Commit Graph

2833 Commits

Author SHA1 Message Date
Martin Kroeker
a1fb7670f7 Merge pull request #1510 from ararslan/aa/travis-macos
Add macOS to the Travis testing matrix
2018-04-07 12:07:12 +02:00
Martin Kroeker
6c99c97489 Merge pull request #1509 from ararslan/aa/dragonfly
Add DragonFly to exports/Makefile
2018-04-07 12:06:57 +02:00
Alex Arslan
6a0930560e Add macOS to the Travis testing matrix 2018-04-06 17:53:58 -07:00
Alex Arslan
24f8d5b624 Add DragonFly to exports/Makefile
Its exclusion was an oversight on my part.
2018-04-06 17:30:10 -07:00
Martin Kroeker
77b4dbd53b Merge pull request #1506 from martin-frbg/issue1497
Fix thread races and infinite looping on systems with many cpus
2018-04-05 23:46:36 +02:00
Martin Kroeker
bc4c3bca01 Merge pull request #1507 from martin-frbg/threads_usage
Underline importance of NUM_THREADS setting for BUFFER allocation
2018-04-05 08:54:07 +02:00
Martin Kroeker
6b0a9d135c Merge pull request #1508 from ararslan/aa/wording
Minor changes to wording and formatting in the README
2018-04-05 08:53:38 +02:00
Alex Arslan
137ccd9dd9 Minor changes to wording and formatting in the README
The wording in some places is not grammatically correct. This change
also provides minor adjustments to the Markdown formatting which provide
modest improvements to readability.
2018-04-04 14:30:32 -07:00
Martin Kroeker
84923dedb7 Merge pull request #1505 from ararslan/aa/compiler
Compile with cc rather than gcc whenever possible
2018-04-04 22:45:33 +02:00
Martin Kroeker
8ec28ff461 Remove unguarded use of _Atomic and fix tabbing 2018-04-04 22:40:30 +02:00
Martin Kroeker
ca8ca796d3 Underline importance of NUM_THREADS setting for BUFFER allocation
following augray's suggestion from #1451, and incorporating ashwinyes' comments from #1141 on the importance of NUM_THREADS even for single-threaded builds.
2018-04-04 22:26:51 +02:00
Alex Arslan
8f811a9312 Reinstate macOS logic 2018-04-04 11:41:45 -07:00
Alex Arslan
36a17536ca Compile with cc rather than gcc whenever possible 2018-04-04 11:26:54 -07:00
Martin Kroeker
bb9876db33 Fix thread races and infinite looping on systems with many cpus
On systems with more than 64 cpus, blas_quickdivide will sometimes return zero which creates bogus workloads when used for the stride calculation. This then leads to threads spinning incessantly waiting for a status change that never happens, as seen in #1497.
This patch also fixes several data races that were found by helgrind and/or tsan while debugging the issue.
2018-04-04 18:16:52 +02:00
Martin Kroeker
d636b418af Merge pull request #1504 from ararslan/aa/openbsd
Allow building on OpenBSD
2018-04-04 15:26:46 +02:00
Martin Kroeker
a460c92577 Merge pull request #1501 from martin-frbg/issue875
Add workaround for old gcc and clang versions
2018-04-04 15:26:21 +02:00
Alex Arslan
33f838393c Add OpenBSD and DragonFly to community supported platforms 2018-04-03 16:42:01 -07:00
Alex Arslan
a41d241a0e Add support for DragonFly BSD 2018-04-03 16:39:29 -07:00
Alex Arslan
8da6b6ae52 Allow building on OpenBSD
With this change, OpenBLAS builds and all tests pass on OpenBSD 6.2
using Clang. Tested on x86-64 only, with and without DYNAMIC_ARCH=1.
