[Ltrace-devel] ltrace v0.7.0-git: Session on a Freetz MIPSEL router box

[Edgar E. Iglesias]

On Thu, Oct 11, 2012 at 12:55:39PM +0200, Petr Machata wrote:
> "Edgar E. Iglesias" <edgar.iglesias at gmail.com> writes:
> 
> > On Thu, Oct 11, 2012 at 04:44:15PM +0800, "Jackie Yeh [?????????]" wrote:
> >> Your patches works!
> >> I have to apply patch 1~6, only 1~4 doesn't work.
> >> 
> >> One question for the patches:
> >> For these patches I copy-and-paste, then save to a patch file, but most of them 
> >> doesn't work at all -- some characters are missing and some long line are cut. 
> >> So I actually patch them manually.
> >> Do you have some better way to apply these patches?
> 
> If your mail client allows "display as raw" or "show source", or some
> other type of option, then that might be the right way to do it.  Gmail
> inteface has "Show Original" (or some such, my interface is translated)
> that would seemingly serve this purpose, if that's what you are using.
> 
> >> Another question I'm not sure it's ltrace bug or my application's problem:
> >> <quote>
> >> _ZThn8_NK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE(0x603710, 8, 
> >> 0x603710, 0callstack_push_symfunc: Error: call nesting too deep!
> >> </quote>
> >> 
> >> Initially the MAX_CALLDEPTH is defined as 64, after I increase it to 256, the 
> >> "call nesting too deep" error keeps happened, then the application
> >> stops.
> >
> > These are likely bugs in ltrace. I still have many cases that fail that I plan
> > to work on. Hopefully I'll have more time and can post more fixes
> > continously.
> >
> > Regarding emailed patches, I usually save the files into a mail folder and use
> > the "git am" command. I plan to setup a public git tree soon, that will make it
> > easier for others to just clone or pull from it.
> 
> If you create an account on alioth, I'll happily grant you repo access.
> You could push to your own branch and I'd just merge after reviewing.

Thanks Petr,

Jackie, I've pushed the mipsel-cpic patches into the edgar/mipsel-cpic
branch. That should make it easier to grab them.

> (I'll get around to your sprintf bug and the recent patch set later
> today.)

Petr, I've put the sprintf patches into edgar/sprintf-fix. I've changed
the vect fix to overallocate 8 slots. And fixed the powerpc backend.

In the process of fixing the powerpc backend I noticed that it doesn't
build (with my toolchain). This is the fix:

-- >8 --

Subject: [PATCH] powerpc: Include sys/wait.h to fix the build

Signed-off-by: Edgar E. Iglesias <edgar at axis.com>
---
 sysdeps/linux-gnu/ppc/trace.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/sysdeps/linux-gnu/ppc/trace.c b/sysdeps/linux-gnu/ppc/trace.c
index 2bb317f..1285c35 100644
--- a/sysdeps/linux-gnu/ppc/trace.c
+++ b/sysdeps/linux-gnu/ppc/trace.c
@@ -29,6 +29,7 @@
 #include <errno.h>
 #include <signal.h>
 #include <string.h>
+#include <sys/wait.h>
 
 #include "backend.h"
 #include "breakpoint.h"
-- 
1.7.8.6