On Wed, Mar 26, 2008 at 07:29:33PM -0400, Mike Frysinger wrote:
Signed-off-by: Mike Frysinger [EMAIL PROTECTED]
---
linux/errnoent.h | 18 ++
1 files changed, 10 insertions(+), 8 deletions(-)
No objections arrived so far, tested and applied.
--
ldv
pgpErvyvD5gn9.pgp
On Sat, Apr 19, 2008 at 04:28:54PM -0400, Mike Frysinger wrote:
On Saturday 19 April 2008, Dmitry V. Levin wrote:
On Wed, Mar 26, 2008 at 07:20:11PM -0400, Mike Frysinger wrote:
[...]
double hrm. this was svn acting up on me again. sometimes it fails to
detect that a file actually has
Hi,
On Fri, Jun 27, 2008 at 11:01:19AM +0200, Jan Kratochvil wrote:
there was committed a typo and strace now displays funny output on ia64:
$ ./strace echo x
unlink(/bin/echo) = 1
unlink(0x6001c578) = 1
times(NULL) = -1
On Mon, Jun 30, 2008 at 03:25:58PM +0200, Jan Kratochvil wrote:
[...]
this testcase hangs on ia64:
$ echo 'int main () { return vfork (); }' vfork.c; gcc -o vfork vfork.c;
strace -f ./vfork
execve(./vfork, [./vfork], [/* 41 vars */]) = 1
...
clone(Process 2929 attached (waiting for
should be ignored, otherwise
any number of specified -F options should be treated as one -f option.
OK to commit?
--
ldv
2008-07-24 Dmitry V. Levin [EMAIL PROTECTED]
* strace.c (main): Fix -F option backwards compatibility.
--- strace/strace.c
+++ strace/strace.c
@@ -622,6 +622,7
2008-04-19 Dmitry V. Levin [EMAIL PROTECTED]
* syscall.c (is_restart_error): New function.
* defs.h (is_restart_error): Declare it.
* linux/dummy.h (sys_nanosleep): Uncouple from sys_adjtime.
* time.c (sys_nanosleep): New function, based on is_restart_error
2008-04-19 Dmitry V. Levin [EMAIL PROTECTED]
* desc.c (sys_fcntl): Do not initialize auxstr for failed syscall.
* process.c (sys_fork, sys_rfork) [USE_PROCFS]: Likewise.
* signal.c (sys_signal): Likewise.
* stream.c (internal_stream_ioctl): Likewise
2008-04-19 Dmitry V. Levin [EMAIL PROTECTED]
* desc.c (decode_select): Decode timeout for interrupted syscall.
* syscall.c (trace_syscall): Fix ERESTART* decode output.
---
strace/desc.c|7 +++
strace/syscall.c |8
2 files changed, 11 insertions(+), 4
2008-04-19 Dmitry V. Levin [EMAIL PROTECTED]
* stream.c (decode_poll): Decode timeout for interrupted syscall.
---
strace/stream.c |7 +++
1 files changed, 7 insertions(+), 0 deletions(-)
--- a/strace/stream.c
+++ b/strace/stream.c
@@ -353,6 +353,13 @@ decode_poll(struct tcb
2008-04-19 Dmitry V. Levin [EMAIL PROTECTED]
* linux/syscallent.h: Use sys_prctl() decoder for prctl syscall.
* linux/alpha/syscallent.h: Likewise.
* linux/arm/syscallent.h: Likewise.
* linux/m68k/syscallent.h: Likewise.
* linux/powerpc/syscallent.h
2008-04-19 Dmitry V. Levin [EMAIL PROTECTED]
* process.c (prctl_options): Add more constants.
---
strace/process.c |6 ++
1 files changed, 6 insertions(+), 0 deletions(-)
--- a/strace/process.c
+++ b/strace/process.c
@@ -204,6 +204,12 @@ static const struct xlat prctl_options
2008-04-19 Dmitry V. Levin [EMAIL PROTECTED]
* process.c (sys_prctl): Decode PR_SET_PDEATHSIG, PR_GET_PDEATHSIG,
PR_SET_DUMPABLE, PR_GET_DUMPABLE, PR_SET_KEEPCAPS, PR_GET_KEEPCAPS.
Fix PR_GET_UNALIGN decoder.
