[linux] 01/01: Update fix for CVE-2017-1000364 (Closes: #865303)
debian-kernel at lists.debian.org
debian-kernel at lists.debian.org
Mon Jun 26 16:48:55 UTC 2017
This is an automated email from the git hooks/post-receive script.
benh pushed a commit to branch jessie-security
in repository linux.
commit d6dd962b76a4a2a95d7bd09a003988b389539713
Author: Ben Hutchings <ben at decadent.org.uk>
Date: Mon Jun 26 16:09:55 2017 +0100
Update fix for CVE-2017-1000364 (Closes: #865303)
---
debian/changelog | 8 +
...d-me-mm-allow-to-configure-stack-gap-size.patch | 79 --
.../all/mm-allow-to-configure-stack-gap-size.patch | 71 --
.../mm-do-not-collapse-stack-gap-into-THP.patch | 52 --
.../bugfix/all/mm-enlarge-stack-guard-gap.patch | 473 ------------
...mm-fix-new-crash-in-unmapped_area_topdown.patch | 46 ++
.../mm-larger-stack-guard-gap-between-vmas.patch | 807 +++++++++++++++++++++
...-the-stack-gap-for-unpopulated-growing-vm.patch | 45 --
debian/patches/series | 7 +-
9 files changed, 863 insertions(+), 725 deletions(-)
diff --git a/debian/changelog b/debian/changelog
index 3cca09b..02519d7 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,11 @@
+linux (3.16.43-2+deb8u2) jessie-security; urgency=high
+
+ * Revert previous fixes for CVE-2017-1000364 (Closes: #865303)
+ * mm: larger stack guard gap, between vmas (CVE-2017-1000364)
+ * mm: fix new crash in unmapped_area_topdown()
+
+ -- Ben Hutchings <ben at decadent.org.uk> Mon, 26 Jun 2017 16:09:55 +0100
+
linux (3.16.43-2+deb8u1) jessie-security; urgency=high
[ Ben Hutchings ]
diff --git a/debian/patches/bugfix/all/fold-me-mm-allow-to-configure-stack-gap-size.patch b/debian/patches/bugfix/all/fold-me-mm-allow-to-configure-stack-gap-size.patch
deleted file mode 100644
index 6447d59..0000000
--- a/debian/patches/bugfix/all/fold-me-mm-allow-to-configure-stack-gap-size.patch
+++ /dev/null
@@ -1,79 +0,0 @@
-From: Michal Hocko <mhocko at suse.com>
-Date: Fri, 16 Jun 2017 00:06:28 +0200
-Subject: fold me "mm: allow to configure stack gap size"
-Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2017-1000364
-
-- do not rely on is_stack when reporting the gap. show_map_vma has
- all the information we need
-[carnil: backport for 3.16, adjust context, use of vm_is_stack in 3.16 rather
-is_stack]
----
- fs/proc/task_mmu.c | 19 +++++++++++++------
- 1 file changed, 13 insertions(+), 6 deletions(-)
-
---- a/fs/proc/task_mmu.c
-+++ b/fs/proc/task_mmu.c
-@@ -251,7 +251,7 @@ static int do_maps_open(struct inode *in
- }
-
- static void
--show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
-+show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid, bool *has_gap)
- {
- struct mm_struct *mm = vma->vm_mm;
- struct file *file = vma->vm_file;
-@@ -278,11 +278,17 @@ show_map_vma(struct seq_file *m, struct
- start = vma->vm_start;
- end = vma->vm_end;
- if (vma->vm_flags & VM_GROWSDOWN) {
-- if (stack_guard_area(vma, start))
-+ if (stack_guard_area(vma, start)) {
- start = min(end, start + stack_guard_gap);
-+ if (has_gap)
-+ *has_gap = true;
-+ }
- } else if (vma->vm_flags & VM_GROWSUP) {
-- if (stack_guard_area(vma, end))
-+ if (stack_guard_area(vma, end)) {
- end = max(start, end - stack_guard_gap);
-+ if (has_gap)
-+ *has_gap = true;
-+ }
- }
-
- seq_setwidth(m, 25 + sizeof(void *) * 6 - 1);
-@@ -359,7 +365,7 @@ static int show_map(struct seq_file *m,
- struct proc_maps_private *priv = m->private;
- struct task_struct *task = priv->task;
-
-- show_map_vma(m, vma, is_pid);
-+ show_map_vma(m, vma, is_pid, NULL);
-
- if (m->count < m->size) /* vma is copied successfully */
- m->version = (vma != get_gate_vma(task->mm))
-@@ -598,6 +604,7 @@ static int show_smap(struct seq_file *m,
- .mm = vma->vm_mm,
- .private = &mss,
- };
-+ bool has_gap = false;
-
- memset(&mss, 0, sizeof mss);
- mss.vma = vma;
-@@ -605,7 +612,7 @@ static int show_smap(struct seq_file *m,
- if (vma->vm_mm && !is_vm_hugetlb_page(vma))
- walk_page_range(vma->vm_start, vma->vm_end, &smaps_walk);
-
-- show_map_vma(m, vma, is_pid);
-+ show_map_vma(m, vma, is_pid, &has_gap);
-
- seq_printf(m,
- "Size: %8lu kB\n"
-@@ -642,7 +649,7 @@ static int show_smap(struct seq_file *m,
- seq_printf(m, "Nonlinear: %8lu kB\n",
- mss.nonlinear >> 10);
-
-- if (vm_is_stack(m->private, vma, is_pid))
-+ if (has_gap)
- seq_printf(m, "Stack_Gap: %8lu kB\n", stack_guard_gap >>10);
-
- show_smap_vma_flags(m, vma);
diff --git a/debian/patches/bugfix/all/mm-allow-to-configure-stack-gap-size.patch b/debian/patches/bugfix/all/mm-allow-to-configure-stack-gap-size.patch
deleted file mode 100644
index c516800..0000000
--- a/debian/patches/bugfix/all/mm-allow-to-configure-stack-gap-size.patch
+++ /dev/null
@@ -1,71 +0,0 @@
-From: Michal Hocko <mhocko at suse.com>
-Date: Wed, 14 Jun 2017 08:17:02 +0200
-Subject: mm: allow to configure stack gap size
-Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2017-1000364
-
-Add a kernel command line option (stack_guard_gap) to specify the stack
-gap size (in page unites) and export the value in /proc/<pid>/smaps for
-stack vmas. This might be used for special applications like CRIU/RR.
-
-Suggested-by: Linus Torvalds <torvalds at linux-foundation.org>
-Signed-off-by: Michal Hocko <mhocko at suse.com>
-[carnil: backport to 3.16
- - context adjustment
- - adjust location for documentation
- - is_stack -> vm_is_stack]
----
- Documentation/kernel-parameters.txt | 7 +++++++
- fs/proc/task_mmu.c | 3 +++
- mm/mmap.c | 13 +++++++++++++
- 3 files changed, 23 insertions(+)
-
---- a/Documentation/kernel-parameters.txt
-+++ b/Documentation/kernel-parameters.txt
-@@ -3150,6 +3150,13 @@ bytes respectively. Such letter suffixes
- spia_pedr=
- spia_peddr=
-
-+ stack_guard_gap= [MM]
-+ override the default stack gap protection. The value
-+ is in page units and it defines how many pages prior
-+ to (for stacks growing down) resp. after (for stacks
-+ growing up) the main stack are reserved for no other
-+ mapping. Default value is 256 pages.
-+
- stacktrace [FTRACE]
- Enabled the stack tracer on boot up.
