Commit Graph

  • *
  • *
  • *
  • *
  • | *
  • | |\
  • | | *
  • | |/
  • | | *
  • | | *
  • | * |
  • | |\ \
  • | | * |
  • | |/ /
  • | | *
  • | | *
  • | | | *
  • | | |/
  • | |/|
  • | * |
  • | * |
  • | |\ \
  • | | | | *
  • | | | | *
  • | | | | *
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | | * | |
  • | |/ / /
  • | | | *
  • | | | *
  • | | |/
  • | |/|
  • | * |
  • | * |
  • | |\ \
  • | * \ \
  • | |\ \ \
  • | | * | |
  • | |/ / /
  • | * | |
  • | |\ \ \
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | * |
  • | | * | | |
  • | | | |_|/
  • | | |/| |
  • | | | | *
  • | | | | *
  • | | | | *
  • | * | | |
  • | |\ \ \ \
  • | | |/ / /
  • | |/| | |
  • | | | | | *
  • | | |_|_|/
  • | |/| | |
  • | | * | |
  • | |/ / /
  • | | | *
  • | | | *
  • | | | *
  • | | | *
  • | | |/
  • | |/|
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • * | |
  • | | *
  • | | *
  • * | |
  • | | *
  • | |/
  • * |
  • * |
  • * |
  • * |
  • * |
  • * |
  • * |
  • * |
  • * |
  • * |
  • * |
  • | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | |/
  • | | *
  • | | *
  • | | *
  • | | *
  • e665617e9 Update .travis.yml by Martin Kroeker 2021-10-23 23:31:49 +0200
  • 113badd7a Update param.h by Martin Kroeker 2021-10-23 20:49:22 +0200
  • 01c9b25fc Update Makefile.arm64 by Martin Kroeker 2021-10-20 17:30:09 +0200
  • 3e77b3fbf try march=neoverse instead of mtune by Martin Kroeker 2021-10-20 17:06:34 +0200
  • 059d3a04c Merge pull request #3420 from martin-frbg/issue3419 by Martin Kroeker 2021-10-20 12:00:06 +0200
  • 2845f54eb (refs/pull/3420/head) Remove dangerous optimization from previous #3252 - buffer is never unused here by Martin Kroeker 2021-10-20 10:50:02 +0200
  • 70b5169bb Update azure-pipelines.yml by Martin Kroeker 2021-10-20 10:06:03 +0200
  • 264ddaa4c Update azure-pipelines.yml by Martin Kroeker 2021-10-20 09:00:23 +0200
  • c6208bbb4 Merge pull request #3418 from martin-frbg/issue2927-2 by Martin Kroeker 2021-10-20 08:23:53 +0200
  • 6975cbe1f (refs/pull/3418/head) Enable SVE for A64FX by Martin Kroeker 2021-10-19 23:23:40 +0200
  • a1b5196df Update azure-pipelines.yml by Martin Kroeker 2021-10-19 22:59:39 +0200
  • 59d748e63 Update azure-pipelines.yml by Martin Kroeker 2021-10-19 19:32:40 +0200
  • 85f3f64b4 (refs/pull/3417/head) [cmake] Only build for targets newer than TARGET by Mehdi Chinoune 2021-10-18 16:43:05 +0100
  • 22bf5c27b Add basic support for the Fujitsu A64FX (#3415) by Martin Kroeker 2021-10-18 15:00:19 +0200
  • 8cbf61792 Merge pull request #3416 from guowangy/spr-bf16 by Martin Kroeker 2021-10-18 14:59:21 +0200
  • 218b00cac (refs/pull/3415/head) remove stray digit from id by Martin Kroeker 2021-10-18 13:45:02 +0200
  • 4801c5729 rebase to avoid removing SapphireRapids again by Martin Kroeker 2021-10-18 08:52:32 +0200
  • b85ca5ec3 fix duplicate id by Martin Kroeker 2021-10-18 08:32:03 +0200
  • 63a103ba6 (refs/pull/3416/head) sbgemm: spr: disable small matrix path by default by Wangyang Guo 2021-10-12 01:18:37 -0700
  • 82194ea9d sbgemm: spr: implement otcopy_16 by Wangyang Guo 2021-09-23 01:08:40 -0700
  • 8632380a9 sbgemm: spr: reuse ncopy_16 from cooperlake as incopy by Wangyang Guo 2021-09-18 01:11:31 -0700
  • 6bc8204ce sbgemm: spr: optimization for tmp_c buffer by Wangyang Guo 2021-09-17 23:59:32 -0700
  • f018aa342 sbgemm: spr: kernel handle alpha != 1.0 by Wangyang Guo 2021-09-17 00:48:52 -0700
