OpenBLAS

OpenBLAS copied to clipboard

Modifying GEMM for strided GEMM computation

11

Hi, We want to implement strided GEMM implementation where the dot product for each output element is computed using only even indexed elements from rows/columns of matrix A/B. Seems like...

nitish2112

openblas and openmp

7

I have tried to use openblas with Tim Davis' SuiteSparse package. I have download openblas from either redhat on my dell desktop or from ubuntu on my thinkpad lap; in...

bill-hager

Use getauxval to implement dynamic arch on ARM and AArch64

2

I've just came across [this post](https://community.arm.com/groups/android-community/blog/2014/10/10/runtime-detection-of-cpu-features-on-an-armv8-a-cpu) the other day which suggests that even though the feature registers cannot be accessed in EL0, the kernel puts this info in the auxiliary...

yuyichao

ARMV7 (with hard float flag) did not run with correct result

15

hello, Resently, we are using openblas to setup caffe env in our ARMv7 platform, but we got a problem when run openblas with hard float flag. we compiled the openblas...

gangm

Julia + OpenBLAS vs. MATLAB + MKL - Matrix Operations Benchmark

22

Hi, I did some tests with MATLAB and Julia: [Matlab & Julia Matrix Operations Benchmark](https://github.com/RoyiAvital/MatlabJuliaMatrixOperationsBenchmark) I think they (At least to some part) reflect OpenBLAS vs. Intel MKL. Hence I...

RoyiAvital

Non-deterministic output with multiple OpenMP runtimes in the same process

21

I observe non-deterministic output when calling into cblas_sgemv concurrently from two different threads each of which are bound to a different copy of an OpenMP runtime in the same process....

lightsighter

openBLAS nested parallelism

44

Hi, we are trying to run two instances of cblas_dgemm in parallel. If the total number of threads is 16, we would like each instance to run using 8 threads....

SanazGheibi

Stop treating Cygwin as some kind of Windows variant

11

An unfortunate legacy brought over from GotoBlas is use of Windows APIs on Cygwin, and general treatment of Cygwin as just some variant of Windows. At one time that might...

embray

performance on AMD Ryzen and Threadripper

121

This report comes right from #1425 where the discussion drifted off from thread safety in openblas v. 0.2.20 to performance on AMD Ryzen and Threadripper processors (in this particular case...

tkswe88

Use of long double in OpenBLAS

14

Pull Request #1709 surprised me by showing that long double variables are used in some of the OpenBLAS code. Although gcc on x86_64 uses 80 bit Intel extended precision for...

brianborchers