-
---- a/fs/proc/task_mmu.c
-+++ b/fs/proc/task_mmu.c
-@@ -642,6 +642,9 @@ static int show_smap(struct seq_file *m,
- seq_printf(m, "Nonlinear: %8lu kB\n",
- mss.nonlinear >> 10);
-
-+ if (vm_is_stack(m->private, vma, is_pid))
-+ seq_printf(m, "Stack_Gap: %8lu kB\n", stack_guard_gap >>10);
-+
- show_smap_vma_flags(m, vma);
-
- if (m->count < m->size) /* vma is copied successfully */
---- a/mm/mmap.c
-+++ b/mm/mmap.c
-@@ -2261,6 +2261,19 @@ int expand_downwards(struct vm_area_stru
- /* enforced gap between the expanding stack and other mappings. */
- unsigned long stack_guard_gap = 256UL<<PAGE_SHIFT;
-
-+static int __init cmdline_parse_stack_guard_gap(char *p)
-+{
-+ unsigned long val;
-+ char *endptr;
-+
-+ val = simple_strtoul(p, &endptr, 10);
-+ if (!*endptr)
-+ stack_guard_gap = val << PAGE_SHIFT;
-+
-+ return 0;
-+}
-+__setup("stack_guard_gap=", cmdline_parse_stack_guard_gap);
-+
- /*
- * Note how expand_stack() refuses to expand the stack all the way to
- * abut the next virtual mapping, *unless* that mapping itself is also
diff --git a/debian/patches/bugfix/all/mm-do-not-collapse-stack-gap-into-THP.patch b/debian/patches/bugfix/all/mm-do-not-collapse-stack-gap-into-THP.patch
deleted file mode 100644
index 574e02d..0000000
--- a/debian/patches/bugfix/all/mm-do-not-collapse-stack-gap-into-THP.patch
+++ /dev/null
@@ -1,52 +0,0 @@
-From: Michal Hocko <mhocko at suse.com>
-Date: Wed, 14 Jun 2017 08:18:00 +0200
-Subject: mm: do not collapse stack gap into THP
-Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2017-1000364
-
-Oleg has noticed that khugepaged will happilly collapse stack vma (as
-long as it is not an early stack - see is_vma_temporary_stack) and
-it might effectively remove the stack gap area as well because a larger
-part of the stack vma is usually populated. The same applies to the
-page fault handler.
-
-Fix this by checking stack_guard_area when revalidating a VMA
-in hugepage_vma_revalidate. We do not want to hook/replace
-is_vma_temporary_stack() check because THP might be still useful for
-stack, all we need is excluding the gap from collapsing into a THP.
-
-Also check the to-be-created THP in do_huge_pmd_anonymous_page to
-make sure it is completely outside of the gap area because we we could
-create THP covering the gap area.
-
-Noticed-by: Oleg Nesterov <oleg at redhat.com>
-Signed-off-by: Michal Hocko <mhocko at suse.com>
-[carnil: backport to 3.16: move stack_guard_area check to huge_memory.c
-in collapse_huge_page]
----
- mm/huge_memory.c | 3 +++
- mm/khugepaged.c | 4 ++++
- 2 files changed, 7 insertions(+)
-
---- a/mm/huge_memory.c
-+++ b/mm/huge_memory.c
-@@ -792,6 +792,9 @@ int do_huge_pmd_anonymous_page(struct mm
-
- if (haddr < vma->vm_start || haddr + HPAGE_PMD_SIZE > vma->vm_end)
- return VM_FAULT_FALLBACK;
-+ if (stack_guard_area(vma, haddr) ||
-+ stack_guard_area(vma, haddr + HPAGE_PMD_SIZE))
-+ return VM_FAULT_FALLBACK;
- if (unlikely(anon_vma_prepare(vma)))
- return VM_FAULT_OOM;
- if (unlikely(khugepaged_enter(vma, vma->vm_flags)))
-@@ -2435,6 +2438,10 @@ static void collapse_huge_page(struct mm
- goto out;
- if (!hugepage_vma_check(vma))
- goto out;
-+
-+ /* never try to collapse stack gap */
-+ if (stack_guard_area(vma, hstart) || stack_guard_area(vma, hend))
-+ goto out;
- pmd = mm_find_pmd(mm, address);
- if (!pmd)
- goto out;
diff --git a/debian/patches/bugfix/all/mm-enlarge-stack-guard-gap.patch b/debian/patches/bugfix/all/mm-enlarge-stack-guard-gap.patch
deleted file mode 100644
index 9131d42..0000000
--- a/debian/patches/bugfix/all/mm-enlarge-stack-guard-gap.patch
+++ /dev/null
@@ -1,473 +0,0 @@
-From: Michal Hocko <mhocko at suse.com>
-Date: Wed, 14 Jun 2017 08:16:54 +0200
-Subject: mm: enlarge stack guard gap
-Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2017-1000364
-
-Stack guard page is a useful feature to reduce a risk of stack smashing
-into a different mapping. We have been using a single page gap which
-is sufficient to prevent having stack adjacent to a different mapping.
-But this seems to be insufficient in the light of the stack usage in
-the userspace. E.g. glibc uses as large as 64kB alloca() in many
-commonly used functions. Others use constructs liks gid_t
-buffer[NGROUPS_MAX] which is 256kB or stack strings with MAX_ARG_STRLEN.
-
-This will become especially dangerous for suid binaries and the default
-no limit for the stack size limit because those applications can be
-tricked to consume a large portion of the stack and a single glibc call
-could jump over the guard page. These attacks are not theoretical,
-unfortunatelly.
-
-Make those attacks less probable by increasing the stack guard gap
-to 1MB (on systems with 4k pages but make it depend on the page size
-because systems with larger base pages might cap stack allocations in
-the PAGE_SIZE units) which should cover larger alloca() and VLA stack
-allocations. It is obviously not a full fix because the problem is
-somehow inherent but it should reduce attack space a lot. One could
-argue that the gap size should be configurable from the userspace but
-that can be done later on top when somebody finds that the new 1MB is
-not suitable or even wrong for some special case applications.
-
-Implementation wise, get rid of check_stack_guard_page and move all the
-guard page specific code to expandable_stack_area which always tries to
-guarantee the gap. do_anonymous_page then just calls expand_stack. Also
-get rid of stack_guard_page_{start,end} and replace them with
-stack_guard_area to handle stack population and /proc/<pid>/[s]maps.
-
-This should clean up the code which is quite scattered currently
-and therefore justify the change.
-
-Signed-off-by: Michal Hocko <mhocko at suse.com>
-[carnil: backport for 3.16:
- - vmf->address -> address;
- - context adjustment;
- - backport for pre-"FOLL_MLOCK"-"FOLL_POPULATE"-rename]
----
- arch/ia64/mm/fault.c | 2 +-
- fs/exec.c | 8 ++-
- fs/proc/task_mmu.c | 11 ++--
- include/linux/mm.h | 40 +++-----------
- mm/gup.c | 4 +-
- mm/memory.c | 38 ++-----------
- mm/mmap.c | 152 +++++++++++++++++++++++++++++++++++++++++----------
- 7 files changed, 152 insertions(+), 103 deletions(-)
-
---- a/arch/ia64/mm/fault.c
-+++ b/arch/ia64/mm/fault.c
-@@ -224,7 +224,7 @@ retry:
- */
- if (address > vma->vm_end + PAGE_SIZE - sizeof(long))
- goto bad_area;
-- if (expand_upwards(vma, address))
-+ if (expand_upwards(vma, address, 0))
- goto bad_area;
- }
- goto good_area;
---- a/fs/exec.c
-+++ b/fs/exec.c
-@@ -193,7 +193,7 @@ static struct page *get_arg_page(struct
-
- #ifdef CONFIG_STACK_GROWSUP
- if (write) {
-- ret = expand_downwards(bprm->vma, pos);
-+ ret = expand_downwards(bprm->vma, pos, 0);
- if (ret < 0)
- return NULL;
- }
-@@ -207,6 +207,12 @@ static struct page *get_arg_page(struct
- unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start;
- struct rlimit *rlim;
-
-+ /*
-+ * GRWOSUP doesn't really have any gap at this stage because we grow
-+ * the stack down now. See the expand_downwards above.