  • a52456b16 sbgemm: spr: oncopy: use tile load/store instead by Wangyang Guo 2021-09-16 20:08:42 -0700
  • f2485352a sbgemm: spr: only load A once in tail_k handling by Wangyang Guo 2021-09-16 01:04:01 -0700
  • 9ab33228b sbgemm: spr: process k2 and odd k at the same time by Wangyang Guo 2021-09-15 23:59:38 -0700
  • 7b2f5cb3b sbgemm: spr: enlarge P to 256 for performance by Wangyang Guo 2021-09-15 20:29:49 -0700
  • 10d52646e sbgemm: spr: oncopy: avoid handling too much pointer at a time by Wangyang Guo 2021-09-15 19:36:02 -0700
  • 88154ed02 sbgemm: spr: reduce tile conf loading by seperate tail k handling by Wangyang Guo 2021-09-15 01:11:15 -0700
  • 0abbcd19c sbgemm: spr: tuning for blocking params by Wangyang Guo 2021-09-13 01:44:53 -0700
  • a70bfb52d sbgemm: spr: kernel works for NN case when alpha is 1.0 by Wangyang Guo 2021-09-12 19:22:58 -0700
  • 6051c8674 sbgemm: spr: kernel works for m32 in NN case by Wangyang Guo 2021-09-10 01:14:05 -0700
  • d0b253ac6 sbgemm: spr: implement oncopy_16 by Wangyang Guo 2021-09-08 19:41:12 -0700
  • 1d48b7cb1 sbgemm: spr: add dummy source files by Wangyang Guo 2021-09-06 19:48:23 -0700
  • 6adea5243 Add initial support for Fujitsu A64FX as generic ARMV8 by Martin Kroeker 2021-10-18 00:28:57 +0200
  • 3cb1ec2a9 Add basic autodetection support for Fujitsu A64FX by Martin Kroeker 2021-10-18 00:27:54 +0200
  • b57acdf2d Add march/mtune flags for clang builds on ARM64 as well (#3414) by Martin Kroeker 2021-10-18 00:26:14 +0200
  • 02ea3db8e Merge pull request #3404 from guowangy/spr-build by Martin Kroeker 2021-10-17 23:05:11 +0200
  • 4e4f78442 Merge pull request #3413 from MehdiChinoune/cmake-readibiltiy by Martin Kroeker 2021-10-17 22:46:48 +0200
  • 556788281 (refs/pull/3413/head) [NFC] Improve CMakeLists.txt file readibility by Mehdi Chinoune 2021-10-17 05:19:30 +0100
  • f34850646 Merge pull request #3411 from MehdiChinoune/both_shared_static by Martin Kroeker 2021-10-17 20:07:14 +0200
  • 8c6673c27 (refs/pull/3414/head) fix typo and remove clang entry for EMAG8180 for now by Martin Kroeker 2021-10-17 16:26:38 +0200
  • eeb76d63b Fix syntax by Martin Kroeker 2021-10-17 15:46:44 +0200
  • f06c3c7d2 Add march/mtune flags for clang as well by Martin Kroeker 2021-10-17 14:57:35 +0200
  • 059e972d3 Update azure-pipelines.yml by Martin Kroeker 2021-10-16 22:09:17 +0200
  • 28a77a869 (refs/pull/3411/head) Support building both static and shared libraries by Mehdi Chinoune 2021-10-16 08:33:47 +0100
  • 3791d320d Update azure-pipelines.yml by Martin Kroeker 2021-10-16 16:45:14 +0200
  • a66356730 Update azure-pipelines.yml by Martin Kroeker 2021-10-16 15:30:15 +0200
  • 3f7232d8b Update azure-pipelines.yml by Martin Kroeker 2021-10-16 13:58:47 +0200
  • 481b3dc4b Merge pull request #3410 from MehdiChinoune/mingw-clang-64 by Martin Kroeker 2021-10-16 13:52:41 +0200
  • a7ba58ee2 (refs/pull/3412/head) Silent warning from cmake about version<2.8.12 by Mehdi Chinoune 2021-10-16 09:25:26 +0100
  • efd7ac241 (refs/pull/3410/head) Fix MinGW/Clang 64 bits detection. by مهدي شينون (Mehdi Chinoune) 2021-10-16 07:55:10 +0100
  • 6c92f0921 Update azure-pipelines.yml by Martin Kroeker 2021-10-15 16:24:40 +0200
  • a5c733f9a Update azure-pipelines.yml by Martin Kroeker 2021-10-15 15:08:38 +0200
  • 5264ea396 Update azure-pipelines.yml by Martin Kroeker 2021-10-15 13:55:02 +0200
  • bed8e250b Fix missing gfortran in OSX cmake-clang-gfortran job by Martin Kroeker 2021-10-15 12:35:22 +0200
  • 4f1aae7ed Update param.h by Martin Kroeker 2021-10-15 09:40:58 +0200
