Joseph Myers [Sat, 1 Sep 2012 21:32:04 +0000 (21:32 +0000)]
Remove __ASSUME_POSIX_CPU_TIMERS.
Joseph Myers [Fri, 31 Aug 2012 19:49:31 +0000 (19:49 +0000)]
Fix sem_post race (bug 14532).
H.J. Lu [Wed, 29 Aug 2012 13:45:37 +0000 (06:45 -0700)]
Use LD_SO to set $ld_so_name/$ld_so_version
Siddhesh Poyarekar [Wed, 29 Aug 2012 04:33:39 +0000 (10:03 +0530)]
Don't make ttyname* fail if proc filesystem is not available
The ttyname and ttyname_r functions on Linux now fall back to
searching for the tty file descriptor in /dev/pts or /dev if /proc is
not available. This allows creation of chroots without the procfs
mounted on /proc.
Fixes BZ #14516.
Joseph Myers [Mon, 27 Aug 2012 19:36:04 +0000 (19:36 +0000)]
Remove __ASSUME_STAT64_SYSCALL.
Mike Frysinger [Mon, 27 Aug 2012 17:30:45 +0000 (13:30 -0400)]
update NEWS
Guido Guenther [Mon, 27 Aug 2012 16:20:25 +0000 (12:20 -0400)]
mips: clear condition bit 23 in fpu too
We've been carrying this in Gentoo for quite a long time to fix some test
failures that people hit.
Original message:
> make[4]: *** [/glibc/glibc-package-2.3/mips-linux/obj/math/test-fpucw.out] Error 1
This test fails since the read back fpu control word is 0x80000 instead
of 0x0. I wonder if this patch is correct:
...
which additionally masks out the condition bit 23 - note that the other
condition bits (25-31) are masked out too?
URL: http://sourceware.org/ml/libc-alpha/2002-10/msg00392.html
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Mike Frysinger [Fri, 17 Aug 2012 17:01:28 +0000 (13:01 -0400)]
alpha: add conditional fdatasync support
While most arches have had the fdatasync syscall for a long time, the
alpha port didn't add it until the 2.6.22 release.
This is heavily based on Aurelien Jarno's initial work.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Joseph Myers [Mon, 27 Aug 2012 16:04:19 +0000 (16:04 +0000)]
Fix sign of zero on strtod underflow (bug 14519).
Joseph Myers [Mon, 27 Aug 2012 16:01:27 +0000 (16:01 +0000)]
Fix strtod rounding (bug 3479).
Joseph Myers [Mon, 27 Aug 2012 15:59:24 +0000 (15:59 +0000)]
Fix strtod integer/buffer overflow (bug 14459).
Marek Polacek [Sat, 25 Aug 2012 16:32:23 +0000 (18:32 +0200)]
Fix some typos in comments.
Roland McGrath [Thu, 23 Aug 2012 23:55:40 +0000 (16:55 -0700)]
Fix tst-rfc3484* build failures from USE_NSCD move to config.h.
Steve McIntyre [Thu, 23 Aug 2012 20:38:25 +0000 (20:38 +0000)]
Check floating-point ABI in ARM VALID_ELF_HEADER.
Joseph Myers [Thu, 23 Aug 2012 11:34:37 +0000 (11:34 +0000)]
Make dl-fxstatat64.c include of fxstatat64.c use <> not "".
Roland McGrath [Wed, 22 Aug 2012 23:06:53 +0000 (16:06 -0700)]
Fix shadow, gshadow, networks, protocols, rpc, aliases, and nscd routines for USE_NSCD case.
Joseph Myers [Wed, 22 Aug 2012 21:03:12 +0000 (21:03 +0000)]
Clean up MIPS stat64 handling.
Roland McGrath [Wed, 22 Aug 2012 20:35:55 +0000 (13:35 -0700)]
Add --disable-build-nscd configure option.
Roland McGrath [Wed, 22 Aug 2012 20:31:12 +0000 (13:31 -0700)]
BZ#13696: Add --disable-nscd configure option.
Dmitry V. Levin [Wed, 22 Aug 2012 19:58:18 +0000 (12:58 -0700)]
Fix sed configure check for newer sed --version output.
Roland McGrath [Wed, 22 Aug 2012 19:53:38 +0000 (12:53 -0700)]
Fix libc-start change for IRELless machines.
