->vx_info;
-+ struct _vx_sched_pc *sched_pc;
-+ int delta_min[2];
-+ int ret;
-+
-+ if (!vx_info_flags(vxi, VXF_SCHED_PRIO, 0))
-+ return 1;
-+
-+ sched_pc = &vx_per_cpu(vxi, sched_pc, cpu);
-+ vxm_sched_info(sched_pc, vxi, cpu);
-+ ret = vx_tokens_recalc(sched_pc,
-+ &rq->norm_time, &rq->idle_time, delta_min);
-+ vxm_tokens_recalc(sched_pc, rq, vxi, cpu);
-+ return 1;
++ if (!vxi)
++ return;
++ vx_cpu(vxi, sched_pc).sys_ticks += cputime;
+}
+
-+#endif /* CONFIG_VSERVER_HARDCPU */
-+
---- linux-2.6.22/kernel/sched_mon.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/sched_mon.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,200 @@
-+
-+#include <linux/vserver/monitor.h>
-+
-+#ifdef CONFIG_VSERVER_MONITOR
-+
-+#ifdef CONFIG_VSERVER_HARDCPU
-+#define HARDCPU(x) (x)
+#else
-+#define HARDCPU(x) (0)
++#warning duplicate inclusion
+#endif
+--- linux-2.6.25.4/include/linux/vs_socket.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_socket.h 2008-04-23 14:32:00.000000000 -0400
+@@ -0,0 +1,67 @@
++#ifndef _VS_SOCKET_H
++#define _VS_SOCKET_H
+
-+#ifdef CONFIG_VSERVER_IDLETIME
-+#define IDLETIME(x) (x)
-+#else
-+#define IDLETIME(x) (0)
-+#endif
++#include "vserver/debug.h"
++#include "vserver/base.h"
++#include "vserver/cacct.h"
++#include "vserver/context.h"
++#include "vserver/tag.h"
+
-+struct _vx_mon_entry *vxm_advance(int cpu);
+
++/* socket accounting */
+
-+static inline
-+void __vxm_basic(struct _vx_mon_entry *entry, xid_t xid, int type)
-+{
-+ entry->type = type;
-+ entry->xid = xid;
-+}
++#include <linux/socket.h>
+
-+static inline
-+void __vxm_sync(int cpu)
++static inline int vx_sock_type(int family)
+{
-+ struct _vx_mon_entry *entry = vxm_advance(cpu);
-+
-+ __vxm_basic(entry, 0, VXM_SYNC);
-+ entry->ev.sec = xtime.tv_sec;
-+ entry->ev.nsec = xtime.tv_nsec;
++ switch (family) {
++ case PF_UNSPEC:
++ return VXA_SOCK_UNSPEC;
++ case PF_UNIX:
++ return VXA_SOCK_UNIX;
++ case PF_INET:
++ return VXA_SOCK_INET;
++ case PF_INET6:
++ return VXA_SOCK_INET6;
++ case PF_PACKET:
++ return VXA_SOCK_PACKET;
++ default:
++ return VXA_SOCK_OTHER;
++ }
+}
+
-+static inline
-+void __vxm_task(struct task_struct *p, int type)
-+{
-+ struct _vx_mon_entry *entry = vxm_advance(task_cpu(p));
-+
-+ __vxm_basic(entry, p->xid, type);
-+ entry->ev.tsk.pid = p->pid;
-+ entry->ev.tsk.state = p->state;
-+}
++#define vx_acc_sock(v, f, p, s) \
++ __vx_acc_sock(v, f, p, s, __FILE__, __LINE__)
+
-+static inline
-+void __vxm_sched(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
++static inline void __vx_acc_sock(struct vx_info *vxi,
++ int family, int pos, int size, char *file, int line)
+{
-+ struct _vx_mon_entry *entry = vxm_advance(cpu);
++ if (vxi) {
++ int type = vx_sock_type(family);
+
-+ __vxm_basic(entry, vxi->vx_id, (VXM_SCHED | s->flags));
-+ entry->sd.tokens = s->tokens;
-+ entry->sd.norm_time = s->norm_time;
-+ entry->sd.idle_time = s->idle_time;
++ atomic_long_inc(&vxi->cacct.sock[type][pos].count);
++ atomic_long_add(size, &vxi->cacct.sock[type][pos].total);
++ }
+}
+
-+static inline
-+void __vxm_rqinfo1(struct rq *q, int cpu)
-+{
-+ struct _vx_mon_entry *entry = vxm_advance(cpu);
++#define vx_sock_recv(sk, s) \
++ vx_acc_sock((sk)->sk_vx_info, (sk)->sk_family, 0, s)
++#define vx_sock_send(sk, s) \
++ vx_acc_sock((sk)->sk_vx_info, (sk)->sk_family, 1, s)
++#define vx_sock_fail(sk, s) \
++ vx_acc_sock((sk)->sk_vx_info, (sk)->sk_family, 2, s)
+
-+ entry->type = VXM_RQINFO_1;
-+ entry->xid = ((unsigned long)q >> 16) & 0xffff;
-+ entry->q1.running = q->nr_running;
-+ entry->q1.onhold = HARDCPU(q->nr_onhold);
-+ entry->q1.iowait = atomic_read(&q->nr_iowait);
-+ entry->q1.uintr = q->nr_uninterruptible;
-+ entry->q1.idle_tokens = IDLETIME(q->idle_tokens);
-+}
+
-+static inline
-+void __vxm_rqinfo2(struct rq *q, int cpu)
-+{
-+ struct _vx_mon_entry *entry = vxm_advance(cpu);
++#define sock_vx_init(s) do { \
++ (s)->sk_xid = 0; \
++ (s)->sk_vx_info = NULL; \
++ } while (0)
+
-+ entry->type = VXM_RQINFO_2;
-+ entry->xid = (unsigned long)q & 0xffff;
-+ entry->q2.norm_time = q->norm_time;
-+ entry->q2.idle_time = q->idle_time;
-+ entry->q2.idle_skip = IDLETIME(q->idle_skip);
-+}
++#define sock_nx_init(s) do { \
++ (s)->sk_nid = 0; \
++ (s)->sk_nx_info = NULL; \
++ } while (0)
+
-+static inline
-+void __vxm_update(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
-+{
-+ struct _vx_mon_entry *entry = vxm_advance(cpu);
++#else
++#warning duplicate inclusion
++#endif
+--- linux-2.6.25.4/include/linux/vs_tag.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_tag.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,43 @@
++#ifndef _VS_TAG_H
++#define _VS_TAG_H
+
-+ __vxm_basic(entry, vxi->vx_id, VXM_UPDATE);
-+ entry->ev.tokens = s->tokens;
-+}
++#include <linux/vserver/tag.h>
+
-+static inline
-+void __vxm_update1(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
-+{
-+ struct _vx_mon_entry *entry = vxm_advance(cpu);
++/* check conditions */
+
-+ __vxm_basic(entry, vxi->vx_id, VXM_UPDATE_1);
-+ entry->u1.tokens_max = s->tokens_max;
-+ entry->u1.fill_rate = s->fill_rate[0];
-+ entry->u1.interval = s->interval[0];
-+}
++#define DX_ADMIN 0x0001
++#define DX_WATCH 0x0002
++#define DX_HOSTID 0x0008
+
-+static inline
-+void __vxm_update2(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
-+{
-+ struct _vx_mon_entry *entry = vxm_advance(cpu);
++#define DX_IDENT 0x0010
+
-+ __vxm_basic(entry, vxi->vx_id, VXM_UPDATE_2);
-+ entry->u2.tokens_min = s->tokens_min;
-+ entry->u2.fill_rate = s->fill_rate[1];
-+ entry->u2.interval = s->interval[1];
-+}
++#define DX_ARG_MASK 0x0010
+
+
-+#define vxm_activate_task(p,q) __vxm_task(p, VXM_ACTIVATE)
-+#define vxm_activate_idle(p,q) __vxm_task(p, VXM_IDLE)
-+#define vxm_deactivate_task(p,q) __vxm_task(p, VXM_DEACTIVATE)
-+#define vxm_hold_task(p,q) __vxm_task(p, VXM_HOLD)
-+#define vxm_unhold_task(p,q) __vxm_task(p, VXM_UNHOLD)
++#define dx_task_tag(t) ((t)->tag)
+
-+static inline
-+void vxm_migrate_task(struct task_struct *p, struct rq *rq, int dest)
-+{
-+ __vxm_task(p, VXM_MIGRATE);
-+ __vxm_rqinfo1(rq, task_cpu(p));
-+ __vxm_rqinfo2(rq, task_cpu(p));
-+}
++#define dx_current_tag() dx_task_tag(current)
+
-+static inline
-+void vxm_idle_skip(struct rq *rq, int cpu)
-+{
-+ __vxm_rqinfo1(rq, cpu);
-+ __vxm_rqinfo2(rq, cpu);
-+}
++#define dx_check(c, m) __dx_check(dx_current_tag(), c, m)
+
-+static inline
-+void vxm_need_resched(struct task_struct *p, int slice, int cpu)
-+{
-+ if (slice)
-+ return;
++#define dx_weak_check(c, m) ((m) ? dx_check(c, m) : 1)
+
-+ __vxm_task(p, VXM_RESCHED);
-+}
+
-+static inline
-+void vxm_sync(unsigned long now, int cpu)
++/*
++ * check current context for ADMIN/WATCH and
++ * optionally against supplied argument
++ */
++static inline int __dx_check(tag_t cid, tag_t id, unsigned int mode)
+{
-+ if (!CONFIG_VSERVER_MONITOR_SYNC ||
-+ (now % CONFIG_VSERVER_MONITOR_SYNC))
-+ return;
-+
-+ __vxm_sync(cpu);
++ if (mode & DX_ARG_MASK) {
++ if ((mode & DX_IDENT) && (id == cid))
++ return 1;
++ }
++ return (((mode & DX_ADMIN) && (cid == 0)) ||
++ ((mode & DX_WATCH) && (cid == 1)) ||
++ ((mode & DX_HOSTID) && (id == 0)));
+}
+
-+#define vxm_sched_info(s,v,c) __vxm_sched(s,v,c)
-+
-+static inline
-+void vxm_tokens_recalc(struct _vx_sched_pc *s, struct rq *rq,
-+ struct vx_info *vxi, int cpu)
-+{
-+ __vxm_sched(s, vxi, cpu);
-+ __vxm_rqinfo2(rq, cpu);
-+}
++#else
++#warning duplicate inclusion
++#endif
+--- linux-2.6.25.4/include/linux/vs_time.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/include/linux/vs_time.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,19 @@
++#ifndef _VS_TIME_H
++#define _VS_TIME_H
+
-+static inline
-+void vxm_update_sched(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
-+{
-+ __vxm_sched(s, vxi, cpu);
-+ __vxm_update(s, vxi, cpu);
-+ __vxm_update1(s, vxi, cpu);
-+ __vxm_update2(s, vxi, cpu);
-+}
+
-+static inline
-+void vxm_rq_max_min(struct rq *rq, int cpu)
-+{
-+ __vxm_rqinfo1(rq, cpu);
-+ __vxm_rqinfo2(rq, cpu);
-+}
++/* time faking stuff */
+
-+#else /* CONFIG_VSERVER_MONITOR */
++#ifdef CONFIG_VSERVER_VTIME
+
-+#define vxm_activate_task(t,q) do { } while (0)
-+#define vxm_activate_idle(t,q) do { } while (0)
-+#define vxm_deactivate_task(t,q) do { } while (0)
-+#define vxm_hold_task(t,q) do { } while (0)
-+#define vxm_unhold_task(t,q) do { } while (0)
-+#define vxm_migrate_task(t,q,d) do { } while (0)
-+#define vxm_idle_skip(q,c) do { } while (0)
-+#define vxm_need_resched(t,s,c) do { } while (0)
-+#define vxm_sync(s,c) do { } while (0)
-+#define vxm_sched_info(s,v,c) do { } while (0)
-+#define vxm_tokens_recalc(s,q,v,c) do { } while (0)
-+#define vxm_update_sched(s,v,c) do { } while (0)
-+#define vxm_rq_max_min(q,c) do { } while (0)
++extern void vx_gettimeofday(struct timeval *tv);
++extern int vx_settimeofday(struct timespec *ts);
+
-+#endif /* CONFIG_VSERVER_MONITOR */
++#else
++#define vx_gettimeofday(t) do_gettimeofday(t)
++#define vx_settimeofday(t) do_settimeofday(t)
++#endif
+
---- linux-2.6.22/kernel/signal.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/signal.c 2007-07-07 03:52:54 +0200
-@@ -26,6 +26,8 @@
- #include <linux/freezer.h>
- #include <linux/pid_namespace.h>
- #include <linux/nsproxy.h>
-+#include <linux/vs_context.h>
-+#include <linux/vs_pid.h>
++#else
++#warning duplicate inclusion
++#endif
+--- linux-2.6.25.4/include/net/addrconf.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/net/addrconf.h 2008-04-19 15:14:52.000000000 -0400
+@@ -75,10 +75,12 @@ extern struct inet6_ifaddr *ipv6_ge
+
+ extern int ipv6_get_saddr(struct dst_entry *dst,
+ struct in6_addr *daddr,
+- struct in6_addr *saddr);
++ struct in6_addr *saddr,
++ struct nx_info *nxi);
+ extern int ipv6_dev_get_saddr(struct net_device *dev,
+ struct in6_addr *daddr,
+- struct in6_addr *saddr);
++ struct in6_addr *saddr,
++ struct nx_info *nxi);
+ extern int ipv6_get_lladdr(struct net_device *dev,
+ struct in6_addr *addr,
+ unsigned char banned_flags);
+--- linux-2.6.25.4/include/net/af_unix.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/net/af_unix.h 2008-04-19 15:14:52.000000000 -0400
+@@ -4,6 +4,7 @@
+ #include <linux/socket.h>
+ #include <linux/un.h>
+ #include <linux/mutex.h>
++#include <linux/vs_base.h>
+ #include <net/sock.h>
- #include <asm/param.h>
- #include <asm/uaccess.h>
-@@ -523,19 +525,34 @@ static int check_kill_permission(int sig
- if (!valid_signal(sig))
- return error;
+ extern void unix_inflight(struct file *fp);
+--- linux-2.6.25.4/include/net/inet_sock.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/net/inet_sock.h 2008-04-19 15:14:52.000000000 -0400
+@@ -24,7 +24,7 @@
+ #include <net/flow.h>
+ #include <net/sock.h>
+ #include <net/request_sock.h>
+-#include <net/route.h>
++// #include <net/route.h>
-+ if ((info != SEND_SIG_NOINFO) &&
-+ (is_si_special(info) || !SI_FROMUSER(info)))
-+ goto skip;
-+
-+ vxdprintk(VXD_CBIT(misc, 7),
-+ "check_kill_permission(%d,%p,%p[#%u,%u])",
-+ sig, info, t, vx_task_xid(t), t->pid);
-+
- error = audit_signal_info(sig, t); /* Let audit system see the signal */
- if (error)
- return error;
+ /** struct ip_options - IP Options
+ *
+@@ -193,9 +193,4 @@ static inline int inet_sk_ehashfn(const
+ }
- error = -EPERM;
-- if ((info == SEND_SIG_NOINFO || (!is_si_special(info) && SI_FROMUSER(info)))
-- && ((sig != SIGCONT) ||
-+ if (((sig != SIGCONT) ||
- (process_session(current) != process_session(t)))
- && (current->euid ^ t->suid) && (current->euid ^ t->uid)
- && (current->uid ^ t->suid) && (current->uid ^ t->uid)
- && !capable(CAP_KILL))
- return error;
-+ error = -ESRCH;
-+ if (!vx_check(vx_task_xid(t), VS_WATCH_P | VS_IDENT)) {
-+ vxdprintk(current->xid || VXD_CBIT(misc, 7),
-+ "signal %d[%p] xid mismatch %p[#%u,%u] xid=#%u",
-+ sig, info, t, vx_task_xid(t), t->pid, current->xid);
-+ return error;
-+ }
-+skip:
- return security_task_kill(t, info, sig, 0);
- }
+-static inline int inet_iif(const struct sk_buff *skb)
+-{
+- return ((struct rtable *)skb->dst)->rt_iif;
+-}
+-
+ #endif /* _INET_SOCK_H */
+--- linux-2.6.25.4/include/net/inet_timewait_sock.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/net/inet_timewait_sock.h 2008-04-19 15:25:34.000000000 -0400
+@@ -15,15 +15,14 @@
+ #ifndef _INET_TIMEWAIT_SOCK_
+ #define _INET_TIMEWAIT_SOCK_
-@@ -1043,7 +1060,7 @@ int kill_pid_info(int sig, struct siginf
++// #include <net/inet_sock.h>
++#include <net/sock.h>
- p = pid_task(pid, PIDTYPE_PID);
- error = -ESRCH;
-- if (p)
-+ if (p && vx_check(vx_task_xid(p), VS_IDENT))
- error = group_send_sig_info(sig, info, p);
+ #include <linux/list.h>
+ #include <linux/module.h>
+ #include <linux/timer.h>
+ #include <linux/types.h>
+ #include <linux/workqueue.h>
+-
+-#include <net/inet_sock.h>
+-#include <net/sock.h>
+ #include <net/tcp_states.h>
+ #include <net/timewait_sock.h>
- if (unlikely(sig_needs_tasklist(sig)))
-@@ -1057,7 +1074,7 @@ kill_proc_info(int sig, struct siginfo *
- {
- int error;
- rcu_read_lock();
-- error = kill_pid_info(sig, info, find_pid(pid));
-+ error = kill_pid_info(sig, info, find_pid(vx_rmap_pid(pid)));
- rcu_read_unlock();
- return error;
- }
-@@ -1118,7 +1135,8 @@ static int kill_something_info(int sig,
+@@ -116,6 +115,10 @@ struct inet_timewait_sock {
+ #define tw_hash __tw_common.skc_hash
+ #define tw_prot __tw_common.skc_prot
+ #define tw_net __tw_common.skc_net
++#define tw_xid __tw_common.skc_xid
++#define tw_vx_info __tw_common.skc_vx_info
++#define tw_nid __tw_common.skc_nid
++#define tw_nx_info __tw_common.skc_nx_info
+ int tw_timeout;
+ volatile unsigned char tw_substate;
+ /* 3 bits hole, try to pack */
+--- linux-2.6.25.4/include/net/route.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/net/route.h 2008-04-21 12:39:35.000000000 -0400
+@@ -34,7 +34,7 @@
+ #include <linux/ip.h>
+ #include <linux/cache.h>
+ #include <linux/security.h>
+-#include <net/sock.h>
++#include <linux/in.h>
- read_lock(&tasklist_lock);
- for_each_process(p) {
-- if (p->pid > 1 && p->tgid != current->tgid) {
-+ if (vx_check(vx_task_xid(p), VS_ADMIN_P|VS_IDENT) &&
-+ p->pid > 1 && p->tgid != current->tgid) {
- int err = group_send_sig_info(sig, info, p);
- ++count;
- if (err != -EPERM)
-@@ -1815,6 +1833,11 @@ relock:
- if (current == child_reaper(current))
- continue;
+ #ifndef __KERNEL__
+ #warning This file is not supposed to be used outside of kernel.
+@@ -86,6 +86,11 @@ struct ip_rt_acct
+ __u32 i_packets;
+ };
-+ /* virtual init is protected against user signals */
-+ if ((info->si_code == SI_USER) &&
-+ vx_current_initpid(current->pid))
-+ continue;
++static inline int inet_iif(const struct sk_buff *skb)
++{
++ return ((struct rtable *)skb->dst)->rt_iif;
++}
+
- if (sig_kernel_stop(signr)) {
- /*
- * The default action is to stop all threads in
---- linux-2.6.22/kernel/softirq.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/softirq.c 2007-06-15 02:37:04 +0200
-@@ -18,6 +18,7 @@
- #include <linux/rcupdate.h>
- #include <linux/smp.h>
- #include <linux/tick.h>
-+#include <linux/vs_context.h>
+ struct rt_cache_stat
+ {
+ unsigned int in_hit;
+@@ -136,6 +141,9 @@ static inline void ip_rt_put(struct rtab
+ dst_release(&rt->u.dst);
+ }
- #include <asm/irq.h>
- /*
---- linux-2.6.22/kernel/sys.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/sys.c 2007-06-15 05:28:51 +0200
-@@ -35,6 +35,7 @@
- #include <linux/compat.h>
- #include <linux/syscalls.h>
- #include <linux/kprobes.h>
-+#include <linux/vs_pid.h>
++#include <linux/vs_base.h>
++#include <linux/vs_inet.h>
++
+ #define IPTOS_RT_MASK (IPTOS_TOS_MASK & ~3)
- #include <asm/uaccess.h>
- #include <asm/io.h>
-@@ -638,7 +639,10 @@ static int set_one_prio(struct task_stru
- goto out;
- }
- if (niceval < task_nice(p) && !can_nice(p, niceval)) {
-- error = -EACCES;
-+ if (vx_flags(VXF_IGNEG_NICE, 0))
-+ error = 0;
-+ else
-+ error = -EACCES;
- goto out;
- }
- no_nice = security_task_setnice(p, niceval);
-@@ -686,6 +690,8 @@ asmlinkage long sys_setpriority(int whic
- else
- pgrp = task_pgrp(current);
- do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
-+ if (!vx_check(p->xid, VS_ADMIN_P | VS_IDENT))
-+ continue;
- error = set_one_prio(p, niceval, error);
- } while_each_pid_task(pgrp, PIDTYPE_PGID, p);
- break;
-@@ -694,7 +700,8 @@ asmlinkage long sys_setpriority(int whic
- if (!who)
- who = current->uid;
- else
-- if ((who != current->uid) && !(user = find_user(who)))
-+ if ((who != current->uid) &&
-+ !(user = find_user(vx_current_xid(), who)))
- goto out_unlock; /* No processes for this user */
+ extern const __u8 ip_tos2prio[16];
+@@ -145,6 +153,9 @@ static inline char rt_tos2priority(u8 to
+ return ip_tos2prio[IPTOS_TOS(tos)>>1];
+ }
- do_each_thread(g, p)
-@@ -746,6 +753,8 @@ asmlinkage long sys_getpriority(int whic
- else
- pgrp = task_pgrp(current);
- do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
-+ if (!vx_check(p->xid, VS_ADMIN_P | VS_IDENT))
-+ continue;
- niceval = 20 - task_nice(p);
- if (niceval > retval)
- retval = niceval;
-@@ -756,7 +765,8 @@ asmlinkage long sys_getpriority(int whic
- if (!who)
- who = current->uid;
- else
-- if ((who != current->uid) && !(user = find_user(who)))
-+ if ((who != current->uid) &&
-+ !(user = find_user(vx_current_xid(), who)))
- goto out_unlock; /* No processes for this user */
++extern int ip_v4_find_src(struct net *net, struct nx_info *,
++ struct rtable **, struct flowi *);
++
+ static inline int ip_route_connect(struct rtable **rp, __be32 dst,
+ __be32 src, u32 tos, int oif, u8 protocol,
+ __be16 sport, __be16 dport, struct sock *sk,
+@@ -162,7 +173,21 @@ static inline int ip_route_connect(struc
- do_each_thread(g, p)
-@@ -869,6 +879,9 @@ void kernel_power_off(void)
- machine_power_off();
- }
- EXPORT_SYMBOL_GPL(kernel_power_off);
+ int err;
+ struct net *net = sk->sk_net;
+- if (!dst || !src) {
++ struct nx_info *nx_info = current->nx_info;
+
-+long vs_reboot(unsigned int, void __user *);
++ if (sk)
++ nx_info = sk->sk_nx_info;
+
- /*
- * Reboot system call: for obvious reasons only root may call it,
- * and even root needs to set up some magic numbers in the registers
-@@ -899,6 +912,9 @@ asmlinkage long sys_reboot(int magic1, i
- if ((cmd == LINUX_REBOOT_CMD_POWER_OFF) && !pm_power_off)
- cmd = LINUX_REBOOT_CMD_HALT;
-
-+ if (!vx_check(0, VS_ADMIN|VS_WATCH))
-+ return vs_reboot(cmd, arg);
++ vxdprintk(VXD_CBIT(net, 4),
++ "ip_route_connect(%p) %p,%p;%lx",
++ sk, nx_info, sk->sk_socket,
++ (sk->sk_socket?sk->sk_socket->flags:0));
+
- lock_kernel();
- switch (cmd) {
- case LINUX_REBOOT_CMD_RESTART:
-@@ -1078,7 +1094,7 @@ static int set_user(uid_t new_ruid, int
- {
- struct user_struct *new_user;
++ err = ip_v4_find_src(net, nx_info, rp, &fl);
++ if (err)
++ return err;
++
++ if (!fl.fl4_dst || !fl.fl4_src) {
+ err = __ip_route_output_key(net, rp, &fl);
+ if (err)
+ return err;
+--- linux-2.6.25.4/include/net/sock.h 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/include/net/sock.h 2008-04-19 15:14:52.000000000 -0400
+@@ -123,6 +123,10 @@ struct sock_common {
+ unsigned int skc_hash;
+ struct proto *skc_prot;
+ struct net *skc_net;
++ xid_t skc_xid;
++ struct vx_info *skc_vx_info;
++ nid_t skc_nid;
++ struct nx_info *skc_nx_info;
+ };
-- new_user = alloc_uid(new_ruid);
-+ new_user = alloc_uid(vx_current_xid(), new_ruid);
- if (!new_user)
- return -EAGAIN;
+ /**
+@@ -205,6 +209,10 @@ struct sock {
+ #define sk_hash __sk_common.skc_hash
+ #define sk_prot __sk_common.skc_prot
+ #define sk_net __sk_common.skc_net
++#define sk_xid __sk_common.skc_xid
++#define sk_vx_info __sk_common.skc_vx_info
++#define sk_nid __sk_common.skc_nid
++#define sk_nx_info __sk_common.skc_nx_info
+ unsigned char sk_shutdown : 2,
+ sk_no_check : 2,
+ sk_userlocks : 4;
+--- linux-2.6.25.4/init/main.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/init/main.c 2008-04-21 10:46:10.000000000 -0400
+@@ -58,6 +58,7 @@
+ #include <linux/kthread.h>
+ #include <linux/sched.h>
+ #include <linux/signal.h>
++#include <linux/vserver/percpu.h>
+
+ #include <asm/io.h>
+ #include <asm/bugs.h>
+@@ -370,12 +371,14 @@ EXPORT_SYMBOL(__per_cpu_offset);
-@@ -1433,15 +1449,18 @@ asmlinkage long sys_setpgid(pid_t pid, p
+ static void __init setup_per_cpu_areas(void)
{
- struct task_struct *p;
- struct task_struct *group_leader = current->group_leader;
-+ pid_t rpgid;
- int err = -EINVAL;
-
- if (!pid)
-- pid = group_leader->pid;
-+ pid = vx_map_pid(group_leader->pid);
- if (!pgid)
- pgid = pid;
- if (pgid < 0)
- return -EINVAL;
+- unsigned long size, i;
++ unsigned long size, vspc, i;
+ char *ptr;
+ unsigned long nr_possible_cpus = num_possible_cpus();
-+ rpgid = vx_rmap_pid(pgid);
++ vspc = PERCPU_PERCTX * CONFIG_VSERVER_CONTEXTS;
+
- /* From this point forward we keep holding onto the tasklist lock
- * so that our parent does not change from under us. -DaveM
- */
-@@ -1475,20 +1494,20 @@ asmlinkage long sys_setpgid(pid_t pid, p
+ /* Copy section for each CPU (we discard the original) */
+- size = ALIGN(PERCPU_ENOUGH_ROOM, PAGE_SIZE);
++ size = ALIGN(PERCPU_ENOUGH_ROOM + vspc, PAGE_SIZE);
+ ptr = alloc_bootmem_pages(size * nr_possible_cpus);
+
+ for_each_possible_cpu(i) {
+--- linux-2.6.25.4/ipc/mqueue.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/ipc/mqueue.c 2008-04-19 15:14:52.000000000 -0400
+@@ -31,6 +31,8 @@
+ #include <linux/mutex.h>
+ #include <linux/nsproxy.h>
+ #include <linux/pid.h>
++#include <linux/vs_context.h>
++#include <linux/vs_limit.h>
- if (pgid != pid) {
- struct task_struct *g =
-- find_task_by_pid_type(PIDTYPE_PGID, pgid);
-+ find_task_by_pid_type(PIDTYPE_PGID, rpgid);
+ #include <net/sock.h>
+ #include "util.h"
+@@ -71,6 +73,7 @@ struct mqueue_inode_info {
+ struct sigevent notify;
+ struct pid* notify_owner;
+ struct user_struct *user; /* user who created, for accounting */
++ struct vx_info *vxi;
+ struct sock *notify_sock;
+ struct sk_buff *notify_cookie;
- if (!g || task_session(g) != task_session(group_leader))
- goto out;
+@@ -119,6 +122,7 @@ static struct inode *mqueue_get_inode(st
+ struct mqueue_inode_info *info;
+ struct task_struct *p = current;
+ struct user_struct *u = p->user;
++ struct vx_info *vxi = p->vx_info;
+ unsigned long mq_bytes, mq_msg_tblsz;
+
+ inode->i_fop = &mqueue_file_operations;
+@@ -133,6 +137,7 @@ static struct inode *mqueue_get_inode(st
+ info->notify_owner = NULL;
+ info->qsize = 0;
+ info->user = NULL; /* set when all is ok */
++ info->vxi = NULL;
+ memset(&info->attr, 0, sizeof(info->attr));
+ info->attr.mq_maxmsg = DFLT_MSGMAX;
+ info->attr.mq_msgsize = DFLT_MSGSIZEMAX;
+@@ -147,22 +152,26 @@ static struct inode *mqueue_get_inode(st
+ spin_lock(&mq_lock);
+ if (u->mq_bytes + mq_bytes < u->mq_bytes ||
+ u->mq_bytes + mq_bytes >
+- p->signal->rlim[RLIMIT_MSGQUEUE].rlim_cur) {
++ p->signal->rlim[RLIMIT_MSGQUEUE].rlim_cur ||
++ !vx_ipcmsg_avail(vxi, mq_bytes)) {
+ spin_unlock(&mq_lock);
+ goto out_inode;
+ }
+ u->mq_bytes += mq_bytes;
++ vx_ipcmsg_add(vxi, u, mq_bytes);
+ spin_unlock(&mq_lock);
+
+ info->messages = kmalloc(mq_msg_tblsz, GFP_KERNEL);
+ if (!info->messages) {
+ spin_lock(&mq_lock);
+ u->mq_bytes -= mq_bytes;
++ vx_ipcmsg_sub(vxi, u, mq_bytes);
+ spin_unlock(&mq_lock);
+ goto out_inode;
+ }
+ /* all is ok */
+ info->user = get_uid(u);
++ info->vxi = get_vx_info(vxi);
+ } else if (S_ISDIR(mode)) {
+ inc_nlink(inode);
+ /* Some things misbehave if size == 0 on a directory */
+@@ -253,10 +262,14 @@ static void mqueue_delete_inode(struct i
+ (info->attr.mq_maxmsg * info->attr.mq_msgsize));
+ user = info->user;
+ if (user) {
++ struct vx_info *vxi = info->vxi;
++
+ spin_lock(&mq_lock);
+ user->mq_bytes -= mq_bytes;
++ vx_ipcmsg_sub(vxi, user, mq_bytes);
+ queues_count--;
+ spin_unlock(&mq_lock);
++ put_vx_info(vxi);
+ free_uid(user);
}
+ }
+@@ -743,7 +756,7 @@ asmlinkage long sys_mq_unlink(const char
+ if (inode)
+ atomic_inc(&inode->i_count);
-- err = security_task_setpgid(p, pgid);
-+ err = security_task_setpgid(p, rpgid);
- if (err)
- goto out;
+- err = vfs_unlink(dentry->d_parent->d_inode, dentry);
++ err = vfs_unlink(dentry->d_parent->d_inode, dentry, NULL);
+ out_err:
+ dput(dentry);
-- if (process_group(p) != pgid) {
-+ if (process_group(p) != rpgid) {
- detach_pid(p, PIDTYPE_PGID);
-- p->signal->pgrp = pgid;
-- attach_pid(p, PIDTYPE_PGID, find_pid(pgid));
-+ p->signal->pgrp = rpgid;
-+ attach_pid(p, PIDTYPE_PGID, find_pid(rpgid));
- }
-
- err = 0;
-@@ -1501,7 +1520,7 @@ out:
- asmlinkage long sys_getpgid(pid_t pid)
- {
- if (!pid)
-- return process_group(current);
-+ return vx_rmap_pid(process_group(current));
- else {
- int retval;
- struct task_struct *p;
-@@ -1513,7 +1532,7 @@ asmlinkage long sys_getpgid(pid_t pid)
- if (p) {
- retval = security_task_getpgid(p);
- if (!retval)
-- retval = process_group(p);
-+ retval = vx_rmap_pid(process_group(p));
- }
- read_unlock(&tasklist_lock);
- return retval;
-@@ -1864,7 +1883,7 @@ asmlinkage long sys_sethostname(char __u
- int errno;
- char tmp[__NEW_UTS_LEN];
+--- linux-2.6.25.4/ipc/msg.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/ipc/msg.c 2008-04-21 10:41:47.000000000 -0400
+@@ -37,6 +37,7 @@
+ #include <linux/rwsem.h>
+ #include <linux/nsproxy.h>
+ #include <linux/ipc_namespace.h>
++#include <linux/vs_base.h>
-- if (!capable(CAP_SYS_ADMIN))
-+ if (!vx_capable(CAP_SYS_ADMIN, VXC_SET_UTSNAME))
- return -EPERM;
- if (len < 0 || len > __NEW_UTS_LEN)
- return -EINVAL;
-@@ -1909,7 +1928,7 @@ asmlinkage long sys_setdomainname(char _
- int errno;
- char tmp[__NEW_UTS_LEN];
+ #include <asm/current.h>
+ #include <asm/uaccess.h>
+@@ -168,6 +169,7 @@ static int newque(struct ipc_namespace *
-- if (!capable(CAP_SYS_ADMIN))
-+ if (!vx_capable(CAP_SYS_ADMIN, VXC_SET_UTSNAME))
- return -EPERM;
- if (len < 0 || len > __NEW_UTS_LEN)
- return -EINVAL;
-@@ -1976,7 +1995,7 @@ asmlinkage long sys_setrlimit(unsigned i
- return -EINVAL;
- old_rlim = current->signal->rlim + resource;
- if ((new_rlim.rlim_max > old_rlim->rlim_max) &&
-- !capable(CAP_SYS_RESOURCE))
-+ !vx_capable(CAP_SYS_RESOURCE, VXC_SET_RLIMIT))
- return -EPERM;
- if (resource == RLIMIT_NOFILE && new_rlim.rlim_max > NR_OPEN)
- return -EPERM;
---- linux-2.6.22/kernel/sysctl.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/sysctl.c 2007-06-15 02:37:04 +0200
-@@ -89,6 +89,7 @@ static int ngroups_max = NGROUPS_MAX;
- #ifdef CONFIG_KMOD
- extern char modprobe_path[];
- #endif
-+extern char vshelper_path[];
- #ifdef CONFIG_CHR_DEV_SG
- extern int sg_big_buff;
- #endif
-@@ -343,6 +344,15 @@ static ctl_table kern_table[] = {
- .strategy = &sysctl_string,
- },
- #endif
-+ {
-+ .ctl_name = KERN_VSHELPER,
-+ .procname = "vshelper",
-+ .data = &vshelper_path,
-+ .maxlen = 256,
-+ .mode = 0644,
-+ .proc_handler = &proc_dostring,
-+ .strategy = &sysctl_string,
-+ },
- #ifdef CONFIG_CHR_DEV_SG
- {
- .ctl_name = KERN_SG_BIG_BUFF,
---- linux-2.6.22/kernel/time.c 2007-07-09 13:20:03 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/time.c 2007-06-15 02:37:04 +0200
-@@ -60,7 +60,7 @@ asmlinkage long sys_time(time_t __user *
- time_t i;
+ msq->q_perm.mode = msgflg & S_IRWXUGO;
+ msq->q_perm.key = key;
++ msq->q_perm.xid = vx_current_xid();
+
+ msq->q_perm.security = NULL;
+ retval = security_msg_queue_alloc(msq);
+--- linux-2.6.25.4/ipc/namespace.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/ipc/namespace.c 2008-04-21 10:44:58.000000000 -0400
+@@ -9,6 +9,8 @@
+ #include <linux/rcupdate.h>
+ #include <linux/nsproxy.h>
+ #include <linux/slab.h>
++#include <linux/vs_base.h>
++#include <linux/vserver/global.h>
+
+ #include "util.h"
+
+@@ -25,6 +27,7 @@ static struct ipc_namespace *clone_ipc_n
+ shm_init_ns(ns);
+
+ kref_init(&ns->kref);
++ atomic_inc(&vs_global_ipc_ns);
+ return ns;
+ }
+
+@@ -82,5 +85,6 @@ void free_ipc_ns(struct kref *kref)
+ sem_exit_ns(ns);
+ msg_exit_ns(ns);
+ shm_exit_ns(ns);
++ atomic_dec(&vs_global_ipc_ns);
+ kfree(ns);
+ }
+--- linux-2.6.25.4/ipc/sem.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/ipc/sem.c 2008-04-21 10:45:22.000000000 -0400
+@@ -83,6 +83,8 @@
+ #include <linux/rwsem.h>
+ #include <linux/nsproxy.h>
+ #include <linux/ipc_namespace.h>
++#include <linux/vs_base.h>
++#include <linux/vs_limit.h>
+
+ #include <asm/uaccess.h>
+ #include "util.h"
+@@ -252,6 +254,7 @@ static int newary(struct ipc_namespace *
+
+ sma->sem_perm.mode = (semflg & S_IRWXUGO);
+ sma->sem_perm.key = key;
++ sma->sem_perm.xid = vx_current_xid();
+
+ sma->sem_perm.security = NULL;
+ retval = security_sem_alloc(sma);
+@@ -267,6 +270,9 @@ static int newary(struct ipc_namespace *
+ return id;
+ }
+ ns->used_sems += nsems;
++ /* FIXME: obsoleted? */
++ vx_semary_inc(sma);
++ vx_nsems_add(sma, nsems);
+
+ sma->sem_perm.id = sem_buildid(id, sma->sem_perm.seq);
+ sma->sem_base = (struct sem *) &sma[1];
+--- linux-2.6.25.4/ipc/shm.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/ipc/shm.c 2008-04-21 10:45:38.000000000 -0400
+@@ -39,6 +39,8 @@
+ #include <linux/nsproxy.h>
+ #include <linux/mount.h>
+ #include <linux/ipc_namespace.h>
++#include <linux/vs_context.h>
++#include <linux/vs_limit.h>
+
+ #include <asm/uaccess.h>
+
+@@ -202,7 +204,12 @@ static void shm_open(struct vm_area_stru
+ */
+ static void shm_destroy(struct ipc_namespace *ns, struct shmid_kernel *shp)
+ {
+- ns->shm_tot -= (shp->shm_segsz + PAGE_SIZE - 1) >> PAGE_SHIFT;
++ struct vx_info *vxi = lookup_vx_info(shp->shm_perm.xid);
++ int numpages = (shp->shm_segsz + PAGE_SIZE - 1) >> PAGE_SHIFT;
++
++ vx_ipcshm_sub(vxi, shp, numpages);
++ ns->shm_tot -= numpages;
++
+ shm_rmid(ns, shp);
+ shm_unlock(shp);
+ if (!is_file_hugepages(shp->shm_file))
+@@ -212,6 +219,7 @@ static void shm_destroy(struct ipc_names
+ shp->mlock_user);
+ fput (shp->shm_file);
+ security_shm_free(shp);
++ put_vx_info(vxi);
+ ipc_rcu_putref(shp);
+ }
+
+@@ -383,11 +391,15 @@ static int newseg(struct ipc_namespace *
+ if (ns->shm_tot + numpages > ns->shm_ctlall)
+ return -ENOSPC;
+
++ if (!vx_ipcshm_avail(current->vx_info, numpages))
++ return -ENOSPC;
++
+ shp = ipc_rcu_alloc(sizeof(*shp));
+ if (!shp)
+ return -ENOMEM;
+
+ shp->shm_perm.key = key;
++ shp->shm_perm.xid = vx_current_xid();
+ shp->shm_perm.mode = (shmflg & S_IRWXUGO);
+ shp->mlock_user = NULL;
+
+@@ -441,6 +453,7 @@ static int newseg(struct ipc_namespace *
+ ns->shm_tot += numpages;
+ error = shp->shm_perm.id;
+ shm_unlock(shp);
++ vx_ipcshm_add(current->vx_info, key, numpages);
+ return error;
+
+ no_id:
+--- linux-2.6.25.4/kernel/capability.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/capability.c 2008-04-19 15:14:52.000000000 -0400
+@@ -13,6 +13,7 @@
+ #include <linux/security.h>
+ #include <linux/syscalls.h>
+ #include <linux/pid_namespace.h>
++#include <linux/vs_context.h>
+ #include <asm/uaccess.h>
+
+ /*
+@@ -171,6 +172,8 @@ static inline int cap_set_pg(int pgrp_nr
+
+ pgrp = find_vpid(pgrp_nr);
+ do_each_pid_task(pgrp, PIDTYPE_PGID, g) {
++ if (!vx_check(g->xid, VS_ADMIN_P | VS_IDENT))
++ continue;
+ target = g;
+ while_each_thread(g, target) {
+ if (!security_capset_check(target, effective,
+@@ -335,8 +338,12 @@ int __capable(struct task_struct *t, int
+ return 0;
+ }
+
++#include <linux/vserver/base.h>
+ int capable(int cap)
+ {
++ /* here for now so we don't require task locking */
++ if (vs_check_bit(VXC_CAP_MASK, cap) && !vx_mcaps(1L << cap))
++ return 0;
+ return __capable(current, cap);
+ }
+ EXPORT_SYMBOL(capable);
+--- linux-2.6.25.4/kernel/cgroup.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/cgroup.c 2008-05-21 14:30:41.000000000 -0400
+@@ -2833,7 +2833,7 @@ int cgroup_clone(struct task_struct *tsk
+ }
+
+ /* Create the cgroup directory, which also creates the cgroup */
+- ret = vfs_mkdir(inode, dentry, S_IFDIR | 0755);
++ ret = vfs_mkdir(inode, dentry, S_IFDIR | 0755, NULL);
+ child = __d_cgrp(dentry);
+ dput(dentry);
+ if (ret) {
+--- linux-2.6.25.4/kernel/compat.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/compat.c 2008-04-19 15:14:52.000000000 -0400
+@@ -846,7 +846,7 @@ asmlinkage long compat_sys_time(compat_t
+ compat_time_t i;
struct timeval tv;
- do_gettimeofday(&tv);
@@ -17542,7 +15754,7 @@
i = tv.tv_sec;
if (tloc) {
-@@ -91,7 +91,7 @@ asmlinkage long sys_stime(time_t __user
+@@ -870,7 +870,7 @@ asmlinkage long compat_sys_stime(compat_
if (err)
return err;
@@ -17551,527 +15763,1757 @@
return 0;
}
-@@ -101,7 +101,7 @@ asmlinkage long sys_gettimeofday(struct
- {
- if (likely(tv != NULL)) {
- struct timeval ktv;
-- do_gettimeofday(&ktv);
-+ vx_gettimeofday(&ktv);
- if (copy_to_user(tv, &ktv, sizeof(ktv)))
- return -EFAULT;
- }
-@@ -175,7 +175,7 @@ int do_sys_settimeofday(struct timespec
- /* SMP safe, again the code in arch/foo/time.c should
- * globally block out interrupts when it runs.
- */
-- return do_settimeofday(tv);
-+ return vx_settimeofday(tv);
- }
- return 0;
- }
-@@ -388,7 +388,7 @@ void getnstimeofday(struct timespec *tv)
- {
- struct timeval x;
-
-- do_gettimeofday(&x);
-+ vx_gettimeofday(&x);
- tv->tv_sec = x.tv_sec;
- tv->tv_nsec = x.tv_usec * NSEC_PER_USEC;
- }
---- linux-2.6.22/kernel/timer.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/timer.c 2007-06-15 02:37:04 +0200
-@@ -36,6 +36,10 @@
- #include <linux/delay.h>
- #include <linux/tick.h>
- #include <linux/kallsyms.h>
-+#include <linux/vs_base.h>
-+#include <linux/vs_cvirt.h>
+--- linux-2.6.25.4/kernel/exit.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/exit.c 2008-04-19 15:14:52.000000000 -0400
+@@ -44,6 +44,11 @@
+ #include <linux/resource.h>
+ #include <linux/blkdev.h>
+ #include <linux/task_io_accounting_ops.h>
++#include <linux/vs_limit.h>
++#include <linux/vs_context.h>
++#include <linux/vs_network.h>
+#include <linux/vs_pid.h>
-+#include <linux/vserver/sched.h>
++#include <linux/vserver/global.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
-@@ -921,12 +925,6 @@ asmlinkage unsigned long sys_alarm(unsig
-
- #endif
+@@ -468,9 +473,11 @@ static void close_files(struct files_str
+ filp_close(file, files);
+ cond_resched();
+ }
++ vx_openfd_dec(i);
+ }
+ i++;
+ set >>= 1;
++ cond_resched();
+ }
+ }
+ }
+@@ -1014,6 +1021,10 @@ NORET_TYPE void do_exit(long code)
+ if (tsk->splice_pipe)
+ __free_pipe_info(tsk->splice_pipe);
--#ifndef __alpha__
--
--/*
-- * The Alpha uses getxpid, getxuid, and getxgid instead. Maybe this
-- * should be moved into arch/i386 instead?
-- */
++ /* needs to stay after exit_notify() */
++ exit_vx_info(tsk, code);
++ exit_nx_info(tsk);
++
+ preempt_disable();
+ /* causes final put_task_struct in finish_task_switch(). */
+ tsk->state = TASK_DEAD;
+--- linux-2.6.25.4/kernel/fork.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/fork.c 2008-04-21 11:09:01.000000000 -0400
+@@ -53,6 +53,11 @@
+ #include <linux/tty.h>
+ #include <linux/proc_fs.h>
+ #include <linux/blkdev.h>
++#include <linux/vs_context.h>
++#include <linux/vs_network.h>
++#include <linux/vs_limit.h>
++#include <linux/vs_memory.h>
++#include <linux/vserver/global.h>
- /**
- * sys_getpid - return the thread group id of the current process
-@@ -939,7 +937,7 @@ asmlinkage unsigned long sys_alarm(unsig
- */
- asmlinkage long sys_getpid(void)
- {
-- return current->tgid;
-+ return vx_map_tgid(current->tgid);
+ #include <asm/pgtable.h>
+ #include <asm/pgalloc.h>
+@@ -113,6 +118,8 @@ void free_task(struct task_struct *tsk)
+ prop_local_destroy_single(&tsk->dirties);
+ free_thread_info(tsk->stack);
+ rt_mutex_debug_task_free(tsk);
++ clr_vx_info(&tsk->vx_info);
++ clr_nx_info(&tsk->nx_info);
+ free_task_struct(tsk);
}
+ EXPORT_SYMBOL(free_task);
+@@ -229,6 +236,8 @@ static int dup_mmap(struct mm_struct *mm
+ mm->free_area_cache = oldmm->mmap_base;
+ mm->cached_hole_size = ~0UL;
+ mm->map_count = 0;
++ __set_mm_counter(mm, file_rss, 0);
++ __set_mm_counter(mm, anon_rss, 0);
+ cpus_clear(mm->cpu_vm_mask);
+ mm->mm_rb = RB_ROOT;
+ rb_link = &mm->mm_rb.rb_node;
+@@ -240,7 +249,7 @@ static int dup_mmap(struct mm_struct *mm
- /*
-@@ -955,10 +953,23 @@ asmlinkage long sys_getppid(void)
- rcu_read_lock();
- pid = rcu_dereference(current->real_parent)->tgid;
- rcu_read_unlock();
-+ return vx_map_pid(pid);
+ if (mpnt->vm_flags & VM_DONTCOPY) {
+ long pages = vma_pages(mpnt);
+- mm->total_vm -= pages;
++ vx_vmpages_sub(mm, pages);
+ vm_stat_account(mm, mpnt->vm_flags, mpnt->vm_file,
+ -pages);
+ continue;
+@@ -351,8 +360,8 @@ static struct mm_struct * mm_init(struct
+ : MMF_DUMP_FILTER_DEFAULT;
+ mm->core_waiters = 0;
+ mm->nr_ptes = 0;
+- set_mm_counter(mm, file_rss, 0);
+- set_mm_counter(mm, anon_rss, 0);
++ __set_mm_counter(mm, file_rss, 0);
++ __set_mm_counter(mm, anon_rss, 0);
+ spin_lock_init(&mm->page_table_lock);
+ rwlock_init(&mm->ioctx_list_lock);
+ mm->ioctx_list = NULL;
+@@ -362,6 +371,7 @@ static struct mm_struct * mm_init(struct
+
+ if (likely(!mm_alloc_pgd(mm))) {
+ mm->def_flags = 0;
++ set_vx_info(&mm->mm_vx_info, p->vx_info);
+ return mm;
+ }
+
+@@ -395,6 +405,7 @@ void __mmdrop(struct mm_struct *mm)
+ BUG_ON(mm == &init_mm);
+ mm_free_pgd(mm);
+ destroy_context(mm);
++ clr_vx_info(&mm->mm_vx_info);
+ free_mm(mm);
+ }
+ EXPORT_SYMBOL_GPL(__mmdrop);
+@@ -511,6 +522,7 @@ static struct mm_struct *dup_mm(struct t
+ goto fail_nomem;
+
+ memcpy(mm, oldmm, sizeof(*mm));
++ mm->mm_vx_info = NULL;
+
+ /* Initializing for Swap token stuff */
+ mm->token_priority = 0;
+@@ -542,6 +554,7 @@ fail_nocontext:
+ * If init_new_context() failed, we cannot use mmput() to free the mm
+ * because it calls destroy_context()
+ */
++ clr_vx_info(&mm->mm_vx_info);
+ mm_free_pgd(mm);
+ free_mm(mm);
+ return NULL;
+@@ -612,6 +625,7 @@ static struct fs_struct *__copy_fs_struc
+ fs->altroot.dentry = NULL;
+ }
+ read_unlock(&old->lock);
++ atomic_inc(&vs_global_fs);
+ }
+ return fs;
+ }
+@@ -730,6 +744,8 @@ static struct files_struct *dup_fd(struc
+ struct file *f = *old_fds++;
+ if (f) {
+ get_file(f);
++ /* TODO: sum it first for check and performance */
++ vx_openfd_inc(open_files - i);
+ } else {
+ /*
+ * The fd may be claimed in the fd bitmap but not yet
+@@ -1011,6 +1027,8 @@ static struct task_struct *copy_process(
+ int retval;
+ struct task_struct *p;
+ int cgroup_callbacks_done = 0;
++ struct vx_info *vxi;
++ struct nx_info *nxi;
+
+ if ((clone_flags & (CLONE_NEWNS|CLONE_FS)) == (CLONE_NEWNS|CLONE_FS))
+ return ERR_PTR(-EINVAL);
+@@ -1045,12 +1063,28 @@ static struct task_struct *copy_process(
+ DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled);
+ DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled);
+ #endif
++ init_vx_info(&p->vx_info, current->vx_info);
++ init_nx_info(&p->nx_info, current->nx_info);
++
++ /* check vserver memory */
++ if (p->mm && !(clone_flags & CLONE_VM)) {
++ if (vx_vmpages_avail(p->mm, p->mm->total_vm))
++ vx_pages_add(p->vx_info, RLIMIT_AS, p->mm->total_vm);
++ else
++ goto bad_fork_free;
++ }
++ if (p->mm && vx_flags(VXF_FORK_RSS, 0)) {
++ if (!vx_rss_avail(p->mm, get_mm_counter(p->mm, file_rss)))
++ goto bad_fork_cleanup_vm;
++ }
+ retval = -EAGAIN;
++ if (!vx_nproc_avail(1))
++ goto bad_fork_cleanup_vm;
+ if (atomic_read(&p->user->processes) >=
+ p->signal->rlim[RLIMIT_NPROC].rlim_cur) {
+ if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_RESOURCE) &&
+ p->user != current->nsproxy->user_ns->root_user)
+- goto bad_fork_free;
++ goto bad_fork_cleanup_vm;
+ }
+
+ atomic_inc(&p->user->__count);
+@@ -1352,6 +1386,18 @@ static struct task_struct *copy_process(
+
+ total_forks++;
+ spin_unlock(¤t->sighand->siglock);
++
++ /* p is copy of current */
++ vxi = p->vx_info;
++ if (vxi) {
++ claim_vx_info(vxi, p);
++ atomic_inc(&vxi->cvirt.nr_threads);
++ atomic_inc(&vxi->cvirt.total_forks);
++ vx_nproc_inc(p);
++ }
++ nxi = p->nx_info;
++ if (nxi)
++ claim_nx_info(nxi, p);
+ write_unlock_irq(&tasklist_lock);
+ proc_fork_connector(p);
+ cgroup_post_fork(p);
+@@ -1398,6 +1444,9 @@ bad_fork_cleanup_count:
+ put_group_info(p->group_info);
+ atomic_dec(&p->user->processes);
+ free_uid(p->user);
++bad_fork_cleanup_vm:
++ if (p->mm && !(clone_flags & CLONE_VM))
++ vx_pages_sub(p->vx_info, RLIMIT_AS, p->mm->total_vm);
+ bad_fork_free:
+ free_task(p);
+ fork_out:
+--- linux-2.6.25.4/kernel/kthread.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/kthread.c 2008-04-19 15:14:52.000000000 -0400
+@@ -13,6 +13,7 @@
+ #include <linux/file.h>
+ #include <linux/module.h>
+ #include <linux/mutex.h>
++#include <linux/vs_pid.h>
+ #include <asm/semaphore.h>
+
+ #define KTHREAD_NICE_LEVEL (-5)
+@@ -99,7 +100,7 @@ static void create_kthread(struct kthrea
+ struct sched_param param = { .sched_priority = 0 };
+ wait_for_completion(&create->started);
+ read_lock(&tasklist_lock);
+- create->result = find_task_by_pid(pid);
++ create->result = find_task_by_real_pid(pid);
+ read_unlock(&tasklist_lock);
+ /*
+ * root may have changed our (kthreadd's) priority or CPU mask.
+--- linux-2.6.25.4/kernel/Makefile 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/Makefile 2008-04-19 15:14:52.000000000 -0400
+@@ -11,6 +11,8 @@ obj-y = sched.o fork.o exec_domain.o
+ hrtimer.o rwsem.o nsproxy.o srcu.o \
+ notifier.o ksysfs.o pm_qos_params.o
+
++obj-y += vserver/
++
+ obj-$(CONFIG_SYSCTL) += sysctl_check.o
+ obj-$(CONFIG_STACKTRACE) += stacktrace.o
+ obj-y += time/
+--- linux-2.6.25.4/kernel/nsproxy.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/nsproxy.c 2008-04-19 15:14:52.000000000 -0400
+@@ -20,6 +20,8 @@
+ #include <linux/mnt_namespace.h>
+ #include <linux/utsname.h>
+ #include <linux/pid_namespace.h>
++#include <linux/vserver/global.h>
++#include <linux/vserver/debug.h>
+ #include <net/net_namespace.h>
+ #include <linux/ipc_namespace.h>
+
+@@ -38,6 +40,9 @@ static inline struct nsproxy *clone_nspr
+ if (ns) {
+ memcpy(ns, orig, sizeof(struct nsproxy));
+ atomic_set(&ns->count, 1);
++ vxdprintk(VXD_CBIT(space, 2), "clone_nsproxy(%p[%u] = %p[1]",
++ orig, atomic_read(&orig->count), ns);
++ atomic_inc(&vs_global_nsproxy);
+ }
+ return ns;
+ }
+@@ -47,47 +52,51 @@ static inline struct nsproxy *clone_nspr
+ * Return the newly created nsproxy. Do not attach this to the task,
+ * leave it to the caller to do proper locking and attach it to task.
