Aymen Qader
b036235f37
Add Arm®v9-A architecture SME target
Add a new target, ARMV9SME, for Arm®v9-A architecture systems that
support the Scalable Matrix Extension (SME) [1].
Initially inherits ARMV8SVE settings with updated compiler flags. This
target can only be built with an SME-capable toolchain such as GCC 14 or
LLVM 19.
Includes some initial FEAT_SME2 feature detection on Linux targets via
hwcaps. Target is disabled in DYNAMIC_ARCH builds by default.
This is intended as a base target for SME2 kernels.
[1] https://developer.arm.com/documentation/109246/0100/SME-Overview/SME-and-SME2
1 year ago
Martin Kroeker
89f02ed394
Merge pull request #5014 from martin-frbg/issue5013
Add some missed lapack 3.11+ symbols to gensymbol
1 year ago
Martin Kroeker
61d5aec7c1
remove typo
1 year ago
Martin Kroeker
5aea097df0
add missing lapack 3.11+ symbols
1 year ago
Martin Kroeker
72f7b7011c
Merge pull request #5009 from martin-frbg/pybenchdoc
DOCS, pybench : Add build notes for Windows and flang from gh Discussion 5008
1 year ago
Martin Kroeker
0f8ff82592
Add build notes for Windows and flang from gh Discussion 5008
1 year ago
Martin Kroeker
81666de4ef
Merge pull request #5007 from martin-frbg/issue5006
Revert the NRM2 kernels for NeoverseN2 and ARMV8SVE targets to the generic NEON version
1 year ago
Martin Kroeker
230e665bca
Merge pull request #4996 from iha-taisei/sdgemv_sve_unroll
Loop-unrolled transposed [SD]GEMV kernels for A64FX and Neoverse V1
1 year ago
Martin Kroeker
3345007d8f
retire the thunderx2 NRM2 kernels due to reported inaccuracies and NAN
1 year ago
Martin Kroeker
5fe983db29
retire the thunderx2 nrm2 kernels for now due to NAN and inaccuracies
1 year ago
Martin Kroeker
5dc4d7dd7e
Merge pull request #5005 from martin-frbg/evbarm
Improve support for NetBSD on arm64 (evbarm)
1 year ago
Martin Kroeker
4ba471dd5a
Merge pull request #5003 from mathomp4/bugfix/nag-pic
Fixes for NAG Compiler
1 year ago
Martin Kroeker
a791912cbb
handle uname returning evbarm on NetBSD
1 year ago
Martin Kroeker
1a6ecda398
utilize /proc/cpuinfo on NetBSD too
1 year ago
Matthew Thompson
c4e8bac5a5
Fix indent
1 year ago
Matthew Thompson
d3b2036d49
Move to use ERROR STOP instead of ABORT
1 year ago
Matthew Thompson
35334ed2ea
Fixes for Fortran Standards violations for lapack-netlib
1 year ago
Matthew Thompson
be19966d3b
Fixes for NAG CMake
1 year ago
Martin Kroeker
9c5d20187b
Merge pull request #4999 from dg0yt/macro-failed
Fix redefinition of FAILED
1 year ago
Matthew Thompson
2eaf285de5
Use F_COMPILER name
1 year ago
Matthew Thompson
a8b1705dbd
CMake build has wrong PIC flag for NAG
1 year ago
Martin Kroeker
5f65846691
Merge pull request #4998 from dg0yt/arm-type-function
arm: Declare symbols as .type function
1 year ago
Kai Pastor
93eb42fdc8
Fix redefinition of FAILED
1 year ago
Kai Pastor
dc905636d1
arm: Declare symbols as .type function
1 year ago
Iha, Taisei
4918beecbe
Loop-unrolled transposed [SD]GEMV kernels for A64FX and Neoverse V1
1 year ago
Martin Kroeker
0578a89afd
Merge pull request #4993 from martin-frbg/issue4991
Translate CMAKE_SYSTEM_NAME in compilations on or for IOS
1 year ago
Martin Kroeker
57a51d74c9
translate CMAKE_SYSTEM_NAME in compilations on or for IOS
1 year ago
Martin Kroeker
35f2e6afe6
Merge pull request #4992 from mmuetzel/ci-msys2
CI (MinGW): Remove CLANG32 environment from build matrix.
