qiurui
/
glibc
mirror of git://sourceware.org/git/glibc.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
19,698 Commits 249 Branches 1,650 Tags 318 MiB
Commit Graph

19698 Commits

Author SHA1 Message Date
H.J. Lu 56526b14bb Mention BZ #14278 in NEWS 2012-06-21 10:30:14 -07:00
H.J. Lu 1f1509082e Fix a typo in i386 SYSCALL_PIC_SETUP 2012-06-21 10:17:07 -07:00
Jeff Law 28363bbf70 2012-06-21 Jeff Law <law@redhat.com> 
[BZ #13882]
        * elf/dl-deps.c (_dl_map_object_deps): Fix cycle detection.  Use
        uint16_t for elements in the "seen" array to avoid char overflows.
        * elf/dl-fini.c (_dl_sort_fini): Likewise.
        * elf/dl-open.c (dl_open_worker): Likewise.
 2012-06-21 09:26:41 -06:00
Carlos O'Donell 09615db4a9 Add ChangeLog ... 
... for feca660a79.
 2012-06-21 08:04:35 -07:00
Carlos O'Donell feca660a79 Regenerate libc.pot. 
Enhance scripts/list-sources.sh to search glibc-ports
for translatable strings.
 2012-06-21 07:58:27 -07:00
Andreas Jaeger 8d847132bc Fix typo in name 2012-06-21 16:45:27 +02:00
Andreas Jaeger 2174c6dd85 Avoid -Wconversion warning for htons 2012-06-21 15:49:33 +02:00
Petr Baudis 416bf84422 locales/tr_CY: LC_TIME with copy cannot contain additional data 
Luckily, tr_TR also sets first_weekday and first_workday to Monday,
so this is just not necessary.
 2012-06-20 17:30:05 +02:00
Petr Baudis a324404dd9 locales/mag_IN: Fix syntax error 2012-06-20 17:29:08 +02:00
Siddhesh Poyarekar 18b5e737de Account for the extra stack size when rlimit is small enough 
When rlimit is small enough to be used as the stacksize to be returned
in pthread_getattr_np, cases where a stack is made executable due to a
DSO load get stack size that is larger than what the kernel
allows. This is because in such a case the stack size does not account
for the pages that have auxv and program arguments.

Additionally, the stacksize for the process derived from this should
be truncated to align to page size to avoid going beyond rlimit.
 2012-06-20 15:08:22 +05:30
H.J. Lu a2f34833b1 Use x86-64 fpu_control.h for i386 and x86_64 2012-06-18 11:43:51 -07:00
H.J. Lu 3ee947b124 Use __volatile__ to access fpu control word 2012-06-18 11:09:58 -07:00
Richard Henderson 62c8cc6dbe Update news for fixed alpha bz 2012-06-15 14:42:57 -07:00
H.J. Lu 1fb6e570ab Mention BZ 14050/14117 in NEWS 2012-06-15 14:33:57 -07:00
H.J. Lu 25f1282ae5 Use i386 bits/mathinline.h for i386 and x86_64 2012-06-15 13:56:26 -07:00
H.J. Lu ed1825f858 Disable x87 inline functions for x86-64 2012-06-15 13:53:27 -07:00
Chung-Lin Tang 36d54b744e Make endian.h usable for assembler 2012-06-15 21:19:12 +02:00
Andreas Jaeger 929cc14e20 Mention ix86 shared headers 2012-06-15 21:13:21 +02:00
H.J. Lu f8308a72ea Mention BZ #14229 in NEWS 2012-06-15 05:38:30 -07:00
Joseph Myers 10285c2113 Fix pow (0.0, -Inf) (bug 14241). 2012-06-15 11:27:51 +00:00
Andreas Jaeger 06c5abbdf8 Update manual/contrib.texi 
Update entry of Liubov Dmitrieva and add entries for Will Schmidt and Tulio Magno
Quites Machado Filho.
 2012-06-15 09:14:17 +02:00
Cyril Hrubis a3aeac403e Add __wur to GNU version of strerror_r. 
Not using the result of the GNU strerror_r() is always a mistake.

