Commit Graph

  • *
  • |\
  • | *
  • | *
  • | *
  • | *
  • | |\
  • | |/
  • |/|
  • * |
  • |\ \
  • * \ \
  • |\ \ \
  • * \ \ \
  • |\ \ \ \
  • * \ \ \ \
  • |\ \ \ \ \
  • | * | | | |
  • | | |_|_|/
  • | |/| | |
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | | | | *
  • | | * | | |
  • | |/ / / /
  • |/| | | |
  • | | | | *
  • | | | * |
  • | |_|/ /
  • |/| | |
  • | | | *
  • | | | *
  • | | | *
  • | | | *
  • | | | *
  • | | | |\
  • | |_|_|/
  • |/| | |
  • | | | *
  • | | | *
  • * | | |
  • | | | *
  • | | | *
  • | | | *
  • | | | *
  • | | | *
  • | | | *
  • | | |/
  • | |/|
  • | | | *
  • | | | *
  • | | | *
  • | | | *
  • | * | |
  • | |\ \ \
  • | |/ / /
  • |/| | |
  • | * | |
  • | |\ \ \
  • | | * | |
  • | |/ / /
  • | | | *
  • | | | *
  • | | | *
  • * | | |
  • | * | |
  • | | | | *
  • | | * | |
  • | |/ / /
  • |/| | |
  • | | * |
  • | | | *
  • | | | *
  • | | | *
  • | | |/
  • | |/|
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • | |/
  • | *
  • | |\
  • | |/
  • |/|
  • * |
  • | | *
  • | | *
  • | | *
  • | | *
  • | | *
  • * | |
  • |\ \ \
  • | | | *
  • | | | *
  • | * | |
  • | * | |
  • | |/ /
  • | | *
  • | | *
  • | | *
  • | |/
  • * |
  • |\ \
  • | * |
  • | * |
  • | |/
  • * |
  • |\ \
  • * \ \
  • |\ \ \
  • | * | |
  • | * | |
  • | | |/
  • | |/|
  • | * |
  • | |\ \
  • | |/ /
  • |/| |
  • | | | *
  • | |_|/
  • |/| |
  • e7f0da929 Merge pull request #2551 from martin-frbg/issue2538-2 by Martin Kroeker 2020-04-12 22:34:41 +0200
  • e9bfa2291 (refs/pull/2551/head) Fix parameter overflow by Martin Kroeker 2020-04-12 19:47:02 +0200
  • 2a28448a9 Add safeguards for sufficient BUFFER_SIZE by Martin Kroeker 2020-04-12 19:45:36 +0200
  • a33d17743 Increase default BUFFER_SIZE on ARM, ZARCH and newer x86_64, add GEMM_R for POWER8/9 by Martin Kroeker 2020-04-12 19:44:48 +0200
  • f73391c9c Merge pull request #45 from xianyi/develop by Martin Kroeker 2020-04-12 19:39:05 +0200
  • 7905383cb Merge pull request #2547 from sharvil/develop by Martin Kroeker 2020-04-11 00:35:38 +0200
  • a8cbd451b Merge pull request #2541 from bapt/develop by Martin Kroeker 2020-04-11 00:35:07 +0200
  • eecd8c320 Merge pull request #2548 from gxw-loongson/develop by Martin Kroeker 2020-04-11 00:34:04 +0200
  • ea85eb2e0 Merge pull request #2549 from martin-frbg/fixthreadtest by Martin Kroeker 2020-04-10 23:54:40 +0200
  • 66f89c0aa (refs/pull/2549/head) Match thread count to machine capability by Martin Kroeker 2020-04-10 22:06:44 +0200
  • a3fd074bc (refs/pull/2546/head) Lets try an actual read barrier rather than rw - does it fail ? by Martin Kroeker 2020-04-10 21:47:17 +0200
  • ed21cff78 add a read memory barrier for experimenting by Martin Kroeker 2020-04-10 21:46:00 +0200
  • fa38477ae Convert to read barrier, will probably fail now by Martin Kroeker 2020-04-10 15:52:03 +0200