---
strace/process.c | 50
On Wed, Sep 24, 2008 at 06:58:05PM +0200, Denys Vlasenko wrote:
This patch is intended to implement this RFE:
https://bugzilla.redhat.com/show_bug.cgi?id=105371
There are some patches for this, but apparently discussion
on mailing list died out with no results. At least that bug
is still
On Thu, Oct 23, 2008 at 05:50:36AM -0400, Mike Frysinger wrote:
i'm attaching the diff compressed since it wouldnt make it through the mail
admin before
-mike
2008-10-23 Mike Frysinger [EMAIL PROTECTED]
* configure.ac: Add bfin to supported architectures.
* process.c: Skip
2008-10-23 Dmitry V. Levin [EMAIL PROTECTED]
Implement parsers for new linux syscalls.
* desc.c (do_dup2, sys_dup3): New functions.
(sys_dup2): Use do_dup2.
(sys_epoll_create1): New function.
(do_eventfd, sys_eventfd2): New functions.
(sys_eventfd
Hi,
On Sun, Oct 26, 2008 at 04:36:15PM +0100, Rainer M. Canavan wrote:
I'm trying to get a current release of strace to compile and run
on IRIX 6.5. Aside from a number of new and obsolete syscalls that
need to be ifdefed in system.c, there appears to be a problem with
the decoding of
On Thu, Dec 11, 2008 at 06:01:29PM +0530, Kedar Sovani wrote:
strace builds on the ARM platform fail looking for fix-ups in the file
linux/arm/syscallent.h. This patch moves system calls around to match
the asm/unistd.h contents.
Please review patches with subject linux/arm/syscallent.h
On Thu, Jan 01, 2009 at 05:30:39PM +0100, Andreas Schwab wrote:
The third argument of accept is a pointer to int, not long. This also
fixes the decoding of getsockname and getpeername, which is implemented
in terms of this function.
Applied, thanks.
--
ldv
pgpu5DTHic4Tc.pgp
Description:
On Fri, Oct 24, 2008 at 03:36:07AM +0400, Dmitry V. Levin wrote:
2008-10-23 Dmitry V. Levin l...@altlinux.org
Implement parsers for new linux syscalls.
* desc.c (do_dup2, sys_dup3): New functions.
(sys_dup2): Use do_dup2.
(sys_epoll_create1): New function
On Fri, Feb 13, 2009 at 11:37:07AM +0100, Denys Vlasenko wrote:
[...]
Please review.
Sorry for the long delay.
The change looks good indeed. See my comments below, though.
+ if (do_ptrace(PTRACE_SET_SYSCALL, tcp, NULL, new, SET_SYSCALL) != 0)
The SET_SYSCALL is deducable from
On Tue, Feb 24, 2009 at 02:08:47PM +0100, Denys Vlasenko wrote:
On Tue, 2009-02-24 at 03:04 +0300, Dmitry V. Levin wrote:
On Fri, Feb 13, 2009 at 11:37:07AM +0100, Denys Vlasenko wrote:
[...]
Please review.
Sorry for the long delay.
The change looks good indeed. See my comments
On Thu, Mar 05, 2009 at 10:02:53PM -0500, Mike Frysinger wrote:
configure.ac |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/configure.ac b/configure.ac
index ffbfeb4..e0b341a 100644
--- a/configure.ac
+++ b/configure.ac
@@ -173,6 +173,7 @@ AC_PROG_CPP
On Tue, Mar 24, 2009 at 07:07:33PM -0700, Roland McGrath wrote:
Does that bug really still exist? In my build, fopen_for_output is fopen64.
According to changelog entry, it is fixed long time ago:
2007-07-05 Roland McGrath rol...@redhat.com
* strace.c (strace_fopen):
Hi,
On Fri, May 29, 2009 at 07:58:47PM -0400, Mike Frysinger wrote:
Looking at the arches that provide __NR_cacheflush and the ones that
provide asm/cachectl.h shows that ARM falls only into the first category.
I'd prefer regular check with AC_CHECK_HEADERS instead of explicit #ifdef
check for
On Sun, May 31, 2009 at 11:45:32PM -0400, Mike Frysinger wrote:
Convert all of the #ifdef checks for specific headers to AC_CHECK_HEADERS
to avoid nested arch checks. Also reformat AC_CHECK_HEADERS in the
configure.ac file in the process to keep it sorted and easily updated and
prevent
On Tue, Jun 02, 2009 at 06:10:28PM -0700, Roland McGrath wrote:
How would folks feel about switching from CVS to GIT?