Moreover this would generate warning if XSI version was expected but GNU
version was used instead (because some random used header defined
_GNU_SOURCE which was Python.h in this case).
 2012-06-15 09:09:59 +02:00
H.J. Lu 49bdf4c19d Use RAX_LP/RDX_LP on SAVE_PTR in x86_64 strtok.S 2012-06-14 10:00:28 -07:00
Roland McGrath 834f9b8d3c Make sure <build-dir>/manual directory gets created. 2012-06-14 09:44:27 -07:00
Joseph Myers 75fa192131 Allow more spurious underflow exceptions from csqrt and Bessel functions. 2012-06-14 14:35:24 +00:00
Carlos O'Donell 771766df5a Set default charset to UTF-8 for libc.pot. 2012-06-13 16:51:55 -07:00
Paul Eggert 8e0933bb47 Mention 14210 in NEWS. 2012-06-13 10:46:52 -07:00
Paul Pluzhnikov ceb9e56b3d Suppress sign-conversion warning from FD_SET. 
[BZ #14210] See <http://sourceware.org/ml/libc-alpha/2012-05/msg01794.html>.
* debug/fdelt_chk.c (__fdelt_chk): Accept and return long int,
not unsigned long int.
* misc/bits/select2.h (__fdelt_chk, __fdelt_warn, __FD_ELT): Likewise.
 2012-06-13 09:57:18 -07:00
H.J. Lu ddb2897514 Add SSE math and C++ namespace to i386 bits/mathinline.h 2012-06-12 21:10:00 -07:00
Thomas Schwinge d7b4fb26e2 Fix a typo in sysdeps/unix/Makefile. 2012-06-12 10:27:01 +02:00
Roland McGrath eb55f5c21f BZ#14218: Reword argp_parse description slightly. 2012-06-11 16:21:44 -07:00
Thomas Schwinge 99ff6e5c49 SH: No FE_UPWARD, FE_DOWNWARD. 2012-06-09 18:46:01 +02:00
Thomas Schwinge 366af02c96 SH: Do not re-write fpscr after reading it. 2012-06-09 18:36:52 +02:00
Kaz Kojima 793ea8518f SH: refresh *.abilist files. 2012-06-09 18:08:02 +02:00
Carlos Sánchez de La Lama 24a6dbed9b Fix build on non-v9 sparc32. 
nptl/

	[BZ #14205]
	* sysdeps/sparc/sparc32/pthread_spin_lock.S: Do not use v9
	branches.
 2012-06-07 11:57:09 -07:00
H.J. Lu c08010c7cd Use x86_64 fpu/bits/fenv.h for i386 and x86_64 2012-06-06 10:13:19 -07:00
H.J. Lu a728a38fb6 Check __SSE_MATH__ in x86_64 feraiseexcept 2012-06-06 10:11:14 -07:00
Siddhesh Poyarekar 6e230d1183 Fix validation check when converting from ibm930 to utf 
[BZ #14134]

When converting IBM930 code with iconv(), if IBM930 code which
includes invalid multibyte character "0xffff" is specified, then
iconv() segfaults. This is easy to see using the following command:

echo '0x0e 0x43 0x8c 0xff 0xff 0x43 0xbd 0x43 0xbd' | xxd -r |
	iconv -f IBM930 -t UTF-8
 2012-06-06 18:41:55 +05:30
Adhemerval Zanella 1b671feb61 Fix for wrong ldbl128-ibm fmodl commit 2012-06-05 21:33:23 -03:00
Adhemerval Zanella 6043738b36 Fix spurious undeflow for ldbl-128ibm erfl 
For values higher than 25.6283 erflc underflow, so adjust erfl
to return a constant value based argument sign.
 2012-06-05 10:42:49 -03:00
Adhemerval Zanella 34ae0b3270 Fix ldbl128ibm fmodl for subnormals. 2012-06-05 10:16:49 -03:00
Mike Frysinger 1214ec8f4c tst-getcpu: decode errno when sched_getcpu fails 
The tst-getcpu test is failing on ia64 platforms due to sched_getcpu
returning an error.  But the current test only considers -1 when it's
also ENOSYS.  Tweak the logic a bit to make the output clearer.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
 2012-06-04 22:22:02 -04:00
H.J. Lu 3cc0cd4c5b Fix a typo in comment 2012-06-04 13:47:07 -07:00
H.J. Lu 0e4a3cd708 Use i486 bits/string.h for i386 and x86_64 2012-06-04 13:45:43 -07:00
H.J. Lu 6704c645b0 Provide 32-bit inline string functions for >= i486 2012-06-04 13:44:46 -07:00
H.J. Lu 48495318fa Use x86_64 bits/sigcontext.h for i386 and x86_64 2012-06-04 12:41:38 -07:00
H.J. Lu 14adcbfc36 Update x86_64 bits/sigcontext.h from kernel 3.4.0 2012-06-04 12:40:46 -07:00
H.J. Lu 68e408ab5d Use x86_64 bits/{debugreg,reg}.h for i386 and x86-64 2012-06-04 12:32:07 -07:00
H.J. Lu f3d1f93cfb Use x86_64 bits/{io,perm}.h for i386 and x86_64 2012-06-04 12:29:49 -07:00
Tulio Magno Quites Machado Filho d9dc34cd56 Manual for platform-specific features and new __ppc_get_timebase inline. 
[BZ #13743]
A new class of installed headers has been documented for low-level
platform-specific functionality.  PowerPC added the first instance with a
function to provide time base register access (__ppc_get_timebase).  This
is required for applications that measure time at high frequencies with
high precision that can't afford a syscall.
 2012-06-04 13:46:37 -05:00