OpenMathLib/OpenBLAS on Hacker News

…the 2500x one is using OpenBLAS which uses lower level things than what's in C[1]. [1] see eg https://github.com/OpenMathLib/OpenBLAS/tree/develop/kernel/...

comment by fulafel · 2026-08-20 · in MicroGPT-C in pure C hits 10M TPS on Apple M5

Sure they do. https://github.com/OpenMathLib/OpenBLAS https://github.com/FFmpeg/FFmpeg Plenty of assembly in those projects but no mention of it in the README. Most C projects don't acknowledge the assembly they use.

comment by nromiun · 2025-08-07 · in Python performance myths and fairy tales

…same principles they use. [0] https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprogra... [1] https://github.com/OpenMathLib/OpenBLAS [2] https://www.intel.com/content/www/us/en/developer/tools/onea... [3]…

comment by steppi · 2024-04-01 · in LLaMA now goes faster on CPUs

…and when hand-tuned assembly kernels work better, that's what are used. For example, [2]. [0] https://github.com/xianyi/OpenBLAS/issues/1968 [1] https://github.com/xianyi/OpenBLAS/blob/develop/kernel/x86_6... [2]…

comment by steppi · 2023-06-01 · in Learn x86-64 assembly by writing a GUI from scratch

…assembly code, with dispatching code to pick out the appropriate implementations. https://github.com/xianyi/OpenBLAS/blob/02ea3db8e720b0ffb3e2... https://github.com/xianyi/OpenBLAS/blob/02ea3db8e720b0ffb3e2...

comment by MaxBarraclough · 2022-12-06 · in Optimizing compilers reload vector constants needlessly

…is that the server actually had an EPYC CPU, but the hypervisor was configured to say Opteron. [0]: https://github.com/xianyi/OpenBLAS/issues/3638

comment by GrayShade · 2022-05-29 · in Debugging a mysterious Python crash

…https://lists.qt-project.org/pipermail/development/2022-Marc... https://github.com/xianyi/OpenBLAS/issues/3551 https://www.phoronix.com/scan.php?page=news_item&px=OpenBLAS... So the non-shared forks of FOSS are likely to b

comment by pabs3 · 2022-03-07 · in Russia to legalize software piracy

…syscall without calling the atfork handlers. https://github.com/tensorflow/tensorflow/issues/13802 https://github.com/xianyi/OpenBLAS/issues/240 https://trac.sagemath.org/ticket/22021 https://bugs.python.org/issue31814…

comment by albertzeyer · 2022-02-28 · in Fork() is evil; vfork() is goodness; afork() would be better; clone() is stupid

I used OpenBLAS on my cheap last-generation AMD Ryzen 7 4700U laptop like so: git clone https://github.com/xianyi/OpenBLAS && cd OpenBLAS && make PREFIX=/opt/openblas install && curl https://jott.live/code/blas_test.cc | sed -n…

comment by homerowilson · 2021-11-22 · in Benchmarking the Apple M1 Max

…would disagree with this statement. State-of-the-art HPC code is still written in ASM (see e.g., https://github.com/xianyi/OpenBLAS ) [that's what Intel is doing too]

comment by zanellia · 2021-11-17 · in Show HN: Prometeo – a Python-to-C transpiler for high-performance computing

Can you explain what you mean by this? There is a nice description of this: https://github.com/xianyi/OpenBLAS/issues/2543 At a previous employer, we have seen various issues, including crashes, non-determinisms, etc. Usually, these issues…

comment by danieldk · 2021-10-03 · in R, OpenMP, MKL, Disaster

…with OpenBLAS multi-threading. Thus, you must set OpenBLAS to use single thread as following. https://github.com/xianyi/OpenBLAS/wiki/faq#multi-threaded

comment by danieldk · 2020-09-01 · in Intel MKL on AMD Zen

I noticed recently that OpenBLAS has gained some code using the POWER 10 matrix-multiplication units: https://github.com/xianyi/OpenBLAS/tree/develop/kernel/power

comment by gnufx · 2020-08-17 · in Next-Generation IBM POWER10 Processor

…documented example of an OpenBLAS threading issue from someone rather more knowledgeable than I am: https://github.com/xianyi/OpenBLAS/issues/731

comment by itamarst · 2020-02-05 · in The Parallelism Blues: when faster code is slower

Well, if https://github.com/xianyi/OpenBLAS is any indication, you can just recompile the library for ARM. A proper pipeline would recompile everything on the appropriate platform without any developer intervention.

comment by shepardrtc · 2019-02-22 · in Linus Torvalds on Why ARM Won't Win the Server Space

…say, eigen https://github.com/PX4/eigen/blob/master/Eigen/src/SparseCho... not to mention openblas https://github.com/xianyi/OpenBLAS/blob/develop/lapack/lauu2...

comment by ernst_klim · 2018-08-13 · in My Haskell Journey from C#

…about them; it's probably the best option, especially considering its open source status. [0] https://github.com/xianyi/OpenBLAS/issues/991#issuecomment-3...

comment by stochastic_monk · 2018-05-14 · in Performance experiments with matrix multiplication in Rust

…You can study the (micro-)architecture-dependent uses in e.g. https://github.com/flame/blis https://github.com/xianyi/OpenBLAS and https://github.com/hfp/libxsmm and it may be discussed in the papers describing them. libxsmm provides a…

comment by gnufx · 2018-05-01 · in Is software prefetching (__builtin_prefetch) useful for performance?

…freed memory. Ah, good times. Here's the root cause analysis if anyone wants the gory details :-) https://github.com/xianyi/OpenBLAS/issues/716#issuecomment-1...

comment by ChrisFoster · 2017-01-31 · in Don't setenv in multi-threaded code on glibc

…why you see -lgfortran -lf77blas when you compile programs that require numerical tools. [1] https://github.com/xianyi/OpenBLAS

comment by ThatGeoGuy · 2016-12-18 · in Why physicists still use Fortran

…state-of-the-art SGEMV kernel looks like for a modern chip like Haswell, check out OpenBLAS's code: https://github.com/xianyi/OpenBLAS/blob/develop/kernel/x86_6...

comment by gok · 2016-10-14 · in Intel will add deep-learning instructions to its processors

According to Github, 51.6% of the code in OpenBLAS is Fortran: https://github.com/xianyi/OpenBLAS

comment by shepardrtc · 2015-04-21 · in Why GEMM is at the heart of deep learning

…for writing SIMD kernels or wrapping up some unrolled ASM from openblas/blis like this monstrosity https://github.com/xianyi/OpenBLAS/blob/develop/kernel/x86_6... for throughput cpu/ram throughput bounded workloads that aren't allocation…

comment by carterschonwald · 2014-10-16 · in From 60 Frames per Second to 500 in Haskell

…created GotoBLAS, now works on MKL at Intel. [1] http://cython.org [2] http://numba.pydata.org [3] https://github.com/xianyi/OpenBLAS [4] http://en.wikipedia.org/wiki/GotoBLAS

comment by StefanKarpinski · 2013-11-29 · in Why I’m not on the Julia bandwagon (yet) (2012)

…by the Julia programming language [3]. [1] http://eigen.tuxfamily.org/index.php?title=Benchmark [2] https://github.com/xianyi/OpenBLAS [3] http://julialang.org/

comment by chrisaycock · 2012-05-11 · in Writing the Fastest Code, by Hand, for Fun (2005)

There is an actively maintained fork of GotoBLAS2, OpenBLAS: https://github.com/xianyi/OpenBLAS

comment by fdej · 2012-05-11 · in Writing the Fastest Code, by Hand, for Fun (2005)