Commit Graph

3502 Commits

Author SHA1 Message Date
Martin Kroeker 16494cb7c4
Merge pull request #1980 from martin-frbg/issue1979
Report SkylakeX as Haswell if compiler does not support AVX512
2019-01-22 21:10:38 +01:00
Martin Kroeker b56b34a75c
Syntax fix 2019-01-22 18:55:43 +01:00
Martin Kroeker 21eda8b577
Report SkylakeX as Haswell if compiler does not support AVX512
... or make was invoked with NO_AVX512=1
2019-01-22 18:47:12 +01:00
Daniel Cohen Gindi 24288803b3 Adjust test script for correct deployment 2019-01-22 14:38:01 +02:00
Martin Kroeker f0d834b824
Use VERSION_LESS for comparisons involving software version numbers 2019-01-22 12:32:24 +01:00
Daniel Cohen Gindi 63bbd7b0d7 Better support for MSVC/Windows in CMake 2019-01-21 17:47:47 +02:00
maamountki b111829226
[ZARCH] Update max/min functions 2019-01-21 15:56:04 +02:00
Martin Kroeker 010d59bfee
Merge pull request #1973 from martin-frbg/issue1464
Increase Zen SWITCH_RATIO to 16
2019-01-20 20:30:11 +01:00
Martin Kroeker 83b5c6b92d
Fix compilation with NO_AVX=1 set
fixes #1974
2019-01-20 12:18:53 +01:00
Martin Kroeker bbfdd6c0fe
Increase Zen SWITCH_RATIO to 16
following GEMM benchmarks on Ryzen2700X. For #1464
2019-01-19 23:01:31 +01:00
Martin Kroeker cda81cfae0
Shift transition to multithreading towards larger matrix sizes
See #1886 and JuliaRobotics issue 500. trsm benchmarks on Haswell and Zen showed that with these values performance is roughly doubled for matrix sizes between 8x8 and 14x14, and still 10 to 20 percent better near the new cutoff at 32x32.
2019-01-19 00:10:01 +01:00
Martin Kroeker 32b0f1168e
Fix declaration of input arguments in the Sandybridge GER microkernels (#1967)
* Tag arguments 0 and 1 as both input and output
2019-01-18 08:11:39 +01:00
Martin Kroeker b495e54310
Fix declaration of input arguments in the x86_64 SCAL microkernels (#1966)
* Tag arguments 0 and 1 as both input and output (see #1964)
2019-01-18 08:11:07 +01:00
Martin Kroeker d5e6940253
Fix declaration of input arguments in the x86_64 microkernels for DOT and AXPY (#1965)
* Tag operands 0 and 1 as both input and output