(All the other kids are doing it.)
While we consider this, please nobody do any CVS commits or git pushes.
If Dmitry and Denys are happy with git, then I'll declare the deed
On Wed, Jun 03, 2009 at 10:46:46AM -0700, Roland McGrath wrote:
Well, I switched my strace repository from cvs to git about 3 years ago,
so I definitely happy with this transition.
Would it be easier for you if we used your conversion rather than the one I
did?
This is not important for
On Tue, Jul 07, 2009 at 01:24:12PM -0700, Roland McGrath wrote:
Sorry for the long period of neglect. I'd like to get 4.5.19 out soon now.
AFAIK the current state of the git tree is more or less fine as it is.
Is there anything either urgent or trivial to fix/merge in floating around?
On Tue, Jul 07, 2009 at 01:16:39PM -0700, Roland McGrath wrote:
I've disabled the sourceforge cvs repository. Git is it. The same
approval conventions we've been using should now apply to git.
Is there any git post-receive hooks enabled?
--
ldv
pgpQoW68BVqnS.pgp
Description: PGP
On Tue, Jul 07, 2009 at 01:24:12PM -0700, Roland McGrath wrote:
Sorry for the long period of neglect. I'd like to get 4.5.19 out soon now.
AFAIK the current state of the git tree is more or less fine as it is.
Commit v4.5.18-96-geb9e2e8 which reverts a lot of unapproved changes
introduces
On Tue, Jul 07, 2009 at 03:01:52PM -0700, Roland McGrath wrote:
trivial: .cvsignore is obsolete, README-CVS needs to be renamed and updated.
Please go ahead and do it in git.
I pushed these changes to branch ldv/master, please have a look whether
it's OK for master.
BTW, we probably have to
On Tue, Jul 07, 2009 at 03:06:35PM -0700, Roland McGrath wrote:
Is there any git post-receive hooks enabled?
I don't think so. I'm not very clear on what options/constraints there are
for scripts on the sourceforge servers. You can fiddle the server repo
manually including hooks/ or
Standard git pre-commit hook uncovered a lot of spacing issues.
I suggest to fix spacing and install an update hook to prevent
new commits introducing new trailing whitespaces.
http://strace.git.sourceforge.net/git/gitweb.cgi?p=strace;a=commitdiff;h=ldv/master
--
ldv
pgpDa0MITOmib.pgp
On Tue, Jul 07, 2009 at 03:11:56PM -0700, Roland McGrath wrote:
Commit v4.5.18-96-geb9e2e8 which reverts a lot of unapproved changes
introduces regressions. For example, in system without /proc mounted,
command strace -f -o/dev/null autoreconf -fisv executed in strace
directory produces
On Tue, Jul 07, 2009 at 06:30:42PM -0700, Roland McGrath wrote:
[...]
BTW, we probably have to accept some guideline about maintaining ChangeLog
file in git.
I don't have any particular preferences or feelings about this for strace.
Whatever people here think is best will be OK with me as
On Tue, Aug 11, 2009 at 04:12:39PM +0200, Andreas Schwab wrote:
This patch fixes the decoding of futex to include the recent additions
to its interface.
Andreas.
Add more futex decoding.
* process.c (FUTEX_WAIT_REQUEUE_PI, FUTEX_CMP_REQUEUE_PI)
(FUTEX_PRIVATE_FLAG,
Hi,
On Wed, Aug 12, 2009 at 11:11:53PM +0200, Adrien Kunysz wrote:
As everybody and their dog tend to write Perl scripts to reimplement
strace -c from regular strace output, I think it could be interesting to
make strace able to print both regular output and statistics at the end.
Attached
On Fri, Aug 14, 2009 at 12:06:51AM +0200, Adrien Krunch Kunysz wrote:
On Fri, Aug 14, 2009 at 12:56:12AM +0400, Dmitry V. Levin wrote:
Your patch introduces new variable (Cflag) which behaves similar to
already existing variable (cflag). More simple approach would be to reuse
the cflag
On Fri, Aug 14, 2009 at 10:11:31AM -0600, Orion Poplawski wrote:
Seeing this when tracing a proprietary java application on Fedora 11
during application shutdown:
strace: malloc.c:3552: mremap_chunk: Assertion `((size + offset)
(mp_.pagesize-1)) == 0' failed.
