Commit Graph

  • *
  • | *
  • | |\
  • | * \
  • | |\ \
  • | * \ \
  • | |\ \ \
  • | * \ \ \
  • | |\ \ \ \
  • | * | | | |
  • |/| | | | |
  • | | | | | *
  • | * | | | |
  • |/ / / / /
  • | | | | *
  • | | | | *
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | | * |
  • | | * | |
  • | | * | |
  • | | | * |
  • | | * | |
  • | | * | |
  • * | | | |
  • |\ \ \ \ \
  • | * | | | |
  • |/ / / / /
  • | | * | |
  • | | * | |
  • | | | |/
  • | | |/|
  • * | | |
  • |\ \ \ \
  • | |_|/ /
  • |/| | |
  • | * | |
  • |/ / /
  • * | |
  • |\ \ \
  • | | * |
  • | |/ /
  • |/| |
  • | * |
  • |/ /
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | | | *
  • | | | *
  • | | |/
  • * | |
  • |\ \ \
  • * \ \ \
  • |\ \ \ \
  • * \ \ \ \
  • |\ \ \ \ \
  • * \ \ \ \ \
  • |\ \ \ \ \ \
  • | | | | * | |
  • | | | | * | |
  • | | | | | |/
  • | | | | |/|
  • | | | * / |
  • | | | |/ /
  • | | | | | *
  • | | | | | *
  • | | | | |/
  • | | | |/|
  • | | * | |
  • | * | | |
  • | | * | |
  • | | |/ /
  • | * | |
  • | * | |
  • | * | |
  • | * | |
  • | * | |
  • | * | |
  • | |/ /
  • | * |
  • | |\ \
  • | |/ /
  • |/| |
  • * | |
  • |\ \ \
  • * \ \ \
  • |\ \ \ \
  • * \ \ \ \
  • |\ \ \ \ \
  • * \ \ \ \ \
  • |\ \ \ \ \ \
  • | * \ \ \ \ \
  • | |\ \ \ \ \ \
  • | |/ / / / / /
  • |/| | | | | |
  • | * | | | | |
  • | | | | | | | *
  • | |_|_|_|_|_|/
  • |/| | | | | |
  • * | | | | | |
  • |\ \ \ \ \ \ \
  • | | | * | | | |
  • | |_|/ / / / /
  • |/| | | | | |
  • | * | | | | |
  • |/ / / / / /
  • | | | | | | *
  • | | | | | | *
  • | |_|_|_|_|/
  • |/| | | | |
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • | | | | | | *
  • * | | | | | |
  • |\ \ \ \ \ \ \
  • * \ \ \ \ \ \ \
  • |\ \ \ \ \ \ \ \
  • | * | | | | | | |
  • | | | | | | | | | *
  • | | | | | | | |_|/
  • | | | | | | |/| |
  • * | | | | | | | |
  • |\ \ \ \ \ \ \ \ \
  • | |_|_|_|_|_|_|/ /
  • |/| | | | | | | |
  • | * | | | | | | |
  • | | |_|_|_|_|/ /
  • | |/| | | | | |
  • | * | | | | | |
  • | |\ \ \ \ \ \ \
  • | |/ / / / / / /
  • |/| | | | | | |
  • | | | | | * | |
  • | |_|_|_|/ / /
  • |/| | | | | |
  • | | * | | | |
  • | | * | | | |
  • | | * | | | |
  • | | * | | | |
  • | | * | | | |
  • | | * | | | |
  • | |/ / / / /
  • |/| | | | |
  • | | | | | | *
  • | |_|_|_|_|/
  • |/| | | | |
  • | | * | | |
  • | |/ / / /
  • |/| | | |
  • | | * | |
  • | | |\ \ \
  • 213c0e7ab (refs/pull/3021/head) Added special unrolled vectorized versions of "Solve" for specific sizes, in DTRSM and STRSM, to improve performance in Power9 and Power10. by Gordon Fossum 2020-12-04 17:07:06 -0600
  • f21618684 Merge pull request #3018 from martin-frbg/issue3015 by Martin Kroeker 2020-12-04 22:08:17 +0100
  • 441c08c9f Merge pull request #3016 from xiegengxin/complex-asum by Martin Kroeker 2020-12-04 22:07:16 +0100
  • 66302b3c0 Merge pull request #3013 from martin-frbg/gcc46 by Martin Kroeker 2020-12-04 08:54:11 +0100
  • 07e9a1234 Merge pull request #3011 from cyyever/fix_link by Martin Kroeker 2020-12-04 08:50:59 +0100
  • dd1adbdec Merge pull request #3019 from RajalakshmiSR/dgemm_param by Martin Kroeker 2020-12-04 08:49:28 +0100
  • a1eecccda (refs/pull/3018/head) Update f_check by Martin Kroeker 2020-12-03 23:43:17 +0100