+ */
+-static struct nsproxy *create_new_namespaces(unsigned long flags,
+- struct task_struct *tsk, struct fs_struct *new_fs)
++static struct nsproxy *unshare_namespaces(unsigned long flags,
++ struct nsproxy *orig, struct fs_struct *new_fs)
+ {
+ struct nsproxy *new_nsp;
+ int err;
+
+- new_nsp = clone_nsproxy(tsk->nsproxy);
++ vxdprintk(VXD_CBIT(space, 4),
++ "unshare_namespaces(0x%08lx,%p,%p)",
++ flags, orig, new_fs);
++
++ new_nsp = clone_nsproxy(orig);
+ if (!new_nsp)
+ return ERR_PTR(-ENOMEM);
+
+- new_nsp->mnt_ns = copy_mnt_ns(flags, tsk->nsproxy->mnt_ns, new_fs);
++ new_nsp->mnt_ns = copy_mnt_ns(flags, orig->mnt_ns, new_fs);
+ if (IS_ERR(new_nsp->mnt_ns)) {
+ err = PTR_ERR(new_nsp->mnt_ns);
+ goto out_ns;
+ }
+
+- new_nsp->uts_ns = copy_utsname(flags, tsk->nsproxy->uts_ns);
++ new_nsp->uts_ns = copy_utsname(flags, orig->uts_ns);
+ if (IS_ERR(new_nsp->uts_ns)) {
+ err = PTR_ERR(new_nsp->uts_ns);
+ goto out_uts;
+ }
+
+- new_nsp->ipc_ns = copy_ipcs(flags, tsk->nsproxy->ipc_ns);
++ new_nsp->ipc_ns = copy_ipcs(flags, orig->ipc_ns);
+ if (IS_ERR(new_nsp->ipc_ns)) {
+ err = PTR_ERR(new_nsp->ipc_ns);
+ goto out_ipc;
+ }
+
+- new_nsp->pid_ns = copy_pid_ns(flags, task_active_pid_ns(tsk));
++ new_nsp->pid_ns = copy_pid_ns(flags, orig->pid_ns);
+ if (IS_ERR(new_nsp->pid_ns)) {
+ err = PTR_ERR(new_nsp->pid_ns);
+ goto out_pid;
+ }
+
+- new_nsp->user_ns = copy_user_ns(flags, tsk->nsproxy->user_ns);
++ new_nsp->user_ns = copy_user_ns(flags, orig->user_ns);
+ if (IS_ERR(new_nsp->user_ns)) {
+ err = PTR_ERR(new_nsp->user_ns);
+ goto out_user;
+ }
+
+- new_nsp->net_ns = copy_net_ns(flags, tsk->nsproxy->net_ns);
++ new_nsp->net_ns = copy_net_ns(flags, orig->net_ns);
+ if (IS_ERR(new_nsp->net_ns)) {
+ err = PTR_ERR(new_nsp->net_ns);
+ goto out_net;
+@@ -115,6 +124,37 @@ out_ns:
+ return ERR_PTR(err);
+ }
+
++static struct nsproxy *create_new_namespaces(int flags, struct task_struct *tsk,
++ struct fs_struct *new_fs)
++{
++ return unshare_namespaces(flags, tsk->nsproxy, new_fs);
++}
++
++/*
++ * copies the nsproxy, setting refcount to 1, and grabbing a
++ * reference to all contained namespaces.
++ */
++struct nsproxy *copy_nsproxy(struct nsproxy *orig)
++{
++ struct nsproxy *ns = clone_nsproxy(orig);
++
++ if (ns) {
++ if (ns->mnt_ns)
++ get_mnt_ns(ns->mnt_ns);
++ if (ns->uts_ns)
++ get_uts_ns(ns->uts_ns);
++ if (ns->ipc_ns)
++ get_ipc_ns(ns->ipc_ns);
++ if (ns->pid_ns)
++ get_pid_ns(ns->pid_ns);
++ if (ns->user_ns)
++ get_user_ns(ns->user_ns);
++ if (ns->net_ns)
++ get_net(ns->net_ns);
++ }
++ return ns;
++}
++
+ /*
+ * called from clone. This now handles copy for nsproxy and all
+ * namespaces therein.
+@@ -122,9 +162,12 @@ out_ns:
+ int copy_namespaces(unsigned long flags, struct task_struct *tsk)
+ {
+ struct nsproxy *old_ns = tsk->nsproxy;
+- struct nsproxy *new_ns;
++ struct nsproxy *new_ns = NULL;
+ int err = 0;
+
++ vxdprintk(VXD_CBIT(space, 7), "copy_namespaces(0x%08lx,%p[%p])",
++ flags, tsk, old_ns);
++
+ if (!old_ns)
+ return 0;
+
+@@ -155,6 +198,9 @@ int copy_namespaces(unsigned long flags,
+
+ out:
+ put_nsproxy(old_ns);
++ vxdprintk(VXD_CBIT(space, 3),
++ "copy_namespaces(0x%08lx,%p[%p]) = %d [%p]",
++ flags, tsk, old_ns, err, new_ns);
+ return err;
+ }
+
+@@ -171,6 +217,7 @@ void free_nsproxy(struct nsproxy *ns)
+ if (ns->user_ns)
+ put_user_ns(ns->user_ns);
+ put_net(ns->net_ns);
++ atomic_dec(&vs_global_nsproxy);
+ kmem_cache_free(nsproxy_cachep, ns);
+ }
+
+@@ -183,6 +230,10 @@ int unshare_nsproxy_namespaces(unsigned
+ {
+ int err = 0;
+
++ vxdprintk(VXD_CBIT(space, 4),
++ "unshare_nsproxy_namespaces(0x%08lx,[%p])",
++ unshare_flags, current->nsproxy);
++
+ if (!(unshare_flags & (CLONE_NEWNS | CLONE_NEWUTS | CLONE_NEWIPC |
+ CLONE_NEWUSER | CLONE_NEWNET)))
+ return 0;
+--- linux-2.6.25.4/kernel/pid.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/pid.c 2008-04-29 17:56:00.000000000 -0400
+@@ -35,6 +35,8 @@
+ #include <linux/pid_namespace.h>
+ #include <linux/init_task.h>
+ #include <linux/syscalls.h>
++#include <linux/vs_pid.h>
++#include <linux/vserver/global.h>
+
+ #define pid_hashfn(nr, ns) \
+ hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift)
+@@ -303,7 +305,7 @@ EXPORT_SYMBOL_GPL(find_pid_ns);
+
+ struct pid *find_vpid(int nr)
+ {
+- return find_pid_ns(nr, current->nsproxy->pid_ns);
++ return find_pid_ns(vx_rmap_pid(nr), current->nsproxy->pid_ns);
+ }
+ EXPORT_SYMBOL_GPL(find_vpid);
+
+@@ -359,6 +361,9 @@ void transfer_pid(struct task_struct *ol
+ struct task_struct *pid_task(struct pid *pid, enum pid_type type)
+ {
+ struct task_struct *result = NULL;
++
++ if (type == PIDTYPE_REALPID)
++ type = PIDTYPE_PID;
+ if (pid) {
+ struct hlist_node *first;
+ first = rcu_dereference(pid->tasks[type].first);
+@@ -388,14 +393,14 @@ EXPORT_SYMBOL(find_task_by_pid);
+
+ struct task_struct *find_task_by_vpid(pid_t vnr)
+ {
+- return find_task_by_pid_type_ns(PIDTYPE_PID, vnr,
++ return find_task_by_pid_type_ns(PIDTYPE_PID, vx_rmap_pid(vnr),
+ current->nsproxy->pid_ns);
+ }
+ EXPORT_SYMBOL(find_task_by_vpid);
+
+ struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns)
+ {
+- return find_task_by_pid_type_ns(PIDTYPE_PID, nr, ns);
++ return find_task_by_pid_type_ns(PIDTYPE_PID, vx_rmap_pid(nr), ns);
+ }
+ EXPORT_SYMBOL(find_task_by_pid_ns);
+
+@@ -430,7 +435,7 @@ struct pid *find_get_pid(pid_t nr)
+ return pid;
+ }
+
+-pid_t pid_nr_ns(struct pid *pid, struct pid_namespace *ns)
++pid_t pid_unmapped_nr_ns(struct pid *pid, struct pid_namespace *ns)
+ {
+ struct upid *upid;
+ pid_t nr = 0;
+@@ -443,6 +448,11 @@ pid_t pid_nr_ns(struct pid *pid, struct
+ return nr;
+ }
+
++pid_t pid_nr_ns(struct pid *pid, struct pid_namespace *ns)
++{
++ return vx_map_pid(pid_unmapped_nr_ns(pid, ns));
++}
++
+ pid_t pid_vnr(struct pid *pid)
+ {
+ return pid_nr_ns(pid, current->nsproxy->pid_ns);
+--- linux-2.6.25.4/kernel/pid_namespace.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/pid_namespace.c 2008-04-23 11:52:08.000000000 -0400
+@@ -12,6 +12,7 @@
+ #include <linux/pid_namespace.h>
+ #include <linux/syscalls.h>
+ #include <linux/err.h>
++#include <linux/vserver/global.h>
+
+ #define BITS_PER_PAGE (PAGE_SIZE*8)
+
+@@ -84,6 +85,7 @@ static struct pid_namespace *create_pid_
+ goto out_free_map;
+
+ kref_init(&ns->kref);
++ atomic_inc(&vs_global_pid_ns);
+ ns->last_pid = 0;
+ ns->child_reaper = NULL;
+ ns->level = level;
+@@ -112,6 +114,7 @@ static void destroy_pid_namespace(struct
+
+ for (i = 0; i < PIDMAP_ENTRIES; i++)
+ kfree(ns->pidmap[i].page);
++ atomic_dec(&vs_global_pid_ns);
+ kmem_cache_free(pid_ns_cachep, ns);
+ }
+
+--- linux-2.6.25.4/kernel/posix-timers.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/posix-timers.c 2008-04-19 15:14:52.000000000 -0400
+@@ -47,6 +47,7 @@
+ #include <linux/wait.h>
+ #include <linux/workqueue.h>
+ #include <linux/module.h>
++#include <linux/vs_context.h>
+
+ /*
+ * Management arrays for POSIX timers. Timers are kept in slab memory
+@@ -299,6 +300,12 @@ void do_schedule_next_timer(struct sigin
+
+ int posix_timer_event(struct k_itimer *timr,int si_private)
+ {
++ struct vx_info_save vxis;
++ struct vx_info *vxi;
++ int ret;
++
++ vxi = task_get_vx_info(timr->it_process);
++ enter_vx_info(vxi, &vxis);
+ memset(&timr->sigq->info, 0, sizeof(siginfo_t));
+ timr->sigq->info.si_sys_private = si_private;
+ /* Send signal to the process that owns this timer.*/
+@@ -311,11 +318,11 @@ int posix_timer_event(struct k_itimer *t
+
+ if (timr->it_sigev_notify & SIGEV_THREAD_ID) {
+ struct task_struct *leader;
+- int ret = send_sigqueue(timr->it_sigev_signo, timr->sigq,
+- timr->it_process);
+
++ ret = send_sigqueue(timr->it_sigev_signo, timr->sigq,
++ timr->it_process);
+ if (likely(ret >= 0))
+- return ret;
++ goto out;
+
+ timr->it_sigev_notify = SIGEV_SIGNAL;
+ leader = timr->it_process->group_leader;
+@@ -323,8 +330,12 @@ int posix_timer_event(struct k_itimer *t
+ timr->it_process = leader;
+ }
+
+- return send_group_sigqueue(timr->it_sigev_signo, timr->sigq,
++ ret = send_group_sigqueue(timr->it_sigev_signo, timr->sigq,
+ timr->it_process);
++out:
++ leave_vx_info(&vxis);
++ put_vx_info(vxi);
++ return ret;
+ }
+ EXPORT_SYMBOL_GPL(posix_timer_event);
+
+--- linux-2.6.25.4/kernel/printk.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/printk.c 2008-04-21 10:59:28.000000000 -0400
+@@ -32,6 +32,7 @@
+ #include <linux/security.h>
+ #include <linux/bootmem.h>
+ #include <linux/syscalls.h>
++#include <linux/vs_cvirt.h>
+
+ #include <asm/uaccess.h>
+
+@@ -297,18 +298,13 @@ int do_syslog(int type, char __user *buf
+ unsigned i, j, limit, count;
+ int do_clear = 0;
+ char c;
+- int error = 0;
++ int error;
+
+ error = security_syslog(type);
+ if (error)
+ return error;
+
+- switch (type) {
+- case 0: /* Close log */
+- break;
+- case 1: /* Open log */
+- break;
+- case 2: /* Read from log */
++ if ((type >= 2) && (type <= 4)) {
+ error = -EINVAL;
+ if (!buf || len < 0)
+ goto out;
+@@ -319,6 +315,16 @@ int do_syslog(int type, char __user *buf
+ error = -EFAULT;
+ goto out;
+ }
++ }
++ if (!vx_check(0, VS_ADMIN|VS_WATCH))
++ return vx_do_syslog(type, buf, len);
++
++ switch (type) {
++ case 0: /* Close log */
++ break;
++ case 1: /* Open log */
++ break;
++ case 2: /* Read from log */
+ error = wait_event_interruptible(log_wait,
+ (log_start - log_end));
+ if (error)
+@@ -343,16 +349,6 @@ int do_syslog(int type, char __user *buf
+ do_clear = 1;
+ /* FALL THRU */
+ case 3: /* Read last kernel messages */
+- error = -EINVAL;
+- if (!buf || len < 0)
+- goto out;
+- error = 0;
+- if (!len)
+- goto out;
+- if (!access_ok(VERIFY_WRITE, buf, len)) {
+- error = -EFAULT;
+- goto out;
+- }
+ count = len;
+ if (count > log_buf_len)
+ count = log_buf_len;
+--- linux-2.6.25.4/kernel/ptrace.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/ptrace.c 2008-04-21 10:50:28.000000000 -0400
+@@ -21,6 +21,7 @@
+ #include <linux/audit.h>
+ #include <linux/pid_namespace.h>
+ #include <linux/syscalls.h>
++#include <linux/vs_context.h>
+
+ #include <asm/pgtable.h>
+ #include <asm/uaccess.h>
+@@ -147,6 +148,11 @@ int __ptrace_may_attach(struct task_stru
+ dumpable = get_dumpable(task->mm);
+ if (!dumpable && !capable(CAP_SYS_PTRACE))
+ return -EPERM;
++ if (!vx_check(task->xid, VS_ADMIN_P|VS_IDENT))
++ return -EPERM;
++ if (!vx_check(task->xid, VS_IDENT) &&
++ !task_vx_flags(task, VXF_STATE_ADMIN, 0))
++ return -EACCES;
+
+ return security_ptrace(current, task);
+ }
+@@ -562,6 +568,10 @@ asmlinkage long sys_ptrace(long request,
+ goto out;
+ }
+
++ ret = -EPERM;
++ if (!vx_check(vx_task_xid(child), VS_WATCH_P | VS_IDENT))
++ goto out_put_task_struct;
++
+ if (request == PTRACE_ATTACH) {
+ ret = ptrace_attach(child);
+ /*
+--- linux-2.6.25.4/kernel/sched.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/sched.c 2008-05-21 14:30:41.000000000 -0400
+@@ -66,6 +66,8 @@
+ #include <linux/unistd.h>
+ #include <linux/pagemap.h>
+ #include <linux/hrtimer.h>
++#include <linux/vs_sched.h>
++#include <linux/vs_cvirt.h>
+
+ #include <asm/tlb.h>
+ #include <asm/irq_regs.h>
+@@ -375,6 +377,16 @@ struct root_domain {
+ static struct root_domain def_root_domain;
+
+ #endif
++ unsigned long norm_time;
++ unsigned long idle_time;
++#ifdef CONFIG_VSERVER_IDLETIME
++ int idle_skip;
++#endif
++#ifdef CONFIG_VSERVER_HARDCPU
++ struct list_head hold_queue;
++ unsigned long nr_onhold;
++ int idle_tokens;
++#endif
+
+ /*
+ * This is the main, per-CPU runqueue data structure.
+@@ -1366,6 +1378,7 @@ static void set_load_weight(struct task_
+
+ static void enqueue_task(struct rq *rq, struct task_struct *p, int wakeup)
+ {
++ // BUG_ON(p->state & TASK_ONHOLD);
+ sched_info_queued(p);
+ p->sched_class->enqueue_task(rq, p, wakeup);
+ p->se.on_rq = 1;
+@@ -1556,6 +1569,9 @@ struct migration_req {
+ struct completion done;
+ };
+
++#include "sched_mon.h"
++
++
+ /*
+ * The task's runqueue lock must be held.
+ * Returns true if you have to wait for migration thread.
+@@ -1565,6 +1581,7 @@ migrate_task(struct task_struct *p, int
+ {
+ struct rq *rq = task_rq(p);
+
++ vxm_migrate_task(p, rq, dest_cpu);
+ /*
+ * If the task is not on a runqueue (and not running), then
+ * it is sufficient to simply update the task's cpu field.
+@@ -1926,6 +1943,12 @@ static int try_to_wake_up(struct task_st
+ /* might preempt at this point */
+ rq = task_rq_lock(p, &flags);
+ old_state = p->state;
++
++ /* we need to unhold suspended tasks
++ if (old_state & TASK_ONHOLD) {
++ vx_unhold_task(p, rq);
++ old_state = p->state;
++ } */
+ if (!(old_state & state))
+ goto out;
+ if (p->se.on_rq)
+@@ -3697,13 +3720,16 @@ unsigned long long task_sched_runtime(st
+ void account_user_time(struct task_struct *p, cputime_t cputime)
+ {
+ struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
++ struct vx_info *vxi = p->vx_info; /* p is _always_ current */
+ cputime64_t tmp;
++ int nice = (TASK_NICE(p) > 0);
+
+ p->utime = cputime_add(p->utime, cputime);
++ vx_account_user(vxi, cputime, nice);
+
+ /* Add user time to cpustat. */
+ tmp = cputime_to_cputime64(cputime);
+- if (TASK_NICE(p) > 0)
++ if (nice)
+ cpustat->nice = cputime64_add(cpustat->nice, tmp);
+ else
+ cpustat->user = cputime64_add(cpustat->user, tmp);
+@@ -3748,6 +3774,7 @@ void account_system_time(struct task_str
+ cputime_t cputime)
+ {
+ struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
++ struct vx_info *vxi = p->vx_info; /* p is _always_ current */
+ struct rq *rq = this_rq();
+ cputime64_t tmp;
+
+@@ -3755,6 +3782,7 @@ void account_system_time(struct task_str
+ return account_guest_time(p, cputime);
+
+ p->stime = cputime_add(p->stime, cputime);
++ vx_account_system(vxi, cputime, (p == rq->idle));
+
+ /* Add system time to cpustat. */
+ tmp = cputime_to_cputime64(cputime);
+@@ -4500,7 +4528,7 @@ asmlinkage long sys_nice(int increment)
+ nice = 19;
+
+ if (increment < 0 && !can_nice(current, nice))
+- return -EPERM;
++ return vx_flags(VXF_IGNEG_NICE, 0) ? 0 : -EPERM;
+
+ retval = security_task_setnice(current, nice);
+ if (retval)
+--- linux-2.6.25.4/kernel/sched_fair.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/sched_fair.c 2008-04-19 15:14:52.000000000 -0400
+@@ -537,6 +537,8 @@ enqueue_entity(struct cfs_rq *cfs_rq, st
+ check_spread(cfs_rq, se);
+ if (se != cfs_rq->curr)
+ __enqueue_entity(cfs_rq, se);
++ if (entity_is_task(se))
++ vx_activate_task(task_of(se));
+ account_entity_enqueue(cfs_rq, se);
+ }
+
+@@ -580,6 +582,8 @@ dequeue_entity(struct cfs_rq *cfs_rq, st
+
+ if (se != cfs_rq->curr)
+ __dequeue_entity(cfs_rq, se);
++ if (entity_is_task(se))
++ vx_deactivate_task(task_of(se));
+ account_entity_dequeue(cfs_rq, se);
+ }
+
+--- linux-2.6.25.4/kernel/sched_hard.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/sched_hard.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,324 @@
++
++#ifdef CONFIG_VSERVER_IDLELIMIT
++
++/*
++ * vx_idle_resched - reschedule after maxidle
++ */
++static inline
++void vx_idle_resched(struct rq *rq)
++{
++ /* maybe have a better criterion for paused */
++ if (!--rq->idle_tokens && !list_empty(&rq->hold_queue))
++ set_need_resched();
++}
++
++#else /* !CONFIG_VSERVER_IDLELIMIT */
++
++#define vx_idle_resched(rq)
++
++#endif /* CONFIG_VSERVER_IDLELIMIT */
++
++
++
++#ifdef CONFIG_VSERVER_IDLETIME
++
++#define vx_set_rq_min_skip(rq, min) \
++ (rq)->idle_skip = (min)
++
++#define vx_save_min_skip(ret, min, val) \
++ __vx_save_min_skip(ret, min, val)
++
++static inline
++void __vx_save_min_skip(int ret, int *min, int val)
++{
++ if (ret > -2)
++ return;
++ if ((*min > val) || !*min)
++ *min = val;
++}
++
++static inline
++int vx_try_skip(struct rq *rq, int cpu)
++{
++ /* artificially advance time */
++ if (rq->idle_skip > 0) {
++ vxdprintk(list_empty(&rq->hold_queue),
++ "hold queue empty on cpu %d", cpu);
++ rq->idle_time += rq->idle_skip;
++ vxm_idle_skip(rq, cpu);
++ return 1;
++ }
++ return 0;
++}
++
++#else /* !CONFIG_VSERVER_IDLETIME */
++
++#define vx_set_rq_min_skip(rq, min) \
++ ({ int dummy = (min); dummy; })
++
++#define vx_save_min_skip(ret, min, val)
++
++static inline
++int vx_try_skip(struct rq *rq, int cpu)
++{
++ return 0;
++}
++
++#endif /* CONFIG_VSERVER_IDLETIME */
++
++
++
++#ifdef CONFIG_VSERVER_HARDCPU
++
++#define vx_set_rq_max_idle(rq, max) \
++ (rq)->idle_tokens = (max)
++
++#define vx_save_max_idle(ret, min, val) \
++ __vx_save_max_idle(ret, min, val)
++
++static inline
++void __vx_save_max_idle(int ret, int *min, int val)
++{
++ if (*min > val)
++ *min = val;
++}
++
++
++/*
++ * vx_hold_task - put a task on the hold queue
++ */
++static inline
++void vx_hold_task(struct task_struct *p, struct rq *rq)
++{
++ __deactivate_task(p, rq);
++ p->state |= TASK_ONHOLD;
++ /* a new one on hold */
++ rq->nr_onhold++;
++ vxm_hold_task(p, rq);
++ list_add_tail(&p->run_list, &rq->hold_queue);
++}
++
++/*
++ * vx_unhold_task - put a task back to the runqueue
++ */
++static inline
++void vx_unhold_task(struct task_struct *p, struct rq *rq)
++{
++ list_del(&p->run_list);
++ /* one less waiting */
++ rq->nr_onhold--;
++ p->state &= ~TASK_ONHOLD;
++ enqueue_task(p, rq->expired);
++ inc_nr_running(p, rq);
++ vxm_unhold_task(p, rq);
++
++ if (p->static_prio < rq->best_expired_prio)
++ rq->best_expired_prio = p->static_prio;
++}
++
++unsigned long nr_onhold(void)
++{
++ unsigned long i, sum = 0;
++
++ for_each_online_cpu(i)
++ sum += cpu_rq(i)->nr_onhold;
++
++ return sum;
++}
++
++
++
++static inline
++int __vx_tokens_avail(struct _vx_sched_pc *sched_pc)
++{
++ return sched_pc->tokens;
++}
++
++static inline
++void __vx_consume_token(struct _vx_sched_pc *sched_pc)
++{
++ sched_pc->tokens--;
++}
++
++static inline
++int vx_need_resched(struct task_struct *p, int slice, int cpu)
++{
++ struct vx_info *vxi = p->vx_info;
++
++ if (vx_info_flags(vxi, VXF_SCHED_HARD|VXF_SCHED_PRIO, 0)) {
++ struct _vx_sched_pc *sched_pc =
++ &vx_per_cpu(vxi, sched_pc, cpu);
++ int tokens;
++
++ /* maybe we can simplify that to decrement
++ the token counter unconditional? */
++
++ if ((tokens = __vx_tokens_avail(sched_pc)) > 0)
++ __vx_consume_token(sched_pc);
++
++ /* for tokens > 0, one token was consumed */
++ if (tokens < 2)
++ slice = 0;
++ }
++ vxm_need_resched(p, slice, cpu);
++ return (slice == 0);
++}
++
++
++#define vx_set_rq_time(rq, time) do { \
++ rq->norm_time = time; \
++} while (0)
++
++
++static inline
++void vx_try_unhold(struct rq *rq, int cpu)
++{
++ struct vx_info *vxi = NULL;
++ struct list_head *l, *n;
++ int maxidle = HZ;
++ int minskip = 0;
++
++ /* nothing to do? what about pause? */
++ if (list_empty(&rq->hold_queue))
++ return;
++
++ list_for_each_safe(l, n, &rq->hold_queue) {
++ int ret, delta_min[2];
++ struct _vx_sched_pc *sched_pc;
++ struct task_struct *p;
++
++ p = list_entry(l, struct task_struct, run_list);
++ /* don't bother with same context */
++ if (vxi == p->vx_info)
++ continue;
++
++ vxi = p->vx_info;
++ /* ignore paused contexts */
++ if (vx_info_flags(vxi, VXF_SCHED_PAUSE, 0))
++ continue;
++
++ sched_pc = &vx_per_cpu(vxi, sched_pc, cpu);
++
++ /* recalc tokens */
++ vxm_sched_info(sched_pc, vxi, cpu);
++ ret = vx_tokens_recalc(sched_pc,
++ &rq->norm_time, &rq->idle_time, delta_min);
++ vxm_tokens_recalc(sched_pc, rq, vxi, cpu);
++
++ if (ret > 0) {
++ /* we found a runable context */
++ vx_unhold_task(p, rq);
++ break;
++ }
++ vx_save_max_idle(ret, &maxidle, delta_min[0]);
++ vx_save_min_skip(ret, &minskip, delta_min[1]);
++ }
++ vx_set_rq_max_idle(rq, maxidle);
++ vx_set_rq_min_skip(rq, minskip);
++ vxm_rq_max_min(rq, cpu);
++}
++
++
++static inline
++int vx_schedule(struct task_struct *next, struct rq *rq, int cpu)
++{
++ struct vx_info *vxi = next->vx_info;
++ struct _vx_sched_pc *sched_pc;
++ int delta_min[2];
++ int flags, ret;
++
++ if (!vxi)
++ return 1;
++
++ flags = vxi->vx_flags;
++
++ if (unlikely(vs_check_flags(flags, VXF_SCHED_PAUSE, 0)))
++ goto put_on_hold;
++ if (!vs_check_flags(flags, VXF_SCHED_HARD | VXF_SCHED_PRIO, 0))
++ return 1;
++
++ sched_pc = &vx_per_cpu(vxi, sched_pc, cpu);
++#ifdef CONFIG_SMP
++ /* update scheduler params */
++ if (cpu_isset(cpu, vxi->sched.update)) {
++ vx_update_sched_param(&vxi->sched, sched_pc);
++ vxm_update_sched(sched_pc, vxi, cpu);
++ cpu_clear(cpu, vxi->sched.update);
++ }
++#endif
++ vxm_sched_info(sched_pc, vxi, cpu);
++ ret = vx_tokens_recalc(sched_pc,
++ &rq->norm_time, &rq->idle_time, delta_min);
++ vxm_tokens_recalc(sched_pc, rq, vxi, cpu);
++
++ if (!vs_check_flags(flags, VXF_SCHED_HARD, 0))
++ return 1;
++
++ if (unlikely(ret < 0)) {
++ vx_save_max_idle(ret, &rq->idle_tokens, delta_min[0]);
++ vx_save_min_skip(ret, &rq->idle_skip, delta_min[1]);
++ vxm_rq_max_min(rq, cpu);
++ put_on_hold:
++ vx_hold_task(next, rq);
++ return 0;
++ }
++ return 1;
++}
++
++
++#else /* CONFIG_VSERVER_HARDCPU */
++
++static inline
++void vx_hold_task(struct task_struct *p, struct rq *rq)
++{
++ return;
++}
++
++static inline
++void vx_unhold_task(struct task_struct *p, struct rq *rq)
++{
++ return;
++}
++
++unsigned long nr_onhold(void)
++{
++ return 0;
++}
++
++
++static inline
++int vx_need_resched(struct task_struct *p, int slice, int cpu)
++{
++ return (slice == 0);
+}
-
-- return pid;
-+#ifdef __alpha__
+
-+/*
-+ * The Alpha uses getxpid, getxuid, and getxgid instead.
-+ */
+
-+asmlinkage long do_getxpid(long *ppid)
++#define vx_set_rq_time(rq, time)
++
++static inline
++void vx_try_unhold(struct rq *rq, int cpu)
+{
-+ *ppid = sys_getppid();
-+ return sys_getpid();
- }
-
-+#else /* _alpha_ */
++ return;
++}
+
- asmlinkage long sys_getuid(void)
- {
- /* Only we change this so SMP safe */
-@@ -1118,6 +1129,8 @@ int do_sysinfo(struct sysinfo *info)
- tp.tv_nsec = tp.tv_nsec - NSEC_PER_SEC;
- tp.tv_sec++;
- }
-+ if (vx_flags(VXF_VIRT_UPTIME, 0))
-+ vx_vsi_uptime(&tp, NULL);
- info->uptime = tp.tv_sec + (tp.tv_nsec ? 1 : 0);
-
- info->loads[0] = avenrun[0] << (SI_LOAD_SHIFT - FSHIFT);
---- linux-2.6.22/kernel/user.c 2007-02-06 03:01:56 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/user.c 2007-06-15 02:37:04 +0200
-@@ -23,8 +23,8 @@
- #define UIDHASH_BITS (CONFIG_BASE_SMALL ? 3 : 8)
- #define UIDHASH_SZ (1 << UIDHASH_BITS)
- #define UIDHASH_MASK (UIDHASH_SZ - 1)
--#define __uidhashfn(uid) (((uid >> UIDHASH_BITS) + uid) & UIDHASH_MASK)
--#define uidhashentry(uid) (uidhash_table + __uidhashfn((uid)))
-+#define __uidhashfn(xid,uid) ((((uid) >> UIDHASH_BITS) + ((uid)^(xid))) & UIDHASH_MASK)
-+#define uidhashentry(xid,uid) (uidhash_table + __uidhashfn((xid),(uid)))
-
- static struct kmem_cache *uid_cachep;
- static struct list_head uidhash_table[UIDHASH_SZ];
-@@ -66,7 +66,7 @@ static inline void uid_hash_remove(struc
- list_del(&up->uidhash_list);
- }
-
--static inline struct user_struct *uid_hash_find(uid_t uid, struct list_head *hashent)
-+static inline struct user_struct *uid_hash_find(xid_t xid, uid_t uid, struct list_head *hashent)
- {
- struct list_head *up;
-
-@@ -75,7 +75,7 @@ static inline struct user_struct *uid_ha
-
- user = list_entry(up, struct user_struct, uidhash_list);
-
-- if(user->uid == uid) {
-+ if(user->uid == uid && user->xid == xid) {
- atomic_inc(&user->__count);
- return user;
- }
-@@ -90,13 +90,13 @@ static inline struct user_struct *uid_ha
- *
- * If the user_struct could not be found, return NULL.
- */
--struct user_struct *find_user(uid_t uid)
-+struct user_struct *find_user(xid_t xid, uid_t uid)
- {
- struct user_struct *ret;
- unsigned long flags;
-
- spin_lock_irqsave(&uidhash_lock, flags);
-- ret = uid_hash_find(uid, uidhashentry(uid));
-+ ret = uid_hash_find(xid, uid, uidhashentry(xid, uid));
- spin_unlock_irqrestore(&uidhash_lock, flags);
- return ret;
- }
-@@ -120,13 +120,13 @@ void free_uid(struct user_struct *up)
- }
- }
-
--struct user_struct * alloc_uid(uid_t uid)
-+struct user_struct * alloc_uid(xid_t xid, uid_t uid)
- {
-- struct list_head *hashent = uidhashentry(uid);
-+ struct list_head *hashent = uidhashentry(xid, uid);
- struct user_struct *up;
-
- spin_lock_irq(&uidhash_lock);
-- up = uid_hash_find(uid, hashent);
-+ up = uid_hash_find(xid, uid, hashent);
- spin_unlock_irq(&uidhash_lock);
-
- if (!up) {
-@@ -136,6 +136,7 @@ struct user_struct * alloc_uid(uid_t uid
- if (!new)
- return NULL;
- new->uid = uid;
-+ new->xid = xid;
- atomic_set(&new->__count, 1);
- atomic_set(&new->processes, 0);
- atomic_set(&new->files, 0);
-@@ -158,7 +159,7 @@ struct user_struct * alloc_uid(uid_t uid
- * on adding the same user already..
- */
- spin_lock_irq(&uidhash_lock);
-- up = uid_hash_find(uid, hashent);
-+ up = uid_hash_find(xid, uid, hashent);
- if (up) {
- key_put(new->uid_keyring);
- key_put(new->session_keyring);
-@@ -215,7 +216,7 @@ static int __init uid_cache_init(void)
-
- /* Insert the root user immediately (init already runs as root) */
- spin_lock_irq(&uidhash_lock);
-- uid_hash_insert(&root_user, uidhashentry(0));
-+ uid_hash_insert(&root_user, uidhashentry(0,0));
- spin_unlock_irq(&uidhash_lock);
-
- return 0;
---- linux-2.6.22/kernel/utsname.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/utsname.c 2007-06-15 02:37:04 +0200
-@@ -13,6 +13,7 @@
- #include <linux/uts.h>
- #include <linux/utsname.h>
- #include <linux/version.h>
-+#include <linux/vserver/global.h>
-
- /*
- * Clone a new ns copying an original utsname, setting refcount to 1
-@@ -27,6 +28,7 @@ static struct uts_namespace *clone_uts_n
- if (ns) {
- memcpy(&ns->name, &old_ns->name, sizeof(ns->name));
- kref_init(&ns->kref);
-+ atomic_inc(&vs_global_uts_ns);
- }
- return ns;
- }
-@@ -58,5 +60,6 @@ void free_uts_ns(struct kref *kref)
- struct uts_namespace *ns;
-
- ns = container_of(kref, struct uts_namespace, kref);
-+ atomic_dec(&vs_global_uts_ns);
- kfree(ns);
- }
---- linux-2.6.22/kernel/vserver/Kconfig 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/Kconfig 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,284 @@
-+#
-+# Linux VServer configuration
-+#
++static inline
++int vx_schedule(struct task_struct *next, struct rq *rq, int cpu)
++{
++ struct vx_info *vxi = next->vx_info;
++ struct _vx_sched_pc *sched_pc;
++ int delta_min[2];
++ int ret;
+
-+menu "Linux VServer"
++ if (!vx_info_flags(vxi, VXF_SCHED_PRIO, 0))
++ return 1;
+
-+config VSERVER_LEGACY
-+ bool "Enable Legacy Kernel API"
-+ depends on EXPERIMENTAL
-+ default n
-+ help
-+ This enables the legacy API used in vs1.xx, maintaining
-+ compatibility with older vserver tools, and guest images
-+ that are configured using the legacy method.
-+
-+config VSERVER_LEGACY_VERSION
-+ bool "Show a Legacy Version ID"
-+ depends on VSERVER_LEGACY
-+ default n
-+ help
-+ This shows a special legacy version to very old tools
-+ which do not handle the current version correctly.
++ sched_pc = &vx_per_cpu(vxi, sched_pc, cpu);
++ vxm_sched_info(sched_pc, vxi, cpu);
++ ret = vx_tokens_recalc(sched_pc,
++ &rq->norm_time, &rq->idle_time, delta_min);
++ vxm_tokens_recalc(sched_pc, rq, vxi, cpu);
++ return 1;
++}
+
-+ Warning: recent tools are not able to utilize the
-+ newer APIs when this is enabled, so some features will
-+ not be available. Better avoid it, unless you really,
-+ really need it for backwards compatibility.
-+
-+config VSERVER_DYNAMIC_IDS
-+ bool "Enable dynamic context IDs"
-+ depends on EXPERIMENTAL && VSERVER_LEGACY
-+ default n
-+ help
-+ This enables support of in kernel dynamic context IDs,
-+ which is deprecated and will probably be removed in the
-+ next release.
++#endif /* CONFIG_VSERVER_HARDCPU */
+
-+config VSERVER_LEGACYNET
-+ bool "Enable Legacy Networking Kernel API"
-+ depends on EXPERIMENTAL
-+ default n
-+ help
-+ This enables the legacy networking API which is used
-+ by older tools (pre 0.30.210) to set up the network
-+ context (chbind).
+--- linux-2.6.25.4/kernel/sched_mon.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/sched_mon.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,200 @@
+
-+config VSERVER_REMAP_SADDR
-+ bool "Remap Source IP Address"
-+ depends on EXPERIMENTAL
-+ default n
-+ help
-+ This allows to remap the source IP address of 'local'
-+ connections from 127.0.0.1 to the first assigned
-+ guest IP.
++#include <linux/vserver/monitor.h>
+
-+config VSERVER_COWBL
-+ bool "Enable COW Immutable Link Breaking"
-+ default y
-+ help
-+ This enables the COW (Copy-On-Write) link break code.
-+ It allows you to treat unified files like normal files
-+ when writing to them (which will implicitely break the
-+ link and create a copy of the unified file)
++#ifdef CONFIG_VSERVER_MONITOR
+
-+config VSERVER_VTIME
-+ bool "Enable Virtualized Guest Time"
-+ depends on EXPERIMENTAL
-+ default n
-+ help
-+ This enables per guest time offsets to allow for
-+ adjusting the system clock individually per guest.
-+ this adds some overhead to the time functions and
-+ therefore should not be enabled without good reason.
++#ifdef CONFIG_VSERVER_HARDCPU
++#define HARDCPU(x) (x)
++#else
++#define HARDCPU(x) (0)
++#endif
+
-+config VSERVER_PROC_SECURE
-+ bool "Enable Proc Security"
-+ depends on PROC_FS
-+ default y
-+ help
-+ This configures ProcFS security to initially hide
-+ non-process entries for all contexts except the main and
-+ spectator context (i.e. for all guests), which is a secure
-+ default.
++#ifdef CONFIG_VSERVER_IDLETIME
++#define IDLETIME(x) (x)
++#else
++#define IDLETIME(x) (0)
++#endif
+
-+ (note: on 1.2x the entries were visible by default)
++struct _vx_mon_entry *vxm_advance(int cpu);
+
-+config VSERVER_HARDCPU
-+ bool "Enable Hard CPU Limits"
-+ default y
-+ help
-+ Activate the Hard CPU Limits
+
-+ This will compile in code that allows the Token Bucket
-+ Scheduler to put processes on hold when a context's
-+ tokens are depleted (provided that its per-context
-+ sched_hard flag is set).
++static inline
++void __vxm_basic(struct _vx_mon_entry *entry, xid_t xid, int type)
++{
++ entry->type = type;
++ entry->xid = xid;
++}
+
-+ Processes belonging to that context will not be able
-+ to consume CPU resources again until a per-context
-+ configured minimum of tokens has been reached.
++static inline
++void __vxm_sync(int cpu)
++{
++ struct _vx_mon_entry *entry = vxm_advance(cpu);
+
-+config VSERVER_IDLETIME
-+ bool "Avoid idle CPUs by skipping Time"
-+ depends on VSERVER_HARDCPU
-+ default y
-+ help
-+ This option allows the scheduler to artificially
-+ advance time (per cpu) when otherwise the idle
-+ task would be scheduled, thus keeping the cpu
-+ busy and sharing the available resources among
-+ certain contexts.
++ __vxm_basic(entry, 0, VXM_SYNC);
++ entry->ev.sec = xtime.tv_sec;
++ entry->ev.nsec = xtime.tv_nsec;
++}
+
-+config VSERVER_IDLELIMIT
-+ bool "Limit the IDLE task"
-+ depends on VSERVER_HARDCPU
-+ default n
-+ help
-+ Limit the idle slices, so the the next context
-+ will be scheduled as soon as possible.
++static inline
++void __vxm_task(struct task_struct *p, int type)
++{
++ struct _vx_mon_entry *entry = vxm_advance(task_cpu(p));
+
-+ This might improve interactivity and latency, but
-+ will also marginally increase scheduling overhead.
++ __vxm_basic(entry, p->xid, type);
++ entry->ev.tsk.pid = p->pid;
++ entry->ev.tsk.state = p->state;
++}
+
-+choice
-+ prompt "Persistent Inode Tagging"
-+ default TAGGING_ID24
-+ help
-+ This adds persistent context information to filesystems
-+ mounted with the tagxid option. Tagging is a requirement
-+ for per-context disk limits and per-context quota.
++static inline
++void __vxm_sched(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
++{
++ struct _vx_mon_entry *entry = vxm_advance(cpu);
++
++ __vxm_basic(entry, vxi->vx_id, (VXM_SCHED | s->flags));
++ entry->sd.tokens = s->tokens;
++ entry->sd.norm_time = s->norm_time;
++ entry->sd.idle_time = s->idle_time;
++}
+
++static inline
++void __vxm_rqinfo1(struct rq *q, int cpu)
++{
++ struct _vx_mon_entry *entry = vxm_advance(cpu);
+
-+config TAGGING_NONE
-+ bool "Disabled"
-+ help
-+ do not store per-context information in inodes.
++ entry->type = VXM_RQINFO_1;
++ entry->xid = ((unsigned long)q >> 16) & 0xffff;
++ entry->q1.running = q->nr_running;
++ entry->q1.onhold = HARDCPU(q->nr_onhold);
++ entry->q1.iowait = atomic_read(&q->nr_iowait);
++ entry->q1.uintr = q->nr_uninterruptible;
++ entry->q1.idle_tokens = IDLETIME(q->idle_tokens);
++}
++
++static inline
++void __vxm_rqinfo2(struct rq *q, int cpu)
++{
++ struct _vx_mon_entry *entry = vxm_advance(cpu);
++
++ entry->type = VXM_RQINFO_2;
++ entry->xid = (unsigned long)q & 0xffff;
++ entry->q2.norm_time = q->norm_time;
++ entry->q2.idle_time = q->idle_time;
++ entry->q2.idle_skip = IDLETIME(q->idle_skip);
++}
++
++static inline
++void __vxm_update(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
++{
++ struct _vx_mon_entry *entry = vxm_advance(cpu);
++
++ __vxm_basic(entry, vxi->vx_id, VXM_UPDATE);
++ entry->ev.tokens = s->tokens;
++}
++
++static inline
++void __vxm_update1(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
++{
++ struct _vx_mon_entry *entry = vxm_advance(cpu);
++
++ __vxm_basic(entry, vxi->vx_id, VXM_UPDATE_1);
++ entry->u1.tokens_max = s->tokens_max;
++ entry->u1.fill_rate = s->fill_rate[0];
++ entry->u1.interval = s->interval[0];
++}
+
-+config TAGGING_UID16
-+ bool "UID16/GID32"
-+ help
-+ reduces UID to 16 bit, but leaves GID at 32 bit.
++static inline
++void __vxm_update2(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
++{
++ struct _vx_mon_entry *entry = vxm_advance(cpu);
+
-+config TAGGING_GID16
-+ bool "UID32/GID16"
-+ help
-+ reduces GID to 16 bit, but leaves UID at 32 bit.
++ __vxm_basic(entry, vxi->vx_id, VXM_UPDATE_2);
++ entry->u2.tokens_min = s->tokens_min;
++ entry->u2.fill_rate = s->fill_rate[1];
++ entry->u2.interval = s->interval[1];
++}
+
-+config TAGGING_ID24
-+ bool "UID24/GID24"
-+ help
-+ uses the upper 8bit from UID and GID for XID tagging
-+ which leaves 24bit for UID/GID each, which should be
-+ more than sufficient for normal use.
+
-+config TAGGING_INTERN
-+ bool "UID32/GID32"
-+ help
-+ this uses otherwise reserved inode fields in the on
-+ disk representation, which limits the use to a few
-+ filesystems (currently ext2 and ext3)
++#define vxm_activate_task(p,q) __vxm_task(p, VXM_ACTIVATE)
++#define vxm_activate_idle(p,q) __vxm_task(p, VXM_IDLE)
++#define vxm_deactivate_task(p,q) __vxm_task(p, VXM_DEACTIVATE)
++#define vxm_hold_task(p,q) __vxm_task(p, VXM_HOLD)
++#define vxm_unhold_task(p,q) __vxm_task(p, VXM_UNHOLD)
+
-+endchoice
++static inline
++void vxm_migrate_task(struct task_struct *p, struct rq *rq, int dest)
++{
++ __vxm_task(p, VXM_MIGRATE);
++ __vxm_rqinfo1(rq, task_cpu(p));
++ __vxm_rqinfo2(rq, task_cpu(p));
++}
+
-+config TAG_NFSD
-+ bool "Tag NFSD User Auth and Files"
-+ default n
-+ help
-+ Enable this if you do want the in-kernel NFS
-+ Server to use the tagging specified above.
-+ (will require patched clients too)
++static inline
++void vxm_idle_skip(struct rq *rq, int cpu)
++{
++ __vxm_rqinfo1(rq, cpu);
++ __vxm_rqinfo2(rq, cpu);
++}
+
-+config PROPAGATE
-+ bool "Enable Inode Tag Propagation"
-+ default n
-+ depends on EXPERIMENTAL
-+ help
-+ This allows for the tagid= mount option to specify
-+ a tagid which is to be used for the entire mount
-+ tree.
++static inline
++void vxm_need_resched(struct task_struct *p, int slice, int cpu)
++{
++ if (slice)
++ return;
+
-+config VSERVER_PRIVACY
-+ bool "Honor Privacy Aspects of Guests"
-+ default y
-+ help
-+ When enabled, most context checks will disallow
-+ access to structures assigned to a specific context,
-+ like ptys or loop devices.
++ __vxm_task(p, VXM_RESCHED);
++}
+
-+config VSERVER_CONTEXTS
-+ int "Maximum number of Contexts (1-65533)" if EMBEDDED
-+ range 1 65533
-+ default "768" if 64BIT
-+ default "256"
-+ help
-+ This setting will optimize certain data structures
-+ and memory allocations according to the expected
-+ maximum.
++static inline
++void vxm_sync(unsigned long now, int cpu)
++{
++ if (!CONFIG_VSERVER_MONITOR_SYNC ||
++ (now % CONFIG_VSERVER_MONITOR_SYNC))
++ return;
+
-+ note: this is not a strict upper limit.
++ __vxm_sync(cpu);
++}
+
-+config VSERVER_WARN
-+ bool "VServer Warnings"
-+ default y
-+ help
-+ This enables various runtime warnings, which will
-+ notify about potential manipulation attempts or
-+ resource shortage. It is generally considered to
-+ be a good idea to have that enabled.
++#define vxm_sched_info(s,v,c) __vxm_sched(s,v,c)
+
-+config VSERVER_DEBUG
-+ bool "VServer Debugging Code"
-+ default n
-+ help
-+ Set this to yes if you want to be able to activate
-+ debugging output at runtime. It adds a very small
-+ overhead to all vserver related functions and
-+ increases the kernel size by about 20k.
++static inline
++void vxm_tokens_recalc(struct _vx_sched_pc *s, struct rq *rq,
++ struct vx_info *vxi, int cpu)
++{
++ __vxm_sched(s, vxi, cpu);
++ __vxm_rqinfo2(rq, cpu);
++}
+
-+config VSERVER_HISTORY
-+ bool "VServer History Tracing"
-+ depends on VSERVER_DEBUG
-+ default n
-+ help
-+ Set this to yes if you want to record the history of
-+ linux-vserver activities, so they can be replayed in
-+ the event of a kernel panic or oops.
++static inline
++void vxm_update_sched(struct _vx_sched_pc *s, struct vx_info *vxi, int cpu)
++{
++ __vxm_sched(s, vxi, cpu);
++ __vxm_update(s, vxi, cpu);
++ __vxm_update1(s, vxi, cpu);
++ __vxm_update2(s, vxi, cpu);
++}
+
-+config VSERVER_HISTORY_SIZE
-+ int "Per-CPU History Size (32-65536)"
-+ depends on VSERVER_HISTORY
-+ range 32 65536
-+ default 64
-+ help
-+ This allows you to specify the number of entries in
-+ the per-CPU history buffer.
++static inline
++void vxm_rq_max_min(struct rq *rq, int cpu)
++{
++ __vxm_rqinfo1(rq, cpu);
++ __vxm_rqinfo2(rq, cpu);
++}
+
-+config VSERVER_MONITOR
-+ bool "VServer Scheduling Monitor"
-+ depends on VSERVER_DEBUG
-+ default n
-+ help
-+ Set this to yes if you want to record the scheduling
-+ decisions, so that they can be relayed to userspace
-+ for detailed analysis.
++#else /* CONFIG_VSERVER_MONITOR */
+
-+config VSERVER_MONITOR_SIZE
-+ int "Per-CPU Monitor Queue Size (32-65536)"
-+ depends on VSERVER_MONITOR
-+ range 32 65536
-+ default 1024
-+ help
-+ This allows you to specify the number of entries in
-+ the per-CPU scheduling monitor buffer.
++#define vxm_activate_task(t,q) do { } while (0)
++#define vxm_activate_idle(t,q) do { } while (0)
++#define vxm_deactivate_task(t,q) do { } while (0)
++#define vxm_hold_task(t,q) do { } while (0)
++#define vxm_unhold_task(t,q) do { } while (0)
++#define vxm_migrate_task(t,q,d) do { } while (0)
++#define vxm_idle_skip(q,c) do { } while (0)
++#define vxm_need_resched(t,s,c) do { } while (0)
++#define vxm_sync(s,c) do { } while (0)
++#define vxm_sched_info(s,v,c) do { } while (0)
++#define vxm_tokens_recalc(s,q,v,c) do { } while (0)
++#define vxm_update_sched(s,v,c) do { } while (0)
++#define vxm_rq_max_min(q,c) do { } while (0)
+
-+config VSERVER_MONITOR_SYNC
-+ int "Per-CPU Monitor Sync Interval (0-65536)"
-+ depends on VSERVER_MONITOR
-+ range 0 65536
-+ default 256
-+ help
-+ This allows you to specify the interval in ticks
-+ when a time sync entry is inserted.
++#endif /* CONFIG_VSERVER_MONITOR */
+
-+endmenu
+--- linux-2.6.25.4/kernel/signal.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/signal.c 2008-05-15 15:41:03.000000000 -0400
+@@ -26,6 +26,8 @@
+ #include <linux/freezer.h>
+ #include <linux/pid_namespace.h>
+ #include <linux/nsproxy.h>
++#include <linux/vs_context.h>
++#include <linux/vs_pid.h>
+
+ #include <asm/param.h>
+ #include <asm/uaccess.h>
+@@ -530,6 +532,14 @@ static int check_kill_permission(int sig
+ if (!valid_signal(sig))
+ return error;
+
++ if ((info != SEND_SIG_NOINFO) &&
++ (is_si_special(info) || !SI_FROMUSER(info)))
++ goto skip;
+
++ vxdprintk(VXD_CBIT(misc, 7),
++ "check_kill_permission(%d,%p,%p[#%u,%u])",
++ sig, info, t, vx_task_xid(t), t->pid);
+
-+config VSERVER
-+ bool
-+ default y
-+ select UTS_NS
-+ select SYSVIPC
-+ select IPC_NS
+ if (info == SEND_SIG_NOINFO || (!is_si_special(info) && SI_FROMUSER(info))) {
+ error = audit_signal_info(sig, t); /* Let audit system see the signal */
+ if (error)
+@@ -543,6 +553,18 @@ static int check_kill_permission(int sig
+ return error;
+ }
+
++ error = -EPERM;
++ if (t->pid == 1 && current->xid)
++ return error;
+
-+config VSERVER_SECURITY
-+ bool
-+ depends on SECURITY
-+ default y
-+ select SECURITY_CAPABILITIES
++ error = -ESRCH;
++ if (!vx_check(vx_task_xid(t), VS_WATCH_P | VS_IDENT)) {
++ vxdprintk(current->xid || VXD_CBIT(misc, 7),
++ "signal %d[%p] xid mismatch %p[#%u,%u] xid=#%u",
++ sig, info, t, vx_task_xid(t), t->pid, current->xid);
++ return error;
++ }
++skip:
+ return security_task_kill(t, info, sig, 0);
+ }
+
+@@ -1088,7 +1110,7 @@ int kill_pid_info_as_uid(int sig, struct
+
+ read_lock(&tasklist_lock);
+ p = pid_task(pid, PIDTYPE_PID);
+- if (!p) {
++ if (!p || !vx_check(vx_task_xid(p), VS_IDENT)) {
+ ret = -ESRCH;
+ goto out_unlock;
+ }
+@@ -1140,7 +1162,9 @@ static int kill_something_info(int sig,
+ struct task_struct * p;
+
+ for_each_process(p) {
+- if (p->pid > 1 && !same_thread_group(p, current)) {
++ if (vx_check(vx_task_xid(p), VS_ADMIN|VS_IDENT) &&
++ p->pid > 1 && !same_thread_group(p, current) &&
++ !vx_current_initpid(p->pid)) {
+ int err = group_send_sig_info(sig, info, p);
+ ++count;
+ if (err != -EPERM)
+@@ -1842,6 +1866,11 @@ relock:
+ if (is_global_init(current))
+ continue;
+
++ /* virtual init is protected against user signals */
++ if ((info->si_code == SI_USER) &&
++ vx_current_initpid(current->pid))
++ continue;
++
+ if (sig_kernel_stop(signr)) {
+ /*
+ * The default action is to stop all threads in
+--- linux-2.6.25.4/kernel/softirq.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/softirq.c 2008-04-19 15:14:52.000000000 -0400
+@@ -21,6 +21,7 @@
+ #include <linux/rcupdate.h>
+ #include <linux/smp.h>
+ #include <linux/tick.h>
++#include <linux/vs_context.h>
+
+ #include <asm/irq.h>
+ /*
+--- linux-2.6.25.4/kernel/sys.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/sys.c 2008-04-19 15:14:52.000000000 -0400
+@@ -38,6 +38,7 @@
+ #include <linux/syscalls.h>
+ #include <linux/kprobes.h>
+ #include <linux/user_namespace.h>
++#include <linux/vs_pid.h>
+
+ #include <asm/uaccess.h>
+ #include <asm/io.h>
+@@ -116,7 +117,10 @@ static int set_one_prio(struct task_stru
+ goto out;
+ }
+ if (niceval < task_nice(p) && !can_nice(p, niceval)) {
+- error = -EACCES;
++ if (vx_flags(VXF_IGNEG_NICE, 0))
++ error = 0;
++ else
++ error = -EACCES;
+ goto out;
+ }
+ no_nice = security_task_setnice(p, niceval);
+@@ -164,6 +168,8 @@ asmlinkage long sys_setpriority(int whic
+ else
+ pgrp = task_pgrp(current);
+ do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
++ if (!vx_check(p->xid, VS_ADMIN_P | VS_IDENT))
++ continue;
+ error = set_one_prio(p, niceval, error);
+ } while_each_pid_task(pgrp, PIDTYPE_PGID, p);
+ break;
+@@ -224,6 +230,8 @@ asmlinkage long sys_getpriority(int whic
+ else
+ pgrp = task_pgrp(current);
+ do_each_pid_task(pgrp, PIDTYPE_PGID, p) {
++ if (!vx_check(p->xid, VS_ADMIN_P | VS_IDENT))
++ continue;
+ niceval = 20 - task_nice(p);
+ if (niceval > retval)
+ retval = niceval;
+@@ -353,6 +361,9 @@ void kernel_power_off(void)
+ machine_power_off();
+ }
+ EXPORT_SYMBOL_GPL(kernel_power_off);
+
-+config VSERVER_NGNET
-+ bool
-+ depends on EXPERIMENTAL && !VSERVER_LEGACYNET
-+ default y
++long vs_reboot(unsigned int, void __user *);
+
---- linux-2.6.22/kernel/vserver/Makefile 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/Makefile 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,17 @@
-+#
-+# Makefile for the Linux vserver routines.
