phynexis-deps/OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
10,485 Commits 60 Branches 70 Tags
a0febbbc8d7063ddefafe8b409bc4d599b0b421e
Commit Graph
10485 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
moluopro a0febbbc8d docs: update CI job matrix 
Replace retired Cirrus CI entries with the corresponding GitHub Actions jobs for Apple M1, Neoverse N1, Graviton3, and FreeBSD runners.
 2026-07-06 23:21:00 +08:00
moluopro 2497041827 docs: refresh README CI badges 
Replace the retired Cirrus CI badge with the current GitHub Actions badge, label the Azure Pipelines badge, and remove stale public OSUOSL badges whose job pages no longer resolve.
 2026-07-06 23:21:00 +08:00
Martin Kroeker 6468bcbdad Merge pull request #5889 from moluopro/develop 
ARM64 SME: fix zero-scalar handling and clean up direct kernels
 2026-07-06 16:50:48 +02:00
Martin Kroeker c61f95c8d3 Merge pull request #5888 from martin-frbg/lapack1297 
Relax eigenvalue consistency criterion in testsuite (Reference-LAPACK PR 1297)
 2026-07-06 16:34:53 +02:00
Martin Kroeker a2f9fc83df Merge pull request #5885 from martin-frbg/lapack1317 
Increase deflation tolerance in ?LAED to match recent changes in ?SLASD  (Reference-LAPACK PR 1317)
 2026-07-06 13:09:16 +02:00
moluopro 130102e661 Clean up SSYR2K SME direct warnings 
Mark read-only transpose inputs const and limit sve_cntw and ldb definitions to the variants that use them.
 2026-07-06 16:00:21 +08:00
moluopro 3781e2c15d Clean up SSYRK SME direct warnings 
Mark the read-only B input const and limit sve_cntw and ldb definitions to the variants that use them.
 2026-07-06 15:59:53 +08:00
moluopro edd8ab196d Clean up SSYMM SME preprocessing warnings 
Use const source pointers in the symmetric preprocessing helpers and compile only the LU or LL helper needed by each object variant.
 2026-07-06 15:59:25 +08:00
Martin Kroeker aa8b19de74 Relax eigenvalue consistency test (Reference-LAPACK PR 1297) 2026-07-06 09:43:58 +02:00
moluopro 8cde36c9b8 Use integer round-up in ARM64 SME direct kernels 
The padded SME dimensions are integer quantities, so compute them with integer arithmetic and drop the now-unused math.h include.
 2026-07-06 15:33:29 +08:00
moluopro d49cd021dd Handle zero alpha/beta in SSYR2K SME direct kernel 
Avoid loading C when beta is zero, and skip A/B preprocessing for alpha == 0 or K == 0 by reusing the triangular direct kernel with k = 0.
 2026-07-06 15:30:42 +08:00
moluopro 9925b29db5 Handle zero alpha/beta in SSYRK SME direct kernel 
Avoid loading C when beta is zero, and skip A preprocessing for alpha == 0 or K == 0 by reusing the triangular direct kernel with k = 0.
 2026-07-06 15:30:37 +08:00
moluopro 35a841fd49 Handle zero alpha in SSYMM SME direct kernel 
When alpha is zero, avoid preprocessing the symmetric matrix and reuse the SGEMM alpha/beta direct kernel with k = 0 for the beta-only update.
 2026-07-06 15:30:27 +08:00
moluopro ab27636deb Handle zero alpha/beta in SGEMM SME direct kernel 
Avoid loading C when beta is zero; ZA has already been initialized to zero.

