OpenBLAS

OpenBLAS copied to clipboard

[WIP] Set visibility attribute on internal function symbols to hidden

3

cf. https://github.com/MacPython/openblas-libs/issues/79 (remaining symbol conflicts between builds with and without INTERFACE64 as the internally used functions to not get suffixed. Adding suffixes for them would be the second-best solution if...

martin-frbg

[emmake]compile OpenBLAS using emmake

6

I tried to compile OpenBLAS using emmake, but I had the following problems: Is there any other way I can solve the problem? I want to compile the.c file that...

924657644

Multithreaded DGBMV degrades performance on AMD

12

I have a driver cpp file that calls cblas_dgbmv function with proper arguments. When I build OpenBLAS with "make", dgbmv runs automatically with 8 threads (multithreaded dgbmv is invoked in...

selinyldrm

Add the integrated test system for maintaining OpenBLAS on new ISA.

1

Thank you for OpenBLAS developers and users. There are some new architecture supported from PRs. However, the OpenBLAS developer team may not test or verify the code on the new...

xianyi

For multithreaded GEMM in OpenBLAS, do I need to call API in the program to ensure thread affinity?

4

AnonymousYWL

Decouple number of OpenBLAS threads and max number of OpenMP threads

16

It is possible to set the number of threads used by OpenBLAS via `openblas_set_num_threads`. For the "custom thread" solution this works quite well: Independent of what the application may do...

Flamefire

Test failure with Skylake CPU, potentially related to avx512 instructions

12

I have a test failure (dblas3) on the current develop branch. [Here is the test output, DBLAT3.SUMM.](https://github.com/xianyi/OpenBLAS/files/7764787/DBLAT3.SUMM.txt) I build OpenBLAS with cmake, setting CMAKE_BUILD_TYPE=Release and running ctest. The machine has...

quantumsteve

OpenBLAS on Graviton2 (NeoverseN1) markedly slower than ARM libarmpl

12

link to benchmark results copied from #3251: https://developer.arm.com/tools-and-software/server-and-hpc/compile/arm-compiler-for-linux/resources/tutorials/benchmarks current implementation in OpenBLAS is a mix of generic ARMV8 and ThunderX2T99 (initial PR #2465) Graviton2 is/was in our Travis CI setup...

martin-frbg

?getrf performance degrades after some cache limit passed

9

I am using a Windows/Mingw64 built OpenBLAS copy that I further built NumPy and SciPy with. The script I used is ```bash march="x86-64" extra="-fno-asynchronous-unwind-tables" vc_arch="X64" cflags="-O2 -march=$march -mtune=generic $extra" fflags="$cflags...

ilayn

Functions are not guarded against early threading

28

Not exhaustive list, just those used in practice, sort of measurement takes some time Discovered in #1882 - `dgetrf` `zherk` `dsyrk` From #1883 - `axpy` From #1897 (mxnet) `cblas_?syrk` `_potri_`...

brada4