about summary refs log tree commit diff
path: root/sysdeps/x86_64/fpu
Commit message (Collapse)AuthorAgeFilesLines
* Reduce ldouble ULPs for jn tests on x86Andreas Schwab2012-02-081-18/+16
|
* Remove __ELF__ conditionalsMarek Polacek2012-02-079-44/+11
|
* Fix makefile/configure problems with sse2avx changes.Joseph Myers2012-01-301-1/+1
|
* Optimize x86-64 math inline header a bitUlrich Drepper2012-01-281-8/+31
|
* Use -msse2avx option for x86-64 libm functionsUlrich Drepper2012-01-288-49/+26
|
* Simplify use of AVX instructions in internal math macrosUlrich Drepper2012-01-281-59/+20
|
* Really fix AVX testsUlrich Drepper2012-01-266-7/+11
| | | | | | There is no problem with strcmp, it doesn't use the YMM registers. The math routines might since gcc perhaps generates such code. Introduce bit_YMM_USBALE and use it in the math routines.
* Remove pre-ISO C supportUlrich Drepper2012-01-071-3/+3
| | | | No more __const.
* Use __REDIRECT_NTH for __feraiseexcept_renamedUlrich Drepper2011-12-221-1/+1
|
* Define x86_64 feraiseexcept inline only under __USE_EXTERN_INLINES.Rafael Ávila de Espíndola2011-12-211-1/+1
|
* Fix more warningsUlrich Drepper2011-12-032-11/+0
|
* Remove code without too much effectsUlrich Drepper2011-10-2815-113/+5
| | | | | Some of the AVX-specific code is not giving enough speed-up to justify the extra code.
* Add AVX optimized versions for some x86-64 math functionsUlrich Drepper2011-10-2527-20/+276
|
* Use VEX encoding in inline math functions on x86-64 when possibleUlrich Drepper2011-10-251-48/+138
|
* Cleanup FMA4 patchUlrich Drepper2011-10-2522-1/+26
| | | | | Move the FMA4 code into its own section. Avoid some of the duplication of data resulting from the double use of source files.
* Better DLA_FMSUlrich Drepper2011-10-241-0/+17
| | | | | It's better to use __builtin_fma if it works. Use it for gcc 4.6 and higher. Move the x86-64 dla.h to the correct place.
* Optimize accurate 64-bit routines for FMA4 on x86-64Ulrich Drepper2011-10-2431-0/+328
|
* Fix typo in last changeUlrich Drepper2011-10-221-1/+1
|
* Update ULPs for x86-64Ulrich Drepper2011-10-221-4/+12
|
* Don't use NULL in last s_fma{,f} changeUlrich Drepper2011-10-212-2/+2
|
* Check for FMA4 support and generate appropriate fma functionsUlrich Drepper2011-10-202-6/+38
|
* Move fma routines to right placeUlrich Drepper2011-10-202-0/+85
|
* Fix PLT use for feraiseexcept on x86_64Andreas Schwab2011-10-191-1/+1
|
* Use new internal libc_fe* interfaces in more functionsUlrich Drepper2011-10-181-5/+30
|
* Provide combined internal feholdexcept/fesetround interfaceUlrich Drepper2011-10-181-1/+13
|
* Pretty print last change to x86-64 mathinline.hUlrich Drepper2011-10-181-2/+2
|
* Add optimized nearbyint{,f} for x86-64Ulrich Drepper2011-10-186-1/+105
|
* Start optimizing the use of the fenv interfaces in libm itselfUlrich Drepper2011-10-181-0/+42
|
* Fix last changeAndreas Schwab2011-10-181-2/+3
|
* Fix linkage conflict with feraiseexceptAndreas Schwab2011-10-181-6/+10
|
* Relax asm requirements for recently added x86-64 math interfacesUlrich Drepper2011-10-172-10/+10
|
* Makr x86-64 math_private.h more robustUlrich Drepper2011-10-171-10/+10
|
* Provide internal optimizations on x86-64 with SSE4.1Ulrich Drepper2011-10-176-12/+51
| | | | | | | Provide macros so that the internal users can, if possible, directly use the new instructions. Also fix up the mathinline.h header when compiling with SSE4.1 enabled.
* Fix last x86-64 mathinline changeUlrich Drepper2011-10-171-2/+2
| | | | Use correct function names.
* Use rounds{s,d} for x86 rint, ceil, floorUlrich Drepper2011-10-1614-10/+335
|
* Fix lost feraiseexcept symbolAndreas Schwab2011-10-141-1/+2
|
* Fix PLT use in feupdateenv on x86_64Andreas Schwab2011-10-132-5/+5
|
* Optimize libmUlrich Drepper2011-10-1213-47/+152
| | | | | libm is now somewhat integrated with gcc's -ffinite-math-only option and lots of the wrapper functions have been optimized.
* Optimize use of isnan, isinf, finiteUlrich Drepper2011-10-081-0/+21
|
* Fix parse error in bits/mathinline.h with --std=c99Andreas Schwab2011-09-261-8/+8
|
* Add fmax and fmin inlines for x86-64Ulrich Drepper2011-09-151-5/+41
|
* Use correct section to allow mergingUlrich Drepper2011-09-141-2/+2
|
* Optimized lrint and llrint for x86-64Ulrich Drepper2011-09-141-10/+46
|
* Fix jn precisionPetr Baudis2011-09-091-44/+98
|
* Fix typo in x86-64 powlH.J. Lu2011-05-181-2/+2
|
* Correct implementation of fmaf.Jakub Jelinek2010-10-112-2/+4
|
* Avoid PLT call to fegetenv on s390Andreas Schwab2010-02-091-0/+1
|
* Fix x86_64 bits/mathinline.h for -m32 compilation.Jakub Jelinek2009-09-011-0/+12
|
* Fix parse error in bits/mathinline.h with --std=c99Andreas Schwab2009-08-311-2/+2
|
* Optimize out duplicated scalbln code for x86-64.Ulrich Drepper2009-08-252-0/+11
|
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-11-11 15:25:42 +0000