For alpha == 0 or K == 0, skip A preprocessing and reuse the direct kernel with k = 0 to perform only the beta update.
 2026-07-06 15:30:22 +08:00
Martin Kroeker 882526a408 Merge pull request #5886 from martin-frbg/lapack1290 
Avoid intermediate overflow in ?LARFGP (Reference-LAPACK PR 1290)
 2026-07-06 08:59:14 +02:00
Martin Kroeker 170da5f679 Merge pull request #5883 from martin-frbg/lapack1300 
Use QR  for bidiagonal SVD with vectors in ?BDSDC (Reference-LAPACK PR 1300)
 2026-07-06 07:30:06 +02:00
Martin Kroeker ac394175e4 Merge pull request #5887 from martin-frbg/disable_cirrusci 
CI: disable the remaining Cirrus jobs now that the service is gone
 2026-07-06 00:39:32 +02:00
Martin Kroeker cc02866ef1 disable the remaining jobs now that Cirrus is gone 2026-07-06 00:38:31 +02:00
Martin Kroeker 022b1c6a03 Merge branch 'OpenMathLib:develop' into lapack1290 2026-07-05 22:05:59 +02:00
Martin Kroeker 6cd5516da2 Merge branch 'OpenMathLib:develop' into lapack1317 2026-07-05 22:05:40 +02:00
Martin Kroeker baa8c5d0b9 Merge branch 'OpenMathLib:develop' into lapack1300 2026-07-05 22:05:20 +02:00
Martin Kroeker 60d3fbea46 Merge pull request #5884 from martin-frbg/fixtinfo_ci 
CI: update Ubuntu tinfo package version in gh workflow
 2026-07-05 22:04:53 +02:00
Martin Kroeker f3f5fd9116 Avoid intermediate overflow (Reference-LAPACK PR 1290) 2026-07-05 17:43:58 +02:00
Martin Kroeker db20661747 Increase deflation tolerance to match recent ?SLASD changes (Reference-LAPACK PR 1317) 2026-07-05 17:37:17 +02:00
Martin Kroeker 78f06538c7 Merge pull request #5881 from moluopro/develop 
ci, docs: fix documentation text and skip docs-only CI
 2026-07-05 16:43:41 +02:00
Martin Kroeker a55df7c0c6 Update package name in the wget call too 2026-07-05 16:27:41 +02:00
Martin Kroeker a45da9e756 update Ubuntu tinfo package version 2026-07-05 15:04:43 +02:00
Martin Kroeker e70a7a735f Merge pull request #5868 from Vasudeva-bit/macTuneZEN4 
Optimize Zen 4 GEMM macro block sizes (P, Q, R)
 2026-07-05 00:01:20 +02:00
Martin Kroeker 5aa157c2cf Use QR instead of D&C for bidiagonal SVD with vectors (Reference-LAPACK PR 1300) 2026-07-04 23:26:22 +02:00
Martin Kroeker e97eb66c29 Merge pull request #5817 from martin-frbg/bsdqemu 
[WIP] Create QEmu-based FreeBSD runner to replace deprecated Cirrus
 2026-07-04 10:14:18 +02:00
Vasudeva-bit 58846317d9 fix C89 scoping, dynamic R for memory issues, robust zen4/5 check 2026-07-04 11:30:34 +05:30
Martin Kroeker 8f0b0333ba Merge pull request #5876 from kumaraditya303/openmp 
fix thread safety and data races of openmp backend
 2026-07-03 22:22:55 +02:00
moluopro 448622eeea ci: skip non-doc workflows for markdown-only changes 2026-07-04 02:52:02 +08:00
moluopro 45bf57426a ci: restrict docs workflow to docs inputs 2026-07-04 01:20:43 +08:00
moluopro e097d0b32c ci: skip non-doc workflows for docs-only changes 2026-07-04 01:20:37 +08:00
moluopro 1a2386b437 docs: fix miscellaneous documentation text 2026-07-04 00:32:31 +08:00
moluopro 68a988c441 docs: fix distributing guide wording 2026-07-04 00:32:24 +08:00
moluopro ac7fa5c7d3 docs: fix extensions table formatting 2026-07-04 00:32:16 +08:00
Martin Kroeker a8c41f73ed Merge pull request #5878 from moluopro/develop 
docs: fix install guide typos
 2026-07-03 15:28:22 +02:00
Martin Kroeker 34232eb5f4 Merge pull request #5880 from martin-frbg/issue5873-2 
Fix broken conditional preventing ARM64 SME compilation with AppleClang
 2026-07-03 15:24:31 +02:00
Martin Kroeker aedbcbc5a5 Fix conditional 2026-07-03 12:43:42 +02:00
Martin Kroeker e717107ba6 Merge pull request #5874 from jschueller/dev-fix 
Fix ZA tile slice indices in ssyrk SME direct kernel
 2026-07-03 12:40:37 +02:00
Martin Kroeker 8e6088de89 Merge pull request #5877 from hheei/fix/laed3-sqrt-nan-guard 
Fix/laed3 sqrt nan guard
 2026-07-02 20:52:15 +02:00
moluopro 1d7cbce8ba docs: fix install guide typos 2026-07-03 01:50:09 +08:00
hheei aea58b2b91 Add hheei to CONTRIBUTORS.md 2026-07-02 23:03:58 +08:00
hheei 71f2260186 lapack/laed3: guard sqrt(-w[i]) against spuriously positive w[i] 
When eigenvalues are nearly degenerate, numerical accumulation in
the divide-and-conquer merge can push w[i] slightly above zero.
The subsequent sqrt(-w[i]) then produces NaN, which propagates
through the entire eigenvector output of DSYEVD/DSTEDC.

Use fmax(-w[i], 0.0) to safely clamp the sqrt argument, yielding
zero instead of NaN for spuriously positive w[i].  The fmax call
maps to a single vmaxsd instruction on x86_64 with zero overhead.

Add test/test_laed3_nan.c which exercises DSYEVD on a rank-1
perturbed identity matrix designed to create near-degenerate
eigenvalue clusters.
 2026-07-02 23:03:38 +08:00
Kumar Aditya 2d75a67d89 setup CI 2026-07-02 15:50:18 +05:30
Kumar Aditya a2261f9e52 fix thread safety of openmp backend 2026-07-02 15:39:09 +05:30
Julien Schueller 43f0a90e1b Fix ZA tile slice indices in ssyrk SME direct kernel 
The kernel_2x2 function uses 4 ZA tiles (0-3) each with svl slices.
Tiles 0/1 handle rows 0..svl-1 with slice indices 0..svl-1.
Tiles 2/3 handle rows svl..2*svl-1, so their slice indices
must start at 0, i.e. (i - svl) instead of i.

Fix all three tile 2/3 access sites:
- C load into ZA (svwrite_hor_za32_f32_m)
- C writeback for UPPER (svst1_hor_za32)
- C writeback for LOWER (svst1_hor_za32)

Fixes #5873
 2026-07-02 09:29:32 +02:00
Martin Kroeker 0f51e2d22f Merge pull request #5871 from martin-frbg/lapack1163 
Add non-trivial terminating case to ?LARFT (Reference-LAPACK PR 1163)
 2026-07-01 00:42:11 +02:00