Commit Graph

  • 2c552f1074
    Change "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-12 00:11:31 +0200
  • 7ae9e8960e
    Change "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-12 00:08:29 +0200
  • e3a29f6b58
    Change "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-12 00:07:37 +0200
  • 006c7f6671
    Change "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-12 00:06:06 +0200
  • 85154c2e18
    Change "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-12 00:05:05 +0200
  • ae1ab5bfdf
    Change "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-12 00:03:21 +0200
  • 052f31bc3c
    Change "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-12 00:02:16 +0200
  • 3aecafad80
    Change "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-12 00:00:55 +0200
  • 756062afa5
    Rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:56:17 +0200
  • 2061f7fdff
    Rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:54:53 +0200
  • dc8a1afa63
    Rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:53:50 +0200
  • 32733ded04
    Rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:52:45 +0200
  • 3bc8e8c334
    Rename "HALF" and "sh" to "BFLOAT16"and "sb" Martin Kroeker 2020-10-11 23:51:34 +0200
  • 573508f0ee
    Rename common_sh.h to common_sb.h Martin Kroeker 2020-10-11 23:50:54 +0200
  • ca31c32693
    Rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:49:22 +0200
  • 5800758b43
    Rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:44:38 +0200
  • 924fd806d0
    Rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:43:36 +0200
  • 4db09c6cec
    Rename compare_sgemm_shgemm.c to compare_sgemm_sbgemm.c Martin Kroeker 2020-10-11 23:42:45 +0200
  • fd94236042
    Rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:42:07 +0200
  • 68ce719fac
    Rename shdot_microk_cooperlake.c to sbdot_microk_cooperlake.c Martin Kroeker 2020-10-11 23:41:13 +0200
  • d7dd9b396c
    Rename shdot.c to sbdot.c Martin Kroeker 2020-10-11 23:40:43 +0200
  • 9ae80490e0
    rename "HALF" and "sh" to "BFLOAT16" and "sb" Martin Kroeker 2020-10-11 23:39:42 +0200
  • d314d1f49f
    Rename shgemm_kernel_power10.c to sbgemm_kernel_power10.c Martin Kroeker 2020-10-11 23:37:38 +0200
  • f0883740e4
    Merge pull request #96 from xianyi/develop Martin Kroeker 2020-10-11 23:34:36 +0200
  • 1c0b03efb4
    Merge branch 'develop' into develop Martin Kroeker 2020-10-11 23:34:14 +0200
  • c589c3e2a1
    Merge pull request #2882 from martin-frbg/issue2709 Martin Kroeker 2020-10-11 22:22:30 +0200
  • ec638a82bf
    Merge pull request #2852 from martin-frbg/issue2588-cmake Martin Kroeker 2020-10-11 22:21:33 +0200
  • caa0d757ca
    repair TABs Martin Kroeker 2020-10-11 18:29:34 +0200
  • 6154f72d6d
    Copy BUILD_ settings to the LAPACK make.inc Martin Kroeker 2020-10-11 18:25:16 +0200
  • ae8b0d257a
    Set BUILD_ options to 1 instead of just defining them Martin Kroeker 2020-10-11 18:08:21 +0200
  • 1da32cc1fc
    Add cblas_xerbla interface Martin Kroeker 2020-10-11 17:45:41 +0200
  • 8c5e08076e
    If none of the BUILD_ options is set, enable them all Martin Kroeker 2020-10-11 17:33:51 +0200
  • 5f23bdf437
    remove debug output Martin Kroeker 2020-10-11 17:23:08 +0200
  • b593e6b650
    Merge pull request #2885 from martin-frbg/ifexists Martin Kroeker 2020-10-11 15:45:24 +0200
  • 082c86a538
    Merge pull request #2884 from martin-frbg/sse_fixup Martin Kroeker 2020-10-11 15:14:03 +0200
