floraachy
/
OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
5,592 Commits 47 Branches 64 Tags 71 MiB
Commit Graph

5592 Commits

Author SHA1 Message Date
Martin Kroeker 3de80b3f5a
Merge pull request #2713 from RajalakshmiSR/p10-gcc10 
Change minimum gcc version for POWER10
 2020-07-10 10:43:33 +02:00
Rajalakshmi Srinivasaraghavan af1e140e35 Change minimum gcc version for POWER10 
As the MMA patches for POWER10 are backported to gcc10.2, changing
the minimum gcc version needed to build OpenBLAS for POWER10.
 2020-07-09 21:46:06 -05:00
Martin Kroeker d4a0299e16
Do not build lapack-test on MSVC for now (same as with BLAS test) 2020-07-09 13:57:27 +02:00
Martin Kroeker f766024749
enable fortran for cmake 2020-07-09 13:44:25 +02:00
Martin Kroeker c502760bef
Modify for building with OpenBLAS 2020-07-09 13:13:16 +02:00
Martin Kroeker 29b5887d5f
Modify for building with OpenBLAS 2020-07-09 13:12:35 +02:00
Martin Kroeker 60188a8c82
Append crude hack for enabling lapack tests in the OpenBLAS build 2020-07-09 11:44:31 +02:00
Martin Kroeker 1d63631afe
Add lapack-test 2020-07-09 11:42:02 +02:00
Martin Kroeker e82bb953a7
Merge pull request #2708 from RajalakshmiSR/p10_future 
Changing mcpu option as power10
 2020-07-08 12:26:44 +02:00
Martin Kroeker ed7e155c35
Merge branch 'develop' into aix 2020-07-07 18:52:06 +02:00
Rajalakshmi Srinivasaraghavan 45d819ca82 Changing mcpu option as power10 
As compiler enabled mcpu option as power10, changing it from future.
 2020-07-07 11:25:20 -05:00
Martin Kroeker 8751a69271
Obtain actual cpu count on AIX and suppress spurious NO_AVX512 on non-x86 2020-07-07 15:46:32 +02:00
Jussi Enkovaara 10a2923f64 fixes #2238 
Always obey omp_get_max_threads() when build with USE_OPENMP
 2020-07-07 13:35:43 +03:00
Martin Kroeker 5ff83a4261
Merge pull request #2670 from mhillenibm/dumpfullversion_on_gcc7 
RFC: Use -dumpfullversion to get minor version on gcc-7 and newer
 2020-07-07 00:12:28 +02:00
Martin Kroeker 5bc9680a86
Merge pull request #2703 from martin-frbg/issue2702 
Compatibility fix for gcc < 4.7
 2020-07-02 22:32:51 +02:00
Martin Kroeker 4ab3651591
Option -mavx2 requires at least gcc 4.7 2020-07-02 17:00:15 +02:00
Martin Kroeker a83680b40b
Merge pull request #69 from xianyi/develop 
rebase
 2020-07-02 16:56:00 +02:00
Martin Kroeker c3aa036e99
Merge pull request #2693 from EGuesnet/AIX-build-on-POWER8-32bits 
AIX build on POWER8 32bits
 2020-07-01 08:29:52 +02:00
EGuesnet 634e1305f9
Update cgemm_kernel_8x4_power8.S 2020-06-30 15:16:39 +02:00
Martin Kroeker c467516132
Merge pull request #2688 from martin-frbg/cometlake 
Add autodetection of Intel Comet Lake H and S models
 2020-06-27 17:47:24 +02:00
Martin Kroeker 83f4746825
Add support for Comet Lake H and S 2020-06-27 14:41:24 +02:00
Martin Kroeker 584ef8d4ae
Add support for Comet Lake H & S 2020-06-27 14:36:37 +02:00
Martin Kroeker 8dfda02e89
Merge pull request #68 from xianyi/develop 
rebase
 2020-06-27 14:29:29 +02:00
Martin Kroeker 28d69e0097
Merge pull request #2687 from martin-frbg/utfbom 
Strip UTF8 byte order marker from source files
 2020-06-26 22:53:09 +02:00
Martin Kroeker c2467c9619
Merge pull request #2686 from RajalakshmiSR/p10_shgemm 
powerpc: Optimized SHGEMM kernel for POWER10
 2020-06-26 22:52:45 +02:00
Martin Kroeker f86e749df4
Merge pull request #2683 from mtreinish/add-comet-lake-support 
Add cpu detection support for comet lake U
 2020-06-26 12:11:03 +02:00
Martin Kroeker d199c2787d
Merge pull request #2680 from kavanabhat/aix_makefile_fix 
Fix for #2671
 2020-06-26 11:27:28 +02:00
Martin Kroeker e30ad0e521
Strip UTF8 byte order marker from source 2020-06-26 09:00:43 +02:00
Rajalakshmi Srinivasaraghavan d23419accc powerpc: Optimized SHGEMM kernel for POWER10 
This patch introduces new optimized version of SHGEMM kernel
using power10 Matrix-Multiply Assist (MMA) feature introduced in
POWER ISA v3.1. This patch makes use of new POWER10 compute instructions
for matrix multiplication operation.

