Commit Graph

  • f7c7dff720
    [appveyor] set branch xoviat 2017-10-31 16:51:49 -0500
  • 8eefc4fc5e
    [appveyor] set branch xoviat 2017-10-31 16:41:19 -0500
  • b51ef0c7a2
    [appveyor] use -q xoviat 2017-10-31 16:40:33 -0500
  • 9e81791bc5
    [appveyor] fix environment xoviat 2017-10-31 16:29:01 -0500
  • e06443e5ac
    [appveyor] suppress failure xoviat 2017-10-31 16:23:50 -0500
  • f61b7b4b21
    [appveyor] enable clcache xoviat 2017-10-31 16:15:16 -0500
  • ead0aa7569 Merge b2b9d17146 into 5f998efd7b Pavel Grunt 2017-10-30 11:44:41 +0000
  • b2b9d17146 cmake: Allow to build without fortran Pavel Grunt 2017-10-30 10:45:54 +0100
  • ab87ee6b48 Merge pull request #1329 from martin-frbg/dsdot Martin Kroeker 2017-10-25 19:13:38 +0200
  • a07807caac Eliminate loop code when called as/from dsdot Martin Kroeker 2017-10-25 16:45:41 +0200
  • b71f4fe681 Merge pull request #1334 from ashwinyes/develop_aarch64_20171024_addlocallabels Martin Kroeker 2017-10-24 19:50:03 +0200
  • a0128aa489 ARM64: Convert all labels to local labels Ashwin Sekhar T K 2017-10-24 10:47:11 +0000
  • 627133f9ad Merge pull request #1333 from martin-frbg/haswell32 Martin Kroeker 2017-10-24 11:25:03 +0200
  • 0e2cf102e1 Fix 32bit HASWELL Martin Kroeker 2017-10-24 10:07:44 +0200
  • 5e3e91d0fc Split the microkernel workload into chunks of 32 floats for dsdot mode to limit loss of precision Martin Kroeker 2017-10-22 18:18:51 +0200
  • 28c3fa8950 Add dsdot Martin Kroeker 2017-10-16 23:29:03 +0200
  • 8ac87c1cb6 Implement DSDOT with unchanged sdot microkernels Martin Kroeker 2017-10-16 23:27:51 +0200
  • 17b993c0a3 Update izamin.c Abdelrauf 2017-10-13 19:29:27 +0400
  • b7cee00455 Merge pull request #1327 from martin-frbg/cmake-relapack Martin Kroeker 2017-10-12 20:26:35 +0200
  • 962b20a9bb Optionally add ReLAPACK to LIB_COMPONENTS Martin Kroeker 2017-10-12 17:02:01 +0200
  • fbf83f4833 Add cmake build list file for ReLAPACK Martin Kroeker 2017-10-12 17:00:00 +0200
  • 78cec6209c Add ReLAPACK option Martin Kroeker 2017-10-12 16:58:37 +0200
  • c460027dbe Merge pull request #1325 from grisuthedragon/patch-1 Martin Kroeker 2017-10-10 12:14:34 +0200
  • bfa9b9f6b2 Update README.md Martin Köhler 2017-10-10 10:12:04 +0200
  • c7a8512d12 Cmake fixes for DYNAMIC_ARCH builds and whitespace in path names (#1323) Martin Kroeker 2017-10-09 23:34:18 +0200
  • 8c96cc5c59 Add files via upload Martin Kroeker 2017-10-09 21:24:00 +0200
  • 8c194ea387 Add files via upload Martin Kroeker 2017-10-09 18:31:04 +0200
  • f9198a9ab8 revert quoting of path names to see if it caused the appveyor fault Martin Kroeker 2017-10-09 11:56:54 +0200
  • 00aefcc19b Fix common_lapack header inclusion and DYNAMIC_ARCH build Martin Kroeker 2017-10-09 10:29:45 +0200
  • 122f451866 Correct template for ?laswp_(plus,minus) and add DYNAMIC_ARCH versions Martin Kroeker 2017-10-09 09:29:35 +0200
  • d5cc12daaa Fix common_lapack header inclusion Martin Kroeker 2017-10-09 09:28:52 +0200
  • 62ab563641 Put quotes around path names that may contain whitespace Martin Kroeker 2017-10-09 09:27:31 +0200
  • a26344bdda Correct template for ?laswp_(plus,minus) and add DYNAMIC_ARCH versions Martin Kroeker 2017-10-09 00:08:26 +0200
  • 4b584b0bc5 Fix common_lapack header inclusion and DYNAMIC_ARCH Martin Kroeker 2017-10-09 00:05:59 +0200
  • b73a23153e Put quotes around path names that may contain whitespace Martin Kroeker 2017-10-08 23:59:18 +0200
  • db72ad8f6a Merge pull request #1320 from timmoon10/develop Martin Kroeker 2017-10-08 23:31:33 +0200