  • e396ec8b56
    Allow building support for only a subset of variable types Martin Kroeker 2020-10-11 15:11:15 +0200
  • 68e6823d36
    Adapt for supporting only a subset of variable types Martin Kroeker 2020-10-11 15:01:32 +0200
  • 887e00fd7f
    Adapt for supporting only a subset of variable types Martin Kroeker 2020-10-11 14:58:57 +0200
  • 886a8e3190
    Adapt for supporting only a subset of variable types Martin Kroeker 2020-10-11 14:57:32 +0200
  • 0f7d73ff6d
    Allow supporting only a subset of variable types Martin Kroeker 2020-10-11 14:53:26 +0200
  • 6b6adf8a4a
    Allow compiling only a subset of kernels for specific variable types Martin Kroeker 2020-10-11 14:52:09 +0200
  • a6570108c5
    Add Makefile support for enabling only some variable types Martin Kroeker 2020-10-11 14:49:58 +0200
  • ef552bc578
    Add Makefile support for enabling only some variable types Martin Kroeker 2020-10-11 14:49:06 +0200
  • efe1ad4700
    Add Makefile support for enabling only some variable types Martin Kroeker 2020-10-11 14:48:23 +0200
  • b27ca78a21
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:46:24 +0200
  • 93454022a9
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:45:40 +0200
  • 20cf1d773f
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:44:56 +0200
  • 5c657fffad
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:44:13 +0200
  • b262058059
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:43:13 +0200
  • bc319cee82
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:42:26 +0200
  • e5966f8606
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:41:43 +0200
  • 9df12eb08f
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:40:51 +0200
  • cf53970bcb
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:40:06 +0200
  • dcd51d5c72
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:39:19 +0200
  • b8f95354c7
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:38:25 +0200
  • d33de97d60
    Adapt to having only a subset of variable types supported Martin Kroeker 2020-10-11 14:36:45 +0200
  • 6a83c591d6
    Adapt for having only a subset of variable types Martin Kroeker 2020-10-11 14:34:12 +0200
  • f6d2827d0c
    Adapt ctests to having only a subset of types in the build Martin Kroeker 2020-10-11 14:32:00 +0200
  • 08f4749eb4
    Adapt tests to having only a subset of types in the build Martin Kroeker 2020-10-11 14:25:24 +0200
  • 63d7dad04c
    Adapt utests for builds supportin only some variable types Martin Kroeker 2020-10-11 14:15:35 +0200
  • ac653c94f3
    Merge branch 'develop' into issue2588-cmake Martin Kroeker 2020-10-11 13:57:07 +0200
  • 190b74dd24
    Add files via upload Martin Kroeker 2020-10-11 13:26:05 +0200
  • 9d43140d61
    Improve check for conflicting config_kernel.h Martin Kroeker 2020-10-11 12:58:17 +0200
  • 8ef600f1a3
    Merge pull request #95 from xianyi/develop Martin Kroeker 2020-10-11 12:53:18 +0200
  • 88928650c4
    Merge pull request #2883 from martin-frbg/issue2872 Martin Kroeker 2020-10-11 10:30:33 +0200
  • 7a53128481
    Add whitelist of DYNAMIC_ARCH kernels for which -msse3 needs to be enabled Martin Kroeker 2020-10-11 01:06:46 +0200
  • 0c773b8205
    Do not rely on HAVE_SSE3 in DYNAMIC_ARCH builds Martin Kroeker 2020-10-11 01:04:57 +0200
  • fbda20c856
    Merge pull request #94 from xianyi/develop Martin Kroeker 2020-10-11 01:03:00 +0200
  • 82a497ec5d