  • 41fe6e864 (refs/pull/3019/head) POWER10: Update param.h by Rajalakshmi Srinivasaraghavan 2020-12-03 14:40:11 -0600
  • 74b585058 Add libomp to the LAPACK(-test) dependencies in clang/gfortran builds by Martin Kroeker 2020-12-03 21:28:10 +0100
  • da0c94c76 Avoid linking both GNU libgomp and LLVM libomp in clang/gfortran builds by Martin Kroeker 2020-12-03 21:25:57 +0100
  • a6692dc12 (refs/pull/3013/head) use gfortran-10 with xcode 12 by Martin Kroeker 2020-12-03 14:32:21 +0100
  • 72a553f5b Update .travis.yml by Martin Kroeker 2020-12-03 09:17:27 +0100
  • dcbb3b5ef fix misplaced lines by Martin Kroeker 2020-12-02 23:13:13 +0100
  • 57456c248 fix gfortran requirement in osx interface64 test by Martin Kroeker 2020-12-02 15:56:21 +0100
  • c36131356 Disable deprecated 32bit xcode by Martin Kroeker 2020-12-02 07:49:43 +0100
  • 0cb7a403b (refs/pull/3016/head) fix error declare function blas_level1_thread_with_return_value by Gengxin Xie 2020-12-02 09:51:52 +0800
  • 77a538d4b Update an overlooked instance of xcode 10.0 as well by Martin Kroeker 2020-12-01 22:05:35 +0100
  • 9621062eb Update OSX xcode version to 11.5 by Martin Kroeker 2020-12-01 12:23:30 +0100
  • b766c1e9b Improve the performance of zasum and casum with AVX512 intrinsic by Gengxin Xie 2020-12-01 16:49:26 +0800
  • 22574b474 Suppress -mfma as well for gcc 4.6 by Martin Kroeker 2020-11-30 21:41:51 +0100
  • f66202299 Move the version check to avoid overwriting unprocessed compiler data by Martin Kroeker 2020-11-30 17:24:27 +0100
  • 5e81e8147 Merge pull request #3014 from RajalakshmiSR/dgemvnp10 by Martin Kroeker 2020-11-30 08:18:24 +0100
  • 7d46e31de (refs/pull/3014/head) POWER10: Optimize dgemv_n by Rajalakshmi Srinivasaraghavan 2020-11-29 15:28:28 -0600
  • 62a2eb884 Add SSE flags for x86 by Martin Kroeker 2020-11-29 15:33:07 +0100
  • 2e99e2699 Add workaround for gcc 4.6 miscompiling assembly kernels with -mavx by Martin Kroeker 2020-11-29 15:32:17 +0100
  • 006b13299 Merge pull request #3012 from martin-frbg/restore-getarch by Martin Kroeker 2020-11-29 13:27:47 +0100
  • ca17d3dc3 (refs/pull/3012/head) Restore RISCV entries accidentally trashed by my PR 3005 by Martin Kroeker 2020-11-29 13:19:51 +0100
  • 52ed2741c Merge pull request #3010 from ggouaillardet/topic/fj_compilers by Martin Kroeker 2020-11-29 11:36:43 +0100
  • 3b4c01611 (refs/pull/3011/head) link math lib on FreeBSD by cyy 2020-11-29 17:17:07 +0800
  • 358100ec1 (refs/pull/3010/head) add Fujitsu compilers by Gilles Gouaillardet 2020-11-29 13:57:57 +0900
  • 3e6d10612 (refs/pull/3008/head) Do not pass -mavx for gcc 4.6 by Martin Kroeker 2020-11-28 23:22:26 +0100
  • 24c52ff34 Add -msse2 by Martin Kroeker 2020-11-27 19:45:56 +0100
  • 6e8439143 remove DYNAMIC_ARCH restriction on -msse3 by Martin Kroeker 2020-11-27 13:30:38 +0100
  • 30db556da export NO_AVX2 by Martin Kroeker 2020-11-27 10:39:04 +0100
  • c903518e8 Downgrade HASWELL/ZEN targets to SANDYBRIDGE if no AVX2 support by Martin Kroeker 2020-11-27 10:07:53 +0100
  • ca793f6db Make -mavx2 -mfma conditional on compiler support by Martin Kroeker 2020-11-27 10:05:47 +0100
  • 18a5520a3 Add check for pre-AVX2 gcc versions on x86 by Martin Kroeker 2020-11-27 10:04:45 +0100
  • 953c4ae1a (refs/pull/3007/head) remove quiet to debug piledriver build failure by Martin Kroeker 2020-11-23 17:07:24 +0100
