floraachy/OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
8,147 Commits 47 Branches 64 Tags
58659730a4c2aa03e2eb0427f0170810d2fedffb
Commit Graph

8147 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Martin Kroeker
58659730a4 Merge pull request #4595 from martin-frbg/issue4583-2 
Temporarily revert S/DNRM2 on NeoverseN1 and Apple M to the older NEON kernel
 2024-04-02 17:06:03 +02:00
Martin Kroeker
9ead81bd39 Revert S/DNRM2 to the base NEON kernel to fix precision loss 2024-04-02 15:59:20 +02:00
Martin Kroeker
29995b2969 Merge pull request #4594 from mattip/openblas.pc.il 
fix typo
 2024-04-02 08:27:43 +02:00
Matti Picus
5b33e648b4 fix typo 2024-04-02 08:54:56 +11:00
Martin Kroeker
b1e8ba5017 Merge pull request #4587 from mseminatore/win_queue_fix 
Address Windows thread server re-entrant queue bug #4582
 2024-03-29 14:36:21 +01:00
Martin Kroeker
8267fcfda8 Merge pull request #4588 from XiWeiGu/loongarch_fixed_dzamax 
loongarch: Fixed dzamax
 2024-03-29 13:48:27 +01:00
Mark Seminatore
b0ad8a78ff code to fix lost work in case of re-entrant calls to exec_blas_async() 2024-03-28 15:24:52 -07:00
Martin Kroeker
e1638ea43a Merge pull request #4586 from martin-frbg/potrf-para 
use atomic acq/rel operations in potrf_parallel as in the corresponding getrf_parallel
 2024-03-28 14:51:20 +01:00
Martin Kroeker
2dda40d280 use atomic operations as in the corresponding getrf 2024-03-28 11:33:31 +01:00
gxw
96607cbb98 loongarch: Fixed dzamax 
Initialize the registers to prevent sporadic errors.
 2024-03-25 23:17:53 -04:00
Martin Kroeker
9af2a9dc3b Merge pull request #4579 from ChipKerchner/fixInializerPriority 
Fix global (static) constructor priority so that OpenBLAS gets initialized before other libraries.  Other unit test AIX fix.
 2024-03-25 22:36:29 +01:00
Chip Kerchner
0e0d0bce1a Fix global (static) constructor priorty so that OpenBLAS gets initialized before other libraries. Other unit test AIX fix. 2024-03-25 15:11:55 -05:00
Martin Kroeker
4059a75c9c Merge pull request #4578 from jerryz123/patch-1 
Fix README formatting error
 2024-03-25 17:17:01 +01:00
Jerry Zhao
0b814ab8b9 Fix README formatting error 2024-03-25 08:14:00 -07:00
Martin Kroeker
87f83ebe9c Merge pull request #4575 from martin-frbg/fixup4503 
Restore outer loop of blas_buffer_inuse setup for parallel OpenMP
 2024-03-24 20:00:09 +01:00
Martin Kroeker
88b5330ae7 Restore outer loop of blas_buffer_inuse setup 2024-03-24 18:33:21 +01:00
Martin Kroeker
52b71a1673 Filter out FFLAGS that flang-new from LLVM18 no longer supports (#4569) 
* Filter out FFLAGS that flang-new from LLVM18 no longer supports
 2024-03-22 17:02:39 +01:00
Martin Kroeker
3d2a9e4a61 Merge pull request #4567 from RajalakshmiSR/P9AIX 
POWER9: Use default param values from POWER8 on AIX
 2024-03-21 00:12:37 +01:00
Martin Kroeker
9ad9b52884 Merge pull request #4568 from martin-frbg/fixbenchloops 
Fix bad assignment of OPENBLAS_LOOPS  variable in several benchmarks
 2024-03-21 00:09:34 +01:00
Martin Kroeker
3f1ec74fe7 Fix OPENBLAS_LOOPS assignment 2024-03-20 19:22:48 +01:00
Martin Kroeker
fe39c891a6 Fix OPENBLAS_LOOPS assignment 2024-03-20 19:21:37 +01:00
Martin Kroeker
ffcbaca167 Fix OPENBLAS_LOOPS assignment 2024-03-20 19:20:16 +01:00
Martin Kroeker
05d0438c25 Fix OPENBLAS_LOOPS assignment 2024-03-20 19:19:11 +01:00
Rajalakshmi Srinivasaraghavan
f5b2a877e2 POWER9: Use default param values from POWER8 on AIX 
AIX uses KERNEL.POWER8 optimization on POWER9 and changing
the default GEMM parameters in param.h to use POWER8 values
on POWER9.