  • eb0ae9e71 Move the memory barrier to the allocation loop by Martin Kroeker 2020-04-10 14:11:37 +0200
  • 79cb24dc1 Update dgemm_thread_safety.cpp by Martin Kroeker 2020-04-10 12:45:58 +0200
  • 504809086 Update dgemv_thread_safety.cpp by Martin Kroeker 2020-04-10 12:45:39 +0200
  • 82375a5a1 Add back the cpp_thread_test by Martin Kroeker 2020-04-10 12:34:43 +0200
  • d3eba36c5 Reduce thread count if necessary by Martin Kroeker 2020-04-10 12:32:37 +0200
  • a6f45ab7f Reduce thread count if necessary by Martin Kroeker 2020-04-10 12:31:12 +0200
  • d8bdd4f23 revert previous, num_buffers is not a makefile variable by Martin Kroeker 2020-04-09 23:25:33 +0200
  • ff23bd09f Update gemm.c by Martin Kroeker 2020-04-09 23:24:21 +0200
  • 1d12a33a9 print num_buffers at end of build just to be sure by Martin Kroeker 2020-04-09 23:09:34 +0200
  • c00b96000 Update .drone.yml by Martin Kroeker 2020-04-09 21:38:56 +0200
  • 417eb2851 Update .drone.yml by Martin Kroeker 2020-04-09 21:38:17 +0200
  • 54973cca1 Update .drone.yml by Martin Kroeker 2020-04-09 20:35:27 +0200
  • 5d2cf4ec1 Update gemm.c by Martin Kroeker 2020-04-09 19:13:56 +0200
  • 4ffe9d788 Update .drone.yml by Martin Kroeker 2020-04-09 18:04:12 +0200
  • f10c9a99a Delete azure-pipelines.yml by Martin Kroeker 2020-04-09 16:52:00 +0200
  • b7fa8fe69 Delete appveyor.yml by Martin Kroeker 2020-04-09 16:51:37 +0200
  • 71b8e284e Delete .travis.yml by Martin Kroeker 2020-04-09 16:51:20 +0200
  • 8290b6787 Update .drone.yml by Martin Kroeker 2020-04-09 16:50:49 +0200
  • 67de70813 remove thread count from pragma as drone.io HW varies by Martin Kroeker 2020-04-09 15:47:52 +0200
  • 35036d9b6 reduce NUM_PARALLEL to 1 by Martin Kroeker 2020-04-09 14:43:59 +0200
  • ce9585310 limit dgemm benchmark to just 10,10,0 by Martin Kroeker 2020-04-09 14:39:04 +0200
  • 8d07cf9b6 (refs/pull/2548/head) Fix compilation problem on loongson platform by gxw 2020-04-09 19:25:13 +0800
  • 11528f3af Update gemm.c by Martin Kroeker 2020-04-08 22:19:18 +0200
  • 7b4773b24 (refs/pull/2547/head) Add API to set thread affinity on Linux. by Sharvil Nanavati 2020-04-08 12:47:41 -0700
  • 9ed53824d Update gemm.c by Martin Kroeker 2020-04-08 20:26:58 +0200
  • 3778b9165 Update gemm.c by Martin Kroeker 2020-04-08 17:25:28 +0200
  • 626e98028 Update gemm.c by Martin Kroeker 2020-04-08 15:24:22 +0200
  • aa170123e fix accidental deletion by Martin Kroeker 2020-04-08 14:58:37 +0200
  • 353e996d1 Merge branch 'develop' into dronethunder2 by Martin Kroeker 2020-04-08 14:45:32 +0200
  • bc792904e use modified gemm benchmark to trigger race condition by Martin Kroeker 2020-04-08 14:43:19 +0200
  • d8735bb66 parallelize gemm benchmark to trigger races by Martin Kroeker 2020-04-08 14:41:21 +0200
  • 69f277f8e Add another memory barrier for ARM and a multicore test run on ThunderX to help detect such issues (#2544) by Martin Kroeker 2020-04-08 11:04:51 +0200
