Commit Graph

  • ba31c8f5f9
    Merge pull request #2853 from Qiyu8/usimd-daxpy Martin Kroeker 2020-09-27 23:19:59 +0200
  • e961d4d609
    Merge pull request #2864 from martin-frbg/lapack445 Martin Kroeker 2020-09-27 23:11:17 +0200
  • 7ed25e9e10
    FIx underflow/rounding errors in LAPACK (S,D)LANV2 Martin Kroeker 2020-09-27 22:59:20 +0200
  • 7b169379e0
    Merge pull request #2863 from martin-frbg/readmefixes Martin Kroeker 2020-09-27 22:50:25 +0200
  • 7f539fb850
    Update cpu list, outline cmake build, clarify scope of set_num_threads extension Martin Kroeker 2020-09-27 22:48:41 +0200
  • caf7a12295
    Merge pull request #90 from xianyi/develop Martin Kroeker 2020-09-27 22:35:45 +0200
  • 72b5b73647
    Merge pull request #2850 from xiaojiayuan111/develop Martin Kroeker 2020-09-27 12:12:35 +0200
  • 881c15179f remove default support for FMA4 on zen architect Qiyu8 2020-09-27 09:35:50 +0800
  • 896bbd55e1
    Add support for building only selected variable types Martin Kroeker 2020-09-26 23:25:55 +0200
  • c5a32288c6
    Work around sgemm_r/dgemm_r not being properly defined with BUILD_COMPLEX/BUILD_COMPLEX16 Martin Kroeker 2020-09-26 23:24:37 +0200
  • dfaafd3b55
    Merge pull request #2854 from martin-frbg/travis-graviton Martin Kroeker 2020-09-23 21:59:18 +0200
  • f2e9a24e1a
    Add AWS Graviton2 build Martin Kroeker 2020-09-23 19:02:20 +0200
  • 98153875e9
    Adapt tests to having only a subset of types in the library Martin Kroeker 2020-09-22 23:28:57 +0200
  • 0eaae30e8c
    Adapt tests to having only a subset of types in the build Martin Kroeker 2020-09-22 23:28:03 +0200
  • dfbc62ef7e
    Support building only a subset of types Martin Kroeker 2020-09-22 23:25:59 +0200
  • b475b4bd0d
    Support building only a subset of types Martin Kroeker 2020-09-22 23:25:04 +0200
  • 357bff06b5
    Add BUILD_vartype defines Martin Kroeker 2020-09-22 23:24:22 +0200
  • 988a6f429e
    Add BUILD_vartype defines Martin Kroeker 2020-09-22 23:23:33 +0200
  • e5e2fbd593
    Support building only selected types Martin Kroeker 2020-09-22 23:21:30 +0200
  • 3287848c8f
    Support building only seleced types Martin Kroeker 2020-09-22 23:20:51 +0200
  • 26611af8e1
    fix grouping of sources used for more than one type Martin Kroeker 2020-09-22 23:20:05 +0200
  • b886bd672b
    add defines for building a subset of types Martin Kroeker 2020-09-22 23:18:55 +0200
  • 61fae59298
    Merge pull request #88 from xianyi/develop Martin Kroeker 2020-09-22 23:15:33 +0200
  • 33d22f99f1
    Merge pull request #2851 from martin-frbg/travis-xcode12 Martin Kroeker 2020-09-22 21:44:55 +0200
  • 5ba01dd1a8
    Add an OSX build with xcode12 Martin Kroeker 2020-09-22 17:26:19 +0200
  • 14f7dad3b7 performance improved Qiyu8 2020-09-22 16:52:15 +0800
  • 06cf73a239 fix a bug of trmm y00512012 2020-09-22 16:47:10 +0800
  • ebe64a3c03 fix a bug of trmm y00512012 2020-09-22 15:39:59 +0800
  • 325b539c26 Optimize the performance of daxpy by using universal intrinsics Qiyu8 2020-09-22 10:38:35 +0800
  • 0f112077e6
    Merge pull request #2847 from mhillenibm/fixup_cscal Martin Kroeker 2020-09-21 22:22:43 +0200
  • 22aa81f3e5 s390x: fix cscal and zscal implementations Marius Hillenbrand 2020-09-14 18:36:31 +0200
  • 77ea73f5e5 s390x: for clang use fp-contract=on instead of fast Marius Hillenbrand 2020-09-16 15:55:38 +0200
  • f91057cbad s390x: move common vector definitions and utils into header Marius Hillenbrand 2020-09-15 10:54:37 +0200
  • 992d7ca63d
    Merge pull request #2845 from martin-frbg/lapack443 Martin Kroeker 2020-09-18 23:18:41 +0200
  • 7e4d5c237c