 2024-03-20 10:17:49 -05:00
Martin Kroeker
b4a1153648 Merge pull request #4566 from XiWeiGu/fix_loongarch_lsx 
LoongArch: Fixed  LSX opt
 2024-03-19 10:21:21 +01:00
gxw
50869f6ca8 loongarch: Fixed zrot LSX opt 2024-03-19 10:08:11 +08:00
gxw
b5eb9d6bac loongarch: Fixed {sc/dz}amax LSX opt 2024-03-19 09:56:11 +08:00
gxw
ad13e04669 loongarch: Fixed {s/d/sc/dz}amin LSX opt 2024-03-19 09:18:44 +08:00
gxw
bbf82cb624 loongarch: Fixed {s/d}axpby LSX opt 2024-03-18 17:51:42 +08:00
gxw
ac460eb42a loongarch: Fixed i{c/z}amin LSX opt 2024-03-18 17:15:58 +08:00
Martin Kroeker
56d114b245 Merge pull request #4565 from martin-frbg/issue4564 
Fix argument lists of RELAPACK_?gemmt for good
 2024-03-17 20:41:58 +01:00
Martin Kroeker
2e9ce9bb07 Fix argument lists of RELAPACK_?gemmt for good 2024-03-17 19:20:19 +01:00
Martin Kroeker
79cb121ab9 Merge pull request #4563 from XiWeiGu/loongarch_fix_lasx 
Loongarch: Fixed LASX opt
 2024-03-16 10:34:32 +01:00
gxw
60e251a1f8 loongarch: Fixed {sc/dz}amax LASX opt 2024-03-16 14:52:17 +08:00
gxw
a10dde5554 loongarch: Fixed {s/d/sc/dz}amin LASX opt 2024-03-16 14:52:14 +08:00
gxw
6534d378b7 loongarch: Fixed {s/d/c/z}sum LASX opt 2024-03-16 14:52:10 +08:00
gxw
6159cffc58 loongarch: Fixed i{s/c/z}amin LASX opt 2024-03-16 14:52:06 +08:00
gxw
7d755912b9 loongarch: Fixed {s/d/c/z}axpby LASX opt 2024-03-16 14:51:56 +08:00
Martin Kroeker
66bde6243e Merge pull request #4503 from shivammonaka/OpenMP-Locks 
OpenMP locks instead of busy-waiting with NUM_PARALLEL
 2024-03-14 20:56:04 +01:00
Martin Kroeker
dc0338af47 Merge pull request #4560 from martin-frbg/issue4551-3 
Add support for negative increments to the ?NRM2 kernels for RISC-V RVV targets
 2024-03-13 14:48:56 +01:00
Martin Kroeker
cf80bd8500 Update nrm2_rvv.c 2024-03-13 13:07:26 +01:00
Martin Kroeker
9baa757905 Update nrm2_vector.c 2024-03-13 11:40:14 +01:00
Martin Kroeker
18a6db6862 Update nrm2_vector.c 2024-03-13 11:10:26 +01:00
Martin Kroeker
855bbdda4f Merge pull request #4556 from ChipKerchner/updateREADMEAIX 
Update README for build instructions on AIX and OpenXL.
 2024-03-12 23:12:50 +01:00
Martin Kroeker
3752e73919 handle incx < 0 2024-03-12 20:44:01 +01:00
Martin Kroeker
db70c7f7fb handle incx < 0 2024-03-12 20:42:11 +01:00
Martin Kroeker
dee8557d58 handle incx < 0 2024-03-12 20:40:29 +01:00
Martin Kroeker
d9dff17aec handle incx < 0 2024-03-12 20:38:23 +01:00
Martin Kroeker
5802e7a62f Merge pull request #4559 from martin-frbg/issue4551-2 
Remove another unwanted early exit in the ThunderX2/NeoN1/AppleM ?NRM2 kernels
 2024-03-12 20:34:20 +01:00
Martin Kroeker
552c521353 remove another early exit for incx < 0 2024-03-12 18:49:27 +01:00