floraachy/OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
8,525 Commits 47 Branches 64 Tags
b613754143d68a2635f0232f29a04318f15e34d3
Commit Graph

8525 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Martin Kroeker
b613754143 Update scal..c 2024-07-24 14:31:29 +02:00
Martin Kroeker
f5d04318e3 Merge branch 'OpenMathLib:develop' into scalfixes 2024-07-21 13:43:43 +02:00
Martin Kroeker
73f8866ffb make NAN handling depend on DUMMY2 parameter 2024-07-21 13:42:47 +02:00
Martin Kroeker
dfbc2348a8 fix NAN handling 2024-07-20 18:27:15 +02:00
Martin Kroeker
c064319ecb fix alpha=NAN case 2024-07-20 17:42:31 +02:00
Martin Kroeker
c2ffd90e8c make NAN handling depend on dummy2 parameter 2024-07-20 17:31:00 +02:00
Martin Kroeker
a815594fd1 Merge pull request #4801 from markdryan/markdryan/riscv-dynamic-arch 
Add autodetection for riscv64
 2024-07-19 17:12:07 +02:00
Martin Kroeker
dd6c33d34d make NAN handling depend on dummy2 parameter 2024-07-19 16:14:55 +02:00
Martin Kroeker
5a845ef1f4 Merge pull request #4809 from penghongbo/reorder_gemm_gemvt 
Change computational order in GEMV and GEMM Power6 kernel
 2024-07-19 13:06:42 +02:00
Hong Bo Peng
db98f8753f Try to fix LAPACK testing failures on P7. 
1. Remove the FADD insn from the GEMV Transpose code.
  2. Remove the FADD insn from GEMM and ZGEMM code.
  3. Reorder the compution of the Imaginary part in ZGEMM code.
 2024-07-19 02:08:19 -04:00
Martin Kroeker
2020569705 fix NAN handling and make it depend on dummy2 parameter 2024-07-17 23:55:54 +02:00
Martin Kroeker
3870995f01 make NAN handling depend on dummy2 parameter 2024-07-17 23:54:24 +02:00
Martin Kroeker
7284c533b5 make NAN handling depend on dummy2 parameter 2024-07-17 23:50:40 +02:00
Martin Kroeker
73751218a4 make NAN handling depend on dummy2 parameter 2024-07-17 23:41:26 +02:00
Martin Kroeker
b9bfc8ce09 make NAN handling depend on dummy2 parameter 2024-07-17 23:29:50 +02:00
Martin Kroeker
eb4879e04c make NAN handling depend on the dummy2 parameter 2024-07-17 23:24:19 +02:00
Martin Kroeker
ee87cb90d0 Merge pull request #4803 from iha-taisei/SVESupportSDGEMV 
A64FX: Add support for SVE to SGEMV/DGEMV kernels.
 2024-07-17 23:14:21 +02:00
Martin Kroeker
e9f6aa46a4 Merge pull request #4800 from vlad0x00/patch-2 
Add missing parentheses
 2024-07-16 16:32:04 +02:00
Martin Kroeker
b1aa2e1768 Merge pull request #4802 from markdryan/markdryan/rvv_axpby_incy0 
Fix axpby_rvv kernels for cases where inc_y = 0
 2024-07-16 14:22:38 +02:00
iha fujitsu
0985fdc82b A64FX: Add support for SVE to SGEMV/DGEMV kernels. 2024-07-16 17:31:33 +09:00
Vladimir Nikolić
56e1782ffb Add another missing parenthesis 2024-07-15 15:15:23 -07:00
Vladimir Nikolić
127ea5d0d9 Add missing parenthesis 2024-07-15 15:12:21 -07:00
Martin Kroeker
a3c10c6c25 Merge pull request #4799 from martin-frbg/issue4762 
Improve the error message for (p)thread creation failure
 2024-07-15 20:57:56 +02:00
Martin Kroeker
a373d0f107 Improve the error message for thread creation failure 2024-07-15 18:32:21 +02:00
Mark Ryan
67bf4b6998 Fix axpby_rvv kernels for cases where inc_y = 0 
The following openblas_utest tests fail when the RISCV64_ZVL128B is
enabled.

TEST 89/103 axpby:zaxpby_inc_0 [FAIL]
TEST 92/103 axpby:caxpby_inc_0 [FAIL]
TEST 95/103 axpby:daxpby_inc_0 [FAIL]
TEST 98/103 axpby:saxpby_inc_0 [FAIL]

The issue is that the vectorized kernels do not work when inc_y == 0.
This patch updates the kernels to fall back to the scalar algorithms
when inc_y == 0, fixing the failing tests.