Joseph Myers [Wed, 22 Aug 2012 19:17:56 +0000 (19:17 +0000)]
Move bug number to correct section of NEWS.
Jeff Law [Wed, 22 Aug 2012 17:41:40 +0000 (11:41 -0600)]
* sysdeps/posix/getaddrinfo.c (gaih_inet): Only use gethostbyname4_r
if the family is PF_UNSPEC.
Mike Frysinger [Wed, 22 Aug 2012 04:21:31 +0000 (00:21 -0400)]
rename V variable to lib-version
Due to the rise of kbuild type build systems (as the Linux kernel is a
popularly emulated environment), the V variable has become common as a
knob for controlling verbosity. Unfortunately, if you run `make V=1`
with glibc during install, it fails with weird errors due to the glibc
build already using this variable for versioning information.
Granted, overriding this variable in the glibc context makes no sense
so people shouldn't be doing it, but when paired with build frameworks
that like to use one set of options for all packages, glibc starts to
stick out as an oddball (in that it fails). Considering it's easy
enough to rename (it's used in just one place), let's do so.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Roland McGrath [Wed, 22 Aug 2012 16:25:21 +0000 (09:25 -0700)]
Leading-tabify ChangeLog.
Roland McGrath [Wed, 22 Aug 2012 16:24:51 +0000 (09:24 -0700)]
Fix ChangeLog format.
Will Schmidt [Wed, 22 Aug 2012 16:04:42 +0000 (11:04 -0500)]
Add versions of wcscpy, wcschr, wcsrchr for power6/power7.
Initially based on the versions found in wcsmbs/* ; these files have
been changed by hand unrolling, and adding some additional variables
to allow some read-ahead to occur, which then relieves some of the
wait-for-increment/wait-for-load/wait-for-compare-results pressure
that was slowing down every iteration through the while-loop.
For 64-bit Power7, These changes give an approx 20% throughput boost
for the wcschr and wcsrchr functions; and approx 40% boost for the
wcscpy function. 32-bit improvements appear to be slightly better
with ~ %30 and ~ %45 respectively. Results for Power6 closely match
those for power7.
Maxim Kuvyrkov [Tue, 29 May 2012 05:46:07 +0000 (22:46 -0700)]
Micro-optimize critical path of strstr, strcase and memmem.
Maxim Kuvyrkov [Tue, 29 May 2012 06:32:12 +0000 (23:32 -0700)]
Use pointers for traversing arrays in strstr, strcasestr and memmem.
Maxim Kuvyrkov [Tue, 29 May 2012 05:50:15 +0000 (22:50 -0700)]
Detect EOL on-the-fly in strstr, strcasestr and memmem.
Maxim Kuvyrkov [Tue, 29 May 2012 07:04:12 +0000 (00:04 -0700)]
Optimize first-character loop of strstr, strcasestr and memmem.
Roland McGrath [Tue, 21 Aug 2012 22:01:27 +0000 (15:01 -0700)]
Support static IFUNC calls irrespective of USE_MULTIARCH.
Joseph Myers [Tue, 21 Aug 2012 21:07:22 +0000 (21:07 +0000)]
Remove __ASSUME_FADVISE64_64_SYSCALL.
Will Schmidt [Tue, 21 Aug 2012 19:20:55 +0000 (14:20 -0500)]
[Powerpc] Tune/optimize powerpc{32,64}/power7/memchr.S.
Assorted tweaking, twisting and tuning to squeeze a few additional cycles
out of the memchr code. Changes include bypassing the shift pairs
(sld,srd) when they are not required, and unrolling the small_loop that
handles short and trailing strings.
Per scrollpipe data measuring aligned strings for 64-bit, these changes
save between five and eight cycles (9-13% overall) for short strings (<32),
Longer aligned strings see slight improvement of 1-3% due to bypassing the
shifts and the instruction rearranging.
Roland McGrath [Mon, 20 Aug 2012 19:44:59 +0000 (12:44 -0700)]
Fix typos in manual wrt syslog.
Roland McGrath [Mon, 20 Aug 2012 16:51:04 +0000 (09:51 -0700)]
Fix conditional on using DSOCAPS to match condition on defining it.
Joseph Myers [Mon, 20 Aug 2012 14:39:53 +0000 (14:39 +0000)]
Remove __ASSUME_SWAPCONTEXT_SYSCALL.
Joseph Myers [Mon, 20 Aug 2012 14:38:48 +0000 (14:38 +0000)]
Define __ASSUME_UTIMES for s390.