Version-Release number of
On Fri, Aug 14, 2009 at 12:34:06PM +0200, Paolo Bonzini wrote:
This second patch in the series adds support for accept4 as a socketcall
sub-call. Besides the need to renumber all system calls, this poses
no problem.
OK for me.
--
ldv
pgpQIBR4FL5F8.pgp
Description: PGP signature
On Mon, Aug 17, 2009 at 03:08:07PM +0200, Paolo Bonzini wrote:
On 08/17/2009 03:06 PM, Dmitry V. Levin wrote:
On Fri, Aug 14, 2009 at 12:34:05PM +0200, Paolo Bonzini wrote:
This patch changes the paccept syscall to accept4 for x86-64, since
the former was dropped in Linux kernel commit
Hi,
On Fri, Aug 14, 2009 at 12:34:05PM +0200, Paolo Bonzini wrote:
This patch changes the paccept syscall to accept4 for x86-64, since
the former was dropped in Linux kernel commit 2d4c8266.
At the same time, it adds support for pretty-printing its arguments.
The pretty printing part of the
On Mon, Aug 17, 2009 at 05:33:25PM +0400, Dmitry V. Levin wrote:
On Mon, Aug 17, 2009 at 03:08:07PM +0200, Paolo Bonzini wrote:
On 08/17/2009 03:06 PM, Dmitry V. Levin wrote:
On Fri, Aug 14, 2009 at 12:34:05PM +0200, Paolo Bonzini wrote:
This patch changes the paccept syscall to accept4
On Mon, Sep 07, 2009 at 11:54:31AM -0700, Roland McGrath wrote:
It's again been weeks since I said let's release it now and nobody has
followed up on the release work. Let's do it already!
Two months ago I said that I'd like to see reintroduced one change (which
was lost with big revert
On Sun, Sep 13, 2009 at 10:53:12PM -0400, Mike Frysinger wrote:
Latest Linux 2.6.31 defines errno 132 as ERFKILL.
This and other two changes are OK, but commit log requirements
(described in README-hacking) are not fully satisfied.
Mike, could you resubmit these changes in ready-to-apply form,
On Tue, Sep 15, 2009 at 12:16:50AM +0400, Dmitry V. Levin wrote:
On Mon, Sep 07, 2009 at 06:25:13PM -0700, Roland McGrath wrote:
Two months ago I said that I'd like to see reintroduced one change (which
was lost with big revert commit) made by Denys: Replace many more bare
ptrace calls
On Wed, Sep 09, 2009 at 02:13:19PM -0400, Carlos O'Donell wrote:
Architectures using function pointers, like hppa, may need to
manipulate the function pointer to compute the result of a comparison.
However, the SA_HANDLER function pointer exists only in the address
space of the traced process,
On Fri, Sep 18, 2009 at 03:48:49PM +0400, Dmitry V. Levin wrote:
On Fri, Sep 04, 2009 at 11:57:21AM +1000, Steve Bennett wrote:
strace -f on an ARM OABI platform results in the child
processing dying with SIGSEGV.
The attached patch fixes this problem by taking into account
Hi,
On Tue, Sep 29, 2009 at 04:10:48AM +0400, Dmitry V. Levin wrote:
On Tue, Sep 22, 2009 at 01:04:45AM +0200, Frederik Schüler wrote:
[...]
I attached my current diff, you might want to crosscheck, waiting for
debian
hppa porters to confirm the issues are fixed.
Is there any
Hi,
On Fri, Oct 02, 2009 at 11:42:46AM +0200, Edgar E. Iglesias wrote:
I hope this can make it in before 4.5.19, it's critical for the CRIS port.
Applied, thank you for the patch.
--
ldv
pgp1VMoFf0Q4A.pgp
Description: PGP signature
On Mon, Oct 05, 2009 at 11:10:09PM +0200, Frederik Schüler wrote:
[...]
hppa issues fixed and tested, patch attached.
[...]
