Commit Graph

  • b976a0bf40
    Remove previous workaround for compiler flags related to cpu capabilities in x86_64 DYNAMIC_ARCH builds Martin Kroeker 2020-11-07 20:39:56 +0100
  • a04f532edf
    Reset cpu property flags between build cycles in DYNAMIC_ARCH mode Martin Kroeker 2020-11-07 20:37:03 +0100
  • ccb9731c7b
    Fix propagation of cpu properties to compiler options Martin Kroeker 2020-11-07 20:30:15 +0100
  • a29338aaa6
    Remove extraneous quotes that caused a cmake policy warning Martin Kroeker 2020-11-07 20:27:42 +0100
  • 438a8e5624
    Fix placement of getarch call and spurious cpu property accumulation in DYNAMIC_ARCH builds Martin Kroeker 2020-11-07 20:26:12 +0100
  • e5967810b7
    Merge pull request #110 from xianyi/develop Martin Kroeker 2020-11-07 20:22:41 +0100
  • ff74319ea5
    Merge pull request #2977 from martin-frbg/issue2976 Martin Kroeker 2020-11-07 14:41:34 +0100
  • 8257aa8b5a
    Fix macro name used in ifdef Martin Kroeker 2020-11-07 12:20:08 +0100
  • 28d2dfe2b3
    Fix macro name used in ifdef Martin Kroeker 2020-11-07 12:17:49 +0100
  • de932375c7
    use parallel make in the ARM server build&test as well Martin Kroeker 2020-11-05 17:17:17 +0100
  • 40be13d5ea
    Switch the Epyc build to parallel make to catch build races Martin Kroeker 2020-11-05 16:37:54 +0100
  • e27aefcaec
    Add ieee and recursive options for Sun Fortran Martin Kroeker 2020-11-05 16:29:30 +0100
  • 725ffbf041 fix typo Gengxin Xie 2020-11-05 16:25:17 +0800
  • d9ba49165a Improve the performance of rot by using AVX512 and AVX2 intrinsic Gengxin Xie 2020-09-27 10:38:19 +0800
  • 9fd4ad70d7
    rebase and fix solaris options Martin Kroeker 2020-11-04 23:02:50 +0100
  • 60ab9c783f
    Merge pull request #2966 from martin-frbg/issue2964 Martin Kroeker 2020-11-04 16:02:46 +0100
  • 8cc73fee98
    Export NO_EXPRECISION after overriding for DYNAMIC_ARCH with GENERIC target Martin Kroeker 2020-11-03 23:47:04 +0100
  • 0155cd53a3
    Add -msse3 where needed for DYNAMIC_ARCH builds Martin Kroeker 2020-11-03 23:45:49 +0100
  • a9f9354296
    Fix target test Martin Kroeker 2020-11-02 23:17:46 +0100
  • b9bc76aec4
    Add files via upload Martin Kroeker 2020-11-02 22:43:50 +0100
  • f071245939
    Merge pull request #2967 from RajalakshmiSR/dgemm88 Martin Kroeker 2020-11-02 18:54:36 +0100
  • 60997ddd73
    allow setting soname without suffix or prefix Aisha Tammy 2020-11-02 13:04:53 +0000
  • e5f8c2bf8a
    typo fix Martin Kroeker 2020-11-01 22:25:43 +0100
  • 6baf8af658
    Disable EXPRECISION for the combination of DYNAMIC_CORE and GENERIC target Martin Kroeker 2020-11-01 22:11:48 +0100
  • 40a93c232b
    Disable EXPRECISION for DYNAMIC_ARCH in combination with TARGET=GENERIC Martin Kroeker 2020-11-01 21:58:26 +0100
  • fab952bee4
    Merge pull request #2962 from brada4/develop Martin Kroeker 2020-11-01 14:24:40 +0100
  • 1cf04a6f0e
    Merge pull request #2963 from martin-frbg/issue2959 Martin Kroeker 2020-11-01 09:14:54 +0100
  • dd7a9cc5bf POWER10: Change dgemm unroll factors Rajalakshmi Srinivasaraghavan 2020-10-31 18:28:57 -0500
  • 7f26be4802
    Reunify BUFFERSIZE across arm64 platforms to avoid segfaults in DYNAMIC_ARCH Martin Kroeker 2020-11-01 00:00:43 +0100
  • 9fab65e90a add openbsd gfortran User User-User 2020-11-01 00:38:08 +0200
  • 8452ec30a5
    Use generic kernels for dot and c/zswap Martin Kroeker 2020-10-31 22:50:43 +0100
  • 87336eecbc
    Do not use CREAL/CIMAG macros on Sparc Martin Kroeker 2020-10-31 22:48:04 +0100
  • a51a735415
    Add support for SolarisStudio12 compiler and separate x86 and sparc options Martin Kroeker 2020-10-31 22:41:34 +0100
