Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2011-10-26 | Fix uses of math_force_eval | Andreas Schwab | 3 | -6/+6 | |
2011-10-26 | Make math_force_eval accept non-addressable arguments | Andreas Schwab | 1 | -3/+4 | |
2011-10-25 | Add AVX optimized versions for some x86-64 math functions | Ulrich Drepper | 29 | -92/+335 | |
2011-10-25 | Optimize x86-64 SSE4.2+ strcmp a bit more | Ulrich Drepper | 1 | -175/+130 | |
2011-10-25 | Use math_force_eval in more places | Ulrich Drepper | 20 | -310/+235 | |
2011-10-25 | Use VEX encoding in inline math functions on x86-64 when possible | Ulrich Drepper | 1 | -48/+138 | |
2011-10-25 | Cleanup FMA4 patch | Ulrich Drepper | 46 | -198/+444 | |
Move the FMA4 code into its own section. Avoid some of the duplication of data resulting from the double use of source files. | |||||
2011-10-24 | Better DLA_FMS | Ulrich Drepper | 1 | -1/+8 | |
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. | |||||
2011-10-24 | Use inline asm for DLA_FMS because of broken old compilers | Ulrich Drepper | 1 | -2/+6 | |
2011-10-24 | Optimize accurate 64-bit routines for FMA4 on x86-64 | Ulrich Drepper | 43 | -176/+533 | |
2011-10-24 | Finish renamed DLA_FMA -> DLA_FMS | Ulrich Drepper | 8 | -9/+9 | |
2011-10-23 | Optimized remquo for 64-bit platforms | Ulrich Drepper | 1 | -0/+112 | |
2011-10-23 | Optimize f_mod | Ulrich Drepper | 2 | -6/+109 | |
Branch prediction for the 32-bit implementation and a new optimized 64-bit implementation. | |||||
2011-10-23 | Fix strnlen change | Ulrich Drepper | 1 | -0/+1 | |
2011-10-23 | Add optimized wcslen and strnlen for x86-32 | Liubov Dmitrieva | 8 | -40/+723 | |
2011-10-23 | Fix WS | Ulrich Drepper | 1 | -1/+1 | |
2011-10-23 | Optimized strnlen and wcscmp for x86-64 | Liubov Dmitrieva | 8 | -317/+987 | |
2011-10-23 | Improve x86-32 SSSE3 memcpy | Michael Zolotukhin | 1 | -219/+708 | |
2011-10-23 | Fix WS | Ulrich Drepper | 2 | -10/+10 | |
2011-10-23 | Fix signedness in wcscmp comparison | Liubov Dmitrieva | 2 | -116/+151 | |
2011-10-23 | Clean up FMA use | Ulrich Drepper | 2 | -9/+6 | |
The macro's name should reflect that subtraction is being done. And use __builtin_fma, it seems to work after all. | |||||
2011-10-23 | Clean up last dla.h change | Ulrich Drepper | 11 | -21/+18 | |
2011-10-23 | Add missing _finite aliases | Andreas Schwab | 1 | -0/+2 | |
2011-10-22 | Fix typo in last change | Ulrich Drepper | 1 | -1/+1 | |
2011-10-22 | Update ULPs for x86-64 | Ulrich Drepper | 1 | -4/+12 | |
2011-10-22 | Improve DLA_FMA for FMA4 | Ulrich Drepper | 1 | -2/+2 | |
2011-10-22 | Start using fma in the libm implementation | Ulrich Drepper | 9 | -150/+192 | |
2011-10-22 | Fix compilation due to __nan defines | Andreas Schwab | 1 | -0/+1 | |
2011-10-22 | Optimization to some complex math functions | Ulrich Drepper | 3 | -3/+3 | |
Also, change last reference to __sqrt into __sqrt to easily. | |||||
2011-10-21 | Fix compilation problems in x86-64 init-arch | Ulrich Drepper | 1 | -1/+2 | |
2011-10-21 | Add some __builtin_expect to generic __ieee754_log | Ulrich Drepper | 1 | -3/+3 | |
2011-10-21 | Don't use NULL in last s_fma{,f} change | Ulrich Drepper | 2 | -2/+2 | |
2011-10-20 | Check for FMA4 support and generate appropriate fma functions | Ulrich Drepper | 6 | -27/+103 | |
2011-10-20 | Move fma routines to right place | Ulrich Drepper | 2 | -0/+0 | |
2011-10-19 | Optimize x86-64 rawmemchr and add test | Ulrich Drepper | 1 | -4/+3 | |
2011-10-19 | Add optimized str{,n}casecmp for AVX on x86-64 | Ulrich Drepper | 3 | -1798/+1870 | |
2011-10-19 | Fix PLT use for feraiseexcept on x86_64 | Andreas Schwab | 1 | -1/+1 | |
2011-10-18 | Use new internal libc_fe* interfaces in more functions | Ulrich Drepper | 6 | -56/+63 | |
2011-10-18 | Provide combined internal feholdexcept/fesetround interface | Ulrich Drepper | 2 | -6/+14 | |
2011-10-18 | Pretty print last change to x86-64 mathinline.h | Ulrich Drepper | 1 | -2/+2 | |
2011-10-18 | Add optimized nearbyint{,f} for x86-64 | Ulrich Drepper | 6 | -1/+105 | |
2011-10-18 | Start optimizing the use of the fenv interfaces in libm itself | Ulrich Drepper | 3 | -31/+58 | |
2011-10-18 | Fix last change | Andreas Schwab | 1 | -2/+3 | |
2011-10-18 | Fix linkage conflict with feraiseexcept | Andreas Schwab | 1 | -6/+10 | |
2011-10-17 | Small optimization of generic exp2 | Ulrich Drepper | 1 | -11/+12 | |
2011-10-17 | Add optimized 64-bit frexp | Ulrich Drepper | 1 | -0/+67 | |
2011-10-17 | Relax asm requirements for recently added x86-64 math interfaces | Ulrich Drepper | 2 | -10/+10 | |
2011-10-17 | Makr x86-64 math_private.h more robust | Ulrich Drepper | 1 | -10/+10 | |
2011-10-17 | Provide internal optimizations on x86-64 with SSE4.1 | Ulrich Drepper | 6 | -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. | |||||
2011-10-17 | Fix last x86-64 mathinline change | Ulrich Drepper | 1 | -2/+2 | |
Use correct function names. |