GNU bug report logs -
#8230
touch dumps core on solaris 10
Previous Next
To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 8230 in the body.
You can then email your comments to 8230 AT debbugs.gnu.org in the normal way.
Toggle the display of automated, internal messages from the tracker.
Report forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Fri, 11 Mar 2011 01:54:01 GMT)
Full text and
rfc822 format available.
Acknowledgement sent
to
Ben Walton <bwalton <at> artsci.utoronto.ca>:
New bug report received and forwarded. Copy sent to
bug-coreutils <at> gnu.org.
(Fri, 11 Mar 2011 01:54:02 GMT)
Full text and
rfc822 format available.
Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):
Hi All,
Since 8.8, touch has dumped core on solaris 10 (built on solaris 9).
The command that triggers the core dump is a simple: touch somefile
This is seemingly related to the recommend patch cluster of Jan 10,
2011 as it runs just fine on older releases of solaris 10.
On affected boxes, the file is touched but in the subsequent dup2, it
goes into a loop until it dies. I tried to attach truss output, but
the MTA at gnu.org rejected it twice. The URL's below have both sparc
and i386 traces.
http://web.chass.utoronto.ca/bwalton/gtouch.i386
http://web.chass.utoronto.ca/bwalton/gtouch.sparc
Please let me know if other inputs would be of use in debugging this.
I've taken a brief look but don't see anything immediately obvious.
I'm hoping to spend more time on it this evening.
Thanks
-Ben
--
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302
Information forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Fri, 11 Mar 2011 04:34:02 GMT)
Full text and
rfc822 format available.
Message #8 received at 8230 <at> debbugs.gnu.org (full text, mbox):
This reminds me of an earlier thread:
http://lists.gnu.org/archive/html/coreutils/2010-11/msg00102.html
I don't know that that problem ever got resolved, but
it is odd that you started seeing it only in the January 10
cluster, but that other report was in November.
What's the output of the following commands?
/sbin/uname -a
showrev -p
Information forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Fri, 11 Mar 2011 13:46:02 GMT)
Full text and
rfc822 format available.
Message #11 received at 8230 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
[adding bug-gnulib]
On 03/10/2011 06:53 PM, Ben Walton wrote:
>
> Hi All,
>
> Since 8.8, touch has dumped core on solaris 10 (built on solaris 9).
> The command that triggers the core dump is a simple: touch somefile
>
> This is seemingly related to the recommend patch cluster of Jan 10,
> 2011 as it runs just fine on older releases of solaris 10.
>
> On affected boxes, the file is touched but in the subsequent dup2, it
> goes into a loop until it dies. I tried to attach truss output, but
> the MTA at gnu.org rejected it twice.
Can you show the config.log, especially the portion where it is checking
for dup2 and fcntl POSIX compliance? This smells like a bug in gnulib's
fcntl() implementation.
>
> Please let me know if other inputs would be of use in debugging this.
If you compiled with debugging, then it might help to show a gdb session
of rpl_fcntl in lib/fcntl.c while stepping enough to prove the loop, as
well as a backtrace that leads to that loop, might be worthwhile.
--
Eric Blake eblake <at> redhat.com +1-801-349-2682
Libvirt virtualization library http://libvirt.org
[signature.asc (application/pgp-signature, attachment)]
Information forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Fri, 11 Mar 2011 22:42:02 GMT)
Full text and
rfc822 format available.
Message #14 received at 8230 <at> debbugs.gnu.org (full text, mbox):
Regarding <http://debbugs.gnu.org/cgi/bugreport.cgi?bug=8230>:
Eric Blake wrote:
> If you compiled with debugging, then it might help to show a gdb session
> of rpl_fcntl in lib/fcntl.c while stepping enough to prove the loop, as
> well as a backtrace that leads to that loop, might be worthwhile.