--- a/syscall.c
+++ b/syscall.c
@@ -2119,12 +2119,19 @@ syscall_enter(struct tcb *tcp)
{
int i;
+ long args[] = { PT_GR26-4*0,
On Tue, Oct 06, 2009 at 01:30:07AM +0400, Dmitry V. Levin wrote:
On Mon, Oct 05, 2009 at 11:10:09PM +0200, Frederik Schüler wrote:
I did the checks myself now, thanks to debian porters machines.
armel build fixes confirmed.
hppa issues fixed and tested, patch attached.
I hope
On Tue, Oct 06, 2009 at 04:27:41PM -0400, Carlos O'Donell wrote:
[...]
I have verified that upstream strace HEAD works correctly on hppa.
Good news. Looks like everything is ready for 4.5.19 release?
--
ldv
pgpaN6COQs6ry.pgp
Description: PGP signature
On Tue, Oct 06, 2009 at 06:05:32PM -0700, Roland McGrath wrote:
I committed some nits to make for a happy make distcheck (d087571). After
that, everything looks good to me.
I noticed a compilation warning on x86:
util.c: In function 'do_ptrace':
util.c:260: warning: passing argument 4 of
On Wed, Oct 07, 2009 at 01:05:10AM -0400, Mike Frysinger wrote:
* linux/ioctlsort.c: Check for __powerpc__.
On Wed, Oct 07, 2009 at 01:05:11AM -0400, Mike Frysinger wrote:
* linux/bfin/ioctlent.h: Sync list latest kernel sources.
* linux/bfin/syscallent.h: Likewise.
On Wed, Oct 07, 2009 at
Hi,
On Wed, Oct 07, 2009 at 10:25:10PM +0200, Jakub Bogusz wrote:
[...]
OK, updated patch attached.
There are two issues remained which should be addressed.
First, redundancy should be avoided.
You patch introduces 4 very similar copies of sembuf parser.
Second, user input should not be
On Wed, Oct 07, 2009 at 05:25:01AM -0400, Mike Frysinger wrote:
When getdents finishes processing, it returns 0. Strace uses this to then
try and do malloc(0), but on some systems this will always return NULL.
Since the code won't read the pointer in question if len is 0, then don't
abort on
On Wed, Oct 07, 2009 at 04:10:53AM -0400, Mike Frysinger wrote:
Systems that lack a MMU cannot use fork() to create the child process.
First we detect if the toolchain has the fork() symbol and if it does not,
we just always use vfork(). If it does, then we try that first. If it
fails due to
On Wed, Oct 07, 2009 at 07:48:50PM -0400, Mike Frysinger wrote:
On Wednesday 07 October 2009 19:32:39 Dmitry V. Levin wrote:
On Wed, Oct 07, 2009 at 05:25:01AM -0400, Mike Frysinger wrote:
When getdents finishes processing, it returns 0. Strace uses this to
then try and do malloc(0
On Wed, Oct 07, 2009 at 07:03:14PM -0700, Roland McGrath wrote:
Mike Frysinger already submitted a fix for this issue, I'm going to check
it on x86 and x86-64, and merge it if everything is OK.
Ok.
Another small issue is file timestamps in release tarball.
Since git does not store
On Wed, Oct 07, 2009 at 03:00:50PM +0200, Frederik Schüler wrote:
[...]
and on sparc:
gcc -DHAVE_CONFIG_H -I. -I.. -Ilinux/sparc -I../linux/sparc -Ilinux
-I../linux
-Wall -g -O2 -MT syscall.o -MD -MP -MF .deps/syscall.Tpo -c -o syscall.o
../syscall.c
[...]
../syscall.c: In function
On Fri, Oct 09, 2009 at 10:49:30AM -0700, Roland McGrath wrote:
Yes, there are no problem, at least for those who just build strace from
tarball. But there are people who still list tarball contents before use.