-+#
+ /*
+ * Reboot system call: for obvious reasons only root may call it,
+ * and even root needs to set up some magic numbers in the registers
+@@ -383,6 +394,9 @@ asmlinkage long sys_reboot(int magic1, i
+ if ((cmd == LINUX_REBOOT_CMD_POWER_OFF) && !pm_power_off)
+ cmd = LINUX_REBOOT_CMD_HALT;
+
++ if (!vx_check(0, VS_ADMIN|VS_WATCH))
++ return vs_reboot(cmd, arg);
+
+ lock_kernel();
+ switch (cmd) {
+ case LINUX_REBOOT_CMD_RESTART:
+@@ -1343,7 +1357,7 @@ asmlinkage long sys_sethostname(char __u
+ int errno;
+ char tmp[__NEW_UTS_LEN];
+
+- if (!capable(CAP_SYS_ADMIN))
++ if (!vx_capable(CAP_SYS_ADMIN, VXC_SET_UTSNAME))
+ return -EPERM;
+ if (len < 0 || len > __NEW_UTS_LEN)
+ return -EINVAL;
+@@ -1388,7 +1402,7 @@ asmlinkage long sys_setdomainname(char _
+ int errno;
+ char tmp[__NEW_UTS_LEN];
+
+- if (!capable(CAP_SYS_ADMIN))
++ if (!vx_capable(CAP_SYS_ADMIN, VXC_SET_UTSNAME))
+ return -EPERM;
+ if (len < 0 || len > __NEW_UTS_LEN)
+ return -EINVAL;
+@@ -1455,7 +1469,7 @@ asmlinkage long sys_setrlimit(unsigned i
+ return -EINVAL;
+ old_rlim = current->signal->rlim + resource;
+ if ((new_rlim.rlim_max > old_rlim->rlim_max) &&
+- !capable(CAP_SYS_RESOURCE))
++ !vx_capable(CAP_SYS_RESOURCE, VXC_SET_RLIMIT))
+ return -EPERM;
+ if (resource == RLIMIT_NOFILE && new_rlim.rlim_max > sysctl_nr_open)
+ return -EPERM;
+--- linux-2.6.25.4/kernel/sysctl.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/sysctl.c 2008-04-19 15:14:52.000000000 -0400
+@@ -107,6 +107,7 @@ static int ngroups_max = NGROUPS_MAX;
+ #ifdef CONFIG_KMOD
+ extern char modprobe_path[];
+ #endif
++extern char vshelper_path[];
+ #ifdef CONFIG_CHR_DEV_SG
+ extern int sg_big_buff;
+ #endif
+@@ -492,6 +493,15 @@ static struct ctl_table kern_table[] = {
+ .strategy = &sysctl_string,
+ },
+ #endif
++ {
++ .ctl_name = KERN_VSHELPER,
++ .procname = "vshelper",
++ .data = &vshelper_path,
++ .maxlen = 256,
++ .mode = 0644,
++ .proc_handler = &proc_dostring,
++ .strategy = &sysctl_string,
++ },
+ #ifdef CONFIG_CHR_DEV_SG
+ {
+ .ctl_name = KERN_SG_BIG_BUFF,
+--- linux-2.6.25.4/kernel/sysctl_check.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/sysctl_check.c 2008-04-21 13:42:56.000000000 -0400
+@@ -39,6 +39,7 @@ static const struct trans_ctl_table tran
+
+ { KERN_PANIC, "panic" },
+ { KERN_REALROOTDEV, "real-root-dev" },
++ { KERN_VSHELPER, "vshelper", },
+
+ { KERN_SPARC_REBOOT, "reboot-cmd" },
+ { KERN_CTLALTDEL, "ctrl-alt-del" },
+@@ -1217,6 +1218,22 @@ static const struct trans_ctl_table tran
+ {}
+ };
+
++static struct trans_ctl_table trans_vserver_table[] = {
++ { 1, "debug_switch" },
++ { 2, "debug_xid" },
++ { 3, "debug_nid" },
++ { 4, "debug_tag" },
++ { 5, "debug_net" },
++ { 6, "debug_limit" },
++ { 7, "debug_cres" },
++ { 8, "debug_dlim" },
++ { 9, "debug_quota" },
++ { 10, "debug_cvirt" },
++ { 11, "debug_space" },
++ { 12, "debug_misc" },
++ {}
++};
+
-+obj-y += vserver.o
+ static const struct trans_ctl_table trans_root_table[] = {
+ { CTL_KERN, "kernel", trans_kern_table },
+ { CTL_VM, "vm", trans_vm_table },
+@@ -1233,6 +1250,7 @@ static const struct trans_ctl_table tran
+ { CTL_SUNRPC, "sunrpc", trans_sunrpc_table },
+ { CTL_PM, "pm", trans_pm_table },
+ { CTL_FRV, "frv", trans_frv_table },
++ { CTL_VSERVER, "vserver", trans_vserver_table },
+ {}
+ };
+
+--- linux-2.6.25.4/kernel/time.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/time.c 2008-04-19 15:14:52.000000000 -0400
+@@ -60,6 +60,7 @@ EXPORT_SYMBOL(sys_tz);
+ asmlinkage long sys_time(time_t __user * tloc)
+ {
+ time_t i = get_seconds();
++/* FIXME: do_gettimeofday(&tv) -> vx_gettimeofday(&tv) */
+
+ if (tloc) {
+ if (put_user(i,tloc))
+@@ -89,7 +90,7 @@ asmlinkage long sys_stime(time_t __user
+ if (err)
+ return err;
+
+- do_settimeofday(&tv);
++ vx_settimeofday(&tv);
+ return 0;
+ }
+
+@@ -100,7 +101,7 @@ asmlinkage long sys_gettimeofday(struct
+ {
+ if (likely(tv != NULL)) {
+ struct timeval ktv;
+- do_gettimeofday(&ktv);
++ vx_gettimeofday(&ktv);
+ if (copy_to_user(tv, &ktv, sizeof(ktv)))
+ return -EFAULT;
+ }
+@@ -175,7 +176,7 @@ int do_sys_settimeofday(struct timespec
+ /* SMP safe, again the code in arch/foo/time.c should
+ * globally block out interrupts when it runs.
+ */
+- return do_settimeofday(tv);
++ return vx_settimeofday(tv);
+ }
+ return 0;
+ }
+@@ -307,7 +308,7 @@ void getnstimeofday(struct timespec *tv)
+ {
+ struct timeval x;
+
+- do_gettimeofday(&x);
++ vx_gettimeofday(&x);
+ tv->tv_sec = x.tv_sec;
+ tv->tv_nsec = x.tv_usec * NSEC_PER_USEC;
+ }
+--- linux-2.6.25.4/kernel/timer.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/timer.c 2008-04-19 15:14:52.000000000 -0400
+@@ -37,6 +37,10 @@
+ #include <linux/delay.h>
+ #include <linux/tick.h>
+ #include <linux/kallsyms.h>
++#include <linux/vs_base.h>
++#include <linux/vs_cvirt.h>
++#include <linux/vs_pid.h>
++#include <linux/vserver/sched.h>
+
+ #include <asm/uaccess.h>
+ #include <asm/unistd.h>
+@@ -955,12 +959,6 @@ asmlinkage unsigned long sys_alarm(unsig
+
+ #endif
+
+-#ifndef __alpha__
+-
+-/*
+- * The Alpha uses getxpid, getxuid, and getxgid instead. Maybe this
+- * should be moved into arch/i386 instead?
+- */
+
+ /**
+ * sys_getpid - return the thread group id of the current process
+@@ -989,10 +987,23 @@ asmlinkage long sys_getppid(void)
+ rcu_read_lock();
+ pid = task_tgid_vnr(current->real_parent);
+ rcu_read_unlock();
++ return vx_map_pid(pid);
++}
+
+- return pid;
++#ifdef __alpha__
+
-+vserver-y := switch.o context.o space.o sched.o network.o inode.o \
-+ limit.o cvirt.o cacct.o signal.o helper.o init.o dlimit.o
++/*
++ * The Alpha uses getxpid, getxuid, and getxgid instead.
++ */
+
-+vserver-$(CONFIG_PROC_FS) += proc.o
-+vserver-$(CONFIG_VSERVER_DEBUG) += sysctl.o debug.o
-+vserver-$(CONFIG_VSERVER_LEGACY) += legacy.o
-+vserver-$(CONFIG_VSERVER_LEGACYNET) += legacynet.o
-+vserver-$(CONFIG_VSERVER_HISTORY) += history.o
-+vserver-$(CONFIG_VSERVER_MONITOR) += monitor.o
++asmlinkage long do_getxpid(long *ppid)
++{
++ *ppid = sys_getppid();
++ return sys_getpid();
+ }
+
++#else /* _alpha_ */
+
---- linux-2.6.22/kernel/vserver/cacct.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/cacct.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,44 @@
+ asmlinkage long sys_getuid(void)
+ {
+ /* Only we change this so SMP safe */
+@@ -1160,6 +1171,8 @@ int do_sysinfo(struct sysinfo *info)
+ tp.tv_nsec = tp.tv_nsec - NSEC_PER_SEC;
+ tp.tv_sec++;
+ }
++ if (vx_flags(VXF_VIRT_UPTIME, 0))
++ vx_vsi_uptime(&tp, NULL);
+ info->uptime = tp.tv_sec + (tp.tv_nsec ? 1 : 0);
+
+ info->loads[0] = avenrun[0] << (SI_LOAD_SHIFT - FSHIFT);
+--- linux-2.6.25.4/kernel/user.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/user.c 2008-04-23 16:24:56.000000000 -0400
+@@ -219,14 +219,15 @@ static struct kobj_type uids_ktype = {
+ };
+
+ /* create /sys/kernel/uids/<uid>/cpu_share file for this user */
+-static int uids_user_create(struct user_struct *up)
++static int uids_user_create(struct user_namespace *ns, struct user_struct *up)
+ {
+ struct kobject *kobj = &up->kobj;
+ int error;
+
+ memset(kobj, 0, sizeof(struct kobject));
+ kobj->kset = uids_kset;
+- error = kobject_init_and_add(kobj, &uids_ktype, NULL, "%d", up->uid);
++ error = kobject_init_and_add(kobj, &uids_ktype, NULL,
++ "%p:%d", ns, up->uid);
+ if (error) {
+ kobject_put(kobj);
+ goto done;
+@@ -248,7 +249,7 @@ int __init uids_sysfs_init(void)
+ if (!uids_kset)
+ return -ENOMEM;
+
+- return uids_user_create(&root_user);
++ return uids_user_create(NULL, &root_user);
+ }
+
+ /* work function to remove sysfs directory for a user and free up
+@@ -308,7 +309,8 @@ static inline void free_user(struct user
+ #else /* CONFIG_USER_SCHED && CONFIG_SYSFS */
+
+ int uids_sysfs_init(void) { return 0; }
+-static inline int uids_user_create(struct user_struct *up) { return 0; }
++static inline int uids_user_create(struct user_namespace *ns,
++ struct user_struct *up) { return 0; }
+ static inline void uids_mutex_lock(void) { }
+ static inline void uids_mutex_unlock(void) { }
+
+@@ -399,7 +401,7 @@ struct user_struct * alloc_uid(struct us
+ if (sched_create_user(new) < 0)
+ goto out_put_keys;
+
+- if (uids_user_create(new))
++ if (uids_user_create(ns, new))
+ goto out_destoy_sched;
+
+ /*
+--- linux-2.6.25.4/kernel/user_namespace.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/user_namespace.c 2008-04-19 15:14:52.000000000 -0400
+@@ -9,6 +9,7 @@
+ #include <linux/version.h>
+ #include <linux/nsproxy.h>
+ #include <linux/user_namespace.h>
++#include <linux/vserver/global.h>
+
+ /*
+ * Clone a new ns copying an original user ns, setting refcount to 1
+@@ -26,6 +27,7 @@ static struct user_namespace *clone_user
+ return ERR_PTR(-ENOMEM);
+
+ kref_init(&ns->kref);
++ atomic_inc(&vs_global_user_ns);
+
+ for (n = 0; n < UIDHASH_SZ; ++n)
+ INIT_HLIST_HEAD(ns->uidhash_table + n);
+@@ -71,5 +73,6 @@ void free_user_ns(struct kref *kref)
+
+ ns = container_of(kref, struct user_namespace, kref);
+ release_uids(ns);
++ atomic_dec(&vs_global_user_ns);
+ kfree(ns);
+ }
+--- linux-2.6.25.4/kernel/utsname.c 2008-04-17 10:37:25.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/utsname.c 2008-04-19 15:14:52.000000000 -0400
+@@ -14,6 +14,7 @@
+ #include <linux/utsname.h>
+ #include <linux/version.h>
+ #include <linux/err.h>
++#include <linux/vserver/global.h>
+
+ /*
+ * Clone a new ns copying an original utsname, setting refcount to 1
+@@ -32,6 +33,7 @@ static struct uts_namespace *clone_uts_n
+ memcpy(&ns->name, &old_ns->name, sizeof(ns->name));
+ up_read(&uts_sem);
+ kref_init(&ns->kref);
++ atomic_inc(&vs_global_uts_ns);
+ return ns;
+ }
+
+@@ -62,5 +64,6 @@ void free_uts_ns(struct kref *kref)
+ struct uts_namespace *ns;
+
+ ns = container_of(kref, struct uts_namespace, kref);
++ atomic_dec(&vs_global_uts_ns);
+ kfree(ns);
+ }
+--- linux-2.6.25.4/kernel/vserver/cacct.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/cacct.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,42 @@
+/*
+ * linux/kernel/vserver/cacct.c
+ *
@@ -18084,9 +17526,7 @@
+ */
+
+#include <linux/types.h>
-+#include <linux/sched.h>
+#include <linux/vs_context.h>
-+#include <linux/vserver/switch.h>
+#include <linux/vserver/cacct_cmd.h>
+#include <linux/vserver/cacct_int.h>
+
@@ -18116,8 +17556,8 @@
+ return 0;
+}
+
---- linux-2.6.22/kernel/vserver/cacct_init.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/cacct_init.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/kernel/vserver/cacct_init.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/cacct_init.h 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,25 @@
+
+
@@ -18144,8 +17584,8 @@
+ return;
+}
+
---- linux-2.6.22/kernel/vserver/cacct_proc.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/cacct_proc.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/kernel/vserver/cacct_proc.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/cacct_proc.h 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,53 @@
+#ifndef _VX_CACCT_PROC_H
+#define _VX_CACCT_PROC_H
@@ -18200,9 +17640,9 @@
+}
+
+#endif /* _VX_CACCT_PROC_H */
---- linux-2.6.22/kernel/vserver/context.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/context.c 2007-06-16 06:44:23 +0200
-@@ -0,0 +1,1106 @@
+--- linux-2.6.25.4/kernel/vserver/context.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/context.c 2008-04-23 22:26:24.000000000 -0400
+@@ -0,0 +1,1005 @@
+/*
+ * linux/kernel/vserver/context.c
+ *
@@ -18226,18 +17666,17 @@
+ * V0.14 changed vcmds to vxi arg
+ * V0.15 added context stat
+ * V0.16 have __create claim() the vxi
++ * V0.17 removed older and legacy stuff
+ *
+ */
+
+#include <linux/slab.h>
+#include <linux/types.h>
-+#include <linux/mnt_namespace.h>
++#include <linux/security.h>
+#include <linux/pid_namespace.h>
+
-+#include <linux/sched.h>
+#include <linux/vserver/context.h>
+#include <linux/vserver/network.h>
-+#include <linux/vserver/legacy.h>
+#include <linux/vserver/debug.h>
+#include <linux/vserver/limit.h>
+#include <linux/vserver/limit_int.h>
@@ -18245,11 +17684,9 @@
+
+#include <linux/vs_context.h>
+#include <linux/vs_limit.h>
++#include <linux/vs_pid.h>
+#include <linux/vserver/context_cmd.h>
+
-+#include <linux/err.h>
-+#include <asm/errno.h>
-+
+#include "cvirt_init.h"
+#include "cacct_init.h"
+#include "limit_init.h"
@@ -18301,6 +17738,7 @@
+ /* prepare reaper */
+ get_task_struct(init_pid_ns.child_reaper);
+ new->vx_reaper = init_pid_ns.child_reaper;
++ new->vx_badness_bias = 0;
+
+ /* rest of init goes here */
+ vx_info_init_limit(&new->limit);
@@ -18317,9 +17755,9 @@
+ }
+
+ new->vx_flags = VXF_INIT_SET;
-+ new->vx_bcaps = CAP_INIT_EFF_SET;
++ cap_set_init_eff(new->vx_bcaps);
+ new->vx_ccaps = 0;
-+ new->vx_cap_bset = cap_bset;
++ // new->vx_cap_bset = current->cap_bset;
+
+ new->reboot_cmd = 0;
+ new->exit_code = 0;
@@ -18396,14 +17834,16 @@
+{
+ unsigned long flags;
+
-+ /* context shutdown is mandatory */
-+ BUG_ON(!vx_info_state(vxi, VXS_SHUTDOWN));
-+
++ /* check for reference counts first */
+ BUG_ON(atomic_read(&vxi->vx_usecnt));
+ BUG_ON(atomic_read(&vxi->vx_tasks));
+
++ /* context must not be hashed */
+ BUG_ON(vx_info_state(vxi, VXS_HASHED));
+
++ /* context shutdown is mandatory */
++ BUG_ON(!vx_info_state(vxi, VXS_SHUTDOWN));
++
+ BUG_ON(vxi->vx_nsproxy);
+ BUG_ON(vxi->vx_fs);
+
@@ -18512,1359 +17952,1700 @@
+}
+
+
-+/* __vx_dynamic_id()
-+
-+ * find unused dynamic xid
-+ * requires the hash_lock to be held */
-+
-+static inline xid_t __vx_dynamic_id(void)
-+{
-+ static xid_t seq = MAX_S_CONTEXT;
-+ xid_t barrier = seq;
-+
-+ vxd_assert_lock(&vx_info_hash_lock);
-+ do {
-+ if (++seq > MAX_S_CONTEXT)
-+ seq = MIN_D_CONTEXT;
-+ if (!__lookup_vx_info(seq)) {
-+ vxdprintk(VXD_CBIT(xid, 4),
-+ "__vx_dynamic_id: [#%d]", seq);
-+ return seq;
-+ }
-+ } while (barrier != seq);
-+ return 0;
-+}
-+
-+#ifdef CONFIG_VSERVER_LEGACY
-+
-+/* __loc_vx_info()
++/* __create_vx_info()
+
-+ * locate or create the requested context
-+ * get() it and if new hash it */
++ * create the requested context
++ * get(), claim() and hash it */
+
-+static struct vx_info *__loc_vx_info(int id, int *err)
++static struct vx_info *__create_vx_info(int id)
+{
+ struct vx_info *new, *vxi = NULL;
+
-+ vxdprintk(VXD_CBIT(xid, 1), "loc_vx_info(%d)*", id);
++ vxdprintk(VXD_CBIT(xid, 1), "create_vx_info(%d)*", id);
+
-+ if (!(new = __alloc_vx_info(id))) {
-+ *err = -ENOMEM;
-+ return NULL;
-+ }
++ if (!(new = __alloc_vx_info(id)))
++ return ERR_PTR(-ENOMEM);
+
+ /* required to make dynamic xids unique */
+ spin_lock(&vx_info_hash_lock);
+
-+ /* dynamic context requested */
-+ if (id == VX_DYNAMIC_ID) {
-+#ifdef CONFIG_VSERVER_DYNAMIC_IDS
-+ id = __vx_dynamic_id();
-+ if (!id) {
-+ printk(KERN_ERR "no dynamic context available.\n");
-+ goto out_unlock;
-+ }
-+ new->vx_id = id;
-+#else
-+ printk(KERN_ERR "dynamic contexts disabled.\n");
-+ goto out_unlock;
-+#endif
-+ }
-+ /* existing context requested */
-+ else if ((vxi = __lookup_vx_info(id))) {
-+ /* context in setup is not available */
-+ if (vxi->vx_flags & VXF_STATE_SETUP) {
-+ vxdprintk(VXD_CBIT(xid, 0),
-+ "loc_vx_info(%d) = %p (not available)", id, vxi);
-+ vxi = NULL;
-+ *err = -EBUSY;
-+ } else {
-+ vxdprintk(VXD_CBIT(xid, 0),
-+ "loc_vx_info(%d) = %p (found)", id, vxi);
-+ get_vx_info(vxi);
-+ *err = 0;
-+ }
++ /* static context requested */
++ if ((vxi = __lookup_vx_info(id))) {
++ vxdprintk(VXD_CBIT(xid, 0),
++ "create_vx_info(%d) = %p (already there)", id, vxi);
++ if (vx_info_flags(vxi, VXF_STATE_SETUP, 0))
++ vxi = ERR_PTR(-EBUSY);
++ else
++ vxi = ERR_PTR(-EEXIST);
+ goto out_unlock;
+ }
-+
-+ /* new context requested */
++ /* new context */
+ vxdprintk(VXD_CBIT(xid, 0),
-+ "loc_vx_info(%d) = %p (new)", id, new);
++ "create_vx_info(%d) = %p (new)", id, new);
++ claim_vx_info(new, NULL);
+ __hash_vx_info(get_vx_info(new));
+ vxi = new, new = NULL;
-+ *err = 1;
+
+out_unlock:
+ spin_unlock(&vx_info_hash_lock);
-+ vxh_loc_vx_info(vxi, id);
++ vxh_create_vx_info(IS_ERR(vxi) ? NULL : vxi, id);
+ if (new)
+ __dealloc_vx_info(new);
+ return vxi;
+}
+
++
++/* exported stuff */
++
++
++void unhash_vx_info(struct vx_info *vxi)
++{
++ __shutdown_vx_info(vxi);
++ spin_lock(&vx_info_hash_lock);
++ __unhash_vx_info(vxi);
++ spin_unlock(&vx_info_hash_lock);
++ __wakeup_vx_info(vxi);
++}
++
++
++/* lookup_vx_info()
++
++ * search for a vx_info and get() it
++ * negative id means current */
++
++struct vx_info *lookup_vx_info(int id)
++{
++ struct vx_info *vxi = NULL;
++
++ if (id < 0) {
++ vxi = get_vx_info(current->vx_info);
++ } else if (id > 1) {
++ spin_lock(&vx_info_hash_lock);
++ vxi = get_vx_info(__lookup_vx_info(id));
++ spin_unlock(&vx_info_hash_lock);
++ }
++ return vxi;
++}
++
++/* xid_is_hashed()
++
++ * verify that xid is still hashed */
++
++int xid_is_hashed(xid_t xid)
++{
++ int hashed;
++
++ spin_lock(&vx_info_hash_lock);
++ hashed = (__lookup_vx_info(xid) != NULL);
++ spin_unlock(&vx_info_hash_lock);
++ return hashed;
++}
++
++#ifdef CONFIG_PROC_FS
++
++/* get_xid_list()
++
++ * get a subset of hashed xids for proc
++ * assumes size is at least one */
++
++int get_xid_list(int index, unsigned int *xids, int size)
++{
++ int hindex, nr_xids = 0;
++
++ /* only show current and children */
++ if (!vx_check(0, VS_ADMIN | VS_WATCH)) {
++ if (index > 0)
++ return 0;
++ xids[nr_xids] = vx_current_xid();
++ return 1;
++ }
++
++ for (hindex = 0; hindex < VX_HASH_SIZE; hindex++) {
++ struct hlist_head *head = &vx_info_hash[hindex];
++ struct hlist_node *pos;
++
++ spin_lock(&vx_info_hash_lock);
++ hlist_for_each(pos, head) {
++ struct vx_info *vxi;
++
++ if (--index > 0)
++ continue;
++
++ vxi = hlist_entry(pos, struct vx_info, vx_hlist);
++ xids[nr_xids] = vxi->vx_id;
++ if (++nr_xids >= size) {
++ spin_unlock(&vx_info_hash_lock);
++ goto out;
++ }
++ }
++ /* keep the lock time short */
++ spin_unlock(&vx_info_hash_lock);
++ }
++out:
++ return nr_xids;
++}
++#endif
++
++#ifdef CONFIG_VSERVER_DEBUG
++
++void dump_vx_info_inactive(int level)
++{
++ struct hlist_node *entry, *next;
++
++ hlist_for_each_safe(entry, next, &vx_info_inactive) {
++ struct vx_info *vxi =
++ list_entry(entry, struct vx_info, vx_hlist);
++
++ dump_vx_info(vxi, level);
++ }
++}
++
+#endif
+
-+/* __create_vx_info()
++#if 0
++int vx_migrate_user(struct task_struct *p, struct vx_info *vxi)
++{
++ struct user_struct *new_user, *old_user;
++
++ if (!p || !vxi)
++ BUG();
++
++ if (vx_info_flags(vxi, VXF_INFO_PRIVATE, 0))
++ return -EACCES;
++
++ new_user = alloc_uid(vxi->vx_id, p->uid);
++ if (!new_user)
++ return -ENOMEM;
++
++ old_user = p->user;
++ if (new_user != old_user) {
++ atomic_inc(&new_user->processes);
++ atomic_dec(&old_user->processes);
++ p->user = new_user;
++ }
++ free_uid(old_user);
++ return 0;
++}
++#endif
++
++#if 0
++void vx_mask_cap_bset(struct vx_info *vxi, struct task_struct *p)
++{
++ // p->cap_effective &= vxi->vx_cap_bset;
++ p->cap_effective =
++ cap_intersect(p->cap_effective, vxi->cap_bset);
++ // p->cap_inheritable &= vxi->vx_cap_bset;
++ p->cap_inheritable =
++ cap_intersect(p->cap_inheritable, vxi->cap_bset);
++ // p->cap_permitted &= vxi->vx_cap_bset;
++ p->cap_permitted =
++ cap_intersect(p->cap_permitted, vxi->cap_bset);
++}
++#endif
++
++
++#include <linux/file.h>
++
++static int vx_openfd_task(struct task_struct *tsk)
++{
++ struct files_struct *files = tsk->files;
++ struct fdtable *fdt;
++ const unsigned long *bptr;
++ int count, total;
++
++ /* no rcu_read_lock() because of spin_lock() */
++ spin_lock(&files->file_lock);
++ fdt = files_fdtable(files);
++ bptr = fdt->open_fds->fds_bits;
++ count = fdt->max_fds / (sizeof(unsigned long) * 8);
++ for (total = 0; count > 0; count--) {
++ if (*bptr)
++ total += hweight_long(*bptr);
++ bptr++;
++ }
++ spin_unlock(&files->file_lock);
++ return total;
++}
++
++
++/* for *space compatibility */
++
++asmlinkage long sys_unshare(unsigned long);
++
++/*
++ * migrate task to new context
++ * gets vxi, puts old_vxi on change
++ * optionally unshares namespaces (hack)
++ */
++
++int vx_migrate_task(struct task_struct *p, struct vx_info *vxi, int unshare)
++{
++ struct vx_info *old_vxi;
++ int ret = 0;
++
++ if (!p || !vxi)
++ BUG();
++
++ vxdprintk(VXD_CBIT(xid, 5),
++ "vx_migrate_task(%p,%p[#%d.%d])", p, vxi,
++ vxi->vx_id, atomic_read(&vxi->vx_usecnt));
++
++ if (vx_info_flags(vxi, VXF_INFO_PRIVATE, 0) &&
++ !vx_info_flags(vxi, VXF_STATE_SETUP, 0))
++ return -EACCES;
++
++ if (vx_info_state(vxi, VXS_SHUTDOWN))
++ return -EFAULT;
++
++ old_vxi = task_get_vx_info(p);
++ if (old_vxi == vxi)
++ goto out;
++
++// if (!(ret = vx_migrate_user(p, vxi))) {
++ {
++ int openfd;
++
++ task_lock(p);
++ openfd = vx_openfd_task(p);
++
++ if (old_vxi) {
++ atomic_dec(&old_vxi->cvirt.nr_threads);
++ atomic_dec(&old_vxi->cvirt.nr_running);
++ __rlim_dec(&old_vxi->limit, RLIMIT_NPROC);
++ /* FIXME: what about the struct files here? */
++ __rlim_sub(&old_vxi->limit, VLIMIT_OPENFD, openfd);
++ /* account for the executable */
++ __rlim_dec(&old_vxi->limit, VLIMIT_DENTRY);
++ }
++ atomic_inc(&vxi->cvirt.nr_threads);
++ atomic_inc(&vxi->cvirt.nr_running);
++ __rlim_inc(&vxi->limit, RLIMIT_NPROC);
++ /* FIXME: what about the struct files here? */
++ __rlim_add(&vxi->limit, VLIMIT_OPENFD, openfd);
++ /* account for the executable */
++ __rlim_inc(&vxi->limit, VLIMIT_DENTRY);
+
-+ * create the requested context
-+ * get(), claim() and hash it */
++ if (old_vxi) {
++ release_vx_info(old_vxi, p);
++ clr_vx_info(&p->vx_info);
++ }
++ claim_vx_info(vxi, p);
++ set_vx_info(&p->vx_info, vxi);
++ p->xid = vxi->vx_id;
+
-+static struct vx_info *__create_vx_info(int id)
-+{
-+ struct vx_info *new, *vxi = NULL;
++ vxdprintk(VXD_CBIT(xid, 5),
++ "moved task %p into vxi:%p[#%d]",
++ p, vxi, vxi->vx_id);
+
-+ vxdprintk(VXD_CBIT(xid, 1), "create_vx_info(%d)*", id);
++ // vx_mask_cap_bset(vxi, p);
++ task_unlock(p);
+
-+ if (!(new = __alloc_vx_info(id)))
-+ return ERR_PTR(-ENOMEM);
++ /* hack for *spaces to provide compatibility */
++ if (unshare) {
++ struct nsproxy *old_nsp, *new_nsp;
+
-+ /* required to make dynamic xids unique */
-+ spin_lock(&vx_info_hash_lock);
++ ret = unshare_nsproxy_namespaces(
++ CLONE_NEWUTS | CLONE_NEWIPC | CLONE_NEWUSER,
++ &new_nsp, NULL);
++ if (ret)
++ goto out;
+
-+ /* dynamic context requested */
-+ if (id == VX_DYNAMIC_ID) {
-+#ifdef CONFIG_VSERVER_DYNAMIC_IDS
-+ id = __vx_dynamic_id();
-+ if (!id) {
-+ printk(KERN_ERR "no dynamic context available.\n");
-+ vxi = ERR_PTR(-EAGAIN);
-+ goto out_unlock;
++ old_nsp = xchg(&p->nsproxy, new_nsp);
++ vx_set_space(vxi, CLONE_NEWUTS | CLONE_NEWIPC | CLONE_NEWUSER);
++ put_nsproxy(old_nsp);
+ }
-+ new->vx_id = id;
-+#else
-+ printk(KERN_ERR "dynamic contexts disabled.\n");
-+ vxi = ERR_PTR(-EINVAL);
-+ goto out_unlock;
-+#endif
-+ }
-+ /* static context requested */
-+ else if ((vxi = __lookup_vx_info(id))) {
-+ vxdprintk(VXD_CBIT(xid, 0),
-+ "create_vx_info(%d) = %p (already there)", id, vxi);
-+ if (vx_info_flags(vxi, VXF_STATE_SETUP, 0))
-+ vxi = ERR_PTR(-EBUSY);
-+ else
-+ vxi = ERR_PTR(-EEXIST);
-+ goto out_unlock;
-+ }
-+#ifdef CONFIG_VSERVER_DYNAMIC_IDS
-+ /* dynamic xid creation blocker */
-+ else if (id >= MIN_D_CONTEXT) {
-+ vxdprintk(VXD_CBIT(xid, 0),
-+ "create_vx_info(%d) (dynamic rejected)", id);
-+ vxi = ERR_PTR(-EINVAL);
-+ goto out_unlock;
+ }
-+#endif
-+
-+ /* new context */
-+ vxdprintk(VXD_CBIT(xid, 0),
-+ "create_vx_info(%d) = %p (new)", id, new);
-+ claim_vx_info(new, NULL);
-+ __hash_vx_info(get_vx_info(new));
-+ vxi = new, new = NULL;
-+
-+out_unlock:
-+ spin_unlock(&vx_info_hash_lock);
-+ vxh_create_vx_info(IS_ERR(vxi) ? NULL : vxi, id);
-+ if (new)
-+ __dealloc_vx_info(new);
-+ return vxi;
++out:
++ put_vx_info(old_vxi);
++ return ret;
+}
+
++int vx_set_reaper(struct vx_info *vxi, struct task_struct *p)
++{
++ struct task_struct *old_reaper;
+
-+/* exported stuff */
++ if (!vxi)
++ return -EINVAL;
+
++ vxdprintk(VXD_CBIT(xid, 6),
++ "vx_set_reaper(%p[#%d],%p[#%d,%d])",
++ vxi, vxi->vx_id, p, p->xid, p->pid);
+
-+void unhash_vx_info(struct vx_info *vxi)
-+{
-+ __shutdown_vx_info(vxi);
-+ spin_lock(&vx_info_hash_lock);
-+ __unhash_vx_info(vxi);
-+ spin_unlock(&vx_info_hash_lock);
-+ __wakeup_vx_info(vxi);
++ old_reaper = vxi->vx_reaper;
++ if (old_reaper == p)
++ return 0;
++
++ /* set new child reaper */
++ get_task_struct(p);
++ vxi->vx_reaper = p;
++ put_task_struct(old_reaper);
++ return 0;
+}
+
++int vx_set_init(struct vx_info *vxi, struct task_struct *p)
++{
++ if (!vxi)
++ return -EINVAL;
+
-+/* lookup_vx_info()
++ vxdprintk(VXD_CBIT(xid, 6),
++ "vx_set_init(%p[#%d],%p[#%d,%d,%d])",
++ vxi, vxi->vx_id, p, p->xid, p->pid, p->tgid);
+
-+ * search for a vx_info and get() it
-+ * negative id means current */
++ vxi->vx_flags &= ~VXF_STATE_INIT;
++ vxi->vx_initpid = p->tgid;
++ return 0;
++}
+
-+struct vx_info *lookup_vx_info(int id)
++void vx_exit_init(struct vx_info *vxi, struct task_struct *p, int code)
+{
-+ struct vx_info *vxi = NULL;
++ vxdprintk(VXD_CBIT(xid, 6),
++ "vx_exit_init(%p[#%d],%p[#%d,%d,%d])",
++ vxi, vxi->vx_id, p, p->xid, p->pid, p->tgid);
+
-+ if (id < 0) {
-+ vxi = get_vx_info(current->vx_info);
-+ } else if (id > 1) {
-+ spin_lock(&vx_info_hash_lock);
-+ vxi = get_vx_info(__lookup_vx_info(id));
-+ spin_unlock(&vx_info_hash_lock);
-+ }
-+ return vxi;
++ vxi->exit_code = code;
++ vxi->vx_initpid = 0;
+}
+
-+/* xid_is_hashed()
-+
-+ * verify that xid is still hashed */
+
-+int xid_is_hashed(xid_t xid)
++void vx_set_persistent(struct vx_info *vxi)
+{
-+ int hashed;
++ vxdprintk(VXD_CBIT(xid, 6),
++ "vx_set_persistent(%p[#%d])", vxi, vxi->vx_id);
+
-+ spin_lock(&vx_info_hash_lock);
-+ hashed = (__lookup_vx_info(xid) != NULL);
-+ spin_unlock(&vx_info_hash_lock);
-+ return hashed;
++ get_vx_info(vxi);
++ claim_vx_info(vxi, NULL);
+}
+
-+#ifdef CONFIG_VSERVER_LEGACY
-+
-+struct vx_info *lookup_or_create_vx_info(int id)
++void vx_clear_persistent(struct vx_info *vxi)
+{
-+ int err;
++ vxdprintk(VXD_CBIT(xid, 6),
++ "vx_clear_persistent(%p[#%d])", vxi, vxi->vx_id);
+
-+ return __loc_vx_info(id, &err);
++ release_vx_info(vxi, NULL);
++ put_vx_info(vxi);
+}
+
-+#endif
++void vx_update_persistent(struct vx_info *vxi)
++{
++ if (vx_info_flags(vxi, VXF_PERSISTENT, 0))
++ vx_set_persistent(vxi);
++ else
++ vx_clear_persistent(vxi);
++}
+
-+#ifdef CONFIG_PROC_FS
+
-+/* get_xid_list()
++/* task must be current or locked */
+
-+ * get a subset of hashed xids for proc
-+ * assumes size is at least one */
++void exit_vx_info(struct task_struct *p, int code)
++{
++ struct vx_info *vxi = p->vx_info;
+
-+int get_xid_list(int index, unsigned int *xids, int size)
++ if (vxi) {
++ atomic_dec(&vxi->cvirt.nr_threads);
++ vx_nproc_dec(p);
++
++ vxi->exit_code = code;
++ release_vx_info(vxi, p);
++ }
++}
++
++void exit_vx_info_early(struct task_struct *p, int code)
+{
-+ int hindex, nr_xids = 0;
++ struct vx_info *vxi = p->vx_info;
+
-+ /* only show current and children */
-+ if (!vx_check(0, VS_ADMIN | VS_WATCH)) {
-+ if (index > 0)
-+ return 0;
-+ xids[nr_xids] = vx_current_xid();
-+ return 1;
++ if (vxi) {
++ if (vxi->vx_initpid == p->tgid)
++ vx_exit_init(vxi, p, code);
++ if (vxi->vx_reaper == p)
++ vx_set_reaper(vxi, init_pid_ns.child_reaper);
+ }
++}
+
-+ for (hindex = 0; hindex < VX_HASH_SIZE; hindex++) {
-+ struct hlist_head *head = &vx_info_hash[hindex];
-+ struct hlist_node *pos;
+
-+ spin_lock(&vx_info_hash_lock);
-+ hlist_for_each(pos, head) {
-+ struct vx_info *vxi;
++/* vserver syscall commands below here */
+
-+ if (--index > 0)
-+ continue;
++/* taks xid and vx_info functions */
+
-+ vxi = hlist_entry(pos, struct vx_info, vx_hlist);
-+ xids[nr_xids] = vxi->vx_id;
-+ if (++nr_xids >= size) {
-+ spin_unlock(&vx_info_hash_lock);
-+ goto out;
-+ }
-+ }
-+ /* keep the lock time short */
-+ spin_unlock(&vx_info_hash_lock);
-+ }
-+out:
-+ return nr_xids;
-+}
-+#endif
++#include <asm/uaccess.h>
+
-+#ifdef CONFIG_VSERVER_DEBUG
+
-+void dump_vx_info_inactive(int level)
++int vc_task_xid(uint32_t id)
+{
-+ struct hlist_node *entry, *next;
++ xid_t xid;
+
-+ hlist_for_each_safe(entry, next, &vx_info_inactive) {
-+ struct vx_info *vxi =
-+ list_entry(entry, struct vx_info, vx_hlist);
++ if (id) {
++ struct task_struct *tsk;
+
-+ dump_vx_info(vxi, level);
-+ }
++ read_lock(&tasklist_lock);
++ tsk = find_task_by_real_pid(id);
++ xid = (tsk) ? tsk->xid : -ESRCH;
++ read_unlock(&tasklist_lock);
++ } else
++ xid = vx_current_xid();
++ return xid;
+}
+
-+#endif
+
-+int vx_migrate_user(struct task_struct *p, struct vx_info *vxi)
++int vc_vx_info(struct vx_info *vxi, void __user *data)
+{
-+ struct user_struct *new_user, *old_user;
++ struct vcmd_vx_info_v0 vc_data;
+
-+ if (!p || !vxi)
-+ BUG();
++ vc_data.xid = vxi->vx_id;
++ vc_data.initpid = vxi->vx_initpid;
+
-+ if (vx_info_flags(vxi, VXF_INFO_PRIVATE, 0))
-+ return -EACCES;
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ return -EFAULT;
++ return 0;
++}
+
-+ new_user = alloc_uid(vxi->vx_id, p->uid);
-+ if (!new_user)
-+ return -ENOMEM;
+
-+ old_user = p->user;
-+ if (new_user != old_user) {
-+ atomic_inc(&new_user->processes);
-+ atomic_dec(&old_user->processes);
-+ p->user = new_user;
-+ }
-+ free_uid(old_user);
++int vc_ctx_stat(struct vx_info *vxi, void __user *data)
++{
++ struct vcmd_ctx_stat_v0 vc_data;
++
++ vc_data.usecnt = atomic_read(&vxi->vx_usecnt);
++ vc_data.tasks = atomic_read(&vxi->vx_tasks);
++
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ return -EFAULT;
+ return 0;
+}
+
-+void vx_mask_cap_bset(struct vx_info *vxi, struct task_struct *p)
++
++/* context functions */
++
++int vc_ctx_create(uint32_t xid, void __user *data)
+{
-+ p->cap_effective &= vxi->vx_cap_bset;
-+ p->cap_inheritable &= vxi->vx_cap_bset;
-+ p->cap_permitted &= vxi->vx_cap_bset;
-+}
++ struct vcmd_ctx_create vc_data = { .flagword = VXF_INIT_SET };
++ struct vx_info *new_vxi;
++ int ret;
+
++ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
+
-+#include <linux/file.h>
++ if ((xid > MAX_S_CONTEXT) || (xid < 2))
++ return -EINVAL;
+
-+static int vx_openfd_task(struct task_struct *tsk)
-+{
-+ struct files_struct *files = tsk->files;
-+ struct fdtable *fdt;
-+ const unsigned long *bptr;
-+ int count, total;
++ new_vxi = __create_vx_info(xid);
++ if (IS_ERR(new_vxi))
++ return PTR_ERR(new_vxi);
+
-+ /* no rcu_read_lock() because of spin_lock() */
-+ spin_lock(&files->file_lock);
-+ fdt = files_fdtable(files);
-+ bptr = fdt->open_fds->fds_bits;
-+ count = fdt->max_fds / (sizeof(unsigned long) * 8);
-+ for (total = 0; count > 0; count--) {
-+ if (*bptr)
-+ total += hweight_long(*bptr);
-+ bptr++;
-+ }
-+ spin_unlock(&files->file_lock);
-+ return total;
-+}
++ /* initial flags */
++ new_vxi->vx_flags = vc_data.flagword;
+
++ ret = -ENOEXEC;
++ if (vs_state_change(new_vxi, VSC_STARTUP))
++ goto out;
+
-+/* for *space compatibility */
++ ret = vx_migrate_task(current, new_vxi, (!data));
++ if (ret)
++ goto out;
+
-+asmlinkage long sys_unshare(unsigned long);
++ /* return context id on success */
++ ret = new_vxi->vx_id;
+
-+/*
-+ * migrate task to new context
-+ * gets vxi, puts old_vxi on change
-+ * optionally unshares namespaces (hack)
-+ */
++ /* get a reference for persistent contexts */
++ if ((vc_data.flagword & VXF_PERSISTENT))
++ vx_set_persistent(new_vxi);
++out:
++ release_vx_info(new_vxi, NULL);
++ put_vx_info(new_vxi);
++ return ret;
++}
+
-+int vx_migrate_task(struct task_struct *p, struct vx_info *vxi, int unshare)
++
++int vc_ctx_migrate(struct vx_info *vxi, void __user *data)
+{
-+ struct vx_info *old_vxi;
-+ int ret = 0;
++ struct vcmd_ctx_migrate vc_data = { .flagword = 0 };
++ int ret;
+
-+ if (!p || !vxi)
-+ BUG();
++ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
+
-+ vxdprintk(VXD_CBIT(xid, 5),
-+ "vx_migrate_task(%p,%p[#%d.%d])", p, vxi,
-+ vxi->vx_id, atomic_read(&vxi->vx_usecnt));
++ ret = vx_migrate_task(current, vxi, 0);
++ if (ret)
++ return ret;
++ if (vc_data.flagword & VXM_SET_INIT)
++ ret = vx_set_init(vxi, current);
++ if (ret)
++ return ret;
++ if (vc_data.flagword & VXM_SET_REAPER)
++ ret = vx_set_reaper(vxi, current);
++ return ret;
++}
+
-+ if (vx_info_flags(vxi, VXF_INFO_PRIVATE, 0) &&
-+ !vx_info_flags(vxi, VXF_STATE_SETUP, 0))
-+ return -EACCES;
+
-+ if (vx_info_state(vxi, VXS_SHUTDOWN))
++int vc_get_cflags(struct vx_info *vxi, void __user *data)
++{
++ struct vcmd_ctx_flags_v0 vc_data;
++
++ vc_data.flagword = vxi->vx_flags;
++
++ /* special STATE flag handling */
++ vc_data.mask = vs_mask_flags(~0ULL, vxi->vx_flags, VXF_ONE_TIME);
++
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
+ return -EFAULT;
++ return 0;
++}
+
-+ old_vxi = task_get_vx_info(p);
-+ if (old_vxi == vxi)
-+ goto out;
++int vc_set_cflags(struct vx_info *vxi, void __user *data)
++{
++ struct vcmd_ctx_flags_v0 vc_data;
++ uint64_t mask, trigger;
+
-+ if (!(ret = vx_migrate_user(p, vxi))) {
-+ int openfd;
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
+
-+ task_lock(p);
-+ openfd = vx_openfd_task(p);
++ /* special STATE flag handling */
++ mask = vs_mask_mask(vc_data.mask, vxi->vx_flags, VXF_ONE_TIME);
++ trigger = (mask & vxi->vx_flags) ^ (mask & vc_data.flagword);
+
-+ if (old_vxi) {
-+ atomic_dec(&old_vxi->cvirt.nr_threads);
-+ atomic_dec(&old_vxi->cvirt.nr_running);
-+ __rlim_dec(&old_vxi->limit, RLIMIT_NPROC);
-+ /* FIXME: what about the struct files here? */
-+ __rlim_sub(&old_vxi->limit, VLIMIT_OPENFD, openfd);
-+ /* account for the executable */
-+ __rlim_dec(&old_vxi->limit, VLIMIT_DENTRY);
-+ }
-+ atomic_inc(&vxi->cvirt.nr_threads);
-+ atomic_inc(&vxi->cvirt.nr_running);
-+ __rlim_inc(&vxi->limit, RLIMIT_NPROC);
-+ /* FIXME: what about the struct files here? */
-+ __rlim_add(&vxi->limit, VLIMIT_OPENFD, openfd);
-+ /* account for the executable */
-+ __rlim_inc(&vxi->limit, VLIMIT_DENTRY);
++ if (vxi == current->vx_info) {
++ /* if (trigger & VXF_STATE_SETUP)
++ vx_mask_cap_bset(vxi, current); */
++ if (trigger & VXF_STATE_INIT) {
++ int ret;
+
-+ if (old_vxi) {
-+ release_vx_info(old_vxi, p);
-+ clr_vx_info(&p->vx_info);
++ ret = vx_set_init(vxi, current);
++ if (ret)
++ return ret;
++ ret = vx_set_reaper(vxi, current);
++ if (ret)
++ return ret;
+ }
-+ claim_vx_info(vxi, p);
-+ set_vx_info(&p->vx_info, vxi);
-+ p->xid = vxi->vx_id;
++ }
+
-+ vxdprintk(VXD_CBIT(xid, 5),
-+ "moved task %p into vxi:%p[#%d]",
-+ p, vxi, vxi->vx_id);
++ vxi->vx_flags = vs_mask_flags(vxi->vx_flags,
++ vc_data.flagword, mask);
++ if (trigger & VXF_PERSISTENT)
++ vx_update_persistent(vxi);
+
-+ vx_mask_cap_bset(vxi, p);
-+ task_unlock(p);
++ return 0;
++}
+
-+ /* hack for *spaces to provide compatibility */
-+ if (unshare) {
-+ struct nsproxy *old_nsp = p->nsproxy;
-+ struct nsproxy *new_nsp;
+
-+ ret = unshare_nsproxy_namespaces(
-+ CLONE_NEWUTS | CLONE_NEWIPC,
-+ &new_nsp, NULL);
-+ if (ret)
-+ goto out;
++static inline uint64_t caps_from_cap_t(kernel_cap_t c)
++{
++ uint64_t v = c.cap[0] | ((uint64_t)c.cap[1] << 32);
+
-+ new_nsp = xchg(&p->nsproxy, new_nsp);
-+ vx_set_space(vxi, CLONE_NEWUTS | CLONE_NEWIPC);
-+ put_nsproxy(old_nsp);
-+ put_nsproxy(new_nsp);
-+ }
-+ }
-+out:
-+ put_vx_info(old_vxi);
-+ return ret;
++ // printk("caps_from_cap_t(%08x:%08x) = %016llx\n", c.cap[1], c.cap[0], v);
++ return v;
+}
+
-+int vx_set_reaper(struct vx_info *vxi, struct task_struct *p)
++static inline kernel_cap_t cap_t_from_caps(uint64_t v)
+{
-+ struct task_struct *old_reaper;
++ kernel_cap_t c = __cap_empty_set;
+
-+ if (!vxi)
-+ return -EINVAL;
++ c.cap[0] = v & 0xFFFFFFFF;
++ c.cap[1] = (v >> 32) & 0xFFFFFFFF;
+
-+ vxdprintk(VXD_CBIT(xid, 6),
-+ "vx_set_reaper(%p[#%d],%p[#%d,%d])",
-+ vxi, vxi->vx_id, p, p->xid, p->pid);
++ // printk("cap_t_from_caps(%016llx) = %08x:%08x\n", v, c.cap[1], c.cap[0]);
++ return c;
++}
+
-+ old_reaper = vxi->vx_reaper;
-+ if (old_reaper == p)
-+ return 0;
+
-+ /* set new child reaper */
-+ get_task_struct(p);
-+ vxi->vx_reaper = p;
-+ put_task_struct(old_reaper);
++static int do_get_caps(struct vx_info *vxi, uint64_t *bcaps, uint64_t *ccaps)
++{
++ if (bcaps)
++ *bcaps = caps_from_cap_t(vxi->vx_bcaps);
++ if (ccaps)
++ *ccaps = vxi->vx_ccaps;
++
+ return 0;
+}
+
-+int vx_set_init(struct vx_info *vxi, struct task_struct *p)
++int vc_get_ccaps(struct vx_info *vxi, void __user *data)
+{
-+ if (!vxi)
-+ return -EINVAL;
++ struct vcmd_ctx_caps_v1 vc_data;
++ int ret;
+
-+ vxdprintk(VXD_CBIT(xid, 6),
-+ "vx_set_init(%p[#%d],%p[#%d,%d,%d])",
-+ vxi, vxi->vx_id, p, p->xid, p->pid, p->tgid);
++ ret = do_get_caps(vxi, NULL, &vc_data.ccaps);
++ if (ret)
++ return ret;
++ vc_data.cmask = ~0ULL;
+
-+ vxi->vx_flags &= ~VXF_STATE_INIT;
-+ vxi->vx_initpid = p->tgid;
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ return -EFAULT;
+ return 0;
+}
+
-+void vx_exit_init(struct vx_info *vxi, struct task_struct *p, int code)
++static int do_set_caps(struct vx_info *vxi,
++ uint64_t bcaps, uint64_t bmask, uint64_t ccaps, uint64_t cmask)
+{
-+ vxdprintk(VXD_CBIT(xid, 6),
-+ "vx_exit_init(%p[#%d],%p[#%d,%d,%d])",
-+ vxi, vxi->vx_id, p, p->xid, p->pid, p->tgid);
++ uint64_t bcold = caps_from_cap_t(vxi->vx_bcaps);
+
-+ vxi->exit_code = code;
-+ vxi->vx_initpid = 0;
-+}
++#if 0
++ printk("do_set_caps(%16llx, %16llx, %16llx, %16llx)\n",
++ bcaps, bmask, ccaps, cmask);
++#endif
++ vxi->vx_bcaps = cap_t_from_caps(
++ vs_mask_flags(bcold, bcaps, bmask));
++ vxi->vx_ccaps = vs_mask_flags(vxi->vx_ccaps, ccaps, cmask);
+
++ return 0;
++}
+
-+void vx_set_persistent(struct vx_info *vxi)
++int vc_set_ccaps(struct vx_info *vxi, void __user *data)
+{
-+ vxdprintk(VXD_CBIT(xid, 6),
-+ "vx_set_persistent(%p[#%d])", vxi, vxi->vx_id);
++ struct vcmd_ctx_caps_v1 vc_data;
+
-+ get_vx_info(vxi);
-+ claim_vx_info(vxi, NULL);
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ return do_set_caps(vxi, 0, 0, vc_data.ccaps, vc_data.cmask);
+}
+
-+void vx_clear_persistent(struct vx_info *vxi)
++int vc_get_bcaps(struct vx_info *vxi, void __user *data)
+{
-+ vxdprintk(VXD_CBIT(xid, 6),
-+ "vx_clear_persistent(%p[#%d])", vxi, vxi->vx_id);
++ struct vcmd_bcaps vc_data;
++ int ret;
+
-+ release_vx_info(vxi, NULL);
-+ put_vx_info(vxi);
++ ret = do_get_caps(vxi, &vc_data.bcaps, NULL);
++ if (ret)
++ return ret;
++ vc_data.bmask = ~0ULL;
++
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ return -EFAULT;
++ return 0;
+}
+
-+void vx_update_persistent(struct vx_info *vxi)
++int vc_set_bcaps(struct vx_info *vxi, void __user *data)
+{
-+ if (vx_info_flags(vxi, VXF_PERSISTENT, 0))
-+ vx_set_persistent(vxi);
-+ else
-+ vx_clear_persistent(vxi);
-+}
++ struct vcmd_bcaps vc_data;
+
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
+
-+/* task must be current or locked */
++ return do_set_caps(vxi, vc_data.bcaps, vc_data.bmask, 0, 0);
++}
+
-+void exit_vx_info(struct task_struct *p, int code)
++
++int vc_get_badness(struct vx_info *vxi, void __user *data)
+{
-+ struct vx_info *vxi = p->vx_info;
++ struct vcmd_badness_v0 vc_data;
+
-+ if (vxi) {
-+ atomic_dec(&vxi->cvirt.nr_threads);
-+ vx_nproc_dec(p);
++ vc_data.bias = vxi->vx_badness_bias;
+
-+ vxi->exit_code = code;
-+ release_vx_info(vxi, p);
-+ }
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ return -EFAULT;
++ return 0;
+}
+
-+void exit_vx_info_early(struct task_struct *p, int code)
++int vc_set_badness(struct vx_info *vxi, void __user *data)
+{
-+ struct vx_info *vxi = p->vx_info;
++ struct vcmd_badness_v0 vc_data;
+
-+ if (vxi) {
-+ if (vxi->vx_initpid == p->tgid)
-+ vx_exit_init(vxi, p, code);
-+ if (vxi->vx_reaper == p)
-+ vx_set_reaper(vxi, init_pid_ns.child_reaper);
-+ }
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ vxi->vx_badness_bias = vc_data.bias;
++ return 0;
+}
+
++#include <linux/module.h>
+
-+/* vserver syscall commands below here */
++EXPORT_SYMBOL_GPL(free_vx_info);
+
-+/* taks xid and vx_info functions */
+--- linux-2.6.25.4/kernel/vserver/cvirt.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/cvirt.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,301 @@
++/*
++ * linux/kernel/vserver/cvirt.c
++ *
++ * Virtual Server: Context Virtualization
++ *
++ * Copyright (C) 2004-2007 Herbert Pötzl
++ *
++ * V0.01 broken out from limit.c
++ * V0.02 added utsname stuff
++ * V0.03 changed vcmds to vxi arg
++ *
++ */
++
++#include <linux/types.h>
++#include <linux/utsname.h>
++#include <linux/vs_cvirt.h>
++#include <linux/vserver/switch.h>
++#include <linux/vserver/cvirt_cmd.h>
+
+#include <asm/uaccess.h>
+
+
-+int vc_task_xid(uint32_t id, void __user *data)
++void vx_vsi_uptime(struct timespec *uptime, struct timespec *idle)
+{
-+ xid_t xid;
-+
-+ if (id) {
-+ struct task_struct *tsk;
-+
-+ if (!vx_check(0, VS_ADMIN | VS_WATCH))
-+ return -EPERM;
++ struct vx_info *vxi = current->vx_info;
+
-+ read_lock(&tasklist_lock);
-+ tsk = find_task_by_real_pid(id);
-+ xid = (tsk) ? tsk->xid : -ESRCH;
-+ read_unlock(&tasklist_lock);
-+ } else
-+ xid = vx_current_xid();
-+ return xid;
++ set_normalized_timespec(uptime,
++ uptime->tv_sec - vxi->cvirt.bias_uptime.tv_sec,
++ uptime->tv_nsec - vxi->cvirt.bias_uptime.tv_nsec);
++ if (!idle)
++ return;
++ set_normalized_timespec(idle,
++ idle->tv_sec - vxi->cvirt.bias_idle.tv_sec,
++ idle->tv_nsec - vxi->cvirt.bias_idle.tv_nsec);
++ return;
+}
+
-+
-+int vc_vx_info(struct vx_info *vxi, void __user *data)
++uint64_t vx_idle_jiffies(void)
+{
-+ struct vcmd_vx_info_v0 vc_data;
-+
-+ vc_data.xid = vxi->vx_id;
-+ vc_data.initpid = vxi->vx_initpid;
-+
-+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
-+ return -EFAULT;
-+ return 0;
++ return init_task.utime + init_task.stime;
+}
+
+
-+int vc_ctx_stat(struct vx_info *vxi, void __user *data)
++
++static inline uint32_t __update_loadavg(uint32_t load,
++ int wsize, int delta, int n)
+{
-+ struct vcmd_ctx_stat_v0 vc_data;
++ unsigned long long calc, prev;
+
-+ vc_data.usecnt = atomic_read(&vxi->vx_usecnt);
-+ vc_data.tasks = atomic_read(&vxi->vx_tasks);
++ /* just set it to n */
++ if (unlikely(delta >= wsize))
++ return (n << FSHIFT);
+
-+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
-+ return -EFAULT;
-+ return 0;
++ calc = delta * n;
++ calc <<= FSHIFT;
++ prev = (wsize - delta);
++ prev *= load;
++ calc += prev;
++ do_div(calc, wsize);
++ return calc;
+}
+
+
-+/* context functions */
-+
-+int vc_ctx_create(uint32_t xid, void __user *data)
++void vx_update_load(struct vx_info *vxi)
+{
-+ struct vcmd_ctx_create vc_data = { .flagword = VXF_INIT_SET };
-+ struct vx_info *new_vxi;
-+ int ret;
-+
-+ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
-+
-+ if ((xid > MAX_S_CONTEXT) && (xid != VX_DYNAMIC_ID))
-+ return -EINVAL;
-+ if (xid < 2)
-+ return -EINVAL;
++ uint32_t now, last, delta;
++ unsigned int nr_running, nr_uninterruptible;
++ unsigned int total;
++ unsigned long flags;
+
-+ new_vxi = __create_vx_info(xid);
-+ if (IS_ERR(new_vxi))
-+ return PTR_ERR(new_vxi);
++ spin_lock_irqsave(&vxi->cvirt.load_lock, flags);
+
-+ /* initial flags */
-+ new_vxi->vx_flags = vc_data.flagword;
++ now = jiffies;
++ last = vxi->cvirt.load_last;
++ delta = now - last;
+
-+ ret = -ENOEXEC;
-+ if (vs_state_change(new_vxi, VSC_STARTUP))
++ if (delta < 5*HZ)
+ goto out;
+
-+ ret = vx_migrate_task(current, new_vxi, (!data));
-+ if (ret)
-+ goto out;
++ nr_running = atomic_read(&vxi->cvirt.nr_running);
++ nr_uninterruptible = atomic_read(&vxi->cvirt.nr_uninterruptible);
++ total = nr_running + nr_uninterruptible;
+
-+ /* return context id on success */
-+ ret = new_vxi->vx_id;
++ vxi->cvirt.load[0] = __update_loadavg(vxi->cvirt.load[0],
++ 60*HZ, delta, total);
++ vxi->cvirt.load[1] = __update_loadavg(vxi->cvirt.load[1],
++ 5*60*HZ, delta, total);
++ vxi->cvirt.load[2] = __update_loadavg(vxi->cvirt.load[2],
++ 15*60*HZ, delta, total);
+
-+ /* get a reference for persistent contexts */
-+ if ((vc_data.flagword & VXF_PERSISTENT))
-+ vx_set_persistent(new_vxi);
++ vxi->cvirt.load_last = now;
+out:
-+ release_vx_info(new_vxi, NULL);
-+ put_vx_info(new_vxi);
-+ return ret;
++ atomic_inc(&vxi->cvirt.load_updates);
++ spin_unlock_irqrestore(&vxi->cvirt.load_lock, flags);
+}
+
+
-+int vc_ctx_migrate(struct vx_info *vxi, void __user *data)
++/*
++ * Commands to do_syslog:
++ *
++ * 0 -- Close the log. Currently a NOP.