Joseph Myers [Mon, 20 Aug 2012 14:37:27 +0000 (14:37 +0000)]
Remove __ASSUME_MMAP2_SYSCALL.
Andreas Krebbel [Mon, 20 Aug 2012 13:10:13 +0000 (15:10 +0200)]
S/390: Remove 32 bit getrlimit.c.
Joseph Myers [Sun, 19 Aug 2012 15:34:59 +0000 (15:34 +0000)]
Add bug number to NEWS.
Andreas Jaeger [Sat, 18 Aug 2012 16:01:46 +0000 (18:01 +0200)]
Fix last patch: Add missing DUMMY variable
Mike Frysinger [Sat, 18 Aug 2012 04:37:30 +0000 (00:37 -0400)]
add attribute_hidden to __have_{sock_cloexec,pipe2,dup3
These internal knobs are not exposed as part of the public ABI, so mark
them hidden to avoid generating relocations against them.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Mike Frysinger [Sat, 18 Aug 2012 02:03:56 +0000 (22:03 -0400)]
split assume pipe2/dup3/sock_cloexec knobs
We can't assume sock_cloexec and pipe2 are bound together as the former
defines are found in glibc only while the latter are a combo of kernel
headers and glibc. So if we do a runtime detection of SOCK_CLOEXEC, but
pipe2() is a stub inside of glibc, we hit a problem. For example:
main()
{
getgrnam("portage");
if (!popen("ls", "r"))
perror("popen()");
}
getgrnam() will detect that the kernel supports SOCK_CLOEXEC and then set
both __have_sock_cloexec and __have_pipe2 to true. But if glibc was built
against older kernel headers where __NR_pipe2 does not exist, glibc will
have a ENOSYS stub for it. So popen() will always fail as glibc assumes
pipe2() works.
While this isn't too much of an issue for some arches as they added the
functionality to the kernel at the same time, not all arches are that
lucky.
Since the code already has dedicated names for each feature, delete the
defines wiring these three features together and make each one a proper
dedicated knob.
We've been carrying this in Gentoo since glibc-2.9.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Mike Frysinger [Sat, 18 Aug 2012 02:39:07 +0000 (22:39 -0400)]
update linux nice.c include path too after recent file shuffle
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Marek Polacek [Fri, 17 Aug 2012 21:44:53 +0000 (23:44 +0200)]
Quash warning in s_sincosl.
Roland McGrath [Fri, 17 Aug 2012 21:16:12 +0000 (14:16 -0700)]
Adjust old #include's of sysdeps/unix/time.c to point to sysdeps/posix/time.c instead.
Roland McGrath [Fri, 17 Aug 2012 21:05:53 +0000 (14:05 -0700)]
Make sysheaders available in config.make.
Roland McGrath [Fri, 17 Aug 2012 20:52:32 +0000 (13:52 -0700)]
Move some things from sysdeps/unix to sysdeps/posix.
Roland McGrath [Fri, 17 Aug 2012 20:42:27 +0000 (13:42 -0700)]
Fix rule dependency in last change.
Jeff Law [Fri, 17 Aug 2012 20:32:19 +0000 (14:32 -0600)]
2012-08-17 Jeff Law <law@redhat.com>
* intl/Makefile (codeset_mo): New variable.
($(codeset_mo)): New target.
(tst-codeset.out): Depend on that. Remove explicit rule.
(tst-gettext3.out, tst-gettext5.out): Likewise.
(LOCPATH-ENV, tst-codeset-ENV): New variables.
(tst-gettext3-ENV, tst-gettext5-ENV): Likewise.
* intl/tst-codeset.sh: Remove.
* intl/tst-gettext3.sh: Likewise.
* intl/tst-gettext5.sh: Likewise.
Roland McGrath [Fri, 17 Aug 2012 19:56:33 +0000 (12:56 -0700)]
Merge unix/inet/syscalls.list into unix/syscalls.list.
Roland McGrath [Fri, 17 Aug 2012 18:49:00 +0000 (11:49 -0700)]
Fix typo in last change.
Roland McGrath [Fri, 17 Aug 2012 18:47:06 +0000 (11:47 -0700)]
Fix getaddrinfo for [!_STATBUF_ST_NSEC] case.
Roland McGrath [Fri, 17 Aug 2012 18:29:45 +0000 (11:29 -0700)]
Make malloc build for no-threads configurations.