Signed-off-by: Mark Ryan <markdryan@rivosinc.com>
 2024-07-15 14:24:47 +00:00
Mark Ryan
3b715e6162 Add autodetection for riscv64 
Implement DYNAMIC_ARCH support for riscv64.  Three cpu types are
supported, riscv64_generic, riscv64_zvl256b, riscv64_zvl128b.
The two non-generic kernels require CPU support for RVV 1.0 to
function correctly.  Detecting that a riscv64 device supports
RVV 1.0 is a little complicated as there are some boards on the
market that advertise support for V via hwcap but only support
RVV 0.7.1, which is not binary compatible with RVV 1.0.  The
approach taken is to first try hwprobe.  If hwprobe is not
available, we fall back to hwcap + an additional check to distinguish
between RVV 1.0 and RVV 0.7.1.

Tested on a VM with VLEN=256, a CanMV K230 with VLEN=128 (with only
the big core enabled), a Lichee Pi with RVV 0.7.1 and a VF2 with no
vector.

A compiler with RVV 1.0 support must be used to build OpenBLAS for
riscv64 when DYNAMIC_ARCH=1.

Signed-off-by: Mark Ryan <markdryan@rivosinc.com>
 2024-07-15 14:24:22 +00:00
Martin Kroeker
6013b36b16 Merge pull request #4796 from martin-frbg/ppcbuf 
Suffix BUFFER_SIZEs on POWER as UL to prevent int overflow in computations
 2024-07-12 11:06:45 +02:00
Martin Kroeker
9789034281 Merge branch 'OpenMathLib:develop' into ppcbuf 2024-07-12 11:05:46 +02:00
Martin Kroeker
5d08ec7ff3 Merge pull request #4782 from martin-frbg/azurewincl 
Fix NAN handling in ARM/generic SCAL; have AzureCI Windows show errors on failure
 2024-07-11 23:55:15 +02:00
Martin Kroeker
dfc11ef248 Merge pull request #4791 from ChipKerchner/vectorizeSBGEMMincopy 
Vectorize SBGEMM incopy for Power10 - 4x faster.
 2024-07-11 21:38:57 +02:00
Martin Kroeker
2fefdfa2b8 Merge branch 'OpenMathLib:develop' into azurewincl 2024-07-11 21:38:21 +02:00
Martin Kroeker
475bd2452b Suffix BUFFERSIZEs as UL to prevent int overflow in computations 2024-07-11 20:13:57 +02:00
Martin Kroeker
b70227ad62 Merge pull request #4795 from pkubaj/patch-1 
Fix build on FreeBSD/powerpc64*
 2024-07-11 19:00:07 +02:00
Martin Kroeker
8277828fdc Merge pull request #4785 from rgommers/docs-install 
Rewrite "Install OpenBLAS" docs page
 2024-07-11 18:49:07 +02:00
Martin Kroeker
f0fc7249f1 Merge pull request #4792 from martin-frbg/issue4790 
Fix core assignment in cpu detection for Intel family 15
 2024-07-11 17:38:43 +02:00
Martin Kroeker
362856fece Merge pull request #4778 from JAicewizard/develop 
Add support for RISCV64_GENERIC in cmake
 2024-07-11 15:12:46 +02:00
Martin Kroeker
1d77647d1b Merge pull request #4769 from drupol/fix-buffersize-value 
openblas: fix `BUFFERSIZE` value
 2024-07-11 14:45:50 +02:00
Piotr Kubaj
4c12090776 Fix build on FreeBSD/powerpc64* 2024-07-10 22:21:48 +00:00
Chip Kerchner
f708944fea Add all 4 variations of the SBGEMM to compare_sgemm_sbgemm 2024-07-10 13:07:48 -05:00
Martin Kroeker
e706bc1ec0 Fix core assignment for Intel family 15 2024-07-09 20:22:56 +02:00
Chip Kerchner
cb154832f8 Vectorize SBGEMM incopy - 4x faster. 2024-07-09 13:10:03 -05:00
Martin Kroeker
a5c04e326a Update scal.c 2024-07-04 22:28:01 +02:00
Ralf Gommers
268dcd8f45 docs: convert remaining install sections (Android, iOS, FreeBSD, Cortex-M) 2024-07-04 19:16:30 +02:00
Ralf Gommers
452014341e docs: rework building from source on Windows section 2024-07-04 19:16:30 +02:00
Ralf Gommers
4547908901 docs: rewrite "Install OpenBLAS" page (part 1: binaries, basic from source) 2024-07-04 19:15:51 +02:00
Martin Kroeker
e1eef56e05 Merge pull request #4783 from martin-frbg/cpuid_meteor 
Add another CPUID for Intel Meteor Lake
 2024-07-04 18:09:27 +02:00
Martin Kroeker
536200bc9e fix handling of INF or NAN 2024-07-04 17:47:19 +02:00
Martin Kroeker
3063d03021 Add another CPUID for Meteor Lake 2024-07-04 16:05:05 +02:00
Martin Kroeker
b422742899 collect error output from ctest, if any 2024-07-04 15:42:34 +02:00
Jaap Aarts
cea4abcac0 Fix compiling on mingw 2024-07-04 14:56:16 +02:00