5f200dca5
Merge pull request #5166 from martin-frbg/issue5158 by
2025-03-03 16:02:02 +0100
8b98db13e
Merge pull request #5167 from taoye9/fix_sbgemv_n_kernel_typo by
2025-03-03 14:47:53 +0100
6b8b35cdf
(refs/pull/5167/head)
fix minior issues of redeclaration of float x0,x1 in sbgemv_n_neon.c by
2025-03-03 11:55:27 +0000
38ee7c930
(refs/pull/5168/head)
Add dispatch of SBGEMVNKERNEL for NEOVERSEN2 and NEOVERSEV2 by
2025-03-03 11:30:45 +0000
d3325b23b
deploy: 217324d880 by
2025-03-03 07:10:42 +0000
217324d88
Merge pull request #5162 from taoye9/add_sbgemv_tests by
2025-03-03 08:10:12 +0100
289ba9cf7
deploy: e4630ed15a by
2025-03-03 00:02:04 +0000
e4630ed15
Merge pull request #5160 from taoye9/sbgemv_n_neon by
2025-03-02 23:50:42 +0100
35914aa9a
(refs/pull/5166/head)
Expose the option to build without LAPACKE to ccmake by
2025-03-02 22:54:59 +0100
2b941c44b
(refs/pull/5160/head)
Merge branch 'develop' into sbgemv_n_neon by
2025-03-02 22:39:32 +0100
55ee48b7f
deploy: c797e27a1c by
2025-03-02 21:23:52 +0000
c797e27a1
Merge pull request #5159 from annop-w/sbgemv_t_bfdot by
2025-03-02 22:23:19 +0100
4346b9155
(refs/pull/5162/head)
add beta and alpha testcase for sbgemv by
2025-02-28 13:17:46 +0000
35bdbca15
Add sbgemv_n_neon kernel for arm64. by
2025-02-27 18:15:17 +0000
747ec6d22
(refs/pull/5161/head)
add beta and alpha testcase for sbgemv by
2025-02-28 13:17:46 +0000
bb540dccd
Add sbgemv_n_neon kernel for arm64. by
2025-02-27 18:15:17 +0000
edaf51dd9
(refs/pull/5159/head)
Add sbgemv_t_bfdot kernel for ARM64 by
2025-02-26 12:47:11 +0000
8094b18ab
(refs/pull/4963/merge)
Merge 2f251c16fc into ef9e3f7159 by
2025-02-27 12:18:01 +0530
949b09f13
deploy: ef9e3f7159 by
2025-02-25 13:01:49 +0000
ef9e3f715
Merge pull request #5149 from martin-frbg/fixup5077-5088 by
2025-02-25 14:01:13 +0100
09ba09946
(refs/pull/5149/head)
make throttling code conditional on SMP by
2025-02-25 12:10:48 +0100
030ae1fd9
(refs/pull/5150/head)
Redefined threading logic for WoA by
2025-02-25 15:40:39 +0530
918e65c47
deploy: 1533fe49be by
2025-02-24 15:07:42 +0000
1533fe49b
Merge pull request #5144 from taoye9/dispatch_neoversve2_to_neoversven2 by
2025-02-24 16:07:06 +0100
db3a7a056
deploy: c03a81b927 by
2025-02-23 11:16:45 +0000
c03a81b92
Merge pull request #5141 from michalowski-arm/fork-throttle by
2025-02-23 12:16:09 +0100
c32dabd7d
deploy: 643966d9c7 by
2025-02-22 20:57:40 +0000
643966d9c
Merge pull request #5146 from martin-frbg/issue5123 by
2025-02-22 21:57:09 +0100
77fba0f40
(refs/pull/5146/head)
Fix "dummy2" flag handling by
2025-02-22 20:09:21 +0100
692794751
(refs/pull/5145/head)
Run CI on Github-hosted Arm instances too by
2025-02-21 21:56:24 +0000
f0bea79a6
(refs/pull/5144/head)
dispatch NEOVERSEV2 to NEOVERSEN2 under dynamic setting by
2025-02-21 10:03:50 +0000
5515d5086
deploy: 20d1118865 by
2025-02-21 08:21:14 +0000
20d111886
Merge pull request #5143 from martin-frbg/issue5111 by
2025-02-21 09:20:39 +0100
75b958a01
(refs/pull/5143/head)
Transform the B array back if necessary before returning by