Roland McGrath [Fri, 17 Aug 2012 18:17:04 +0000 (11:17 -0700)]
Change type of constant to avoid a warning.
Roland McGrath [Fri, 17 Aug 2012 16:55:17 +0000 (09:55 -0700)]
Split sys/param.h out into common file and sysdeps bits/param.h file.
Roland McGrath [Fri, 17 Aug 2012 16:49:37 +0000 (09:49 -0700)]
Add a cast to silence a warning.
Roland McGrath [Fri, 17 Aug 2012 16:35:36 +0000 (09:35 -0700)]
Make libio compile without _IO_MTSAFE_IO.
Roland McGrath [Fri, 17 Aug 2012 16:35:15 +0000 (09:35 -0700)]
Clean up definition of _LIBC_REENTRANT and _IO_MTSAFE_IO.
Gary Benson [Fri, 17 Aug 2012 10:44:37 +0000 (11:44 +0100)]
Also set r->r_map when unmapping the first object in a namespace.
When unmapping the first object in a namespace, the runtime linker
did not update the externally visible pointer. This resulted in
debuggers seeing pointers to memory that had been freed.
Roland McGrath [Thu, 16 Aug 2012 22:38:45 +0000 (15:38 -0700)]
Do not pollute name space with internal_*netgrent functions.
Joseph Myers [Thu, 16 Aug 2012 21:37:04 +0000 (21:37 +0000)]
Update longlong.h from GCC.
Roland McGrath [Thu, 16 Aug 2012 21:08:47 +0000 (14:08 -0700)]
Suppress -Wunused-but-set warnings in GMP code.
Carlos O'Donell [Thu, 16 Aug 2012 15:36:24 +0000 (08:36 -0700)]
Fifth argument of la_pltenter() is not constant.
The original runtime linker auditing interface described
by Solaris allows the 5th argument of la_pltenter() to be
modified. This patch cleans up the ldsodefs.h definitions
such that the 5th argument is not constant.
At one point the 5th argument *was* constant but this was
changed with commit
2413fdba7a02ba8916f75d17199a6e9133a8f7b0.
This patch updates alpha, ia64, mips, sh and sparc with similar
changes.
Joseph Myers [Thu, 16 Aug 2012 14:03:43 +0000 (14:03 +0000)]
Remove __ASSUME_POSIX_TIMERS.
Roland McGrath [Thu, 16 Aug 2012 00:05:10 +0000 (17:05 -0700)]
Fix dl-load.c for [!_LIBC_REENTRANT] case.
Roland McGrath [Thu, 16 Aug 2012 00:03:56 +0000 (17:03 -0700)]
Define __rtld_lock_initialize in stub bits/libc-lock.h file.
Roland McGrath [Thu, 16 Aug 2012 00:02:22 +0000 (17:02 -0700)]
Add a missing #include to dl-sym.c.
Maxim Kuvyrkov [Wed, 15 Aug 2012 23:44:30 +0000 (16:44 -0700)]
Optimize __libc_lock_lock and __libc_lock_trylock for MIPS.
Maxim Kuvyrkov [Wed, 15 Aug 2012 23:29:06 +0000 (16:29 -0700)]
Add generic versions of pthread_spin_lock and pthread_spin_trylock.
Roland McGrath [Wed, 15 Aug 2012 23:03:58 +0000 (16:03 -0700)]
Suppress regcomp.c warnings in 32-bit builds.
Roland McGrath [Wed, 15 Aug 2012 22:47:52 +0000 (15:47 -0700)]
Define O_DSYNC and O_RSYNC in generic/4.4 bits/fcntl.h file.
Roland McGrath [Wed, 15 Aug 2012 22:40:06 +0000 (15:40 -0700)]
Move stub lseek.c to the right directory.
Roland McGrath [Wed, 15 Aug 2012 22:35:16 +0000 (15:35 -0700)]
Fix stub clock_nanosleep #include.
Roland McGrath [Wed, 15 Aug 2012 20:59:59 +0000 (13:59 -0700)]
Fix last sha512.c change to avoid compiler warning.
Andreas Jaeger [Wed, 15 Aug 2012 19:18:44 +0000 (21:18 +0200)]
Remove executable permissions checked in by accident
Liubov Dmitrieva [Wed, 15 Aug 2012 19:06:55 +0000 (21:06 +0200)]
Fix segmentation fault in strncasecmp for i686
2012-08-15 Liubov Dmitrieva <liubov.dmitrieva@gmail.com>
[BZ #14195]
* sysdeps/i386/i686/multiarch/strcmp-sssse3.S: Fix
segmentation fault for a case of two empty input strings.
