glibc.git/benchtests, branch master Unnamed repository; edit this file 'description' to name the repository. benchtests: Fix bench-build after cd748a63ab 2025-11-21T16:22:34+00:00 Adhemerval Zanella adhemerval.zanella@linaro.org 2025-11-21T16:21:25+00:00 aa6066087f5a4d709ff148591d4a58564e2b09bf The benchtests does not define _LIBC. 
The benchtests does not define _LIBC.
bench-malloc-thread: Add libm for powf 2025-11-21T16:13:24+00:00 Adhemerval Zanella adhemerval.zanella@linaro.org 2025-11-20T18:30:15+00:00 52524467afc2398140c65a8ca4c937a085e592b5 The compiler might not constant fold the call, which issues linker error. Reviewed-by: Sam James <sam@gentoo.org> 
The compiler might not constant fold the call, which issues
linker error.

Reviewed-by: Sam James <sam@gentoo.org>
benchtests: Remove clang warnings 2025-11-21T16:13:21+00:00 Adhemerval Zanella adhemerval.zanella@linaro.org 2025-11-20T18:30:14+00:00 4d695077b893c4e65c451e4d7ff91e01c05d4efa clangs warns of the implicit cast of RAND_MAX to float: error: implicit conversion from 'int' to 'float' changes value from 2147483647 to 2147483648 [-Werror,-Wimplicit-const-int-float-conversion] So make it explicit. Reviewed-by: Sam James <sam@gentoo.org> 
clangs warns of the implicit cast of RAND_MAX to float:

  error: implicit conversion from 'int' to 'float' changes value from
  2147483647 to 2147483648 [-Werror,-Wimplicit-const-int-float-conversion]

So make it explicit.

Reviewed-by: Sam James <sam@gentoo.org>
benchtests: Add attribute_optimize 2025-11-21T16:13:15+00:00 Adhemerval Zanella adhemerval.zanella@linaro.org 2025-11-20T18:30:13+00:00 9849a61b34b4693cf3fea32aefd6e3793477ab1e Similar to tst-printf-bz18872.sh, add the attribute_optimize to avoid build failures with compilers that do not support "GCC optimize" pragma. Reviewed-by: Sam James <sam@gentoo.org> 
Similar to tst-printf-bz18872.sh, add the attribute_optimize to avoid
build failures with compilers that do not support "GCC optimize" pragma.

Reviewed-by: Sam James <sam@gentoo.org>
benchtests: Use __f128 on ilogbf128-inputs constants 2025-11-21T16:13:14+00:00 Adhemerval Zanella adhemerval.zanella@linaro.org 2025-11-20T18:30:12+00:00 46be3e521c0fab7e3156e1ab8323128557d7cebc The f128 is not a valid floating constant suffix on clang. Reviewed-by: Sam James <sam@gentoo.org> 
The f128 is not a valid floating constant suffix on clang.

Reviewed-by: Sam James <sam@gentoo.org>
benchtests: Add fmaf benchtests 2025-11-21T16:13:05+00:00 Adhemerval Zanella adhemerval.zanella@linaro.org 2025-11-13T12:58:21+00:00 2885d8a3ab7658033b9933022099ae5e1c3479d4 Random inputs in the range [0,10]. Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com> 
Random inputs in the range [0,10].

Reviewed-by: Wilco Dijkstra  <Wilco.Dijkstra@arm.com>
benchtests: Add fma benchtests 2025-11-21T16:12:58+00:00 Adhemerval Zanella adhemerval.zanella@linaro.org 2025-11-13T12:58:18+00:00 8ce2b5dd272342e2affe7f2245abb43c1398ebae Random inputs in the range [0,10]. Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com> 
Random inputs in the range [0,10].

Reviewed-by: Wilco Dijkstra  <Wilco.Dijkstra@arm.com>
htl: Drop pthread-functions infrastructure 2025-11-13T22:23:13+00:00 Samuel Thibault samuel.thibault@ens-lyon.org 2025-11-13T22:08:34+00:00 23b8e6ae4fb4ccb28fcf92d86dbb351df3a591b7 All previously forwarded functions are now called directly (either via local call in libc, or through a __export).t 
All previously forwarded functions are now called directly (either via local
call in libc, or through a __export).t
hurd: Drop remnants of cthreads 2025-11-12T00:11:11+00:00 Samuel Thibault samuel.thibault@ens-lyon.org 2025-11-11T21:32:39+00:00 f851a7434696b70ce7c266ade1de2469619e6f52 These are not used in GNU/Hurd since very long now. 
These are not used in GNU/Hurd since very long now.
benchtests: Add benchmarks for frexp functions 2025-11-10T11:58:16+00:00 Osama Abdelkader osama.abdelkader@gmail.com 2025-11-04T17:55:05+00:00 ff041e8f8e66371bc13103abdf18fa676b9c214a Add benchmark support for frexp, frexpf, and frexpl to measure the performance improvement of the fast path optimization. - Created frexp-inputs, frexpf-inputs, frexpl-inputs with random test values - Added frexp, frexpf, frexpl to bench-math list - Added CFLAGS to disable builtins for accurate benchmarking These benchmarks will be used to quantify the performance gains from the fast path optimization for normal floating-point numbers. Signed-off-by: Osama Abdelkader <osama.abdelkader@gmail.com> 
Add benchmark support for frexp, frexpf, and frexpl to measure the
performance improvement of the fast path optimization.

- Created frexp-inputs, frexpf-inputs, frexpl-inputs with random test values
- Added frexp, frexpf, frexpl to bench-math list
- Added CFLAGS to disable builtins for accurate benchmarking

These benchmarks will be used to quantify the performance gains from the
fast path optimization for normal floating-point numbers.

Signed-off-by: Osama Abdelkader <osama.abdelkader@gmail.com>