    Fix workspace query in xGELQ (Reference-LAPACK PR443) Martin Kroeker 2020-09-18 09:19:46 +0200
  • 8d12027a79
    Merge pull request #86 from xianyi/develop Martin Kroeker 2020-09-18 09:17:49 +0200
  • b1e0bcceec
    Merge pull request #2844 from RajalakshmiSR/daxpy_p10 Martin Kroeker 2020-09-17 23:46:32 +0200
  • be43d2cb96 Optimize daxpy/zaxpy for POWER10 Rajalakshmi Srinivasaraghavan 2020-09-17 12:56:28 -0500
  • 2855e6000c
    Merge pull request #2841 from martin-frbg/cpp_gemvtest Martin Kroeker 2020-09-17 17:29:56 +0200
  • 144a03446d
    Merge pull request #2843 from mhillenibm/fixup_merge_dynamic_zarch Martin Kroeker 2020-09-17 17:28:43 +0200
  • 75d440caa0 s390x/DYNAMIC_ARCH: fixup broken merge and reapply simplification Marius Hillenbrand 2020-09-17 16:45:07 +0200
  • 6abca76c4e
    Add option for running only the less demanding GEMV version of the thread safety tests Martin Kroeker 2020-09-17 13:49:24 +0200
  • 84c00c3c6e
    Support running just the GEMV version of the thread safety test Martin Kroeker 2020-09-17 13:46:41 +0200
  • 8c5c991bd7
    Add cpp_thread_test options Martin Kroeker 2020-09-17 13:45:40 +0200
  • 2e3b15d68b
    Add CMakeLists.txt Martin Kroeker 2020-09-17 13:43:55 +0200
  • eaf7f825bd
    Merge pull request #85 from xianyi/develop Martin Kroeker 2020-09-17 13:42:47 +0200
  • 4c10a1673d
    Merge pull request #2840 from martin-frbg/fixup2833 Martin Kroeker 2020-09-16 18:55:50 +0200
  • c4aeeeb9f4
    Activate all BUILD_ options if none was specified Martin Kroeker 2020-09-15 23:15:34 +0200
  • 3843bd188c
    Merge pull request #84 from xianyi/develop Martin Kroeker 2020-09-15 23:13:30 +0200
  • ddec244a5a
    Merge pull request #2838 from austinpagan/gordon_trmm Martin Kroeker 2020-09-15 21:17:48 +0200
  • dfeca46098 Adding performance patch for trmm, just like #2836 fossum 2020-09-15 08:59:50 -0500
  • f8950f40a2
    Merge pull request #2836 from austinpagan/gordon_trsm Martin Kroeker 2020-09-15 11:26:37 +0200
  • 274d6e015b Fixing a performance bug in trsm_[LR].c. fossum 2020-09-14 13:10:48 -0500
  • 91c84e1c01
    Merge pull request #2796 from Guobing-Chen/BF16_dot_coversion_apis Martin Kroeker 2020-09-14 15:00:19 +0200
  • 1ee1e7b495
    Merge pull request #2833 from martin-frbg/issue2830 Martin Kroeker 2020-09-14 07:24:23 +0200
  • ba644378dc
    Copy BUILD_ options available to the compiler flags Martin Kroeker 2020-09-14 00:03:33 +0200
  • 9e11c2d62f
    Add BUILD_SINGLE etc Martin Kroeker 2020-09-13 23:55:11 +0200
  • 4d250d0cdf
    Rearrange ifdefs Martin Kroeker 2020-09-13 23:29:01 +0200
  • de139337b8
    Remove spurious tests for complex ASUM and NRM2 Martin Kroeker 2020-09-13 22:20:41 +0200
  • ec2948f147
    Make tests conditional on BUILD_DOUBLE Martin Kroeker 2020-09-13 22:17:46 +0200
  • ce89398636
    Make tests for individual variable types conditional on the respective BUILD_ option Martin Kroeker 2020-09-13 21:52:18 +0200
  • 593ce9e237
    Make building individual tests depend on BUILD_SINGLE etc defines Martin Kroeker 2020-09-13 21:50:12 +0200
  • 74e358bcd5
    Remove spurious complex16 tests Martin Kroeker 2020-09-13 21:49:01 +0200
  • 26792d2096
    Copy BUILD_* directives to the compiler options to allow ifdef in tests Martin Kroeker 2020-09-13 21:47:55 +0200
  • 6b52c7e172
    Merge pull request #2832 from martin-frbg/issue2831 Martin Kroeker 2020-09-13 21:20:30 +0200
  • 746ad3bd19
    Fix vendor match for GCC gfortran Martin Kroeker 2020-09-13 18:40:59 +0200
  • 55d4d470ec
    Merge pull request #83 from xianyi/develop Martin Kroeker 2020-09-13 18:30:11 +0200
  • a270894730