++ * 1 -- Open the log. Currently a NOP.
++ * 2 -- Read from the log.
++ * 3 -- Read all messages remaining in the ring buffer.
++ * 4 -- Read and clear all messages remaining in the ring buffer
++ * 5 -- Clear ring buffer.
++ * 6 -- Disable printk's to console
++ * 7 -- Enable printk's to console
++ * 8 -- Set level of messages printed to console
++ * 9 -- Return number of unread characters in the log buffer
++ * 10 -- Return size of the log buffer
++ */
++int vx_do_syslog(int type, char __user *buf, int len)
+{
-+ struct vcmd_ctx_migrate vc_data = { .flagword = 0 };
-+ int ret;
++ int error = 0;
++ int do_clear = 0;
++ struct vx_info *vxi = current->vx_info;
++ struct _vx_syslog *log;
+
-+ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
++ if (!vxi)
++ return -EINVAL;
++ log = &vxi->cvirt.syslog;
+
-+ ret = vx_migrate_task(current, vxi, 0);
-+ if (ret)
-+ return ret;
-+ if (vc_data.flagword & VXM_SET_INIT)
-+ ret = vx_set_init(vxi, current);
-+ if (ret)
-+ return ret;
-+ if (vc_data.flagword & VXM_SET_REAPER)
-+ ret = vx_set_reaper(vxi, current);
-+ return ret;
++ switch (type) {
++ case 0: /* Close log */
++ case 1: /* Open log */
++ break;
++ case 2: /* Read from log */
++ error = wait_event_interruptible(log->log_wait,
++ (log->log_start - log->log_end));
++ if (error)
++ break;
++ spin_lock_irq(&log->logbuf_lock);
++ spin_unlock_irq(&log->logbuf_lock);
++ break;
++ case 4: /* Read/clear last kernel messages */
++ do_clear = 1;
++ /* fall through */
++ case 3: /* Read last kernel messages */
++ return 0;
++
++ case 5: /* Clear ring buffer */
++ return 0;
++
++ case 6: /* Disable logging to console */
++ case 7: /* Enable logging to console */
++ case 8: /* Set level of messages printed to console */
++ break;
++
++ case 9: /* Number of chars in the log buffer */
++ return 0;
++ case 10: /* Size of the log buffer */
++ return 0;
++ default:
++ error = -EINVAL;
++ break;
++ }
++ return error;
+}
+
+
-+int vc_get_cflags(struct vx_info *vxi, void __user *data)
++/* virtual host info names */
++
++static char *vx_vhi_name(struct vx_info *vxi, int id)
+{
-+ struct vcmd_ctx_flags_v0 vc_data;
++ struct nsproxy *nsproxy;
++ struct uts_namespace *uts;
+
-+ vc_data.flagword = vxi->vx_flags;
+
-+ /* special STATE flag handling */
-+ vc_data.mask = vs_mask_flags(~0UL, vxi->vx_flags, VXF_ONE_TIME);
++ if (id == VHIN_CONTEXT)
++ return vxi->vx_name;
+
-+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
-+ return -EFAULT;
-+ return 0;
++ nsproxy = vxi->vx_nsproxy;
++ if (!nsproxy)
++ return NULL;
++
++ uts = nsproxy->uts_ns;
++ if (!uts)
++ return NULL;
++
++ switch (id) {
++ case VHIN_SYSNAME:
++ return uts->name.sysname;
++ case VHIN_NODENAME:
++ return uts->name.nodename;
++ case VHIN_RELEASE:
++ return uts->name.release;
++ case VHIN_VERSION:
++ return uts->name.version;
++ case VHIN_MACHINE:
++ return uts->name.machine;
++ case VHIN_DOMAINNAME:
++ return uts->name.domainname;
++ default:
++ return NULL;
++ }
++ return NULL;
+}
+
-+int vc_set_cflags(struct vx_info *vxi, void __user *data)
++int vc_set_vhi_name(struct vx_info *vxi, void __user *data)
+{
-+ struct vcmd_ctx_flags_v0 vc_data;
-+ uint64_t mask, trigger;
++ struct vcmd_vhi_name_v0 vc_data;
++ char *name;
+
+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
+ return -EFAULT;
+
-+ /* special STATE flag handling */
-+ mask = vs_mask_mask(vc_data.mask, vxi->vx_flags, VXF_ONE_TIME);
-+ trigger = (mask & vxi->vx_flags) ^ (mask & vc_data.flagword);
-+
-+ if (vxi == current->vx_info) {
-+ if (trigger & VXF_STATE_SETUP)
-+ vx_mask_cap_bset(vxi, current);
-+ if (trigger & VXF_STATE_INIT) {
-+ int ret;
-+
-+ ret = vx_set_init(vxi, current);
-+ if (ret)
-+ return ret;
-+ ret = vx_set_reaper(vxi, current);
-+ if (ret)
-+ return ret;
-+ }
-+ }
-+
-+ vxi->vx_flags = vs_mask_flags(vxi->vx_flags,
-+ vc_data.flagword, mask);
-+ if (trigger & VXF_PERSISTENT)
-+ vx_update_persistent(vxi);
++ name = vx_vhi_name(vxi, vc_data.field);
++ if (!name)
++ return -EINVAL;
+
++ memcpy(name, vc_data.name, 65);
+ return 0;
+}
+
-+static int do_get_caps(struct vx_info *vxi, uint64_t *bcaps, uint64_t *ccaps)
++int vc_get_vhi_name(struct vx_info *vxi, void __user *data)
+{
-+ if (bcaps)
-+ *bcaps = vxi->vx_bcaps;
-+ if (ccaps)
-+ *ccaps = vxi->vx_ccaps;
-+
-+ return 0;
-+}
++ struct vcmd_vhi_name_v0 vc_data;
++ char *name;
+
-+int vc_get_ccaps_v0(struct vx_info *vxi, void __user *data)
-+{
-+ struct vcmd_ctx_caps_v0 vc_data;
-+ int ret;
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
+
-+ ret = do_get_caps(vxi, &vc_data.bcaps, &vc_data.ccaps);
-+ if (ret)
-+ return ret;
-+ vc_data.cmask = ~0UL;
++ name = vx_vhi_name(vxi, vc_data.field);
++ if (!name)
++ return -EINVAL;
+
++ memcpy(vc_data.name, name, 65);
+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
+ return -EFAULT;
+ return 0;
+}
+
-+int vc_get_ccaps(struct vx_info *vxi, void __user *data)
++
++int vc_virt_stat(struct vx_info *vxi, void __user *data)
+{
-+ struct vcmd_ctx_caps_v1 vc_data;
-+ int ret;
++ struct vcmd_virt_stat_v0 vc_data;
++ struct _vx_cvirt *cvirt = &vxi->cvirt;
++ struct timespec uptime;
+
-+ ret = do_get_caps(vxi, NULL, &vc_data.ccaps);
-+ if (ret)
-+ return ret;
-+ vc_data.cmask = ~0UL;
++ do_posix_clock_monotonic_gettime(&uptime);
++ set_normalized_timespec(&uptime,
++ uptime.tv_sec - cvirt->bias_uptime.tv_sec,
++ uptime.tv_nsec - cvirt->bias_uptime.tv_nsec);
++
++ vc_data.offset = timeval_to_ns(&cvirt->bias_tv);
++ vc_data.uptime = timespec_to_ns(&uptime);
++ vc_data.nr_threads = atomic_read(&cvirt->nr_threads);
++ vc_data.nr_running = atomic_read(&cvirt->nr_running);
++ vc_data.nr_uninterruptible = atomic_read(&cvirt->nr_uninterruptible);
++ vc_data.nr_onhold = atomic_read(&cvirt->nr_onhold);
++ vc_data.nr_forks = atomic_read(&cvirt->total_forks);
++ vc_data.load[0] = cvirt->load[0];
++ vc_data.load[1] = cvirt->load[1];
++ vc_data.load[2] = cvirt->load[2];
+
+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
+ return -EFAULT;
+ return 0;
+}
+
-+static int do_set_caps(struct vx_info *vxi,
-+ uint64_t bcaps, uint64_t bmask, uint64_t ccaps, uint64_t cmask)
-+{
-+ vxi->vx_bcaps = vs_mask_flags(vxi->vx_bcaps, bcaps, bmask);
-+ vxi->vx_ccaps = vs_mask_flags(vxi->vx_ccaps, ccaps, cmask);
-+
-+ return 0;
-+}
-+
-+int vc_set_ccaps_v0(struct vx_info *vxi, void __user *data)
-+{
-+ struct vcmd_ctx_caps_v0 vc_data;
+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
++#ifdef CONFIG_VSERVER_VTIME
+
-+ /* simulate old &= behaviour for bcaps */
-+ return do_set_caps(vxi, 0, ~vc_data.bcaps,
-+ vc_data.ccaps, vc_data.cmask);
-+}
++/* virtualized time base */
+
-+int vc_set_ccaps(struct vx_info *vxi, void __user *data)
++void vx_gettimeofday(struct timeval *tv)
+{
-+ struct vcmd_ctx_caps_v1 vc_data;
++ do_gettimeofday(tv);
++ if (!vx_flags(VXF_VIRT_TIME, 0))
++ return;
+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
++ tv->tv_sec += current->vx_info->cvirt.bias_tv.tv_sec;
++ tv->tv_usec += current->vx_info->cvirt.bias_tv.tv_usec;
+
-+ return do_set_caps(vxi, 0, 0, vc_data.ccaps, vc_data.cmask);
++ if (tv->tv_usec >= USEC_PER_SEC) {
++ tv->tv_sec++;
++ tv->tv_usec -= USEC_PER_SEC;
++ } else if (tv->tv_usec < 0) {
++ tv->tv_sec--;
++ tv->tv_usec += USEC_PER_SEC;
++ }
+}
+
-+int vc_get_bcaps(struct vx_info *vxi, void __user *data)
++int vx_settimeofday(struct timespec *ts)
+{
-+ struct vcmd_bcaps vc_data;
-+ int ret;
++ struct timeval tv;
+
-+ ret = do_get_caps(vxi, &vc_data.bcaps, NULL);
-+ if (ret)
-+ return ret;
-+ vc_data.bmask = ~0UL;
++ if (!vx_flags(VXF_VIRT_TIME, 0))
++ return do_settimeofday(ts);
+
-+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
-+ return -EFAULT;
++ do_gettimeofday(&tv);
++ current->vx_info->cvirt.bias_tv.tv_sec =
++ ts->tv_sec - tv.tv_sec;
++ current->vx_info->cvirt.bias_tv.tv_usec =
++ (ts->tv_nsec/NSEC_PER_USEC) - tv.tv_usec;
+ return 0;
+}
+
-+int vc_set_bcaps(struct vx_info *vxi, void __user *data)
-+{
-+ struct vcmd_bcaps vc_data;
++#endif
+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
+--- linux-2.6.25.4/kernel/vserver/cvirt_init.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/cvirt_init.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,69 @@
+
-+ return do_set_caps(vxi, vc_data.bcaps, vc_data.bmask, 0, 0);
-+}
+
-+#include <linux/module.h>
++extern uint64_t vx_idle_jiffies(void);
+
-+EXPORT_SYMBOL_GPL(free_vx_info);
++static inline void vx_info_init_cvirt(struct _vx_cvirt *cvirt)
++{
++ uint64_t idle_jiffies = vx_idle_jiffies();
++ uint64_t nsuptime;
+
---- linux-2.6.22/kernel/vserver/cvirt.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/cvirt.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,304 @@
-+/*
-+ * linux/kernel/vserver/cvirt.c
-+ *
-+ * Virtual Server: Context Virtualization
-+ *
-+ * Copyright (C) 2004-2007 Herbert Pötzl
-+ *
-+ * V0.01 broken out from limit.c
-+ * V0.02 added utsname stuff
-+ * V0.03 changed vcmds to vxi arg
-+ *
-+ */
++ do_posix_clock_monotonic_gettime(&cvirt->bias_uptime);
++ nsuptime = (unsigned long long)cvirt->bias_uptime.tv_sec
++ * NSEC_PER_SEC + cvirt->bias_uptime.tv_nsec;
++ cvirt->bias_clock = nsec_to_clock_t(nsuptime);
++ cvirt->bias_tv.tv_sec = 0;
++ cvirt->bias_tv.tv_usec = 0;
+
-+#include <linux/sched.h>
-+#include <linux/sysctl.h>
-+#include <linux/types.h>
-+#include <linux/vs_context.h>
-+#include <linux/vs_cvirt.h>
-+#include <linux/vserver/switch.h>
-+#include <linux/vserver/cvirt_cmd.h>
++ jiffies_to_timespec(idle_jiffies, &cvirt->bias_idle);
++ atomic_set(&cvirt->nr_threads, 0);
++ atomic_set(&cvirt->nr_running, 0);
++ atomic_set(&cvirt->nr_uninterruptible, 0);
++ atomic_set(&cvirt->nr_onhold, 0);
+
-+#include <asm/errno.h>
-+#include <asm/uaccess.h>
++ spin_lock_init(&cvirt->load_lock);
++ cvirt->load_last = jiffies;
++ atomic_set(&cvirt->load_updates, 0);
++ cvirt->load[0] = 0;
++ cvirt->load[1] = 0;
++ cvirt->load[2] = 0;
++ atomic_set(&cvirt->total_forks, 0);
+
++ spin_lock_init(&cvirt->syslog.logbuf_lock);
++ init_waitqueue_head(&cvirt->syslog.log_wait);
++ cvirt->syslog.log_start = 0;
++ cvirt->syslog.log_end = 0;
++ cvirt->syslog.con_start = 0;
++ cvirt->syslog.logged_chars = 0;
++}
+
-+void vx_vsi_uptime(struct timespec *uptime, struct timespec *idle)
++static inline
++void vx_info_init_cvirt_pc(struct _vx_cvirt_pc *cvirt_pc, int cpu)
+{
-+ struct vx_info *vxi = current->vx_info;
++ // cvirt_pc->cpustat = { 0 };
++}
+
-+ set_normalized_timespec(uptime,
-+ uptime->tv_sec - vxi->cvirt.bias_uptime.tv_sec,
-+ uptime->tv_nsec - vxi->cvirt.bias_uptime.tv_nsec);
-+ if (!idle)
-+ return;
-+ set_normalized_timespec(idle,
-+ idle->tv_sec - vxi->cvirt.bias_idle.tv_sec,
-+ idle->tv_nsec - vxi->cvirt.bias_idle.tv_nsec);
++static inline void vx_info_exit_cvirt(struct _vx_cvirt *cvirt)
++{
++ int value;
++
++ vxwprintk_xid((value = atomic_read(&cvirt->nr_threads)),
++ "!!! cvirt: %p[nr_threads] = %d on exit.",
++ cvirt, value);
++ vxwprintk_xid((value = atomic_read(&cvirt->nr_running)),
++ "!!! cvirt: %p[nr_running] = %d on exit.",
++ cvirt, value);
++ vxwprintk_xid((value = atomic_read(&cvirt->nr_uninterruptible)),
++ "!!! cvirt: %p[nr_uninterruptible] = %d on exit.",
++ cvirt, value);
++ vxwprintk_xid((value = atomic_read(&cvirt->nr_onhold)),
++ "!!! cvirt: %p[nr_onhold] = %d on exit.",
++ cvirt, value);
+ return;
+}
+
-+uint64_t vx_idle_jiffies(void)
++static inline
++void vx_info_exit_cvirt_pc(struct _vx_cvirt_pc *cvirt_pc, int cpu)
+{
-+ return init_task.utime + init_task.stime;
++ return;
+}
+
+--- linux-2.6.25.4/kernel/vserver/cvirt_proc.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/cvirt_proc.h 2008-04-21 13:01:29.000000000 -0400
+@@ -0,0 +1,135 @@
++#ifndef _VX_CVIRT_PROC_H
++#define _VX_CVIRT_PROC_H
++
++#include <linux/nsproxy.h>
++#include <linux/mnt_namespace.h>
++#include <linux/ipc_namespace.h>
++#include <linux/utsname.h>
++#include <linux/ipc.h>
++
++
++static inline
++int vx_info_proc_nsproxy(struct nsproxy *nsproxy, char *buffer)
++{
++ struct mnt_namespace *ns;
++ struct uts_namespace *uts;
++ struct ipc_namespace *ipc;
++ struct path path;
++ char *pstr, *root;
++ int length = 0;
++
++ if (!nsproxy)
++ goto out;
++
++ length += sprintf(buffer + length,
++ "NSProxy:\t%p [%p,%p,%p]\n",
++ nsproxy, nsproxy->mnt_ns,
++ nsproxy->uts_ns, nsproxy->ipc_ns);
+
++ ns = nsproxy->mnt_ns;
++ if (!ns)
++ goto skip_ns;
+
-+static inline uint32_t __update_loadavg(uint32_t load,
-+ int wsize, int delta, int n)
-+{
-+ unsigned long long calc, prev;
++ pstr = kmalloc(PATH_MAX, GFP_KERNEL);
++ if (!pstr)
++ goto skip_ns;
+
-+ /* just set it to n */
-+ if (unlikely(delta >= wsize))
-+ return (n << FSHIFT);
++ path.mnt = ns->root;
++ path.dentry = ns->root->mnt_root;
++ root = d_path(&path, pstr, PATH_MAX - 2);
++ length += sprintf(buffer + length,
++ "Namespace:\t%p [#%u]\n"
++ "RootPath:\t%s\n",
++ ns, atomic_read(&ns->count),
++ root);
++ kfree(pstr);
++skip_ns:
+
-+ calc = delta * n;
-+ calc <<= FSHIFT;
-+ prev = (wsize - delta);
-+ prev *= load;
-+ calc += prev;
-+ do_div(calc, wsize);
-+ return calc;
-+}
++ uts = nsproxy->uts_ns;
++ if (!uts)
++ goto skip_uts;
+
++ length += sprintf(buffer + length,
++ "SysName:\t%.*s\n"
++ "NodeName:\t%.*s\n"
++ "Release:\t%.*s\n"
++ "Version:\t%.*s\n"
++ "Machine:\t%.*s\n"
++ "DomainName:\t%.*s\n",
++ __NEW_UTS_LEN, uts->name.sysname,
++ __NEW_UTS_LEN, uts->name.nodename,
++ __NEW_UTS_LEN, uts->name.release,
++ __NEW_UTS_LEN, uts->name.version,
++ __NEW_UTS_LEN, uts->name.machine,
++ __NEW_UTS_LEN, uts->name.domainname);
++skip_uts:
+
-+void vx_update_load(struct vx_info *vxi)
-+{
-+ uint32_t now, last, delta;
-+ unsigned int nr_running, nr_uninterruptible;
-+ unsigned int total;
-+ unsigned long flags;
++ ipc = nsproxy->ipc_ns;
++ if (!ipc)
++ goto skip_ipc;
+
-+ spin_lock_irqsave(&vxi->cvirt.load_lock, flags);
++ length += sprintf(buffer + length,
++ "SEMS:\t\t%d %d %d %d %d\n"
++ "MSG:\t\t%d %d %d\n"
++ "SHM:\t\t%lu %lu %d %d\n",
++ ipc->sem_ctls[0], ipc->sem_ctls[1],
++ ipc->sem_ctls[2], ipc->sem_ctls[3],
++ ipc->used_sems,
++ ipc->msg_ctlmax, ipc->msg_ctlmnb, ipc->msg_ctlmni,
++ (unsigned long)ipc->shm_ctlmax,
++ (unsigned long)ipc->shm_ctlall,
++ ipc->shm_ctlmni, ipc->shm_tot);
++skip_ipc:
++out:
++ return length;
++}
+
-+ now = jiffies;
-+ last = vxi->cvirt.load_last;
-+ delta = now - last;
+
-+ if (delta < 5*HZ)
-+ goto out;
++#include <linux/sched.h>
+
-+ nr_running = atomic_read(&vxi->cvirt.nr_running);
-+ nr_uninterruptible = atomic_read(&vxi->cvirt.nr_uninterruptible);
-+ total = nr_running + nr_uninterruptible;
++#define LOAD_INT(x) ((x) >> FSHIFT)
++#define LOAD_FRAC(x) LOAD_INT(((x) & (FIXED_1 - 1)) * 100)
+
-+ vxi->cvirt.load[0] = __update_loadavg(vxi->cvirt.load[0],
-+ 60*HZ, delta, total);
-+ vxi->cvirt.load[1] = __update_loadavg(vxi->cvirt.load[1],
-+ 5*60*HZ, delta, total);
-+ vxi->cvirt.load[2] = __update_loadavg(vxi->cvirt.load[2],
-+ 15*60*HZ, delta, total);
++static inline
++int vx_info_proc_cvirt(struct _vx_cvirt *cvirt, char *buffer)
++{
++ int length = 0;
++ int a, b, c;
+
-+ vxi->cvirt.load_last = now;
-+out:
-+ atomic_inc(&vxi->cvirt.load_updates);
-+ spin_unlock_irqrestore(&vxi->cvirt.load_lock, flags);
++ length += sprintf(buffer + length,
++ "BiasUptime:\t%lu.%02lu\n",
++ (unsigned long)cvirt->bias_uptime.tv_sec,
++ (cvirt->bias_uptime.tv_nsec / (NSEC_PER_SEC / 100)));
++
++ a = cvirt->load[0] + (FIXED_1 / 200);
++ b = cvirt->load[1] + (FIXED_1 / 200);
++ c = cvirt->load[2] + (FIXED_1 / 200);
++ length += sprintf(buffer + length,
++ "nr_threads:\t%d\n"
++ "nr_running:\t%d\n"
++ "nr_unintr:\t%d\n"
++ "nr_onhold:\t%d\n"
++ "load_updates:\t%d\n"
++ "loadavg:\t%d.%02d %d.%02d %d.%02d\n"
++ "total_forks:\t%d\n",
++ atomic_read(&cvirt->nr_threads),
++ atomic_read(&cvirt->nr_running),
++ atomic_read(&cvirt->nr_uninterruptible),
++ atomic_read(&cvirt->nr_onhold),
++ atomic_read(&cvirt->load_updates),
++ LOAD_INT(a), LOAD_FRAC(a),
++ LOAD_INT(b), LOAD_FRAC(b),
++ LOAD_INT(c), LOAD_FRAC(c),
++ atomic_read(&cvirt->total_forks));
++ return length;
+}
+
++static inline
++int vx_info_proc_cvirt_pc(struct _vx_cvirt_pc *cvirt_pc,
++ char *buffer, int cpu)
++{
++ int length = 0;
++ return length;
++}
+
++#endif /* _VX_CVIRT_PROC_H */
+--- linux-2.6.25.4/kernel/vserver/debug.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/debug.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,32 @@
+/*
-+ * Commands to do_syslog:
++ * kernel/vserver/debug.c
++ *
++ * Copyright (C) 2005-2007 Herbert Pötzl
++ *
++ * V0.01 vx_info dump support
+ *
-+ * 0 -- Close the log. Currently a NOP.
-+ * 1 -- Open the log. Currently a NOP.
-+ * 2 -- Read from the log.
-+ * 3 -- Read all messages remaining in the ring buffer.
-+ * 4 -- Read and clear all messages remaining in the ring buffer
-+ * 5 -- Clear ring buffer.
-+ * 6 -- Disable printk's to console
-+ * 7 -- Enable printk's to console
-+ * 8 -- Set level of messages printed to console
-+ * 9 -- Return number of unread characters in the log buffer
-+ * 10 -- Return size of the log buffer
+ */
-+int vx_do_syslog(int type, char __user *buf, int len)
-+{
-+ int error = 0;
-+ int do_clear = 0;
-+ struct vx_info *vxi = current->vx_info;
-+ struct _vx_syslog *log;
-+
-+ if (!vxi)
-+ return -EINVAL;
-+ log = &vxi->cvirt.syslog;
+
-+ switch (type) {
-+ case 0: /* Close log */
-+ case 1: /* Open log */
-+ break;
-+ case 2: /* Read from log */
-+ error = wait_event_interruptible(log->log_wait,
-+ (log->log_start - log->log_end));
-+ if (error)
-+ break;
-+ spin_lock_irq(&log->logbuf_lock);
-+ spin_unlock_irq(&log->logbuf_lock);
-+ break;
-+ case 4: /* Read/clear last kernel messages */
-+ do_clear = 1;
-+ /* fall through */
-+ case 3: /* Read last kernel messages */
-+ return 0;
++#include <linux/module.h>
+
-+ case 5: /* Clear ring buffer */
-+ return 0;
++#include <linux/vserver/context.h>
+
-+ case 6: /* Disable logging to console */
-+ case 7: /* Enable logging to console */
-+ case 8: /* Set level of messages printed to console */
-+ break;
+
-+ case 9: /* Number of chars in the log buffer */
-+ return 0;
-+ case 10: /* Size of the log buffer */
-+ return 0;
-+ default:
-+ error = -EINVAL;
-+ break;
++void dump_vx_info(struct vx_info *vxi, int level)
++{
++ printk("vx_info %p[#%d, %d.%d, %4x]\n", vxi, vxi->vx_id,
++ atomic_read(&vxi->vx_usecnt),
++ atomic_read(&vxi->vx_tasks),
++ vxi->vx_state);
++ if (level > 0) {
++ __dump_vx_limit(&vxi->limit);
++ __dump_vx_sched(&vxi->sched);
++ __dump_vx_cvirt(&vxi->cvirt);
++ __dump_vx_cacct(&vxi->cacct);
+ }
-+ return error;
++ printk("---\n");
+}
+
+
-+/* virtual host info names */
-+
-+static char *vx_vhi_name(struct vx_info *vxi, int id)
-+{
-+ struct nsproxy *nsproxy;
-+ struct uts_namespace *uts;
-+
++EXPORT_SYMBOL_GPL(dump_vx_info);
+
-+ if (id == VHIN_CONTEXT)
-+ return vxi->vx_name;
+--- linux-2.6.25.4/kernel/vserver/device.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/device.c 2008-04-21 12:35:24.000000000 -0400
+@@ -0,0 +1,443 @@
++/*
++ * linux/kernel/vserver/device.c
++ *
++ * Linux-VServer: Device Support
++ *
++ * Copyright (C) 2006 Herbert Pötzl
++ * Copyright (C) 2007 Daniel Hokka Zakrisson
++ *
++ * V0.01 device mapping basics
++ * V0.02 added defaults
++ *
++ */
+
-+ nsproxy = vxi->vx_nsproxy;
-+ if (!nsproxy)
-+ return NULL;
++#include <linux/slab.h>
++#include <linux/rcupdate.h>
++#include <linux/fs.h>
++#include <linux/namei.h>
++#include <linux/hash.h>
+
-+ uts = nsproxy->uts_ns;
-+ if (!uts)
-+ return NULL;
++#include <asm/errno.h>
++#include <asm/uaccess.h>
++#include <linux/vserver/base.h>
++#include <linux/vserver/debug.h>
++#include <linux/vserver/context.h>
++#include <linux/vserver/device.h>
++#include <linux/vserver/device_cmd.h>
+
-+ switch (id) {
-+ case VHIN_SYSNAME:
-+ return uts->name.sysname;
-+ case VHIN_NODENAME:
-+ return uts->name.nodename;
-+ case VHIN_RELEASE:
-+ return uts->name.release;
-+ case VHIN_VERSION:
-+ return uts->name.version;
-+ case VHIN_MACHINE:
-+ return uts->name.machine;
-+ case VHIN_DOMAINNAME:
-+ return uts->name.domainname;
-+ default:
-+ return NULL;
-+ }
-+ return NULL;
-+}
+
-+int vc_set_vhi_name(struct vx_info *vxi, void __user *data)
-+{
-+ struct vcmd_vhi_name_v0 vc_data;
-+ char *name;
++#define DMAP_HASH_BITS 4
+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
+
-+ name = vx_vhi_name(vxi, vc_data.field);
-+ if (!name)
-+ return -EINVAL;
++struct vs_mapping {
++ union {
++ struct hlist_node hlist;
++ struct list_head list;
++ } u;
++#define dm_hlist u.hlist
++#define dm_list u.list
++ xid_t xid;
++ dev_t device;
++ struct vx_dmap_target target;
++};
+
-+ memcpy(name, vc_data.name, 65);
-+ return 0;
-+}
+
-+int vc_get_vhi_name(struct vx_info *vxi, void __user *data)
-+{
-+ struct vcmd_vhi_name_v0 vc_data;
-+ char *name;
++static struct hlist_head dmap_main_hash[1 << DMAP_HASH_BITS];
+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
++static spinlock_t dmap_main_hash_lock = SPIN_LOCK_UNLOCKED;
+
-+ name = vx_vhi_name(vxi, vc_data.field);
-+ if (!name)
-+ return -EINVAL;
++static struct vx_dmap_target dmap_defaults[2] = {
++ { .flags = DATTR_OPEN },
++ { .flags = DATTR_OPEN },
++};
+
-+ memcpy(vc_data.name, name, 65);
-+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
-+ return -EFAULT;
++
++struct kmem_cache *dmap_cachep __read_mostly;
++
++int __init dmap_cache_init(void)
++{
++ dmap_cachep = kmem_cache_create("dmap_cache",
++ sizeof(struct vs_mapping), 0,
++ SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
+ return 0;
+}
+
++__initcall(dmap_cache_init);
+
-+int vc_virt_stat(struct vx_info *vxi, void __user *data)
++
++static inline unsigned int __hashval(dev_t dev, int bits)
+{
-+ struct vcmd_virt_stat_v0 vc_data;
-+ struct _vx_cvirt *cvirt = &vxi->cvirt;
-+ struct timespec uptime;
++ return hash_long((unsigned long)dev, bits);
++}
+
-+ do_posix_clock_monotonic_gettime(&uptime);
-+ set_normalized_timespec(&uptime,
-+ uptime.tv_sec - cvirt->bias_uptime.tv_sec,
-+ uptime.tv_nsec - cvirt->bias_uptime.tv_nsec);
+
-+ vc_data.offset = timeval_to_ns(&cvirt->bias_tv);
-+ vc_data.uptime = timespec_to_ns(&uptime);
-+ vc_data.nr_threads = atomic_read(&cvirt->nr_threads);
-+ vc_data.nr_running = atomic_read(&cvirt->nr_running);
-+ vc_data.nr_uninterruptible = atomic_read(&cvirt->nr_uninterruptible);
-+ vc_data.nr_onhold = atomic_read(&cvirt->nr_onhold);
-+ vc_data.nr_forks = atomic_read(&cvirt->total_forks);
-+ vc_data.load[0] = cvirt->load[0];
-+ vc_data.load[1] = cvirt->load[1];
-+ vc_data.load[2] = cvirt->load[2];
++/* __hash_mapping()
++ * add the mapping to the hash table
++ */
++static inline void __hash_mapping(struct vx_info *vxi, struct vs_mapping *vdm)
++{
++ spinlock_t *hash_lock = &dmap_main_hash_lock;
++ struct hlist_head *head, *hash = dmap_main_hash;
++ int device = vdm->device;
+
-+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
-+ return -EFAULT;
-+ return 0;
++ spin_lock(hash_lock);
++ vxdprintk(VXD_CBIT(misc, 8), "__hash_mapping: %p[#%d] %08x:%08x",
++ vxi, vxi ? vxi->vx_id : 0, device, vdm->target.target);
++
++ head = &hash[__hashval(device, DMAP_HASH_BITS)];
++ hlist_add_head(&vdm->dm_hlist, head);
++ spin_unlock(hash_lock);
+}
+
+
-+#ifdef CONFIG_VSERVER_VTIME
++static inline int __mode_to_default(umode_t mode)
++{
++ switch (mode) {
++ case S_IFBLK:
++ return 0;
++ case S_IFCHR:
++ return 1;
++ default:
++ BUG();
++ }
++}
+
-+/* virtualized time base */
+
-+void vx_gettimeofday(struct timeval *tv)
++/* __set_default()
++ * set a default
++ */
++static inline void __set_default(struct vx_info *vxi, umode_t mode,
++ struct vx_dmap_target *vdmt)
+{
-+ do_gettimeofday(tv);
-+ if (!vx_flags(VXF_VIRT_TIME, 0))
-+ return;
++ spinlock_t *hash_lock = &dmap_main_hash_lock;
++ spin_lock(hash_lock);
+
-+ tv->tv_sec += current->vx_info->cvirt.bias_tv.tv_sec;
-+ tv->tv_usec += current->vx_info->cvirt.bias_tv.tv_usec;
++ if (vxi)
++ vxi->dmap.targets[__mode_to_default(mode)] = *vdmt;
++ else
++ dmap_defaults[__mode_to_default(mode)] = *vdmt;
+
-+ if (tv->tv_usec >= USEC_PER_SEC) {
-+ tv->tv_sec++;
-+ tv->tv_usec -= USEC_PER_SEC;
-+ } else if (tv->tv_usec < 0) {
-+ tv->tv_sec--;
-+ tv->tv_usec += USEC_PER_SEC;
-+ }
++
++ spin_unlock(hash_lock);
++
++ vxdprintk(VXD_CBIT(misc, 8), "__set_default: %p[#%u] %08x %04x",
++ vxi, vxi ? vxi->vx_id : 0, vdmt->target, vdmt->flags);
+}
+
-+int vx_settimeofday(struct timespec *ts)
++
++/* __remove_default()
++ * remove a default
++ */
++static inline int __remove_default(struct vx_info *vxi, umode_t mode)
+{
-+ struct timeval tv;
++ spinlock_t *hash_lock = &dmap_main_hash_lock;
++ spin_lock(hash_lock);
+
-+ if (!vx_flags(VXF_VIRT_TIME, 0))
-+ return do_settimeofday(ts);
++ if (vxi)
++ vxi->dmap.targets[__mode_to_default(mode)].flags = 0;
++ else /* remove == reset */
++ dmap_defaults[__mode_to_default(mode)].flags = DATTR_OPEN | mode;
+
-+ do_gettimeofday(&tv);
-+ current->vx_info->cvirt.bias_tv.tv_sec =
-+ ts->tv_sec - tv.tv_sec;
-+ current->vx_info->cvirt.bias_tv.tv_usec =
-+ (ts->tv_nsec/NSEC_PER_USEC) - tv.tv_usec;
++ spin_unlock(hash_lock);
+ return 0;
+}
+
-+#endif
+
---- linux-2.6.22/kernel/vserver/cvirt_init.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/cvirt_init.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,69 @@
++/* __find_mapping()
++ * find a mapping in the hash table
++ *
++ * caller must hold hash_lock
++ */
++static inline int __find_mapping(xid_t xid, dev_t device, umode_t mode,
++ struct vs_mapping **local, struct vs_mapping **global)
++{
++ struct hlist_head *hash = dmap_main_hash;
++ struct hlist_head *head = &hash[__hashval(device, DMAP_HASH_BITS)];
++ struct hlist_node *pos;
++ struct vs_mapping *vdm;
+
++ *local = NULL;
++ if (global)
++ *global = NULL;
+
-+extern uint64_t vx_idle_jiffies(void);
++ hlist_for_each(pos, head) {
++ vdm = hlist_entry(pos, struct vs_mapping, dm_hlist);
+
-+static inline void vx_info_init_cvirt(struct _vx_cvirt *cvirt)
++ if ((vdm->device == device) &&
++ !((vdm->target.flags ^ mode) & S_IFMT)) {
++ if (vdm->xid == xid) {
++ *local = vdm;
++ return 1;
++ } else if (global && vdm->xid == 0)
++ *global = vdm;
++ }
++ }
++
++ if (global && *global)
++ return 0;
++ else
++ return -ENOENT;
++}
++
++
++/* __lookup_mapping()
++ * find a mapping and store the result in target and flags
++ */
++static inline int __lookup_mapping(struct vx_info *vxi,
++ dev_t device, dev_t *target, int *flags, umode_t mode)
+{
-+ uint64_t idle_jiffies = vx_idle_jiffies();
-+ uint64_t nsuptime;
++ spinlock_t *hash_lock = &dmap_main_hash_lock;
++ struct vs_mapping *vdm, *global;
++ struct vx_dmap_target *vdmt;
++ int ret = 0;
++ xid_t xid = vxi->vx_id;
++ int index;
+
-+ do_posix_clock_monotonic_gettime(&cvirt->bias_uptime);
-+ nsuptime = (unsigned long long)cvirt->bias_uptime.tv_sec
-+ * NSEC_PER_SEC + cvirt->bias_uptime.tv_nsec;
-+ cvirt->bias_clock = nsec_to_clock_t(nsuptime);
-+ cvirt->bias_tv.tv_sec = 0;
-+ cvirt->bias_tv.tv_usec = 0;
++ spin_lock(hash_lock);
++ if (__find_mapping(xid, device, mode, &vdm, &global) > 0) {
++ ret = 1;
++ vdmt = &vdm->target;
++ goto found;
++ }
+
-+ jiffies_to_timespec(idle_jiffies, &cvirt->bias_idle);
-+ atomic_set(&cvirt->nr_threads, 0);
-+ atomic_set(&cvirt->nr_running, 0);
-+ atomic_set(&cvirt->nr_uninterruptible, 0);
-+ atomic_set(&cvirt->nr_onhold, 0);
++ index = __mode_to_default(mode);
++ if (vxi && vxi->dmap.targets[index].flags) {
++ ret = 2;
++ vdmt = &vxi->dmap.targets[index];
++ } else if (global) {
++ ret = 3;
++ vdmt = &global->target;
++ goto found;
++ } else {
++ ret = 4;
++ vdmt = &dmap_defaults[index];
++ }
+
-+ spin_lock_init(&cvirt->load_lock);
-+ cvirt->load_last = jiffies;
-+ atomic_set(&cvirt->load_updates, 0);
-+ cvirt->load[0] = 0;
-+ cvirt->load[1] = 0;
-+ cvirt->load[2] = 0;
-+ atomic_set(&cvirt->total_forks, 0);
++found:
++ if (target && (vdmt->flags & DATTR_REMAP))
++ *target = vdmt->target;
++ else if (target)
++ *target = device;
++ if (flags)
++ *flags = vdmt->flags;
+
-+ spin_lock_init(&cvirt->syslog.logbuf_lock);
-+ init_waitqueue_head(&cvirt->syslog.log_wait);
-+ cvirt->syslog.log_start = 0;
-+ cvirt->syslog.log_end = 0;
-+ cvirt->syslog.con_start = 0;
-+ cvirt->syslog.logged_chars = 0;
-+}
++ spin_unlock(hash_lock);
+
-+static inline
-+void vx_info_init_cvirt_pc(struct _vx_cvirt_pc *cvirt_pc, int cpu)
-+{
-+ // cvirt_pc->cpustat = { 0 };
++ return ret;
+}
+
-+static inline void vx_info_exit_cvirt(struct _vx_cvirt *cvirt)
++
++/* __remove_mapping()
++ * remove a mapping from the hash table
++ */
++static inline int __remove_mapping(struct vx_info *vxi, dev_t device,
++ umode_t mode)
+{
-+ int value;
++ spinlock_t *hash_lock = &dmap_main_hash_lock;
++ struct vs_mapping *vdm = NULL;
++ int ret = 0;
+
-+ vxwprintk((value = atomic_read(&cvirt->nr_threads)),
-+ "!!! cvirt: %p[nr_threads] = %d on exit.",
-+ cvirt, value);
-+ vxwprintk((value = atomic_read(&cvirt->nr_running)),
-+ "!!! cvirt: %p[nr_running] = %d on exit.",
-+ cvirt, value);
-+ vxwprintk((value = atomic_read(&cvirt->nr_uninterruptible)),
-+ "!!! cvirt: %p[nr_uninterruptible] = %d on exit.",
-+ cvirt, value);
-+ vxwprintk((value = atomic_read(&cvirt->nr_onhold)),
-+ "!!! cvirt: %p[nr_onhold] = %d on exit.",
-+ cvirt, value);
-+ return;
++ spin_lock(hash_lock);
++
++ ret = __find_mapping((vxi ? vxi->vx_id : 0), device, mode, &vdm,
++ NULL);
++ vxdprintk(VXD_CBIT(misc, 8), "__remove_mapping: %p[#%d] %08x %04x",
++ vxi, vxi ? vxi->vx_id : 0, device, mode);
++ if (ret < 0)
++ goto out;
++ hlist_del(&vdm->dm_hlist);
++
++out:
++ spin_unlock(hash_lock);
++ if (vdm)
++ kmem_cache_free(dmap_cachep, vdm);
++ return ret;
+}
+
-+static inline
-+void vx_info_exit_cvirt_pc(struct _vx_cvirt_pc *cvirt_pc, int cpu)
++
++
++int vs_map_device(struct vx_info *vxi,
++ dev_t device, dev_t *target, umode_t mode)
+{
-+ return;
-+}
++ int ret, flags = DATTR_MASK;
+
---- linux-2.6.22/kernel/vserver/cvirt_proc.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/cvirt_proc.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,133 @@
-+#ifndef _VX_CVIRT_PROC_H
-+#define _VX_CVIRT_PROC_H
++ if (!vxi) {
++ if (target)
++ *target = device;
++ goto out;
++ }
++ ret = __lookup_mapping(vxi, device, target, &flags, mode);
++ vxdprintk(VXD_CBIT(misc, 8), "vs_map_device: %08x target: %08x flags: %04x mode: %04x mapped=%d",
++ device, target ? *target : 0, flags, mode, ret);
++out:
++ return (flags & DATTR_MASK);
++}
+
-+#include <linux/nsproxy.h>
-+#include <linux/mnt_namespace.h>
-+#include <linux/utsname.h>
-+#include <linux/ipc.h>
+
+
-+static inline
-+int vx_info_proc_nsproxy(struct nsproxy *nsproxy, char *buffer)
++static int do_set_mapping(struct vx_info *vxi,
++ dev_t device, dev_t target, int flags, umode_t mode)
+{
-+ struct mnt_namespace *ns;
-+ struct uts_namespace *uts;
-+ struct ipc_namespace *ipc;
-+ struct vfsmount *mnt;
-+ char *path, *root;
-+ int length = 0;
++ if (device) {
++ struct vs_mapping *new;
+
-+ if (!nsproxy)
-+ goto out;
++ new = kmem_cache_alloc(dmap_cachep, GFP_KERNEL);
++ if (!new)
++ return -ENOMEM;
+
-+ length += sprintf(buffer + length,
-+ "NSProxy:\t%p [%p,%p,%p]\n",
-+ nsproxy, nsproxy->mnt_ns,
-+ nsproxy->uts_ns, nsproxy->ipc_ns);
++ INIT_HLIST_NODE(&new->dm_hlist);
++ new->device = device;
++ new->target.target = target;
++ new->target.flags = flags | mode;
++ new->xid = (vxi ? vxi->vx_id : 0);
+
-+ ns = nsproxy->mnt_ns;
-+ if (!ns)
-+ goto skip_ns;
++ vxdprintk(VXD_CBIT(misc, 8), "do_set_mapping: %08x target: %08x flags: %04x", device, target, flags);
++ __hash_mapping(vxi, new);
++ } else {
++ struct vx_dmap_target new = {
++ .target = target,
++ .flags = flags | mode,
++ };
++ __set_default(vxi, mode, &new);
++ }
++ return 0;
++}
+
-+ path = kmalloc(PATH_MAX, GFP_KERNEL);
-+ if (!path)
-+ goto skip_ns;
+
-+ mnt = ns->root;
-+ root = d_path(mnt->mnt_root, mnt->mnt_parent, path, PATH_MAX - 2);
-+ length += sprintf(buffer + length,
-+ "Namespace:\t%p [#%u]\n"
-+ "RootPath:\t%s\n",
-+ ns, atomic_read(&ns->count),
-+ root);
-+ kfree(path);
-+skip_ns:
++static int do_unset_mapping(struct vx_info *vxi,
++ dev_t device, dev_t target, int flags, umode_t mode)
++{
++ int ret = -EINVAL;
+
-+ uts = nsproxy->uts_ns;
-+ if (!uts)
-+ goto skip_uts;
++ if (device) {
++ ret = __remove_mapping(vxi, device, mode);
++ if (ret < 0)
++ goto out;
++ } else {
++ ret = __remove_default(vxi, mode);
++ if (ret < 0)
++ goto out;
++ }
+
-+ length += sprintf(buffer + length,
-+ "SysName:\t%.*s\n"
-+ "NodeName:\t%.*s\n"
-+ "Release:\t%.*s\n"
-+ "Version:\t%.*s\n"
-+ "Machine:\t%.*s\n"
-+ "DomainName:\t%.*s\n",
-+ __NEW_UTS_LEN, uts->name.sysname,
-+ __NEW_UTS_LEN, uts->name.nodename,
-+ __NEW_UTS_LEN, uts->name.release,
-+ __NEW_UTS_LEN, uts->name.version,
-+ __NEW_UTS_LEN, uts->name.machine,
-+ __NEW_UTS_LEN, uts->name.domainname);
-+skip_uts:
++out:
++ return ret;
++}
++
++
++static inline int __user_device(const char __user *name, dev_t *dev,
++ umode_t *mode)
++{
++ struct nameidata nd;
++ int ret;
++
++ if (!name) {
++ *dev = 0;
++ return 0;
++ }
++ ret = user_path_walk_link(name, &nd);
++ if (ret)
++ return ret;
++ if (nd.path.dentry->d_inode) {
++ *dev = nd.path.dentry->d_inode->i_rdev;
++ *mode = nd.path.dentry->d_inode->i_mode;
++ }
++ path_put(&nd.path);
++ return 0;
++}
+
-+ ipc = nsproxy->ipc_ns;
-+ if (!ipc)
-+ goto skip_ipc;
++static inline int __mapping_mode(dev_t device, dev_t target,
++ umode_t device_mode, umode_t target_mode, umode_t *mode)
++{
++ if (device)
++ *mode = device_mode & S_IFMT;
++ else if (target)
++ *mode = target_mode & S_IFMT;
++ else
++ return -EINVAL;
+
-+ length += sprintf(buffer + length,
-+ "SEMS:\t\t%d %d %d %d %d\n"
-+ "MSG:\t\t%d %d %d\n"
-+ "SHM:\t\t%lu %lu %d %d\n",
-+ ipc->sem_ctls[0], ipc->sem_ctls[1],
-+ ipc->sem_ctls[2], ipc->sem_ctls[3],
-+ ipc->used_sems,
-+ ipc->msg_ctlmax, ipc->msg_ctlmnb, ipc->msg_ctlmni,
-+ (unsigned long)ipc->shm_ctlmax,
-+ (unsigned long)ipc->shm_ctlall,
-+ ipc->shm_ctlmni, ipc->shm_tot);
-+skip_ipc:
-+out:
-+ return length;
++ /* if both given, device and target mode have to match */
++ if (device && target &&
++ ((device_mode ^ target_mode) & S_IFMT))
++ return -EINVAL;
++ return 0;
+}
+
+
-+#include <linux/sched.h>
++static inline int do_mapping(struct vx_info *vxi, const char __user *device_path,
++ const char __user *target_path, int flags, int set)
++{
++ dev_t device = ~0, target = ~0;
++ umode_t device_mode = 0, target_mode = 0, mode;
++ int ret;
+
-+#define LOAD_INT(x) ((x) >> FSHIFT)
-+#define LOAD_FRAC(x) LOAD_INT(((x) & (FIXED_1 - 1)) * 100)
++ ret = __user_device(device_path, &device, &device_mode);
++ if (ret)
++ return ret;
++ ret = __user_device(target_path, &target, &target_mode);
++ if (ret)
++ return ret;
+
-+static inline
-+int vx_info_proc_cvirt(struct _vx_cvirt *cvirt, char *buffer)
++ ret = __mapping_mode(device, target,
++ device_mode, target_mode, &mode);
++ if (ret)
++ return ret;
++
++ if (set)
++ return do_set_mapping(vxi, device, target,
++ flags, mode);
++ else
++ return do_unset_mapping(vxi, device, target,
++ flags, mode);
++}
++
++
++int vc_set_mapping(struct vx_info *vxi, void __user *data)
+{
-+ int length = 0;
-+ int a, b, c;
++ struct vcmd_set_mapping_v0 vc_data;
+
-+ length += sprintf(buffer + length,
-+ "BiasUptime:\t%lu.%02lu\n",
-+ (unsigned long)cvirt->bias_uptime.tv_sec,
-+ (cvirt->bias_uptime.tv_nsec / (NSEC_PER_SEC / 100)));
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
+
-+ a = cvirt->load[0] + (FIXED_1 / 200);
-+ b = cvirt->load[1] + (FIXED_1 / 200);
-+ c = cvirt->load[2] + (FIXED_1 / 200);
-+ length += sprintf(buffer + length,
-+ "nr_threads:\t%d\n"
-+ "nr_running:\t%d\n"
-+ "nr_unintr:\t%d\n"
-+ "nr_onhold:\t%d\n"
-+ "load_updates:\t%d\n"
-+ "loadavg:\t%d.%02d %d.%02d %d.%02d\n"
-+ "total_forks:\t%d\n",
-+ atomic_read(&cvirt->nr_threads),
-+ atomic_read(&cvirt->nr_running),
-+ atomic_read(&cvirt->nr_uninterruptible),
-+ atomic_read(&cvirt->nr_onhold),
-+ atomic_read(&cvirt->load_updates),
-+ LOAD_INT(a), LOAD_FRAC(a),
-+ LOAD_INT(b), LOAD_FRAC(b),
-+ LOAD_INT(c), LOAD_FRAC(c),
-+ atomic_read(&cvirt->total_forks));
-+ return length;