  • 95ce6d31f Update param.h by Martin Kroeker 2021-10-14 23:52:56 +0200
  • 763042b98 Update param.h by Martin Kroeker 2021-10-14 22:16:47 +0200
  • 9991bb578 Update param.h by Martin Kroeker 2021-10-14 17:45:14 +0200
  • f7193770a Update param.h by Martin Kroeker 2021-10-14 16:25:33 +0200
  • d2df5bd72 Update param.h by Martin Kroeker 2021-10-14 14:51:04 +0200
  • af4d4e55d Update param.h by Martin Kroeker 2021-10-14 13:51:19 +0200
  • f5e7fe0ec Update param.h by Martin Kroeker 2021-10-14 10:37:09 +0200
  • 93cec29c8 Update param.h by Martin Kroeker 2021-10-14 08:59:14 +0200
  • fca825906 Update param.h by Martin Kroeker 2021-10-13 22:30:26 +0200
  • 656b17b4b Update param.h by Martin Kroeker 2021-10-13 21:43:47 +0200
  • c684cae97 Update param.h by Martin Kroeker 2021-10-13 20:59:13 +0200
  • a7a05b78f Update param.h by Martin Kroeker 2021-10-13 14:52:56 +0200
  • 49878cad5 Update param.h by Martin Kroeker 2021-10-13 13:59:16 +0200
  • bb05550b1 Update .travis.yml by Martin Kroeker 2021-10-13 11:40:03 +0200
  • 699c0a036 Update param.h by Martin Kroeker 2021-10-12 15:01:05 +0200
  • 3ce413d1d Update param.h by Martin Kroeker 2021-10-12 13:30:58 +0200
  • 1049dfefa Update param.h by Martin Kroeker 2021-10-12 11:17:52 +0200
  • 1eca91f31 (refs/pull/3404/head) Fix build error in legacy gcc by Wangyang Guo 2021-10-12 02:01:20 -0700
  • 4280dff10 Add NO_AVX=1 fallbacks to Sapphire Rapids build by Wangyang Guo 2021-10-12 01:39:09 -0700
  • 3e409b156 Update param.h by Martin Kroeker 2021-10-12 10:32:24 +0200
  • 3dc6052c7 initial support for Sapphire Rapids platform by Wangyang Guo 2021-09-03 00:39:50 -0700
  • 4217096c9 Update param.h by Martin Kroeker 2021-10-11 22:15:41 +0200
  • ceb535c1e Update param.h by Martin Kroeker 2021-10-11 21:55:37 +0200
  • 2b3d2ef78 Update param.h by Martin Kroeker 2021-10-11 21:30:54 +0200
  • 17376df24 Update param.h by Martin Kroeker 2021-10-11 21:08:00 +0200
  • 2cc76cc84 Update param.h by Martin Kroeker 2021-10-11 18:51:59 +0200
  • 1489e977b Update param.h by Martin Kroeker 2021-10-11 17:56:47 +0200
  • 0a92a783b Update param.h by Martin Kroeker 2021-10-11 17:32:54 +0200
  • 4224f7ee5 Update param.h by Martin Kroeker 2021-10-11 17:19:25 +0200
  • 98548457e Update param.h by Martin Kroeker 2021-10-11 17:07:51 +0200
  • eda222a14 Update .travis.yml by Martin Kroeker 2021-10-11 14:55:04 +0200
  • fa7e4d86f try 512/512 for neoverse dgemm by Martin Kroeker 2021-10-11 13:03:24 +0200
  • 8a87e80c7 Update conda in Appveyor CI and move jobs from Appveyor to Azure (#3400) by Martin Kroeker 2021-10-10 23:24:52 +0200
  • 7e0c93585 (refs/pull/3400/head) Add clang/flang and mingw builds on Windows by Martin Kroeker 2021-10-10 22:19:48 +0200
  • 724128319 Fix clang/cl builds and move them to Azure by Martin Kroeker 2021-10-10 22:18:51 +0200
  • b83b0808d Update appveyor.yml by Martin Kroeker 2021-10-10 18:07:53 +0200
  • 8789d8502 Update appveyor.yml by Martin Kroeker 2021-10-10 16:00:32 +0200
  • 32dab3e0b Update appveyor.yml by Martin Kroeker 2021-10-10 15:51:31 +0200
  • d4ff5eafb Update appveyor.yml by Martin Kroeker 2021-10-10 15:48:12 +0200
  • 17c237c5a (refs/pull/3357/head) Update appveyor.yml by Martin Kroeker 2021-10-10 15:36:04 +0200
  • 58455df95 Update azure-pipelines.yml by Martin Kroeker 2021-10-10 15:34:39 +0200
  • 0c9efc81d Update azure-pipelines.yml by Martin Kroeker 2021-10-10 14:07:49 +0200
  • f42e74536 Update CMakeLists.txt by Martin Kroeker 2021-10-10 13:07:21 +0200