    Merge pull request #2829 from mhillenibm/clang_s390x Martin Kroeker 2020-09-08 23:36:41 +0200
  • 047b8d7aff Add an s390 build with clang to the Travis configuration Marius Hillenbrand 2020-09-08 19:30:37 +0200
  • f7731a358a Update CONTRIBUTERS.md - clang build fixes for IBM z Marius Hillenbrand 2020-09-08 15:15:15 +0200
  • a55fe06f25 s390x/DYNAMIC_ARCH: define a HW_CAP flag to support slightly older glibc versions Marius Hillenbrand 2020-09-07 17:13:03 +0200
  • 4f34bcfb5e s390x/DYNAMIC_ARCH: pass supported arch levels from Makefile to run-time code Marius Hillenbrand 2020-09-07 17:04:03 +0200
  • 0629d8ebdb s390x/DYNAMIC_ARCH: generalize detecting supported archs for clang Marius Hillenbrand 2020-09-04 16:32:45 +0200
  • 15da2f9acb
    Merge pull request #2828 from martin-frbg/lapack438 Martin Kroeker 2020-09-08 10:25:19 +0200
  • 7d9c77f421
    Correct dimension argument to xLASET Martin Kroeker 2020-09-07 22:03:46 +0200
  • c8f029a518
    Merge pull request #82 from xianyi/develop Martin Kroeker 2020-09-07 21:59:13 +0200
  • e72430fe46
    Merge pull request #2803 from xiegengxin/AVX2-asum Martin Kroeker 2020-09-06 18:32:15 +0200
  • 6e0f6c5f00
    Merge pull request #2824 from martin-frbg/asumbench Martin Kroeker 2020-09-06 10:05:47 +0200
  • 6f8fad87c5
    Use POSIX2001 clock.gettime for higher resolution Martin Kroeker 2020-09-05 19:44:01 +0200
  • ed0f2d3dd7
    Merge pull request #2816 from martin-frbg/silicon Martin Kroeker 2020-09-05 19:17:59 +0200
  • 43a31b7786
    Merge pull request #2823 from martin-frbg/fix2778 Martin Kroeker 2020-09-05 17:29:38 +0200
  • 8a2a137a9e
    Correct argument to SLASET (Improves fix from PR2778) Martin Kroeker 2020-09-05 13:06:31 +0200
  • 0d1f30a297
    Merge pull request #81 from xianyi/develop Martin Kroeker 2020-09-05 12:47:03 +0200
  • 70a254d507
    Merge pull request #2822 from martin-frbg/issue2821 Martin Kroeker 2020-09-05 12:39:32 +0200
  • 330044d821
    Fix potentiol domain error in sqrt Martin Kroeker 2020-09-05 09:44:33 +0200
  • 97636b2c8a
    Merge pull request #2819 from h-vetinari/carry_lapack_437 Martin Kroeker 2020-09-04 23:50:43 +0200
  • 4d36711547
    Merge pull request #2820 from RajalakshmiSR/clang Martin Kroeker 2020-09-04 23:09:31 +0200
  • 718f67421a POWER9: Fix mcpu option with clang Rajalakshmi Srinivasaraghavan 2020-09-04 10:36:19 -0500
  • 3426519ae2 adapt ?ggsv?-functions to ambient code style in LAPACKE/include/lapack.h H. Vetinari 2020-09-02 22:46:47 +0200
  • 1c6c71fa85 Follow-up to lapack#434 & lapack#409: add missing 'const' in signatures H. Vetinari 2020-09-02 22:41:50 +0200
  • 860247b5da Follow-up to lapack#434 & lapack#409: fix signature mismatches H. Vetinari 2020-09-02 22:38:56 +0200
  • c61771e335
    Merge pull request #2778 from martin-frbg/lapackeig Martin Kroeker 2020-09-04 10:06:02 +0200
  • deaeb6c5b8 Add bfloat16 based dot and conversion with single/double Chen, Guobing 2020-08-27 06:42:28 +0800
  • c7ef7174e4
    Merge pull request #2817 from martin-frbg/lapack436 Martin Kroeker 2020-09-03 17:10:23 +0200
  • 775a87242d
    Rename KERNEL.SILICON to KERNEL.VORTEX Martin Kroeker 2020-09-03 08:44:20 +0200
  • af5bc95503
    Rename SILICON to VORTEX and fix duplicate numbering Martin Kroeker 2020-09-03 08:43:26 +0200
  • ea3a58c844
    Rename SILICON to VORTEX Martin Kroeker 2020-09-03 08:38:53 +0200
  • 17dca035de
    rename SILICON to VORTEX Martin Kroeker 2020-09-03 08:38:08 +0200
  • 1b0f17eeed align to 64, using SSE when input size is small Gengxin Xie 2020-09-01 15:41:48 +0800
  • c31b72965e
    Fix data type of work array in zgesvdq prototype Martin Kroeker 2020-09-02 23:44:44 +0200