-+ */
-+ if (!IS_ENABLED(CONFIG_STACK_GROWSUP))
-+ size -= stack_guard_gap;
- acct_arg_size(bprm, size / PAGE_SIZE);
-
- /*
---- a/fs/proc/task_mmu.c
-+++ b/fs/proc/task_mmu.c
-@@ -276,11 +276,14 @@ show_map_vma(struct seq_file *m, struct
-
- /* We don't show the stack guard page in /proc/maps */
- start = vma->vm_start;
-- if (stack_guard_page_start(vma, start))
-- start += PAGE_SIZE;
- end = vma->vm_end;
-- if (stack_guard_page_end(vma, end))
-- end -= PAGE_SIZE;
-+ if (vma->vm_flags & VM_GROWSDOWN) {
-+ if (stack_guard_area(vma, start))
-+ start += stack_guard_gap;
-+ } else if (vma->vm_flags & VM_GROWSUP) {
-+ if (stack_guard_area(vma, end))
-+ end -= stack_guard_gap;
-+ }
-
- seq_setwidth(m, 25 + sizeof(void *) * 6 - 1);
- seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu ",
---- a/include/linux/mm.h
-+++ b/include/linux/mm.h
-@@ -1263,34 +1263,6 @@ int set_page_dirty_lock(struct page *pag
- int clear_page_dirty_for_io(struct page *page);
- int get_cmdline(struct task_struct *task, char *buffer, int buflen);
-
--/* Is the vma a continuation of the stack vma above it? */
--static inline int vma_growsdown(struct vm_area_struct *vma, unsigned long addr)
--{
-- return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN);
--}
--
--static inline int stack_guard_page_start(struct vm_area_struct *vma,
-- unsigned long addr)
--{
-- return (vma->vm_flags & VM_GROWSDOWN) &&
-- (vma->vm_start == addr) &&
-- !vma_growsdown(vma->vm_prev, addr);
--}
--
--/* Is the vma a continuation of the stack vma below it? */
--static inline int vma_growsup(struct vm_area_struct *vma, unsigned long addr)
--{
-- return vma && (vma->vm_start == addr) && (vma->vm_flags & VM_GROWSUP);
--}
--
--static inline int stack_guard_page_end(struct vm_area_struct *vma,
-- unsigned long addr)
--{
-- return (vma->vm_flags & VM_GROWSUP) &&
-- (vma->vm_end == addr) &&
-- !vma_growsup(vma->vm_next, addr);
--}
--
- extern pid_t
- vm_is_stack(struct task_struct *task, struct vm_area_struct *vma, int in_group);
-
-@@ -1936,16 +1908,22 @@ void page_cache_async_readahead(struct a
-
- unsigned long max_sane_readahead(unsigned long nr);
-
-+extern unsigned long stack_guard_gap;
- /* Generic expand stack which grows the stack according to GROWS{UP,DOWN} */
- extern int expand_stack(struct vm_area_struct *vma, unsigned long address);
-+extern int stack_guard_area(struct vm_area_struct *vma, unsigned long address);
-
- /* CONFIG_STACK_GROWSUP still needs to to grow downwards at some places */
- extern int expand_downwards(struct vm_area_struct *vma,
-- unsigned long address);
-+ unsigned long address, unsigned long gap);
-+unsigned long expandable_stack_area(struct vm_area_struct *vma,
-+ unsigned long address, unsigned long *gap);
-+
- #if VM_GROWSUP
--extern int expand_upwards(struct vm_area_struct *vma, unsigned long address);
-+extern int expand_upwards(struct vm_area_struct *vma,
-+ unsigned long address, unsigned long gap);
- #else
-- #define expand_upwards(vma, address) (0)
-+ #define expand_upwards(vma, address, gap) (0)
- #endif
-
- /* Look up the first VMA which satisfies addr < vm_end, NULL if none. */
---- a/mm/gup.c
-+++ b/mm/gup.c
-@@ -267,9 +267,7 @@ static int faultin_page(struct task_stru
- int ret;
-
- /* For mlock, just skip the stack guard page. */
-- if ((*flags & FOLL_MLOCK) &&
-- (stack_guard_page_start(vma, address) ||
-- stack_guard_page_end(vma, address + PAGE_SIZE)))
-+ if ((*flags & FOLL_MLOCK) && stack_guard_area(vma, address))
- return -ENOENT;
- if (*flags & FOLL_WRITE)
- fault_flags |= FAULT_FLAG_WRITE;
---- a/mm/memory.c
-+++ b/mm/memory.c
-@@ -2588,39 +2588,7 @@ out_release:
- return ret;
- }
-
--/*
-- * This is like a special single-page "expand_{down|up}wards()",
-- * except we must first make sure that 'address{-|+}PAGE_SIZE'
-- * doesn't hit another vma.
-- */
--static inline int check_stack_guard_page(struct vm_area_struct *vma, unsigned long address)
--{
-- address &= PAGE_MASK;
-- if ((vma->vm_flags & VM_GROWSDOWN) && address == vma->vm_start) {
-- struct vm_area_struct *prev = vma->vm_prev;
--
-- /*
-- * Is there a mapping abutting this one below?
-- *
-- * That's only ok if it's the same stack mapping
-- * that has gotten split..
-- */
-- if (prev && prev->vm_end == address)
-- return prev->vm_flags & VM_GROWSDOWN ? 0 : -ENOMEM;
--
-- return expand_downwards(vma, address - PAGE_SIZE);
-- }
-- if ((vma->vm_flags & VM_GROWSUP) && address + PAGE_SIZE == vma->vm_end) {
-- struct vm_area_struct *next = vma->vm_next;
-
-- /* As VM_GROWSDOWN but s/below/above/ */
-- if (next && next->vm_start == address + PAGE_SIZE)
-- return next->vm_flags & VM_GROWSUP ? 0 : -ENOMEM;
--
-- return expand_upwards(vma, address + PAGE_SIZE);
-- }
-- return 0;
--}
-
- /*
- * We enter with non-exclusive mmap_sem (to exclude vma changes,
-@@ -2642,8 +2610,10 @@ static int do_anonymous_page(struct mm_s
- return VM_FAULT_SIGBUS;
-
- /* Check if we need to add a guard page to the stack */
-- if (check_stack_guard_page(vma, address) < 0)
-- return VM_FAULT_SIGSEGV;
-+ if (stack_guard_area(vma, address)) {
-+ if (expand_stack(vma, address) < 0)
-+ return VM_FAULT_SIGSEGV;
-+ }
-
- /* Use the zero-page for reads */
- if (!(flags & FAULT_FLAG_WRITE)) {
---- a/mm/mmap.c
-+++ b/mm/mmap.c
-@@ -2079,7 +2079,8 @@ find_vma_prev(struct mm_struct *mm, unsi
- * update accounting. This is shared with both the
- * grow-up and grow-down cases.
- */
--static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow)
-+static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow,
-+ unsigned long gap)
- {
- struct mm_struct *mm = vma->vm_mm;
- struct rlimit *rlim = current->signal->rlim;
-@@ -2092,7 +2093,7 @@ static int acct_stack_growth(struct vm_a
- /* Stack limit test */
- actual_size = size;
- if (size && (vma->vm_flags & (VM_GROWSUP | VM_GROWSDOWN)))
-- actual_size -= PAGE_SIZE;
-+ actual_size -= gap;
- if (actual_size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur))
- return -ENOMEM;
-
-@@ -2132,19 +2133,13 @@ static int acct_stack_growth(struct vm_a
- * PA-RISC uses this for its stack; IA64 for its Register Backing Store.
- * vma is the last one with address > vma->vm_end. Have to extend vma.
- */
--int expand_upwards(struct vm_area_struct *vma, unsigned long address)
-+int expand_upwards(struct vm_area_struct *vma, unsigned long address, unsigned long gap)
- {
- int error = 0;
-
- if (!(vma->vm_flags & VM_GROWSUP))
- return -EFAULT;
-
-- /* Guard against wrapping around to address 0. */
-- if (address < PAGE_ALIGN(address+4))
-- address = PAGE_ALIGN(address+4);
-- else
-- return -ENOMEM;
--
- /* We must make sure the anon_vma is allocated. */
- if (unlikely(anon_vma_prepare(vma)))
- return -ENOMEM;
-@@ -2165,7 +2160,7 @@ int expand_upwards(struct vm_area_struct
-
- error = -ENOMEM;
- if (vma->vm_pgoff + (size >> PAGE_SHIFT) >= vma->vm_pgoff) {
-- error = acct_stack_growth(vma, size, grow);
-+ error = acct_stack_growth(vma, size, grow, gap);
- if (!error) {
- /*
- * vma_gap_update() doesn't support concurrent
-@@ -2203,7 +2198,7 @@ int expand_upwards(struct vm_area_struct
- * vma is the first one with address < vma->vm_start. Have to extend vma.
- */
- int expand_downwards(struct vm_area_struct *vma,
-- unsigned long address)
-+ unsigned long address, unsigned long gap)
- {
- int error;
-
-@@ -2232,7 +2227,7 @@ int expand_downwards(struct vm_area_stru
-
- error = -ENOMEM;
- if (grow <= vma->vm_pgoff) {
-- error = acct_stack_growth(vma, size, grow);
-+ error = acct_stack_growth(vma, size, grow, gap);
- if (!error) {
- /*
- * vma_gap_update() doesn't support concurrent
-@@ -2263,29 +2258,72 @@ int expand_downwards(struct vm_area_stru
- return error;
- }
-
-+/* enforced gap between the expanding stack and other mappings. */
-+unsigned long stack_guard_gap = 256UL<<PAGE_SHIFT;
-+
- /*
- * Note how expand_stack() refuses to expand the stack all the way to
- * abut the next virtual mapping, *unless* that mapping itself is also
-- * a stack mapping. We want to leave room for a guard page, after all
-+ * a stack mapping. We want to leave room for a guard area, after all
- * (the guard page itself is not added here, that is done by the
- * actual page faulting logic)
-- *
-- * This matches the behavior of the guard page logic (see mm/memory.c:
-- * check_stack_guard_page()), which only allows the guard page to be
-- * removed under these circumstances.