    restore PRESCOTT default for DYNAMIC_LIST Martin Kroeker 2020-10-11 00:43:09 +0200
  • de27e4f5fb
    Stop DYNAMIC_ARCH build if the toplevel source contains a stray config_kernel.h from a gmake build Martin Kroeker 2020-10-11 00:40:22 +0200
  • e1b7123bbe
    Merge pull request #2867 from Qiyu8/usimd-floatdot Martin Kroeker 2020-10-10 12:10:25 +0200
  • f32d34a015 add sse3 compiler flag Qiyu8 2020-10-10 10:36:15 +0800
  • 599777ecb7
    Merge pull request #2879 from martin-frbg/issue2839 Martin Kroeker 2020-10-06 23:26:52 +0200
  • 7812486091
    Use generic C for D/Z nrm2 kernels on Windows to work around fpu exception bug Martin Kroeker 2020-10-06 21:33:16 +0200
  • a5b164946c add fninit to reset fpu registers before assembler routines Matti Picus 2020-10-05 22:13:25 +0300
  • a5feea6611
    make BLAS3_MEM_ALLOC_THRESHOLD configurable on non-Windows Martin Kroeker 2020-10-04 23:01:06 +0200
  • dc8e4e1959
    Reduce the BLAS3 heap allocation threshold to 32 and mark it as configurable Martin Kroeker 2020-10-04 22:59:24 +0200
  • cccd1438da
    Merge pull request #93 from xianyi/develop Martin Kroeker 2020-10-04 22:57:11 +0200
  • f032d8966e
    Merge pull request #2874 from Flamefire/memory_fixes Martin Kroeker 2020-10-04 15:16:51 +0200
  • f6e4cf2f9d
    Merge pull request #2876 from Flamefire/omp_fork_fix Martin Kroeker 2020-10-03 22:52:17 +0200
  • 9828343e12
    Merge pull request #2878 from brada4/asms Martin Kroeker 2020-10-03 22:51:49 +0200
  • d2333e7842 aarch64 fix std=c18 compilation User User-User 2020-10-03 18:00:34 +0300
  • 3094fc6c83
    Lazyly reinit threads after a fork in OMP mode Alexander Grund 2020-10-01 15:41:42 +0200
  • 3c05f54df8
    Avoid out of bounds access on invalid memory free Alexander Grund 2020-10-01 10:48:45 +0200
  • dee7c49938
    Fix TABs and trailing space Alexander Grund 2020-10-01 10:43:16 +0200
  • d3c0d6811b
    Merge pull request #2873 from martin-frbg/issue2871 Martin Kroeker 2020-10-01 06:38:22 +0200
  • 9637cd1fd1
    Merge pull request #2865 from thisch/backticks Martin Kroeker 2020-10-01 06:38:06 +0200
  • 2367726578
    Remove redundant status message Martin Kroeker 2020-09-30 23:28:49 +0200
  • 5464eb13ea
    Change ifdef linux to __linux for C11 compatibility Martin Kroeker 2020-09-30 22:59:41 +0200
  • e1574cbc83
    Change ifdef linux to __linux for C11 compatibility Martin Kroeker 2020-09-30 22:50:21 +0200
  • 0b2bb5696a
    Change ifdef linux to __linux for C11 compatibility Martin Kroeker 2020-09-30 22:47:25 +0200
  • a7d5d0078d
    Change ifdef linux to __linux for C11 compatibility Martin Kroeker 2020-09-30 22:46:25 +0200
  • be40440ec5
    Change ifdef linux to __linux for C11 compatibility Martin Kroeker 2020-09-30 22:45:18 +0200
  • 2bf70c8e3b
    Change ifdef linux to __linux for C11 compatibility Martin Kroeker 2020-09-30 22:43:25 +0200
  • 60e6c68e38 Adapt ARM architect Qiyu8 2020-09-29 16:36:14 +0800
  • 64629cb5c7
    Merge pull request #91 from xianyi/develop Martin Kroeker 2020-09-28 22:48:53 +0200
  • 1b1a757f5f Optimize the performance of dot by using universal intrinsics in X86/ARM Qiyu8 2020-09-28 20:36:53 +0800
  • 0d98ce202c
    Merge pull request #2866 from RajalakshmiSR/p10_dcopy Martin Kroeker 2020-09-28 07:22:54 +0200
  • 2df4235e00 Optimize dcopy/zcopy for POWER10 Rajalakshmi Srinivasaraghavan 2020-09-27 21:42:32 -0500
  • fe8cd5ae7e Consolidate usage of backticks for build options Thomas Hisch 2020-09-28 00:42:17 +0200