Commit Graph

9225 Commits

Author SHA1 Message Date
pengxu
57bb46bedf Loongarch64: fixed rot_lasx 2025-04-30 16:42:22 +08:00
pengxu
6dc4ca2391 Loongarch64: fixed icamax_lasx 2025-04-30 16:42:12 +08:00
pengxu
b528b1b8ea Loongarch64: fixed iamax_lasx 2025-04-30 16:41:58 +08:00
pengxu
ba9569e382 Loongarch64: fixed dot_lasx 2025-04-30 16:41:48 +08:00
pengxu
dc5fa29851 Loongarch64: fixed cscal_lasx 2025-04-30 16:41:39 +08:00
pengxu
a98dd6d911 Loongarch64: fixed copy_lasx 2025-04-30 16:41:28 +08:00
pengxu
d49319c2d2 Loongarch64: fixed cnrm2_lasx 2025-04-30 16:41:18 +08:00
pengxu
74c97ef814 Loongarch64: fixed cdot_lasx 2025-04-30 16:41:05 +08:00
pengxu
be525521ad Loongarch64: fixed asum_lasx 2025-04-30 16:40:55 +08:00
pengxu
0cd5ca5527 Loongarch64: fixed amax_lasx 2025-04-30 16:40:44 +08:00
Martin Kroeker
70dff3b84f Merge pull request #5242 from abhishek-iitmadras/abhishekk_dot
optimise dot using thread throttling for NEOVERSE V1
2025-04-23 13:53:55 -07:00
abhishek-fujitsu
0c239c9d48 update contribution list 2025-04-23 22:35:05 +05:30
abhishek-fujitsu
9c02cdb073 optimise dot using thread throttling for NEOVERSE V1 2025-04-23 22:35:05 +05:30
Martin Kroeker
d0e8fd6d40 Merge pull request #5239 from annop-w/gemv_n_sve
Use SVE kernel for S/DGEMVN for SVE machines
2025-04-22 10:19:49 -07:00
Martin Kroeker
ddfefd9bf8 Merge pull request #5240 from iha-taisei/fixedIssue5231
Fix: Potential out-of-bounds read in non-transposed [SD]GEMV kernels for A64FX and Neoverse V1.
2025-04-22 06:04:26 -07:00
Iha, Taisei
08b5c18d70 fixed a potential out-of-bounds on gemv. 2025-04-22 19:56:44 +09:00
Annop Wongwathanarat
e11744a411 Use SVE kernel for S/DGEMVN for SVE machines 2025-04-22 09:40:13 +00:00
Martin Kroeker
db0abfa907 Merge pull request #5238 from martin-frbg/revert5125
remove non-vectorized SGEMV transpose reduce path for POWER8, restoring optimizations frpm PR4880
2025-04-22 02:12:19 -07:00
Martin Kroeker
7389b6c483 Merge pull request #5237 from martin-frbg/revert5219
Fix and reinstate the Cooper Lake/Sapphire Rapids microkernel for non-transpose SBGEMV
2025-04-21 23:36:23 -07:00
Martin Kroeker
4ec62d7f73 remove non-vectorized code path for power8, restoring PR4880 2025-04-21 23:14:10 +02:00
Martin Kroeker
1df8738f27 Merge pull request #5235 from quickwritereader/issue_unaligned_ppc64le
Explicit unaligned vector load/stores in PPC64LE GEMV kernels
2025-04-21 14:03:56 -07:00
Martin Kroeker
99d9f1ff38 Fix conditional 2025-04-21 22:55:45 +02:00
Martin Kroeker
96d80801bc Reinstate the CooperLake microkernel 2025-04-21 22:53:26 +02:00
Martin Kroeker
f5bc97c37e Merge pull request #5227 from zanpeeters/develop
Wrong output from getarch on Apple M4
2025-04-21 10:28:53 -07:00
Martin Kroeker
050c3b26ae Merge pull request #5236 from ywwry66/apple_workaround
Follow-up to #5233, fixing "Argument list too long"
2025-04-21 03:04:14 -07:00
Ruiyang Wu
9aa7a0b2a7 Follow-up to d659f3c 2025-04-21 00:22:11 -04:00
Martin Kroeker
94fceaeac5 Merge pull request #5233 from ywwry66/apple_workaround
Fix "Argument list too long" compilation error for Intel macOS
