OpenBLAS

OpenBLAS copied to clipboard

test for #2923 [skip actions] [skip azp] [skip appveyor]

martin-frbg

WIN64: OpenBLAS ctest with errors when linked against libopenblas.dll but works if build statically.

5

This is with Windows 64 bit, Msys2 gcc-12. (Both: UCRT and MSVCRT) I tested for curiosity what happens if the test programs located in ctest are dynamically linked against libopenblas.dll....

carlkl

undefined reference to `cblas_dgemm'

4

Hi, I am trying to run matrix operations in C, I am coding on VScode, when I compile, it gives this message: ``` >> gcc testtest.c -I "C:\\Users\\experiments\\OpenBlasInstall\\include" C:/msys64/mingw64/bin/../lib/gcc/x86_64-w64-mingw32/12.2.0/../../../../x86_64-w64-mingw32/bin/ld.exe: C:\Users\User\AppData\Local\Temp\ccsrFgiV.o:testtest.c:(.text+0x29a):...

johnny2231

This allows Julia to set a default number of threads (usually `1`) to be used when no other thread counts are specified [0], to short-circuit the default OpenBLAS thread initialization...

staticfloat

cmake: Allow specifying optimization level in COMMON_OPT

9

When building with `make` a default optimization level of `-O2` is used. See: https://github.com/xianyi/OpenBLAS/blob/00534523ad999d89945d23b7df0eafc69c31f1b3/Makefile.system#L1551-L1557 By default, `cmake` uses `-O3` for `Release` builds currently: https://github.com/Kitware/CMake/blob/v3.24.1/Modules/Compiler/GNU.cmake#L59 This means binaries are built with...

mmuetzel

We need this modification if we want to compile `ReLAPACK` and also not override the .o and symbols generated by LAPACK (`INCLUDE_ALL=0` in `relapack/config.h`).

amontoison

Clarifying naming of compute kernels prior to submission of PR

10

I'm currently implementing the FP16 kernels as a follow up to #3754 and #2767 I want to be 100% clear on OpenBLAS nomenclature prior to submitting my PR. ### Baseline...

FCLC

introduce groundwork for FP16 (half precision IEEE 754 FLOAT) support in both AVX512 and f16c versions

4

This PR provides for the basic infrastructure needed for detecting the presence of hardware for supporting any and all available FP16 capabilities available on x86_64 hardware. FP16 differs from BF16...

FCLC

Denormal floats handling

42

Dear @xianyi and everyone! Is there a way to forbid denormals in the OpenBLAS? I tried to execute the following code (MSVC2015) before calling any OpenBLAS routines: unsigned int current_word...

Arech

ARMv7 vfp version

1

Using the `ARMv7` target results in the compiler option `-mfpu=vfpv3`, yet some ARMv7 support `vfpv4`, for example theAllwinner H3 used in various single board computer. Could an additional ARM target...

gramian