After compiling with debugging on Solaris 9, and running on Solaris 10,
I get this stack trace from the end of the endless recursion:
#3 0x00019e98 in futimens (fd=0, times=0x0) at futimens.c:36
#4 0xff2b6310 in futimesat () from /lib/libc.so.1
#5 0x0001fd38 in fdutimens (fd=0, file=0x0, timespec=0x0) at utimens.c:339
#6 0x00019e98 in futimens (fd=0, times=0x0) at futimens.c:36
#7 0xff2b6310 in futimesat () from /lib/libc.so.1
#8 0x0001fd38 in fdutimens (fd=0, file=0x0, timespec=0x0) at utimens.c:339
#9 0x00019e98 in futimens (fd=0, times=0x0) at futimens.c:36
#10 0xff2b6310 in futimesat () from /lib/libc.so.1
and this one from the beginning:
#0 futimens (fd=0, times=0x0) at futimens.c:36
#1 0xff2b6310 in futimesat () from /lib/libc.so.1
#2 0x0001fd38 in fdutimens (fd=0, file=0x0, timespec=0x0) at utimens.c:339
#3 0x00019e98 in futimens (fd=0, times=0x0) at futimens.c:36
#4 0x00013c6c in fdutimensat (fd=0, dir=-3041965,
file=0xffbffd4a <error reading variable>, ts=0x0, atflag=0)
at fdutimensat.c:48
#5 0x00012be8 in touch (file=0xffbffd4a <error reading variable>)
at touch.c:168
#6 0x00013b48 in main (argc=2, argv=0xffbffc5c) at touch.c:434
So the situation is this:
- On Solaris 9, futimesat() exists in libc, but futimens() doesn't.
- On Solaris 10, both exist, and futimesat() invokes futimens(), but the
latter is redefined in the executable with code from gnulib.
In gnulib, we have invocations
fdutimensat -> futimens -> fdutimens -> {futimesat,utimes}
and Solaris 10 libc has an invocation
futimens <-------------- futimesat
There is also a second cycle: gnulib has invocations
utimensat -> utimens -> futimesat
and Solaris 10 libc has an invocation
utimensat <------------ futimesat
I propose to change gnulib so that it uses
#define futimens rpl_futimens
#define utimensat rpl_utimensat
also on Solaris 9 (with the usual idioms for the C++ and the gnulib namespace).
I have verified that when these two macro definitions are added to config.h,
the 'touch' program no longer goes into an infinite recursion.
Bruno
--
In memoriam Eulogius of Córdoba <http://en.wikipedia.org/wiki/Eulogius_of_Córdoba>
Information forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Sat, 12 Mar 2011 02:06:02 GMT)
Full text and
rfc822 format available.
Message #17 received at 8230 <at> debbugs.gnu.org (full text, mbox):
On 03/11/2011 02:41 PM, Bruno Haible wrote:
> I propose to change gnulib so that it uses
> #define futimens rpl_futimens
> #define utimensat rpl_utimensat
> also on Solaris 9 (with the usual idioms for the C++ and the gnulib namespace).
That sounds good, but why make Solaris 9 a special case?
Wouldn't it be simpler to do it for all platforms where gnulib
defines futimens or utimensat functions?
Information forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Sat, 12 Mar 2011 12:13:02 GMT)
Full text and
rfc822 format available.
Message #20 received at 8230 <at> debbugs.gnu.org (full text, mbox):
Paul Eggert wrote:
> That sounds good, but why make Solaris 9 a special case?
> Wouldn't it be simpler to do it for all platforms where gnulib
> defines futimens or utimensat functions?
The functions of the two cycles are available on the following platforms:
Cycle #1:
futimens AIX 7, Cygwin 1.7, Solaris >= 10
fdutimens none
futimesat Cygwin 1.7, Solaris >= 9
Cycle #2:
utimensat AIX 7, Cygwin 1.7, Solaris >= 10
utimens none
futimesat Cygwin 1.7, Solaris >= 9
On AIX, there's no problem, because AIX does not use the ELF format, therefore
a function definition in an executable cannot change how libc behaves. And on
Cygwin, all functions appeared at the same time.
I'm applying this patch:
2011-03-12 Bruno Haible <bruno <at> clisp.org>
futimens, utimensat: Avoid endless recursion on Solaris 10.
* lib/sys_stat.in.h (futimens, utimensat): Define with rpl_ prefix on
Solaris.
Reported by Ben Walton <bwalton <at> artsci.utoronto.ca> via Eric Blake
in <http://debbugs.gnu.org/cgi/bugreport.cgi?bug=8230>.