2025-04-20 13:38:43 -07:00
Ruiyang Wu
d659f3c3f6 Fix "Argument list too long" compilation error for Intel macOS 2025-04-20 11:57:55 -04:00
Martin Kroeker
2e4309315c Merge pull request #5219 from martin-frbg/sbgemvn_cooper
Temporarily disable the Cooper Lake/Sapphire Rapids microkernel for non-transpose SBGEMV
2025-04-20 07:29:20 -07:00
Martin Kroeker
afc1dc69cd Merge pull request #5234 from RevySR/bump-xuantie-qemu
Bump xuantie qemu for c910v
2025-04-20 07:28:21 -07:00
Ubuntu
0cc2485594 Explicit unaligned vector load/stores in PPC64LE GEMV kernels 2025-04-20 08:00:29 +00:00
Han Gao
1f687b2f60 Bump xuantie qemu for c910v
Signed-off-by: Han Gao <rabenda.cn@gmail.com>
2025-04-20 14:20:49 +08:00
Martin Kroeker
dd38b4e811 Merge pull request #5225 from annop-w/gemv_n
Improve performance for SGEMVN on NEONVERSEN1
2025-04-17 01:54:10 -07:00
Martin Kroeker
3a088de2d1 Merge pull request #5228 from martin-frbg/cmakecrossarm
Update and amend parameters for Neoverse cpus in CMake crossbuilds
2025-04-17 00:51:26 -07:00
Martin Kroeker
0241d516f6 Merge pull request #5220 from iha-taisei/sdgemv_n_unroll
Further performance improvements to non-transposed [SD]GEMV kernels for A64FX and Neoverse V1.
2025-04-16 12:55:55 -07:00
Martin Kroeker
afb664527f Merge pull request #5221 from tetsuzo-usui/tune_symv_for_arm64
Add AArch64-optimized SYMV kernels
2025-04-16 10:01:38 -07:00
Annop Wongwathanarat
d535728803 Improve performance for SGEMVN on NEONVERSEN1 2025-04-16 09:54:30 +00:00
Martin Kroeker
d9369bda1e Update and amend parameters for Neoverse cpus 2025-04-16 01:09:57 -07:00
zanpeeters
acef78c778 Reset buffer length before every call to sysctlbyname. 2025-04-15 17:17:17 -07:00
zanpeeters
d1c2528aed Add L1_DATA_LINESIZE for ifdef __APPLE__ 2025-04-15 17:14:19 -07:00
zanpeeters
7b66330dea hw.perflevel[01].cpusperl changed to hw.perflevel[01].cpusperl2 2025-04-15 17:12:03 -07:00
Usui, Tetsuzo
d711906e3e Add symv kernels for arm64 2025-04-11 20:39:52 +09:00
Iha, Taisei
f1e628b889 Further performance improvements to [SD]GEMV. 2025-04-11 20:00:33 +09:00
Martin Kroeker
39718cd28e Merge pull request #5218 from martin-frbg/lapacke_mangling
lapacke_mangling.h is no longer generated, so don't delete on make clean
2025-04-10 22:40:10 -07:00
Martin Kroeker
211dfd0754 disable the CooperLake microkernel as it produces wrong results 2025-04-10 22:21:57 +02:00
Martin Kroeker
fd3afef122 lapacke_mangling.h is no longer generated, so don't delete on make clean 2025-04-10 22:09:19 +02:00
Martin Kroeker
b30dc9701f Merge pull request #5215 from annop-w/gemv_t
Use SVE kernel for S/DGEMVT for SVE machines
2025-04-10 13:06:07 -07:00
Martin Kroeker
2893d0add4 Merge pull request #5211 from guoyuanplct/develop
Optimizing the Implementation of GEMV on the RISC-V V Extension
2025-04-10 09:43:03 -07:00
Martin Kroeker
ed1e470663 Merge pull request #5217 from haampie/hs/fix/darwin-gcc
test_potrs.c: do not use GCC pragma on darwin-aarch64
2025-04-10 08:19:46 -07:00
Harmen Stoppels
3d6d026fe1 no-gcse when loongarch64 2025-04-10 15:44:31 +02:00