++ return do_mapping(vxi, vc_data.device, vc_data.target,
++ vc_data.flags, 1);
+}
+
-+static inline
-+int vx_info_proc_cvirt_pc(struct _vx_cvirt_pc *cvirt_pc,
-+ char *buffer, int cpu)
++int vc_unset_mapping(struct vx_info *vxi, void __user *data)
+{
-+ int length = 0;
-+ return length;
++ struct vcmd_set_mapping_v0 vc_data;
++
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ return do_mapping(vxi, vc_data.device, vc_data.target,
++ vc_data.flags, 0);
+}
+
-+#endif /* _VX_CVIRT_PROC_H */
---- linux-2.6.22/kernel/vserver/debug.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/debug.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,35 @@
-+/*
-+ * kernel/vserver/debug.c
-+ *
-+ * Copyright (C) 2005-2007 Herbert Pötzl
-+ *
-+ * V0.01 vx_info dump support
-+ *
-+ */
+
-+#include <linux/errno.h>
-+#include <linux/kernel.h>
-+#include <linux/module.h>
-+#include <linux/vs_base.h>
++#ifdef CONFIG_COMPAT
+
-+#include <linux/vserver/context.h>
++int vc_set_mapping_x32(struct vx_info *vxi, void __user *data)
++{
++ struct vcmd_set_mapping_v0_x32 vc_data;
+
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
+
-+void dump_vx_info(struct vx_info *vxi, int level)
++ return do_mapping(vxi, compat_ptr(vc_data.device_ptr),
++ compat_ptr(vc_data.target_ptr), vc_data.flags, 1);
++}
++
++int vc_unset_mapping_x32(struct vx_info *vxi, void __user *data)
+{
-+ printk("vx_info %p[#%d, %d.%d, %4x]\n", vxi, vxi->vx_id,
-+ atomic_read(&vxi->vx_usecnt),
-+ atomic_read(&vxi->vx_tasks),
-+ vxi->vx_state);
-+ if (level > 0) {
-+ __dump_vx_limit(&vxi->limit);
-+ __dump_vx_sched(&vxi->sched);
-+ __dump_vx_cvirt(&vxi->cvirt);
-+ __dump_vx_cacct(&vxi->cacct);
-+ }
-+ printk("---\n");
++ struct vcmd_set_mapping_v0_x32 vc_data;
++
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ return do_mapping(vxi, compat_ptr(vc_data.device_ptr),
++ compat_ptr(vc_data.target_ptr), vc_data.flags, 0);
+}
+
++#endif /* CONFIG_COMPAT */
+
-+EXPORT_SYMBOL_GPL(dump_vx_info);
+
---- linux-2.6.22/kernel/vserver/dlimit.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/dlimit.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,527 @@
+--- linux-2.6.25.4/kernel/vserver/dlimit.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/dlimit.c 2008-04-21 12:36:09.000000000 -0400
+@@ -0,0 +1,521 @@
+/*
+ * linux/kernel/vserver/dlimit.c
+ *
@@ -19877,18 +19658,12 @@
+ *
+ */
+
-+#include <linux/fs.h>
-+// #include <linux/mnt_namespace.h>
-+#include <linux/namei.h>
+#include <linux/statfs.h>
-+#include <linux/compat.h>
-+#include <linux/vserver/switch.h>
-+#include <linux/vs_context.h>
++#include <linux/sched.h>
+#include <linux/vs_tag.h>
+#include <linux/vs_dlimit.h>
+#include <linux/vserver/dlimit_cmd.h>
+
-+#include <asm/errno.h>
+#include <asm/uaccess.h>
+
+/* __alloc_dl_info()
@@ -20059,9 +19834,9 @@
+ struct dl_info *dli;
+
+ ret = -EINVAL;
-+ if (!nd.dentry->d_inode)
++ if (!nd.path.dentry->d_inode)
+ goto out_release;
-+ if (!(sb = nd.dentry->d_inode->i_sb))
++ if (!(sb = nd.path.dentry->d_inode->i_sb))
+ goto out_release;
+
+ if (add) {
@@ -20088,7 +19863,7 @@
+ if (add && dli)
+ __dealloc_dl_info(dli);
+ out_release:
-+ path_release(&nd);
++ path_put(&nd.path);
+ }
+ return ret;
+}
@@ -20155,9 +19930,9 @@
+ struct dl_info *dli;
+
+ ret = -EINVAL;
-+ if (!nd.dentry->d_inode)
++ if (!nd.path.dentry->d_inode)
+ goto out_release;
-+ if (!(sb = nd.dentry->d_inode->i_sb))
++ if (!(sb = nd.path.dentry->d_inode->i_sb))
+ goto out_release;
+ if ((reserved != CDLIM_KEEP &&
+ reserved > 100) ||
@@ -20197,7 +19972,7 @@
+ ret = 0;
+
+ out_release:
-+ path_release(&nd);
++ path_put(&nd.path);
+ }
+ return ret;
+}
@@ -20248,9 +20023,9 @@
+ struct dl_info *dli;
+
+ ret = -EINVAL;
-+ if (!nd.dentry->d_inode)
++ if (!nd.path.dentry->d_inode)
+ goto out_release;
-+ if (!(sb = nd.dentry->d_inode->i_sb))
++ if (!(sb = nd.path.dentry->d_inode->i_sb))
+ goto out_release;
+
+ ret = -ESRCH;
@@ -20275,7 +20050,7 @@
+
+ ret = 0;
+ out_release:
-+ path_release(&nd);
++ path_put(&nd.path);
+ }
+ return ret;
+}
@@ -20392,9 +20167,9 @@
+EXPORT_SYMBOL_GPL(locate_dl_info);
+EXPORT_SYMBOL_GPL(rcu_free_dl_info);
+
---- linux-2.6.22/kernel/vserver/helper.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/helper.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,208 @@
+--- linux-2.6.25.4/kernel/vserver/helper.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/helper.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,199 @@
+/*
+ * linux/kernel/vserver/helper.c
+ *
@@ -20406,17 +20181,12 @@
+ *
+ */
+
-+#include <linux/errno.h>
+#include <linux/kmod.h>
-+#include <linux/sched.h>
+#include <linux/reboot.h>
+#include <linux/vs_context.h>
+#include <linux/vs_network.h>
+#include <linux/vserver/signal.h>
+
-+#include <asm/uaccess.h>
-+#include <asm/unistd.h>
-+
+
+char vshelper_path[255] = "/sbin/vshelper";
+
@@ -20492,11 +20262,7 @@
+ return 0;
+ }
+
-+#ifndef CONFIG_VSERVER_LEGACY
-+ ret = do_vshelper(vshelper_path, argv, envp, 1);
-+#else
+ ret = do_vshelper(vshelper_path, argv, envp, 0);
-+#endif
+ vxi->vx_state &= ~VXS_HELPER;
+ __wakeup_vx_info(vxi);
+ return (ret) ? -EPERM : 0;
@@ -20603,9 +20369,9 @@
+ return do_vshelper(vshelper_path, argv, envp, 1);
+}
+
---- linux-2.6.22/kernel/vserver/history.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/history.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,264 @@
+--- linux-2.6.25.4/kernel/vserver/history.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/history.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,258 @@
+/*
+ * kernel/vserver/history.c
+ *
@@ -20619,14 +20385,8 @@
+ *
+ */
+
-+#include <linux/errno.h>
+#include <linux/module.h>
-+#include <linux/types.h>
-+#include <linux/ctype.h>
-+
+#include <asm/uaccess.h>
-+#include <asm/atomic.h>
-+#include <asm/unistd.h>
+
+#include <linux/vserver/context.h>
+#include <linux/vserver/debug.h>
@@ -20827,52 +20587,280 @@
+ return ret ? ret : (*index < end);
+}
+
-+int vc_read_history(uint32_t id, void __user *data)
++int vc_read_history(uint32_t id, void __user *data)
++{
++ struct vcmd_read_history_v0 vc_data;
++ int ret;
++
++ if (id >= NR_CPUS)
++ return -EINVAL;
++
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ ret = do_read_history((struct __user _vx_hist_entry *)vc_data.data,
++ id, &vc_data.index, &vc_data.count);
++
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ return -EFAULT;
++ return ret;
++}
++
++#ifdef CONFIG_COMPAT
++
++int vc_read_history_x32(uint32_t id, void __user *data)
++{
++ struct vcmd_read_history_v0_x32 vc_data;
++ int ret;
++
++ if (id >= NR_CPUS)
++ return -EINVAL;
++
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ ret = do_read_history((struct __user _vx_hist_entry *)
++ compat_ptr(vc_data.data_ptr),
++ id, &vc_data.index, &vc_data.count);
++
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ return -EFAULT;
++ return ret;
++}
++
++#endif /* CONFIG_COMPAT */
++
+--- linux-2.6.25.4/kernel/vserver/inet.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/inet.c 2008-04-21 12:49:53.000000000 -0400
+@@ -0,0 +1,225 @@
++
++#include <linux/in.h>
++#include <linux/inetdevice.h>
++#include <linux/vs_inet.h>
++#include <linux/vs_inet6.h>
++#include <linux/vserver/debug.h>
++#include <net/route.h>
++#include <net/addrconf.h>
++
++
++int nx_v4_addr_conflict(struct nx_info *nxi1, struct nx_info *nxi2)
++{
++ int ret = 0;
++
++ if (!nxi1 || !nxi2 || nxi1 == nxi2)
++ ret = 1;
++ else {
++ struct nx_addr_v4 *ptr;
++
++ for (ptr = &nxi1->v4; ptr; ptr = ptr->next) {
++ if (v4_nx_addr_in_nx_info(nxi2, ptr, -1)) {
++ ret = 1;
++ break;
++ }
++ }
++ }
++
++ vxdprintk(VXD_CBIT(net, 2),
++ "nx_v4_addr_conflict(%p,%p): %d",
++ nxi1, nxi2, ret);
++
++ return ret;
++}
++
++
++#ifdef CONFIG_IPV6
++
++int nx_v6_addr_conflict(struct nx_info *nxi1, struct nx_info *nxi2)
++{
++ int ret = 0;
++
++ if (!nxi1 || !nxi2 || nxi1 == nxi2)
++ ret = 1;
++ else {
++ struct nx_addr_v6 *ptr;
++
++ for (ptr = &nxi1->v6; ptr; ptr = ptr->next) {
++ if (v6_nx_addr_in_nx_info(nxi2, ptr, -1)) {
++ ret = 1;
++ break;
++ }
++ }
++ }
++
++ vxdprintk(VXD_CBIT(net, 2),
++ "nx_v6_addr_conflict(%p,%p): %d",
++ nxi1, nxi2, ret);
++
++ return ret;
++}
++
++#endif
++
++int v4_dev_in_nx_info(struct net_device *dev, struct nx_info *nxi)
++{
++ struct in_device *in_dev;
++ struct in_ifaddr **ifap;
++ struct in_ifaddr *ifa;
++ int ret = 0;
++
++ if (!dev)
++ goto out;
++ in_dev = in_dev_get(dev);
++ if (!in_dev)
++ goto out;
++
++ for (ifap = &in_dev->ifa_list; (ifa = *ifap) != NULL;
++ ifap = &ifa->ifa_next) {
++ if (v4_addr_in_nx_info(nxi, ifa->ifa_local, NXA_MASK_SHOW)) {
++ ret = 1;
++ break;
++ }
++ }
++ in_dev_put(in_dev);
++out:
++ return ret;
++}
++
++
++#ifdef CONFIG_IPV6
++
++int v6_dev_in_nx_info(struct net_device *dev, struct nx_info *nxi)
++{
++ struct inet6_dev *in_dev;
++ struct inet6_ifaddr **ifap;
++ struct inet6_ifaddr *ifa;
++ int ret = 0;
++
++ if (!dev)
++ goto out;
++ in_dev = in6_dev_get(dev);
++ if (!in_dev)
++ goto out;
++
++ for (ifap = &in_dev->addr_list; (ifa = *ifap) != NULL;
++ ifap = &ifa->if_next) {
++ if (v6_addr_in_nx_info(nxi, &ifa->addr, -1)) {
++ ret = 1;
++ break;
++ }
++ }
++ in6_dev_put(in_dev);
++out:
++ return ret;
++}
++
++#endif
++
++int dev_in_nx_info(struct net_device *dev, struct nx_info *nxi)
+{
-+ struct vcmd_read_history_v0 vc_data;
-+ int ret;
++ int ret = 1;
+
-+ if (id >= NR_CPUS)
-+ return -EINVAL;
++ if (!nxi)
++ goto out;
++ if (nxi->v4.type && v4_dev_in_nx_info(dev, nxi))
++ goto out;
++#ifdef CONFIG_IPV6
++ ret = 2;
++ if (nxi->v6.type && v6_dev_in_nx_info(dev, nxi))
++ goto out;
++#endif
++ ret = 0;
++out:
++ vxdprintk(VXD_CBIT(net, 3),
++ "dev_in_nx_info(%p,%p[#%d]) = %d",
++ dev, nxi, nxi ? nxi->nx_id : 0, ret);
++ return ret;
++}
+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
++int ip_v4_find_src(struct net *net, struct nx_info *nxi,
++ struct rtable **rp, struct flowi *fl)
++{
++ if (!nxi)
++ return 0;
+
-+ ret = do_read_history((struct __user _vx_hist_entry *)vc_data.data,
-+ id, &vc_data.index, &vc_data.count);
++ /* FIXME: handle lback only case */
++ if (!NX_IPV4(nxi))
++ return -EPERM;
+
-+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
-+ return -EFAULT;
-+ return ret;
-+}
++ vxdprintk(VXD_CBIT(net, 4),
++ "ip_v4_find_src(%p[#%u]) " NIPQUAD_FMT " -> " NIPQUAD_FMT,
++ nxi, nxi ? nxi->nx_id : 0,
++ NIPQUAD(fl->fl4_src), NIPQUAD(fl->fl4_dst));
+
-+#ifdef CONFIG_COMPAT
++ /* single IP is unconditional */
++ if (nx_info_flags(nxi, NXF_SINGLE_IP, 0) &&
++ (fl->fl4_src == INADDR_ANY))
++ fl->fl4_src = nxi->v4.ip[0].s_addr;
++
++ if (fl->fl4_src == INADDR_ANY) {
++ struct nx_addr_v4 *ptr;
++ __be32 found = 0;
++ int err;
+
-+int vc_read_history_x32(uint32_t id, void __user *data)
-+{
-+ struct vcmd_read_history_v0_x32 vc_data;
-+ int ret;
++ err = __ip_route_output_key(net, rp, fl);
++ if (!err) {
++ found = (*rp)->rt_src;
++ ip_rt_put(*rp);
++ vxdprintk(VXD_CBIT(net, 4),
++ "ip_v4_find_src(%p[#%u]) rok[%u]: " NIPQUAD_FMT,
++ nxi, nxi ? nxi->nx_id : 0, fl->oif, NIPQUAD(found));
++ if (v4_addr_in_nx_info(nxi, found, NXA_MASK_BIND))
++ goto found;
++ }
+
-+ if (id >= NR_CPUS)
-+ return -EINVAL;
++ for (ptr = &nxi->v4; ptr; ptr = ptr->next) {
++ __be32 primary = ptr->ip[0].s_addr;
++ __be32 mask = ptr->mask.s_addr;
++ __be32 neta = primary & mask;
++
++ vxdprintk(VXD_CBIT(net, 4), "ip_v4_find_src(%p[#%u]) chk: "
++ NIPQUAD_FMT "/" NIPQUAD_FMT "/" NIPQUAD_FMT,
++ nxi, nxi ? nxi->nx_id : 0, NIPQUAD(primary),
++ NIPQUAD(mask), NIPQUAD(neta));
++ if ((found & mask) != neta)
++ continue;
+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
++ fl->fl4_src = primary;
++ err = __ip_route_output_key(net, rp, fl);
++ vxdprintk(VXD_CBIT(net, 4),
++ "ip_v4_find_src(%p[#%u]) rok[%u]: " NIPQUAD_FMT,
++ nxi, nxi ? nxi->nx_id : 0, fl->oif, NIPQUAD(primary));
++ if (!err) {
++ found = (*rp)->rt_src;
++ ip_rt_put(*rp);
++ if (found == primary)
++ goto found;
++ }
++ }
++ /* still no source ip? */
++ found = IN_LOOPBACK(fl->fl4_dst)
++ ? IPI_LOOPBACK : nxi->v4.ip[0].s_addr;
++ found:
++ /* assign src ip to flow */
++ fl->fl4_src = found;
+
-+ ret = do_read_history((struct __user _vx_hist_entry *)
-+ compat_ptr(vc_data.data_ptr),
-+ id, &vc_data.index, &vc_data.count);
++ } else {
++ if (!v4_addr_in_nx_info(nxi, fl->fl4_src, NXA_MASK_BIND))
++ return -EPERM;
++ }
+
-+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
-+ return -EFAULT;
-+ return ret;
++ if (nx_info_flags(nxi, NXF_LBACK_REMAP, 0)) {
++ if (IN_LOOPBACK(fl->fl4_dst))
++ fl->fl4_dst = nxi->v4_lback.s_addr;
++ if (IN_LOOPBACK(fl->fl4_src))
++ fl->fl4_src = nxi->v4_lback.s_addr;
++ } else if (IN_LOOPBACK(fl->fl4_dst) &&
++ !nx_info_flags(nxi, NXF_LBACK_ALLOW, 0))
++ return -EPERM;
++
++ return 0;
+}
+
-+#endif /* CONFIG_COMPAT */
++EXPORT_SYMBOL_GPL(ip_v4_find_src);
+
---- linux-2.6.22/kernel/vserver/init.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/init.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,47 @@
+--- linux-2.6.25.4/kernel/vserver/init.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/init.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,45 @@
+/*
+ * linux/kernel/init.c
+ *
@@ -20884,9 +20872,7 @@
+ *
+ */
+
-+#include <linux/errno.h>
+#include <linux/init.h>
-+#include <linux/module.h>
+
+int vserver_register_sysctl(void);
+void vserver_unregister_sysctl(void);
@@ -20920,9 +20906,9 @@
+module_init(init_vserver);
+module_exit(exit_vserver);
+
---- linux-2.6.22/kernel/vserver/inode.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/inode.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,369 @@
+--- linux-2.6.25.4/kernel/vserver/inode.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/inode.c 2008-04-21 16:52:16.000000000 -0400
+@@ -0,0 +1,409 @@
+/*
+ * linux/kernel/vserver/inode.c
+ *
@@ -20931,22 +20917,22 @@
+ * Copyright (C) 2004-2007 Herbert Pötzl
+ *
+ * V0.01 separated from vcontext V0.05
++ * V0.02 moved to tag (instead of xid)
+ *
+ */
+
-+#include <linux/sched.h>
++#include <linux/tty.h>
+#include <linux/proc_fs.h>
+#include <linux/devpts_fs.h>
-+#include <linux/namei.h>
++#include <linux/fs.h>
++#include <linux/file.h>
+#include <linux/mount.h>
+#include <linux/parser.h>
-+#include <linux/compat.h>
+#include <linux/vserver/inode.h>
+#include <linux/vserver/inode_cmd.h>
+#include <linux/vs_base.h>
+#include <linux/vs_tag.h>
+
-+#include <asm/errno.h>
+#include <asm/uaccess.h>
+
+
@@ -20995,10 +20981,10 @@
+ return 0;
+}
+
-+int vc_get_iattr(uint32_t id, void __user *data)
++int vc_get_iattr(void __user *data)
+{
+ struct nameidata nd;
-+ struct vcmd_ctx_iattr_v1 vc_data = { .xid = -1 };
++ struct vcmd_ctx_iattr_v1 vc_data = { .tag = -1 };
+ int ret;
+
+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
@@ -21006,9 +20992,9 @@
+
+ ret = user_path_walk_link(vc_data.name, &nd);
+ if (!ret) {
-+ ret = __vc_get_iattr(nd.dentry->d_inode,
-+ &vc_data.xid, &vc_data.flags, &vc_data.mask);
-+ path_release(&nd);
++ ret = __vc_get_iattr(nd.path.dentry->d_inode,
++ &vc_data.tag, &vc_data.flags, &vc_data.mask);
++ path_put(&nd.path);
+ }
+ if (ret)
+ return ret;
@@ -21020,22 +21006,20 @@
+
+#ifdef CONFIG_COMPAT
+
-+int vc_get_iattr_x32(uint32_t id, void __user *data)
++int vc_get_iattr_x32(void __user *data)
+{
+ struct nameidata nd;
-+ struct vcmd_ctx_iattr_v1_x32 vc_data = { .xid = -1 };
++ struct vcmd_ctx_iattr_v1_x32 vc_data = { .tag = -1 };
+ int ret;
+
-+ if (!vx_check(0, VS_ADMIN))
-+ return -ENOSYS;
+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
+ return -EFAULT;
+
+ ret = user_path_walk_link(compat_ptr(vc_data.name_ptr), &nd);
+ if (!ret) {
-+ ret = __vc_get_iattr(nd.dentry->d_inode,
-+ &vc_data.xid, &vc_data.flags, &vc_data.mask);
-+ path_release(&nd);
++ ret = __vc_get_iattr(nd.path.dentry->d_inode,
++ &vc_data.tag, &vc_data.flags, &vc_data.mask);
++ path_put(&nd.path);
+ }
+ if (ret)
+ return ret;
@@ -21048,6 +21032,30 @@
+#endif /* CONFIG_COMPAT */
+
+
++int vc_fget_iattr(uint32_t fd, void __user *data)
++{
++ struct file *filp;
++ struct vcmd_ctx_fiattr_v0 vc_data = { .tag = -1 };
++ int ret;
++
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ filp = fget(fd);
++ if (!filp || !filp->f_dentry || !filp->f_dentry->d_inode)
++ return -EBADF;
++
++ ret = __vc_get_iattr(filp->f_dentry->d_inode,
++ &vc_data.tag, &vc_data.flags, &vc_data.mask);
++
++ fput(filp);
++
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ ret = -EFAULT;
++ return ret;
++}
++
++
+static int __vc_set_iattr(struct dentry *de, uint32_t *tag, uint32_t *flags, uint32_t *mask)
+{
+ struct inode *in = de->d_inode;
@@ -21124,7 +21132,7 @@
+ return error;
+}
+
-+int vc_set_iattr(uint32_t id, void __user *data)
++int vc_set_iattr(void __user *data)
+{
+ struct nameidata nd;
+ struct vcmd_ctx_iattr_v1 vc_data;
@@ -21137,9 +21145,9 @@
+
+ ret = user_path_walk_link(vc_data.name, &nd);
+ if (!ret) {
-+ ret = __vc_set_iattr(nd.dentry,
-+ &vc_data.xid, &vc_data.flags, &vc_data.mask);
-+ path_release(&nd);
++ ret = __vc_set_iattr(nd.path.dentry,
++ &vc_data.tag, &vc_data.flags, &vc_data.mask);
++ path_put(&nd.path);
+ }
+
+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
@@ -21149,7 +21157,7 @@
+
+#ifdef CONFIG_COMPAT
+
-+int vc_set_iattr_x32(uint32_t id, void __user *data)
++int vc_set_iattr_x32(void __user *data)
+{
+ struct nameidata nd;
+ struct vcmd_ctx_iattr_v1_x32 vc_data;
@@ -21162,9 +21170,9 @@
+
+ ret = user_path_walk_link(compat_ptr(vc_data.name_ptr), &nd);
+ if (!ret) {
-+ ret = __vc_set_iattr(nd.dentry,
-+ &vc_data.xid, &vc_data.flags, &vc_data.mask);
-+ path_release(&nd);
++ ret = __vc_set_iattr(nd.path.dentry,
++ &vc_data.tag, &vc_data.flags, &vc_data.mask);
++ path_put(&nd.path);
+ }
+
+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
@@ -21174,63 +21182,63 @@
+
+#endif /* CONFIG_COMPAT */
+
-+#ifdef CONFIG_VSERVER_LEGACY
++int vc_fset_iattr(uint32_t fd, void __user *data)
++{
++ struct file *filp;
++ struct vcmd_ctx_fiattr_v0 vc_data;
++ int ret;
+
-+#define PROC_DYNAMIC_FIRST 0xF0000000UL
++ if (!capable(CAP_LINUX_IMMUTABLE))
++ return -EPERM;
++ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
+
-+int vx_proc_ioctl(struct inode *inode, struct file *filp,
-+ unsigned int cmd, unsigned long arg)
-+{
-+ struct proc_dir_entry *entry;
-+ int error = 0;
-+ int flags;
++ filp = fget(fd);
++ if (!filp || !filp->f_dentry || !filp->f_dentry->d_inode)
++ return -EBADF;
+
-+ if (inode->i_ino < PROC_DYNAMIC_FIRST)
-+ return -ENOTTY;
++ ret = __vc_set_iattr(filp->f_dentry, &vc_data.tag,
++ &vc_data.flags, &vc_data.mask);
+
-+ entry = PROC_I(inode)->pde;
-+ if (!entry)
-+ return -ENOTTY;
-+
-+ switch(cmd) {
-+ case FIOC_GETXFLG: {
-+ /* fixme: if stealth, return -ENOTTY */
-+ error = -EPERM;
-+ flags = entry->vx_flags;
-+ if (capable(CAP_CONTEXT))
-+ error = put_user(flags, (int __user *) arg);
-+ break;
-+ }
-+ case FIOC_SETXFLG: {
-+ /* fixme: if stealth, return -ENOTTY */
-+ error = -EPERM;
-+ if (!capable(CAP_CONTEXT))
-+ break;
-+ error = -EROFS;
-+ if (IS_RDONLY(inode))
-+ break;
-+ error = -EFAULT;
-+ if (get_user(flags, (int __user *) arg))
-+ break;
-+ error = 0;
-+ entry->vx_flags = flags;
-+ break;
-+ }
-+ default:
-+ return -ENOTTY;
-+ }
-+ return error;
++ fput(filp);
++
++ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
++ return -EFAULT;
++ return ret;
+}
-+#endif /* CONFIG_VSERVER_LEGACY */
+
++
++enum { Opt_notagcheck, Opt_tag, Opt_notag, Opt_tagid, Opt_err };
++
++static match_table_t tokens = {
++ {Opt_notagcheck, "notagcheck"},
+#ifdef CONFIG_PROPAGATE
++ {Opt_notag, "notag"},
++ {Opt_tag, "tag"},
++ {Opt_tagid, "tagid=%u"},
++#endif
++ {Opt_err, NULL}
++};
+
-+int dx_parse_tag(char *string, tag_t *tag, int remove)
++
++static void __dx_parse_remove(char *string, char *opt)
++{
++ char *p = strstr(string, opt);
++ char *q = p;
++
++ if (p) {
++ while (*q != '\0' && *q != ',')
++ q++;
++ while (*q)
++ *p++ = *q++;
++ while (*p)
++ *p++ = '\0';
++ }
++}
++
++static inline
++int __dx_parse_tag(char *string, tag_t *tag, int remove)
+{
-+ static match_table_t tokens = {
-+ {1, "tagid=%u"},
-+ {0, NULL}
-+ };
+ substring_t args[MAX_OPT_ARGS];
+ int token, option = 0;
+
@@ -21238,29 +21246,47 @@
+ return 0;
+
+ token = match_token(string, tokens, args);
-+ if (token && tag && !match_int(args, &option))
-+ *tag = option;
+
+ vxdprintk(VXD_CBIT(tag, 7),
+ "dx_parse_tag(»%s«): %d:#%d",
+ string, token, option);
+
-+ if ((token == 1) && remove) {
-+ char *p = strstr(string, "tagid=");
-+ char *q = p;
-+
-+ if (p) {
-+ while (*q != '\0' && *q != ',')
-+ q++;
-+ while (*q)
-+ *p++ = *q++;
-+ while (*p)
-+ *p++ = '\0';
-+ }
++ switch (token) {
++ case Opt_tag:
++ if (tag)
++ *tag = 0;
++ if (remove)
++ __dx_parse_remove(string, "tag");
++ return MNT_TAGID;
++ case Opt_notag:
++ if (remove)
++ __dx_parse_remove(string, "notag");
++ return MNT_NOTAG;
++ case Opt_notagcheck:
++ if (remove)
++ __dx_parse_remove(string, "notagcheck");
++ return MNT_NOTAGCHECK;
++ case Opt_tagid:
++ if (tag && !match_int(args, &option))
++ *tag = option;
++ if (remove)
++ __dx_parse_remove(string, "tagid");
++ return MNT_TAGID;
+ }
-+ return token;
++ return 0;
++}
++
++int dx_parse_tag(char *string, tag_t *tag, int remove)
++{
++ int retval, flags = 0;
++
++ while ((retval = __dx_parse_tag(string, tag, remove)))
++ flags |= retval;
++ return flags;
+}
+
++#ifdef CONFIG_PROPAGATE
++
+void __dx_propagate_tag(struct nameidata *nd, struct inode *inode)
+{
+ tag_t new_tag = 0;
@@ -21269,7 +21295,7 @@
+
+ if (!nd)
+ return;
-+ mnt = nd->mnt;
++ mnt = nd->path.mnt;
+ if (!mnt)
+ return;
+
@@ -21292,213 +21318,263 @@
+
+#endif /* CONFIG_PROPAGATE */
+
---- linux-2.6.22/kernel/vserver/legacy.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/legacy.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,114 @@
-+/*
-+ * linux/kernel/vserver/legacy.c
-+ *
-+ * Virtual Server: Legacy Funtions
-+ *
-+ * Copyright (C) 2001-2003 Jacques Gelinas
-+ * Copyright (C) 2003-2007 Herbert Pötzl
-+ *
-+ * V0.01 broken out from vcontext.c V0.05
-+ * V0.02 updated to spaces *sigh*
-+ *
-+ */
+--- linux-2.6.25.4/kernel/vserver/Kconfig 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/Kconfig 2008-05-21 15:02:48.000000000 -0400
+@@ -0,0 +1,252 @@
++#
++# Linux VServer configuration
++#
+
-+#include <linux/sched.h>
-+#include <linux/vs_context.h>
-+#include <linux/vs_network.h>
-+#include <linux/vserver/legacy.h>
-+#include <linux/vserver/space.h>
-+// #include <linux/mnt_namespace.h>
++menu "Linux VServer"
+
-+#include <asm/errno.h>
-+#include <asm/uaccess.h>
++config VSERVER_AUTO_LBACK
++ bool "Automatically Assign Loopback IP"
++ default y
++ help
++ Automatically assign a guest specific loopback
++ IP and add it to the kernel network stack on
++ startup.
+
++config VSERVER_AUTO_SINGLE
++ bool "Automatic Single IP Special Casing"
++ depends on EXPERIMENTAL
++ default y
++ help
++ This allows network contexts with a single IP to
++ automatically remap 0.0.0.0 bindings to that IP,
++ avoiding further network checks and improving
++ performance.
+
-+extern int vx_set_init(struct vx_info *, struct task_struct *);
++ (note: such guests do not allow to change the ip
++ on the fly and do not show loopback addresses)
+
-+static int vx_set_initpid(struct vx_info *vxi, int pid)
-+{
-+ struct task_struct *init;
++config VSERVER_COWBL
++ bool "Enable COW Immutable Link Breaking"
++ default y
++ help
++ This enables the COW (Copy-On-Write) link break code.
++ It allows you to treat unified files like normal files
++ when writing to them (which will implicitely break the
++ link and create a copy of the unified file)
+
-+ init = find_task_by_real_pid(pid);
-+ if (!init)
-+ return -ESRCH;
-+ return vx_set_init(vxi, init);
-+}
++config VSERVER_VTIME
++ bool "Enable Virtualized Guest Time"
++ depends on EXPERIMENTAL
++ default n
++ help
++ This enables per guest time offsets to allow for
++ adjusting the system clock individually per guest.
++ this adds some overhead to the time functions and
++ therefore should not be enabled without good reason.
+
-+int vc_new_s_context(uint32_t ctx, void __user *data)
-+{
-+ int ret = -ENOMEM;
-+ struct vcmd_new_s_context_v1 vc_data;
-+ struct vx_info *new_vxi;
++config VSERVER_DEVICE
++ bool "Enable Guest Device Mapping"
++ depends on EXPERIMENTAL
++ default n
++ help
++ This enables generic device remapping.
+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
++config VSERVER_PROC_SECURE
++ bool "Enable Proc Security"
++ depends on PROC_FS
++ default y
++ help
++ This configures ProcFS security to initially hide
++ non-process entries for all contexts except the main and
++ spectator context (i.e. for all guests), which is a secure
++ default.
+
-+ /* legacy hack, will be removed soon */
-+ if (ctx == -2) {
-+ /* assign flags and initpid */
-+ if (!current->vx_info)
-+ return -EINVAL;
-+ ret = 0;
-+ if (vc_data.flags & VX_INFO_INIT)
-+ ret = vx_set_initpid(current->vx_info, current->tgid);
-+ if (ret == 0) {
-+ /* We keep the same vx_id, but lower the capabilities */
-+ current->vx_info->vx_bcaps &= (~vc_data.remove_cap);
-+ ret = vx_current_xid();
-+ current->vx_info->vx_flags |= vc_data.flags;
-+ }
-+ return ret;
-+ }
++ (note: on 1.2x the entries were visible by default)
+
-+ if (!vx_check(0, VS_ADMIN) || !capable(CAP_SYS_ADMIN)
-+ /* might make sense in the future, or not ... */
-+ || vx_flags(VX_INFO_PRIVATE, 0))
-+ return -EPERM;
++config VSERVER_HARDCPU
++ bool "Enable Hard CPU Limits"
++ default y
++ help
++ Activate the Hard CPU Limits
+
-+ /* ugly hack for Spectator */
-+ if (ctx == 1) {
-+ current->xid = 1;
-+ return 0;
-+ }
++ This will compile in code that allows the Token Bucket
++ Scheduler to put processes on hold when a context's
++ tokens are depleted (provided that its per-context
++ sched_hard flag is set).
++
++ Processes belonging to that context will not be able
++ to consume CPU resources again until a per-context
++ configured minimum of tokens has been reached.
++
++config VSERVER_IDLETIME
++ bool "Avoid idle CPUs by skipping Time"
++ depends on VSERVER_HARDCPU
++ default y
++ help
++ This option allows the scheduler to artificially
++ advance time (per cpu) when otherwise the idle
++ task would be scheduled, thus keeping the cpu
++ busy and sharing the available resources among
++ certain contexts.
++
++config VSERVER_IDLELIMIT
++ bool "Limit the IDLE task"
++ depends on VSERVER_HARDCPU
++ default n
++ help
++ Limit the idle slices, so the the next context
++ will be scheduled as soon as possible.
++
++ This might improve interactivity and latency, but
++ will also marginally increase scheduling overhead.
++
++choice
++ prompt "Persistent Inode Tagging"
++ default TAGGING_ID24
++ help
++ This adds persistent context information to filesystems
++ mounted with the tagxid option. Tagging is a requirement
++ for per-context disk limits and per-context quota.
++
++
++config TAGGING_NONE
++ bool "Disabled"
++ help
++ do not store per-context information in inodes.
++
++config TAGGING_UID16
++ bool "UID16/GID32"
++ help
++ reduces UID to 16 bit, but leaves GID at 32 bit.
+
-+ if (((ctx > MAX_S_CONTEXT) && (ctx != VX_DYNAMIC_ID)) ||
-+ (ctx == 0))
-+ return -EINVAL;
++config TAGGING_GID16
++ bool "UID32/GID16"
++ help
++ reduces GID to 16 bit, but leaves UID at 32 bit.
+
-+ if ((ctx == VX_DYNAMIC_ID) || (ctx < MIN_D_CONTEXT))
-+ new_vxi = lookup_or_create_vx_info(ctx);
-+ else
-+ new_vxi = lookup_vx_info(ctx);
++config TAGGING_ID24
++ bool "UID24/GID24"
++ help
++ uses the upper 8bit from UID and GID for XID tagging
++ which leaves 24bit for UID/GID each, which should be
++ more than sufficient for normal use.
+
-+ if (!new_vxi)
-+ return -EINVAL;
++config TAGGING_INTERN
++ bool "UID32/GID32"
++ help
++ this uses otherwise reserved inode fields in the on
++ disk representation, which limits the use to a few
++ filesystems (currently ext2 and ext3)
+
-+ ret = -EPERM;
-+ if (!vx_info_flags(new_vxi, VXF_STATE_SETUP, 0) &&
-+ vx_info_flags(new_vxi, VX_INFO_PRIVATE, 0))
-+ goto out_put;
++endchoice
+
-+ ret = vx_migrate_task(current, new_vxi,
-+ vx_info_flags(new_vxi, VXF_STATE_SETUP, 0));
-+ new_vxi->vx_flags &= ~VXF_STATE_SETUP;
-+
-+ if (ret == 0) {
-+ current->vx_info->vx_bcaps &= (~vc_data.remove_cap);
-+ new_vxi->vx_flags |= vc_data.flags;
-+ if (vc_data.flags & VX_INFO_INIT)
-+ vx_set_initpid(new_vxi, current->tgid);
-+ if (vc_data.flags & VX_INFO_NAMESPACE)
-+ vx_set_space(new_vxi, CLONE_NEWNS | CLONE_FS);
-+ if (vc_data.flags & VX_INFO_NPROC)
-+ __rlim_set(&new_vxi->limit, RLIMIT_NPROC,
-+ current->signal->rlim[RLIMIT_NPROC].rlim_max);
-+
-+ /* tweak some defaults for legacy */
-+ new_vxi->vx_flags |= (VXF_HIDE_NETIF | VXF_INFO_INIT);
-+ ret = new_vxi->vx_id;
-+ }
-+out_put:
-+ put_vx_info(new_vxi);
-+ return ret;
-+}
++config TAG_NFSD
++ bool "Tag NFSD User Auth and Files"
++ default n
++ help
++ Enable this if you do want the in-kernel NFS
++ Server to use the tagging specified above.
++ (will require patched clients too)
+
---- linux-2.6.22/kernel/vserver/legacynet.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/legacynet.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,85 @@
++config VSERVER_PRIVACY
++ bool "Honor Privacy Aspects of Guests"
++ default n
++ help
++ When enabled, most context checks will disallow
++ access to structures assigned to a specific context,
++ like ptys or loop devices.
+
-+/*
-+ * linux/kernel/vserver/legacynet.c
-+ *
-+ * Virtual Server: Legacy Network Funtions
-+ *
-+ * Copyright (C) 2001-2003 Jacques Gelinas
-+ * Copyright (C) 2003-2007 Herbert Pötzl
-+ *
-+ * V0.01 broken out from legacy.c
-+ *
-+ */
++config VSERVER_CONTEXTS
++ int "Maximum number of Contexts (1-65533)" if EMBEDDED
++ range 1 65533
++ default "768" if 64BIT
++ default "256"
++ help
++ This setting will optimize certain data structures
++ and memory allocations according to the expected
++ maximum.
+
-+#include <linux/sched.h>
-+#include <linux/vs_context.h>
-+#include <linux/vs_network.h>
-+#include <linux/vserver/legacy.h>
-+// #include <linux/mnt_namespace.h>
-+#include <linux/err.h>
++ note: this is not a strict upper limit.
+
-+#include <asm/errno.h>
-+#include <asm/uaccess.h>
++config VSERVER_WARN
++ bool "VServer Warnings"
++ default y
++ help
++ This enables various runtime warnings, which will
++ notify about potential manipulation attempts or
++ resource shortage. It is generally considered to
++ be a good idea to have that enabled.
+
++config VSERVER_DEBUG
++ bool "VServer Debugging Code"
++ default n
++ help
++ Set this to yes if you want to be able to activate
++ debugging output at runtime. It adds a very small
++ overhead to all vserver related functions and
++ increases the kernel size by about 20k.
+
-+extern struct nx_info *create_nx_info(void);
++config VSERVER_HISTORY
++ bool "VServer History Tracing"
++ depends on VSERVER_DEBUG
++ default n
++ help
++ Set this to yes if you want to record the history of
++ linux-vserver activities, so they can be replayed in
++ the event of a kernel panic or oops.
+
-+/* set ipv4 root (syscall) */
++config VSERVER_HISTORY_SIZE
++ int "Per-CPU History Size (32-65536)"
++ depends on VSERVER_HISTORY
++ range 32 65536
++ default 64
++ help
++ This allows you to specify the number of entries in
++ the per-CPU history buffer.
+
-+int vc_set_ipv4root(uint32_t nbip, void __user *data)
-+{
-+ int i, err = -EPERM;
-+ struct vcmd_set_ipv4root_v3 vc_data;
-+ struct nx_info *new_nxi, *nxi = current->nx_info;
++config VSERVER_MONITOR
++ bool "VServer Scheduling Monitor"
++ depends on VSERVER_DISABLED
++ default n
++ help
++ Set this to yes if you want to record the scheduling
++ decisions, so that they can be relayed to userspace
++ for detailed analysis.
+
-+ if (nbip < 0 || nbip > NB_IPV4ROOT)
-+ return -EINVAL;
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
++config VSERVER_MONITOR_SIZE
++ int "Per-CPU Monitor Queue Size (32-65536)"
++ depends on VSERVER_MONITOR
++ range 32 65536
++ default 1024
++ help
++ This allows you to specify the number of entries in
++ the per-CPU scheduling monitor buffer.
+
-+ if (!nxi || nxi->ipv4[0] == 0 || capable(CAP_NET_ADMIN))
-+ /* We are allowed to change everything */
-+ err = 0;
-+ else if (nxi) {
-+ int found = 0;
-+
-+ /* We are allowed to select a subset of the currently
-+ installed IP numbers. No new one are allowed
-+ We can't change the broadcast address though */
-+ for (i = 0; i < nbip; i++) {
-+ int j;
-+ __u32 nxip = vc_data.nx_mask_pair[i].ip;
-+ for (j = 0; j < nxi->nbipv4; j++) {
-+ if (nxip == nxi->ipv4[j]) {
-+ found++;
-+ break;
-+ }
-+ }
-+ }
-+ if ((found == nbip) &&
-+ (vc_data.broadcast == nxi->v4_bcast))
-+ err = 0;
-+ }
-+ if (err)
-+ return err;
++config VSERVER_MONITOR_SYNC
++ int "Per-CPU Monitor Sync Interval (0-65536)"
++ depends on VSERVER_MONITOR
++ range 0 65536
++ default 256
++ help
++ This allows you to specify the interval in ticks
++ when a time sync entry is inserted.
+
-+ new_nxi = create_nx_info();
-+ if (IS_ERR(new_nxi))
-+ return -EINVAL;
++endmenu
+
-+ new_nxi->nbipv4 = nbip;
-+ for (i = 0; i < nbip; i++) {
-+ new_nxi->ipv4[i] = vc_data.nx_mask_pair[i].ip;
-+ new_nxi->mask[i] = vc_data.nx_mask_pair[i].mask;
-+ }
-+ new_nxi->v4_bcast = vc_data.broadcast;
-+ if (nxi)
-+ printk("!!! switching nx_info %p->%p\n", nxi, new_nxi);
+
-+ nx_migrate_task(current, new_nxi);
-+ release_nx_info(new_nxi, NULL);
-+ put_nx_info(new_nxi);
-+ return 0;
-+}
++config VSERVER
++ bool
++ default y
++ select NAMESPACES
++ select UTS_NS
++ select IPC_NS
++ select PID_NS
++ select USER_NS
++ select SYSVIPC
+
++config VSERVER_SECURITY
++ bool
++ depends on SECURITY
++ default y
++ select SECURITY_CAPABILITIES
+
---- linux-2.6.22/kernel/vserver/limit.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/limit.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/kernel/vserver/limit.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/limit.c 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,319 @@
+/*
+ * linux/kernel/vserver/limit.c
@@ -21512,14 +21588,12 @@
+ *
+ */
+
++#include <linux/sched.h>
+#include <linux/module.h>
-+#include <linux/vs_context.h>
+#include <linux/vs_limit.h>
+#include <linux/vserver/limit.h>
-+#include <linux/vserver/switch.h>
+#include <linux/vserver/limit_cmd.h>
+
-+#include <asm/errno.h>
+#include <asm/uaccess.h>
+
+
@@ -21812,15 +21886,17 @@
+ if (!vxi)
+ return 0;
+
++ points = vxi->vx_badness_bias;
++
+ v = __vx_cres_array_fixup(&vxi->limit, VLA_RSS);
+ w = __rlim_soft(&vxi->limit, RLIMIT_RSS);
-+ points = (v > w) ? (v - w) : 0;
++ points += (v > w) ? (v - w) : 0;
+
+ return points;
+}
+
---- linux-2.6.22/kernel/vserver/limit_init.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/limit_init.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/kernel/vserver/limit_init.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/limit_init.h 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,33 @@
+
+
@@ -21848,15 +21924,15 @@
+ if ((1 << lim) & VLIM_NOCHECK)
+ continue;
+ value = __rlim_get(limit, lim);
-+ vxwprintk(value,
++ vxwprintk_xid(value,
+ "!!! limit: %p[%s,%d] = %ld on exit.",
+ limit, vlimit_name[lim], lim, (long)value);
+ }
+#endif
+}
+
---- linux-2.6.22/kernel/vserver/limit_proc.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/limit_proc.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/kernel/vserver/limit_proc.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/limit_proc.h 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,57 @@
+#ifndef _VX_LIMIT_PROC_H
+#define _VX_LIMIT_PROC_H
@@ -21915,9 +21991,30 @@
+#endif /* _VX_LIMIT_PROC_H */
+
+
---- linux-2.6.22/kernel/vserver/monitor.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/monitor.c 2007-06-15 05:59:57 +0200
-@@ -0,0 +1,143 @@
+--- linux-2.6.25.4/kernel/vserver/Makefile 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/Makefile 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,18 @@
++#
++# Makefile for the Linux vserver routines.
