Martin Kroeker
3318a2b904
override CDOT and ZDOT with the generic C kernel
1 year ago
davidz-ampere
84730068af
reduce duplicate kernel code
1 year ago
Martin Kroeker
85337c5160
Merge pull request #5310 from nakagawa-fj/bugfix/identify_cpu_part_for_arm64
Bug Fix: Problem with identifying some ARM64 processors
1 year ago
Martin Kroeker
53cd6e7ff7
Update Changelog.txt
1 year ago
Masato Nakagawa
1dd396033a
Fix:Problem with identifying some ARM64 processors.
1 year ago
Srangrang
3b1ac29b77
disable BUILD_HFLOAT16
1 year ago
davidz-ampere
be68ef03b4
Add support for Ampere processors
1 year ago
Martin Kroeker
f1097d1cba
Merge pull request #5306 from martin-frbg/lapack1131
Fix missing initialization leading to bypassing corner cases in C/ZGEQP3RK (Reference-LAPACK PR #1131 )
1 year ago
Martin Kroeker
3fe7f196e6
Update the Changelog for version 0.3.30
1 year ago
Martin Kroeker
bad47bd024
Fix too strict leading dimensions check in LAPACKE_?gesdd_work (Reference-LAPACK PR #1126 ) ( #5307 )
* relax leading dimensions check (Reference-LAPACK PR #1126 )
1 year ago
Martin Kroeker
7f3093a0ad
Merge pull request #5305 from martin-frbg/lapack1135
Fix 2nd dimension used by LAPACKE_c/zunmlq in NaN check and transposition (Reference-LAPACK PR #1135 )
1 year ago
Martin Kroeker
1804ff58d7
fix missing initialization
1 year ago
Martin Kroeker
906b9df316
fix missing initialization
1 year ago
Martin Kroeker
f4e5177050
fix dimension used in nancheck (Reference-LAPACK PR 1135)
1 year ago
Martin Kroeker
2a6beac88f
fix dimension used in transposition (Reference-LAPACK PR 1135)
1 year ago
Martin Kroeker
d8a2324699
fix dimension used in nancheck (Reference-LAPACK PR 1135)
1 year ago
Martin Kroeker
874744976c
fix dimension used in nancheck (Reference-LAPACK PR 1135)
1 year ago
Srangrang
9f13b2c6ac
style: modify HALF to BFLOAT16 in benchmark folder
1 year ago
Srangrang
ec14e1648c
fix: resolve non-RISCV host build failed issue
- adjust interface to disable "small matrix" pathway
- separate HFLOAT16 from BFLOAT16
- remove SHGEMM_UNROLL_M and SHGEMM_UNROLL_N equal conditions
Related to PR#5290
Co-authored-by Martin
1 year ago
Martin Kroeker
0ea173ec8c
Merge pull request #5304 from martin-frbg/fixgemmtr_if
fix source file used for sbgemmt/sbgemmtr in CMake builds
1 year ago
Martin Kroeker
5e393f207c
fix source file used for sbgemmt/sbgemmtr
1 year ago
Martin Kroeker
162591af47
Merge branch 'OpenMathLib:develop' into issue5289
1 year ago
Martin Kroeker
dbd5643d37
Merge pull request #5302 from martin-frbg/zscal_mips_3
mips64 SICORTEX: temporarily change default C/ZSCAL to the non-asm implementation
1 year ago
Martin Kroeker
31ef2cbbb3
Exit if memory allocation keeps failing, instead of looping forever
1 year ago
Martin Kroeker
e338d34ce1
fix path
1 year ago
Martin Kroeker
d36093d084
temporarily change default C/ZSCAL to the non-asm implementation
1 year ago
Martin Kroeker
cc4b04a684
Merge pull request #5301 from martin-frbg/zscal_mips_2
kernel/mips(64): Fix cscal and zscal
1 year ago
Martin Kroeker
b3c90564d7
resync with the generic arm version for inf/nan handling
1 year ago
Martin Kroeker
6bdc7f9eb7
Merge pull request #5300 from martin-frbg/fixup5296
kernel/riscv64: Fix cscal/zscal for riscv64_generic
1 year ago
Martin Kroeker
63272b6c82
Merge pull request #5299 from martin-frbg/x86_64-ssezscal
Disable the default SSE kernels for x86_64 CSCAL/ZSCAL for now
1 year ago
Martin Kroeker
73af02b89f
use dummy2 as Inf/NAN handling flag
1 year ago
Martin Kroeker
549a9f1dbb
Disable the default SSE kernels for CSCAL/ZSCAL for now
1 year ago
Martin Kroeker
ca1ce84ee5
Merge pull request #5298 from martin-frbg/fixup5281
Fix PR5281 "kernel/arm64: fix cscal/zscal"
1 year ago
Martin Kroeker
58eeb9041c
fix handling of dummy2
1 year ago
Martin Kroeker
7c77537b25
Merge pull request #5297 from martin-frbg/zscal_x86_sparc
kernel/(x86|sparc): Fix cscal and zscal by reverting to the generic C kernels
1 year ago
Martin Kroeker
63287e1855
Merge pull request #5296 from martin-frbg/zscal_riscv
kernel/riscv64: Fix cscal and zscal
1 year ago
Martin Kroeker
d2855d3dab
Merge pull request #5285 from martin-frbg/zscal_zarch
kernel/zarch: Fix cscal and zscal
1 year ago
Martin Kroeker
1408be5fe0
Merge pull request #5282 from martin-frbg/zscal_power
kernel/power: Fixed cscal and zscal
1 year ago
Martin Kroeker
1589d0b21e
Merge pull request #5281 from martin-frbg/zscal_arm64
kernel/arm64: fixed cscal and zscal
1 year ago
Martin Kroeker
a86419fb66
Merge pull request #5280 from martin-frbg/zscal_x86_64
kernel/x86_64: fixed cscal and zscal
1 year ago
Martin Kroeker
11ff18bb0f
Merge pull request #5081 from XiWeiGu/kernel_generic_fixed_cscal_zscal
kernel/generic: Fixed cscal and zscal
1 year ago
Martin Kroeker
2e2691b34b
Merge pull request #5078 from XiWeiGu/la64_fixed_cscal_zscal
LoongArch64: fixed cscal and zscal
1 year ago
Martin Kroeker
f4194fc65f
Merge branch 'develop' into la64_fixed_cscal_zscal
1 year ago
Martin Kroeker
e12132abd4
Use generic C/ZSCAL kernels to address inf/nan handling for now
1 year ago
Martin Kroeker
1cefbea7ea
Use generic SCAL kernels to address inf/nan handling for now
1 year ago
Sharif Inamdar
8279e68805
Optimize gemv_n_sve_v1x3 kernel
- Calculate predicate outside the loop
- Divide matrix in blocks of 3
1 year ago
Martin Kroeker
f18b7a46bf
add dummy2 flag handling for inf/nan agnostic zeroing
1 year ago
Martin Kroeker
fe220a0d7d
Merge pull request #5291 from guoyuanplct/develop
kernel/riscv64:fixed the performance problem in RISCV64_ZVL256 when OPENBLAS_K is small
1 year ago
Martin Kroeker
bbdc265798
Merge pull request #5294 from arnej27959/arnej/fix-arm64-register
Accumulate results in output register explicitly
1 year ago
Arne Juul
5442aff218
Accumulate results in output register explicitly
1 year ago