  • 0e0681f53 (refs/pull/2544/head) Experimental barrier by Martin Kroeker 2020-04-08 09:24:03 +0200
  • 29a50dd04 increase nthreads to 96 by Martin Kroeker 2020-04-08 01:04:40 +0200
  • aa8269d47 Add g++ as dependency for dgemm_tester by Martin Kroeker 2020-04-08 00:00:15 +0200
  • e1ec040b9 Try dgemm_tester instead of lapack-test by Martin Kroeker 2020-04-07 23:50:41 +0200
  • 9a4959997 Add python dependency for lapack test by Martin Kroeker 2020-04-07 22:36:16 +0200
  • 8639c8a68 Try to get an all-core lapack test to identify barrier issues by Martin Kroeker 2020-04-07 21:48:38 +0200
  • 330d6b1ee (refs/pull/2539/head) Update common_param.h by Martin Kroeker 2020-04-07 00:10:14 +0200
  • fd99b3e05 workaround for sign change warning by Martin Kroeker 2020-04-06 23:15:13 +0200
  • aab5380aa typo fix by Martin Kroeker 2020-04-06 22:14:44 +0200
  • 6f2e18d5e Comment out SGEMM_R for POWER8 again, try if declaring P and Q as UL is sufficient to avoid int overflow by Martin Kroeker 2020-04-06 20:51:14 +0200
  • 3a6d51c2f Merge pull request #44 from xianyi/develop by Martin Kroeker 2020-04-04 22:48:53 +0200
  • 1c7771df9 Merge pull request #43 from martin-frbg/revert-42-z12ci by Martin Kroeker 2020-04-04 22:46:58 +0200
  • a56c9ec52 Revert "Add IBM Z to Travis configuration (#42)" by Martin Kroeker 2020-04-04 22:45:01 +0200
  • 66caf61a2 Try predefining GEMM_R for POWER8 by Martin Kroeker 2020-04-04 19:31:38 +0200
  • 188e9239a Increase BUFFER_SIZE and remove remnants of arm64 source by Martin Kroeker 2020-04-04 15:27:32 +0200
  • 0b8d69f7a Restore correct version by Martin Kroeker 2020-04-04 00:00:10 +0200
  • 4ae6d1a01 Add a Z13 build to the Travis configuration (#2542) by Martin Kroeker 2020-04-03 16:02:11 +0200
  • 7972beb37 Add IBM Z to Travis configuration (#42) by Martin Kroeker 2020-04-03 15:59:18 +0200
  • e19d10622 (refs/pull/2542/head) Update .travis.yml by Martin Kroeker 2020-04-03 14:43:30 +0200
  • 41e802443 (refs/pull/2541/head) libname: treat FreeBSD and DragonFly like linux and sunos by Baptiste Daroussin 2020-04-03 06:20:42 +0200
  • 07d59c045 print the current values when buffer_size is too small by Martin Kroeker 2020-04-02 23:27:10 +0200
  • fdcf50f99 Add arch entry for s390x by Martin Kroeker 2020-04-02 22:24:43 +0200
  • 4666cc442 Update .travis.yml by Martin Kroeker 2020-04-02 21:38:14 +0200
  • b474c65db Add IBM Z to Travis configuration by Martin Kroeker 2020-04-02 19:54:34 +0200
  • f03b667dd Increase BUFFER_SIZE for POWER8/9 by Martin Kroeker 2020-04-02 18:20:27 +0200
  • 053712eb1 Increase BUFFER_SIZE by Martin Kroeker 2020-04-02 15:12:50 +0200
  • db6db050d Increase BUFFER_SIZE for POWER8/9 by Martin Kroeker 2020-04-02 15:11:53 +0200
  • b21ca5c96 Increase BUFFER_SIZE for POWER8/9 by Martin Kroeker 2020-04-02 14:33:49 +0200
  • cab855d56 Increase default BUFFER_SIZE for Haswell, Zen and SKX by Martin Kroeker 2020-04-02 14:26:53 +0200