  • 9fb80b9e4 try to update the ancient binutils in Ubuntu Precise for fma support by Martin Kroeker 2020-11-23 14:57:36 +0100
  • 3788b6d15 Merge pull request #3005 from martin-frbg/ssefix by Martin Kroeker 2020-11-23 08:35:32 +0100
  • bc5b1ddf0 Merge pull request #3004 from martin-frbg/bsd_getauxval by Martin Kroeker 2020-11-23 08:35:12 +0100
  • 2f42d2310 Merge pull request #3002 from martin-frbg/issue3000 by Martin Kroeker 2020-11-22 22:51:26 +0100
  • b72dd007d Merge pull request #3001 from martin-frbg/issue2996 by Martin Kroeker 2020-11-22 22:50:41 +0100
  • 11ebe5fa2 (refs/pull/3005/head) Avoid redefinition warning by Martin Kroeker 2020-11-22 21:16:07 +0100
  • 01f01dae9 Add -msse if supported by Martin Kroeker 2020-11-22 21:15:08 +0100
  • e7bf8ced6 (refs/pull/3004/head) Build fix for systems that do not support getauxval by Martin Kroeker 2020-11-22 20:20:28 +0100
  • 5df09f845 (refs/pull/3003/head) define inf if needed by Martin Kroeker 2020-11-22 19:35:43 +0100
  • c38bb5d51 Add utest for NRM2 behaviour with an inf value in the input by Martin Kroeker 2020-11-22 19:08:40 +0100
  • 025629492 (refs/pull/3002/head) Fix syntax mixup by Martin Kroeker 2020-11-22 17:41:44 +0100
  • 2b114c3f3 (refs/pull/3001/head) Restore proper Makefile by Martin Kroeker 2020-11-22 17:16:22 +0100
  • 60e1fddca Ensure that the same (large) BUFFERSIZE is used for all cpus in DYNAMIC_ARCH builds by Martin Kroeker 2020-11-22 16:48:22 +0100
  • ebb878869 Use ifneq instead of ifdef for CROSS option by Martin Kroeker 2020-11-22 16:33:34 +0100
  • 857afcc41 Use ifeq instead of ifdef for user-definable build options by Martin Kroeker 2020-11-22 16:31:44 +0100
  • 5fa305172 Use ifeq instead of ifdef for user-definable options by Martin Kroeker 2020-11-22 16:29:56 +0100
  • d3ff1f889 Convert ifndefs to ifneq by Martin Kroeker 2020-11-22 16:27:17 +0100
  • 65eb7afaf Change ifndef CROSS to ifneq by Martin Kroeker 2020-11-22 16:25:36 +0100
  • 8a6b17f97 Change ifndefs to ifneq by Martin Kroeker 2020-11-22 16:19:31 +0100
  • 0f863f96e Merge pull request #112 from xianyi/develop by Martin Kroeker 2020-11-22 16:17:19 +0100
  • 437702e0e Merge pull request #2965 from epsilon-0/develop by Martin Kroeker 2020-11-22 12:25:33 +0100
  • f1bf040b2 Merge pull request #2988 from xiegengxin/smp-asum by Martin Kroeker 2020-11-22 12:24:13 +0100
  • 613e3b2ba Merge pull request #2997 from Flamefire/reproduce_crash by Martin Kroeker 2020-11-22 12:22:57 +0100
  • 05a0ea234 Merge branch 'risc-v' into develop by Xianyi Zhang 2020-11-22 16:05:32 +0800
  • 703784949 Merge branch 'develop' into risc-v by Xianyi Zhang 2020-11-22 16:04:50 +0800
  • c6c9c24d1 Update doc for C910. by Xianyi Zhang 2020-11-22 16:02:19 +0800
  • bed01f47c (refs/pull/2999/head) Cast arguments of `_mm512_abs_pd` to `__m512` by Mosè Giordano 2020-11-21 15:02:59 +0000
  • 6dd71af0c Merge pull request #2995 from Flamefire/fix_thread_buffer_init by Martin Kroeker 2020-11-20 09:42:10 +0100
  • a05dc6e62 (refs/pull/2997/head) Add reproducer test for crash after fork by Alexander Grund 2020-11-19 15:24:57 +0100
  • 60005eb47 (refs/pull/2995/head) Don't overwrite blas_thread_buffer if already set by Alexander Grund 2020-11-19 14:39:00 +0100
