Closed
Description
Hi!
Sorry for that if it is my fault, but I'm getting a segfault on the clean install of 1.7.0-b4 on m1 mac with the eig for the simplest matrices. mwe:
julia> using LinearAlgebra;
julia> A=rand(10, 10);
julia> eigen(A)
signal (11): Segmentation fault: 11
in expression starting at REPL[3]:1
.Ldgemm_tcopy_L8_M8_20 at /Applications/Julia-1.7.app/Contents/Resources/julia/lib/julia/libopenblas64_.dylib (unknown line)
Allocations: 1721957 (Pool: 1721200; Big: 757); GC: 2
Version info:
Julia Version 1.7.0-beta4
Commit d0c90f37ba (2021-08-24 12:35 UTC)
Platform Info:
OS: macOS (arm64-apple-darwin20.5.0)
CPU: Apple M1
WORD_SIZE: 64
LIBM: libopenlibm
LLVM: libLLVM-12.0.1 (ORCJIT, cyclone)
have a suspicion that something happened upon the transition from macOS 12b5 to b6.
Sorry if this is a waste of time!
Activity
jeremiahpslewis commentedon Sep 9, 2021
I can confirm that it segfaults on my machine with the same setup.
mryodo commentedon Sep 16, 2021
the issue is still preset in
v1.7.0-rc1's binaryprobably should close in favour of JuliaLang/julia#42184
mryodo commentedon Sep 17, 2021
okay, so JuliaLang/julia#42293 by @jeremiahpslewis (kudos) fixes the build problem with
juliaoncmake12unfortunately,
OpenBLASproblem is still present even after build; one maybe wants to pull from OpenMathLib/OpenBLAS#3383 (apparently I am not able to figure out how)domschl commentedon Sep 17, 2021
Ideally Julia shouldn't use OpenBLAS on Apple M1. By using Apple's proprietary BLAS implementation (via Accelerate), a speed-up of 2x - 4x can be observed, since Apple's implementation uses their undocumented matrix multiplier hardware.
That will probably not be possible to achieve via OpenBLAS in the foreseeable future...
jeremiahpslewis commentedon Sep 17, 2021
@domschl This is super interesting, but probably worth opening a new issue on?
KristofferC commentedon Sep 17, 2021
Julia can easily change the BLAS used (on 1.7+) see https://github.com/staticfloat/libblastrampoline and https://discourse.julialang.org/t/does-mac-m1-in-multithreads-is-slower-that-in-single-thread/61114/12?u=kristoffer.carlsson
mryodo commentedon Sep 19, 2021
Thanks, it's quite an interesting discussion...
Although the need for the wrapper for the simple
*imply that it would be hard to push this through out all the packages, no?If so, it seems like some kind of a at least temporary workaround for
openblasis still a good idea?domschl commentedon Sep 19, 2021
I made a new issue #869 for Apple M1 BLAS HW-acceleration discussions as suggested so that the discussion can continue there...
mryodo commentedon Sep 22, 2021
Okay, let me close this one: the segfault is gone in
macos12-beta7(guess, they did it again, ***) on the latestdevversion.If somebody still has the same problem, feel free to comment here (also, even before temporary workaround through
LBTwere working sort of fine)Have a nice day, check the discussion above!
2 remaining items