Maybe it's some kind of a taste, but for me it would be a bit more fair if
On Fri, Oct 09, 2009 at 02:48:49PM -0700, Roland McGrath wrote:
I propose to build release tarballs from clean checked out source, and
to set file timestamps right after checkout, e.g.
git clone /path/to/strace.git
cd strace
git-set-file-times
autoreconf -i
./configure
On Sat, Oct 10, 2009 at 03:01:56AM +0400, Dmitry V. Levin wrote:
On Fri, Oct 09, 2009 at 02:48:49PM -0700, Roland McGrath wrote:
I propose to build release tarballs from clean checked out source, and
to set file timestamps right after checkout, e.g.
git clone /path/to/strace.git
Hi,
On Fri, Oct 09, 2009 at 04:48:55PM -0400, Mike Frysinger wrote:
On Friday 09 October 2009 07:42:19 Frederik Schüler wrote:
thanks fot the patch, with a small fix, the 32bit strace binary on sparc
builds now [...] but the 64bit binary fails:
gcc -m64 -DHAVE_CONFIG_H -I. -I..
Hi,
On Mon, Oct 12, 2009 at 04:27:23PM +0200, Frederik Schüler wrote:
Now it builds both the 32 and 64bit strace binaries on sparc64, thanks!
Does it also work there? ;)
--
ldv
pgpZDJQmsXeQn.pgp
Description: PGP signature
On Fri, Oct 16, 2009 at 11:41:21AM +0200, Andreas Schwab wrote:
From 61b7435ca146c8a75f867ae0e1fd35d74d7f86a5 Mon Sep 17 00:00:00 2001
From: Andreas Schwab sch...@redhat.com
Date: Fri, 16 Oct 2009 11:37:13 +0200
Subject: [PATCH] Fix decoding of newfstatat syscall on x86-64
* file.c
Hi,
On Wed, Oct 14, 2009 at 04:46:24PM +0200, Bernhard Reutner-Fischer wrote:
With this patchlet i'm able to build on a uClibc setup with LFS off.
--- a/acinclude.m4
+++ b/acinclude.m4
[...]
+ AC_DEFINE([HAVE_STATFS64], 1,
[...]
--- a/file.c
+++ b/file.c
[...]
-#ifdef LINUX
+#if
On Mon, Oct 12, 2009 at 06:59:04PM -0700, Roland McGrath wrote:
I've done scratch builds for Fedora 10 and 12 (x86_64, i386, ppc64, ppc),
available here:
http://koji.fedoraproject.org/scratch/roland/task_1742936/
http://koji.fedoraproject.org/scratch/roland/task_1742937/
(The build system
On Thu, Oct 22, 2009 at 11:12:53AM -0500, Haszlakiewicz, Eric wrote:
I'm trying to get strace to produce the full command line output for a
command, but no matter what I try it always truncates the arguments to
execve.
Please read about -s option in the strace(1) manual.
Is there a way to
On Tue, Oct 27, 2009 at 04:59:24PM +0100, Andreas Schwab wrote:
[..]
* defs.h: Remove references to __STDC__ and P macros.
* strace.c: Likewise.
I'm OK for removing this junk, but please be consistent and use the same
formatting style. For example,
[...]
@@ -158,15 +158,11 @@
#include
* configure.ac (AC_CHECK_HEADERS): In check for linux/netlink.h, include
sys/socket.h instead of linux/socket.h beforehand.
---
configure.ac |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/configure.ac b/configure.ac
index c8450cd..96c1c24 100644
--- a/configure.ac
+++
On Wed, Nov 04, 2009 at 12:55:46PM +0100, Andreas Schwab wrote:
This patch refactors the printing of 64bit syscall arguments.
I like the change, but I have two questions about particular hunks of
the patch.
@@ -631,7 +625,6 @@ sys_lseek64(struct tcb *tcp)
{
if (entering(tcp)) {
On Wed, Nov 04, 2009 at 05:20:42PM +0100, Andreas Schwab wrote:
Here is an updated patch.
It looks OK for me now.
--
ldv
pgpu0vGGdmkfM.pgp
Description: PGP signature
--
Let Crystal Reports handle the reporting -
On Wed, Nov 11, 2009 at 01:59:16PM +0100, Andreas Schwab wrote:
In general, if a thread is forcefully killed strace should not panic.
OK for me.
--
ldv
pgpL3jg5CPl8Q.pgp
Description: PGP signature
--
Let Crystal
On Fri, Nov 06, 2009 at 08:48:55PM +0300, Dmitry V. Levin wrote:
On Wed, Oct 28, 2009 at 02:45:26PM +0300, Dmitry V. Levin wrote:
On Tue, Oct 27, 2009 at 04:59:24PM +0100, Andreas Schwab wrote:
[..]
* defs.h: Remove references to __STDC__ and P macros.