1 year ago
Markus Mützel
f5e6b5b5c9
CI (MinGW): Remove CLANG32 environment from build matrix.
The CLANG32 environment is in the process of being removed from MSYS2
currently:
https://www.msys2.org/news/#2024-09-23-starting-to-drop-the-clang32-environment
Remove it from the build matrix ahead of its complete removal from MSYS2.
1 year ago
Martin Kroeker
8e8003a2d1
Merge pull request #4180 from mmuetzel/cmake
CI (MinGW): Remove work-around needed for old versions of LLVM Flang
1 year ago
Martin Kroeker
71963a7bc4
Merge pull request #4985 from CheryDan/RISCV/sched
added optimizations for RISC-V YIELDING
1 year ago
Markus Mützel
7452af4471
CI (MinGW): Remove work-around with NO_AVX512 that was needed for older versions of LLVM Flang.
2 years ago
Martin Kroeker
82088cb266
Merge pull request #4986 from martin-frbg/readme_compilers
Add compiler version notes and mention the f2c fallback LAPACK in the README
1 year ago
Martin Kroeker
8481301f1a
Merge pull request #4987 from martin-frbg/issue3973
Update build instructions for WoA (use LLVM19 and its flang-new)
1 year ago
Martin Kroeker
009c1e0387
fix download link for the current WoA binary of LLVM
1 year ago
Martin Kroeker
760a5371f3
Update build instructions for WoA (use LLVM19 and its flang-new)
1 year ago
Martin Kroeker
3a63bbabd1
Add compiler version notes and mention the f2c fallback LAPACK
1 year ago
Martin Kroeker
c520ed1916
Merge pull request #4984 from rgommers/docs-link
doc: update README to link to the html docs and fix links
1 year ago
daichengrong
0b3db03d4b
added optimizations for RISC-V YIELDING
1 year ago
Ralf Gommers
a0131e56e0
doc: update README to link to the html docs and fix links
Also some minor formatting improvements and linking the home page.
1 year ago
Martin Kroeker
18014b04c8
Merge pull request #4979 from martin-frbg/issue4978-2
Remove any optimization flags from DEBUG builds on POWER architecture
1 year ago
Martin Kroeker
9db51f790a
Remove any optimization flags from DEBUG builds on POWER architecture
1 year ago
Martin Kroeker
e334b79b47
Merge pull request #4977 from martin-frbg/issue4973
Add dummy implementations of openblas_get/set_affinity for OpenMP builds
1 year ago
Martin Kroeker
4060dd43e3
Add dummy implementations of openblas_get/set_affinity
1 year ago
Martin Kroeker
2e2f952bfb
Merge pull request #4975 from martin-frbg/fixup4974
Update Cray compiler options and calling convention in CMake
1 year ago
Martin Kroeker
cea9df3643
Update Cray compiler options and calling convention
1 year ago
Martin Kroeker
3e7e312d7d
Merge pull request #4974 from cenewcombe/develop
Corrections for Cray and Nvidia Fortran compiler calling conventions
1 year ago
Caroline Newcombe
10cf06dce1
Merge branch 'OpenMathLib:develop' into develop
1 year ago
Caroline Newcombe
760bf7aa37
Update Fortran return for complex data types (Cray and Nvidia compilers)
1 year ago
Martin Kroeker
2057cd75fd
Merge pull request #4971 from martin-frbg/m4init
Add initial autodetection support for Apple M4 as VORTEX with HAVE_SME
1 year ago