--- lib/sys_stat.in.h.orig Sat Mar 12 12:51:25 2011
+++ lib/sys_stat.in.h Sat Mar 12 12:50:23 2011
@@ -355,7 +355,11 @@
#if @GNULIB_FUTIMENS@
-# if @REPLACE_FUTIMENS@
+/* Use the rpl_ prefix also on Solaris <= 9, because on Solaris 9 our futimens
+ implementation relies on futimesat, which on Solaris 10 makes an invocation
+ to futimens that is meant to invoke the libc's futimens(), not gnulib's
+ futimens(). */
+# if @REPLACE_FUTIMENS@ || defined __sun
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
# undef futimens
# define futimens rpl_futimens
@@ -368,7 +372,9 @@
# endif
_GL_CXXALIAS_SYS (futimens, int, (int fd, struct timespec const times[2]));
# endif
+# if @HAVE_FUTIMENS@
_GL_CXXALIASWARN (futimens);
+# endif
#elif defined GNULIB_POSIXCHECK
# undef futimens
# if HAVE_RAW_DECL_FUTIMENS
@@ -612,7 +618,11 @@
#if @GNULIB_UTIMENSAT@
-# if @REPLACE_UTIMENSAT@
+/* Use the rpl_ prefix also on Solaris <= 9, because on Solaris 9 our utimensat
+ implementation relies on futimesat, which on Solaris 10 makes an invocation
+ to utimensat that is meant to invoke the libc's utimensat(), not gnulib's
+ utimensat(). */
+# if @REPLACE_UTIMENSAT@ || defined __sun
# if !(defined __cplusplus && defined GNULIB_NAMESPACE)
# undef utimensat
# define utimensat rpl_utimensat
@@ -631,7 +641,9 @@
_GL_CXXALIAS_SYS (utimensat, int, (int fd, char const *name,
struct timespec const times[2], int flag));
# endif
+# if @HAVE_UTIMENSAT@
_GL_CXXALIASWARN (utimensat);
+# endif
#elif defined GNULIB_POSIXCHECK
# undef utimensat
# if HAVE_RAW_DECL_UTIMENSAT
--
In memoriam Zoran Djindjić <http://en.wikipedia.org/wiki/Zoran_Đinđić>
Information forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Sun, 13 Mar 2011 02:16:02 GMT)
Full text and
rfc822 format available.
Message #23 received at 8230 <at> debbugs.gnu.org (full text, mbox):
Excerpts from Bruno Haible's message of Sat Mar 12 07:11:51 -0500 2011:
Hi Bruno,
> I'm applying this patch:
Thanks for saving me the legwork on this. The patch does correct the
problem. I appreciate the quick turnaround on this.
Thanks
-Ben
--
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302
Information forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Sun, 13 Mar 2011 09:12:02 GMT)
Full text and
rfc822 format available.
Message #26 received at 8230 <at> debbugs.gnu.org (full text, mbox):
Ben Walton wrote:
> Thanks for saving me the legwork on this. The patch does correct the
> problem. I appreciate the quick turnaround on this.
Thanks to both of you.
I've just updated coreutils to use the latest from gnulib,
so this will be fixed in coreutils-8.11.
However, contrary to most NEWS-worthy bugs, I have not tried
to determine when this one was introduced. From the initial report,
I'm assuming it was introduced in coreutils-8.8, and wrote that in NEWS.
Ben, can you confirm that touch from coreutils-8.7 did not have this problem?
I'll wait for confirmation before pushing.
From d8ed00cb61fede223b1450e53b7d6c21edff4b43 Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering <at> redhat.com>
Date: Sun, 13 Mar 2011 09:56:53 +0100
Subject: [PATCH] touch: update to latest gnulib to fix Solaris 10 touch segfault
* gnulib: Update to latest, to address http://debbugs.gnu.org/8230.
When built on Solaris 9 and run on Solaris 10, touch would segfault.
Reported by Ben Walton.
* bootstrap: Resync from gnulib.
* tests/init.sh: Likewise.
* NEWS (Bug fixes): Mention this.
