OpenBLAS

OpenBLAS copied to clipboard

Stack overflow when compiled with 4096 NUM_THREADS

7

We decided to set a very high NUM_THREADS for building OpenBLAS - 4096. That leads to a stack overflow in `getrf_parallel`. I believe this setting should have led to `ALLOC_HEAP`...

ViralBShah

martin-frbg

Slowdown when using openblas-pthreads alongside openmp based parallel code

26

Hi, I have a code which mixes BLAS calls (gemm) and OpenMP based parallel loops (they are not nested). When OpenBLAS is built using OpenMP everything is fine but when...

jeremiedbb

WIP: allow threading backend to be replaced by caller

13

This PR adds a new function `openblas_set_threads_callback` that allows the caller to register a callback in order to change the threading backend at runtime — instead of spawning its own...

stevengj

MKL_GETRF equivalnet in OpenBLAS

10

Hi, I was converting my code from MKL to OpenBlAS, but I am not able to find an equvalent for dgetrf with partal pivoting and dgetrf witout pivoting. Just I...

nnaron

test failed for ctbsv on blas tester

9

bin/xcl2blastst -R tbsv -U 1 U -A 1 N -D 2 U N -n 4 -X 2 2 1 -q 3 bin/xcl2blastst -R tbsv -U 1 U -A 1 N...

qqqil

Lacking quad precision support

7

Support for quad precision seems to be broken or maybe entirely missing. Setting QUAD_PRECISION=1 in Makefile.rule and running make results in compilation error caused by missing common_quad.h.

dexter2206

wi24rd

Stack overflow segfault in getrf_parallel using Breeze / netlib

5

Hi all, This issue is very similar to https://github.com/xianyi/OpenBLAS/issues/1082. We are still running into this issue as seen in https://github.com/luhenry/netlib/issues/2. This is with `libopenblas-dev` `0.3.5+ds-3` on Debian 10. From looking...

luhenry

Performance: speed up GEMM by pre-faulting memory

22

I've found that at least in some situations, GEMM can be particularly slow if the output array is not already backed by physical pages, and it can be 4x faster...

bmerry