Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Disable use of FMA instructions in branred | Andreas Schwab | 2012-03-21 | 4 | -8/+1 |
| | |||||
* | Remove inaccurate x86 cexp implementations (bug 13883). | Joseph Myers | 2012-03-21 | 1 | -0/+22 |
| | |||||
* | Fix pow of zero and infinity to large powers. | Joseph Myers | 2012-03-21 | 1 | -0/+41 |
| | |||||
* | Check __x86_64__ instead of __WORDSIZE in mathinline.h | H.J. Lu | 2012-03-20 | 1 | -5/+3 |
| | |||||
* | Support x86-64 __jmp_buf with __WORDSIZE != 64 | H.J. Lu | 2012-03-20 | 1 | -0/+2 |
| | |||||
* | Use atomic64_t with 64bit atomic macros | H.J. Lu | 2012-03-19 | 1 | -7/+7 |
| | |||||
* | Check __x86_64__ instead of __WORDSIZE for fenv_t | H.J. Lu | 2012-03-19 | 1 | -4/+1 |
| | |||||
* | Check __x86_64__ instead of __WORDSIZE in mathdef.h | H.J. Lu | 2012-03-19 | 1 | -4/+2 |
| | |||||
* | Use int64_t in x86_64/fpu/math_private.h | H.J. Lu | 2012-03-19 | 1 | -2/+2 |
| | |||||
* | Cast _Unwind_GetCFA return to _Unwind_Ptr first | H.J. Lu | 2012-03-19 | 1 | -1/+3 |
| | |||||
* | Fix clog overflow/underflow (bug 13629). | Joseph Myers | 2012-03-19 | 1 | -0/+67 |
| | |||||
* | Optimize private 387 fenv access; share code between i386 and x86_64. | Richard Henderson | 2012-03-19 | 1 | -119/+1 |
| | |||||
* | i386/x86_64: Optimize feholdexcept. | Richard Henderson | 2012-03-19 | 1 | -14/+7 |
| | |||||
* | Create and use libc_feupdateenv_test. | Richard Henderson | 2012-03-19 | 1 | -4/+24 |
| | | | | | We can reduce the number of STMXCSR, and often we can avoid the call to __feraiseexcept. | ||||
* | Create and use SET_RESTORE_ROUND{,_NOEX,_53BIT}{,F,L}. | Richard Henderson | 2012-03-19 | 1 | -0/+23 |
| | |||||
* | Convert libc_feholdexcept et al from macros to inline functions. | Richard Henderson | 2012-03-19 | 1 | -57/+57 |
| | |||||
* | Make inline __isnan, __isinf_ns, __finite generic. | Richard Henderson | 2012-03-19 | 1 | -27/+1 |
| | | | | | | For code generation to stay identical on x86_64, this requires that we define the fp word manipulation macros before including the generic header. | ||||
* | Define x86-64 ffsl alias only if __LP64__ is defined | H.J. Lu | 2012-03-16 | 1 | -1/+3 |
| | |||||
* | Disable Bessel function TLOSS errors in POSIX mode. | Joseph Myers | 2012-03-16 | 1 | -2/+39 |
| | |||||
* | Fix tan, tanl for large inputs. | Joseph Myers | 2012-03-16 | 3 | -46/+28 |
| | |||||
* | * sysdeps/x86_64/elf/start.S: Include <sysdep.h>. | Jan Kratochvil | 2012-03-16 | 1 | -1/+6 |
| | | | | (_start): Add cfi_startproc, cfi_undefined for rip and cfi_endproc. | ||||
* | Implement ldbl-96 sinl / cosl / sincosl (bug 13851). | Joseph Myers | 2012-03-16 | 7 | -158/+40 |
| | |||||
* | Merge branch 'master' into bug13658-branch | Andreas Jaeger | 2012-03-14 | 4 | -77/+418 |
|\ | |||||
| * | Fix csqrt overflow/underflow (bug 13841). | Joseph Myers | 2012-03-14 | 1 | -0/+39 |