  • d244b271c1
    Increase BUFFER_SIZE to match GEMM parameters Martin Kroeker 2020-10-31 22:39:15 +0100
  • 9666df7054
    Change C++-style comment to C-style Martin Kroeker 2020-10-31 22:38:02 +0100
  • 9efc3f0815
    Merge pull request #109 from xianyi/develop Martin Kroeker 2020-10-31 22:33:52 +0100
  • aa21cb5217
    Merge pull request #2960 from thrasibule/avx2_detection Martin Kroeker 2020-10-31 20:24:21 +0100
  • 1f564d729b fix avx2 detection Guillaume Horel 2020-10-31 10:00:48 -0400
  • 9349dcd206
    Merge pull request #2956 from RajalakshmiSR/caxpy_p10 Martin Kroeker 2020-10-30 08:54:10 +0100
  • b435491885 Optimize caxpy for POWER10 Rajalakshmi Srinivasaraghavan 2020-10-29 14:57:51 -0500
  • 9a058f2451
    Merge pull request #2940 from Qiyu8/optimize-benchmark Martin Kroeker 2020-10-29 20:28:37 +0100
  • 074927a7d0
    Merge pull request #2954 from Guobing-Chen/BF16_gemv_support Martin Kroeker 2020-10-29 09:22:33 +0100
  • 60b22e3462
    Merge pull request #2955 from Guobing-Chen/Fix_cooperlake_build_issue Martin Kroeker 2020-10-29 09:22:07 +0100
  • c5e62dad69 Fix cooperlake compile issue Chen, Guobing 2020-10-29 03:37:51 +0800
  • a7b1f9b1bb Implementation of BF16 based gemv Chen, Guobing 2020-10-28 08:49:12 +0800
  • 67f39ad813
    Merge pull request #2939 from thrasibule/Makefile_cleanup Martin Kroeker 2020-10-28 09:38:40 +0100
  • 6e13a7e99e
    Merge pull request #2951 from martin-frbg/cleanup_make Martin Kroeker 2020-10-28 09:37:56 +0100
  • 2207a16235
    Merge pull request #2952 from martin-frbg/issue2931 Martin Kroeker 2020-10-28 09:37:32 +0100
  • 5d643929dd
    Merge pull request #2948 from martin-frbg/issue2947 Martin Kroeker 2020-10-28 09:37:09 +0100
  • e8cbf0fc50
    Output predefined HAVE_ entries to Makefile.conf for ARM with specified TARGET Martin Kroeker 2020-10-27 23:01:19 +0100
  • b937d78a6d
    Try to read cpu information from /sys/devices/system/cpu/cpu0 if HWCAP_CPUID fails Martin Kroeker 2020-10-27 17:51:32 +0100
  • e2f9005db8
    Merge pull request #2950 from RajalakshmiSR/saxpy Martin Kroeker 2020-10-27 00:02:18 +0100
  • 6a1f3e40af
    Remove debug printout of object list Martin Kroeker 2020-10-26 21:37:04 +0100
  • 878b6d1f41
    Remove spurious expr in flang version check Martin Kroeker 2020-10-26 21:35:40 +0100
  • c24ba8b1dd Optimize saxpy for POWER10 Rajalakshmi Srinivasaraghavan 2020-10-26 13:24:59 -0500
  • f917c26e83 Refractoring remaining benchmark cases. Qiyu8 2020-10-26 10:25:05 +0800
  • 76203e2120
    Merge pull request #2946 from martin-frbg/issue2945 Martin Kroeker 2020-10-26 00:43:44 +0100
  • eec517af0e
    Expressly enable neon for use with intrinsics if available Martin Kroeker 2020-10-26 00:21:56 +0100
  • fd7da56965
    Move definitions that are neither needed nor supported on SUNOS Martin Kroeker 2020-10-25 12:01:50 +0100
  • 2f9fc9be30
    Update version to 0.3.12.dev Martin Kroeker 2020-10-24 23:29:05 +0200
  • 81fcfd5ed3
    Update version to 0.3.12.dev Martin Kroeker 2020-10-24 23:28:29 +0200
  • addf7593ae
    Merge pull request #2944 from xianyi/release-0.3.0 Martin Kroeker 2020-10-24 13:10:51 +0200
  • c5f280a7f0
    Fix typos v0.3.12 Martin Kroeker 2020-10-24 13:03:28 +0200
  • 6e3a05f2c9
    Merge pull request #2943 from xianyi/develop Martin Kroeker 2020-10-24 12:52:59 +0200
  • 89db73569b
    Update Changelog with 0.3.12 changes Martin Kroeker 2020-10-24 12:50:04 +0200
  • e1c18e4eeb
    Update version to 0.3.12 for release Martin Kroeker 2020-10-24 12:15:33 +0200
  • 26f658c9d2
    Update version to 0.3.12 for release Martin Kroeker 2020-10-24 12:14:45 +0200
  • dc35477317