2025-02-20 23:54:12 +0100
650a062e1
(refs/pull/5141/head)
Add thread throttling profile for SGEMV on `NEOVERSEV2` by
2025-02-20 10:19:40 +0000
b723c1b7b
Add thread throttling profile for SGEMM on `NEOVERSEV2` by
2025-02-20 10:18:47 +0000
ceb8f1e34
Merge pull request #5140 from martin-frbg/issue5139 by
2025-02-19 18:17:15 +0100
806073ccb
(refs/pull/4080/head)
utest: test fork safety on OpenMP >= 5 by
2023-06-10 21:20:58 +0300
f677f4f29
blas_thread_shutdown: release OpenMP resources too by
2023-06-10 20:35:59 +0300
c5f0dcf72
c_check: test for omp_pause_resource_all() by
2025-02-19 16:55:49 +0300
f1fa37057
(refs/pull/5140/head)
fix missing endif by
2025-02-19 15:22:26 +0100
6d1444be3
Add ARM64 options for NVIDIA HPC by
2025-02-19 14:26:43 +0100
f71ac9297
deploy: eb84aac7ad by
2025-02-19 09:57:25 +0000
eb84aac7a
Merge pull request #5084 from quic/topic/sgemm_direct_sme1 by
2025-02-19 10:56:49 +0100
ef6ffcb56
deploy: abbd78aa59 by
2025-02-18 08:54:07 +0000
abbd78aa5
Merge pull request #5138 from martin-frbg/issue5131 by
2025-02-18 09:53:31 +0100
ebcab9097
(refs/pull/5138/head)
Handle flang-new runtime library linking on Linux like classic-flang by
2025-02-17 23:12:58 +0100
4626f4fa3
deploy: ed1584666c by
2025-02-17 06:37:54 +0000
ed1584666
Merge pull request #5137 from martin-frbg/issue5136 by
2025-02-17 07:37:07 +0100
b9ae246f2
(refs/pull/5137/head)
define USE_TRMM for RISCV64 targets as well by
2025-02-16 23:18:04 +0100
86cf9d8a2
Merge pull request #5133 from OpenMathLib/revert-4920-issue4917 by
2025-02-16 19:16:43 +0100
e3a46cc7a
deploy: 0b3c56968d by
2025-02-16 18:16:40 +0000
0b3c56968
Merge pull request #5135 from martin-frbg/ghwf-n2 by
2025-02-16 19:16:10 +0100
c1bb90a82
(refs/pull/5135/head)
remove the express NeoverseN2 target from the Cobalt100 job by
2025-02-16 14:23:07 +0100
041b617e4
(refs/pull/5134/head)
revert change from PR 4920 by
2025-02-15 23:23:30 +0100
77c638db6
(refs/pull/5133/head, revert-4920-issue4917)
Revert "Fix potential inaccuracy in multithreaded level3 related to SWITCH_RATIO" by
2025-02-15 20:37:48 +0100
8b2c70515
(refs/pull/5129/head)
add `NEOVERSEV2` in DYNAMIC_ARCH to avoid `NEOVERSEV2` SBGEMM falling to `NEOVERSEV1` SBGEMM kernel by
2025-02-12 12:08:59 +0000
f66ca05b3
(refs/pull/5084/head)
Merge branch 'develop' into topic/sgemm_direct_sme1 by
2025-02-13 14:54:37 +0530
d23eb3b93
Support for SME1 based sgemm_direct kernel for cblas_sgemm level 3 API by
2024-12-05 11:41:05 +0530
a64b75a2e
Merge pull request #5127 from Harishmcw/gesv-threshold by
2025-02-12 22:02:37 +0100
453efbd10
Merge pull request #5128 from martin-frbg/issue5120 by
2025-02-12 21:02:06 +0100
877d5a5be
(refs/pull/5128/head)
Add -O2 to flang flags when building on WoA in Release mode by
2025-02-12 17:01:06 +0100
8d487ef6e
Merge pull request #5124 from XiWeiGu/LoongArch64-LA264-lapack-fixed by
2025-02-12 14:58:30 +0100
daf16b822
(refs/pull/5127/head)
Adjusted GESV threading logic for optimal performance on WoA by
2025-02-12 12:10:57 +0530
