OpenBLAS

OpenBLAS copied to clipboard

Commit Hash Access in Compiled API

5

The [recently added](https://github.com/xianyi/OpenBLAS/pull/1890) version number availability for `openblas_get_config()` is [already proving useful](https://github.com/numpy/numpy/pull/12523) in our project to verify binaries built via linking to openblas (i.e., make sure an old system-level openblas...

tylerjereddy

Building on ORNL Summit (POWER9) with PGI compiler

54

After toying around with `Makefile.power` and `Makefile.system` for a while, I've successfully built OpenBLAS 0.3.10 on POWER9 at Summit (ORNL) with GCC 6.4.0 (the default GCC version, at the time...

wyphan

Issue with Power8/32bit tests

23

Hi, Following issue 2693 [https://github.com/xianyi/OpenBLAS/pull/2693] by @EGuesnet, once fixed by our patch , we have found another issue when tests are run during build phase, still with Power8/32bit, thus using...

trex58

Thread scaling of concurrent small dgemm operations

3

My app performs many small dgemms, each invoked by a separate thread (via a task pool). As recommended I compiled OpenBlas 3.10 with USE_THREAD=0 and USE_LOCKING=1. This is on Cavium...

robertjharrison

ARM results in error

8

Dear developers, thank you for your great work on openBLAS. using it on ARM 32 bit platforms and Ubuntu 14.04, we found some erroneous results used with Torch7: The Lua...

culurciello

cblas_sgemm return all zero

7

Hi Xianyi, We tried to run a matrix multiplication with cblas_sgemm or cblas_dgemm on android. We tried with A = [1 3 4 6], B = [3 5 9 1],...

odieXin

Could you elaborate on the combination of OpenBLAS with multi-threading?

9

https://github.com/xianyi/OpenBLAS/wiki/Faq/4bded95e8dc8aadc70ce65267d1093ca7bdefc4c#multi-threaded says: > If your application is already multi-threaded, it will conflict with OpenBLAS multi-threading. Thus, you must set OpenBLAS to use single thread as following ... That is good...

nh2

[WIP]Optimize GEMM for TSV110

3

Hi, fellows! I benchmarked sgemm performance on [email protected] x 24Cores (square matrix from 128 to 4096 ). There seems to be a gap between the best case (28.3 GFLOPS) and...

craft-zhang

ARMV8 Target implies 64-bit compilation and some Makefile specifics

20

Hi, I've got three issues when compiling OpenBLAS. 1) When compiling using `TARGET=ARMV8` or more explicitly `TARGET=CORTEXA72`, this will lead to a 64-bit binary compilation. The Raspberry Pi 4 with...

MaxiBoether

Adding packed gemm APIs

8

Hi Xianyi, Came across the following article. https://www.codeproject.com/Articles/1169319/Reducing-Packing-Overhead-in-Matrix-Matrix-Multipl This talks about introducing new packed APIs of the following form in MKL. `dest = sgemm_alloc (identifier, m, n, k)` `sgemm_pack (identifier,...

ashwinyes