2018-04-02 10:48:22 -07:00
Martin Kroeker
01c4b82f04 Update memory.c 2018-03-31 22:32:06 +02:00
Martin Kroeker
93db123f7e Update memory.c 2018-03-29 13:13:49 +02:00
Martin Kroeker
752fdb5dd8 Add workaround for old gcc and clang versions
Old gcc and clang do not handle constructor arguments, finally fix #875 as discussed there, using the fedora patch
2018-03-29 11:56:56 +02:00
Martin Kroeker
07ed01e97f Merge pull request #1500 from martin-frbg/issue1474
Correct index variables used in MFlops calculation
2018-03-28 09:15:34 +02:00
Martin Kroeker
35c5a32309 Correct index variables used in MFlops calculation
Fixes #1474
2018-03-27 21:52:29 +02:00
Martin Kroeker
c7b55b6082 Merge pull request #1499 from quickwritereader/develop
Implemented missing vsx simd  kernels for power8 blas1/2 double. z13 modifications
2018-03-27 21:43:23 +02:00
Martin Kroeker
840e01061f Merge pull request #1491 from martin-frbg/ddot_mt
Add multithreading support for Haswell DDOT
2018-03-27 21:43:05 +02:00
QWR QWR
28ca97015d power8:Added initial zgemv_(t|n) ,i(d|z)amax,i(d|z)amin,dgemv_t(transposed),zrot
z13: improved zgemv_(t|n)_4,zscal,zaxpy
2018-03-27 14:54:41 +00:00
Martin Kroeker
73c5ca74fa Merge pull request #1495 from martin-frbg/aff
Disable CPU affinity by default again
2018-03-19 18:03:25 +01:00
Martin Kroeker
e453555d97 Disable CPU affinity by default again
This setting must have been changed unintentionally by my PR #1214 (probably leftover from unrelated tests)
2018-03-19 18:02:23 +01:00
Martin Kroeker
6a6ffaff1e Merge pull request #1494 from martin-frbg/x86_dsdot
Use generic/dot.c instead of the inferior arm/dot.c for x86 DSDOT
2018-03-17 15:26:47 +01:00
Martin Kroeker
28ac9ea5a6 Use generic/dot.c instead of the inferior arm/dot.c for x86 DSDOT
to resolve dsdot utest failure seen in #1492
2018-03-17 13:49:15 +01:00
Martin Kroeker
a55694dd5b Declare dot_compute static to avoid conflicts in multiarch builds 2018-03-16 22:23:36 +01:00
Martin Kroeker
85a41e9cdb Add multithreading support for Haswell DDOT
copied from ashwinyes' implementation in dot_thunderx2t99.c
2018-03-16 16:58:47 +01:00
Martin Kroeker
2c7392f07b Merge pull request #1482 from martin-frbg/haswell_axpy
Re-enable DAXPY AVX microkernels  for x86_64
2018-03-04 22:21:18 +01:00
Martin Kroeker
81215711a2 Re-enable DAXPY microkernels for x86_64
as the inaccuracies seen in the original testcase for #1332 appear to be due to an artefact that amplifies the very small rounding differences between FMA and discrete multiply+add
2018-03-04 19:37:03 +01:00
Martin Kroeker
809fd0d451 Rewrite ROTMG to address cases not covered by the netlib algorithm (#1480)
* Rewrite ROTMG based on the new implementation in GONUM based on the algorithm proposed by Tim Hopkins, see issue 1452 for the reference
* Correct ROTMG utest for issue1452 and add another from gonum, also correct transposition of expected and observed values in error messages
2018-03-04 17:39:56 +01:00
Martin Kroeker
72e65157df Merge pull request #1481 from martin-frbg/utest-fixup
Fix transposition of expected and computed values in error message
2018-03-03 22:43:56 +01:00
Martin Kroeker
69a8aa6de2 Fix transposition of expected and computed values in error message 2018-03-03 18:01:51 +01:00
Martin Kroeker
0ab5bf1746 Merge pull request #1476 from xsacha/patch-1
Fix CMake cross-compiling
2018-02-28 18:47:57 +01:00
Martin Kroeker
22167170b3 Merge pull request #1477 from quickwritereader/develop
Power8 blas3 copy-pack routines
2018-02-28 18:46:54 +01:00
Martin Kroeker
69d9f36ff4 Merge pull request #1468 from martin-frbg/martin-frbg-patch-1
Limit the additional locking from PRs 1052,1299 to non-OpenMP cases
2018-02-28 18:40:31 +01:00
Sacha
f81815e48a Fix CMake cross-compiling
Without specifying thread count, NUM_THREADS would not be defined and CMake would fail.
This is because core count cannot be determined when cross-compiling.
2018-02-28 10:25:25 +10:00
Martin Kroeker
5f855d965d Merge pull request #1475 from ashwinyes/develop_20180227_utest_dsdot_fixes
ARM64: Fix utest dsdot errors
2018-02-27 14:04:16 +01:00
Ashwin Sekhar T K
fa9ca65c0e ARM64: Fix utest dsdot errors 2018-02-27 10:47:55 +00:00
Martin Kroeker
719b68f077 Merge pull request #1473 from martin-frbg/p2align
Replace .align with .p2aligns in dscal.c and the Nehalem microkernels as well
2018-02-27 08:28:20 +01:00
Martin Kroeker
fe9f15f2d8 Merge pull request #1472 from martin-frbg/utest-fixes
Fix limited DSDOT precision on arm,aarch64 and zarch
2018-02-26 22:48:07 +01:00
Martin Kroeker
497f0c3d8a Replace .align with .p2align in the Nehalem microkernels 2018-02-26 20:58:33 +01:00
Martin Kroeker
ea37db828e Convert .align to .p2align for OSX compatibility 2018-02-26 20:48:03 +01:00
Martin Kroeker
e6a0a3de73 Merge pull request #1471 from martin-frbg/p2align
Use .p2align instead of .align for portability on Haswell and Sandybridge
2018-02-26 12:28:01 +01:00
Martin Kroeker
6e70287776 Use generic/dot.c for DSDOT on ARMV5 and above
The default arm/dot.c is less precise when used for DSDOT, as shown by utest
2018-02-25 19:57:23 +01:00