Package Details: python-jaxlib-cuda 0.6.0-1

Git Clone URL: https://aur.archlinux.org/python-jaxlib-cuda.git (read-only, click to copy)
Package Base: python-jaxlib-cuda
Description: XLA library for JAX
Upstream URL: https://github.com/jax-ml/jax/
Keywords: deep-learning google jax maching-learning xla
Licenses: Apache-2.0
Groups: jax
Conflicts: python-jaxlib
Provides: python-jaxlib
Submitter: daskol
Maintainer: daskol
Last Packager: daskol
Votes: 9
Popularity: 0.008610
First Submitted: 2023-02-12 23:18 (UTC)
Last Updated: 2025-04-20 20:54 (UTC)

Latest Comments

« First ‹ Previous 1 2 3 4 5 6 Next › Last »

medaminezghal commented on 2024-11-05 10:02 (UTC)

@actionless both sources for package jax and jaxlib are the same. So you can use jax sources to compile the jax library

medaminezghal commented on 2024-11-05 09:58 (UTC)

@actionless the latest version in pypi is 0.4.35 may be in the source files they combine both packages jax and jaxlib

actionless commented on 2024-11-05 09:49 (UTC)

what link should i use to try it, i see there the latest jaxlib 0.4.32:

https://github.com/jax-ml/jax/tags

medaminezghal commented on 2024-11-05 09:42 (UTC)

@actionless does it work for the latest version (0.4.35)?

actionless commented on 2024-11-05 08:59 (UTC)

yup, i've posted a patch below

but it probably need some additional bazel command for downloading the files without build - otherwise it would be nothing to patch yet (so build would succeed only after restart after 1st fail)

daskol commented on 2024-11-05 08:54 (UTC)

@medaminezghal I will try one more time but last time it seems impossible without patching bazel's BUILD files (see discussion)

medaminezghal commented on 2024-11-05 08:19 (UTC)

@daskol could you update this package to the latest version (0.4.35)?

daskol commented on 2024-09-26 19:46 (UTC) (edited on 2024-10-23 22:50 (UTC) by daskol)

@truncs Yes, there is a building issues with clang18 and local cuda.

UPD It turns out that there are multiple issues related to building jaxlib with CUDA support (see jax-ml/jax#23689).

  1. For some reason local CUPTI does not materialize in targte build directory (xla/tsl).
  2. An old issue between libc++ and libstdc++ when inaccurate ordering of system search paths (-isystem option) leads to broken #include_next inclusion.

truncs commented on 2024-09-26 18:40 (UTC) (edited on 2024-09-26 18:41 (UTC) by truncs)

Seems like jaxlib and jax versions are not in sync. 

  3  aur/python-jax                       0.4.31-1        -> 0.4.33-1
  2  aur/python-orbax-checkpoint          0.5.20-1        -> 0.6.4-1
  1  aur/yay                              12.4.1-1        -> 12.4.2-1
==> Packages to exclude: (eg: "1 2 3", "1-3", "^4" or repo name)
 -> Excluding packages may cause partial upgrades and break systems
==> 
 -> could not find all required packages: python-jaxlib >=0.4.33

actionless commented on 2024-08-13 00:57 (UTC)

i've finally solved previously mentioned problem by adding this before the last line of build() func:

    sed -i \
        -e 's|/opt/cuda/targets/x86_64-linux/include/cudnn_adv.h|/usr/include/cudnn_adv.h|g' \
        -e 's|/opt/cuda/targets/x86_64-linux/include/cudnn_backend.h|/usr/include/cudnn_backend.h|g' \
        -e 's|/opt/cuda/targets/x86_64-linux/include/cudnn_cnn.h|/usr/include/cudnn_cnn.h|g' \
        -e 's|/opt/cuda/targets/x86_64-linux/include/cudnn_graph.h|/usr/include/cudnn_graph.h|g' \
        -e 's|/opt/cuda/targets/x86_64-linux/include/cudnn_ops.h|/usr/include/cudnn_ops.h|g' \
        -e 's|/opt/cuda/targets/x86_64-linux/include/cudnn_version.h|/usr/include/cudnn_version.h|g' \
        bazel-jax-jaxlib-v0.4.31/external/local_config_cuda/cuda/BUILD

« First ‹ Previous 1 2 3 4 5 6 Next › Last »