4bbcf1afa
deploy: 9a3948df82 by
2025-02-12 11:50:57 +0000
e8b11a126
Merge pull request #5125 from martin-frbg/issue5122 by
2025-02-12 12:50:44 +0100
9a3948df8
Merge pull request #5126 from martin-frbg/cirrusbsd4 by
2025-02-12 12:50:21 +0100
7f1f776f5
(refs/pull/5126/head)
Update FreeBSD jobs to 14.2 by
2025-02-12 11:23:02 +0100
81eed868b
(refs/pull/5125/head)
Restore the non-vectorized code from before PR4880 for POWER8 by
2025-02-12 09:07:20 +0100
98b5ef929
Restore the non-vectorized code from before PR4880 for POWER8 by
2025-02-12 09:04:22 +0100
2c4a5cc6e
(refs/pull/5124/head)
LoongArch64: Fixed snrm2_lsx.S and cnrm2_lsx.S by
2025-02-12 14:59:39 +0800
9e75d6b3d
LoongArch64: Fixed swap_lsx.S by
2025-02-12 14:57:35 +0800
e8c740368
LoongArch64: Fixed rot_lsx.S ane crot_lsx.S by
2025-02-12 14:52:49 +0800
c2212d0ab
LoongArch64: Fixed copy_lsx.S by
2025-02-07 18:02:04 +0800
7f1ebc7ae
LoongArch64: Fixed iamax_lsx.S by
2025-02-06 16:52:06 +0800
31d326f89
LoongArch64: Fixed dot_lsx.S by
2025-01-20 10:45:20 +0800
5d6356bc1
LoongArch64: Fixed amax_lsx.S by
2025-01-20 10:45:01 +0800
f42ce7067
Merge pull request #5116 from martin-frbg/issue5110 by
2025-02-09 23:17:20 +0100
7478c1026
(refs/pull/5116/head)
Merge branch 'OpenMathLib:develop' into issue5110 by
2025-02-09 21:40:02 +0100
f9e49a1a1
deploy: c54f5417cc by
2025-02-09 20:40:00 +0000
c54f5417c
Merge pull request #5118 from martin-frbg/zrot_utestext by
2025-02-09 21:39:30 +0100
57208b8bc
(refs/pull/5118/head)
Disable tests with incx,incy=0 (undefined behavior) by
2025-02-09 20:17:29 +0100
3a4a9b21e
Disable tests with incx,incy=0 (undefined behavior) by
2025-02-09 20:16:03 +0100
60d0be0e9
Update nrm2.c by
2025-02-08 23:42:21 +0100
0fd5448b2
Handle INCX=0 by
2025-02-08 19:33:05 +0100
1b85b6a39
Merge pull request #5108 from taoye9/sbgemm_neoversev1 by
2025-02-07 20:30:41 +0100
d2e32fac5
deploy: cae480683a by
2025-02-07 08:38:28 +0000
cae480683
Merge pull request #5113 from martin-frbg/issue5112 by
2025-02-07 09:37:53 +0100
db7e5f1fa
(refs/pull/5113/head)
Update gemmt.c by
2025-02-06 21:26:20 +0100
ff30ac966
Update Makefile by
2025-02-06 19:51:23 +0100
7c3e169b6
Update gemmt.c by
2025-02-06 19:21:08 +0100
09414a418
Ensure that GEMMTR name appears in XERBLA if gemmt was called as such by
2025-02-06 18:52:00 +0100
ed00b0853
(refs/pull/5130/head)
fix regression issue by
2025-02-05 23:41:58 -0500
c748e6a33
(refs/pull/5108/head)
optimized sbgemm kernel for neoverse-v1 (sve-256) by
2024-12-02 17:03:10 +0000
4379a6fbe
* checkpoint sbgemm for SVE-256 by
2024-11-05 16:22:45 +0000
dd3a6acd5
deploy: c139b63342 by
2025-02-02 07:13:18 +0000
c139b6334
Merge pull request #5107 from jhgit/develop by
2025-02-02 08:12:45 +0100
6cd9bbe53
(refs/pull/5107/head)
fix signedness of pointer to integer type passed to blas_lock() by
2025-02-01 17:16:05 -0700
e6f54572d
deploy: 5de5072940 by
2025-01-30 15:56:05 +0000
5de507294
Improve flang-new identification and add CI job for it on OSX-x86_64 (#5103) by
2025-01-30 16:55:26 +0100