about summary refs log tree commit diff
path: root/sysdeps/i386
Commit message (Collapse)AuthorAgeFilesLines
* Fix alignment in x86 destructor calls.Ulrich Drepper2011-01-121-0/+1
|
* Clean up some bits/select.h headers.Ulrich Drepper2011-01-091-25/+8
|
* Remove `.ctors' and `.dtors' output sectionsH.J. Lu2010-12-241-0/+2
|
* Make PowerPC64 default to nonexecutable stackRyan S. Arnold2010-12-191-1/+7
|
* Undo accidental checkin.Ulrich Drepper2010-12-141-25/+30
|
* Fix -D_FORTIFY_SOURCE memmove and bcopJakub Jelinek2010-12-091-30/+25
|
* Remove dead code from x86-32 SSSE3 strncmp.H.J. Lu2010-12-011-8/+1
|
* Relax libm tests for x86-32.Ulrich Drepper2010-11-301-2/+2
|
* Fix warnings in __bswap_16.Ulrich Drepper2010-11-101-7/+8
|
* 32bit memset-sse2.S fails with uneven cache sizeUlrich Drepper2010-11-051-0/+4
| | | | | | | | | 32bit memset-sse2.S assumes cache size is multiple of 128 bytes. If it isn't true, memset-sse2.S will fail. For example, a processor can have 24576 KB L3 cache and 20 cores. That is 2516582 byte per core. Half of it is 1258291, which isn't helpful for vector instructions. This patch rounds cache sizes to multiple of 256 bytes and adds "raw" cache sizes.
* Provide FP_FAST_FMA{,F,L} definitions for x86/x86-64.Ulrich Drepper2010-10-191-1/+15
|
* Implement fmal, some fma bugfixesJakub Jelinek2010-10-153-94/+0
|
* Implement accurate fma.Jakub Jelinek2010-10-132-5/+3
|
* Correct implementation of fmaf.Jakub Jelinek2010-10-113-3/+5
|
* Add missing i686/multiarch/Makefile bit for shared x86_64 code.Roland McGrath2010-08-271-0/+1
|
* Fixup x86 after x86-64 varshift change.Ulrich Drepper2010-08-272-1/+1
|
* Unroll 32bit SSE strlen and handle slow bsfH.J. Lu2010-08-254-80/+480
|
* Clean up SSE variable shiftsRichard Henderson2010-08-243-1/+3
|
* Fix ifunc thunk for strspn on x86 in static libcAndreas Schwab2010-08-191-2/+2
|
* Add -fno-asynchronous-unwind-tables to initfini.s for i386H.J. Lu2010-08-151-0/+4
|
* x86: Add support for frame pointer less mcountAndi Kleen2010-08-071-0/+25
|
* Add missing file from last commit.Andreas Schwab2010-07-211-0/+2
|
* Add strcasestr-nonascii to i386 buildAndreas Schwab2010-07-211-1/+2
|
* Fix bugs in x86-32 strcmp-sse4.S and strcmp-ssse3.SH.J. Lu2010-04-152-7/+8
|
* Add x86-32 FMA supportH.J. Lu2010-04-148-0/+179
|
* Fix sysdeps/i386/i686/multiarch/memcmp-ssse3.SH.J. Lu2010-03-181-0/+15
|
* Pass -mtune=i686 to assembler when compiling for i686H.J. Lu2010-03-043-0/+43
|
* Fix unwind info in x86 strcmp-sse4.SH.J. Lu2010-02-261-1/+3
|
* Fix issues in x86 memcpy-ssse3-rep.SH.J. Lu2010-02-241-106/+140
|
* Fix issues in x86 memcpy-ssse3.SH.J. Lu2010-02-241-40/+73
|
* Fix issues in x86 memset-sse2.S/memset-sse2-rep.SH.J. Lu2010-02-242-15/+18
|
* Use CPUID_OFFSET instead of FEATURE_OFFSETH.J. Lu2010-02-182-2/+2
|
* Align x86 memcmp-sse4.S and fix unwind info.H.J. Lu2010-02-171-7/+23
|
* Fix unwind info in x86 memcmp-ssse3.H.J. Lu2010-02-171-5/+1
|
* Simplify x86 strcmp-sse4 unwind info.H.J. Lu2010-02-171-15/+7
|
* Fix and cleanup unwind info in x86 strcmp-ssse<F3>.Ulrich Drepper2010-02-171-73/+14
|
* Remove commented-out code.Ulrich Drepper2010-02-151-6/+0
|
* Fix up new x86 string functions.Ulrich Drepper2010-02-154-38/+151
|
* 32bit memcmp/strcmp/strncmp optimized for SSSE3/SSS4.2H.J. Lu2010-02-1511-1/+5743
|
* More compact unwind info.Ulrich Drepper2010-02-091-5/+3
|
* Fix i386 __mpn_lshift unwind infoUlrich Drepper2010-02-091-1/+3
|
* Avoid PLT call to fegetenv on s390Andreas Schwab2010-02-091-0/+1
|
* Support __memmove_ssse3_rep when SHARED isn't defined.H.J. Lu2010-01-191-0/+3
|
* Fix whitespace problems.Ulrich Drepper2010-01-126-20/+20
|
* Optimize 32bit memset/memcpy with SSE2/SSSE3.H.J. Lu2010-01-1231-12/+6232
|
* Define bit_SSE2 and index_SSE2.H.J. Lu2009-12-133-8/+8
|
* Correct errno handling in expm1.Andreas Schwab2009-10-193-1/+14
|
* configure tweaks, support $libc_add_on_config_subdirsRoland McGrath2009-09-152-156/+138
|
* Fix IA-32 strstr in multiarch configuration as well.Ulrich Drepper2009-09-031-1/+2
|
* Remove ENABLE_SSSE3_ON_ATOM.H.J. Lu2009-08-281-2/+0
| | | | | It turns that SSSE3 isn't slow on Atom. The problem is bsf. This patch removes ENABLE_SSSE3_ON_ATOM.