  • 97ecd4996a Merge pull request #1319 from martin-frbg/issue601 Martin Kroeker 2017-10-08 23:31:06 +0200
  • 1eb43cccad Merge pull request #1317 from martin-frbg/power8-asm Martin Kroeker 2017-10-08 23:30:46 +0200
  • 9d92f526dd Comment out a code block that performs out-of-bounds memory accesses Martin Kroeker 2017-10-06 23:51:32 +0200
  • bd05387cf2 Merge 2fbd66c4ab into 5f998efd7b Alexander Karatarakis 2017-10-06 21:43:57 +0000
  • 514d237257 Merge pull request #1279 from xsacha/develop Martin Kroeker 2017-10-06 21:13:45 +0200
  • 30486a356c Reduce number of data partitions in n. Tim Moon 2017-10-04 12:37:49 -0700
  • e1b2502840 Merge pull request #1316 from timmoon10/develop Martin Kroeker 2017-10-04 20:35:00 +0200
  • 9de52b489a Cleaning up and documenting multi-threaded GEMM code. Tim Moon 2017-10-03 16:32:08 -0700
  • 860dcfc703 Use 2D thread distribution for small GEMMs. Tim Moon 2017-10-03 13:43:39 -0700
  • f96afd94b0 Fix out-of-bounds accesses where the data should be zero anyway Martin Kroeker 2017-10-01 01:06:39 +0200
  • ebe84215e4 Merge pull request #1318 from pv/potrf-smoketest Martin Kroeker 2017-09-30 21:31:28 +0200
  • 845e6d750f Add trivial smoketest for xpotrf Pauli Virtanen 2017-09-30 18:40:03 +0200
  • a89d6711c6 Increasing flexibility of GEMM benchmark. Tim Moon 2017-09-28 12:56:29 -0700
  • 9c017a2218 Save and restore VSX registers Martin Kroeker 2017-09-28 12:17:09 +0200
  • 0e6b11b708 Merge https://github.com/timmoon10/OpenBLAS into develop Tim Moon 2017-09-27 19:26:38 -0700
  • 6aaa107865 Reducing threads for multi-threaded GEMMs on small matrices. Tim Moon 2017-09-27 19:25:33 -0700
  • 00c42dc815 Merge pull request #1314 from martin-frbg/nofortran-fix-2 Martin Kroeker 2017-09-26 10:34:18 +0200
  • 79e754e548 Rewrite NOFORTRAN conditionals Martin Kroeker 2017-09-25 23:45:14 +0200
  • 2ccd7f6e0c Merge pull request #1310 from sva-img/develop Martin Kroeker 2017-09-22 09:34:54 +0200
  • e3d844b062 Added mips I6500 core Shivraj Patil 2017-09-22 11:57:43 +0530
  • def146efed Merge pull request #1308 from sebastien-villemot/develop Martin Kroeker 2017-09-19 14:04:37 +0200
  • 7543e578a4 Add support for TARGET=ZARCH_GENERIC and TARGET=Z13 Sébastien Villemot 2017-09-19 12:16:42 +0200
  • 601c71fe54 Merge pull request #1304 from martin-frbg/aix-build-fixes Martin Kroeker 2017-09-18 10:16:40 +0200
  • 3810a6fd99 (Plain make) build system fixes for AIX Martin Kroeker 2017-09-18 01:29:21 +0200
  • 742f54c235 Merge pull request #1303 from martin-frbg/imatcopy-rowscols Martin Kroeker 2017-09-14 21:46:26 +0200
  • d674fbb4c7 Fix cols/rows mixup in omatcopy 2nd step for BlasTrans cases Martin Kroeker 2017-09-14 19:59:05 +0200
  • 2922c15f36 Merge pull request #1302 from martin-frbg/nofortran-fix Martin Kroeker 2017-09-14 11:54:20 +0200
  • 3a245a376f Remove default FEXTRALIBS in NOFORTRAN case Martin Kroeker 2017-09-14 09:21:04 +0200
  • 46c9357c72 Merge pull request #1288 from quickwritereader/develop Martin Kroeker 2017-09-09 23:47:17 +0200
  • 1c3e2d3dd5 Merge pull request #1293 from embray/cygwin/install Martin Kroeker 2017-09-09 23:46:27 +0200
  • f66d908282 Merge pull request #1299 from martin-frbg/race_fixes Martin Kroeker 2017-09-09 23:41:53 +0200
  • ba1f91f17b Convert another caller of "allocation" to LOCK_COMMAND Martin Kroeker 2017-09-09 20:30:33 +0200
  • f460776f0f Fix thread data races Martin Kroeker 2017-09-09 19:07:06 +0200
  • e882f3d6f3 Fix thread data race in memory.c Martin Kroeker 2017-09-09 18:58:38 +0200
  • 2fbd66c4ab Fix build issues when path contains whitespace Alexander Karatarakis 2017-09-07 12:08:24 -0700