- */
- #ifdef CONFIG_STACK_GROWSUP
-+unsigned long expandable_stack_area(struct vm_area_struct *vma,
-+ unsigned long address, unsigned long *gap)
-+{
-+ struct vm_area_struct *next = vma->vm_next;
-+ unsigned long guard_gap = stack_guard_gap;
-+ unsigned long guard_addr;
-+
-+ address = ALIGN(address, PAGE_SIZE);;
-+ if (!next)
-+ goto out;
-+
-+ if (next->vm_flags & VM_GROWSUP) {
-+ guard_gap = min(guard_gap, next->vm_start - address);
-+ goto out;
-+ }
-+
-+ if (next->vm_start - address < guard_gap)
-+ return -ENOMEM;
-+out:
-+ if (TASK_SIZE - address < guard_gap)
-+ guard_gap = TASK_SIZE - address;
-+ guard_addr = address + guard_gap;
-+ *gap = guard_gap;
-+
-+ return guard_addr;
-+}
-+
- int expand_stack(struct vm_area_struct *vma, unsigned long address)
- {
-+ unsigned long gap;
-+
-+ address = expandable_stack_area(vma, address, &gap);
-+ if (IS_ERR_VALUE(address))
-+ return -ENOMEM;
-+ return expand_upwards(vma, address, gap);
-+}
-+
-+int stack_guard_area(struct vm_area_struct *vma, unsigned long address)
-+{
- struct vm_area_struct *next;
-
-- address &= PAGE_MASK;
-+ if (!(vma->vm_flags & VM_GROWSUP))
-+ return 0;
-+
-+ /*
-+ * strictly speaking there is a guard gap between disjoint stacks
-+ * but the gap is not canonical (it might be smaller) and it is
-+ * reasonably safe to assume that we can ignore that gap for stack
-+ * POPULATE or /proc/<pid>[s]maps purposes
-+ */
- next = vma->vm_next;
-- if (next && next->vm_start == address + PAGE_SIZE) {
-- if (!(next->vm_flags & VM_GROWSUP))
-- return -ENOMEM;
-- }
-- return expand_upwards(vma, address);
-+ if (next && next->vm_flags & VM_GROWSUP)
-+ return 0;
-+
-+ return vma->vm_end - address <= stack_guard_gap;
- }
-
- struct vm_area_struct *
-@@ -2304,17 +2342,73 @@ find_extend_vma(struct mm_struct *mm, un
- return prev;
- }
- #else
-+unsigned long expandable_stack_area(struct vm_area_struct *vma,
-+ unsigned long address, unsigned long *gap)
-+{
-+ struct vm_area_struct *prev = vma->vm_prev;
-+ unsigned long guard_gap = stack_guard_gap;
-+ unsigned long guard_addr;
-+
-+ address &= PAGE_MASK;
-+ if (!prev)
-+ goto out;
-+
-+ /*
-+ * Is there a mapping abutting this one below?
-+ *
-+ * That's only ok if it's the same stack mapping
-+ * that has gotten split or there is sufficient gap
-+ * between mappings
-+ */
-+ if (prev->vm_flags & VM_GROWSDOWN) {
-+ guard_gap = min(guard_gap, address - prev->vm_end);
-+ goto out;
-+ }
-+
-+ if (address - prev->vm_end < guard_gap)
-+ return -ENOMEM;
-+
-+out:
-+ /* make sure we won't underflow */
-+ if (address < mmap_min_addr)
-+ return -ENOMEM;
-+ if (address - mmap_min_addr < guard_gap)
-+ guard_gap = address - mmap_min_addr;
-+
-+ guard_addr = address - guard_gap;
-+ *gap = guard_gap;
-+
-+ return guard_addr;
-+}
-+
- int expand_stack(struct vm_area_struct *vma, unsigned long address)
- {
-+ unsigned long gap;
-+
-+ address = expandable_stack_area(vma, address, &gap);
-+ if (IS_ERR_VALUE(address))
-+ return -ENOMEM;
-+ return expand_downwards(vma, address, gap);
-+}
-+
-+int stack_guard_area(struct vm_area_struct *vma, unsigned long address)
-+{
- struct vm_area_struct *prev;
-
-- address &= PAGE_MASK;
-+ if (!(vma->vm_flags & VM_GROWSDOWN))
-+ return 0;
-+
-+ /*
-+ * strictly speaking there is a guard gap between disjoint stacks
-+ * but the gap is not canonical (it might be smaller) and it is
-+ * reasonably safe to assume that we can ignore that gap for stack
-+ * POPULATE or /proc/<pid>[s]maps purposes
-+ */
- prev = vma->vm_prev;
-- if (prev && prev->vm_end == address) {
-- if (!(prev->vm_flags & VM_GROWSDOWN))
-- return -ENOMEM;
-- }
-- return expand_downwards(vma, address);
-+ if (prev && prev->vm_flags & VM_GROWSDOWN)
-+ return 0;
-+
-+ return address - vma->vm_start < stack_guard_gap;
- }
-
- struct vm_area_struct *
diff --git a/debian/patches/bugfix/all/mm-fix-new-crash-in-unmapped_area_topdown.patch b/debian/patches/bugfix/all/mm-fix-new-crash-in-unmapped_area_topdown.patch
new file mode 100644
index 0000000..3765867
--- /dev/null
+++ b/debian/patches/bugfix/all/mm-fix-new-crash-in-unmapped_area_topdown.patch
@@ -0,0 +1,46 @@
+From: Hugh Dickins <hughd at google.com>
+Date: Tue, 20 Jun 2017 02:10:44 -0700
+Subject: mm: fix new crash in unmapped_area_topdown()
+Origin: https://git.kernel.org/linus/f4cb767d76cf7ee72f97dd76f6cfa6c76a5edc89
+Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2017-1000364
+
+Trinity gets kernel BUG at mm/mmap.c:1963! in about 3 minutes of
+mmap testing. That's the VM_BUG_ON(gap_end < gap_start) at the
+end of unmapped_area_topdown(). Linus points out how MAP_FIXED
+(which does not have to respect our stack guard gap intentions)
+could result in gap_end below gap_start there. Fix that, and
+the similar case in its alternative, unmapped_area().
+
+Cc: stable at vger.kernel.org
+Fixes: 1be7107fbe18 ("mm: larger stack guard gap, between vmas")
+Reported-by: Dave Jones <davej at codemonkey.org.uk>
+Debugged-by: Linus Torvalds <torvalds at linux-foundation.org>
+Signed-off-by: Hugh Dickins <hughd at google.com>
+Acked-by: Michal Hocko <mhocko at suse.com>
+Signed-off-by: Linus Torvalds <torvalds at linux-foundation.org>
+---
+ mm/mmap.c | 6 ++++--
+ 1 file changed, 4 insertions(+), 2 deletions(-)
+
+--- a/mm/mmap.c
++++ b/mm/mmap.c
+@@ -1750,7 +1750,8 @@ check_current:
+ /* Check if current node has a suitable gap */
+ if (gap_start > high_limit)
+ return -ENOMEM;
+- if (gap_end >= low_limit && gap_end - gap_start >= length)
++ if (gap_end >= low_limit &&
++ gap_end > gap_start && gap_end - gap_start >= length)
+ goto found;
+
+ /* Visit right subtree if it looks promising */
+@@ -1853,7 +1854,8 @@ check_current:
+ gap_end = vm_start_gap(vma);
+ if (gap_end < low_limit)
+ return -ENOMEM;
+- if (gap_start <= high_limit && gap_end - gap_start >= length)
++ if (gap_start <= high_limit &&
++ gap_end > gap_start && gap_end - gap_start >= length)
+ goto found;
+
+ /* Visit left subtree if it looks promising */
diff --git a/debian/patches/bugfix/all/mm-larger-stack-guard-gap-between-vmas.patch b/debian/patches/bugfix/all/mm-larger-stack-guard-gap-between-vmas.patch
new file mode 100644
index 0000000..7d4fd1b
--- /dev/null
+++ b/debian/patches/bugfix/all/mm-larger-stack-guard-gap-between-vmas.patch
@@ -0,0 +1,807 @@
+From: Hugh Dickins <hughd at google.com>
+Date: Mon, 19 Jun 2017 04:03:24 -0700
+Subject: mm: larger stack guard gap, between vmas
+Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2017-1000364
+
+commit 1be7107fbe18eed3e319a6c3e83c78254b693acb upstream.
