Giter VIP home page Giter VIP logo

Comments (12)

bluenote-1577 avatar bluenote-1577 commented on August 18, 2024 1

I'm honestly not sure what's going on here... here's what I've tried

  1. Conda install on CentOS 8 - Works fine but has AVX2
  2. Building on ARM64 architecture - Works fine, no AVX2, but not from conda (bioconda doesn't support aarch64 yet)

I need to track down a machine with no AVX2 running on x86-64. If the binary works for you, then I'd be surprised if compiling doesn't work.

I'll leave this open until I figure it out, or if someone else has a similar problem

from sylph.

fplazaonate avatar fplazaonate commented on August 18, 2024 1

Hi @bluenote-1577,
You can build different executables and choose on the fly which one is the most appropriate.
Look at what's implemented for raptor: https://github.com/bioconda/bioconda-recipes/tree/master/recipes/raptor

from sylph.

bluenote-1577 avatar bluenote-1577 commented on August 18, 2024

Hi @fplaza,

Thanks for bringing this to my attention. Since the executable works, I assume you're on linux? I just tried the conda install on my Ubuntu 16 system and it appears to be fine. If you could provide your distribution and any system specifications that would be helpful for me to try and reproduce the crash.

Thanks!

from sylph.

fplazaonate avatar fplazaonate commented on August 18, 2024

I am using CentOS 8.
The CPU does not have avx2 extensions.

from sylph.

edwardbirdlab avatar edwardbirdlab commented on August 18, 2024

I also seem to have the same issue on CentOS 7 without AVX2.

The executable works for me as well, but the bioconda installation crashes immediately with "Illegal instruction (core dumped)".
I am running on older hardware with AVX1.

from sylph.

bluenote-1577 avatar bluenote-1577 commented on August 18, 2024

@edwardbirdlab thanks for confirming. Does the pre-built binary work for you too?

from sylph.

edwardbirdlab avatar edwardbirdlab commented on August 18, 2024

Yes, the pre-built binary works without any issue.

from sylph.

dawnmy avatar dawnmy commented on August 18, 2024

Hi, I am testing sylph 0.6.0 installed via conda on my Linux server, but got the same error 1724781 illegal hardware instruction (core dumped) sylph when running it. The machine has avx and avx2. The previous version 0.5.1 worked fine on the same machine.

from sylph.

bluenote-1577 avatar bluenote-1577 commented on August 18, 2024

@dawnmy thanks for bringing this up. I can reproduce this bug as well even on my computer.

I think this is a bioconda issue, not a sylph issue: I did not change any hardware instruction behaviour between versions. I will update to version v0.6.1 and hope that bioconda fixes this. If not, it looks like I will have to ask the conda folks for help.

from sylph.

bluenote-1577 avatar bluenote-1577 commented on August 18, 2024

@dawnmy I updated sylph to v0.6.1 on conda. It works okay on my computer now. Let me know on your end

from sylph.

dawnmy avatar dawnmy commented on August 18, 2024

Thank you for the quick update. It works!

from sylph.

bluenote-1577 avatar bluenote-1577 commented on August 18, 2024

I figured out what the issue is: -target=native rust flag is being passed to the bioconda build. So the binary will use native instructions on the bioconda machine that is building sylph. This is why AVX2 is needed for the current conda build.

In the future, I will probably have to disable -target=native for conda. This will maybe make sylph slower, unfortunately...

I would prefer AVX2 to be required for conda, but not AVX512, which -target=native may force.

from sylph.

Related Issues (14)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.