---
NEWS | 3 +++
THANKS.in | 1 +
bootstrap | 6 +++---
gnulib | 2 +-
tests/init.sh | 34 +++++++++++++++++-----------------
5 files changed, 25 insertions(+), 21 deletions(-)
diff --git a/NEWS b/NEWS
index 5770410..9ceaa06 100644
--- a/NEWS
+++ b/NEWS
@@ -14,6 +14,9 @@ GNU coreutils NEWS -*- outline -*-
sort no longer spawns 7 worker threads to sort 16 lines
[bug introduced in coreutils-8.6]
+ touch built on Solaris 9 would segfault when run on Solaris 10
+ [bug introduced in coreutils-8.8]
+
wc would dereference a NULL pointer upon an early out-of-memory error
[bug introduced in coreutils-7.1]
diff --git a/THANKS.in b/THANKS.in
index fe53c44..fbc4153 100644
--- a/THANKS.in
+++ b/THANKS.in
@@ -71,6 +71,7 @@ Barry Kelly http://barrkel.blogspot.com/
Bauke Jan Douma bjdouma <at> xs4all.nl
Ben Elliston bje <at> air.net.au
Ben Harris bjh21 <at> netbsd.org
+Ben Walton bwalton <at> artsci.utoronto.ca
Bengt Martensson bengt <at> mathematik.uni-Bremen.de
Benjamin Cutler cutlerbc <at> simla.colostate.edu
Bernard Giroud bernard.giroud <at> creditlyonnais.ch
diff --git a/bootstrap b/bootstrap
index e9ec11e..f004ad3 100755
--- a/bootstrap
+++ b/bootstrap
@@ -1,6 +1,6 @@
#! /bin/sh
# Print a version string.
-scriptversion=2011-01-21.16; # UTC
+scriptversion=2011-03-03.12; # UTC
# Bootstrap this package from checked-out sources.
@@ -874,7 +874,7 @@ grep -E '^[ ]*AC_CONFIG_HEADERS?\>' configure.ac >/dev/null ||
for command in \
libtool \
- "${ACLOCAL-aclocal} --force -I m4 $ACLOCAL_FLAGS" \
+ "${ACLOCAL-aclocal} --force -I '$m4_base' $ACLOCAL_FLAGS" \
"${AUTOCONF-autoconf} --force" \
"${AUTOHEADER-autoheader} --force" \
"${AUTOMAKE-automake} --add-missing --copy --force-missing"
@@ -885,7 +885,7 @@ do
command="${LIBTOOLIZE-libtoolize} -c -f"
fi
echo "$0: $command ..."