+
+Stack guard page is a useful feature to reduce a risk of stack smashing
+into a different mapping. We have been using a single page gap which
+is sufficient to prevent having stack adjacent to a different mapping.
+But this seems to be insufficient in the light of the stack usage in
+userspace. E.g. glibc uses as large as 64kB alloca() in many commonly
+used functions. Others use constructs liks gid_t buffer[NGROUPS_MAX]
+which is 256kB or stack strings with MAX_ARG_STRLEN.
+
+This will become especially dangerous for suid binaries and the default
+no limit for the stack size limit because those applications can be
+tricked to consume a large portion of the stack and a single glibc call
+could jump over the guard page. These attacks are not theoretical,
+unfortunatelly.
+
+Make those attacks less probable by increasing the stack guard gap
+to 1MB (on systems with 4k pages; but make it depend on the page size
+because systems with larger base pages might cap stack allocations in
+the PAGE_SIZE units) which should cover larger alloca() and VLA stack
+allocations. It is obviously not a full fix because the problem is
+somehow inherent, but it should reduce attack space a lot.
+
+One could argue that the gap size should be configurable from userspace,
+but that can be done later when somebody finds that the new 1MB is wrong
+for some special case applications. For now, add a kernel command line
+option (stack_guard_gap) to specify the stack gap size (in page units).
+
+Implementation wise, first delete all the old code for stack guard page:
+because although we could get away with accounting one extra page in a
+stack vma, accounting a larger gap can break userspace - case in point,
+a program run with "ulimit -S -v 20000" failed when the 1MB gap was
+counted for RLIMIT_AS; similar problems could come with RLIMIT_MLOCK
+and strict non-overcommit mode.
+
+Instead of keeping gap inside the stack vma, maintain the stack guard
+gap as a gap between vmas: using vm_start_gap() in place of vm_start
+(or vm_end_gap() in place of vm_end if VM_GROWSUP) in just those few
+places which need to respect the gap - mainly arch_get_unmapped_area(),
+and and the vma tree's subtree_gap support for that.
+
+Original-patch-by: Oleg Nesterov <oleg at redhat.com>
+Original-patch-by: Michal Hocko <mhocko at suse.com>
+Signed-off-by: Hugh Dickins <hughd at google.com>
+Acked-by: Michal Hocko <mhocko at suse.com>
+Tested-by: Helge Deller <deller at gmx.de> # parisc
+Signed-off-by: Linus Torvalds <torvalds at linux-foundation.org>
+---
+Here's what I thought the 3.16.45 version of the stack guard gap
+patch would look like. This one is easy and uncontroversial (if
+Willy and I get to agreeing about FOLL_MLOCK). Whereas 3.2...
+
+Hugh
+
+--- a/Documentation/kernel-parameters.txt
++++ b/Documentation/kernel-parameters.txt
+@@ -3150,6 +3150,13 @@ bytes respectively. Such letter suffixes
+ spia_pedr=
+ spia_peddr=
+
++ stack_guard_gap= [MM]
++ override the default stack gap protection. The value
++ is in page units and it defines how many pages prior
++ to (for stacks growing down) resp. after (for stacks
++ growing up) the main stack are reserved for no other
++ mapping. Default value is 256 pages.
++
+ stacktrace [FTRACE]
+ Enabled the stack tracer on boot up.
+
+--- a/arch/arc/mm/mmap.c
++++ b/arch/arc/mm/mmap.c
+@@ -64,7 +64,7 @@ arch_get_unmapped_area(struct file *filp
+
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+--- a/arch/arm/mm/mmap.c
++++ b/arch/arm/mm/mmap.c
+@@ -89,7 +89,7 @@ arch_get_unmapped_area(struct file *filp
+
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+@@ -140,7 +140,7 @@ arch_get_unmapped_area_topdown(struct fi
+ addr = PAGE_ALIGN(addr);
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+--- a/arch/frv/mm/elf-fdpic.c
++++ b/arch/frv/mm/elf-fdpic.c
+@@ -74,7 +74,7 @@ unsigned long arch_get_unmapped_area(str
+ addr = PAGE_ALIGN(addr);
+ vma = find_vma(current->mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ goto success;
+ }
+
+--- a/arch/mips/mm/mmap.c
++++ b/arch/mips/mm/mmap.c
+@@ -92,7 +92,7 @@ static unsigned long arch_get_unmapped_a
+
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+--- a/arch/parisc/kernel/sys_parisc.c
++++ b/arch/parisc/kernel/sys_parisc.c
+@@ -88,7 +88,7 @@ unsigned long arch_get_unmapped_area(str
+ unsigned long len, unsigned long pgoff, unsigned long flags)
+ {
+ struct mm_struct *mm = current->mm;
+- struct vm_area_struct *vma;
++ struct vm_area_struct *vma, *prev;
+ unsigned long task_size = TASK_SIZE;
+ int do_color_align, last_mmap;
+ struct vm_unmapped_area_info info;
+@@ -115,9 +115,10 @@ unsigned long arch_get_unmapped_area(str
+ else
+ addr = PAGE_ALIGN(addr);
+
+- vma = find_vma(mm, addr);
++ vma = find_vma_prev(mm, addr, &prev);
+ if (task_size - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)) &&
++ (!prev || addr >= vm_end_gap(prev)))
+ goto found_addr;
+ }
+
+@@ -141,7 +142,7 @@ arch_get_unmapped_area_topdown(struct fi
+ const unsigned long len, const unsigned long pgoff,
+ const unsigned long flags)
+ {
+- struct vm_area_struct *vma;
++ struct vm_area_struct *vma, *prev;
+ struct mm_struct *mm = current->mm;
+ unsigned long addr = addr0;
+ int do_color_align, last_mmap;
+@@ -175,9 +176,11 @@ arch_get_unmapped_area_topdown(struct fi
+ addr = COLOR_ALIGN(addr, last_mmap, pgoff);
+ else
+ addr = PAGE_ALIGN(addr);
+- vma = find_vma(mm, addr);
++
++ vma = find_vma_prev(mm, addr, &prev);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)) &&
++ (!prev || addr >= vm_end_gap(prev)))
+ goto found_addr;
+ }
+
+--- a/arch/powerpc/mm/slice.c
++++ b/arch/powerpc/mm/slice.c
+@@ -103,7 +103,7 @@ static int slice_area_is_free(struct mm_
+ if ((mm->task_size - len) < addr)
+ return 0;
+ vma = find_vma(mm, addr);
+- return (!vma || (addr + len) <= vma->vm_start);
++ return (!vma || (addr + len) <= vm_start_gap(vma));
+ }
+
+ static int slice_low_has_vma(struct mm_struct *mm, unsigned long slice)
+--- a/arch/sh/mm/mmap.c
++++ b/arch/sh/mm/mmap.c
+@@ -63,7 +63,7 @@ unsigned long arch_get_unmapped_area(str
+
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+@@ -113,7 +113,7 @@ arch_get_unmapped_area_topdown(struct fi
+
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+--- a/arch/sparc/kernel/sys_sparc_64.c
++++ b/arch/sparc/kernel/sys_sparc_64.c
+@@ -118,7 +118,7 @@ unsigned long arch_get_unmapped_area(str
+
+ vma = find_vma(mm, addr);
+ if (task_size - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+@@ -181,7 +181,7 @@ arch_get_unmapped_area_topdown(struct fi
+
+ vma = find_vma(mm, addr);
+ if (task_size - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+--- a/arch/sparc/mm/hugetlbpage.c
++++ b/arch/sparc/mm/hugetlbpage.c
+@@ -115,7 +115,7 @@ hugetlb_get_unmapped_area(struct file *f
+ addr = ALIGN(addr, HPAGE_SIZE);
+ vma = find_vma(mm, addr);
+ if (task_size - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+ if (mm->get_unmapped_area == arch_get_unmapped_area)
+--- a/arch/tile/mm/hugetlbpage.c
++++ b/arch/tile/mm/hugetlbpage.c
+@@ -265,7 +265,7 @@ unsigned long hugetlb_get_unmapped_area(
+ addr = ALIGN(addr, huge_page_size(h));
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+ if (current->mm->get_unmapped_area == arch_get_unmapped_area)
+--- a/arch/x86/kernel/sys_x86_64.c