For #1964 (basically a continuation of coding problems first seen in #1292)
2019-01-17 23:20:32 +01:00
Martin Kroeker 24e697eadb
Merge pull request #1970 from quickwritereader/develop
crot fix
2019-01-17 16:42:11 +01:00
Martin Kroeker 3e9fd6359d
Bump xcode version to 10.1 to make sure it handles AVX512 2019-01-17 16:19:03 +01:00
Ubuntu 43a4572038 crot fix 2019-01-17 14:45:31 +00:00
Martin Kroeker 256eb588bb
Merge pull request #1963 from quickwritereader/develop
Blas1 single missing kernels implemented with vector builtins
2019-01-16 18:41:03 +01:00
Abdelrauf a034e65512
Merge branch 'develop' into develop 2019-01-16 19:25:13 +04:00
Ubuntu 8c3386be87 Added missing Blas1 single fp {saxpy, caxpy, cdot, crot(refactored version of srot),isamax ,isamin, icamax, icamin},
Fixed idamin,icamin choosing the first occurance index of equal minimals
2019-01-16 15:16:21 +00:00
Andrew 3e601bd419 disable NaN checks before BLAS calls dgemm.R 2019-01-16 11:54:22 +02:00
Andrew 478d3c4569 disable NaN checks before BLAS calls deig.R (shorten matrix def) 2019-01-16 11:41:46 +02:00
Andrew 3afceb6c2a disable NaN checks before BLAS calls deig.R 2019-01-16 11:38:14 +02:00
Andrew 7af8b21dbb disable NaN checks before BLAS calls dsolve.R (shorter formula) 2019-01-16 11:34:46 +02:00
Martin Kroeker 1e3ada6db4
Merge pull request #1960 from cnjsdfcy/Hygon
Add support for Hygon Dhyana
2019-01-16 10:27:14 +01:00
Andrew 2777a7f506 disable NaN checks before BLAS calls dsolve.R (shorter config part) 2019-01-16 11:23:51 +02:00
Andrew b70fd23836 disable NaN checks before BLAS calls dsolve.R 2019-01-16 11:18:54 +02:00
Andrew def0385caa init 2019-01-16 09:51:29 +02:00
caiyu 29dc72889f Add support for Hygon Dhyana 2019-01-16 14:25:19 +08:00
maamountki b815a04c87
[ZARCH] fix a bug in max/min functions 2019-01-15 21:04:22 +02:00
Martin Kroeker dbc9a060ef
Fix missing braces in support_av() call 2019-01-14 22:41:31 +01:00
Martin Kroeker 00401489c2
Fix missing braces in support_avx() 2019-01-14 22:38:32 +01:00
maamountki 1a7925b3a3
[ZARCH] Update dgemv_n_4.c 2019-01-11 17:43:11 +02:00
maamountki 406f835f00
[ZARCH] update cgemv_n_4.c 2019-01-11 17:39:17 +02:00
maamountki 621dedb37b
[ZARCH] Update cgemv_t_4.c 2019-01-11 17:37:11 +02:00
maamountki b731e8246f
Update sgemv_t_4.c 2019-01-11 17:14:04 +02:00
maamountki ecc31b743f
Update dgemv_t_4.c 2019-01-11 17:13:02 +02:00
maamountki 5d89d6b143
[ZARCH] fix sgemv_n_4.c 2019-01-11 17:08:24 +02:00
maamountki 67432b23c2
[ZARCH] fix cgemv_n_4.c 2019-01-11 16:44:46 +02:00
Martin Kroeker 21c0f2af7b
Merge pull request #1957 from martin-frbg/issue1954
Move TLS key deletion to openblas_quit
2019-01-10 12:04:08 +01:00
Martin Kroeker ad2c386d6a
Move TLS key deletion to openblas_quit
fixes #1954 (as suggested by thrasibule in that issue)
2019-01-10 00:32:50 +01:00
maamountki be66f5d5c2
[ZARCH] fix data prefetch type in sdot 2019-01-09 16:50:07 +02:00
maamountki c2ffef8156
[ZARCH] fix data prefetch type in ddot 2019-01-09 16:49:44 +02:00
maamountki e7455f500c
[ZARCH] fix dsdot.c 2019-01-09 16:33:54 +02:00
maamountki 3eafcfa650
[ZARCH] fix cgemv_n_4.c 2019-01-09 07:43:45 +02:00
Martin Kroeker 8d99dba86b
Merge pull request #1949 from martin-frbg/issue1947
Query AVX2 and AVX512VL support when selecting x86 kernels
2019-01-08 20:44:08 +01:00
Martin Kroeker 1650311246
Bump xcode to 8.3 2019-01-08 14:43:45 +01:00
Martin Kroeker cf5d48e833
Update OSX environment to Sierra
as homebrew seems to have dropped support for El Capitan in their gcc packages
2019-01-08 14:41:48 +01:00
Martin Kroeker 191677b902
Add travis_wait to the OSX brew install phase 2019-01-08 10:46:47 +01:00
Martin Kroeker 31ed19e8b9
Add message for SkylakeX and KNL fallbacks to Haswell 2019-01-05 19:41:13 +01:00