* strace.c: Likewise.
I'm OK
On Fri, Nov 13, 2009 at 11:46:34AM -0500, Mike Frysinger wrote:
* linux/syscallent.h: Move to ...
* linux/i386/syscallent.h: ... here.
* linux/ia64/syscallent.h: Update i386 syscallent.h include.
* linux/sparc/gen.pl: Likewise.
* linux/x86_64/syscallent1.h: Likewise.
I'll push it if there
On Wed, Dec 23, 2009 at 11:26:40AM -0500, Chris Metcalf wrote:
These are not Tilera-specific so I broke them out to a separate
path. One adds support for MREMAP_FIXED for mremap(); the other
just changes the first argument of mbind() to display in hex like
other pointers do.
Thank you, I've
On Wed, Dec 23, 2009 at 11:26:49AM -0500, Chris Metcalf wrote:
This patch adds support for the tile architecture from Tilera (see
http://www.tilera.com). We are shipping strace in our current release;
this patch reflects the version we intend to ship for our 2.1 release.
There are no need to
On Mon, Jan 11, 2010 at 05:54:00PM +0100, Andreas Schwab wrote:
[...]
* strace.c (detach): Call clearbpt when TCB_BPTSET is set.
OK for me.
I can only guess what was the rationale to SIGKILL processes in TCB_BPTSET
state when it was implemented. At least, looking at the current code
I cannot
On Thu, Dec 31, 2009 at 03:31:16PM +, Joachim Nilsson wrote:
diff --git a/configure.ac b/configure.ac
index 9afee58..cef16c7 100644
--- a/configure.ac
+++ b/configure.ac
@@ -8,6 +8,9 @@ AM_INIT_AUTOMAKE([foreign check-news
dist-bzip2 no-dist-gzip])
AM_MAINTAINER_MODE
On Tue, Jan 12, 2010 at 06:13:10PM -0800, Roland McGrath wrote:
I'd rather added
m4_ifdef([AM_SILENT_RULES], [AM_SILENT_RULES])
instead, thus keeping default behaviour unchanged.
Agreed. But I think it's fine to just require the automake version that
has the macro.
I was going to apply
On Fri, Dec 25, 2009 at 11:56:56PM +0300, Dmitry V. Levin wrote:
On Wed, Dec 23, 2009 at 11:26:49AM -0500, Chris Metcalf wrote:
This patch adds support for the tile architecture from Tilera (see
http://www.tilera.com). We are shipping strace in our current release;
this patch reflects
On Sun, Jan 24, 2010 at 05:32:24PM +, Andreas Schwab wrote:
Sorry, I don't understand the sentence.
In the AC_STAT64 test, linux/types.h is included before asm/stat.h,
and your proposal is to replace linux/types.h with sys/types.h.
I'm not sure that any additional headers have to be included
On Sun, Jan 24, 2010 at 11:08:47PM +0100, Andreas Schwab wrote:
$ echo '#include asm/stat.h' |gcc -xc -c - -o /dev/null
In file included from stdin:1:
/usr/include/asm/stat.h:31: error: expected specifier-qualifier-list before
‘ino_t’
OK, just for the record, what is the kernel version and
On Mon, Jan 25, 2010 at 11:38:09AM +0100, Andreas Schwab wrote:
Here is a better patch, pre-POSIX systems might need sys/types.h
before sys/stat.h.
I suppose the proposed change fixes a real spurious failure I'm not
aware of, but I'd like to know for sure.
I'm sorry, but I have to repeat the
On Wed, Jan 27, 2010 at 12:51:01AM +0100, Andreas Schwab wrote:
Dmitry V. Levin ldv-u2l5PoMzF/vg9huczpv...@public.gmane.org writes:
On Mon, Jan 25, 2010 at 11:38:09AM +0100, Andreas Schwab wrote:
Here is a better patch, pre-POSIX systems might need sys/types.h
before sys/stat.h.
I
Hi,
I've pushed for review to
http://strace.git.sourceforge.net/git/gitweb.cgi?p=strace/strace;a=shortlog;h=ldv/cleanup
two cleanup commits:
http://strace.git.sourceforge.net/git/gitweb.cgi?p=strace/strace;a=commitdiff;h=v4.5.19-20-g257e157
- Merge Linux internal_clone() into internal_fork()
Hi,
On Tue, Feb 09, 2010 at 11:03:27AM -0800, David Daney wrote:
syscall.c will not build under mips-n32 or mips-n64.