++++ b/arch/x86/kernel/sys_x86_64.c
+@@ -127,7 +127,7 @@ arch_get_unmapped_area(struct file *filp
+ addr = PAGE_ALIGN(addr);
+ vma = find_vma(mm, addr);
+ if (end - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+@@ -166,7 +166,7 @@ arch_get_unmapped_area_topdown(struct fi
+ addr = PAGE_ALIGN(addr);
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+--- a/arch/x86/mm/hugetlbpage.c
++++ b/arch/x86/mm/hugetlbpage.c
+@@ -156,7 +156,7 @@ hugetlb_get_unmapped_area(struct file *f
+ addr = ALIGN(addr, huge_page_size(h));
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+ if (mm->get_unmapped_area == arch_get_unmapped_area)
+--- a/arch/xtensa/kernel/syscall.c
++++ b/arch/xtensa/kernel/syscall.c
+@@ -86,7 +86,7 @@ unsigned long arch_get_unmapped_area(str
+ /* At this point: (!vmm || addr < vmm->vm_end). */
+ if (TASK_SIZE - len < addr)
+ return -ENOMEM;
+- if (!vmm || addr + len <= vmm->vm_start)
++ if (!vmm || addr + len <= vm_start_gap(vmm))
+ return addr;
+ addr = vmm->vm_end;
+ if (flags & MAP_SHARED)
+--- a/fs/hugetlbfs/inode.c
++++ b/fs/hugetlbfs/inode.c
+@@ -171,7 +171,7 @@ hugetlb_get_unmapped_area(struct file *f
+ addr = ALIGN(addr, huge_page_size(h));
+ vma = find_vma(mm, addr);
+ if (TASK_SIZE - len >= addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)))
+ return addr;
+ }
+
+--- a/fs/proc/task_mmu.c
++++ b/fs/proc/task_mmu.c
+@@ -276,11 +276,7 @@ show_map_vma(struct seq_file *m, struct
+
+ /* We don't show the stack guard page in /proc/maps */
+ start = vma->vm_start;
+- if (stack_guard_page_start(vma, start))
+- start += PAGE_SIZE;
+ end = vma->vm_end;
+- if (stack_guard_page_end(vma, end))
+- end -= PAGE_SIZE;
+
+ seq_setwidth(m, 25 + sizeof(void *) * 6 - 1);
+ seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu ",
+--- a/include/linux/mm.h
++++ b/include/linux/mm.h
+@@ -1263,34 +1263,6 @@ int set_page_dirty_lock(struct page *pag
+ int clear_page_dirty_for_io(struct page *page);
+ int get_cmdline(struct task_struct *task, char *buffer, int buflen);
+
+-/* Is the vma a continuation of the stack vma above it? */
+-static inline int vma_growsdown(struct vm_area_struct *vma, unsigned long addr)
+-{
+- return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN);
+-}
+-
+-static inline int stack_guard_page_start(struct vm_area_struct *vma,
+- unsigned long addr)
+-{
+- return (vma->vm_flags & VM_GROWSDOWN) &&
+- (vma->vm_start == addr) &&
+- !vma_growsdown(vma->vm_prev, addr);
+-}
+-
+-/* Is the vma a continuation of the stack vma below it? */
+-static inline int vma_growsup(struct vm_area_struct *vma, unsigned long addr)
+-{
+- return vma && (vma->vm_start == addr) && (vma->vm_flags & VM_GROWSUP);
+-}
+-
+-static inline int stack_guard_page_end(struct vm_area_struct *vma,
+- unsigned long addr)
+-{
+- return (vma->vm_flags & VM_GROWSUP) &&
+- (vma->vm_end == addr) &&
+- !vma_growsup(vma->vm_next, addr);
+-}
+-
+ extern pid_t
+ vm_is_stack(struct task_struct *task, struct vm_area_struct *vma, int in_group);
+
+@@ -1936,6 +1908,7 @@ void page_cache_async_readahead(struct a
+
+ unsigned long max_sane_readahead(unsigned long nr);
+
++extern unsigned long stack_guard_gap;
+ /* Generic expand stack which grows the stack according to GROWS{UP,DOWN} */
+ extern int expand_stack(struct vm_area_struct *vma, unsigned long address);
+
+@@ -1964,6 +1937,30 @@ static inline struct vm_area_struct * fi
+ return vma;
+ }
+
++static inline unsigned long vm_start_gap(struct vm_area_struct *vma)
++{
++ unsigned long vm_start = vma->vm_start;
++
++ if (vma->vm_flags & VM_GROWSDOWN) {
++ vm_start -= stack_guard_gap;
++ if (vm_start > vma->vm_start)
++ vm_start = 0;
++ }
++ return vm_start;
++}
++
++static inline unsigned long vm_end_gap(struct vm_area_struct *vma)
++{
++ unsigned long vm_end = vma->vm_end;
++
++ if (vma->vm_flags & VM_GROWSUP) {
++ vm_end += stack_guard_gap;
++ if (vm_end < vma->vm_end)
++ vm_end = -PAGE_SIZE;
++ }
++ return vm_end;
++}
++
+ static inline unsigned long vma_pages(struct vm_area_struct *vma)
+ {
+ return (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
+--- a/mm/gup.c
++++ b/mm/gup.c
+@@ -266,11 +266,6 @@ static int faultin_page(struct task_stru
+ unsigned int fault_flags = 0;
+ int ret;
+
+- /* For mlock, just skip the stack guard page. */
+- if ((*flags & FOLL_MLOCK) &&
+- (stack_guard_page_start(vma, address) ||
+- stack_guard_page_end(vma, address + PAGE_SIZE)))
+- return -ENOENT;
+ if (*flags & FOLL_WRITE)
+ fault_flags |= FAULT_FLAG_WRITE;
+ if (nonblocking)
+--- a/mm/memory.c
++++ b/mm/memory.c
+@@ -2589,40 +2589,6 @@ out_release:
+ }
+
+ /*
+- * This is like a special single-page "expand_{down|up}wards()",
+- * except we must first make sure that 'address{-|+}PAGE_SIZE'
+- * doesn't hit another vma.
+- */
+-static inline int check_stack_guard_page(struct vm_area_struct *vma, unsigned long address)
+-{
+- address &= PAGE_MASK;
+- if ((vma->vm_flags & VM_GROWSDOWN) && address == vma->vm_start) {
+- struct vm_area_struct *prev = vma->vm_prev;
+-
+- /*
+- * Is there a mapping abutting this one below?
+- *
+- * That's only ok if it's the same stack mapping
+- * that has gotten split..
+- */
+- if (prev && prev->vm_end == address)
+- return prev->vm_flags & VM_GROWSDOWN ? 0 : -ENOMEM;
+-
+- return expand_downwards(vma, address - PAGE_SIZE);
+- }
+- if ((vma->vm_flags & VM_GROWSUP) && address + PAGE_SIZE == vma->vm_end) {
+- struct vm_area_struct *next = vma->vm_next;
+-
+- /* As VM_GROWSDOWN but s/below/above/ */
+- if (next && next->vm_start == address + PAGE_SIZE)
+- return next->vm_flags & VM_GROWSUP ? 0 : -ENOMEM;
+-
+- return expand_upwards(vma, address + PAGE_SIZE);
+- }
+- return 0;
+-}
+-
+-/*
+ * We enter with non-exclusive mmap_sem (to exclude vma changes,
+ * but allow concurrent faults), and pte mapped but not yet locked.
+ * We return with mmap_sem still held, but pte unmapped and unlocked.
+@@ -2641,10 +2607,6 @@ static int do_anonymous_page(struct mm_s
+ if (vma->vm_flags & VM_SHARED)
+ return VM_FAULT_SIGBUS;
+
+- /* Check if we need to add a guard page to the stack */
+- if (check_stack_guard_page(vma, address) < 0)
+- return VM_FAULT_SIGSEGV;
+-
+ /* Use the zero-page for reads */
+ if (!(flags & FAULT_FLAG_WRITE)) {
+ entry = pte_mkspecial(pfn_pte(my_zero_pfn(address),
+--- a/mm/mmap.c
++++ b/mm/mmap.c
+@@ -266,6 +266,7 @@ SYSCALL_DEFINE1(brk, unsigned long, brk)
+ unsigned long rlim, retval;
+ unsigned long newbrk, oldbrk;
+ struct mm_struct *mm = current->mm;
++ struct vm_area_struct *next;
+ unsigned long min_brk;
+ bool populate;
+
+@@ -311,7 +312,8 @@ SYSCALL_DEFINE1(brk, unsigned long, brk)
+ }
+
+ /* Check against existing mmap mappings. */
+- if (find_vma_intersection(mm, oldbrk, newbrk+PAGE_SIZE))
++ next = find_vma(mm, oldbrk);
++ if (next && newbrk + PAGE_SIZE > vm_start_gap(next))
+ goto out;
+
+ /* Ok, looks good - let it rip. */
+@@ -334,10 +336,22 @@ out:
+
+ static long vma_compute_subtree_gap(struct vm_area_struct *vma)
+ {
+- unsigned long max, subtree_gap;
+- max = vma->vm_start;
+- if (vma->vm_prev)
+- max -= vma->vm_prev->vm_end;
++ unsigned long max, prev_end, subtree_gap;
++
++ /*
++ * Note: in the rare case of a VM_GROWSDOWN above a VM_GROWSUP, we
++ * allow two stack_guard_gaps between them here, and when choosing
++ * an unmapped area; whereas when expanding we only require one.