++#
++
++
++obj-y += vserver.o
++
++vserver-y := switch.o context.o space.o sched.o network.o inode.o \
++ limit.o cvirt.o cacct.o signal.o helper.o init.o \
++ dlimit.o tag.o
++
++vserver-$(CONFIG_INET) += inet.o
++vserver-$(CONFIG_PROC_FS) += proc.o
++vserver-$(CONFIG_VSERVER_DEBUG) += sysctl.o debug.o
++vserver-$(CONFIG_VSERVER_HISTORY) += history.o
++vserver-$(CONFIG_VSERVER_MONITOR) += monitor.o
++vserver-$(CONFIG_VSERVER_DEVICE) += device.o
++
+--- linux-2.6.25.4/kernel/vserver/monitor.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/monitor.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,138 @@
+/*
+ * kernel/vserver/monitor.c
+ *
@@ -21929,15 +22026,10 @@
+ *
+ */
+
-+#include <linux/errno.h>
+#include <linux/module.h>
-+#include <linux/types.h>
-+#include <linux/ctype.h>
+#include <linux/jiffies.h>
-+
+#include <asm/uaccess.h>
+#include <asm/atomic.h>
-+#include <asm/unistd.h>
+
+#include <linux/vserver/monitor.h>
+#include <linux/vserver/debug_cmd.h>
@@ -22061,9 +22153,9 @@
+
+#endif /* CONFIG_COMPAT */
+
---- linux-2.6.22/kernel/vserver/network.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/network.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,771 @@
+--- linux-2.6.25.4/kernel/vserver/network.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/network.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,864 @@
+/*
+ * linux/kernel/vserver/network.c
+ *
@@ -22081,18 +22173,100 @@
+ *
+ */
+
++#include <linux/err.h>
+#include <linux/slab.h>
+#include <linux/rcupdate.h>
-+#include <net/tcp.h>
+
-+#include <asm/errno.h>
-+#include <linux/vserver/base.h>
++#include <linux/vs_network.h>
++#include <linux/vs_pid.h>
+#include <linux/vserver/network_cmd.h>
+
+
+atomic_t nx_global_ctotal = ATOMIC_INIT(0);
+atomic_t nx_global_cactive = ATOMIC_INIT(0);
+
++static struct kmem_cache *nx_addr_v4_cachep = NULL;
++static struct kmem_cache *nx_addr_v6_cachep = NULL;
++
++
++static int __init init_network(void)
++{
++ nx_addr_v4_cachep = kmem_cache_create("nx_v4_addr_cache",
++ sizeof(struct nx_addr_v4), 0,
++ SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
++ nx_addr_v6_cachep = kmem_cache_create("nx_v6_addr_cache",
++ sizeof(struct nx_addr_v6), 0,
++ SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
++ return 0;
++}
++
++
++/* __alloc_nx_addr_v4() */
++
++static inline struct nx_addr_v4 *__alloc_nx_addr_v4(void)
++{
++ struct nx_addr_v4 *nxa = kmem_cache_alloc(
++ nx_addr_v4_cachep, GFP_KERNEL);
++
++ if (!IS_ERR(nxa))
++ memset(nxa, 0, sizeof(*nxa));
++ return nxa;
++}
++
++/* __dealloc_nx_addr_v4() */
++
++static inline void __dealloc_nx_addr_v4(struct nx_addr_v4 *nxa)
++{
++ kmem_cache_free(nx_addr_v4_cachep, nxa);
++}
++
++/* __dealloc_nx_addr_v4_all() */
++
++static inline void __dealloc_nx_addr_v4_all(struct nx_addr_v4 *nxa)
++{
++ while (nxa) {
++ struct nx_addr_v4 *next = nxa->next;
++
++ __dealloc_nx_addr_v4(nxa);
++ nxa = next;
++ }
++}
++
++
++#ifdef CONFIG_IPV6
++
++/* __alloc_nx_addr_v6() */
++
++static inline struct nx_addr_v6 *__alloc_nx_addr_v6(void)
++{
++ struct nx_addr_v6 *nxa = kmem_cache_alloc(
++ nx_addr_v6_cachep, GFP_KERNEL);
++
++ if (!IS_ERR(nxa))
++ memset(nxa, 0, sizeof(*nxa));
++ return nxa;
++}
++
++/* __dealloc_nx_addr_v6() */
++
++static inline void __dealloc_nx_addr_v6(struct nx_addr_v6 *nxa)
++{
++ kmem_cache_free(nx_addr_v6_cachep, nxa);
++}
++
++/* __dealloc_nx_addr_v6_all() */
++
++static inline void __dealloc_nx_addr_v6_all(struct nx_addr_v6 *nxa)
++{
++ while (nxa) {
++ struct nx_addr_v6 *next = nxa->next;
++
++ __dealloc_nx_addr_v6(nxa);
++ nxa = next;
++ }
++}
++
++#endif /* CONFIG_IPV6 */
+
+/* __alloc_nx_info()
+
@@ -22121,6 +22295,9 @@
+
+ /* rest of init goes here */
+
++ new->v4_lback.s_addr = htonl(INADDR_LOOPBACK);
++ new->v4_bcast.s_addr = htonl(INADDR_BROADCAST);
++
+ vxdprintk(VXD_CBIT(nid, 0),
+ "alloc_nx_info(%d) = %p", nid, new);
+ atomic_inc(&nx_global_ctotal);
@@ -22142,6 +22319,8 @@
+ BUG_ON(atomic_read(&nxi->nx_usecnt));
+ BUG_ON(atomic_read(&nxi->nx_tasks));
+
++ __dealloc_nx_addr_v4_all(nxi->v4.next);
++
+ nxi->nx_state |= NXS_RELEASED;
+ kfree(nxi);
+ atomic_dec(&nx_global_ctotal);
@@ -22170,6 +22349,18 @@
+}
+
+
++void __nx_set_lback(struct nx_info *nxi)
++{
++ int nid = nxi->nx_id;
++ __be32 lback = htonl(INADDR_LOOPBACK ^ ((nid & 0xFFFF) << 8));
++
++ nxi->v4_lback.s_addr = lback;
++}
++
++extern int __nx_inet_add_lback(__be32 addr);
++extern int __nx_inet_del_lback(__be32 addr);
++
++
+/* hash table for nx_info hash */
+
+#define NX_HASH_SIZE 13
@@ -22258,29 +22449,6 @@
+}
+
+
-+/* __nx_dynamic_id()
-+
-+ * find unused dynamic nid
-+ * requires the hash_lock to be held */
-+
-+static inline nid_t __nx_dynamic_id(void)
-+{
-+ static nid_t seq = MAX_N_CONTEXT;
-+ nid_t barrier = seq;
-+
-+ vxd_assert_lock(&nx_info_hash_lock);
-+ do {
-+ if (++seq > MAX_N_CONTEXT)
-+ seq = MIN_D_CONTEXT;
-+ if (!__lookup_nx_info(seq)) {
-+ vxdprintk(VXD_CBIT(nid, 4),
-+ "__nx_dynamic_id: [#%d]", seq);
-+ return seq;
-+ }
-+ } while (barrier != seq);
-+ return 0;
-+}
-+
+/* __create_nx_info()
+
+ * create the requested context
@@ -22298,24 +22466,8 @@
+ /* required to make dynamic xids unique */
+ spin_lock(&nx_info_hash_lock);
+
-+ /* dynamic context requested */
-+ if (id == NX_DYNAMIC_ID) {
-+#ifdef CONFIG_VSERVER_DYNAMIC_IDS
-+ id = __nx_dynamic_id();
-+ if (!id) {
-+ printk(KERN_ERR "no dynamic context available.\n");
-+ nxi = ERR_PTR(-EAGAIN);
-+ goto out_unlock;
-+ }
-+ new->nx_id = id;
-+#else
-+ printk(KERN_ERR "dynamic contexts disabled.\n");
-+ nxi = ERR_PTR(-EINVAL);
-+ goto out_unlock;
-+#endif
-+ }
+ /* static context requested */
-+ else if ((nxi = __lookup_nx_info(id))) {
++ if ((nxi = __lookup_nx_info(id))) {
+ vxdprintk(VXD_CBIT(nid, 0),
+ "create_nx_info(%d) = %p (already there)", id, nxi);
+ if (nx_info_flags(nxi, NXF_STATE_SETUP, 0))
@@ -22324,18 +22476,11 @@
+ nxi = ERR_PTR(-EEXIST);
+ goto out_unlock;
+ }
-+ /* dynamic nid creation blocker */
-+ else if (id >= MIN_D_CONTEXT) {
-+ vxdprintk(VXD_CBIT(nid, 0),
-+ "create_nx_info(%d) (dynamic rejected)", id);
-+ nxi = ERR_PTR(-EINVAL);
-+ goto out_unlock;
-+ }
-+
+ /* new context */
+ vxdprintk(VXD_CBIT(nid, 0),
+ "create_nx_info(%d) = %p (new)", id, new);
+ claim_nx_info(new, NULL);
++ __nx_set_lback(new);
+ __hash_nx_info(get_nx_info(new));
+ nxi = new, new = NULL;
+
@@ -22359,15 +22504,6 @@
+ spin_unlock(&nx_info_hash_lock);
+}
+
-+#ifdef CONFIG_VSERVER_LEGACYNET
-+
-+struct nx_info *create_nx_info(void)
-+{
-+ return __create_nx_info(NX_DYNAMIC_ID);
-+}
-+
-+#endif
-+
+/* lookup_nx_info()
+
+ * search for a nx_info and get() it
@@ -22500,102 +22636,6 @@
+}
+
+
-+#ifdef CONFIG_INET
-+
-+#include <linux/netdevice.h>
-+#include <linux/inetdevice.h>
-+
-+int ifa_in_nx_info(struct in_ifaddr *ifa, struct nx_info *nxi)
-+{
-+ if (!nxi)
-+ return 1;
-+ if (!ifa)
-+ return 0;
-+ return addr_in_nx_info(nxi, ifa->ifa_local);
-+}
-+
-+int dev_in_nx_info(struct net_device *dev, struct nx_info *nxi)
-+{
-+ struct in_device *in_dev;
-+ struct in_ifaddr **ifap;
-+ struct in_ifaddr *ifa;
-+ int ret = 0;
-+
-+ if (!nxi)
-+ return 1;
-+
-+ if (!dev)
-+ goto out;
-+ in_dev = in_dev_get(dev);
-+ if (!in_dev)
-+ goto out;
-+
-+ for (ifap = &in_dev->ifa_list; (ifa = *ifap) != NULL;
-+ ifap = &ifa->ifa_next) {
-+ if (addr_in_nx_info(nxi, ifa->ifa_local)) {
-+ ret = 1;
-+ break;
-+ }
-+ }
-+ in_dev_put(in_dev);
-+out:
-+ return ret;
-+}
-+
-+/*
-+ * check if address is covered by socket
-+ *
-+ * sk: the socket to check against
-+ * addr: the address in question (must be != 0)
-+ */
-+static inline int __addr_in_socket(const struct sock *sk, uint32_t addr)
-+{
-+ struct nx_info *nxi = sk->sk_nx_info;
-+ uint32_t saddr = inet_rcv_saddr(sk);
-+
-+ vxdprintk(VXD_CBIT(net, 5),
-+ "__addr_in_socket(%p," NIPQUAD_FMT ") %p:" NIPQUAD_FMT " %p;%lx",
-+ sk, NIPQUAD(addr), nxi, NIPQUAD(saddr), sk->sk_socket,
-+ (sk->sk_socket ? sk->sk_socket->flags : 0));
-+
-+ if (saddr) {
-+ /* direct address match */
-+ return (saddr == addr);
-+ } else if (nxi) {
-+ /* match against nx_info */
-+ return addr_in_nx_info(nxi, addr);
-+ } else {
-+ /* unrestricted any socket */
-+ return 1;
-+ }
-+}
-+
-+
-+int nx_addr_conflict(struct nx_info *nxi, uint32_t addr, const struct sock *sk)
-+{
-+ vxdprintk(VXD_CBIT(net, 2),
-+ "nx_addr_conflict(%p,%p) " NIPQUAD_FMT,
-+ nxi, sk, NIPQUAD(addr));
-+
-+ if (addr) {
-+ /* check real address */
-+ return __addr_in_socket(sk, addr);
-+ } else if (nxi) {
-+ /* check against nx_info */
-+ int i, n = nxi->nbipv4;
-+
-+ for (i = 0; i < n; i++)
-+ if (__addr_in_socket(sk, nxi->ipv4[i]))
-+ return 1;
-+ return 0;
-+ } else {
-+ /* check against any */
-+ return 1;
-+ }
-+}
-+
-+#endif /* CONFIG_INET */
-+
+void nx_set_persistent(struct nx_info *nxi)
+{
+ vxdprintk(VXD_CBIT(nid, 6),
@@ -22629,16 +22669,13 @@
+#include <asm/uaccess.h>
+
+
-+int vc_task_nid(uint32_t id, void __user *data)
++int vc_task_nid(uint32_t id)
+{
+ nid_t nid;
+
+ if (id) {
+ struct task_struct *tsk;
+
-+ if (!nx_check(0, VS_ADMIN | VS_WATCH))
-+ return -EPERM;
-+
+ read_lock(&tasklist_lock);
+ tsk = find_task_by_real_pid(id);
+ nid = (tsk) ? tsk->nid : -ESRCH;
@@ -22672,9 +22709,7 @@
+ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
+ return -EFAULT;
+
-+ if ((nid > MAX_S_CONTEXT) && (nid != NX_DYNAMIC_ID))
-+ return -EINVAL;
-+ if (nid < 2)
++ if ((nid > MAX_S_CONTEXT) || (nid < 2))
+ return -EINVAL;
+
+ new_nxi = __create_nx_info(nid);
@@ -22710,10 +22745,40 @@
+ return nx_migrate_task(current, nxi);
+}
+
++
++
++int do_add_v4_addr(struct nx_info *nxi, __be32 ip, __be32 ip2, __be32 mask,
++ uint16_t type, uint16_t flags)
++{
++ struct nx_addr_v4 *nxa = &nxi->v4;
++
++ if (NX_IPV4(nxi)) {
++ /* locate last entry */
++ for (; nxa->next; nxa = nxa->next);
++ nxa->next = __alloc_nx_addr_v4();
++ nxa = nxa->next;
++
++ if (IS_ERR(nxa))
++ return PTR_ERR(nxa);
++ }
++
++ if (nxi->v4.next)
++ /* remove single ip for ip list */
++ nxi->nx_flags &= ~NXF_SINGLE_IP;
++
++ nxa->ip[0].s_addr = ip;
++ nxa->ip[1].s_addr = ip2;
++ nxa->mask.s_addr = mask;
++ nxa->type = type;
++ nxa->flags = flags;
++ return 0;
++}
++
++
+int vc_net_add(struct nx_info *nxi, void __user *data)
+{
+ struct vcmd_net_addr_v0 vc_data;
-+ int index, pos, ret = 0;
++ int index, ret = 0;
+
+ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
+ return -EFAULT;
@@ -22722,21 +22787,14 @@
+ case NXA_TYPE_IPV4:
+ if ((vc_data.count < 1) || (vc_data.count > 4))
+ return -EINVAL;
-+ break;
+
-+ default:
-+ break;
-+ }
-+
-+ switch (vc_data.type) {
-+ case NXA_TYPE_IPV4:
+ index = 0;
-+ while ((index < vc_data.count) &&
-+ ((pos = nxi->nbipv4) < NB_IPV4ROOT)) {
-+ nxi->ipv4[pos] = vc_data.ip[index];
-+ nxi->mask[pos] = vc_data.mask[index];
++ while (index < vc_data.count) {
++ ret = do_add_v4_addr(nxi, vc_data.ip[index].s_addr, 0,
++ vc_data.mask[index].s_addr, NXA_TYPE_ADDR, 0);
++ if (ret)
++ return ret;
+ index++;
-+ nxi->nbipv4++;
+ }
+ ret = index;
+ break;
@@ -22746,6 +22804,11 @@
+ ret = 1;
+ break;
+
++ case NXA_TYPE_IPV4|NXA_MOD_LBACK:
++ nxi->v4_lback = vc_data.ip[0];
++ ret = 1;
++ break;
++
+ default:
+ ret = -EINVAL;
+ break;
@@ -22762,7 +22825,124 @@
+
+ switch (vc_data.type) {
+ case NXA_TYPE_ANY:
-+ nxi->nbipv4 = 0;
++ __dealloc_nx_addr_v4_all(xchg(&nxi->v4.next, NULL));
++ memset(&nxi->v4, 0, sizeof(nxi->v4));
++ break;
++
++ default:
++ return -EINVAL;
++ }
++ return 0;
++}
++
++
++int vc_net_add_ipv4(struct nx_info *nxi, void __user *data)
++{
++ struct vcmd_net_addr_ipv4_v1 vc_data;
++
++ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ switch (vc_data.type) {
++ case NXA_TYPE_ADDR:
++ case NXA_TYPE_RANGE:
++ case NXA_TYPE_MASK:
++ return do_add_v4_addr(nxi, vc_data.ip.s_addr, 0,
++ vc_data.mask.s_addr, vc_data.type, vc_data.flags);
++
++ case NXA_TYPE_ADDR | NXA_MOD_BCAST:
++ nxi->v4_bcast = vc_data.ip;
++ break;
++
++ case NXA_TYPE_ADDR | NXA_MOD_LBACK:
++ nxi->v4_lback = vc_data.ip;
++ break;
++
++ default:
++ return -EINVAL;
++ }
++ return 0;
++}
++
++int vc_net_remove_ipv4(struct nx_info *nxi, void __user *data)
++{
++ struct vcmd_net_addr_ipv4_v1 vc_data;
++
++ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ switch (vc_data.type) {
++/* case NXA_TYPE_ADDR:
++ break; */
++
++ case NXA_TYPE_ANY:
++ __dealloc_nx_addr_v4_all(xchg(&nxi->v4.next, NULL));
++ memset(&nxi->v4, 0, sizeof(nxi->v4));
++ break;
++
++ default:
++ return -EINVAL;
++ }
++ return 0;
++}
++
++
++#ifdef CONFIG_IPV6
++
++int do_add_v6_addr(struct nx_info *nxi,
++ struct in6_addr *ip, struct in6_addr *mask,
++ uint32_t prefix, uint16_t type, uint16_t flags)
++{
++ struct nx_addr_v6 *nxa = &nxi->v6;
++
++ if (NX_IPV6(nxi)) {
++ /* locate last entry */
++ for (; nxa->next; nxa = nxa->next);
++ nxa->next = __alloc_nx_addr_v6();
++ nxa = nxa->next;
++
++ if (IS_ERR(nxa))
++ return PTR_ERR(nxa);
++ }
++
++ nxa->ip = *ip;
++ nxa->mask = *mask;
++ nxa->prefix = prefix;
++ nxa->type = type;
++ nxa->flags = flags;
++ return 0;
++}
++
++
++int vc_net_add_ipv6(struct nx_info *nxi, void __user *data)
++{
++ struct vcmd_net_addr_ipv6_v1 vc_data;
++
++ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ switch (vc_data.type) {
++ case NXA_TYPE_ADDR:
++ case NXA_TYPE_MASK:
++ return do_add_v6_addr(nxi, &vc_data.ip, &vc_data.mask,
++ vc_data.prefix, vc_data.type, vc_data.flags);
++ default:
++ return -EINVAL;
++ }
++ return 0;
++}
++
++int vc_net_remove_ipv6(struct nx_info *nxi, void __user *data)
++{
++ struct vcmd_net_addr_ipv6_v1 vc_data;
++
++ if (data && copy_from_user(&vc_data, data, sizeof(vc_data)))
++ return -EFAULT;
++
++ switch (vc_data.type) {
++ case NXA_TYPE_ANY:
++ __dealloc_nx_addr_v6_all(xchg(&nxi->v6.next, NULL));
++ memset(&nxi->v6, 0, sizeof(nxi->v6));
+ break;
+
+ default:
@@ -22771,6 +22951,9 @@
+ return 0;
+}
+
++#endif /* CONFIG_IPV6 */
++
++
+int vc_get_nflags(struct nx_info *nxi, void __user *data)
+{
+ struct vcmd_net_flags_v0 vc_data;
@@ -22778,7 +22961,7 @@
+ vc_data.flagword = nxi->nx_flags;
+
+ /* special STATE flag handling */
-+ vc_data.mask = vs_mask_flags(~0UL, nxi->nx_flags, NXF_ONE_TIME);
++ vc_data.mask = vs_mask_flags(~0ULL, nxi->nx_flags, NXF_ONE_TIME);
+
+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
+ return -EFAULT;
@@ -22810,7 +22993,7 @@
+ struct vcmd_net_caps_v0 vc_data;
+
+ vc_data.ncaps = nxi->nx_ncaps;
-+ vc_data.cmask = ~0UL;
++ vc_data.cmask = ~0ULL;
+
+ if (copy_to_user(data, &vc_data, sizeof(vc_data)))
+ return -EFAULT;
@@ -22832,12 +23015,14 @@
+
+#include <linux/module.h>
+
++module_init(init_network);
++
+EXPORT_SYMBOL_GPL(free_nx_info);
+EXPORT_SYMBOL_GPL(unhash_nx_info);
+
---- linux-2.6.22/kernel/vserver/proc.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/proc.c 2007-06-15 19:20:10 +0200
-@@ -0,0 +1,1033 @@
+--- linux-2.6.25.4/kernel/vserver/proc.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/proc.c 2008-04-21 13:42:34.000000000 -0400
+@@ -0,0 +1,1086 @@
+/*
+ * linux/kernel/vserver/proc.c
+ *
@@ -22856,18 +23041,19 @@
+ *
+ */
+
-+#include <linux/errno.h>
+#include <linux/proc_fs.h>
-+#include <linux/sched.h>
++#include <asm/unistd.h>
++
+#include <linux/vs_context.h>
+#include <linux/vs_network.h>
+#include <linux/vs_cvirt.h>
+
-+#include <linux/vserver/switch.h>
-+#include <linux/vserver/global.h>
++#include <linux/in.h>
++#include <linux/inetdevice.h>
++#include <linux/vs_inet.h>
++#include <linux/vs_inet6.h>
+
-+#include <asm/uaccess.h>
-+#include <asm/unistd.h>
++#include <linux/vserver/global.h>
+
+#include "cvirt_proc.h"
+#include "cacct_proc.h"
@@ -22875,6 +23061,19 @@
+#include "sched_proc.h"
+#include "vci_config.h"
+
++
++static inline char *print_cap_t(char *buffer, kernel_cap_t *c)
++{
++ unsigned __capi;
++
++ CAP_FOR_EACH_U32(__capi) {
++ buffer += sprintf(buffer, "%08x",
++ c->cap[(_LINUX_CAPABILITY_U32S-1) - __capi]);
++ }
++ return buffer;
++}
++
++
+static struct proc_dir_entry *proc_virtual;
+
+static struct proc_dir_entry *proc_virtnet;
@@ -22905,14 +23104,16 @@
+ return sprintf(buffer,
+ "#CTotal:\t%d\n"
+ "#CActive:\t%d\n"
-+ "#NSProxy:\t%d\t%d %d %d %d\n",
++ "#NSProxy:\t%d\t%d %d %d %d %d %d\n",
+ atomic_read(&vx_global_ctotal),
+ atomic_read(&vx_global_cactive),
+ atomic_read(&vs_global_nsproxy),
+ atomic_read(&vs_global_fs),
+ atomic_read(&vs_global_mnt_ns),
+ atomic_read(&vs_global_uts_ns),
-+ atomic_read(&vs_global_ipc_ns));
++ atomic_read(&vs_global_ipc_ns),
++ atomic_read(&vs_global_user_ns),
++ atomic_read(&vs_global_pid_ns));
+}
+
+
@@ -22923,31 +23124,37 @@
+ length = sprintf(buffer,
+ "ID:\t%d\n"
+ "Info:\t%p\n"
-+ "Init:\t%d\n",
++ "Init:\t%d\n"
++ "OOM:\t%lld\n",
+ vxi->vx_id,
+ vxi,
-+ vxi->vx_initpid);
++ vxi->vx_initpid,
++ vxi->vx_badness_bias);
+ return length;
+}
+
+int proc_vxi_status(struct vx_info *vxi, char *buffer)
+{
-+ int length;
++ char *orig = buffer;
+
-+ length = sprintf(buffer,
++ buffer += sprintf(buffer,
+ "UseCnt:\t%d\n"
+ "Tasks:\t%d\n"
-+ "Flags:\t%016llx\n"
-+ "BCaps:\t%016llx\n"
-+ "CCaps:\t%016llx\n"
-+ "Spaces:\t%08lx\n",
++ "Flags:\t%016llx\n",
+ atomic_read(&vxi->vx_usecnt),
+ atomic_read(&vxi->vx_tasks),
-+ (unsigned long long)vxi->vx_flags,
-+ (unsigned long long)vxi->vx_bcaps,
++ (unsigned long long)vxi->vx_flags);
++
++ buffer += sprintf(buffer, "BCaps:\t");
++ buffer = print_cap_t(buffer, &vxi->vx_bcaps);
++ buffer += sprintf(buffer, "\n");
++
++ buffer += sprintf(buffer,
++ "CCaps:\t%016llx\n"
++ "Spaces:\t%08lx\n",
+ (unsigned long long)vxi->vx_ccaps,
+ vxi->vx_nsmask);
-+ return length;
++ return buffer - orig;
+}
+
+int proc_vxi_limit(struct vx_info *vxi, char *buffer)
@@ -23009,18 +23216,36 @@
+
+int proc_nxi_info(struct nx_info *nxi, char *buffer)
+{
++ struct nx_addr_v4 *v4a;
++#ifdef CONFIG_IPV6
++ struct nx_addr_v6 *v6a;
++#endif
+ int length, i;
+
+ length = sprintf(buffer,
+ "ID:\t%d\n"
-+ "Info:\t%p\n",
++ "Info:\t%p\n"
++ "Bcast:\t" NIPQUAD_FMT "\n"
++ "Lback:\t" NIPQUAD_FMT "\n",
+ nxi->nx_id,
-+ nxi);
-+ for (i = 0; i < nxi->nbipv4; i++) {
-+ length += sprintf(buffer + length,
-+ "%d:\t" NIPQUAD_FMT "/" NIPQUAD_FMT "\n", i,
-+ NIPQUAD(nxi->ipv4[i]), NIPQUAD(nxi->mask[i]));
-+ }
++ nxi,
++ NIPQUAD(nxi->v4_bcast.s_addr),
++ NIPQUAD(nxi->v4_lback.s_addr));
++
++ if (!NX_IPV4(nxi))
++ goto skip_v4;
++ for (i = 0, v4a = &nxi->v4; v4a; i++, v4a = v4a->next)
++ length += sprintf(buffer + length, "%d:\t" NXAV4_FMT "\n",
++ i, NXAV4(v4a));
++skip_v4:
++#ifdef CONFIG_IPV6
++ if (!NX_IPV6(nxi))
++ goto skip_v6;
++ for (i = 0, v6a = &nxi->v6; v6a; i++, v6a = v6a->next)
++ length += sprintf(buffer + length, "%d:\t" NXAV6_FMT "\n",
++ i, NXAV6(v6a));
++skip_v6:
++#endif
+ return length;
+}
+
@@ -23615,8 +23840,6 @@
+}
+
+
-+
-+#define PROC_NUMBUF 10
+#define PROC_MAXVIDS 32
+
+int proc_virtual_readdir(struct file *filp,
@@ -23783,7 +24006,7 @@
+};
+
+static struct inode_operations proc_virtnet_dir_inode_operations = {
-+ .getattr = proc_virtnet_getattr,
++ .getattr = proc_virtnet_getattr,
+ .lookup = proc_virtnet_lookup,
+};
+
@@ -23825,8 +24048,9 @@
+ if (!vxi)
+ goto out;
+
-+ buffer += sprintf(buffer, "BCaps:\t%016llx\n",
-+ (unsigned long long)vxi->vx_bcaps);
++ buffer += sprintf(buffer, "BCaps:\t");
++ buffer = print_cap_t(buffer, &vxi->vx_bcaps);
++ buffer += sprintf(buffer, "\n");
+ buffer += sprintf(buffer, "CCaps:\t%016llx\n",
+ (unsigned long long)vxi->vx_ccaps);
+ buffer += sprintf(buffer, "CFlags:\t%016llx\n",
@@ -23842,6 +24066,10 @@
+int proc_pid_nx_info(struct task_struct *p, char *buffer)
+{
+ struct nx_info *nxi;
++ struct nx_addr_v4 *v4a;
++#ifdef CONFIG_IPV6
++ struct nx_addr_v6 *v6a;
++#endif
+ char *orig = buffer;
+ int i;
+
@@ -23856,24 +24084,34 @@
+ buffer += sprintf(buffer, "NFlags:\t%016llx\n",
+ (unsigned long long)nxi->nx_flags);
+
-+ for (i = 0; i < nxi->nbipv4; i++){
-+ buffer += sprintf(buffer,
-+ "V4Root[%d]:\t" NIPQUAD_FMT "/" NIPQUAD_FMT "\n", i,
-+ NIPQUAD(nxi->ipv4[i]),
-+ NIPQUAD(nxi->mask[i]));
-+ }
+ buffer += sprintf(buffer,
+ "V4Root[bcast]:\t" NIPQUAD_FMT "\n",
-+ NIPQUAD(nxi->v4_bcast));
-+
++ NIPQUAD(nxi->v4_bcast.s_addr));
++ buffer += sprintf (buffer,
++ "V4Root[lback]:\t" NIPQUAD_FMT "\n",
++ NIPQUAD(nxi->v4_lback.s_addr));
++ if (!NX_IPV4(nxi))
++ goto skip_v4;
++ for (i = 0, v4a = &nxi->v4; v4a; i++, v4a = v4a->next)
++ buffer += sprintf(buffer, "V4Root[%d]:\t" NXAV4_FMT "\n",
++ i, NXAV4(v4a));
++skip_v4:
++#ifdef CONFIG_IPV6
++ if (!NX_IPV6(nxi))
++ goto skip_v6;
++ for (i = 0, v6a = &nxi->v6; v6a; i++, v6a = v6a->next)
++ buffer += sprintf(buffer, "V6Root[%d]:\t" NXAV6_FMT "\n",
++ i, NXAV6(v6a));
++skip_v6:
++#endif
+ put_nx_info(nxi);
+out:
+ return buffer - orig;
+}
+
---- linux-2.6.22/kernel/vserver/sched.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/sched.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,447 @@
+--- linux-2.6.25.4/kernel/vserver/sched.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/sched.c 2008-04-29 18:40:09.000000000 -0400
+@@ -0,0 +1,413 @@
+/*
+ * linux/kernel/vserver/sched.c
+ *
@@ -23884,17 +24122,17 @@
+ * V0.01 adapted Sam Vilains version to 2.6.3
+ * V0.02 removed legacy interface
+ * V0.03 changed vcmds to vxi arg
++ * V0.04 removed older and legacy interfaces
+ *
+ */
+
-+#include <linux/sched.h>
+#include <linux/vs_context.h>
+#include <linux/vs_sched.h>
+#include <linux/vserver/sched_cmd.h>
+
-+#include <asm/errno.h>
+#include <asm/uaccess.h>
+
++
+#define vxd_check_range(val, min, max) do { \
+ vxlprintk((val < min) || (val > max), \
+ "check_range(%ld,%ld,%ld)", \
@@ -24139,6 +24377,7 @@
+ update_mask = vxi->sched.update_mask & VXSM_SET_MASK;
+ update_mask |= (set_mask & (VXSM_SET_MASK | VXSM_IDLE_TIME));
+ vxi->sched.update_mask = update_mask;
++
+#ifdef CONFIG_SMP
+ rmb();
+ if (set_mask & VXSM_CPU_ID) {
@@ -24165,6 +24404,7 @@
+ return 0;
+}
+
++
+#define COPY_IDS(C) C(cpu_id); C(bucket_id)
+#define COPY_PRI(C) C(prio_bias)
+#define COPY_TOK(C) C(tokens); C(tokens_min); C(tokens_max)
@@ -24186,50 +24426,6 @@
+ return do_set_sched(vxi, &vc_data);
+}
+
-+#ifdef CONFIG_VSERVER_LEGACY
-+
-+#define COPY_MASK_V2(name, mask) \
-+ if (vc_data.name != SCHED_KEEP) { \
-+ vc_data_v4.name = vc_data.name; \
-+ vc_data_v4.set_mask |= mask; \
-+ }
-+
-+int vc_set_sched_v2(struct vx_info *vxi, void __user *data)
-+{
-+ struct vcmd_set_sched_v2 vc_data;
-+ struct vcmd_set_sched_v4 vc_data_v4 = { .set_mask = 0 };
-+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
-+
-+ COPY_MASK_V2(fill_rate, VXSM_FILL_RATE);
-+ COPY_MASK_V2(interval, VXSM_INTERVAL);
-+ COPY_MASK_V2(tokens, VXSM_TOKENS);
-+ COPY_MASK_V2(tokens_min, VXSM_TOKENS_MIN);
-+ COPY_MASK_V2(tokens_max, VXSM_TOKENS_MAX);
-+ vc_data_v4.bucket_id = 0;
-+
-+ do_set_sched_v4(vxi, &vc_data_v4);
-+ return 0;
-+}
-+#endif
-+
-+int vc_set_sched_v3(struct vx_info *vxi, void __user *data)
-+{
-+ struct vcmd_set_sched_v3 vc_data;
-+ struct vcmd_set_sched_v4 vc_data_v4;
-+
-+ if (copy_from_user(&vc_data, data, sizeof(vc_data)))
-+ return -EFAULT;
-+
-+ /* structures are binary compatible */
-+ memcpy(&vc_data_v4, &vc_data, sizeof(vc_data));
-+ vc_data_v4.set_mask &= VXSM_V3_MASK;
-+ vc_data_v4.bucket_id = 0;
-+
-+ return do_set_sched_v4(vxi, &vc_data_v4);
-+}
-+
+int vc_set_sched_v4(struct vx_info *vxi, void __user *data)
+{
+ struct vcmd_set_sched_v4 vc_data;
@@ -24253,6 +24449,14 @@
+}
+
+
++#define COPY_PRI(C) C(prio_bias)
++#define COPY_TOK(C) C(tokens); C(tokens_min); C(tokens_max)
++#define COPY_FRI(C) C(fill_rate[0]); C(interval[0]); \
++ C(fill_rate[1]); C(interval[1]);
++
++#define COPY_VALUE(name) vc_data.name = data->name
++
++
+int vc_get_sched(struct vx_info *vxi, void __user *data)
+{
+ struct vcmd_sched_v5 vc_data;
@@ -24321,8 +24525,8 @@
+ return 0;
+}
+
---- linux-2.6.22/kernel/vserver/sched_init.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/sched_init.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/kernel/vserver/sched_init.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/sched_init.h 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,50 @@
+
+static inline void vx_info_init_sched(struct _vx_sched *sched)
@@ -24374,8 +24578,8 @@
+{
+ return;
+}
---- linux-2.6.22/kernel/vserver/sched_proc.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/sched_proc.h 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/kernel/vserver/sched_proc.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/sched_proc.h 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,57 @@
+#ifndef _VX_SCHED_PROC_H
+#define _VX_SCHED_PROC_H
@@ -24434,9 +24638,9 @@
+}
+
+#endif /* _VX_SCHED_PROC_H */
---- linux-2.6.22/kernel/vserver/signal.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/signal.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,134 @@
+--- linux-2.6.25.4/kernel/vserver/signal.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/signal.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,132 @@
+/*
+ * linux/kernel/vserver/signal.c
+ *
@@ -24450,12 +24654,10 @@
+ *
+ */
+
-+#include <linux/sched.h>
-+
-+#include <asm/errno.h>
+#include <asm/uaccess.h>
+
+#include <linux/vs_context.h>
++#include <linux/vs_pid.h>
+#include <linux/vserver/signal_cmd.h>
+
+
@@ -24571,9 +24773,9 @@
+ return ret;
+}
+
---- linux-2.6.22/kernel/vserver/space.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/space.c 2007-06-16 19:12:02 +0200
-@@ -0,0 +1,281 @@
+--- linux-2.6.25.4/kernel/vserver/space.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/space.c 2008-05-15 15:41:12.000000000 -0400
+@@ -0,0 +1,316 @@
+/*
+ * linux/kernel/vserver/space.c
+ *
@@ -24589,34 +24791,36 @@
+ */
+
+#include <linux/utsname.h>
-+#include <linux/sched.h>
-+#include <linux/vs_context.h>
-+#include <linux/vserver/space.h>
-+#include <linux/vserver/space_cmd.h>
-+#include <linux/dcache.h>
-+#include <linux/mount.h>
+#include <linux/nsproxy.h>
-+#include <linux/fs.h>
-+
-+#include <asm/errno.h>
++#include <linux/err.h>
+#include <asm/uaccess.h>
+
++#include <linux/vs_context.h>
++#include <linux/vserver/space.h>
++#include <linux/vserver/space_cmd.h>
+
+atomic_t vs_global_nsproxy = ATOMIC_INIT(0);
+atomic_t vs_global_fs = ATOMIC_INIT(0);
+atomic_t vs_global_mnt_ns = ATOMIC_INIT(0);
+atomic_t vs_global_uts_ns = ATOMIC_INIT(0);
+atomic_t vs_global_ipc_ns = ATOMIC_INIT(0);
++atomic_t vs_global_user_ns = ATOMIC_INIT(0);
++atomic_t vs_global_pid_ns = ATOMIC_INIT(0);
+
+
+/* namespace functions */
+
+#include <linux/mnt_namespace.h>
++#include <linux/user_namespace.h>
++#include <linux/pid_namespace.h>
++#include <linux/ipc_namespace.h>
++#include <net/net_namespace.h>
+
+const struct vcmd_space_mask space_mask = {
+ .mask = CLONE_NEWNS |
+ CLONE_NEWUTS |
+ CLONE_NEWIPC |
++ CLONE_NEWUSER |
+ CLONE_FS
+};
+
@@ -24634,6 +24838,9 @@
+ struct mnt_namespace *old_ns;
+ struct uts_namespace *old_uts;
+ struct ipc_namespace *old_ipc;
++ struct pid_namespace *old_pid;
++ struct user_namespace *old_user;
++ struct net *old_net;
+ struct nsproxy *nsproxy;
+
+ nsproxy = copy_nsproxy(old_nsproxy);
@@ -24664,12 +24871,42 @@
+ } else
+ old_ipc = NULL;
+
++ if (mask & CLONE_NEWUSER) {
++ old_user = nsproxy->user_ns;
++ nsproxy->user_ns = new_nsproxy->user_ns;
++ if (nsproxy->user_ns)
++ get_user_ns(nsproxy->user_ns);
++ } else
++ old_user = NULL;
++
++ if (mask & CLONE_NEWPID) {
++ old_pid = nsproxy->pid_ns;
++ nsproxy->pid_ns = new_nsproxy->pid_ns;
++ if (nsproxy->pid_ns)
++ get_pid_ns(nsproxy->pid_ns);
++ } else
++ old_pid = NULL;
++
++ if (mask & CLONE_NEWNET) {
++ old_net = nsproxy->net_ns;
++ nsproxy->net_ns = new_nsproxy->net_ns;
++ if (nsproxy->net_ns)
++ get_net(nsproxy->net_ns);
++ } else
++ old_net = NULL;
++
+ if (old_ns)
+ put_mnt_ns(old_ns);
+ if (old_uts)
+ put_uts_ns(old_uts);
+ if (old_ipc)
+ put_ipc_ns(old_ipc);
++ if (old_pid)
++ put_pid_ns(old_pid);
++ if (old_user)
++ put_user_ns(old_user);
++ if (old_net)
++ put_net(old_net);
+out:
+ return nsproxy;
+}
@@ -24855,9 +25092,9 @@
+ return 0;
+}
+
---- linux-2.6.22/kernel/vserver/switch.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/switch.c 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,527 @@
+--- linux-2.6.25.4/kernel/vserver/switch.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/switch.c 2008-04-29 18:40:18.000000000 -0400
+@@ -0,0 +1,529 @@
+/*
+ * linux/kernel/vserver/switch.c
+ *
@@ -24873,27 +25110,22 @@
+ * V0.06 added compat32 layer
+ * V0.07 vcmd args and perms
+ * V0.08 added status commands
++ * V0.09 added tag commands
++ * V0.10 added oom bias
++ * V0.11 added device commands
+ *
+ */
+
-+#include <linux/linkage.h>
-+#include <linux/sched.h>
-+#include <linux/compat.h>
-+#include <asm/errno.h>
-+
+#include <linux/vs_context.h>
+#include <linux/vs_network.h>
+#include <linux/vserver/switch.h>
+
+#include "vci_config.h"
+
++
+static inline
+int vc_get_version(uint32_t id)
+{
-+#ifdef CONFIG_VSERVER_LEGACY_VERSION
-+ if (id == 63)
-+ return VCI_LEGACY_VERSION;
-+#endif
+ return VCI_VERSION;
+}
+
@@ -24914,8 +25146,9 @@
+#include <linux/vserver/dlimit_cmd.h>
+#include <linux/vserver/signal_cmd.h>
+#include <linux/vserver/space_cmd.h>
++#include <linux/vserver/tag_cmd.h>
++#include <linux/vserver/device_cmd.h>
+
-+#include <linux/vserver/legacy.h>
+#include <linux/vserver/inode.h>
+#include <linux/vserver/dlimit.h>
+
@@ -24923,9 +25156,13 @@
+#ifdef CONFIG_COMPAT
+#define __COMPAT(name, id, data, compat) \
+ (compat) ? name ## _x32(id, data) : name(id, data)
++#define __COMPAT_NO_ID(name, data, compat) \
++ (compat) ? name ## _x32(data) : name(data)
+#else
+#define __COMPAT(name, id, data, compat) \
+ name(id, data)
++#define __COMPAT_NO_ID(name, data, compat) \
++ name(data)
+#endif
+
+
@@ -24942,16 +25179,18 @@
+ return vc_get_vci(id);
+
+ case VCMD_task_xid:
-+ return vc_task_xid(id, data);
++ return vc_task_xid(id);
+ case VCMD_vx_info:
+ return vc_vx_info(vxi, data);
+
+ case VCMD_task_nid:
-+ return vc_task_nid(id, data);
++ return vc_task_nid(id);
+ case VCMD_nx_info:
+ return vc_nx_info(nxi, data);
+
-+ case VCMD_set_space_v0:
++ case VCMD_task_tag:
++ return vc_task_tag(id);
++
+ /* this is version 1 */
+ case VCMD_set_space:
+ return vc_set_space(vxi, data);
@@ -24994,13 +25233,9 @@
+ case VCMD_get_cflags:
+ return vc_get_cflags(vxi, data);
+
-+ case VCMD_set_ccaps_v0:
-+ return vc_set_ccaps_v0(vxi, data);
+ /* this is version 1 */
+ case VCMD_set_ccaps:
+ return vc_set_ccaps(vxi, data);
-+ case VCMD_get_ccaps_v0:
-+ return vc_get_ccaps_v0(vxi, data);
+ /* this is version 1 */
+ case VCMD_get_ccaps:
+ return vc_get_ccaps(vxi, data);
@@ -25009,6 +25244,11 @@
+ case VCMD_get_bcaps:
+ return vc_get_bcaps(vxi, data);
+
++ case VCMD_set_badness:
++ return vc_set_badness(vxi, data);
++ case VCMD_get_badness:
++ return vc_get_badness(vxi, data);
++
+ case VCMD_set_nflags:
+ return vc_set_nflags(nxi, data);
+ case VCMD_get_nflags:
@@ -25019,12 +25259,6 @@
+ case VCMD_get_ncaps:
+ return vc_get_ncaps(nxi, data);
+
-+#ifdef CONFIG_VSERVER_LEGACY
-+ case VCMD_set_sched_v2:
-+ return vc_set_sched_v2(vxi, data);
-+#endif
-+ case VCMD_set_sched_v3:
-+ return vc_set_sched_v3(vxi, data);
+ case VCMD_set_sched_v4:
+ return vc_set_sched_v4(vxi, data);
+ /* this is version 5 */
@@ -25050,15 +25284,15 @@
+ case VCMD_wait_exit:
+ return vc_wait_exit(vxi, data);
+
-+#ifdef CONFIG_VSERVER_LEGACY
-+ case VCMD_create_context:
-+ return vc_ctx_create(id, NULL);
-+#endif
-+
+ case VCMD_get_iattr:
-+ return __COMPAT(vc_get_iattr, id, data, compat);
++ return __COMPAT_NO_ID(vc_get_iattr, data, compat);
+ case VCMD_set_iattr:
-+ return __COMPAT(vc_set_iattr, id, data, compat);
++ return __COMPAT_NO_ID(vc_set_iattr, data, compat);
++
++ case VCMD_fget_iattr:
++ return vc_fget_iattr(id, data);
++ case VCMD_fset_iattr:
++ return vc_fset_iattr(id, data);
+
+ case VCMD_enter_space_v0:
+ return vc_enter_space(vxi, NULL);
@@ -25081,11 +25315,42 @@
+ return vc_net_create(id, data);
+ case VCMD_net_migrate:
+ return vc_net_migrate(nxi, data);
++
++ case VCMD_tag_migrate:
++ return vc_tag_migrate(id);
++
+ case VCMD_net_add:
+ return vc_net_add(nxi, data);
+ case VCMD_net_remove:
+ return vc_net_remove(nxi, data);
+
++ case VCMD_net_add_ipv4:
++ return vc_net_add_ipv4(nxi, data);
++ case VCMD_net_remove_ipv4:
++ return vc_net_remove_ipv4(nxi, data);
++#ifdef CONFIG_IPV6
++ case VCMD_net_add_ipv6:
++ return vc_net_add_ipv6(nxi, data);
++ case VCMD_net_remove_ipv6:
++ return vc_net_remove_ipv6(nxi, data);
++#endif
++/* case VCMD_add_match_ipv4:
++ return vc_add_match_ipv4(nxi, data);
++ case VCMD_get_match_ipv4:
++ return vc_get_match_ipv4(nxi, data);
++#ifdef CONFIG_IPV6
++ case VCMD_add_match_ipv6:
++ return vc_add_match_ipv6(nxi, data);
++ case VCMD_get_match_ipv6:
++ return vc_get_match_ipv6(nxi, data);
++#endif */
++
++#ifdef CONFIG_VSERVER_DEVICE
++ case VCMD_set_mapping:
++ return __COMPAT(vc_set_mapping, vxi, data, compat);
++ case VCMD_unset_mapping:
++ return __COMPAT(vc_unset_mapping, vxi, data, compat);
++#endif
+#ifdef CONFIG_VSERVER_HISTORY
+ case VCMD_dump_history:
+ return vc_dump_history(id);
@@ -25096,16 +25361,8 @@
+ case VCMD_read_monitor:
+ return __COMPAT(vc_read_monitor, id, data, compat);
+#endif
-+#ifdef CONFIG_VSERVER_LEGACY
-+ case VCMD_new_s_context:
-+ return vc_new_s_context(id, data);
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACYNET
-+ case VCMD_set_ipv4root:
-+ return vc_set_ipv4root(id, data);
-+#endif
+ default:
-+ vxwprintk(1, "unimplemented VCMD_%02d_%d[%d]",
++ vxwprintk_task(1, "unimplemented VCMD_%02d_%d[%d]",
+ VC_CATEGORY(cmd), VC_COMMAND(cmd), VC_VERSION(cmd));
+ }
+ return -ENOSYS;
@@ -25151,9 +25408,9 @@
+ __VCMD(reset_minmax, 2, VCA_VXI, 0);
+ __VCMD(vx_info, 3, VCA_VXI, VCF_INFO);
+ __VCMD(get_bcaps, 3, VCA_VXI, VCF_INFO);
-+ __VCMD(get_ccaps_v0, 3, VCA_VXI, VCF_INFO);
+ __VCMD(get_ccaps, 3, VCA_VXI, VCF_INFO);
+ __VCMD(get_cflags, 3, VCA_VXI, VCF_INFO);
++ __VCMD(get_badness, 3, VCA_VXI, VCF_INFO);
+ __VCMD(get_vhi_name, 3, VCA_VXI, VCF_INFO);
+ __VCMD(get_rlimit, 3, VCA_VXI, VCF_INFO);
+
@@ -25167,7 +25424,10 @@
+ __VCMD(get_ncaps, 3, VCA_NXI, VCF_INFO);
+ __VCMD(get_nflags, 3, VCA_NXI, VCF_INFO);
+
++ __VCMD(task_tag, 2, VCA_NONE, 0);
++
+ __VCMD(get_iattr, 2, VCA_NONE, 0);
++ __VCMD(fget_iattr, 2, VCA_NONE, 0);
+ __VCMD(get_dlimit, 3, VCA_NONE, VCF_INFO);
+ __VCMD(get_sched, 3, VCA_VXI, VCF_INFO);
+ __VCMD(sched_info, 3, VCA_VXI, VCF_INFO | VCF_ZIDOK);
@@ -25185,33 +25445,42 @@
+ __VCMD(net_create, 5, VCA_NONE, 0);
+ __VCMD(net_migrate, 5, VCA_NXI, VCF_ADMIN);
+
++ __VCMD(tag_migrate, 5, VCA_NONE, VCF_ADMIN);
++
+ /* higher admin commands */
+ __VCMD(ctx_kill, 6, VCA_VXI, VCF_ARES);
-+ __VCMD(set_space_v0, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+ __VCMD(set_space, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+
-+ __VCMD(set_ccaps_v0, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+ __VCMD(set_ccaps, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+ __VCMD(set_bcaps, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+ __VCMD(set_cflags, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
++ __VCMD(set_badness, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+
+ __VCMD(set_vhi_name, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+ __VCMD(set_rlimit, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+ __VCMD(set_sched, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
-+ __VCMD(set_sched_v2, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
-+ __VCMD(set_sched_v3, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+ __VCMD(set_sched_v4, 7, VCA_VXI, VCF_ARES | VCF_SETUP);
+
+ __VCMD(set_ncaps, 7, VCA_NXI, VCF_ARES | VCF_SETUP);
+ __VCMD(set_nflags, 7, VCA_NXI, VCF_ARES | VCF_SETUP);
+ __VCMD(net_add, 8, VCA_NXI, VCF_ARES | VCF_SETUP);
+ __VCMD(net_remove, 8, VCA_NXI, VCF_ARES | VCF_SETUP);
-+
++ __VCMD(net_add_ipv4, 8, VCA_NXI, VCF_ARES | VCF_SETUP);
++ __VCMD(net_remove_ipv4, 8, VCA_NXI, VCF_ARES | VCF_SETUP);
++#ifdef CONFIG_IPV6
++ __VCMD(net_add_ipv6, 8, VCA_NXI, VCF_ARES | VCF_SETUP);
++ __VCMD(net_remove_ipv6, 8, VCA_NXI, VCF_ARES | VCF_SETUP);
++#endif
+ __VCMD(set_iattr, 7, VCA_NONE, 0);
++ __VCMD(fset_iattr, 7, VCA_NONE, 0);
+ __VCMD(set_dlimit, 7, VCA_NONE, VCF_ARES);
+ __VCMD(add_dlimit, 8, VCA_NONE, VCF_ARES);
+ __VCMD(rem_dlimit, 8, VCA_NONE, VCF_ARES);
+
++#ifdef CONFIG_VSERVER_DEVICE
++ __VCMD(set_mapping, 8, VCA_VXI, VCF_ARES|VCF_ZIDOK);
++ __VCMD(unset_mapping, 8, VCA_VXI, VCF_ARES|VCF_ZIDOK);
++#endif
+ /* debug level admin commands */
+#ifdef CONFIG_VSERVER_HISTORY
+ __VCMD(dump_history, 9, VCA_NONE, 0);
@@ -25221,14 +25490,6 @@
+ __VCMD(read_monitor, 9, VCA_NONE, 0);
+#endif
+
-+ /* legacy commands */
-+#ifdef CONFIG_VSERVER_LEGACY
-+ __VCMD(new_s_context, 1, VCA_NONE, 0);
-+ __VCMD(create_context, 5, VCA_NONE, 0);
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACYNET
-+ __VCMD(set_ipv4root, 5, VCA_NONE, 0);
-+#endif
+ default:
+ perm = -1;
+ }
@@ -25244,15 +25505,8 @@
+ goto out;
+
+ state = 1;
-+#ifdef CONFIG_VSERVER_LEGACY
-+ if (!capable(CAP_CONTEXT) &&
-+ /* dirty hack for capremove */
-+ !(cmd == VCMD_new_s_context && id == -2))
-+ goto out;
-+#else
+ if (!capable(CAP_CONTEXT))
+ goto out;
-+#endif
+
+ state = 2;
+ /* moved here from the individual commands */
@@ -25269,16 +25523,6 @@
+ state = 4;
+ /* various legacy exceptions */
+ switch (cmd) {
-+#ifdef CONFIG_VSERVER_LEGACY
-+ case VCMD_set_cflags:
-+ case VCMD_set_ccaps_v0:
-+ ret = 0;
-+ if (vx_check(0, VS_WATCH))
-+ goto out;
-+ break;
-+
-+ case VCMD_ctx_create_v0:
-+#endif
+ /* will go away when spectator is a cap */
+ case VCMD_ctx_migrate_v0:
+ case VCMD_ctx_migrate:
@@ -25297,11 +25541,6 @@
+ goto out;
+ }
+ break;
-+
-+ /* legacy special casing */
-+ case VCMD_set_space_v0:
-+ id = -1;
-+ break;
+ }
+
+ /* vcmds are fine by default */
@@ -25385,9 +25624,9 @@
+}
+
+#endif /* CONFIG_COMPAT */
---- linux-2.6.22/kernel/vserver/sysctl.c 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/sysctl.c 2007-06-16 05:31:20 +0200
-@@ -0,0 +1,248 @@
+--- linux-2.6.25.4/kernel/vserver/sysctl.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/sysctl.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,244 @@
+/*
+ * kernel/vserver/sysctl.c
+ *
@@ -25399,20 +25638,13 @@
+ *
+ */
+
-+#include <linux/errno.h>
+#include <linux/module.h>
-+#include <linux/types.h>
+#include <linux/ctype.h>
+#include <linux/sysctl.h>
+#include <linux/parser.h>
-+#include <linux/fs.h>
-+
+#include <asm/uaccess.h>
-+#include <asm/unistd.h>
+
+
-+#define CTL_VSERVER 4242 /* unused? */
-+
+enum {
+ CTL_DEBUG_ERROR = 0,
+ CTL_DEBUG_SWITCH = 1,
@@ -25445,13 +25677,13 @@
+
+
+static struct ctl_table_header *vserver_table_header;
-+static ctl_table vserver_table[];
++static ctl_table vserver_root_table[];
+
+
+void vserver_register_sysctl(void)
+{
+ if (!vserver_table_header) {
-+ vserver_table_header = register_sysctl_table(vserver_table);
++ vserver_table_header = register_sysctl_table(vserver_root_table);
+ }
+
+}
@@ -25522,6 +25754,7 @@
+ return 0;
+}
+
++static int zero;
+
+#define CTL_ENTRY(ctl, name) \
+ { \
@@ -25530,10 +25763,12 @@
+ .data = &vx_ ## name, \
+ .maxlen = sizeof(int), \
+ .mode = 0644, \
-+ .proc_handler = &proc_dodebug \
++ .proc_handler = &proc_dodebug, \
++ .strategy = &sysctl_intvec, \
++ .extra1 = &zero, \
+ }
+
-+static ctl_table debug_table[] = {
++static ctl_table vserver_debug_table[] = {
+ CTL_ENTRY(CTL_DEBUG_SWITCH, debug_switch),
+ CTL_ENTRY(CTL_DEBUG_XID, debug_xid),
+ CTL_ENTRY(CTL_DEBUG_NID, debug_nid),
@@ -25549,12 +25784,12 @@
+ { .ctl_name = 0 }
+};
+
-+static ctl_table vserver_table[] = {
++static ctl_table vserver_root_table[] = {
+ {
+ .ctl_name = CTL_VSERVER,
+ .procname = "vserver",
+ .mode = 0555,
-+ .child = debug_table
++ .child = vserver_debug_table
+ },
+ { .ctl_name = 0 }
+};
@@ -25636,21 +25871,83 @@
+EXPORT_SYMBOL_GPL(vx_debug_space);
+EXPORT_SYMBOL_GPL(vx_debug_misc);
+
---- linux-2.6.22/kernel/vserver/vci_config.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/kernel/vserver/vci_config.h 2007-06-15 02:37:04 +0200
-@@ -0,0 +1,92 @@
+--- linux-2.6.25.4/kernel/vserver/tag.c 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/tag.c 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,63 @@
++/*
++ * linux/kernel/vserver/tag.c
++ *
++ * Virtual Server: Shallow Tag Space
++ *
++ * Copyright (C) 2007 Herbert Pötzl
++ *
++ * V0.01 basic implementation
++ *
++ */
++
++#include <linux/sched.h>
++#include <linux/vserver/debug.h>
++#include <linux/vs_pid.h>
++#include <linux/vs_tag.h>
++
++#include <linux/vserver/tag_cmd.h>
++
++
++int dx_migrate_task(struct task_struct *p, tag_t tag)
++{
++ if (!p)
++ BUG();
++
++ vxdprintk(VXD_CBIT(tag, 5),
++ "dx_migrate_task(%p[#%d],#%d)", p, p->tag, tag);
++
++ task_lock(p);
++ p->tag = tag;
++ task_unlock(p);
++
++ vxdprintk(VXD_CBIT(tag, 5),
++ "moved task %p into [#%d]", p, tag);
++ return 0;
++}
++
++/* vserver syscall commands below here */
++
++/* taks xid and vx_info functions */
++
++
++int vc_task_tag(uint32_t id)
++{
++ tag_t tag;
++
++ if (id) {
++ struct task_struct *tsk;
++ read_lock(&tasklist_lock);
++ tsk = find_task_by_real_pid(id);
++ tag = (tsk) ? tsk->tag : -ESRCH;
++ read_unlock(&tasklist_lock);
++ } else
++ tag = dx_current_tag();
++ return tag;
++}
++
++
++int vc_tag_migrate(uint32_t tag)
++{
++ return dx_migrate_task(current, tag & 0xFFFF);
++}
++
++
+--- linux-2.6.25.4/kernel/vserver/vci_config.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/kernel/vserver/vci_config.h 2008-04-19 15:14:52.000000000 -0400
+@@ -0,0 +1,81 @@
+
+/* interface version */
+
-+#define VCI_VERSION 0x00020200
-+#define VCI_LEGACY_VERSION 0x000100FF
++#define VCI_VERSION 0x00020302
+
+
+enum {
+ VCI_KCBIT_NO_DYNAMIC = 0,
-+ VCI_KCBIT_LEGACY = 1,
-+ VCI_KCBIT_LEGACYNET = 2,
-+ VCI_KCBIT_NGNET = 3,
+
+ VCI_KCBIT_PROC_SECURE = 4,
+ VCI_KCBIT_HARDCPU = 5,
@@ -25660,30 +25957,21 @@
+ VCI_KCBIT_COWBL = 8,
+ VCI_KCBIT_FULLCOWBL = 9,
+ VCI_KCBIT_SPACES = 10,
++ VCI_KCBIT_NETV2 = 11,
+
-+ VCI_KCBIT_LEGACY_VERSION = 15,
+ VCI_KCBIT_DEBUG = 16,
+ VCI_KCBIT_HISTORY = 20,
+ VCI_KCBIT_TAGGED = 24,
++ VCI_KCBIT_PPTAG = 28,
++
++ VCI_KCBIT_MORE = 31,
+};
+
+
+static inline uint32_t vci_kernel_config(void)
+{
+ return
-+ /* various legacy options */
-+#ifndef CONFIG_VSERVER_DYNAMIC_IDS
+ (1 << VCI_KCBIT_NO_DYNAMIC) |
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACY
-+ (1 << VCI_KCBIT_LEGACY) |
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACYNET
-+ (1 << VCI_KCBIT_LEGACYNET) |
-+#endif
-+#ifdef CONFIG_VSERVER_LEGACY_VERSION
-+ (1 << VCI_KCBIT_LEGACY_VERSION) |
-+#endif
+
+ /* configured features */
+#ifdef CONFIG_VSERVER_PROC_SECURE
@@ -25703,6 +25991,7 @@
+ (1 << VCI_KCBIT_FULLCOWBL) |
+#endif
+ (1 << VCI_KCBIT_SPACES) |
++ (1 << VCI_KCBIT_NETV2) |
+
+ /* debug options */
+#ifdef CONFIG_VSERVER_DEBUG
@@ -25728,288 +26017,22 @@
+#else
+ (7 << VCI_KCBIT_TAGGED) |
+#endif
++ (1 << VCI_KCBIT_PPTAG) |
+ 0;
+}
+
---- linux-2.6.22/mm/filemap.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/filemap.c 2007-07-09 13:11:54 +0200
-@@ -1245,6 +1245,31 @@ int file_send_actor(read_descriptor_t *
- return written;
- }
-
-+/* FIXME: It would be as simple as this, if we had a (void __user*) to write.
-+ * We already have a kernel buffer, so it should be even simpler, right? ;)
-+ *
-+ * Yes, sorta. After duplicating the complete path of generic_file_write(),
-+ * at least some special cases could be removed, so the copy is simpler than
-+ * the original. But it remains a copy, so overall complexity increases.
-+ */
-+static ssize_t
-+generic_kernel_file_write(struct file *, const char *, size_t, loff_t *);
-+
-+ssize_t generic_file_sendpage(struct file *file, struct page *page,
-+ int offset, size_t size, loff_t *ppos, int more)
-+{
-+ ssize_t ret;
-+ char *kaddr;
-+
-+ kaddr = kmap(page);
-+ ret = generic_kernel_file_write(file, kaddr + offset, size, ppos);
-+ kunmap(page);
-+
-+ return ret;
-+}
-+
-+EXPORT_SYMBOL(generic_file_sendpage);
-+
- ssize_t generic_file_sendfile(struct file *in_file, loff_t *ppos,
- size_t count, read_actor_t actor, void *target)
- {
-@@ -1924,6 +1949,19 @@ int remove_suid(struct dentry *dentry)
- }
- EXPORT_SYMBOL(remove_suid);
-
-+static inline size_t
-+filemap_copy_from_kernel(struct page *page, unsigned long offset,
-+ const char *buf, unsigned bytes)
-+{
-+ char *kaddr;
-+
-+ kaddr = kmap(page);
-+ memcpy(kaddr + offset, buf, bytes);
-+ kunmap(page);
-+
-+ return bytes;
-+}
-+
- size_t
- __filemap_copy_from_user_iovec_inatomic(char *vaddr,
- const struct iovec *iov, size_t base, size_t bytes)
-@@ -2235,6 +2273,175 @@ zero_length_segment:
- }
- EXPORT_SYMBOL(generic_file_buffered_write);
-
-+static inline void
-+filemap_set_next_kvec(const struct kvec **iovp, size_t *basep, size_t bytes)
-+{
-+ const struct kvec *iov = *iovp;
-+ size_t base = *basep;
-+
-+ while (bytes) {
-+ int copy = min(bytes, iov->iov_len - base);
-+
-+ bytes -= copy;
-+ base += copy;
-+ if (iov->iov_len == base) {
-+ iov++;
-+ base = 0;
-+ }
-+ }
-+ *iovp = iov;
-+ *basep = base;
-+}
-+
-+/*
-+ * TODO:
-+ * This largely tries to copy generic_file_aio_write_nolock(), although it
-+ * doesn't have to be nearly as generic. A real cleanup should either
-+ * merge this into generic_file_aio_write_nolock() as well or keep it special
-+ * and remove as much code as possible.
-+ */
-+static ssize_t
-+generic_kernel_file_aio_write_nolock(struct kiocb *iocb, const struct kvec*iov,
-+ unsigned long nr_segs, loff_t *ppos)
-+{
-+ struct file *file = iocb->ki_filp;
-+ struct address_space * mapping = file->f_mapping;
-+ const struct address_space_operations *a_ops = mapping->a_ops;
-+ size_t ocount; /* original count */
-+ size_t count; /* after file limit checks */
-+ struct inode *inode = mapping->host;
-+ long status = 0;
-+ loff_t pos;
-+ struct page *page;
-+ struct page *cached_page = NULL;
-+ const int isblk = S_ISBLK(inode->i_mode);
-+ ssize_t written;
-+ ssize_t err;
-+ size_t bytes;
-+ struct pagevec lru_pvec;
-+ const struct kvec *cur_iov = iov; /* current kvec */
-+ size_t iov_base = 0; /* offset in the current kvec */
-+ unsigned long seg;
-+ char *buf;
-+
-+ ocount = 0;
-+ for (seg = 0; seg < nr_segs; seg++) {
-+ const struct kvec *iv = &iov[seg];
-+
-+ /*
-+ * If any segment has a negative length, or the cumulative
-+ * length ever wraps negative then return -EINVAL.
-+ */
-+ ocount += iv->iov_len;
-+ if (unlikely((ssize_t)(ocount|iv->iov_len) < 0))
-+ return -EINVAL;
-+ }
-+
-+ count = ocount;
-+ pos = *ppos;
-+ pagevec_init(&lru_pvec, 0);
-+
-+ /* We can write back this queue in page reclaim */
-+ current->backing_dev_info = mapping->backing_dev_info;
-+ written = 0;
-+
-+ err = generic_write_checks(file, &pos, &count, isblk);
-+ if (err)
-+ goto out;
-+
-+
-+ if (count == 0)
-+ goto out;
-+
-+ remove_suid(file->f_dentry);
-+ file_update_time(file);
-+
-+ /* There is no sane reason to use O_DIRECT */
-+ BUG_ON(file->f_flags & O_DIRECT);
-+
-+ buf = iov->iov_base;
-+ do {
-+ unsigned long index;
-+ unsigned long offset;
-+ size_t copied;
-+
-+ offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
-+ index = pos >> PAGE_CACHE_SHIFT;
-+ bytes = PAGE_CACHE_SIZE - offset;
-+ if (bytes > count)
-+ bytes = count;
-+
-+ page = __grab_cache_page(mapping,index,&cached_page,&lru_pvec);
-+ if (!page) {
-+ status = -ENOMEM;
-+ break;
-+ }
-+
-+ status = a_ops->prepare_write(file, page, offset, offset+bytes);
-+ if (unlikely(status)) {
-+ loff_t isize = i_size_read(inode);
-+ /*
-+ * prepare_write() may have instantiated a few blocks
-+ * outside i_size. Trim these off again.
