a495ffc554
Rework multithreading threshold
2023-03-26 00:23:57 +01:00
1659750861
Add ARMV5 to CMAKE prebuild.cmake for cross-compiles ( #3959 )
...
* Add ARMV5
2023-03-26 00:22:06 +01:00
7f56c31e21
Merge pull request #3958 from tttapa/develop
...
Add ARMv6 support to cmake/prebuild.cmake
2023-03-25 19:34:57 +01:00
cf80350b89
Merge pull request #3957 from martin-frbg/issue3954
...
Fix detection of Cray CCE after recent detection changes for gfortran on Cray
2023-03-25 18:33:58 +01:00
a79ae116e2
Add ARMV6 to cmake/prebuild.cmake,
...
based on cpuid_arm.c and param.h
2023-03-25 16:30:42 +01:00
280c8469d4
Fix detection of Cray CCE after detection changes for gfortran on Cray
...
CCE15 no longer produces a .ident line, so look for HPE name in a note section as well
2023-03-25 14:18:21 +01:00
c68afeb567
Merge pull request #3956 from martin-frbg/issue3955
...
Update Reference-LAPACK README to 3.11.0
2023-03-24 23:06:25 +01:00
c2b01c6f81
Update to 3.11.0
2023-03-24 22:54:43 +01:00
02920be53a
Merge pull request #3953 from martin-frbg/issue3801
...
Do not use multithreading for small SYMV/SYR2K workloads
2023-03-24 13:51:37 +01:00
244147495a
Do not use multithreading for small workloads
2023-03-23 23:13:02 +01:00
2885e4d898
Merge pull request #3951 from martin-frbg/issue3940
...
Do not change the global OMP_NUM_THREADS
2023-03-22 15:14:00 +01:00
e09e494f25
Merge pull request #3950 from martin-frbg/issue3856-syr
...
Re-enable overriding the LAPACK SYMV,SYR,SPMV and SPR implementations
2023-03-22 10:42:57 +01:00
dac2d7aef0
Merge pull request #3952 from martin-frbg/cleanup3936
...
Avoid unused variable warning in non-OpenMP builds
2023-03-22 09:00:14 +01:00
743ae5ab18
move declaration of openmp_nthreads under the related ifdef
2023-03-21 21:07:37 +01:00
579bc86671
remove call to omp_set_num_threads
2023-03-21 20:58:56 +01:00
ab32f832a8
fix stray blank on continuation line
2023-03-21 08:29:05 +01:00
e359787e28
restore C/Z SPMV, SPR, SYR,SYMV
2023-03-21 07:43:03 +01:00
6ca6eb7613
Re-enable overriding the LAPACK SYMV,SYR,SPMV and SPR implementations
2023-03-20 23:04:12 +01:00
2b77f28810
Merge pull request #3949 from martin-frbg/lapack750
...
Improve inline documentation of ?GEJSV (Reference-LAPACK PR 750)
2023-03-20 22:48:06 +01:00
164eafd61d
Improve inline documentation of ?GEJSV (Reference-LAPACK PR 750)
2023-03-20 19:51:07 +01:00
871b730dc5
Create SECURITY.md
2023-03-20 15:12:35 +01:00
7719dbecde
Merge pull request #3946 from martin-frbg/lapack682
...
Rewrite ?LAQR5 and S/DHGEQZ , add tests for TRECV3 (Reference-LAPACK PR 682)
2023-03-20 13:48:57 +01:00
147e2fbf87
Add numerical tests for TRECV3 (Reference-LAPACK 682)
2023-03-20 10:04:05 +01:00
2a83ec1f79
Rewrite to use FMA with Householder reflectors
2023-03-20 10:00:42 +01:00
e1c3c34178
Standardize coding style (Reference-LAPACK 682)
2023-03-20 09:59:03 +01:00
f7b9391119
Merge pull request #3945 from ahojnnes/user/joschonb/fix-prebuild-path
...