++ * That's a little inconsistent, but keeps the code here simpler.
++ */
++ max = vm_start_gap(vma);
++ if (vma->vm_prev) {
++ prev_end = vm_end_gap(vma->vm_prev);
++ if (max > prev_end)
++ max -= prev_end;
++ else
++ max = 0;
++ }
+ if (vma->vm_rb.rb_left) {
+ subtree_gap = rb_entry(vma->vm_rb.rb_left,
+ struct vm_area_struct, vm_rb)->rb_subtree_gap;
+@@ -426,7 +440,7 @@ static void validate_mm(struct mm_struct
+ anon_vma_unlock_read(anon_vma);
+ }
+
+- highest_address = vma->vm_end;
++ highest_address = vm_end_gap(vma);
+ vma = vma->vm_next;
+ i++;
+ }
+@@ -594,7 +608,7 @@ void __vma_link_rb(struct mm_struct *mm,
+ if (vma->vm_next)
+ vma_gap_update(vma->vm_next);
+ else
+- mm->highest_vm_end = vma->vm_end;
++ mm->highest_vm_end = vm_end_gap(vma);
+
+ /*
+ * vma->vm_prev wasn't known when we followed the rbtree to find the
+@@ -846,7 +860,7 @@ again: remove_next = 1 + (end > next->
+ vma_gap_update(vma);
+ if (end_changed) {
+ if (!next)
+- mm->highest_vm_end = end;
++ mm->highest_vm_end = vm_end_gap(vma);
+ else if (!adjust_next)
+ vma_gap_update(next);
+ }
+@@ -889,7 +903,7 @@ again: remove_next = 1 + (end > next->
+ else if (next)
+ vma_gap_update(next);
+ else
+- mm->highest_vm_end = end;
++ VM_WARN_ON(mm->highest_vm_end != vm_end_gap(vma));
+ }
+ if (insert && file)
+ uprobe_mmap(insert);
+@@ -1720,7 +1734,7 @@ unsigned long unmapped_area(struct vm_un
+
+ while (true) {
+ /* Visit left subtree if it looks promising */
+- gap_end = vma->vm_start;
++ gap_end = vm_start_gap(vma);
+ if (gap_end >= low_limit && vma->vm_rb.rb_left) {
+ struct vm_area_struct *left =
+ rb_entry(vma->vm_rb.rb_left,
+@@ -1731,7 +1745,7 @@ unsigned long unmapped_area(struct vm_un
+ }
+ }
+
+- gap_start = vma->vm_prev ? vma->vm_prev->vm_end : 0;
++ gap_start = vma->vm_prev ? vm_end_gap(vma->vm_prev) : 0;
+ check_current:
+ /* Check if current node has a suitable gap */
+ if (gap_start > high_limit)
+@@ -1758,8 +1772,8 @@ check_current:
+ vma = rb_entry(rb_parent(prev),
+ struct vm_area_struct, vm_rb);
+ if (prev == vma->vm_rb.rb_left) {
+- gap_start = vma->vm_prev->vm_end;
+- gap_end = vma->vm_start;
++ gap_start = vm_end_gap(vma->vm_prev);
++ gap_end = vm_start_gap(vma);
+ goto check_current;
+ }
+ }
+@@ -1823,7 +1837,7 @@ unsigned long unmapped_area_topdown(stru
+
+ while (true) {
+ /* Visit right subtree if it looks promising */
+- gap_start = vma->vm_prev ? vma->vm_prev->vm_end : 0;
++ gap_start = vma->vm_prev ? vm_end_gap(vma->vm_prev) : 0;
+ if (gap_start <= high_limit && vma->vm_rb.rb_right) {
+ struct vm_area_struct *right =
+ rb_entry(vma->vm_rb.rb_right,
+@@ -1836,7 +1850,7 @@ unsigned long unmapped_area_topdown(stru
+
+ check_current:
+ /* Check if current node has a suitable gap */
+- gap_end = vma->vm_start;
++ gap_end = vm_start_gap(vma);
+ if (gap_end < low_limit)
+ return -ENOMEM;
+ if (gap_start <= high_limit && gap_end - gap_start >= length)
+@@ -1862,7 +1876,7 @@ check_current:
+ struct vm_area_struct, vm_rb);
+ if (prev == vma->vm_rb.rb_right) {
+ gap_start = vma->vm_prev ?
+- vma->vm_prev->vm_end : 0;
++ vm_end_gap(vma->vm_prev) : 0;
+ goto check_current;
+ }
+ }
+@@ -1900,7 +1914,7 @@ arch_get_unmapped_area(struct file *filp
+ unsigned long len, unsigned long pgoff, unsigned long flags)
+ {
+ struct mm_struct *mm = current->mm;
+- struct vm_area_struct *vma;
++ struct vm_area_struct *vma, *prev;
+ struct vm_unmapped_area_info info;
+
+ if (len > TASK_SIZE - mmap_min_addr)
+@@ -1911,9 +1925,10 @@ arch_get_unmapped_area(struct file *filp
+
+ if (addr) {
+ addr = PAGE_ALIGN(addr);
+- vma = find_vma(mm, addr);
++ vma = find_vma_prev(mm, addr, &prev);
+ if (TASK_SIZE - len >= addr && addr >= mmap_min_addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)) &&
++ (!prev || addr >= vm_end_gap(prev)))
+ return addr;
+ }
+
+@@ -1936,7 +1951,7 @@ arch_get_unmapped_area_topdown(struct fi
+ const unsigned long len, const unsigned long pgoff,
+ const unsigned long flags)
+ {
+- struct vm_area_struct *vma;
++ struct vm_area_struct *vma, *prev;
+ struct mm_struct *mm = current->mm;
+ unsigned long addr = addr0;
+ struct vm_unmapped_area_info info;
+@@ -1951,9 +1966,10 @@ arch_get_unmapped_area_topdown(struct fi
+ /* requesting a specific address */
+ if (addr) {
+ addr = PAGE_ALIGN(addr);
+- vma = find_vma(mm, addr);
++ vma = find_vma_prev(mm, addr, &prev);
+ if (TASK_SIZE - len >= addr && addr >= mmap_min_addr &&
+- (!vma || addr + len <= vma->vm_start))
++ (!vma || addr + len <= vm_start_gap(vma)) &&
++ (!prev || addr >= vm_end_gap(prev)))
+ return addr;
+ }
+
+@@ -2079,21 +2095,19 @@ find_vma_prev(struct mm_struct *mm, unsi
+ * update accounting. This is shared with both the
+ * grow-up and grow-down cases.