-+ */
-+ unlock_page(page);
-+ page_cache_release(page);
-+ if (pos + bytes > isize)
-+ vmtruncate(inode, isize);
-+ break;
-+ }
-+
-+ BUG_ON(nr_segs != 1);
-+ copied = filemap_copy_from_kernel(page, offset, buf, bytes);
-+
-+ flush_dcache_page(page);
-+ status = a_ops->commit_write(file, page, offset, offset+bytes);
-+ if (likely(copied > 0)) {
-+ if (!status)
-+ status = copied;
-+
-+ if (status >= 0) {
-+ written += status;
-+ count -= status;
-+ pos += status;
-+ buf += status;
-+ if (unlikely(nr_segs > 1))
-+ filemap_set_next_kvec(&cur_iov,
-+ &iov_base, status);
-+ }
-+ }
-+ if (unlikely(copied != bytes))
-+ if (status >= 0)
-+ status = -EFAULT;
-+ unlock_page(page);
-+ mark_page_accessed(page);
-+ page_cache_release(page);
-+ if (status < 0)
-+ break;
-+ balance_dirty_pages_ratelimited(mapping);
-+ cond_resched();
-+ } while (count);
-+ *ppos = pos;
-+
-+ if (cached_page)
-+ page_cache_release(cached_page);
-+
-+ /*
-+ * For now, when the user asks for O_SYNC, we'll actually give O_DSYNC
-+ */
-+ if (status >= 0) {
-+ if ((file->f_flags & O_SYNC) || IS_SYNC(inode))
-+ status = generic_osync_inode(inode, mapping,
-+ OSYNC_METADATA|OSYNC_DATA);
-+ }
-+
-+ err = written ? written : status;
-+out:
-+ pagevec_lru_add(&lru_pvec);
-+ current->backing_dev_info = 0;
-+ return err;
-+}
-+
- static ssize_t
- __generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov,
- unsigned long nr_segs, loff_t *ppos)
-@@ -2335,6 +2542,36 @@ out:
- return written ? written : err;
- }
-
-+static ssize_t
-+generic_kernel_file_write_nolock(struct file *file, const struct kvec *iov,
-+ unsigned long nr_segs, loff_t *ppos)
-+{
-+ struct kiocb kiocb;
-+ ssize_t ret;
-+
-+ init_sync_kiocb(&kiocb, file);
-+ ret = generic_kernel_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos);
-+ if (ret == -EIOCBQUEUED)
-+ ret = wait_on_sync_kiocb(&kiocb);
-+ return ret;
-+}
-+
-+static ssize_t generic_kernel_file_write(struct file *file, const char *buf,
-+ size_t count, loff_t *ppos)
-+{
-+ struct inode *inode = file->f_mapping->host;
-+ ssize_t err;
-+ struct kvec local_iov = { .iov_base = (char *) buf,
-+ .iov_len = count };
-+
-+ mutex_lock(&inode->i_mutex);
-+ err = generic_kernel_file_write_nolock(file, &local_iov, 1, ppos);
-+ mutex_unlock(&inode->i_mutex);
-+
-+ return err;
-+}
-+
-+
- ssize_t generic_file_aio_write_nolock(struct kiocb *iocb,
- const struct iovec *iov, unsigned long nr_segs, loff_t pos)
- {
---- linux-2.6.22/mm/filemap_xip.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/filemap_xip.c 2007-06-15 05:30:21 +0200
+--- linux-2.6.25.4/mm/filemap_xip.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/filemap_xip.c 2008-04-19 15:14:52.000000000 -0400
@@ -14,6 +14,7 @@
#include <linux/uio.h>
#include <linux/rmap.h>
#include <linux/sched.h>
+#include <linux/vs_memory.h>
#include <asm/tlbflush.h>
- #include "filemap.h"
---- linux-2.6.22/mm/fremap.c 2007-02-06 03:01:56 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/mm/fremap.c 2007-06-15 02:37:04 +0200
+ /*
+--- linux-2.6.25.4/mm/fremap.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/fremap.c 2008-04-19 15:14:52.000000000 -0400
@@ -15,6 +15,7 @@
#include <linux/rmap.h>
#include <linux/module.h>
@@ -26018,17 +26041,8 @@
#include <asm/mmu_context.h>
#include <asm/cacheflush.h>
-@@ -74,6 +75,8 @@ int install_page(struct mm_struct *mm, s
- err = -ENOMEM;
- if (page_mapcount(page) > INT_MAX/2)
- goto unlock;
-+ if (!vx_rss_avail(mm, 1))
-+ goto unlock;
-
- if (pte_none(*pte) || !zap_pte(mm, vma, addr, pte))
- inc_mm_counter(mm, file_rss);
---- linux-2.6.22/mm/hugetlb.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/hugetlb.c 2007-06-17 05:54:20 +0200
+--- linux-2.6.25.4/mm/hugetlb.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/hugetlb.c 2008-04-19 15:14:52.000000000 -0400
@@ -19,6 +19,7 @@
#include <asm/pgtable.h>
@@ -26037,9 +26051,9 @@
#include "internal.h"
const unsigned long hugetlb_zero = 0, hugetlb_infinity = ~0UL;
---- linux-2.6.22/mm/memory.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/memory.c 2007-06-17 05:55:53 +0200
-@@ -498,6 +498,9 @@ static int copy_pte_range(struct mm_stru
+--- linux-2.6.25.4/mm/memory.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/memory.c 2008-04-19 15:14:52.000000000 -0400
+@@ -505,6 +505,9 @@ static int copy_pte_range(struct mm_stru
int progress = 0;
int rss[2];
@@ -26049,8 +26063,8 @@
again:
rss[1] = rss[0] = 0;
dst_pte = pte_alloc_map_lock(dst_mm, dst_pmd, addr, &dst_ptl);
-@@ -2157,6 +2160,11 @@ static int do_swap_page(struct mm_struct
- count_vm_event(PGMAJFAULT);
+@@ -2058,6 +2061,11 @@ static int do_swap_page(struct mm_struct
+ goto out;
}
+ if (!vx_rss_avail(mm, 1)) {
@@ -26058,37 +26072,27 @@
+ goto out;
+ }
+
- delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
mark_page_accessed(page);
lock_page(page);
-@@ -2229,6 +2237,8 @@ static int do_anonymous_page(struct mm_s
- /* Allocate our own private page. */
- pte_unmap(page_table);
-
-+ if (!vx_rss_avail(mm, 1))
-+ goto oom;
- if (unlikely(anon_vma_prepare(vma)))
- goto oom;
- page = alloc_zeroed_user_highpage(vma, address);
-@@ -2302,6 +2312,9 @@ static int do_no_page(struct mm_struct *
+ delayacct_clear_flag(DELAYACCT_PF_SWAPIN);
+@@ -2129,6 +2137,8 @@ static int do_anonymous_page(struct mm_s
+ /* Allocate our own private page. */
pte_unmap(page_table);
- BUG_ON(vma->vm_flags & VM_PFNMAP);
+ if (!vx_rss_avail(mm, 1))
-+ return VM_FAULT_OOM;
-+
- if (vma->vm_file) {
- mapping = vma->vm_file->f_mapping;
- sequence = mapping->truncate_count;
-@@ -2527,6 +2540,7 @@ static inline int handle_pte_fault(struc
++ goto oom;
+ if (unlikely(anon_vma_prepare(vma)))
+ goto oom;
+ page = alloc_zeroed_user_highpage_movable(vma, address);
+@@ -2453,6 +2463,7 @@ static inline int handle_pte_fault(struc
{
pte_t entry;
spinlock_t *ptl;
-+ int ret, type = VXPT_UNKNOWN;
++ int ret = 0, type = VXPT_UNKNOWN;
entry = *pte;
if (!pte_present(entry)) {
-@@ -2555,9 +2569,12 @@ static inline int handle_pte_fault(struc
+@@ -2480,9 +2491,12 @@ static inline int handle_pte_fault(struc
if (unlikely(!pte_same(*pte, entry)))
goto unlock;
if (write_access) {
@@ -26103,20 +26107,20 @@
entry = pte_mkdirty(entry);
}
entry = pte_mkyoung(entry);
-@@ -2576,7 +2593,10 @@ static inline int handle_pte_fault(struc
+@@ -2500,7 +2514,10 @@ static inline int handle_pte_fault(struc
}
unlock:
pte_unmap_unlock(pte, ptl);
-- return VM_FAULT_MINOR;
-+ ret = VM_FAULT_MINOR;
+- return 0;
++ ret = 0;
+out:
+ vx_page_fault(mm, vma, type, ret);
+ return ret;
}
/*
---- linux-2.6.22/mm/mlock.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/mlock.c 2007-06-15 05:30:38 +0200
+--- linux-2.6.25.4/mm/mlock.c 2008-04-17 10:37:25.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/mlock.c 2008-04-19 15:14:52.000000000 -0400
@@ -12,6 +12,7 @@
#include <linux/syscalls.h>
#include <linux/sched.h>
@@ -26173,9 +26177,9 @@
if (!(flags & MCL_CURRENT) || (current->mm->total_vm <= lock_limit) ||
capable(CAP_IPC_LOCK))
ret = do_mlockall(flags);
---- linux-2.6.22/mm/mmap.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/mmap.c 2007-07-07 03:52:54 +0200
-@@ -1144,10 +1144,10 @@ munmap_back:
+--- linux-2.6.25.4/mm/mmap.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/mmap.c 2008-04-19 15:14:52.000000000 -0400
+@@ -1197,10 +1197,10 @@ munmap_back:
kmem_cache_free(vm_area_cachep, vma);
}
out:
@@ -26187,8 +26191,8 @@
+ vx_vmlocked_add(mm, len >> PAGE_SHIFT);
make_pages_present(addr, addr + len);
}
- if (flags & MAP_POPULATE) {
-@@ -1502,9 +1502,9 @@ static int acct_stack_growth(struct vm_a
+ if ((flags & MAP_POPULATE) && !(flags & MAP_NONBLOCK))
+@@ -1549,9 +1549,9 @@ static int acct_stack_growth(struct vm_a
return -ENOMEM;
/* Ok, everything looks good - let it rip */
@@ -26200,7 +26204,7 @@
vm_stat_account(mm, vma->vm_flags, vma->vm_file, grow);
return 0;
}
-@@ -1662,9 +1662,9 @@ static void remove_vma_list(struct mm_st
+@@ -1722,9 +1722,9 @@ static void remove_vma_list(struct mm_st
do {
long nrpages = vma_pages(vma);
@@ -26212,7 +26216,7 @@
vm_stat_account(mm, vma->vm_flags, vma->vm_file, -nrpages);
vma = remove_vma(vma);
} while (vma);
-@@ -1903,6 +1903,8 @@ unsigned long do_brk(unsigned long addr,
+@@ -1967,6 +1967,8 @@ unsigned long do_brk(unsigned long addr,
lock_limit >>= PAGE_SHIFT;
if (locked > lock_limit && !capable(CAP_IPC_LOCK))
return -EAGAIN;
@@ -26221,7 +26225,7 @@
}
/*
-@@ -1929,7 +1931,8 @@ unsigned long do_brk(unsigned long addr,
+@@ -1993,7 +1995,8 @@ unsigned long do_brk(unsigned long addr,
if (mm->map_count > sysctl_max_map_count)
return -ENOMEM;
@@ -26231,8 +26235,8 @@
return -ENOMEM;
/* Can we just expand an old private anonymous mapping? */
-@@ -1955,9 +1958,9 @@ unsigned long do_brk(unsigned long addr,
- (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)];
+@@ -2018,9 +2021,9 @@ unsigned long do_brk(unsigned long addr,
+ vma->vm_page_prot = vm_get_page_prot(flags);
vma_link(mm, vma, prev, rb_link, rb_parent);
out:
- mm->total_vm += len >> PAGE_SHIFT;
@@ -26243,7 +26247,7 @@
make_pages_present(addr, addr + len);
}
return addr;
-@@ -1986,6 +1989,11 @@ void exit_mmap(struct mm_struct *mm)
+@@ -2049,6 +2052,11 @@ void exit_mmap(struct mm_struct *mm)
free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, 0);
tlb_finish_mmu(tlb, 0, end);
@@ -26255,17 +26259,17 @@
/*
* Walk the list again, actually closing and freeing it,
* with preemption enabled, without holding any MM locks.
-@@ -2025,7 +2033,8 @@ int insert_vm_struct(struct mm_struct *
+@@ -2088,7 +2096,8 @@ int insert_vm_struct(struct mm_struct *
if (__vma && __vma->vm_start < vma->vm_end)
return -ENOMEM;
if ((vma->vm_flags & VM_ACCOUNT) &&
-- security_vm_enough_memory(vma_pages(vma)))
-+ (security_vm_enough_memory(vma_pages(vma)) ||
+- security_vm_enough_memory_mm(mm, vma_pages(vma)))
++ (security_vm_enough_memory_mm(mm, vma_pages(vma)) ||
+ !vx_vmpages_avail(mm, vma_pages(vma))))
return -ENOMEM;
vma_link(mm, vma, prev, rb_link, rb_parent);
return 0;
-@@ -2098,6 +2107,8 @@ int may_expand_vm(struct mm_struct *mm,
+@@ -2161,6 +2170,8 @@ int may_expand_vm(struct mm_struct *mm,
if (cur + npages > lim)
return 0;
@@ -26274,18 +26278,17 @@
return 1;
}
-@@ -2168,7 +2179,7 @@ int install_special_mapping(struct mm_st
+@@ -2238,7 +2249,6 @@ int install_special_mapping(struct mm_st
return -ENOMEM;
}
- mm->total_vm += len >> PAGE_SHIFT;
-
-+ // mm->total_vm += len >> PAGE_SHIFT;
+ vx_vmpages_add(mm, len >> PAGE_SHIFT);
return 0;
}
---- linux-2.6.22/mm/mremap.c 2007-02-06 03:01:56 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/mm/mremap.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/mm/mremap.c 2008-04-17 11:31:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/mremap.c 2008-04-19 15:14:52.000000000 -0400
@@ -18,6 +18,7 @@
#include <linux/highmem.h>
#include <linux/security.h>
@@ -26312,7 +26315,7 @@
if (new_len > old_len)
make_pages_present(new_addr + old_len,
new_addr + new_len);
-@@ -337,6 +338,9 @@ unsigned long do_mremap(unsigned long ad
+@@ -341,6 +342,9 @@ unsigned long do_mremap(unsigned long ad
ret = -EAGAIN;
if (locked > lock_limit && !capable(CAP_IPC_LOCK))
goto out;
@@ -26322,7 +26325,7 @@
}
if (!may_expand_vm(mm, (new_len - old_len) >> PAGE_SHIFT)) {
ret = -ENOMEM;
-@@ -365,10 +369,10 @@ unsigned long do_mremap(unsigned long ad
+@@ -369,10 +373,10 @@ unsigned long do_mremap(unsigned long ad
vma_adjust(vma, vma->vm_start,
addr + new_len, vma->vm_pgoff, NULL);
@@ -26335,9 +26338,9 @@
make_pages_present(addr + old_len,
addr + new_len);
}
---- linux-2.6.22/mm/nommu.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/nommu.c 2007-06-15 02:37:04 +0200
-@@ -936,7 +936,7 @@ unsigned long do_mmap_pgoff(struct file
+--- linux-2.6.25.4/mm/nommu.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/nommu.c 2008-04-19 15:14:52.000000000 -0400
+@@ -991,7 +991,7 @@ unsigned long do_mmap_pgoff(struct file
realalloc += kobjsize(vma);
askedalloc += sizeof(*vma);
@@ -26346,7 +26349,7 @@
add_nommu_vma(vma);
-@@ -1061,7 +1061,7 @@ int do_munmap(struct mm_struct *mm, unsi
+@@ -1117,7 +1117,7 @@ int do_munmap(struct mm_struct *mm, unsi
kfree(vml);
update_hiwater_vm(mm);
@@ -26355,7 +26358,7 @@
#ifdef DEBUG
show_process_blocks();
-@@ -1093,7 +1093,7 @@ void exit_mmap(struct mm_struct * mm)
+@@ -1150,7 +1150,7 @@ void exit_mmap(struct mm_struct * mm)
printk("Exit_mmap:\n");
#endif
@@ -26364,17 +26367,17 @@
while ((tmp = mm->context.vmlist)) {
mm->context.vmlist = tmp->next;
---- linux-2.6.22/mm/oom_kill.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/oom_kill.c 2007-06-15 02:37:04 +0200
-@@ -24,6 +24,7 @@
- #include <linux/cpuset.h>
+--- linux-2.6.25.4/mm/oom_kill.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/oom_kill.c 2008-04-19 16:55:20.000000000 -0400
+@@ -26,6 +26,7 @@
#include <linux/module.h>
#include <linux/notifier.h>
+ #include <linux/memcontrol.h>
+#include <linux/vs_memory.h>
int sysctl_panic_on_oom;
- /* #define DEBUG */
-@@ -66,6 +67,12 @@ unsigned long badness(struct task_struct
+ int sysctl_oom_kill_allocating_task;
+@@ -73,6 +74,12 @@ unsigned long badness(struct task_struct
points = mm->total_vm;
/*
@@ -26387,50 +26390,51 @@
* After this unlock we can no longer dereference local variable `mm'
*/
task_unlock(p);
-@@ -156,8 +163,8 @@ unsigned long badness(struct task_struct
+@@ -162,8 +169,8 @@ unsigned long badness(struct task_struct
}
#ifdef DEBUG
-- printk(KERN_DEBUG "OOMkill: task %d (%s) got %d points\n",
+- printk(KERN_DEBUG "OOMkill: task %d (%s) got %lu points\n",
- p->pid, p->comm, points);
+ printk(KERN_DEBUG "OOMkill: task %d:#%u (%s) got %d points\n",
-+ p->pid, p->xid, p->comm, points);
++ task_pid_nr(p), p->xid, p->comm, points);
#endif
return points;
}
-@@ -288,7 +295,8 @@ static void __oom_kill_task(struct task_
+@@ -322,8 +329,8 @@ static void __oom_kill_task(struct task_
}
if (verbose)
-- printk(KERN_ERR "Killed process %d (%s)\n", p->pid, p->comm);
+- printk(KERN_ERR "Killed process %d (%s)\n",
+- task_pid_nr(p), p->comm);
+ printk(KERN_ERR "Killed process %d:#%u (%s)\n",
-+ p->pid, p->xid, p->comm);
++ task_pid_nr(p), p->xid, p->comm);
/*
* We give our sacrificial lamb high priority and access to
-@@ -358,8 +366,8 @@ static int oom_kill_process(struct task_
+@@ -403,8 +410,8 @@ static int oom_kill_process(struct task_
return 0;
}
- printk(KERN_ERR "%s: kill process %d (%s) score %li or a child\n",
-- message, p->pid, p->comm, points);
+- message, task_pid_nr(p), p->comm, points);
+ printk(KERN_ERR "%s: kill process %d:#%u (%s) score %li or a child\n",
-+ message, p->pid, p->xid, p->comm, points);
++ message, task_pid_nr(p), p->xid, p->comm, points);
/* Try to kill a child first */
- list_for_each(tsk, &p->children) {
---- linux-2.6.22/mm/page_alloc.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/page_alloc.c 2007-06-17 05:54:20 +0200
-@@ -41,6 +41,8 @@
- #include <linux/pfn.h>
- #include <linux/backing-dev.h>
+ list_for_each_entry(c, &p->children, sibling) {
+--- linux-2.6.25.4/mm/page_alloc.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/page_alloc.c 2008-05-21 14:30:41.000000000 -0400
+@@ -45,6 +45,8 @@
#include <linux/fault-inject.h>
+ #include <linux/page-isolation.h>
+ #include <linux/memcontrol.h>
+#include <linux/vs_base.h>
+#include <linux/vs_limit.h>
#include <asm/tlbflush.h>
#include <asm/div64.h>
-@@ -1488,6 +1490,9 @@ void si_meminfo(struct sysinfo *val)
+@@ -1762,6 +1764,9 @@ void si_meminfo(struct sysinfo *val)
val->totalhigh = totalhigh_pages;
val->freehigh = nr_free_highpages();
val->mem_unit = PAGE_SIZE;
@@ -26440,7 +26444,7 @@
}
EXPORT_SYMBOL(si_meminfo);
-@@ -1508,6 +1513,9 @@ void si_meminfo_node(struct sysinfo *val
+@@ -1782,6 +1787,9 @@ void si_meminfo_node(struct sysinfo *val
val->freehigh = 0;
#endif
val->mem_unit = PAGE_SIZE;
@@ -26450,19 +26454,19 @@
}
#endif
---- linux-2.6.22/mm/rmap.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/rmap.c 2007-07-07 03:52:54 +0200
-@@ -48,6 +48,7 @@
- #include <linux/rcupdate.h>
+--- linux-2.6.25.4/mm/rmap.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/rmap.c 2008-04-19 16:53:36.000000000 -0400
+@@ -49,6 +49,7 @@
#include <linux/module.h>
#include <linux/kallsyms.h>
+ #include <linux/memcontrol.h>
+#include <linux/vs_memory.h>
#include <asm/tlbflush.h>
---- linux-2.6.22/mm/shmem.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/shmem.c 2007-06-17 05:54:20 +0200
-@@ -55,7 +55,6 @@
+--- linux-2.6.25.4/mm/shmem.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/shmem.c 2008-04-19 15:14:52.000000000 -0400
+@@ -56,7 +56,6 @@
#include <asm/pgtable.h>
/* This magic number is used in glibc for posix shared memory */
@@ -26470,7 +26474,7 @@
#define ENTRIES_PER_PAGE (PAGE_CACHE_SIZE/sizeof(unsigned long))
#define ENTRIES_PER_PAGEPAGE (ENTRIES_PER_PAGE*ENTRIES_PER_PAGE)
-@@ -1734,7 +1733,7 @@ static int shmem_statfs(struct dentry *d
+@@ -1773,7 +1772,7 @@ static int shmem_statfs(struct dentry *d
{
struct shmem_sb_info *sbinfo = SHMEM_SB(dentry->d_sb);
@@ -26479,7 +26483,7 @@
buf->f_bsize = PAGE_CACHE_SIZE;
buf->f_namelen = NAME_MAX;
spin_lock(&sbinfo->stat_lock);
-@@ -2308,7 +2307,7 @@ static int shmem_fill_super(struct super
+@@ -2341,7 +2340,7 @@ static int shmem_fill_super(struct super
sb->s_maxbytes = SHMEM_MAX_BYTES;
sb->s_blocksize = PAGE_CACHE_SIZE;
sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
@@ -26488,9 +26492,9 @@
sb->s_op = &shmem_ops;
sb->s_time_gran = 1;
#ifdef CONFIG_TMPFS_POSIX_ACL
---- linux-2.6.22/mm/slab.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/slab.c 2007-07-09 13:11:54 +0200
-@@ -510,6 +510,8 @@ struct kmem_cache {
+--- linux-2.6.25.4/mm/slab.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/slab.c 2008-04-19 15:14:52.000000000 -0400
+@@ -509,6 +509,8 @@ struct kmem_cache {
#define STATS_INC_FREEMISS(x) do { } while (0)
#endif
@@ -26499,7 +26503,7 @@
#if DEBUG
/*
-@@ -3314,6 +3316,7 @@ retry:
+@@ -3344,6 +3346,7 @@ retry:
obj = slab_get_obj(cachep, slabp, nodeid);
check_slabp(cachep, slabp);
@@ -26507,7 +26511,7 @@
l3->free_objects--;
/* move slabp to correct slabp list: */
list_del(&slabp->list);
-@@ -3386,6 +3389,7 @@ __cache_alloc_node(struct kmem_cache *ca
+@@ -3416,6 +3419,7 @@ __cache_alloc_node(struct kmem_cache *ca
/* ___cache_alloc_node can fall back to other nodes */
ptr = ____cache_alloc_node(cachep, flags, nodeid);
out:
@@ -26515,16 +26519,16 @@
local_irq_restore(save_flags);
ptr = cache_alloc_debugcheck_after(cachep, flags, ptr, caller);
-@@ -3551,6 +3555,7 @@ static inline void __cache_free(struct k
+@@ -3587,6 +3591,7 @@ static inline void __cache_free(struct k
check_irq_off();
objp = cache_free_debugcheck(cachep, objp, __builtin_return_address(0));
+ vx_slab_free(cachep);
- if (cache_free_alien(cachep, objp))
- return;
---- linux-2.6.22/mm/slab_vs.h 1970-01-01 01:00:00 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/mm/slab_vs.h 2007-06-15 02:37:04 +0200
+ /*
+ * Skip calling cache_free_alien() when the platform is not numa.
+--- linux-2.6.25.4/mm/slab_vs.h 1969-12-31 19:00:00.000000000 -0500
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/slab_vs.h 2008-04-19 15:14:52.000000000 -0400
@@ -0,0 +1,27 @@
+
+#include <linux/vserver/context.h>
@@ -26553,9 +26557,9 @@
+ atomic_sub(cachep->buffer_size, ¤t->vx_info->cacct.slab[what]);
+}
+
---- linux-2.6.22/mm/swapfile.c 2007-07-09 13:20:04 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/mm/swapfile.c 2007-06-15 02:37:04 +0200
-@@ -31,6 +31,8 @@
+--- linux-2.6.25.4/mm/swapfile.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/mm/swapfile.c 2008-04-19 15:14:52.000000000 -0400
+@@ -32,6 +32,8 @@
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
#include <linux/swapops.h>
@@ -26564,7 +26568,7 @@
DEFINE_SPINLOCK(swap_lock);
unsigned int nr_swapfiles;
-@@ -1712,6 +1714,8 @@ void si_swapinfo(struct sysinfo *val)
+@@ -1743,6 +1745,8 @@ void si_swapinfo(struct sysinfo *val)
val->freeswap = nr_swap_pages + nr_to_be_unused;
val->totalswap = total_swap_pages + nr_to_be_unused;
spin_unlock(&swap_lock);
@@ -26573,101 +26577,100 @@
}
/*
---- linux-2.6.22/net/core/dev.c 2007-07-09 13:20:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/core/dev.c 2007-07-07 03:52:54 +0200
-@@ -117,6 +117,8 @@
+--- linux-2.6.25.4/net/core/dev.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/core/dev.c 2008-04-19 15:14:52.000000000 -0400
+@@ -119,6 +119,7 @@
#include <linux/err.h>
#include <linux/ctype.h>
#include <linux/if_arp.h>
-+#include <linux/vs_context.h> /* remove with NXF_HIDE_NETIF */
-+#include <linux/vs_network.h>
++#include <linux/vs_inet.h>
- /*
- * The list of packet types we will receive (as opposed to discard)
-@@ -2120,6 +2122,9 @@ static int dev_ifconf(char __user *arg)
+ #include "net-sysfs.h"
+
+@@ -2336,6 +2337,8 @@ static int dev_ifconf(struct net *net, c
total = 0;
- for_each_netdev(dev) {
-+ if (vx_flags(VXF_HIDE_NETIF, 0) &&
-+ !dev_in_nx_info(dev, current->nx_info))
+ for_each_netdev(net, dev) {
++ if (!nx_dev_visible(current->nx_info, dev))
+ continue;
for (i = 0; i < NPROTO; i++) {
if (gifconf_list[i]) {
int done;
-@@ -2183,6 +2188,10 @@ void dev_seq_stop(struct seq_file *seq,
- static void dev_seq_printf_stats(struct seq_file *seq, struct net_device *dev)
+@@ -2404,6 +2407,9 @@ static void dev_seq_printf_stats(struct
{
struct net_device_stats *stats = dev->get_stats(dev);
-+ struct nx_info *nxi = current->nx_info;
-+
-+ if (vx_flags(VXF_HIDE_NETIF, 0) && !dev_in_nx_info(dev, nxi))
-+ return;
++ if (!nx_dev_visible(current->nx_info, dev))
++ return;
++
seq_printf(seq, "%6s:%8lu %7lu %4lu %4lu %4lu %5lu %10lu %9lu "
"%8lu %7lu %4lu %4lu %4lu %5lu %7lu %10lu\n",
---- linux-2.6.22/net/core/rtnetlink.c 2007-07-09 13:20:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/core/rtnetlink.c 2007-06-17 05:54:20 +0200
-@@ -35,6 +35,7 @@
- #include <linux/security.h>
- #include <linux/mutex.h>
- #include <linux/if_addr.h>
-+#include <linux/vs_context.h> /* remove with NXF_HIDE_NETIF */
-
- #include <asm/uaccess.h>
- #include <asm/system.h>
-@@ -537,6 +538,9 @@ static int rtnl_dump_ifinfo(struct sk_bu
+ dev->name, stats->rx_bytes, stats->rx_packets,
+--- linux-2.6.25.4/net/core/rtnetlink.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/core/rtnetlink.c 2008-05-21 14:30:41.000000000 -0400
+@@ -674,6 +674,8 @@ static int rtnl_dump_ifinfo(struct sk_bu
idx = 0;
- for_each_netdev(dev) {
-+ if (vx_info_flags(skb->sk->sk_vx_info, VXF_HIDE_NETIF, 0) &&
-+ !dev_in_nx_info(dev, skb->sk->sk_nx_info))
+ for_each_netdev(net, dev) {
++ if (!nx_dev_visible(skb->sk->sk_nx_info, dev))
+ continue;
if (idx < s_idx)
goto cont;
if (rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK,
---- linux-2.6.22/net/core/sock.c 2007-07-09 13:20:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/core/sock.c 2007-06-15 02:37:04 +0200
-@@ -125,6 +125,9 @@
+@@ -1207,6 +1209,9 @@ void rtmsg_ifinfo(int type, struct net_d
+ struct sk_buff *skb;
+ int err = -ENOBUFS;
+
++ if (!nx_dev_visible(current->nx_info, dev))
++ return;
++
+ skb = nlmsg_new(if_nlmsg_size(dev), GFP_KERNEL);
+ if (skb == NULL)
+ goto errout;
+--- linux-2.6.25.4/net/core/sock.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/core/sock.c 2008-04-23 14:31:31.000000000 -0400
+@@ -126,6 +126,10 @@
#include <linux/ipsec.h>
#include <linux/filter.h>
+#include <linux/vs_socket.h>
+#include <linux/vs_limit.h>
+#include <linux/vs_context.h>
++#include <linux/vs_network.h>
#ifdef CONFIG_INET
#include <net/tcp.h>
-@@ -869,6 +872,8 @@ struct sock *sk_alloc(int family, gfp_t
- sk->sk_prot = sk->sk_prot_creator = prot;
- sock_lock_init(sk);
- }
+@@ -907,6 +911,8 @@ static struct sock *sk_prot_alloc(struct
+ if (!try_module_get(prot->owner))
+ goto out_free_sec;
+ }
+ sock_vx_init(sk);
+ sock_nx_init(sk);
- if (security_sk_alloc(sk, family, priority))
- goto out_free;
-@@ -907,6 +912,11 @@ void sk_free(struct sock *sk)
+ return sk;
+
+@@ -984,6 +990,11 @@ void sk_free(struct sock *sk)
__FUNCTION__, atomic_read(&sk->sk_omem_alloc));
- security_sk_free(sk);
+ put_net(sk->sk_net);
+ vx_sock_dec(sk);
+ clr_vx_info(&sk->sk_vx_info);
+ sk->sk_xid = -1;
+ clr_nx_info(&sk->sk_nx_info);
+ sk->sk_nid = -1;
- if (sk->sk_prot_creator->slab != NULL)
- kmem_cache_free(sk->sk_prot_creator->slab, sk);
- else
-@@ -924,6 +934,8 @@ struct sock *sk_clone(const struct sock
- sock_copy(newsk, sk);
+ sk_prot_free(sk->sk_prot_creator, sk);
+ }
+
+@@ -999,6 +1010,8 @@ struct sock *sk_clone(const struct sock
/* SANITY */
+ get_net(newsk->sk_net);
+ sock_vx_init(newsk);
+ sock_nx_init(newsk);
sk_node_init(&newsk->sk_node);
sock_lock_init(newsk);
bh_lock_sock(newsk);
-@@ -969,6 +981,12 @@ struct sock *sk_clone(const struct sock
+@@ -1045,6 +1058,12 @@ struct sock *sk_clone(const struct sock
newsk->sk_priority = 0;
atomic_set(&newsk->sk_refcnt, 2);
@@ -26680,9 +26683,9 @@
/*
* Increment the counter in the same struct proto as the master
* sock (sk_refcnt_debug_inc uses newsk->sk_prot->socks, that
-@@ -1551,6 +1569,11 @@ void sock_init_data(struct socket *sock,
+@@ -1727,6 +1746,11 @@ void sock_init_data(struct socket *sock,
- sk->sk_stamp = ktime_set(-1L, -1L);
+ sk->sk_stamp = ktime_set(-1L, 0);
+ set_vx_info(&sk->sk_vx_info, current->vx_info);
+ sk->sk_xid = vx_current_xid();
@@ -26690,10 +26693,10 @@
+ set_nx_info(&sk->sk_nx_info, current->nx_info);
+ sk->sk_nid = nx_current_nid();
atomic_set(&sk->sk_refcnt, 1);
+ atomic_set(&sk->sk_drops, 0);
}
-
---- linux-2.6.22/net/ipv4/af_inet.c 2007-07-09 13:20:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/af_inet.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/net/ipv4/af_inet.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/af_inet.c 2008-04-19 16:08:42.000000000 -0400
@@ -115,6 +115,7 @@
#ifdef CONFIG_IP_MROUTE
#include <linux/mroute.h>
@@ -26702,11 +26705,12 @@
DEFINE_SNMP_STAT(struct linux_mib, net_statistics) __read_mostly;
-@@ -308,9 +309,11 @@ lookup_protocol:
+@@ -317,9 +318,12 @@ lookup_protocol:
}
err = -EPERM;
-+ if ((protocol == IPPROTO_ICMP) && vx_ccaps(VXC_RAW_ICMP))
++ if ((protocol == IPPROTO_ICMP) &&
++ nx_capable(answer->capability, NXC_RAW_ICMP))
+ goto override;
if (answer->capability > 0 && !capable(answer->capability))
goto out_rcu_unlock;
@@ -26715,114 +26719,95 @@
sock->ops = answer->ops;
answer_prot = answer->prot;
answer_no_check = answer->no_check;
-@@ -427,6 +430,10 @@ int inet_bind(struct socket *sock, struc
+@@ -433,6 +437,7 @@ int inet_bind(struct socket *sock, struc
+ struct sockaddr_in *addr = (struct sockaddr_in *)uaddr;
+ struct sock *sk = sock->sk;
+ struct inet_sock *inet = inet_sk(sk);
++ struct nx_v4_sock_addr nsa;
unsigned short snum;
int chk_addr_ret;
int err;
-+ __u32 s_addr; /* Address used for validation */
-+ __u32 s_addr1; /* Address used for socket */
-+ __u32 s_addr2; /* Broadcast address for the socket */
-+ struct nx_info *nxi = sk->sk_nx_info;
-
- /* If the socket has its own bind function then use it. (RAW) */
- if (sk->sk_prot->bind) {
-@@ -437,7 +444,40 @@ int inet_bind(struct socket *sock, struc
+@@ -446,7 +451,11 @@ int inet_bind(struct socket *sock, struc
if (addr_len < sizeof(struct sockaddr_in))
goto out;
-- chk_addr_ret = inet_addr_type(addr->sin_addr.s_addr);
-+ s_addr = addr->sin_addr.s_addr;
-+ s_addr1 = s_addr;
-+ s_addr2 = 0xffffffffl;
-+
-+ vxdprintk(VXD_CBIT(net, 3),
-+ "inet_bind(%p)* %p,%p;%lx " NIPQUAD_FMT,
-+ sk, sk->sk_nx_info, sk->sk_socket,
-+ (sk->sk_socket?sk->sk_socket->flags:0),
-+ NIPQUAD(s_addr));
-+ if (nxi) {
-+ __u32 v4_bcast = nxi->v4_bcast;
-+ __u32 ipv4root = nxi->ipv4[0];
-+ int nbipv4 = nxi->nbipv4;
-+
-+ if (s_addr == 0) {
-+ /* bind to any for 1-n */
-+ s_addr = ipv4root;
-+ s_addr1 = (nbipv4 > 1) ? 0 : s_addr;
-+ s_addr2 = v4_bcast;
-+ } else if (s_addr == IPI_LOOPBACK) {
-+ /* rewrite localhost to ipv4root */
-+ s_addr = ipv4root;
-+ s_addr1 = ipv4root;
-+ } else if (s_addr != v4_bcast) {
-+ /* normal address bind */
-+ if (!addr_in_nx_info(nxi, s_addr))
-+ return -EADDRNOTAVAIL;
-+ }
-+ }
-+ chk_addr_ret = inet_addr_type(s_addr);
+- chk_addr_ret = inet_addr_type(&init_net, addr->sin_addr.s_addr);
++ err = v4_map_sock_addr(inet, addr, &nsa);
++ if (err)
++ goto out;
+
-+ vxdprintk(VXD_CBIT(net, 3),
-+ "inet_bind(%p) " NIPQUAD_FMT ", " NIPQUAD_FMT ", " NIPQUAD_FMT,
-+ sk, NIPQUAD(s_addr), NIPQUAD(s_addr1), NIPQUAD(s_addr2));
++ chk_addr_ret = inet_addr_type(&init_net, nsa.saddr);
/* Not specified by any standard per-se, however it breaks too
* many applications when removed. It is unfortunate since
-@@ -449,7 +489,7 @@ int inet_bind(struct socket *sock, struc
+@@ -458,7 +467,7 @@ int inet_bind(struct socket *sock, struc
err = -EADDRNOTAVAIL;
if (!sysctl_ip_nonlocal_bind &&
!inet->freebind &&
-- addr->sin_addr.s_addr != INADDR_ANY &&
-+ s_addr != INADDR_ANY &&
+- addr->sin_addr.s_addr != htonl(INADDR_ANY) &&
++ nsa.saddr != htonl(INADDR_ANY) &&
chk_addr_ret != RTN_LOCAL &&
chk_addr_ret != RTN_MULTICAST &&
chk_addr_ret != RTN_BROADCAST)
-@@ -474,7 +514,8 @@ int inet_bind(struct socket *sock, struc
+@@ -483,7 +492,7 @@ int inet_bind(struct socket *sock, struc
if (sk->sk_state != TCP_CLOSE || inet->num)
goto out_release_sock;
- inet->rcv_saddr = inet->saddr = addr->sin_addr.s_addr;
-+ inet->rcv_saddr = inet->saddr = s_addr1;
-+ inet->rcv_saddr2 = s_addr2;
++ v4_set_sock_addr(inet, &nsa);
if (chk_addr_ret == RTN_MULTICAST || chk_addr_ret == RTN_BROADCAST)
inet->saddr = 0; /* Use device */
---- linux-2.6.22/net/ipv4/devinet.c 2007-07-09 13:20:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/devinet.c 2007-06-17 05:54:20 +0200
-@@ -56,6 +56,7 @@
- #include <linux/sysctl.h>
- #endif
- #include <linux/kmod.h>
-+#include <linux/vs_context.h>
+@@ -676,11 +685,13 @@ int inet_getname(struct socket *sock, st
+ peer == 1))
+ return -ENOTCONN;
+ sin->sin_port = inet->dport;
+- sin->sin_addr.s_addr = inet->daddr;
++ sin->sin_addr.s_addr =
++ nx_map_sock_lback(sk->sk_nx_info, inet->daddr);
+ } else {
+ __be32 addr = inet->rcv_saddr;
+ if (!addr)
+ addr = inet->saddr;
++ addr = nx_map_sock_lback(sk->sk_nx_info, addr);
+ sin->sin_port = inet->sport;
+ sin->sin_addr.s_addr = addr;
+ }
+--- linux-2.6.25.4/net/ipv4/devinet.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/devinet.c 2008-04-19 15:14:52.000000000 -0400
+@@ -421,6 +421,7 @@ struct in_device *inetdev_by_index(struc
+ return in_dev;
+ }
- #include <net/arp.h>
- #include <net/ip.h>
-@@ -676,6 +677,9 @@ int devinet_ioctl(unsigned int cmd, void
++
+ /* Called only from RTNL semaphored context. No locks. */
+
+ struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix,
+@@ -672,6 +673,8 @@ int devinet_ioctl(unsigned int cmd, void
*colon = ':';
if ((in_dev = __in_dev_get_rtnl(dev)) != NULL) {
+ struct nx_info *nxi = current->nx_info;
-+ int hide_netif = vx_flags(VXF_HIDE_NETIF, 0);
+
if (tryaddrmatch) {
/* Matthias Andree */
/* compare label and address (4.4BSD style) */
-@@ -684,6 +688,8 @@ int devinet_ioctl(unsigned int cmd, void
+@@ -680,6 +683,8 @@ int devinet_ioctl(unsigned int cmd, void
This is checked above. */
for (ifap = &in_dev->ifa_list; (ifa = *ifap) != NULL;
ifap = &ifa->ifa_next) {
-+ if (hide_netif && !ifa_in_nx_info(ifa, nxi))
++ if (!nx_v4_ifa_visible(nxi, ifa))
+ continue;
if (!strcmp(ifr.ifr_name, ifa->ifa_label) &&
sin_orig.sin_addr.s_addr ==
ifa->ifa_address) {
-@@ -696,9 +702,12 @@ int devinet_ioctl(unsigned int cmd, void
+@@ -692,9 +697,12 @@ int devinet_ioctl(unsigned int cmd, void
comparing just the label */
if (!ifa) {
for (ifap = &in_dev->ifa_list; (ifa = *ifap) != NULL;
- ifap = &ifa->ifa_next)
+ ifap = &ifa->ifa_next) {
-+ if (hide_netif && !ifa_in_nx_info(ifa, nxi))
++ if (!nx_v4_ifa_visible(nxi, ifa))
+ continue;
if (!strcmp(ifr.ifr_name, ifa->ifa_label))
break;
@@ -26830,66 +26815,80 @@
}
}
-@@ -849,6 +858,9 @@ static int inet_gifconf(struct net_devic
+@@ -846,6 +854,8 @@ static int inet_gifconf(struct net_devic
goto out;
for (; ifa; ifa = ifa->ifa_next) {
-+ if (vx_flags(VXF_HIDE_NETIF, 0) &&
-+ !ifa_in_nx_info(ifa, current->nx_info))
++ if (!nx_v4_ifa_visible(current->nx_info, ifa))
+ continue;
if (!buf) {
done += sizeof(ifr);
continue;
-@@ -1179,6 +1191,7 @@ static int inet_dump_ifaddr(struct sk_bu
+@@ -1171,6 +1181,7 @@ static int inet_dump_ifaddr(struct sk_bu
struct net_device *dev;
struct in_device *in_dev;
struct in_ifaddr *ifa;
+ struct sock *sk = skb->sk;
int s_ip_idx, s_idx = cb->args[0];
- s_ip_idx = ip_idx = cb->args[1];
-@@ -1193,6 +1206,9 @@ static int inet_dump_ifaddr(struct sk_bu
-
- for (ifa = in_dev->ifa_list, ip_idx = 0; ifa;
- ifa = ifa->ifa_next, ip_idx++) {
-+ if (sk && vx_info_flags(sk->sk_vx_info, VXF_HIDE_NETIF, 0) &&
-+ !ifa_in_nx_info(ifa, sk->sk_nx_info))
-+ continue;
- if (ip_idx < s_ip_idx)
- goto cont;
- if (inet_fill_ifaddr(skb, ifa, NETLINK_CB(cb->skb).pid,
---- linux-2.6.22/net/ipv4/fib_hash.c 2007-07-09 13:20:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/fib_hash.c 2007-06-15 02:37:04 +0200
-@@ -34,6 +34,7 @@
- #include <linux/skbuff.h>
- #include <linux/netlink.h>
- #include <linux/init.h>
-+#include <linux/vs_context.h>
-
- #include <net/ip.h>
- #include <net/protocol.h>
-@@ -982,6 +983,8 @@ static unsigned fib_flag_trans(int type,
- return flags;
- }
-
-+extern int dev_in_nx_info(struct net_device *, struct nx_info *);
-+
- /*
- * This outputs /proc/net/route.
- *
-@@ -1012,7 +1015,8 @@ static int fib_seq_show(struct seq_file
+ if (net != &init_net)
+@@ -1188,6 +1199,8 @@ static int inet_dump_ifaddr(struct sk_bu
+
+ for (ifa = in_dev->ifa_list, ip_idx = 0; ifa;
+ ifa = ifa->ifa_next, ip_idx++) {
++ if (sk && !nx_v4_ifa_visible(sk->sk_nx_info, ifa))
++ continue;
+ if (ip_idx < s_ip_idx)
+ continue;
+ if (inet_fill_ifaddr(skb, ifa, NETLINK_CB(cb->skb).pid,
+--- linux-2.6.25.4/net/ipv4/fib_hash.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/fib_hash.c 2008-04-19 15:14:52.000000000 -0400
+@@ -1025,7 +1025,7 @@ static int fib_seq_show(struct seq_file
prefix = f->fn_key;
mask = FZ_MASK(iter->zone);
flags = fib_flag_trans(fa->fa_type, mask, fi);
- if (fi)
-+ if (fi && (!vx_flags(VXF_HIDE_NETIF, 0) ||
-+ dev_in_nx_info(fi->fib_dev, current->nx_info)))
++ if (fi && nx_dev_visible(current->nx_info, fi->fib_dev))
snprintf(bf, sizeof(bf),
"%s\t%08X\t%08X\t%04X\t%d\t%u\t%d\t%08X\t%d\t%u\t%u",
fi->fib_dev ? fi->fib_dev->name : "*", prefix,
---- linux-2.6.22/net/ipv4/inet_connection_sock.c 2007-07-09 13:20:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/inet_connection_sock.c 2007-06-15 02:37:04 +0200
-@@ -37,7 +37,6 @@ int sysctl_local_port_range[2] = { 32768
+--- linux-2.6.25.4/net/ipv4/inet_connection_sock.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/inet_connection_sock.c 2008-04-19 15:14:52.000000000 -0400
+@@ -47,10 +47,40 @@ void inet_get_local_port_range(int *low,
+ }
+ EXPORT_SYMBOL(inet_get_local_port_range);
+
++int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
++{
++ __be32 sk1_rcv_saddr = inet_rcv_saddr(sk1),
++ sk2_rcv_saddr = inet_rcv_saddr(sk2);
++
++ if (inet_v6_ipv6only(sk2))
++ return 0;
++
++ if (sk1_rcv_saddr &&
++ sk2_rcv_saddr &&
++ sk1_rcv_saddr == sk2_rcv_saddr)
++ return 1;
++
++ if (sk1_rcv_saddr &&
++ !sk2_rcv_saddr &&
++ v4_addr_in_nx_info(sk2->sk_nx_info, sk1_rcv_saddr, NXA_MASK_BIND))
++ return 1;
++
++ if (sk2_rcv_saddr &&
++ !sk1_rcv_saddr &&
++ v4_addr_in_nx_info(sk1->sk_nx_info, sk2_rcv_saddr, NXA_MASK_BIND))
++ return 1;
++
++ if (!sk1_rcv_saddr &&
++ !sk2_rcv_saddr &&
++ nx_v4_addr_conflict(sk1->sk_nx_info, sk2->sk_nx_info))
++ return 1;
++
++ return 0;
++}
++
int inet_csk_bind_conflict(const struct sock *sk,
const struct inet_bind_bucket *tb)
{
@@ -26897,21 +26896,94 @@
struct sock *sk2;
struct hlist_node *node;
int reuse = sk->sk_reuse;
-@@ -50,9 +49,8 @@ int inet_csk_bind_conflict(const struct
+@@ -63,9 +93,7 @@ int inet_csk_bind_conflict(const struct
sk->sk_bound_dev_if == sk2->sk_bound_dev_if)) {
if (!reuse || !sk2->sk_reuse ||
sk2->sk_state == TCP_LISTEN) {
- const __be32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
- if (!sk2_rcv_saddr || !sk_rcv_saddr ||
- sk2_rcv_saddr == sk_rcv_saddr)
-+ if (nx_addr_conflict(sk->sk_nx_info,
-+ inet_rcv_saddr(sk), sk2))
++ if (ipv4_rcv_saddr_equal(sk, sk2))
break;
}
}
---- linux-2.6.22/net/ipv4/inet_diag.c 2007-07-09 13:20:05 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/inet_diag.c 2007-06-15 02:37:04 +0200
-@@ -697,6 +697,8 @@ static int inet_diag_dump(struct sk_buff
+--- linux-2.6.25.4/net/ipv4/inet_diag.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/inet_diag.c 2008-04-19 16:05:51.000000000 -0400
+@@ -34,6 +34,8 @@
+ #include <linux/stddef.h>
+
+ #include <linux/inet_diag.h>
++#include <linux/vs_network.h>
++#include <linux/vs_inet.h>
+
+ static const struct inet_diag_handler **inet_diag_table;
+
+@@ -122,8 +124,8 @@ static int inet_csk_diag_fill(struct soc
+
+ r->id.idiag_sport = inet->sport;
+ r->id.idiag_dport = inet->dport;
+- r->id.idiag_src[0] = inet->rcv_saddr;
+- r->id.idiag_dst[0] = inet->daddr;
++ r->id.idiag_src[0] = nx_map_sock_lback(sk->sk_nx_info, inet->rcv_saddr);
++ r->id.idiag_dst[0] = nx_map_sock_lback(sk->sk_nx_info, inet->daddr);
+
+ #if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
+ if (r->idiag_family == AF_INET6) {
+@@ -210,8 +212,8 @@ static int inet_twsk_diag_fill(struct in
+ r->id.idiag_cookie[1] = (u32)(((unsigned long)tw >> 31) >> 1);
+ r->id.idiag_sport = tw->tw_sport;
+ r->id.idiag_dport = tw->tw_dport;
+- r->id.idiag_src[0] = tw->tw_rcv_saddr;
+- r->id.idiag_dst[0] = tw->tw_daddr;
++ r->id.idiag_src[0] = nx_map_sock_lback(tw->tw_nx_info, tw->tw_rcv_saddr);
++ r->id.idiag_dst[0] = nx_map_sock_lback(tw->tw_nx_info, tw->tw_daddr);
+ r->idiag_state = tw->tw_substate;
+ r->idiag_timer = 3;
+ r->idiag_expires = DIV_ROUND_UP(tmo * 1000, HZ);
+@@ -268,6 +270,7 @@ static int inet_diag_get_exact(struct sk
+ err = -EINVAL;
+
+ if (req->idiag_family == AF_INET) {
++ /* TODO: lback */
+ sk = inet_lookup(&init_net, hashinfo, req->id.idiag_dst[0],
+ req->id.idiag_dport, req->id.idiag_src[0],
+ req->id.idiag_sport, req->id.idiag_if);
+@@ -510,6 +513,7 @@ static int inet_csk_diag_dump(struct soc
+ } else
+ #endif
+ {
++ /* TODO: lback */
+ entry.saddr = &inet->rcv_saddr;
+ entry.daddr = &inet->daddr;
+ }
+@@ -546,6 +550,7 @@ static int inet_twsk_diag_dump(struct in
+ } else
+ #endif
+ {
++ /* TODO: lback */
+ entry.saddr = &tw->tw_rcv_saddr;
+ entry.daddr = &tw->tw_daddr;
+ }
+@@ -592,8 +597,8 @@ static int inet_diag_fill_req(struct sk_
+
+ r->id.idiag_sport = inet->sport;
+ r->id.idiag_dport = ireq->rmt_port;
+- r->id.idiag_src[0] = ireq->loc_addr;
+- r->id.idiag_dst[0] = ireq->rmt_addr;
++ r->id.idiag_src[0] = nx_map_sock_lback(sk->sk_nx_info, ireq->loc_addr);
++ r->id.idiag_dst[0] = nx_map_sock_lback(sk->sk_nx_info, ireq->rmt_addr);
+ r->idiag_expires = jiffies_to_msecs(tmo);
+ r->idiag_rqueue = 0;
+ r->idiag_wqueue = 0;
+@@ -663,6 +668,7 @@ static int inet_diag_dump_reqs(struct sk
+ continue;
+
+ if (bc) {
++ /* TODO: lback */
+ entry.saddr =
+ #if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
+ (entry.family == AF_INET6) ?