Wrap variables in cmake/prebuild in quotes to handle spaces in paths
2023-03-19 23:14:18 +01:00
fbfe474583
f
2023-03-19 16:23:44 +01:00
ec4700087e
Wrap variables in cmake/prebuild in quotes to handle spaces in paths
2023-03-19 16:09:47 +01:00
e8db1fe89b
Merge pull request #3943 from martin-frbg/llvm15
...
Add clang option to avoid running out of registers in AVX512 assembly
2023-03-18 11:24:52 +01:00
de937b3194
Add clang option to avoid running out of registers in AVX512 assembly
2023-03-17 21:22:37 +01:00
f3d21039ce
Improve fix from PR3924 ( #3941 )
...
* compare denominator against DBL_MIN rather than a somewhat arbitrary small number near it
2023-03-16 15:09:32 +01:00
8d6813ec41
Merge pull request #3938 from martin-frbg/issue3937
...
Fix CMAKE-based cross-compilation to CortexA53 (wrong DGEMM setting)
2023-03-10 15:33:07 +01:00
19d6795122
Merge pull request #3936 from martin-frbg/issue3933
...
Observe any stricter (OpenMP) thread count limit imposed by openblas_set_num_threads()
2023-03-10 15:32:50 +01:00
50c263716e
Correct DGEMM_UNROLL_M value for A53 in cross-compile
2023-03-10 11:53:24 +01:00
e298d613fa
initialize status variable for openblas_set_num_threads
2023-03-08 23:43:15 +01:00
05aa88268f
add status variable for openblas_set_num_threads
2023-03-08 23:41:57 +01:00
9f9d0012a3
observe thread limit imposed by openblas_set_num_threads()
2023-03-08 23:38:34 +01:00
fe5d3ca8e0
Merge pull request #3935 from martin-frbg/omp_place_num
...
Fix OpenMP thread counting returning places rather than cores
2023-03-08 22:41:18 +01:00
e38ab079a0
Fix OpenMP thread counting returning places rather than cores
2023-03-08 19:17:33 +01:00
9feaaa3f39
Merge pull request #3932 from martin-frbg/issue3931
...
Handle unrecognized ASM compiler (from Arm Compiler 22.1) in CMAKE builds
2023-03-03 12:01:25 +01:00
8272dfc552
Handle unrecognized ASM compiler (from Arm Compiler 22.1)
2023-03-03 00:21:59 +01:00
f616c86404
Merge pull request #3930 from sergei-lewis/dot-kernel-early-bail
...
dot.c early bail fix
2023-03-02 16:46:25 +01:00
cb0a70e0e2
dot.c early bail fix
2023-03-02 09:51:10 +00:00
5925178d03
Merge pull request #3924 from martin-frbg/numpy22025
...
Avoid overflow from division in GETF2 potentially causing NaN
2023-02-27 15:59:44 +01:00
f58080278f
Merge pull request #3923 from xctan/fix-cmake-riscv64
...
Add missing RISC-V architecture in arch.cmake
2023-02-27 09:39:30 +08:00
3d27cbd9a3
avoid overflow in division
2023-02-26 23:44:14 +01:00
a39ced0551
avoid overflow in division
2023-02-26 23:42:20 +01:00
6a0de3aa39
Add missing RISC-V architecture in arch.cmake
...
RISC-V support exists in Makefile.system but is missing in arch.cmake. This patch adds riscv64 platform support to cmake building system just like 039e27545f/Makefile.system (L830-L832)
2023-02-26 20:21:57 +08:00
039e27545f
Merge pull request #3915 from martin-frbg/issue3910
...
Fix DYNAMIC_ARCH builds that select only a subset of precisions
2023-02-24 07:41:33 +01:00
38d6fb4225
Fix dependencies in builds with specified subsets of precision types
2023-02-23 23:12:06 +01:00
Martin Kroeker
Pieter Pas
Johannes Schönberger
Sergei Lewis
Zhang Xianyi
xctan