+ */
+-static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, unsigned long grow)
++static int acct_stack_growth(struct vm_area_struct *vma,
++ unsigned long size, unsigned long grow)
+ {
+ struct mm_struct *mm = vma->vm_mm;
+ struct rlimit *rlim = current->signal->rlim;
+- unsigned long new_start, actual_size;
++ unsigned long new_start;
+
+ /* address space limit tests */
+ if (!may_expand_vm(mm, grow))
+ return -ENOMEM;
+
+ /* Stack limit test */
+- actual_size = size;
+- if (size && (vma->vm_flags & (VM_GROWSUP | VM_GROWSDOWN)))
+- actual_size -= PAGE_SIZE;
+- if (actual_size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur))
++ if (size > ACCESS_ONCE(rlim[RLIMIT_STACK].rlim_cur))
+ return -ENOMEM;
+
+ /* mlock limit tests */
+@@ -2134,17 +2148,30 @@ static int acct_stack_growth(struct vm_a
+ */
+ int expand_upwards(struct vm_area_struct *vma, unsigned long address)
+ {
++ struct vm_area_struct *next;
++ unsigned long gap_addr;
+ int error = 0;
+
+ if (!(vma->vm_flags & VM_GROWSUP))
+ return -EFAULT;
+
+ /* Guard against wrapping around to address 0. */
+- if (address < PAGE_ALIGN(address+4))
+- address = PAGE_ALIGN(address+4);
+- else
++ address &= PAGE_MASK;
++ address += PAGE_SIZE;
++ if (!address)
+ return -ENOMEM;
+
++ /* Enforce stack_guard_gap */
++ gap_addr = address + stack_guard_gap;
++ if (gap_addr < address)
++ return -ENOMEM;
++ next = vma->vm_next;
++ if (next && next->vm_start < gap_addr) {
++ if (!(next->vm_flags & VM_GROWSUP))
++ return -ENOMEM;
++ /* Check that both stack segments have the same anon_vma? */
++ }
++
+ /* We must make sure the anon_vma is allocated. */
+ if (unlikely(anon_vma_prepare(vma)))
+ return -ENOMEM;
+@@ -2185,7 +2212,7 @@ int expand_upwards(struct vm_area_struct
+ if (vma->vm_next)
+ vma_gap_update(vma->vm_next);
+ else
+- vma->vm_mm->highest_vm_end = address;
++ vma->vm_mm->highest_vm_end = vm_end_gap(vma);
+ spin_unlock(&vma->vm_mm->page_table_lock);
+
+ perf_event_mmap(vma);
+@@ -2205,6 +2232,8 @@ int expand_upwards(struct vm_area_struct
+ int expand_downwards(struct vm_area_struct *vma,
+ unsigned long address)
+ {
++ struct vm_area_struct *prev;
++ unsigned long gap_addr;
+ int error;
+
+ address &= PAGE_MASK;
+@@ -2212,6 +2241,17 @@ int expand_downwards(struct vm_area_stru
+ if (error)
+ return error;
+
++ /* Enforce stack_guard_gap */
++ gap_addr = address - stack_guard_gap;
++ if (gap_addr > address)
++ return -ENOMEM;
++ prev = vma->vm_prev;
++ if (prev && prev->vm_end > gap_addr) {
++ if (!(prev->vm_flags & VM_GROWSDOWN))
++ return -ENOMEM;
++ /* Check that both stack segments have the same anon_vma? */
++ }
++
+ /* We must make sure the anon_vma is allocated. */
+ if (unlikely(anon_vma_prepare(vma)))
+ return -ENOMEM;
+@@ -2263,28 +2303,25 @@ int expand_downwards(struct vm_area_stru
+ return error;
+ }
+
+-/*
+- * Note how expand_stack() refuses to expand the stack all the way to
+- * abut the next virtual mapping, *unless* that mapping itself is also
+- * a stack mapping. We want to leave room for a guard page, after all
+- * (the guard page itself is not added here, that is done by the
+- * actual page faulting logic)
+- *
+- * This matches the behavior of the guard page logic (see mm/memory.c:
+- * check_stack_guard_page()), which only allows the guard page to be
+- * removed under these circumstances.
+- */
++/* enforced gap between the expanding stack and other mappings. */
++unsigned long stack_guard_gap = 256UL<<PAGE_SHIFT;
++
++static int __init cmdline_parse_stack_guard_gap(char *p)
++{
++ unsigned long val;
++ char *endptr;
++
++ val = simple_strtoul(p, &endptr, 10);
++ if (!*endptr)
++ stack_guard_gap = val << PAGE_SHIFT;
++
++ return 0;
++}
++__setup("stack_guard_gap=", cmdline_parse_stack_guard_gap);
++
+ #ifdef CONFIG_STACK_GROWSUP
+ int expand_stack(struct vm_area_struct *vma, unsigned long address)
+ {
+- struct vm_area_struct *next;
+-
+- address &= PAGE_MASK;
+- next = vma->vm_next;
+- if (next && next->vm_start == address + PAGE_SIZE) {
+- if (!(next->vm_flags & VM_GROWSUP))
+- return -ENOMEM;
+- }
+ return expand_upwards(vma, address);
+ }
+
+@@ -2306,14 +2343,6 @@ find_extend_vma(struct mm_struct *mm, un
+ #else
+ int expand_stack(struct vm_area_struct *vma, unsigned long address)
+ {
+- struct vm_area_struct *prev;
+-
+- address &= PAGE_MASK;
+- prev = vma->vm_prev;
+- if (prev && prev->vm_end == address) {
+- if (!(prev->vm_flags & VM_GROWSDOWN))
+- return -ENOMEM;
+- }
+ return expand_downwards(vma, address);
+ }
+
+@@ -2409,7 +2438,7 @@ detach_vmas_to_be_unmapped(struct mm_str
+ vma->vm_prev = prev;
+ vma_gap_update(vma);
+ } else
+- mm->highest_vm_end = prev ? prev->vm_end : 0;
++ mm->highest_vm_end = prev ? vm_end_gap(prev) : 0;
+ tail_vma->vm_next = NULL;
+
+ /* Kill the cache */
diff --git a/debian/patches/bugfix/all/mm-proc-cap-the-stack-gap-for-unpopulated-growing-vm.patch b/debian/patches/bugfix/all/mm-proc-cap-the-stack-gap-for-unpopulated-growing-vm.patch
deleted file mode 100644
index 1d187b1..0000000
--- a/debian/patches/bugfix/all/mm-proc-cap-the-stack-gap-for-unpopulated-growing-vm.patch
+++ /dev/null
@@ -1,45 +0,0 @@
-From: Michal Hocko <mhocko at suse.com>
-Date: Wed, 14 Jun 2017 08:17:15 +0200
-Subject: mm, proc: cap the stack gap for unpopulated growing vmas
-Bug-Debian-Security: https://security-tracker.debian.org/tracker/CVE-2017-1000364
-
-Oleg has noticed that show_map_vma has been overly eager to cut the
-the vma range for growing VMAs. This wasn't a big deal with 4kB stack
-gap but now that the gap is much larger we can simply get a bogus VMA
-range in show_map_vma.
-To quote Oleg
-: On ppc PAGE_SIZE == 64K, so stack_guard_gap == 16M, the application does
-: mmap(..., length=4M, ... MAP_GROWSDOWN) and /proc/pid/maps happily reports
-:
-: 30001000000-30000400000 rw-p 00000000 00:00 0
-
-Let's cap the reported range and show an empty range for this peculiar
-case which is what we have been doing for a long time. Note that the
-range will expand as soon as the first page fault happens on this range.
-
-Reported-by: Jan Stancek <jstancek at redhat.com>
-Signed-off-by: Michal Hocko <mhocko at suse.com>
----
- fs/proc/task_mmu.c | 4 ++--
- 1 file changed, 2 insertions(+), 2 deletions(-)
-
-diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
-index 5ef3e14d235f..6f38f96d725f 100644
---- a/fs/proc/task_mmu.c
-+++ b/fs/proc/task_mmu.c
-@@ -303,10 +303,10 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
- end = vma->vm_end;
- if (vma->vm_flags & VM_GROWSDOWN) {
- if (stack_guard_area(vma, start))
-- start += stack_guard_gap;
-+ start = min(end, start + stack_guard_gap);
- } else if (vma->vm_flags & VM_GROWSUP) {
- if (stack_guard_area(vma, end))
-- end -= stack_guard_gap;
-+ end = max(start, end - stack_guard_gap);
- }
-
- seq_setwidth(m, 25 + sizeof(void *) * 6 - 1);
---
-2.11.0
-
diff --git a/debian/patches/series b/debian/patches/series
index 7733634..5e79b32 100644
--- a/debian/patches/series
+++ b/debian/patches/series
@@ -710,11 +710,8 @@ bugfix/all/ipv6-fix-leak-in-ipv6_gso_segment.patch
bugfix/all/sctp-do-not-inherit-ipv6_-mc-ac-fl-_list-from-parent.patch
bugfix/all/ipv6-dccp-do-not-inherit-ipv6_mc_list-from-parent.patch
bugfix/all/ipv6-fix-out-of-bound-writes-in-__ip6_append_data.patch
-bugfix/all/mm-enlarge-stack-guard-gap.patch
-bugfix/all/mm-allow-to-configure-stack-gap-size.patch
-bugfix/all/mm-proc-cap-the-stack-gap-for-unpopulated-growing-vm.patch
-bugfix/all/mm-do-not-collapse-stack-gap-into-THP.patch
-bugfix/all/fold-me-mm-allow-to-configure-stack-gap-size.patch
+bugfix/all/mm-larger-stack-guard-gap-between-vmas.patch
+bugfix/all/mm-fix-new-crash-in-unmapped_area_topdown.patch
# Fix ABI changes
debian/of-fix-abi-changes.patch
--
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/kernel/linux.git
More information about the Kernel-svn-changes
mailing list