+@@ -731,6 +737,8 @@ static int inet_diag_dump(struct sk_buff
sk_for_each(sk, node, &hashinfo->listening_hash[i]) {
struct inet_sock *inet = inet_sk(sk);
@@ -26920,7 +26992,7 @@
if (num < s_num) {
num++;
continue;
-@@ -757,6 +759,8 @@ skip_listen_ht:
+@@ -792,6 +800,8 @@ skip_listen_ht:
sk_for_each(sk, node, &head->chain) {
struct inet_sock *inet = inet_sk(sk);
@@ -26929,7 +27001,7 @@
if (num < s_num)
goto next_normal;
if (!(r->idiag_states & (1 << sk->sk_state)))
-@@ -781,6 +785,8 @@ next_normal:
+@@ -816,6 +826,8 @@ next_normal:
inet_twsk_for_each(tw, node,
&head->twchain) {
@@ -26938,16 +27010,24 @@
if (num < s_num)
goto next_dying;
if (r->id.idiag_sport != tw->tw_sport &&
---- linux-2.6.22/net/ipv4/inet_hashtables.c 2007-05-02 19:25:40 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/inet_hashtables.c 2007-06-15 02:37:04 +0200
-@@ -140,11 +140,10 @@ static struct sock *inet_lookup_listener
+--- linux-2.6.25.4/net/ipv4/inet_hashtables.c 2008-04-17 12:05:44.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/inet_hashtables.c 2008-04-19 15:14:52.000000000 -0400
+@@ -21,6 +21,7 @@
+
+ #include <net/inet_connection_sock.h>
+ #include <net/inet_hashtables.h>
++#include <net/route.h>
+ #include <net/ip.h>
+
+ /*
+@@ -144,11 +145,10 @@ static struct sock *inet_lookup_listener
const __be32 rcv_saddr = inet->rcv_saddr;
int score = sk->sk_family == PF_INET ? 1 : 0;
- if (rcv_saddr) {
- if (rcv_saddr != daddr)
- continue;
-+ if (inet_addr_match(sk->sk_nx_info, daddr, rcv_saddr))
++ if (v4_inet_addr_match(sk->sk_nx_info, daddr, rcv_saddr))
score += 2;
- }
+ else
@@ -26955,68 +27035,61 @@
if (sk->sk_bound_dev_if) {
if (sk->sk_bound_dev_if != dif)
continue;
-@@ -175,7 +174,7 @@ struct sock *__inet_lookup_listener(stru
+@@ -180,7 +180,7 @@ struct sock *__inet_lookup_listener(stru
const struct inet_sock *inet = inet_sk((sk = __sk_head(head)));
if (inet->num == hnum && !sk->sk_node.next &&
- (!inet->rcv_saddr || inet->rcv_saddr == daddr) &&
-+ inet_addr_match(sk->sk_nx_info, daddr, inet->rcv_saddr) &&
++ v4_inet_addr_match(sk->sk_nx_info, daddr, inet->rcv_saddr) &&
(sk->sk_family == PF_INET || !ipv6_only_sock(sk)) &&
- !sk->sk_bound_dev_if)
+ !sk->sk_bound_dev_if && sk->sk_net == net)
goto sherry_cache;
---- linux-2.6.22/net/ipv4/raw.c 2007-07-09 13:20:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/raw.c 2007-06-15 02:37:04 +0200
-@@ -101,6 +101,27 @@ static void raw_v4_unhash(struct sock *s
- write_unlock_bh(&raw_v4_lock);
- }
+--- linux-2.6.25.4/net/ipv4/netfilter/nf_nat_helper.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/netfilter/nf_nat_helper.c 2008-04-19 15:14:52.000000000 -0400
+@@ -18,6 +18,7 @@
+ #include <net/tcp.h>
-+
-+/*
-+ * Check if a given address matches for a socket
-+ *
-+ * nxi: the socket's nx_info if any
-+ * addr: to be verified address
-+ * saddr/baddr: socket addresses
-+ */
-+static inline int raw_addr_match (
-+ struct nx_info *nxi,
-+ uint32_t addr,
-+ uint32_t saddr,
-+ uint32_t baddr)
-+{
-+ if (addr && (saddr == addr || baddr == addr))
-+ return 1;
-+ if (!saddr)
-+ return addr_in_nx_info(nxi, addr);
-+ return 0;
-+}
-+
- struct sock *__raw_v4_lookup(struct sock *sk, unsigned short num,
- __be32 raddr, __be32 laddr,
- int dif)
-@@ -112,7 +133,8 @@ struct sock *__raw_v4_lookup(struct sock
+ #include <linux/netfilter_ipv4.h>
++#include <net/route.h>
+ #include <net/netfilter/nf_conntrack.h>
+ #include <net/netfilter/nf_conntrack_helper.h>
+ #include <net/netfilter/nf_conntrack_ecache.h>
+--- linux-2.6.25.4/net/ipv4/netfilter.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/netfilter.c 2008-04-19 15:14:52.000000000 -0400
+@@ -4,7 +4,7 @@
+ #include <linux/netfilter_ipv4.h>
+ #include <linux/ip.h>
+ #include <linux/skbuff.h>
+-#include <net/route.h>
++// #include <net/route.h>
+ #include <net/xfrm.h>
+ #include <net/ip.h>
+ #include <net/netfilter/nf_queue.h>
+--- linux-2.6.25.4/net/ipv4/raw.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/raw.c 2008-04-29 20:28:52.000000000 -0400
+@@ -126,7 +126,7 @@ static struct sock *__raw_v4_lookup(stru
- if (inet->num == num &&
+ if (sk->sk_net == net && inet->num == num &&
!(inet->daddr && inet->daddr != raddr) &&
- !(inet->rcv_saddr && inet->rcv_saddr != laddr) &&
-+ raw_addr_match(sk->sk_nx_info, laddr,
-+ inet->rcv_saddr, inet->rcv_saddr2) &&
++ v4_sock_addr_match(sk->sk_nx_info, inet, laddr) &&
!(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif))
goto found; /* gotcha */
}
-@@ -314,6 +336,11 @@ static int raw_send_hdrinc(struct sock *
- iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
- }
+@@ -382,6 +382,12 @@ static int raw_send_hdrinc(struct sock *
+ icmp_out_count(((struct icmphdr *)
+ skb_transport_header(skb))->type);
+ err = -EPERM;
-+ if (!nx_check(0, VS_ADMIN) && !capable(CAP_NET_RAW)
-+ && (!addr_in_nx_info(sk->sk_nx_info, iph->saddr)))
++ if (!nx_check(0, VS_ADMIN) && !capable(CAP_NET_RAW) &&
++ sk->sk_nx_info &&
++ !v4_addr_in_nx_info(sk->sk_nx_info, iph->saddr, NXA_MASK_BIND))
+ goto error_free;
+
- err = NF_HOOK(PF_INET, NF_IP_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
+ err = NF_HOOK(PF_INET, NF_INET_LOCAL_OUT, skb, NULL, rt->u.dst.dev,
dst_output);
if (err > 0)
-@@ -325,6 +352,7 @@ out:
+@@ -393,6 +399,7 @@ out:
error_fault:
err = -EFAULT;
@@ -27024,42 +27097,98 @@
kfree_skb(skb);
error:
IP_INC_STATS(IPSTATS_MIB_OUTDISCARDS);
-@@ -491,6 +519,12 @@ static int raw_sendmsg(struct kiocb *ioc
+@@ -560,6 +567,13 @@ static int raw_sendmsg(struct kiocb *ioc
}
security_sk_classify_flow(sk, &fl);
+ if (sk->sk_nx_info) {
-+ err = ip_find_src(sk->sk_nx_info, &rt, &fl);
++ err = ip_v4_find_src(sk->sk_net,
++ sk->sk_nx_info, &rt, &fl);
+
+ if (err)
+ goto done;
+ }
- err = ip_route_output_flow(&rt, &fl, sk, 1);
+ err = ip_route_output_flow(&init_net, &rt, &fl, sk, 1);
}
if (err)
-@@ -795,7 +829,8 @@ static struct sock *raw_get_first(struct
+@@ -622,17 +636,19 @@ static int raw_bind(struct sock *sk, str
+ {
+ struct inet_sock *inet = inet_sk(sk);
+ struct sockaddr_in *addr = (struct sockaddr_in *) uaddr;
++ struct nx_v4_sock_addr nsa = { 0 };
+ int ret = -EINVAL;
+ int chk_addr_ret;
+
+ if (sk->sk_state != TCP_CLOSE || addr_len < sizeof(struct sockaddr_in))
+ goto out;
+- chk_addr_ret = inet_addr_type(sk->sk_net, addr->sin_addr.s_addr);
++ v4_map_sock_addr(inet, addr, &nsa);
++ chk_addr_ret = inet_addr_type(sk->sk_net, nsa.saddr);
+ ret = -EADDRNOTAVAIL;
+- if (addr->sin_addr.s_addr && chk_addr_ret != RTN_LOCAL &&
++ if (nsa.saddr && chk_addr_ret != RTN_LOCAL &&
+ chk_addr_ret != RTN_MULTICAST && chk_addr_ret != RTN_BROADCAST)
+ goto out;
+- inet->rcv_saddr = inet->saddr = addr->sin_addr.s_addr;
++ v4_set_sock_addr(inet, &nsa);
+ if (chk_addr_ret == RTN_MULTICAST || chk_addr_ret == RTN_BROADCAST)
+ inet->saddr = 0; /* Use device */
+ sk_dst_reset(sk);
+@@ -684,7 +700,8 @@ static int raw_recvmsg(struct kiocb *ioc
+ /* Copy the address. */
+ if (sin) {
+ sin->sin_family = AF_INET;
+- sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
++ sin->sin_addr.s_addr =
++ nx_map_sock_lback(sk->sk_nx_info, ip_hdr(skb)->saddr);
+ sin->sin_port = 0;
+ memset(&sin->sin_zero, 0, sizeof(sin->sin_zero));
+ }
+@@ -862,7 +879,8 @@ static struct sock *raw_get_first(struct
struct hlist_node *node;
- sk_for_each(sk, node, &raw_v4_htable[state->bucket])
-- if (sk->sk_family == PF_INET)
-+ if (sk->sk_family == PF_INET &&
+ sk_for_each(sk, node, &state->h->ht[state->bucket])
+- if (sk->sk_net == state->p.net)
++ if ((sk->sk_net == state->p.net) &&
+ nx_check(sk->sk_nid, VS_WATCH_P | VS_IDENT))
goto found;
}
sk = NULL;
-@@ -811,7 +846,8 @@ static struct sock *raw_get_next(struct
+@@ -878,7 +896,8 @@ static struct sock *raw_get_next(struct
sk = sk_next(sk);
try_again:
;
-- } while (sk && sk->sk_family != PF_INET);
-+ } while (sk && (sk->sk_family != PF_INET ||
+- } while (sk && sk->sk_net != state->p.net);
++ } while (sk && ((sk->sk_net != state->p.net) ||
+ !nx_check(sk->sk_nid, VS_WATCH_P | VS_IDENT)));
- if (!sk && ++state->bucket < RAWV4_HTABLE_SIZE) {
- sk = sk_head(&raw_v4_htable[state->bucket]);
---- linux-2.6.22/net/ipv4/tcp.c 2007-07-09 13:20:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/tcp.c 2007-07-07 03:52:54 +0200
-@@ -258,6 +258,7 @@
+ if (!sk && ++state->bucket < RAW_HTABLE_SIZE) {
+ sk = sk_head(&state->h->ht[state->bucket]);
+@@ -937,7 +956,10 @@ static void raw_sock_seq_show(struct seq
+
+ seq_printf(seq, "%4d: %08X:%04X %08X:%04X"
+ " %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p %d",
+- i, src, srcp, dest, destp, sp->sk_state,
++ i,
++ nx_map_sock_lback(current_nx_info(), src), srcp,
++ nx_map_sock_lback(current_nx_info(), dest), destp,
++ sp->sk_state,
+ atomic_read(&sp->sk_wmem_alloc),
+ atomic_read(&sp->sk_rmem_alloc),
+ 0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp),
+--- linux-2.6.25.4/net/ipv4/syncookies.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/syncookies.c 2008-04-19 15:14:52.000000000 -0400
+@@ -20,6 +20,7 @@
+ #include <linux/cryptohash.h>
+ #include <linux/kernel.h>
+ #include <net/tcp.h>
++#include <net/route.h>
+
+ extern int sysctl_tcp_syncookies;
+
+--- linux-2.6.25.4/net/ipv4/tcp.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/tcp.c 2008-04-19 15:14:52.000000000 -0400
+@@ -263,6 +263,7 @@
#include <linux/cache.h>
#include <linux/err.h>
#include <linux/crypto.h>
@@ -27067,9 +27196,9 @@
#include <net/icmp.h>
#include <net/tcp.h>
---- linux-2.6.22/net/ipv4/tcp_ipv4.c 2007-07-09 13:20:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/tcp_ipv4.c 2007-06-17 05:54:20 +0200
-@@ -1975,6 +1975,12 @@ static void *listening_get_next(struct s
+--- linux-2.6.25.4/net/ipv4/tcp_ipv4.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/tcp_ipv4.c 2008-04-19 15:14:52.000000000 -0400
+@@ -1965,6 +1965,12 @@ static void *listening_get_next(struct s
req = req->dl_next;
while (1) {
while (req) {
@@ -27082,7 +27211,7 @@
if (req->rsk_ops->family == st->family) {
cur = req;
goto out;
-@@ -1999,6 +2005,10 @@ get_req:
+@@ -1989,6 +1995,10 @@ get_req:
}
get_sk:
sk_for_each_from(sk, node) {
@@ -27093,9 +27222,9 @@
if (sk->sk_family == st->family) {
cur = sk;
goto out;
-@@ -2050,18 +2060,26 @@ static void *established_get_first(struc
+@@ -2038,18 +2048,26 @@ static void *established_get_first(struc
- read_lock(&tcp_hashinfo.ehash[st->bucket].lock);
+ read_lock_bh(lock);
sk_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) {
- if (sk->sk_family != st->family) {
+ vxdprintk(VXD_CBIT(net, 6),
@@ -27124,7 +27253,7 @@
rc = tw;
goto out;
}
-@@ -2085,7 +2103,8 @@ static void *established_get_next(struct
+@@ -2073,7 +2091,8 @@ static void *established_get_next(struct
tw = cur;
tw = tw_next(tw);
get_tw:
@@ -27134,7 +27263,7 @@
tw = tw_next(tw);
}
if (tw) {
-@@ -2109,6 +2128,11 @@ get_tw:
+@@ -2094,6 +2113,11 @@ get_tw:
sk = sk_next(sk);
sk_for_each_from(sk, node) {
@@ -27146,8 +27275,44 @@
if (sk->sk_family == st->family)
goto found;
}
---- linux-2.6.22/net/ipv4/tcp_minisocks.c 2007-07-09 13:20:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/tcp_minisocks.c 2007-06-15 02:37:04 +0200
+@@ -2266,9 +2290,9 @@ static void get_openreq4(struct sock *sk
+ sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X"
+ " %02X %08X:%08X %02X:%08lX %08X %5d %8d %u %d %p",
+ i,
+- ireq->loc_addr,
++ nx_map_sock_lback(current_nx_info(), ireq->loc_addr),
+ ntohs(inet_sk(sk)->sport),
+- ireq->rmt_addr,
++ nx_map_sock_lback(current_nx_info(), ireq->rmt_addr),
+ ntohs(ireq->rmt_port),
+ TCP_SYN_RECV,
+ 0, 0, /* could print option size, but that is af dependent. */
+@@ -2310,7 +2334,10 @@ static void get_tcp4_sock(struct sock *s
+
+ sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X %02X %08X:%08X %02X:%08lX "
+ "%08X %5d %8d %lu %d %p %u %u %u %u %d",
+- i, src, srcp, dest, destp, sk->sk_state,
++ i,
++ nx_map_sock_lback(current_nx_info(), src), srcp,
++ nx_map_sock_lback(current_nx_info(), dest), destp,
++ sk->sk_state,
+ tp->write_seq - tp->snd_una,
+ sk->sk_state == TCP_LISTEN ? sk->sk_ack_backlog :
+ (tp->rcv_nxt - tp->copied_seq),
+@@ -2345,7 +2372,10 @@ static void get_timewait4_sock(struct in
+
+ sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X"
+ " %02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %p",
+- i, src, srcp, dest, destp, tw->tw_substate, 0, 0,
++ i,
++ nx_map_sock_lback(current_nx_info(), src), srcp,
++ nx_map_sock_lback(current_nx_info(), dest), destp,
++ tw->tw_substate, 0, 0,
+ 3, jiffies_to_clock_t(ttd), 0, 0, 0, 0,
+ atomic_read(&tw->tw_refcnt), tw);
+ }
+--- linux-2.6.25.4/net/ipv4/tcp_minisocks.c 2008-04-17 11:31:40.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/tcp_minisocks.c 2008-04-19 15:14:52.000000000 -0400
@@ -28,6 +28,10 @@
#include <net/inet_common.h>
#include <net/xfrm.h>
@@ -27171,22 +27336,25 @@
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
if (tw->tw_family == PF_INET6) {
struct ipv6_pinfo *np = inet6_sk(sk);
---- linux-2.6.22/net/ipv4/udp.c 2007-07-09 13:20:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv4/udp.c 2007-06-17 06:02:02 +0200
-@@ -221,11 +221,8 @@ int udp_get_port(struct sock *sk, unsign
+--- linux-2.6.25.4/net/ipv4/udp.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv4/udp.c 2008-04-21 12:41:01.000000000 -0400
+@@ -246,14 +246,7 @@ int udp_get_port(struct sock *sk, unsign
+ return __udp_lib_get_port(sk, snum, udp_hash, scmp);
+ }
- int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
- {
+-int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
+-{
- struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);
-
- return ( !ipv6_only_sock(sk2) &&
+- return ( !ipv6_only_sock(sk2) &&
- (!inet1->rcv_saddr || !inet2->rcv_saddr ||
- inet1->rcv_saddr == inet2->rcv_saddr ));
-+ nx_addr_conflict(sk1->sk_nx_info, inet_rcv_saddr(sk1), sk2));
- }
+-}
++extern int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2);
static inline int udp_v4_get_port(struct sock *sk, unsigned short snum)
-@@ -246,15 +243,22 @@ static struct sock *__udp4_lib_lookup(__
+ {
+@@ -273,16 +266,23 @@ static struct sock *__udp4_lib_lookup(st
int badness = -1;
read_lock(&udp_hash_lock);
@@ -27194,22 +27362,23 @@
sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) {
struct inet_sock *inet = inet_sk(sk);
- if (sk->sk_hash == hnum && !ipv6_only_sock(sk)) {
+ if (sk->sk_net == net && sk->sk_hash == hnum &&
+ !ipv6_only_sock(sk)) {
int score = (sk->sk_family == PF_INET ? 1 : 0);
+
if (inet->rcv_saddr) {
if (inet->rcv_saddr != daddr)
continue;
score+=2;
-+ } else if (sk->sk_nx_info) {
-+ if (addr_in_nx_info(sk->sk_nx_info, daddr))
-+ score+=2;
-+ else
++ } else {
++ /* block non nx_info ips */
++ if (!v4_addr_in_nx_info(sk->sk_nx_info,
++ daddr, NXA_MASK_BIND))
+ continue;
}
if (inet->daddr) {
if (inet->daddr != saddr)
-@@ -280,6 +284,7 @@ static struct sock *__udp4_lib_lookup(__
+@@ -308,6 +308,7 @@ static struct sock *__udp4_lib_lookup(st
}
}
}
@@ -27217,38 +27386,42 @@
if (result)
sock_hold(result);
read_unlock(&udp_hash_lock);
-@@ -301,7 +306,8 @@ static inline struct sock *udp_v4_mcast_
+@@ -329,7 +330,7 @@ static inline struct sock *udp_v4_mcast_
if (s->sk_hash != hnum ||
(inet->daddr && inet->daddr != rmt_addr) ||
(inet->dport != rmt_port && inet->dport) ||
- (inet->rcv_saddr && inet->rcv_saddr != loc_addr) ||
-+ (inet->rcv_saddr && inet->rcv_saddr != loc_addr &&
-+ inet->rcv_saddr2 && inet->rcv_saddr2 != loc_addr) ||
++ !v4_sock_addr_match(sk->sk_nx_info, inet, loc_addr) ||
ipv6_only_sock(s) ||
(s->sk_bound_dev_if && s->sk_bound_dev_if != dif))
continue;
-@@ -631,7 +637,20 @@ int udp_sendmsg(struct kiocb *iocb, stru
+@@ -662,7 +663,15 @@ int udp_sendmsg(struct kiocb *iocb, stru
.uli_u = { .ports =
{ .sport = inet->sport,
.dport = dport } } };
+ struct nx_info *nxi = sk->sk_nx_info;
++ struct net *net = sk->sk_net;
+
security_sk_classify_flow(sk, &fl);
-+ if (nxi) {
-+ err = ip_find_src(nxi, &rt, &fl);
-+ if (err)
-+ goto out;
-+ if (daddr == IPI_LOOPBACK && !nx_check(0, VS_ADMIN))
-+ daddr = fl.fl4_dst = nxi->ipv4[0];
-+#ifdef CONFIG_VSERVER_REMAP_SADDR
-+ if (saddr == IPI_LOOPBACK && !nx_check(0, VS_ADMIN))
-+ saddr = fl.fl4_src = nxi->ipv4[0];
-+#endif
-+ }
- err = ip_route_output_flow(&rt, &fl, sk, 1);
++
++ err = ip_v4_find_src(net, nxi, &rt, &fl);
++ if (err)
++ goto out;
++
+ err = ip_route_output_flow(&init_net, &rt, &fl, sk, 1);
if (err) {
if (err == -ENETUNREACH)
-@@ -1551,7 +1570,8 @@ static struct sock *udp_get_first(struct
+@@ -905,7 +914,8 @@ try_again:
+ {
+ sin->sin_family = AF_INET;
+ sin->sin_port = udp_hdr(skb)->source;
+- sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
++ sin->sin_addr.s_addr = nx_map_sock_lback(
++ skb->sk->sk_nx_info, ip_hdr(skb)->saddr);
+ memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
+ }
+ if (inet->cmsg_flags)
+@@ -1516,7 +1526,8 @@ static struct sock *udp_get_first(struct
for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) {
struct hlist_node *node;
sk_for_each(sk, node, state->hashtable + state->bucket) {
@@ -27258,7 +27431,7 @@
goto found;
}
}
-@@ -1568,7 +1588,8 @@ static struct sock *udp_get_next(struct
+@@ -1533,7 +1544,8 @@ static struct sock *udp_get_next(struct
sk = sk_next(sk);
try_again:
;
@@ -27268,44 +27441,423 @@
if (!sk && ++state->bucket < UDP_HTABLE_SIZE) {
sk = sk_head(state->hashtable + state->bucket);
---- linux-2.6.22/net/ipv6/addrconf.c 2007-07-09 13:20:06 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/ipv6/addrconf.c 2007-06-17 05:54:20 +0200
-@@ -2771,7 +2771,10 @@ static void if6_seq_stop(struct seq_file
+@@ -1648,7 +1660,10 @@ static void udp4_format_sock(struct sock
+
+ sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X"
+ " %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p",
+- bucket, src, srcp, dest, destp, sp->sk_state,
++ bucket,
++ nx_map_sock_lback(current_nx_info(), src), srcp,
++ nx_map_sock_lback(current_nx_info(), dest), destp,
++ sp->sk_state,
+ atomic_read(&sp->sk_wmem_alloc),
+ atomic_read(&sp->sk_rmem_alloc),
+ 0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp),
+--- linux-2.6.25.4/net/ipv6/addrconf.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/addrconf.c 2008-04-19 16:44:23.000000000 -0400
+@@ -87,6 +87,8 @@
+
+ #include <linux/proc_fs.h>
+ #include <linux/seq_file.h>
++#include <linux/vs_network.h>
++#include <linux/vs_inet6.h>
+
+ /* Set to 3 to get tracing... */
+ #define ACONF_DEBUG 2
+@@ -918,7 +920,8 @@ static inline int ipv6_saddr_preferred(i
+ }
+
+ int ipv6_dev_get_saddr(struct net_device *daddr_dev,
+- struct in6_addr *daddr, struct in6_addr *saddr)
++ struct in6_addr *daddr, struct in6_addr *saddr,
++ struct nx_info *nxi)
+ {
+ struct ipv6_saddr_score hiscore;
+ struct inet6_ifaddr *ifa_result = NULL;
+@@ -964,6 +967,10 @@ int ipv6_dev_get_saddr(struct net_device
+
+ score.addr_type = __ipv6_addr_type(&ifa->addr);
+
++ /* Use only addresses assigned to the context */
++ if (!v6_ifa_in_nx_info(ifa, nxi))
++ continue;
++
+ /* Rule 0:
+ * - Tentative Address (RFC2462 section 5.4)
+ * - A tentative address is not considered
+@@ -1181,9 +1188,10 @@ record_it:
+
+
+ int ipv6_get_saddr(struct dst_entry *dst,
+- struct in6_addr *daddr, struct in6_addr *saddr)
++ struct in6_addr *daddr, struct in6_addr *saddr,
++ struct nx_info *nxi)
+ {
+- return ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL, daddr, saddr);
++ return ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL, daddr, saddr, nxi);
+ }
+
+ EXPORT_SYMBOL(ipv6_get_saddr);
+@@ -1287,35 +1295,46 @@ struct inet6_ifaddr *ipv6_get_ifaddr(str
+ return ifp;
+ }
+
++extern int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2);
++
+ int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2)
+ {
+ const struct in6_addr *sk_rcv_saddr6 = &inet6_sk(sk)->rcv_saddr;
+ const struct in6_addr *sk2_rcv_saddr6 = inet6_rcv_saddr(sk2);
+- __be32 sk_rcv_saddr = inet_sk(sk)->rcv_saddr;
+ __be32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
+ int sk_ipv6only = ipv6_only_sock(sk);
+ int sk2_ipv6only = inet_v6_ipv6only(sk2);
+ int addr_type = ipv6_addr_type(sk_rcv_saddr6);
+ int addr_type2 = sk2_rcv_saddr6 ? ipv6_addr_type(sk2_rcv_saddr6) : IPV6_ADDR_MAPPED;
+
+- if (!sk2_rcv_saddr && !sk_ipv6only)
++ /* FIXME: needs handling for v4 ANY */
++ if (!sk2_rcv_saddr && !sk_ipv6only && !sk2->sk_nx_info)
+ return 1;
+
+ if (addr_type2 == IPV6_ADDR_ANY &&
+- !(sk2_ipv6only && addr_type == IPV6_ADDR_MAPPED))
++ !(sk2_ipv6only && addr_type == IPV6_ADDR_MAPPED) &&
++ v6_addr_in_nx_info(sk2->sk_nx_info, sk_rcv_saddr6, -1))
+ return 1;
+
+ if (addr_type == IPV6_ADDR_ANY &&
+- !(sk_ipv6only && addr_type2 == IPV6_ADDR_MAPPED))
++ !(sk_ipv6only && addr_type2 == IPV6_ADDR_MAPPED) &&
++ (sk2_rcv_saddr6 && v6_addr_in_nx_info(sk->sk_nx_info, sk2_rcv_saddr6, -1)))
++ return 1;
++
++ if (addr_type == IPV6_ADDR_ANY &&
++ addr_type2 == IPV6_ADDR_ANY &&
++ nx_v6_addr_conflict(sk->sk_nx_info, sk2->sk_nx_info))
+ return 1;
+
+ if (sk2_rcv_saddr6 &&
++ addr_type != IPV6_ADDR_ANY &&
++ addr_type != IPV6_ADDR_ANY &&
+ ipv6_addr_equal(sk_rcv_saddr6, sk2_rcv_saddr6))
+ return 1;
+
+ if (addr_type == IPV6_ADDR_MAPPED &&
+ !sk2_ipv6only &&
+- (!sk2_rcv_saddr || !sk_rcv_saddr || sk_rcv_saddr == sk2_rcv_saddr))
++ ipv4_rcv_saddr_equal(sk, sk2))
+ return 1;
+
+ return 0;
+@@ -2846,7 +2865,10 @@ static void if6_seq_stop(struct seq_file
static int if6_seq_show(struct seq_file *seq, void *v)
{
struct inet6_ifaddr *ifp = (struct inet6_ifaddr *)v;
- seq_printf(seq,
+
-+ /* no ipv6 inside a vserver for now */
-+ if (nx_check(0, VS_ADMIN|VS_WATCH))
++ if (nx_check(0, VS_ADMIN|VS_WATCH) ||
++ v6_addr_in_nx_info(current_nx_info(), &ifp->addr, -1))
+ seq_printf(seq,
NIP6_SEQFMT " %02x %02x %02x %02x %8s\n",
NIP6(ifp->addr),
ifp->idev->dev->ifindex,
-@@ -3254,6 +3257,10 @@ static int inet6_dump_addr(struct sk_buf
+@@ -3337,6 +3359,11 @@ static int inet6_dump_addr(struct sk_buf
+ struct inet6_ifaddr *ifa;
struct ifmcaddr6 *ifmca;
struct ifacaddr6 *ifaca;
-
-+ /* no ipv6 inside a vserver for now */
-+ if (skb->sk && skb->sk->sk_vx_info)
-+ return skb->len;
++ struct nx_info *nxi = skb->sk ? skb->sk->sk_nx_info : NULL;
+
++ /* disable ipv6 on non v6 guests */
++ if (nxi && !nx_info_has_v6(nxi))
++ return skb->len;
+
s_idx = cb->args[0];
s_ip_idx = ip_idx = cb->args[1];
-
-@@ -3577,6 +3584,10 @@ static int inet6_dump_ifinfo(struct sk_b
+@@ -3358,6 +3385,8 @@ static int inet6_dump_addr(struct sk_buf
+ ifa = ifa->if_next, ip_idx++) {
+ if (ip_idx < s_ip_idx)
+ continue;
++ if (!v6_addr_in_nx_info(nxi, &ifa->addr, -1))
++ continue;
+ err = inet6_fill_ifaddr(skb, ifa,
+ NETLINK_CB(cb->skb).pid,
+ cb->nlh->nlmsg_seq,
+@@ -3371,6 +3400,8 @@ static int inet6_dump_addr(struct sk_buf
+ ifmca = ifmca->next, ip_idx++) {
+ if (ip_idx < s_ip_idx)
+ continue;
++ if (!v6_addr_in_nx_info(nxi, &ifa->addr, -1))
++ continue;
+ err = inet6_fill_ifmcaddr(skb, ifmca,
+ NETLINK_CB(cb->skb).pid,
+ cb->nlh->nlmsg_seq,
+@@ -3384,6 +3415,8 @@ static int inet6_dump_addr(struct sk_buf
+ ifaca = ifaca->aca_next, ip_idx++) {
+ if (ip_idx < s_ip_idx)
+ continue;
++ if (!v6_addr_in_nx_info(nxi, &ifa->addr, -1))
++ continue;
+ err = inet6_fill_ifacaddr(skb, ifaca,
+ NETLINK_CB(cb->skb).pid,
+ cb->nlh->nlmsg_seq,
+@@ -3678,6 +3711,11 @@ static int inet6_dump_ifinfo(struct sk_b
+ int s_idx = cb->args[0];
struct net_device *dev;
struct inet6_dev *idev;
-
-+ /* no ipv6 inside a vserver for now */
++ struct nx_info *nxi = skb->sk ? skb->sk->sk_nx_info : NULL;
++
++ /* FIXME: maybe disable ipv6 on non v6 guests?
+ if (skb->sk && skb->sk->sk_vx_info)
-+ return skb->len;
++ return skb->len; */
+
+ if (net != &init_net)
+ return 0;
+@@ -3687,6 +3725,8 @@ static int inet6_dump_ifinfo(struct sk_b
+ for_each_netdev(&init_net, dev) {
+ if (idx < s_idx)
+ goto cont;
++ if (!v6_dev_in_nx_info(dev, nxi))
++ goto cont;
+ if ((idev = in6_dev_get(dev)) == NULL)
+ goto cont;
+ err = inet6_fill_ifinfo(skb, idev, NETLINK_CB(cb->skb).pid,
+--- linux-2.6.25.4/net/ipv6/af_inet6.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/af_inet6.c 2008-04-19 15:14:52.000000000 -0400
+@@ -43,6 +43,8 @@
+ #include <linux/netdevice.h>
+ #include <linux/icmpv6.h>
+ #include <linux/netfilter_ipv6.h>
++#include <linux/vs_inet.h>
++#include <linux/vs_inet6.h>
+
+ #include <net/ip.h>
+ #include <net/ipv6.h>
+@@ -51,6 +53,7 @@
+ #include <net/tcp.h>
+ #include <net/ipip.h>
+ #include <net/protocol.h>
++#include <net/route.h>
+ #include <net/inet_common.h>
+ #include <net/transp_v6.h>
+ #include <net/ip6_route.h>
+@@ -148,9 +151,12 @@ lookup_protocol:
+ }
+
+ err = -EPERM;
++ if ((protocol == IPPROTO_ICMPV6) &&
++ nx_capable(answer->capability, NXC_RAW_ICMP))
++ goto override;
+ if (answer->capability > 0 && !capable(answer->capability))
+ goto out_rcu_unlock;
+-
++override:
+ sock->ops = answer->ops;
+ answer_prot = answer->prot;
+ answer_no_check = answer->no_check;
+@@ -248,6 +254,7 @@ int inet6_bind(struct socket *sock, stru
+ struct sock *sk = sock->sk;
+ struct inet_sock *inet = inet_sk(sk);
+ struct ipv6_pinfo *np = inet6_sk(sk);
++ struct nx_v6_sock_addr nsa;
+ __be32 v4addr = 0;
+ unsigned short snum;
+ int addr_type = 0;
+@@ -259,6 +266,11 @@ int inet6_bind(struct socket *sock, stru
+
+ if (addr_len < SIN6_LEN_RFC2133)
+ return -EINVAL;
+
- read_lock(&dev_base_lock);
- idx = 0;
- for_each_netdev(dev) {
---- linux-2.6.22/net/netlink/af_netlink.c 2007-07-09 13:20:07 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/netlink/af_netlink.c 2007-06-15 05:43:32 +0200
++ err = v6_map_sock_addr(inet, addr, &nsa);
++ if (err)
++ return err;
++
+ addr_type = ipv6_addr_type(&addr->sin6_addr);
+ if ((addr_type & IPV6_ADDR_MULTICAST) && sock->type == SOCK_STREAM)
+ return -EINVAL;
+@@ -282,6 +294,10 @@ int inet6_bind(struct socket *sock, stru
+ err = -EADDRNOTAVAIL;
+ goto out;
+ }
++ if (!v4_addr_in_nx_info(sk->sk_nx_info, v4addr, NXA_MASK_BIND)) {
++ err = -EADDRNOTAVAIL;
++ goto out;
++ }
+ } else {
+ if (addr_type != IPV6_ADDR_ANY) {
+ struct net_device *dev = NULL;
+@@ -307,6 +323,11 @@ int inet6_bind(struct socket *sock, stru
+ }
+ }
+
++ if (!v6_addr_in_nx_info(sk->sk_nx_info, &addr->sin6_addr, -1)) {
++ err = -EADDRNOTAVAIL;
++ goto out;
++ }
++
+ /* ipv4 addr of the socket is invalid. Only the
+ * unspecified and mapped address have a v4 equivalent.
+ */
+@@ -325,6 +346,8 @@ int inet6_bind(struct socket *sock, stru
+ }
+ }
+
++ v6_set_sock_addr(inet, &nsa);
++
+ inet->rcv_saddr = v4addr;
+ inet->saddr = v4addr;
+
+@@ -419,9 +442,11 @@ int inet6_getname(struct socket *sock, s
+ return -ENOTCONN;
+ sin->sin6_port = inet->dport;
+ ipv6_addr_copy(&sin->sin6_addr, &np->daddr);
++ /* FIXME: remap lback? */
+ if (np->sndflow)
+ sin->sin6_flowinfo = np->flow_label;
+ } else {
++ /* FIXME: remap lback? */
+ if (ipv6_addr_any(&np->rcv_saddr))
+ ipv6_addr_copy(&sin->sin6_addr, &np->saddr);
+ else
+--- linux-2.6.25.4/net/ipv6/fib6_rules.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/fib6_rules.c 2008-04-19 15:14:52.000000000 -0400
+@@ -86,7 +86,7 @@ static int fib6_rule_action(struct fib_r
+ r->src.plen && !(flags & RT6_LOOKUP_F_HAS_SADDR)) {
+ struct in6_addr saddr;
+ if (ipv6_get_saddr(&rt->u.dst, &flp->fl6_dst,
+- &saddr))
++ &saddr, NULL))
+ goto again;
+ if (!ipv6_prefix_equal(&saddr, &r->src.addr,
+ r->src.plen))
+--- linux-2.6.25.4/net/ipv6/inet6_hashtables.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/inet6_hashtables.c 2008-04-19 15:14:52.000000000 -0400
+@@ -16,6 +16,7 @@
+
+ #include <linux/module.h>
+ #include <linux/random.h>
++#include <linux/vs_inet6.h>
+
+ #include <net/inet_connection_sock.h>
+ #include <net/inet_hashtables.h>
+@@ -114,6 +115,9 @@ struct sock *inet6_lookup_listener(struc
+ if (!ipv6_addr_equal(&np->rcv_saddr, daddr))
+ continue;
+ score++;
++ } else {
++ if (!v6_addr_in_nx_info(sk->sk_nx_info, daddr, -1))
++ continue;
+ }
+ if (sk->sk_bound_dev_if) {
+ if (sk->sk_bound_dev_if != dif)
+--- linux-2.6.25.4/net/ipv6/ip6_output.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/ip6_output.c 2008-04-19 15:14:52.000000000 -0400
+@@ -920,7 +920,7 @@ static int ip6_dst_lookup_tail(struct so
+ goto out_err_release;
+
+ if (ipv6_addr_any(&fl->fl6_src)) {
+- err = ipv6_get_saddr(*dst, &fl->fl6_dst, &fl->fl6_src);
++ err = ipv6_get_saddr(*dst, &fl->fl6_dst, &fl->fl6_src, sk->sk_nx_info);
+ if (err)
+ goto out_err_release;
+ }
+--- linux-2.6.25.4/net/ipv6/Kconfig 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/Kconfig 2008-04-19 15:14:52.000000000 -0400
+@@ -4,8 +4,8 @@
+
+ # IPv6 as module will cause a CRASH if you try to unload it
+ config IPV6
+- tristate "The IPv6 protocol"
+- default m
++ bool "The IPv6 protocol"
++ default n
+ ---help---
+ This is complemental support for the IP version 6.
+ You will still be able to do traditional IPv4 networking as well.
+--- linux-2.6.25.4/net/ipv6/ndisc.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/ndisc.c 2008-04-19 15:14:52.000000000 -0400
+@@ -563,7 +563,7 @@ static void ndisc_send_na(struct net_dev
+ override = 0;
+ in6_ifa_put(ifp);
+ } else {
+- if (ipv6_dev_get_saddr(dev, daddr, &tmpaddr))
++ if (ipv6_dev_get_saddr(dev, daddr, &tmpaddr, NULL))
+ return;
+ src_addr = &tmpaddr;
+ }
+--- linux-2.6.25.4/net/ipv6/route.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/route.c 2008-04-19 15:14:52.000000000 -0400
+@@ -2122,7 +2122,7 @@ static int rt6_fill_node(struct sk_buff
+ NLA_PUT_U32(skb, RTA_IIF, iif);
+ else if (dst) {
+ struct in6_addr saddr_buf;
+- if (ipv6_get_saddr(&rt->u.dst, dst, &saddr_buf) == 0)
++ if (ipv6_get_saddr(&rt->u.dst, dst, &saddr_buf, (skb->sk ? skb->sk->sk_nx_info : NULL)) == 0)
+ NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf);
+ }
+
+--- linux-2.6.25.4/net/ipv6/tcp_ipv6.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/tcp_ipv6.c 2008-04-19 15:14:52.000000000 -0400
+@@ -68,6 +68,7 @@
+
+ #include <linux/crypto.h>
+ #include <linux/scatterlist.h>
++#include <linux/vs_inet6.h>
+
+ /* Socket used for sending RSTs and ACKs */
+ static struct socket *tcp6_socket;
+@@ -154,8 +155,15 @@ static int tcp_v6_connect(struct sock *s
+ * connect() to INADDR_ANY means loopback (BSD'ism).
+ */
+
+- if(ipv6_addr_any(&usin->sin6_addr))
+- usin->sin6_addr.s6_addr[15] = 0x1;
++ if(ipv6_addr_any(&usin->sin6_addr)) {
++ struct nx_info *nxi = sk->sk_nx_info;
++
++ if (nxi && nx_info_has_v6(nxi))
++ /* FIXME: remap lback? */
++ usin->sin6_addr = nxi->v6.ip;
++ else
++ usin->sin6_addr.s6_addr[15] = 0x1;
++ }
+
+ addr_type = ipv6_addr_type(&usin->sin6_addr);
+
+--- linux-2.6.25.4/net/ipv6/udp.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/udp.c 2008-04-19 15:14:52.000000000 -0400
+@@ -49,6 +49,7 @@
+
+ #include <linux/proc_fs.h>
+ #include <linux/seq_file.h>
++#include <linux/vs_inet6.h>
+ #include "udp_impl.h"
+
+ static inline int udp_v6_get_port(struct sock *sk, unsigned short snum)
+@@ -83,6 +84,10 @@ static struct sock *__udp6_lib_lookup(st
+ if (!ipv6_addr_equal(&np->rcv_saddr, daddr))
+ continue;
+ score++;
++ } else {
++ /* block non nx_info ips */
++ if (!v6_addr_in_nx_info(sk->sk_nx_info, daddr, -1))
++ continue;
+ }
+ if (!ipv6_addr_any(&np->daddr)) {
+ if (!ipv6_addr_equal(&np->daddr, saddr))
+--- linux-2.6.25.4/net/ipv6/xfrm6_policy.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/ipv6/xfrm6_policy.c 2008-04-19 16:13:21.000000000 -0400
+@@ -58,7 +58,7 @@ static int xfrm6_get_saddr(xfrm_address_
+ return -EHOSTUNREACH;
+
+ ipv6_get_saddr(dst, (struct in6_addr *)&daddr->a6,
+- (struct in6_addr *)&saddr->a6);
++ (struct in6_addr *)&saddr->a6, NULL);
+ dst_release(dst);
+ return 0;
+ }
+--- linux-2.6.25.4/net/netlink/af_netlink.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/netlink/af_netlink.c 2008-04-19 15:14:52.000000000 -0400
@@ -56,6 +56,9 @@
#include <linux/audit.h>
#include <linux/selinux.h>
@@ -27314,20 +27866,33 @@
+#include <linux/vs_network.h>
+#include <linux/vs_limit.h>
+ #include <net/net_namespace.h>
#include <net/sock.h>
- #include <net/scm.h>
---- linux-2.6.22/net/socket.c 2007-07-09 13:20:13 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/socket.c 2007-06-15 02:37:04 +0200
-@@ -92,6 +92,8 @@
+--- linux-2.6.25.4/net/sctp/ipv6.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/sctp/ipv6.c 2008-04-19 15:14:52.000000000 -0400
+@@ -316,7 +316,7 @@ static void sctp_v6_get_saddr(struct sct
+ __FUNCTION__, asoc, dst, NIP6(daddr->v6.sin6_addr));
+
+ if (!asoc) {
+- ipv6_get_saddr(dst, &daddr->v6.sin6_addr,&saddr->v6.sin6_addr);
++ ipv6_get_saddr(dst, &daddr->v6.sin6_addr,&saddr->v6.sin6_addr, asoc->base.sk->sk_nx_info);
+ SCTP_DEBUG_PRINTK("saddr from ipv6_get_saddr: " NIP6_FMT "\n",
+ NIP6(saddr->v6.sin6_addr));
+ return;
+--- linux-2.6.25.4/net/socket.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/socket.c 2008-04-19 15:14:52.000000000 -0400
+@@ -93,6 +93,10 @@
#include <net/sock.h>
#include <linux/netfilter.h>
+#include <linux/vs_base.h>
+#include <linux/vs_socket.h>
++#include <linux/vs_inet.h>
++#include <linux/vs_inet6.h>
static int sock_no_open(struct inode *irrelevant, struct file *dontcare);
static ssize_t sock_aio_read(struct kiocb *iocb, const struct iovec *iov,
-@@ -543,7 +545,7 @@ static inline int __sock_sendmsg(struct
+@@ -547,7 +551,7 @@ static inline int __sock_sendmsg(struct
struct msghdr *msg, size_t size)
{
struct sock_iocb *si = kiocb_to_siocb(iocb);
@@ -27336,7 +27901,7 @@
si->sock = sock;
si->scm = NULL;
-@@ -554,7 +556,22 @@ static inline int __sock_sendmsg(struct
+@@ -558,7 +562,22 @@ static inline int __sock_sendmsg(struct
if (err)
return err;
@@ -27360,7 +27925,7 @@
}
int sock_sendmsg(struct socket *sock, struct msghdr *msg, size_t size)
-@@ -623,7 +640,7 @@ EXPORT_SYMBOL_GPL(__sock_recv_timestamp)
+@@ -627,7 +646,7 @@ EXPORT_SYMBOL_GPL(__sock_recv_timestamp)
static inline int __sock_recvmsg(struct kiocb *iocb, struct socket *sock,
struct msghdr *msg, size_t size, int flags)
{
@@ -27369,7 +27934,7 @@
struct sock_iocb *si = kiocb_to_siocb(iocb);
si->sock = sock;
-@@ -636,7 +653,18 @@ static inline int __sock_recvmsg(struct
+@@ -640,7 +659,18 @@ static inline int __sock_recvmsg(struct
if (err)
return err;
@@ -27389,18 +27954,21 @@
}
int sock_recvmsg(struct socket *sock, struct msghdr *msg,
-@@ -1090,6 +1118,10 @@ static int __sock_create(int family, int
+@@ -1105,6 +1135,13 @@ static int __sock_create(struct net *net
if (type < 0 || type >= SOCK_MAX)
return -EINVAL;
-+ /* disable IPv6 inside vservers for now */
-+ if (family == PF_INET6 && !nx_check(0, VS_ADMIN))
-+ return -EAFNOSUPPORT;
++ if (!nx_check(0, VS_ADMIN)) {
++ if (family == PF_INET && !current_nx_info_has_v4())
++ return -EAFNOSUPPORT;
++ if (family == PF_INET6 && !current_nx_info_has_v6())
++ return -EAFNOSUPPORT;
++ }
+
/* Compatibility.
This uglymoron is moved from INET layer to here to avoid
-@@ -1207,6 +1239,7 @@ asmlinkage long sys_socket(int family, i
+@@ -1222,6 +1259,7 @@ asmlinkage long sys_socket(int family, i
if (retval < 0)
goto out;
@@ -27408,7 +27976,7 @@
retval = sock_map_fd(sock);
if (retval < 0)
goto out_release;
-@@ -1239,10 +1272,12 @@ asmlinkage long sys_socketpair(int famil
+@@ -1254,10 +1292,12 @@ asmlinkage long sys_socketpair(int famil
err = sock_create(family, type, protocol, &sock1);
if (err < 0)
goto out;
@@ -27421,8 +27989,8 @@
err = sock1->ops->socketpair(sock1, sock2);
if (err < 0)
---- linux-2.6.22/net/sunrpc/auth.c 2007-05-02 19:25:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/sunrpc/auth.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/net/sunrpc/auth.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/sunrpc/auth.c 2008-04-19 15:14:52.000000000 -0400
@@ -13,6 +13,7 @@
#include <linux/errno.h>
#include <linux/sunrpc/clnt.h>
@@ -27431,7 +27999,7 @@
#ifdef RPC_DEBUG
# define RPCDBG_FACILITY RPCDBG_AUTH
-@@ -263,6 +264,7 @@ rpcauth_lookupcred(struct rpc_auth *auth
+@@ -345,6 +346,7 @@ rpcauth_lookupcred(struct rpc_auth *auth
struct auth_cred acred = {
.uid = current->fsuid,
.gid = current->fsgid,
@@ -27439,7 +28007,7 @@
.group_info = current->group_info,
};
struct rpc_cred *ret;
-@@ -282,6 +284,7 @@ rpcauth_bindcred(struct rpc_task *task)
+@@ -382,6 +384,7 @@ rpcauth_bindcred(struct rpc_task *task)
struct auth_cred acred = {
.uid = current->fsuid,
.gid = current->fsgid,
@@ -27447,8 +28015,8 @@
.group_info = current->group_info,
};
struct rpc_cred *ret;
---- linux-2.6.22/net/sunrpc/auth_unix.c 2007-05-02 19:25:48 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/sunrpc/auth_unix.c 2007-06-15 02:37:04 +0200
+--- linux-2.6.25.4/net/sunrpc/auth_unix.c 2008-04-17 10:37:27.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/sunrpc/auth_unix.c 2008-04-19 15:14:52.000000000 -0400
@@ -11,12 +11,14 @@
#include <linux/module.h>
#include <linux/sunrpc/clnt.h>
@@ -27464,7 +28032,7 @@
gid_t uc_gids[NFS_NGROUPS];
};
#define uc_uid uc_base.cr_uid
-@@ -79,6 +81,7 @@ unx_create_cred(struct rpc_auth *auth, s
+@@ -73,6 +75,7 @@ unx_create_cred(struct rpc_auth *auth, s
if (flags & RPCAUTH_LOOKUP_ROOTCREDS) {
cred->uc_uid = 0;
cred->uc_gid = 0;
@@ -27472,15 +28040,15 @@
cred->uc_gids[0] = NOGROUP;
} else {
int groups = acred->group_info->ngroups;
-@@ -87,6 +90,7 @@ unx_create_cred(struct rpc_auth *auth, s
+@@ -80,6 +83,7 @@ unx_create_cred(struct rpc_auth *auth, s
+ groups = NFS_NGROUPS;
- cred->uc_uid = acred->uid;
cred->uc_gid = acred->gid;
+ cred->uc_tag = acred->tag;
for (i = 0; i < groups; i++)
cred->uc_gids[i] = GROUP_AT(acred->group_info, i);
if (i < NFS_NGROUPS)
-@@ -118,7 +122,8 @@ unx_match(struct auth_cred *acred, struc
+@@ -124,7 +128,8 @@ unx_match(struct auth_cred *acred, struc
int groups;
if (cred->uc_uid != acred->uid
@@ -27490,16 +28058,16 @@
return 0;
groups = acred->group_info->ngroups;
-@@ -144,7 +149,7 @@ unx_marshal(struct rpc_task *task, __be3
+@@ -150,7 +155,7 @@ unx_marshal(struct rpc_task *task, __be3
struct rpc_clnt *clnt = task->tk_client;
- struct unx_cred *cred = (struct unx_cred *) task->tk_msg.rpc_cred;
+ struct unx_cred *cred = container_of(task->tk_msg.rpc_cred, struct unx_cred, uc_base);
__be32 *base, *hold;
- int i;
+ int i, tag;
*p++ = htonl(RPC_AUTH_UNIX);
base = p++;
-@@ -154,9 +159,12 @@ unx_marshal(struct rpc_task *task, __be3
+@@ -160,9 +165,12 @@ unx_marshal(struct rpc_task *task, __be3
* Copy the UTS nodename captured when the client was created.
*/
p = xdr_encode_array(p, clnt->cl_nodename, clnt->cl_nodelen);
@@ -27514,18 +28082,18 @@
hold = p++;
for (i = 0; i < 16 && cred->uc_gids[i] != (gid_t) NOGROUP; i++)
*p++ = htonl((u32) cred->uc_gids[i]);
---- linux-2.6.22/net/sunrpc/clnt.c 2007-07-09 13:20:13 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/sunrpc/clnt.c 2007-06-15 02:37:04 +0200
-@@ -30,6 +30,7 @@
- #include <linux/smp_lock.h>
+--- linux-2.6.25.4/net/sunrpc/clnt.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/sunrpc/clnt.c 2008-04-19 16:09:36.000000000 -0400
+@@ -31,6 +31,7 @@
#include <linux/utsname.h>
#include <linux/workqueue.h>
+ #include <linux/in6.h>
+#include <linux/vs_cvirt.h>
#include <linux/sunrpc/clnt.h>
#include <linux/sunrpc/rpc_pipe_fs.h>
-@@ -249,7 +250,9 @@ struct rpc_clnt *rpc_create(struct rpc_c
- clnt->cl_oneshot = 1;
+@@ -324,7 +325,9 @@ struct rpc_clnt *rpc_create(struct rpc_c
+ clnt->cl_autobind = 1;
if (args->flags & RPC_CLNT_CREATE_DISCRTRY)
clnt->cl_discrtry = 1;
-
@@ -27535,38 +28103,38 @@
return clnt;
}
EXPORT_SYMBOL_GPL(rpc_create);
---- linux-2.6.22/net/unix/af_unix.c 2007-07-09 13:20:13 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/unix/af_unix.c 2007-06-17 05:54:20 +0200
-@@ -115,6 +115,8 @@
+--- linux-2.6.25.4/net/unix/af_unix.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/unix/af_unix.c 2008-04-19 15:55:39.000000000 -0400
+@@ -116,6 +116,8 @@
#include <linux/mount.h>
#include <net/checksum.h>
#include <linux/security.h>
+#include <linux/vs_context.h>
+#include <linux/vs_limit.h>
- int sysctl_unix_max_dgram_qlen __read_mostly = 10;
-
-@@ -252,6 +254,8 @@ static struct sock *__unix_find_socket_b
- sk_for_each(s, node, &unix_socket_table[hash ^ type]) {
- struct unix_sock *u = unix_sk(s);
+ static struct hlist_head unix_socket_table[UNIX_HASH_SIZE + 1];
+ static DEFINE_SPINLOCK(unix_table_lock);
+@@ -255,6 +257,8 @@ static struct sock *__unix_find_socket_b
+ if (s->sk_net != net)
+ continue;
+ if (!nx_check(s->sk_nid, VS_WATCH_P | VS_IDENT))
+ continue;
if (u->addr->len == len &&
!memcmp(u->addr->name, sunname, len))
goto found;
-@@ -807,7 +811,7 @@ static int unix_bind(struct socket *sock
+@@ -819,7 +823,7 @@ static int unix_bind(struct socket *sock
*/
mode = S_IFSOCK |
(SOCK_INODE(sock)->i_mode & ~current->fs->umask);
-- err = vfs_mknod(nd.dentry->d_inode, dentry, mode, 0);
-+ err = vfs_mknod(nd.dentry->d_inode, dentry, mode, 0, NULL);
+- err = vfs_mknod(nd.path.dentry->d_inode, dentry, mode, 0);
++ err = vfs_mknod(nd.path.dentry->d_inode, dentry, mode, 0, NULL);
if (err)
goto out_mknod_dput;
- mutex_unlock(&nd.dentry->d_inode->i_mutex);
---- linux-2.6.22/net/x25/af_x25.c 2007-07-09 13:20:13 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/net/x25/af_x25.c 2007-06-15 02:37:04 +0200
-@@ -500,7 +500,10 @@ static int x25_create(struct socket *soc
+ mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+--- linux-2.6.25.4/net/x25/af_x25.c 2008-04-17 12:05:45.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/net/x25/af_x25.c 2008-04-19 15:14:52.000000000 -0400
+@@ -506,7 +506,10 @@ static int x25_create(struct net *net, s
x25 = x25_sk(sk);
@@ -27578,51 +28146,62 @@
x25_init_timers(sk);
---- linux-2.6.22/security/Kconfig 2006-11-30 21:19:47 +0100
-+++ linux-2.6.22-vs2.2.0-rc5/security/Kconfig 2007-06-15 02:37:04 +0200
-@@ -6,6 +6,7 @@ menu "Security options"
-
- config KEYS
- bool "Enable access key retention support"
-+ depends on !VSERVER_SECURITY
- help
- This option provides support for retaining authentication tokens and
- access keys in the kernel.
---- linux-2.6.22/security/commoncap.c 2007-07-09 13:20:14 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/security/commoncap.c 2007-06-15 02:37:04 +0200
-@@ -22,10 +22,11 @@
- #include <linux/ptrace.h>
- #include <linux/xattr.h>
+--- linux-2.6.25.4/scripts/checksyscalls.sh 2008-04-17 11:31:42.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/scripts/checksyscalls.sh 2008-04-19 15:14:52.000000000 -0400
+@@ -108,7 +108,6 @@ cat << EOF
+ #define __IGNORE_afs_syscall
+ #define __IGNORE_getpmsg
+ #define __IGNORE_putpmsg
+-#define __IGNORE_vserver
+ EOF
+ }
+
+--- linux-2.6.25.4/security/commoncap.c 2008-04-17 12:05:46.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/security/commoncap.c 2008-04-23 22:22:54.000000000 -0400
+@@ -24,6 +24,7 @@
#include <linux/hugetlb.h>
+ #include <linux/mount.h>
+ #include <linux/sched.h>
+#include <linux/vs_context.h>
+ /* Global security state */
+
+@@ -32,7 +33,7 @@ EXPORT_SYMBOL(securebits);
+
int cap_netlink_send(struct sock *sk, struct sk_buff *skb)
{
- NETLINK_CB(skb).eff_cap = current->cap_effective;
-+ cap_t(NETLINK_CB(skb).eff_cap) = vx_mbcap(cap_effective);
++ NETLINK_CB(skb).eff_cap = vx_mbcaps(current->cap_effective);
return 0;
}
-@@ -43,7 +44,7 @@ EXPORT_SYMBOL(cap_netlink_recv);
+@@ -53,9 +54,24 @@ EXPORT_SYMBOL(cap_netlink_recv);
+ */
int cap_capable (struct task_struct *tsk, int cap)
{
++ struct vx_info *vxi = tsk->vx_info;
++
++#if 0
++ printk("cap_capable() VXF_STATE_SETUP = %llx, raised = %x, eff = %08x:%08x\n",
++ vx_info_flags(vxi, VXF_STATE_SETUP, 0),
++ cap_raised(tsk->cap_effective, cap),
++ tsk->cap_effective.cap[1], tsk->cap_effective.cap[0]);
++#endif
++
++ /* special case SETUP */
++ if (vx_info_flags(vxi, VXF_STATE_SETUP, 0) &&
++ cap_raised(tsk->cap_effective, cap))
++ return 0;
++
/* Derived from include/linux/sched.h:capable. */
- if (cap_raised(tsk->cap_effective, cap))
-+ if (vx_cap_raised(tsk->vx_info, tsk->cap_effective, cap))
++ if (vx_cap_raised(vxi, tsk->cap_effective, cap))
return 0;
++
return -EPERM;
}
-@@ -141,7 +142,8 @@ void cap_bprm_apply_creds (struct linux_
- /* Derived from fs/exec.c:compute_creds. */
- kernel_cap_t new_permitted, working;
-
-- new_permitted = cap_intersect (bprm->cap_permitted, cap_bset);
-+ new_permitted = cap_intersect (bprm->cap_permitted,
-+ vx_current_cap_bset());
- working = cap_intersect (bprm->cap_inheritable,
- current->cap_inheritable);
- new_permitted = cap_combine (new_permitted, working);
-@@ -310,7 +312,8 @@ void cap_task_reparent_to_init (struct t
+
+@@ -583,7 +599,8 @@ void cap_task_reparent_to_init (struct t
int cap_syslog (int type)
{
@@ -27632,9 +28211,9 @@
return -EPERM;
return 0;
}
---- linux-2.6.22/security/dummy.c 2007-05-02 19:25:51 +0200
-+++ linux-2.6.22-vs2.2.0-rc5/security/dummy.c 2007-06-15 02:37:04 +0200
-@@ -28,6 +28,7 @@
+--- linux-2.6.25.4/security/dummy.c 2008-04-17 12:05:46.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/security/dummy.c 2008-04-23 20:28:54.000000000 -0400
+@@ -27,6 +27,7 @@
#include <linux/hugetlb.h>
#include <linux/ptrace.h>
#include <linux/file.h>
@@ -27642,12 +28221,22 @@
static int dummy_ptrace (struct task_struct *parent, struct task_struct *child)
{
-@@ -678,7 +679,7 @@ static int dummy_sem_semop (struct sem_a
+@@ -714,7 +715,7 @@ static int dummy_sem_semop (struct sem_a
static int dummy_netlink_send (struct sock *sk, struct sk_buff *skb)
{
- NETLINK_CB(skb).eff_cap = current->cap_effective;
-+ cap_t(NETLINK_CB(skb).eff_cap) = vx_mbcap(cap_effective);
++ NETLINK_CB(skb).eff_cap = vx_mbcaps(current->cap_effective);
return 0;
}
+--- linux-2.6.25.4/security/selinux/hooks.c 2008-05-21 14:30:05.000000000 -0400
++++ linux-2.6.25.4-vs2.3.0.34.10/security/selinux/hooks.c 2008-05-21 14:30:41.000000000 -0400
+@@ -64,7 +64,6 @@
+ #include <linux/dccp.h>
+ #include <linux/quota.h>
+ #include <linux/un.h> /* for Unix socket types */
+-#include <net/af_unix.h> /* for Unix socket types */
+ #include <linux/parser.h>
+ #include <linux/nfs_mount.h>
+ #include <net/ipv6.h>
Modified: dists/sid/linux-2.6/debian/patches/series/1-extra
==============================================================================
--- dists/sid/linux-2.6/debian/patches/series/1-extra (original)
+++ dists/sid/linux-2.6/debian/patches/series/1-extra Sun May 25 09:16:17 2008
@@ -1,5 +1,3 @@
-+ features/all/vserver/vs2.2.0-rc5.patch featureset=vserver featureset=xen-vserver
-+ features/all/vserver/bindmount-dev.patch featureset=vserver featureset=xen-vserver
+ features/all/xen/vmlinuz-target.patch featureset=xen
+ features/all/xen/blkfront-protocol.patch featureset=xen
+ features/all/xen/xenctrl.patch featureset=xen
Added: dists/sid/linux-2.6/debian/patches/series/4-extra
==============================================================================
--- (empty file)
+++ dists/sid/linux-2.6/debian/patches/series/4-extra Sun May 25 09:16:17 2008
@@ -0,0 +1,2 @@
++ features/all/vserver/vs2.3.0.34.10.patch featureset=vserver featureset=xen-vserver
++ features/all/vserver/bindmount-dev.patch featureset=vserver featureset=xen-vserver
More information about the Kernel-svn-changes
mailing list