Tested on simulator and there are no new test failures.
 2020-06-25 22:19:08 -05:00
Matthew Treinish 2f9c10810c
Also set CPUTYPE in get_cpuname() 2020-06-25 15:53:56 -04:00
Matthew Treinish f37e941d52
Add support to driver/others/dynamic.c too 2020-06-25 11:56:49 -04:00
Matthew Treinish 2a91452bdd
Add cpu detection support for comet lake U 
Comet Lake U CPUs have family: 6, model: 6, extended family: 0, and
extended model: 10 were not being correctly detected by GETARCH during
openblas builds and would show CORE=UNKNOWN and LIBCORE=unknown. This
commit adds the necessary information to cpuid_x86 to detect extended
family 10 model 6 and return the proper core information. It's
essentially just a skylake cpu, not skylake x, so I just took the used
the same return fields as skylake.
 2020-06-25 11:32:09 -04:00
Martin Kroeker c854ef5471
Fix variable names in conditional 2020-06-25 13:29:52 +02:00
Martin Kroeker c0afc11742
Fix POWERPC builds on AIX (gcc/gfortran 7) 
1. macro preprocessing for POWER8 and later kernels only
2. default buffer size used by AIX version of m4 is too small
 2020-06-25 13:12:36 +02:00
Martin Kroeker c592f0f80a
Fix utest build on AIX 2020-06-25 12:58:13 +02:00
Martin Kroeker 3f613b1301
Tentative changes for building on AIX 2020-06-25 12:57:00 +02:00
Martin Kroeker 72a0ec8e75
Fix reading of CPU name from prtconf output on AIX 2020-06-25 12:55:10 +02:00
Martin Kroeker 3446e58daf
Fix handling of uname output on AIX 2020-06-25 12:31:35 +02:00
Martin Kroeker 4ca8becc4b
Merge pull request #67 from xianyi/develop 
rebase
 2020-06-25 10:33:03 +02:00
Martin Kroeker dfe819f3bd
Merge pull request #2679 from RajalakshmiSR/P10_GEMM 
POWER10 Optimized GEMM kernels
 2020-06-25 08:31:38 +02:00
Martin Kroeker 4369e52555
Merge pull request #2677 from brada4/develop 
address vs2019 C4293 warning
 2020-06-25 08:31:17 +02:00
Gordon Fossum bb2f52844b powerpc: Optimized ZGEMM kernel for POWER10 
This patch introduces new optimized version of ZGEMM kernel
using power10 Matrix-Multiply Assist (MMA) feature introduced in
POWER ISA v3.1. This patch makes use of new POWER10 compute instructions
for matrix multiplication operation.

Tested on simulator and there are no new test failures.
Cycles count reduced by 30-50%  compared to POWER9 version depending on
M/N/K sizes.
 2020-06-24 14:50:12 -05:00
Rajalakshmi Srinivasaraghavan 571eadb880 powerpc: Optimized SGEMM/DGEMM/CGEMM for POWER10 
This patch introduces new optimized version of SGEMM, CGEMM and DGEMM
using power10 Matrix-Multiply Assist (MMA) feature introduced in
POWER ISA v3.1. This patch makes use of new POWER10 compute instructions
for matrix multiplication operation.

Tested on simulator and there are no new test failures.
Cycles count reduced by 30-50%  compared to POWER9 version depending on
M/N/K sizes.
MMA GCC patch for reference:
https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=8ee2640bfdc62f835ec9740278f948034bc7d9f1
 2020-06-24 14:48:15 -05:00
Kavana Bhat df4ade070f Fix for #2671 2020-06-24 04:25:47 -05:00
User User-User e6b9275034 address vs2019 C4293 2020-06-24 09:12:23 +03:00
Martin Kroeker 53ea5bfece
Merge pull request #66 from xianyi/develop 
rebase
 2020-06-23 10:13:44 +02:00
Martin Kroeker 93592d1260
Merge pull request #2675 from wjc404/develop 
AVX512 DGEMM TCOPY_16 Function
 2020-06-23 09:29:02 +02:00
Martin Kroeker 6eaeb01263
Merge pull request #2658 from RajalakshmiSR/p10 
powerpc: Add support for future processor
 2020-06-23 00:02:37 +02:00
Martin Kroeker 45d542c9d1
Merge pull request #65 from xianyi/develop 
rebase
 2020-06-21 12:41:01 +02:00
wjc404 086d87a302
AVX512 dgemm tcopy_16 function 2020-06-20 00:07:43 +08:00