* string/test-strncasecmp.c (check1): Renamed to...
(bz12205): ...this.
(bz14195): Add new testcase for two empty input strings and N > 0.
(test_main): Call new testcase, adapt for renamed function.
Andreas Jaeger [Wed, 15 Aug 2012 18:51:04 +0000 (20:51 +0200)]
Add testcase for BZ#14090 - md5/sha512 with large sizes
Paul Eggert [Wed, 15 Aug 2012 18:49:45 +0000 (20:49 +0200)]
Fix BZ#14090 - md5/sha512 with large sizes
Roland McGrath [Wed, 15 Aug 2012 18:40:41 +0000 (11:40 -0700)]
Remove local redefinition of MAX macro.
Roland McGrath [Wed, 15 Aug 2012 18:38:57 +0000 (11:38 -0700)]
Clean up x86_64/multiarch/strstr-c.c include order.
Roland McGrath [Wed, 15 Aug 2012 18:26:02 +0000 (11:26 -0700)]
Clean up x86_64/multiarch/memmove.c include order.
Roland McGrath [Wed, 15 Aug 2012 17:09:18 +0000 (10:09 -0700)]
Declare __getdirentries in internal dirent.h.
Roland McGrath [Wed, 15 Aug 2012 17:08:32 +0000 (10:08 -0700)]
Add casts to suppress warnings in system.c under [!_LIBC_REENTRANT].
Roland McGrath [Tue, 14 Aug 2012 23:39:38 +0000 (16:39 -0700)]
Rename ARM bits/atomic.h not to be specific to nptl add-on.
Roland McGrath [Tue, 14 Aug 2012 23:38:23 +0000 (16:38 -0700)]
Separate ARM bits/atomic.h into generic and Linux-specific files.
Joseph Myers [Wed, 15 Aug 2012 09:49:07 +0000 (09:49 +0000)]
Update MIPS dl-lookup.c from generic version.
Mike Frysinger [Wed, 15 Aug 2012 01:38:15 +0000 (21:38 -0400)]
arm: fix up style in reloc fix
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Mike Frysinger [Sun, 12 Aug 2012 19:13:10 +0000 (15:13 -0400)]
i386/x86_64: punt HAVE_CPP_ASM_DEBUGINFO
Pretty sure we require recent enough versions of gcc/binutils to make this
check pointless. I can't any logs in the last few years where this check
didn't return "yes".
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Mike Frysinger [Sun, 12 Aug 2012 19:09:12 +0000 (15:09 -0400)]
ia64: drop HAVE_CPP_ASM_DEBUGINFO
Looks like a wart copied from the i386 code base. The only place I can
find that checks this is the i386 sysdep.h, and even then this looks like
a check that should get thrown away as obsolete ...
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Roland McGrath [Tue, 14 Aug 2012 23:43:24 +0000 (16:43 -0700)]
Add a missing #include.
Roland McGrath [Tue, 14 Aug 2012 23:04:29 +0000 (16:04 -0700)]
Add a missing #include.
Joseph Myers [Tue, 14 Aug 2012 22:34:04 +0000 (22:34 +0000)]
Remove __ASSUME_CLONE_THREAD_FLAGS.
Joseph Myers [Tue, 14 Aug 2012 20:28:33 +0000 (20:28 +0000)]
Add more fixed bug numbers to NEWS.
Andreas Jaeger [Tue, 14 Aug 2012 06:04:51 +0000 (08:04 +0200)]
Another ULPs update.
Andreas Jaeger [Tue, 14 Aug 2012 06:02:08 +0000 (08:02 +0200)]
Update i386 ULPs
Andreas Jaeger [Tue, 14 Aug 2012 05:59:57 +0000 (07:59 +0200)]
Use tabs instead of spaces
Maxim Kuvyrkov [Tue, 14 Aug 2012 02:55:34 +0000 (19:55 -0700)]
Add explicit acquire/release semantics to MIPS' atomic_exchange_and_add.
Maxim Kuvyrkov [Tue, 14 Aug 2012 02:31:00 +0000 (19:31 -0700)]
Add explicit acquire/release semantics to atomic_exchange_and_add.