- $command || exit
+ eval "$command" || exit
done
diff --git a/gnulib b/gnulib
index 6f0680e..68d757e 160000
--- a/gnulib
+++ b/gnulib
@@ -1 +1 @@
-Subproject commit 6f0680eb29a1737d704a1df26aafc00490cd34d8
+Subproject commit 68d757e2cb228590d46961cbf3e9ec7d4460e335
diff --git a/tests/init.sh b/tests/init.sh
index 44be35b..71c6516 100644
--- a/tests/init.sh
+++ b/tests/init.sh
@@ -74,10 +74,10 @@ Exit () { set +e; (exit $1); exit $1; }
# the reason for skip/failure to console, rather than to the .log files.
: ${stderr_fileno_=2}
-warn_() { echo "$@" 1>&$stderr_fileno_; }
-fail_() { warn_ "$ME_: failed test: $@"; Exit 1; }
-skip_() { warn_ "$ME_: skipped test: $@"; Exit 77; }
-framework_failure_() { warn_ "$ME_: set-up failure: $@"; Exit 99; }
+warn_ () { echo "$@" 1>&$stderr_fileno_; }
+fail_ () { warn_ "$ME_: failed test: $@"; Exit 1; }
+skip_ () { warn_ "$ME_: skipped test: $@"; Exit 77; }
+framework_failure_ () { warn_ "$ME_: set-up failure: $@"; Exit 99; }
# Sanitize this shell to POSIX mode, if possible.
DUALCASE=1; export DUALCASE
@@ -193,7 +193,7 @@ fi
test -n "$EXEEXT" && shopt -s expand_aliases
# Enable glibc's malloc-perturbing option.
-# This is cheap and useful for exposing code that depends on the fact that
+# This is useful for exposing code that depends on the fact that
# malloc-related functions often return memory that is mostly zeroed.
# If you have the time and cycles, use valgrind to do an even better job.
: ${MALLOC_PERTURB_=87}
@@ -202,22 +202,22 @@ export MALLOC_PERTURB_
# This is a stub function that is run upon trap (upon regular exit and
# interrupt). Override it with a per-test function, e.g., to unmount
# a partition, or to undo any other global state changes.
-cleanup_() { :; }
+cleanup_ () { :; }
if ( diff --version < /dev/null 2>&1 | grep GNU ) > /dev/null 2>&1; then
- compare() { diff -u "$@"; }
+ compare () { diff -u "$@"; }
elif ( cmp --version < /dev/null 2>&1 | grep GNU ) > /dev/null 2>&1; then
- compare() { cmp -s "$@"; }
+ compare () { cmp -s "$@"; }
else
- compare() { cmp "$@"; }
+ compare () { cmp "$@"; }
fi
# An arbitrary prefix to help distinguish test directories.
-testdir_prefix_() { printf gt; }
+testdir_prefix_ () { printf gt; }
# Run the user-overridable cleanup_ function, remove the temporary
# directory and exit with the incoming value of $?.
-remove_tmp_()
+remove_tmp_ ()
{
__st=$?
cleanup_
@@ -233,7 +233,7 @@ remove_tmp_()
# contains only the specified bytes (see the case stmt below), then print
# a space-separated list of those names and return 0. Otherwise, don't
# print anything and return 1. Naming constraints apply also to DIR.
-find_exe_basenames_()
+find_exe_basenames_ ()
{
feb_dir_=$1
feb_fail_=0
@@ -266,7 +266,7 @@ find_exe_basenames_()
# PROG that simply invokes PROG.exe, then return 0. If any selected
# file name or the directory name, $1, contains an unexpected character,
# define no alias and return 1.
-create_exe_shims_()
+create_exe_shims_ ()
{
case $EXEEXT in
'') return 0 ;;
@@ -288,7 +288,7 @@ create_exe_shims_()
# Use this function to prepend to PATH an absolute name for each
# specified, possibly-$initial_cwd_-relative, directory.
-path_prepend_()
+path_prepend_ ()
{
while test $# != 0; do
path_dir_=$1
@@ -311,7 +311,7 @@ path_prepend_()
export PATH
}
-setup_()
+setup_ ()
{
if test "$VERBOSE" = yes; then
# Test whether set -x may cause the selected shell to corrupt an
@@ -364,7 +364,7 @@ setup_()
# - make only $MAX_TRIES_ attempts
# Helper function. Print $N pseudo-random bytes from a-zA-Z0-9.
-rand_bytes_()
+rand_bytes_ ()
{
n_=$1
@@ -396,7 +396,7 @@ rand_bytes_()
| LC_ALL=C tr -c $chars_ 01234567$chars_$chars_$chars_
}
-mktempd_()
+mktempd_ ()
{
case $# in
2);;
--
1.7.4.1.299.ga459d
Information forwarded
to
owner <at> debbugs.gnu.org, bug-coreutils <at> gnu.org:
bug#8230; Package
coreutils.
(Sun, 13 Mar 2011 13:42:02 GMT)
Full text and
rfc822 format available.
Message #29 received at 8230 <at> debbugs.gnu.org (full text, mbox):
Excerpts from Jim Meyering's message of Sun Mar 13 05:11:33 -0400 2011:
Hi Jim,
> I've just updated coreutils to use the latest from gnulib, so this
> will be fixed in coreutils-8.11.
Great!
> Ben, can you confirm that touch from coreutils-8.7 did not have this
> problem? I'll wait for confirmation before pushing.
I just built 8.7 (I skipped from 8.4 -> 8.8) with my build script and
tested it. It works correctly, so I'd say that yes, the problem was
introduced in 8.8. This was on sparc, I didn't make the verification
on i386, but I will if you'd like.
Thanks
-Ben
--
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302
Reply sent
to
Jim Meyering <jim <at> meyering.net>:
You have taken responsibility.
(Sun, 13 Mar 2011 17:57:02 GMT)
Full text and
rfc822 format available.
Notification sent
to
Ben Walton <bwalton <at> artsci.utoronto.ca>:
bug acknowledged by developer.
(Sun, 13 Mar 2011 17:57:02 GMT)
Full text and
rfc822 format available.
Message #34 received at 8230-done <at> debbugs.gnu.org (full text, mbox):
Ben Walton wrote:
>> Ben, can you confirm that touch from coreutils-8.7 did not have this
>> problem? I'll wait for confirmation before pushing.
>
> I just built 8.7 (I skipped from 8.4 -> 8.8) with my build script and
> tested it. It works correctly, so I'd say that yes, the problem was
> introduced in 8.8. This was on sparc, I didn't make the verification
> on i386, but I will if you'd like.
Thanks for confirming. That should be enough.
I've pushed that change to coreutils, and am marking this issue as closed.
bug archived.
Request was from
Debbugs Internal Request <help-debbugs <at> gnu.org>
to
internal_control <at> debbugs.gnu.org.
(Mon, 11 Apr 2011 11:24:04 GMT)
Full text and
rfc822 format available.
This bug report was last modified 13 years and 233 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.