Mark Ryan
e0b610d01f
Harmonize riscv64 LIBNAME for forced and non-forced targets
The forced values for LIBNAME were either riscv64_generic or c910v
while the non-forced value of LIBNAME was always riscv64.
2 years ago
Mark Ryan
ec2aa32eb0
Fix crash in cpuid_riscv64.c
The crash is reproducible when building OpenBLAS without forcing a
target in a riscv64 container running on an X86_64 machine with an
older version of QEMU, e.g., 7.0.0, registered with binfmt_misc to
run riscv64 binaries. With this setup, cat /proc/cpuinfo in the
container returns the cpu information for the host, which contains a
"model name" string, and we execute the buggy code. The code in
question is searching in an uninitialised buffer for the ':' character
and doesn't check to see whether it was found or not. This can result
in pmodel containing the pointer value 1 and a crash when pmodel is
defererenced. The algorithm to detect the C910V CPU has not been
modified, merely fixed to prevent the crash.
A few additional checks for NULL pointers are added to improve the
robustness of the code and a whitespace error is corrected.
2 years ago
Martin Kroeker
d6a5174e9c
Merge pull request #4447 from RevySR/update-thead-toolchains
Update T-Head toolchains v2.8.0
2 years ago
Han Gao/Revy/Rabenda
304a9b60af
Update T-Head toolchains v2.8.0
Signed-off-by: Han Gao/Revy/Rabenda <rabenda.cn@gmail.com>
2 years ago
Martin Kroeker
f5de4fad27
Merge pull request #4444 from Mousius/part-mapping
Add dynamic support for Arm(R) Neoverse(TM) V2 processor
2 years ago
Chris Sidebottom
aaf65210cc
Add dynamic support for Arm(R) Neoverse(TM) V2 processor
Whilst I figure out how best to map the L2 parameters without
duplicating all of `ARMV8SVE`, lets just map this to `NEOVERSEV1`.
2 years ago
Martin Kroeker
500442cf96
Merge pull request #4442 from pbo-linaro/fix-utest-compilation
Fix utest compilation
2 years ago
Pierrick Bouvier
a4992e09bc
Fix utest compilation
Introduced recently when adding new test cases for ZSCAL
- include cblas is needed for cblas_zscal
- ASSERT macro does not exist
- missing closing )
2 years ago
Martin Kroeker
6f0e0e4021
Merge pull request #4438 from Dirreke/csky-support
Add CSKY support
2 years ago
Martin Kroeker
43cb266178
Merge pull request #4441 from martin-frbg/gemv-threshold
Increase multithreading threshold for S/DGEMV by a factor of 50
2 years ago
Martin Kroeker
d2fc4f3b4d
Increase multithreading threshold by a factor of 50
2 years ago
Martin Kroeker
ec46ca7a43
Support Arm Compiler for Linux as classic flang ( #4436 )
* Support ArmCompilerforLinux as classic flang
2 years ago
Dirreke
ec89466e14
Add CSKY support
2 years ago
Martin Kroeker
452741b67f
Merge pull request #4435 from imciner2/im/sapphire
Fix Clang sapphire rapids march flag
2 years ago
Ian McInerney
8f4e325ea8
Fix Clang sapphire rapids march flag
2 years ago
Martin Kroeker
13c764eaaa
Merge pull request #4434 from martin-frbg/issue4433
Only use mtune=native in ARM64 fallback paths when not cross-compiling
2 years ago
Martin Kroeker
025a1b2c7b
Only use mtune=native when not cross-compiling
2 years ago
Martin Kroeker
2527afaaa2
Merge pull request #4429 from martin-frbg/issue4428
Handle NAN and INF in ARM and generic/s390x ZSCAL
2 years ago
Martin Kroeker
0d2e486edf
Handle NAN and INF
2 years ago
Martin Kroeker
a782103b9c
Merge pull request #4425 from martin-frbg/issue2392
Add BLAS extension openblas_set_num_threads_local()
2 years ago
Martin Kroeker
152a6c43b6
Add blas_omp_threads_local
2 years ago
Martin Kroeker
8a9d492af7
Add default for blas_omp_threads_local
2 years ago
Martin Kroeker
b3341527ad
Merge pull request #4426 from martin-frbg/issue4415
Tweak LAPACK tests for SGS/DGS to avoid spurious errors resulting from FMA-induced inaccuracies
2 years ago
Martin Kroeker
9fab60d32f
Remove matrix dimension 6 from SGS to avoid spurious errors from FMA
2 years ago
Martin Kroeker
bf66af3dc0
remove matrix dimension 6 from DGS to avoid spurious errors from FMA
2 years ago
Martin Kroeker
87d31af2ae
Add openblas_set_num_threads_local()
2 years ago
Martin Kroeker
2e2e538b7c
Add openblas_set_num_threads_local() and use of blas_omp_threads_local in OMP parallel regions
2 years ago
Martin Kroeker
f9b2d7f225
Merge pull request #3253 from wi24rd/patch-1
Fix typo in common.h
2 years ago
Martin Kroeker
5f5b7c4f45
Merge pull request #4423 from martin-frbg/issue4422
Check compiler support for AVX512BF16 and base COL/SPR kernel choice on that
2 years ago
Martin Kroeker
f31bea07dd
Merge pull request #4419 from martin-frbg/issue4413
[WIP] Add fixes and utests for ZSCAL with NaN or Inf arguments
2 years ago
Martin Kroeker
20413ee6ec
Update zscal.c
2 years ago
Martin Kroeker
b57627c27f
Handle NAN and INF
2 years ago
Martin Kroeker
d1ead06bd8
define NAN and INFINITY if needed
2 years ago
Martin Kroeker
995a990e24
Make AVX512 BFLOAT16 kernels conditional on compiler capability
2 years ago
Martin Kroeker
1dada6d65d
Add compiler test and flag for AVX512BF16 capability
2 years ago
Martin Kroeker
7df363e1e2
temporarily disable the MSA C/ZSCAL kernels
2 years ago
Martin Kroeker
3599f2de8b
Merge pull request #4421 from ChipKerchner/power10Copies_DGEMM
Replace two vector loads with one vector pair load and fix endianess of stores - DGEMM PowerPC versions.
2 years ago
Chip-Kerchner
5c5c1a1220
Merge remote-tracking branch 'origin/develop' into power10Copies_DGEMM
2 years ago
Chip-Kerchner
058dd2a4cb
Replace two vector loads with one vector pair load and fix endianess of stores - DGEMM versions.
2 years ago
Martin Kroeker
1c31f56e5a
Handle NAN
2 years ago
Martin Kroeker
7ee1ee38e2
Handle NaN in input
2 years ago
Martin Kroeker
f637e12713
Handle INF and NAN
2 years ago
Martin Kroeker
25b0c48082
Update zscal.c
2 years ago
Martin Kroeker
5e7f714e93
Update zscal.c
2 years ago
Martin Kroeker
cf8b03ae8b
Use NAN rather than SNAN for portability
2 years ago
Martin Kroeker
7a6a24647d
Merge pull request #4420 from martin-frbg/revertstuff
Revert accidental direct commits to develop
2 years ago
Martin Kroeker
f0808d856b
Handle NAN in input
2 years ago
Martin Kroeker
acf17a825d
Handle NAN in input
2 years ago
Martin Kroeker
f052bd4705
revert accidental direct commit to develop
2 years ago
Martin Kroeker
91bbde7f64
revert accidental direct commit to develop
2 years ago