In syscall_enter() there is a call to ptrace() with an undefined
variable 'pid'. This should be replaced with 'tcp-pid'. Attached is a
patch that fixes the issue for me.
On Fri, Feb 12, 2010 at 09:39:12PM +0100, Andreas Schwab wrote:
[...]
+int
+sys_recvmmsg(struct tcb *tcp)
+{
+ if (entering(tcp)) {
+ tprintf(%ld, , tcp-u_arg[0]);
+ printmmsghdr(tcp, tcp-u_arg[1]);
+ tprintf(, %ld, , tcp-u_arg[2]);
+ /*
On Mon, Mar 22, 2010 at 09:43:02AM +, Mark Wielaard wrote:
Dmitry V. Levin l...@... writes:
On Sun, Mar 21, 2010 at 02:41:26PM +0100, Mark Wielaard wrote:
* file.c [LINUX] (sys_inotify_init1): New function.
* linux/syscall.h: Declare new sys_inotify_init1 handler.
* linux/bfin
Hi,
On Sun, Mar 28, 2010 at 02:39:38PM +0100, Adrien Kunysz wrote:
Back in August I posted this patch that implements a new -C option. I
got some feedback and reworked the patch accordingly but I got no follow
up after the third iteration. Can you please consider this again or tell
me what's
On Mon, Mar 29, 2010 at 01:11:19PM +0100, Adrien Kunysz wrote:
On Mon, Mar 29, 2010 at 04:55:23AM +0400, Dmitry V. Levin wrote:
On Sun, Mar 28, 2010 at 06:01:47PM +0100, Adrien Kunysz wrote:
Commit 9906e6da8be98d27bffa2baeeccd40cf1a1b11c1 Fix handling of Linux
systems without struct
On Sat, Feb 13, 2010 at 12:28:26AM +0300, Dmitry V. Levin wrote:
On Fri, Feb 12, 2010 at 09:39:12PM +0100, Andreas Schwab wrote:
[...]
+int
+sys_recvmmsg(struct tcb *tcp)
+{
+ if (entering(tcp)) {
+ tprintf(%ld, , tcp-u_arg[0]);
+ printmmsghdr(tcp, tcp-u_arg[1
On Wed, Mar 31, 2010 at 09:53:38PM +0400, Dmitry V. Levin wrote:
On Wed, Mar 31, 2010 at 12:31:22PM +1100, Anton Blanchard wrote:
The flags field was bogus if we came in via an indirect ipccall, so fix it.
Nice catch. Apparently, your patch fixes regression introduced by
my commit v4.5.18
On Mon, Mar 29, 2010 at 05:10:50AM +0400, Dmitry V. Levin wrote:
On Sun, Mar 28, 2010 at 02:39:38PM +0100, Adrien Kunysz wrote:
Back in August I posted this patch that implements a new -C option. I
got some feedback and reworked the patch accordingly but I got no follow
up after the third
* syscall.c (trace_syscall): Split into trace_syscall_exiting() and
trace_syscall_entering().
---
There is no semantic change.
syscall.c | 303 +++-
1 files changed, 157 insertions(+), 146 deletions(-)
---
strace.1 | 99 ++---
1 files changed, 55 insertions(+), 44 deletions(-)
http://strace.git.sourceforge.net/git/gitweb.cgi?p=strace/strace;a=commitdiff;h=ldv/man
--
ldv
pgpkvG2Sr11sp.pgp
Description: PGP signature
* linux/ioctlent.sh: Search in asm-generic directory as well.
* linux/ioctlent.h: Regenerated.
---
linux/ioctlent.h | 730 +++--
linux/ioctlent.sh |6 +-
2 files changed, 426 insertions(+), 310 deletions(-)
On Fri, Apr 09, 2010 at 03:24:52PM -0700, Roland McGrath wrote:
Sounds OK to me. It would be nice to resolve Andreas's PTRACE_SETOPTIONS
patches first, but not essential if it introduces delays.
Yes, it would be nice, but that change isn't quite obvious, and I have
only two days left. So, I'd
1 - 100 of 1644 matches
Mail list logo