  • dddedbab5d More canonical installation on Cygwin: Erik M. Bray 2017-09-07 14:18:56 +0200
  • 1cfdb2295d Optimized standard Blas Level-1,2 (excluding nrm2 functions) for z13 (double precision) Abdurrauf 2017-04-08 21:51:15 +0400
  • 00740c0e34 Merge pull request #1290 from martin-frbg/imatcopy Martin Kroeker 2017-09-03 13:02:10 +0200
  • 254db9bd7c Use in-place transform shortcut only if matrix is square Martin Kroeker 2017-09-03 09:52:55 +0200
  • f2074f9ac1 Merge pull request #1286 from martin-frbg/baytrail Martin Kroeker 2017-08-27 13:23:57 +0200
  • aece65ea29 Fix coretype detection for Bay Trail Atom Martin Kroeker 2017-08-27 13:06:54 +0200
  • ef64991506 Clean up config file writing. Sacha 2017-08-23 12:47:38 +1000
  • 7a867082d8 Fix open_blas.config which was never working out-of-source. Remove need for gen_config_h.exe. If OpenMP is requested, do not silently ignore when it isn't available. Sacha 2017-08-23 11:16:24 +1000
  • a1b87eac6b Do not require Perl for MSVC if CMake >= 3.4 Sacha Refshauge 2017-08-23 07:19:02 +1000
  • 47ebce4d1a Clean up, fix old typos. Simplify arch usages. Move system arch check to earlier position. Sacha Refshauge 2017-08-21 00:37:29 +1000
  • 69b560751c Improvements to previous commit (cross-compile). Fix typos and bad if statements discovered in 0.2.20. Sacha Refshauge 2017-08-20 22:50:31 +1000
  • 0a7a527a92 Add support for cross compiling. Add support for not having host compiler as CMake cannot detect such a compiler. Add support for not using getarch. Successfully builds Android ARMV8. Any target can be added by supplying the TARGET_CORE config in prebuild.cmake. Sacha Refshauge 2017-08-20 20:08:53 +1000
  • 50715e8945 Merge pull request #1281 from sharkcz/armv8 Martin Kroeker 2017-08-19 20:37:19 +0200
  • 11911fd941 Add kernel/Makefile.LA to CMake Sacha Refshauge 2017-08-20 00:59:14 +1000
  • 408b4fe83f Add a CMake GCC and Clang target to Travis CI Sacha Refshauge 2017-08-20 00:59:00 +1000
  • 4474465438 Remove _static usages for tests Sacha Refshauge 2017-08-20 00:13:46 +1000
  • b9ec72546c Only run utest without NOFORTRAN, same as Makefile. Linux now compiles. Sacha Refshauge 2017-08-20 00:13:24 +1000
  • 37858d1146 Fix threading usage in CMake: s/SMP/USE_THREAD/ Sacha Refshauge 2017-08-19 15:07:42 +1000
  • 1763e01567 fix detection of generic ARMv8 CPUs Dan Horák 2017-08-18 14:53:29 +0200
  • 6aac06587d Fix typos and use CMake OpenMP support. Sacha Refshauge 2017-08-17 17:27:01 +1000
  • 7c1acc07f0 Fix bug that required fortran. Fix bug that needed CXX var. Remove redundant set vars. Fix threading detection. Do not attempt to run code if cross compiling. 2017-08-17 03:32:04 +1000
  • 38d273ea03 Drop some redundant vars and improve arch detection in CMake. 2017-08-17 02:04:36 +1000
  • 7242cdc4ec Allow CMake to determine if it is building static or shared. 2017-08-17 00:51:04 +1000
  • 90a4dab501 Let CMake deal with build type. 2017-08-17 00:35:54 +1000
  • a8a342ccc4 Merge pull request #1277 from cconrads-scicomp/fix-installation-instructions Martin Kroeker 2017-08-10 23:42:23 +0200
  • 9e9a9553db Merge pull request #1276 from cconrads-scicomp/android_-lm_fix Martin Kroeker 2017-08-10 21:35:32 +0200
  • be7c1b6324 Merge pull request #1275 from cconrads-scicomp/recognize-gfortran-on-arm Martin Kroeker 2017-08-10 21:32:09 +0200
  • be8e8228ae Make: show installation instructions after build Christoph Conrads 2017-08-10 14:22:26 -0400
  • 77b197bc93 Make: fix installation instructions Christoph Conrads 2017-08-10 12:47:18 -0400