  • df989d7a5 Add compile-time guard for adequate buffersize by Martin Kroeker 2020-04-02 10:58:05 +0200
  • 5e3e657ca Make BUFFER_SIZE configurable and increase its default value for TSV110 and EMAG8180 by Martin Kroeker 2020-04-02 10:38:35 +0200
  • 7bd8624b7 Merge pull request #41 from xianyi/develop by Martin Kroeker 2020-04-02 10:32:19 +0200
  • 806f89166 Make ARMV7 compile with xcode and add a CI job for it (#2537) by Martin Kroeker 2020-04-02 10:30:37 +0200
  • 41b470244 (refs/pull/2537/head) restore quiet_make by Martin Kroeker 2020-04-02 02:04:31 +0200
  • 07cb1097f Make local labels in macro compatible with the xcode assembler by Martin Kroeker 2020-04-02 00:44:28 +0200
  • 62cf7a82f Update .travis.yml by Martin Kroeker 2020-04-01 23:08:56 +0200
  • f0889ab50 Update .travis.yml by Martin Kroeker 2020-04-01 21:49:14 +0200
  • ac1d704f5 Add no-thumb option for ARMV7 IOS to get it to accept DMB ISH by Martin Kroeker 2020-04-01 20:09:34 +0200
  • f059e614e Merge pull request #2536 from martin-frbg/recurs by Martin Kroeker 2020-04-01 20:00:13 +0200
  • abfc80a5e thread_local appears to be unavailable on ARMV7 iOS by Martin Kroeker 2020-04-01 17:53:40 +0200
  • 2d7209fdb Update .travis.yml by Martin Kroeker 2020-04-01 16:22:01 +0200
  • e13b6773e (refs/pull/2536/head) ifort and pgfort need "recursive" for safe compilation of LAPACK as well by Martin Kroeker 2020-04-01 15:39:16 +0200
  • a05243d0f ifort and pgfort need "recursive" for compiling LAPACK as well by Martin Kroeker 2020-04-01 15:38:07 +0200
  • 2977f652c Update .travis.yml by Martin Kroeker 2020-04-01 14:27:09 +0200
  • 798322bf0 Update .travis.yml by Martin Kroeker 2020-04-01 09:47:20 +0200
  • 1becf4ef5 Add an ARMV7 iOS build by Martin Kroeker 2020-03-31 22:52:05 +0200
  • c6af9bbb3 Merge pull request #2534 from martin-frbg/issue2496 by Martin Kroeker 2020-03-31 20:53:13 +0200
  • 144be81ca (refs/pull/2534/head) fix initialization to zero in the NEON SGEMM_BETA kernel as well by Martin Kroeker 2020-03-31 16:53:56 +0200
  • 07cdd5d05 Fix zero initialization for beta=0 case by Martin Kroeker 2020-03-31 00:21:02 +0200
  • 567d2760e Merge pull request #2520 from wjc404/develop by Martin Kroeker 2020-03-30 20:15:59 +0200
  • 018bb3e43 Merge pull request #2533 from martin-frbg/gemmdirect2 by Martin Kroeker 2020-03-30 20:15:37 +0200
  • 79fd006c5 (refs/pull/2533/head) Expose the support_avx512 function provided in dynamic.c by Martin Kroeker 2020-03-26 21:25:39 +0100
  • 8229c163b Use runtime check for AVX512 (sgemm_direct) capability when using DYNAMIC_ARCH by Martin Kroeker 2020-03-26 21:12:56 +0100
  • a986d42ea Merge pull request #39 from xianyi/develop by Martin Kroeker 2020-03-26 21:06:51 +0100
  • 06ef74c84 (refs/pull/2531/head) Do not deploy import libraries on Windows when NO_STATIC=1 by Harmen Stoppels 2020-03-24 16:45:52 +0100