| | | |||||
| * | Replace FSF snail mail address with URL in miscellaneous files. | Paul Eggert | 2012-03-10 | 1 | -3/+2 |
| | | |||||
| * | Use include_next to chain math_private.h headers. | Richard Henderson | 2012-03-09 | 1 | -1/+1 |
| | | |||||
| * | x86_64: Convert __rint* and __floor* from macros to inlines. | Richard Henderson | 2012-03-09 | 1 | -49/+43 |
| | | |||||
| * | x86_64: Convert __ieee754_sqrt{,f,l} from macros to inlines. | Richard Henderson | 2012-03-09 | 1 | -22/+32 |
| | | |||||
| * | Fix signs of zeros from casinh, cacosh etc. (bug 10716). | Joseph Myers | 2012-03-07 | 1 | -0/+298 |
| | | |||||
| * | Fix up long double fphex. | Marek Polacek | 2012-03-06 | 1 | -1/+2 |
| | | |||||
* | | [BZ #13658] | Andreas Jaeger | 2012-03-07 | 1 | -60/+0 |
|/ | | | | | | * sysdeps/x86_64/fpu/s_sincos.S: Delete. * math/libm-test.inc (sincos_test): Add test for large input. | ||||
* | Fix pow in non-default rounding modes (bug 3976). | Joseph Myers | 2012-03-05 | 1 | -0/+48 |
| | |||||
* | Test cosh, sinh in non-default rounding modes (bug 3976). | Joseph Myers | 2012-03-05 | 1 | -0/+136 |
| | |||||
* | Remove oldish __GNUC_PREREQ. | Marek Polacek | 2012-03-03 | 1 | -40/+4 |
| | |||||
* | Fix sin, cos, tan in non-default rounding modes (bug 3976). | Joseph Myers | 2012-03-02 | 1 | -0/+378 |
| | |||||
* | Fix exp in non-default rounding modes (bug 3976). | Joseph Myers | 2012-03-02 | 1 | -0/+51 |
| | |||||
* | Improve erfc accuracy. | Joseph Myers | 2012-03-01 | 1 | -0/+12 |
| | |||||
* | Add test for bug 5794 (incorrect expm1 overflow). | Joseph Myers | 2012-02-29 | 1 | -0/+5 |
| | |||||
* | Reduce large expected errors from libm tests on x86 and x86_64. | Joseph Myers | 2012-02-29 | 1 | -12/+12 |
| | |||||
* | Optimized expf for x86-64 | Ulrich Drepper | 2012-02-28 | 1 | -0/+340 |
| | |||||
* | Resort ULPs files with gen-libm-test.pl -n in C locale. | Joseph Myers | 2012-02-20 | 1 | -2/+3 |
| | |||||
* | Provide crt[in].S for x86-64. | Marek Polacek | 2012-02-14 | 2 | -70/+88 |
| | |||||
* | Replace FSF snail mail address with URLs. | Paul Eggert | 2012-02-09 | 173 | -520/+346 |
| | |||||
* | Reduce ldouble ULPs for jn tests on x86 | Andreas Schwab | 2012-02-08 | 1 | -18/+16 |
| | |||||
* | Remove __ELF__ conditionals | Marek Polacek | 2012-02-07 | 9 | -44/+11 |
| | |||||
* | Fix makefile/configure problems with sse2avx changes. | Joseph Myers | 2012-01-30 | 1 | -1/+1 |
| | |||||
* | Optimize x86-64 math inline header a bit | Ulrich Drepper | 2012-01-28 | 1 | -8/+31 |
| | |||||
* | Use -msse2avx option for x86-64 libm functions | Ulrich Drepper | 2012-01-28 | 8 | -49/+26 |
| | |||||
* | Simplify use of AVX instructions in internal math macros | Ulrich Drepper | 2012-01-28 | 1 | -59/+20 |
| | |||||
* | Really fix AVX tests | Ulrich Drepper | 2012-01-26 | 8 | -27/+31 |
| | | | | | | 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. |