  • 043f3d6fa (refs/pull/2994/head) POWER10: Use POWER9 as a fallback by Anton Blanchard 2020-11-19 21:04:10 +1100
  • fdf71d66b POWER10: Fix ld version detection by Anton Blanchard 2020-11-19 20:50:42 +1100
  • 8917203eb (refs/pull/2992/head) Update common_thread.h by Martin Kroeker 2020-11-17 20:57:16 +0100
  • 1592c1f70 Compare environment variables for NUM_THREADS against compile-time maximum by Martin Kroeker 2020-11-17 19:21:12 +0100
  • 4639c9ae4 Update common_thread.h by Martin Kroeker 2020-11-17 18:49:59 +0100
  • 811629963 Handle runtime OMP thread count exceeding build-time NUM_THREADS by Martin Kroeker 2020-11-17 18:18:35 +0100
  • 26ce2705f reduce num_threads by Martin Kroeker 2020-11-17 17:56:10 +0100
  • cfe35efbf activate testcase by Martin Kroeker 2020-11-17 15:43:42 +0100
  • 906b23638 typo by Martin Kroeker 2020-11-17 15:18:04 +0100
  • c8a32d0a9 Add alternative OpenMP thread safety test from old issue 602 by Martin Kroeker 2020-11-17 14:47:51 +0100
  • 1748f40cb Add testcase from issue 602 by Martin Kroeker 2020-11-17 14:45:20 +0100
  • e607d8de1 Add C version of testcase from issue 602 by Martin Kroeker 2020-11-17 14:43:26 +0100
  • c1f52d358 Add original testcase from issue 602 by Martin Kroeker 2020-11-17 14:42:15 +0100
  • eead529d3 Create test_dgemm_f90.f by Martin Kroeker 2020-11-17 14:41:29 +0100
  • 4293b4b65 (refs/pull/2974/head) Create test_dgemm_omp.c by Martin Kroeker 2020-11-17 00:02:49 +0100
  • 7e9cb39a2 Merge pull request #2981 from Qiyu8/fix-sum by Martin Kroeker 2020-11-16 08:40:46 +0100
  • be075d53c Merge pull request #2983 from Qiyu8/optimize-srot by Martin Kroeker 2020-11-16 08:38:37 +0100
  • b00a0de13 (refs/pull/2983/head) remove the -mfma flag in when the host has AVX. by Qiyu8 2020-11-16 09:14:56 +0800
  • 1425abc27 (refs/pull/2990/head) Reduce the default BUFFERSIZE for x86_64 to its 0.3.9 value by Martin Kroeker 2020-11-15 19:39:18 +0100
  • d341a0fea Merge pull request #2989 from martin-frbg/cmake-fma by Martin Kroeker 2020-11-13 12:35:09 +0100
  • ec4d77c47 (refs/pull/2989/head) Add -mfma for HAVE_FMA3 in the non-DYNAMIC_ARCH case as well by Martin Kroeker 2020-11-13 09:16:34 +0100
  • 02699226d Merge pull request #111 from xianyi/develop by Martin Kroeker 2020-11-13 09:14:23 +0100
  • d6e7e05bb (refs/pull/2988/head) Improve the performance of dasum and sasum when SMP is defined by Gengxin Xie 2020-11-13 14:20:52 +0800
  • ae0b1dea1 modify system.cmake to enable fma flag by Qiyu8 2020-11-13 10:20:24 +0800
  • e0dac6b53 fix the CI failure of target specific option mismatch by Qiyu8 2020-11-12 20:31:03 +0800
  • e5c2ceb67 fix the CI failure of lack the head by Qiyu8 2020-11-12 17:35:17 +0800
  • a87e537b8 modify macro by Qiyu8 2020-11-11 15:53:48 +0800
  • 5bc0a7583 only FMA3 and vector larger than 128 have positive effects. by Qiyu8 2020-11-11 15:18:01 +0800
  • 8c0b206d4 Optimize the performance of rot by using universal intrinsics by Qiyu8 2020-11-11 14:33:12 +0800
  • 7c71f9448 (refs/pull/2982/head) Revert "Lazyly reinit threads after a fork in OMP mode" by Jonathan Ringer 2020-11-10 17:00:28 -0800
  • c4c591ac5 (refs/pull/2981/head) fix sum optimize issues by Qiyu8 2020-11-10 16:16:38 +0800
  • 1ea6cfefd Refs #2899. Merge branch 'damonyu1989-openblas-open-910' into risc-v by Xianyi Zhang 2020-11-10 09:38:43 +0800