    Merge pull request #2942 from martin-frbg/makebuildtypes Martin Kroeker 2020-10-24 09:26:50 +0200
  • 365f28787c
    Comment out BUILD_SINGLE etc. and add a short explanation Martin Kroeker 2020-10-23 23:32:06 +0200
  • 2f2e9ddb65
    Merge pull request #2941 from martin-frbg/exportsfix Martin Kroeker 2020-10-23 20:47:35 +0200
  • 0d140e61ac
    Fix wrong grouping of dcombssq Martin Kroeker 2020-10-23 15:53:40 +0200
  • 4c45cd6294
    fix missing split of sladiv1/dladiv/ilaenv2stage by build type Martin Kroeker 2020-10-23 15:31:25 +0200
  • 680f744abf
    Merge pull request #108 from xianyi/develop Martin Kroeker 2020-10-23 15:29:48 +0200
  • 6f9460f0f6
    Merge pull request #2937 from martin-frbg/pwr-buffersz Martin Kroeker 2020-10-23 07:15:32 +0200
  • dd6ebdfdab Refactor the performance measurement system Qiyu8 2020-10-23 10:32:03 +0800
  • 1917a4e7b8 reuse variables defined in Makefile.system Guillaume Horel 2020-10-22 22:00:00 -0400
  • 6c970fa998
    Merge pull request #2938 from martin-frbg/2934-3 Martin Kroeker 2020-10-23 00:19:49 +0200
  • b23cb05231
    Fix twisted spelling that broke the gfortran version test again Martin Kroeker 2020-10-23 00:18:29 +0200
  • 1d4c96fa0c
    Increase BUFFERSIZE further Martin Kroeker 2020-10-23 00:12:06 +0200
  • 34c3c407ef
    label always_inline function as inline to silence a gcc warning Martin Kroeker 2020-10-22 22:14:26 +0200
  • 3f84a9ca15
    Merge pull request #2936 from martin-frbg/issue2934-2 Martin Kroeker 2020-10-22 22:08:46 +0200
  • 7e265c50bf
    Merge pull request #2935 from martin-frbg/lapack458 Martin Kroeker 2020-10-22 19:25:58 +0200
  • ee90f30384
    Increase BUFFERSIZE for POWER8-10 and use same value for POWER6 Martin Kroeker 2020-10-22 18:47:07 +0200
  • 2e48d560ba
    Fix compiler version check Martin Kroeker 2020-10-22 16:23:29 +0200
  • ab7f466467
    Merge pull request #106 from xianyi/develop Martin Kroeker 2020-10-22 16:21:09 +0200
  • f95031204e
    Fix macro used in argument conversion (LAPACK PR 458) Martin Kroeker 2020-10-22 16:19:26 +0200
  • 909068facf
    Merge pull request #2932 from RajalakshmiSR/copyp10 Martin Kroeker 2020-10-22 00:29:46 +0200
  • 5b7438fdde
    Merge pull request #2934 from thrasibule/improve_version_check Martin Kroeker 2020-10-22 00:29:02 +0200
  • 47696b43e9 actually check that version is greater than 4.7 Guillaume Horel 2020-10-21 16:42:37 -0400
  • ad745c0bae Optimize scopy/ccopy for POWER10 Rajalakshmi Srinivasaraghavan 2020-10-21 09:53:45 -0500
  • 17c46bf06a
    Merge pull request #2930 from ismail/fix-no-return Martin Kroeker 2020-10-21 11:43:01 +0200
  • 28242096cd
    Merge pull request #2928 from martin-frbg/issue2917 Martin Kroeker 2020-10-21 10:11:02 +0200
  • 4a1d00f589
    Fix build with -Werror=return-type dgemm_tcopy_16_skylakex.c CNAME function should return an int, add a return 0 similar to other files. İsmail Dönmez 2020-10-21 08:43:39 +0200
  • 00813363be
    Enable -mavx2 for flang as well Martin Kroeker 2020-10-20 23:56:30 +0200
  • 336e35469a
    Merge pull request #105 from xianyi/develop Martin Kroeker 2020-10-20 23:48:53 +0200
  • 29668458f7
    Merge pull request #2925 from martin-frbg/issue2911-2 Martin Kroeker 2020-10-20 11:27:36 +0200
  • ee83e29046
    Merge pull request #2926 from bartoldeman/vzeroupper-clobber-all Martin Kroeker 2020-10-20 09:24:47 +0200
  • 1a0f57c8f0
    Fix missing backquotes Martin Kroeker 2020-10-20 08:37:53 +0200
  • b073d759d0 x86_64: clobber all xmm registers after vzeroupper Bart Oldeman 2020-10-20 02:16:47 +0000
  • eddc65c7b7
    Add POWER10 support flag (unconditionally for now) Martin Kroeker 2020-10-20 01:09:49 +0200