Skip to content

AMD requested cherry-picks for release/3.1.x #4794

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Sep 24, 2024

Conversation

bertmaher
Copy link
Collaborator

Per @jataylo :

xinyazhang and others added 3 commits September 24, 2024 10:09
…-lang#4246)

Previously Triton cannot find the HIP runtime bundled in PyTorch if the
wheel is installed with `--user`. Searching the user site-packages directory
may solve this problem.
…ng#4255)

PyTorch have ways to load the libamdhip64.so library, and the original
logic has problems detecting this shared object file because it can be
placed at any location with `venv`/`DT_RUNPATH`/`LD_LIBRARY_PATH`/
`/etc/ld.so.conf.d/*.conf`/`LD_PRELOAD`. Notably the `RUNPATH`
has become the preferred approach for multi-version ROCm installations.

This patch let Triton enumerate the address space with
`dl_iterate_phdr`, and choose the libamdhip64.so that is already loaded
into address space unless `TRITON_LIBHIP_PATH` is already set.

---------

Co-authored-by: Lei Zhang <antiagainst@gmail.com>
The workaround introduced in
triton-lang#4048 "forgot" to skip mfma
layout.
@bertmaher bertmaher requested review from atalman and ptillet and removed request for ptillet September 24, 2024 17:18
Copy link
Collaborator

@atalman atalman left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

@atalman atalman merged commit cf34004 into triton-lang:release/3.1.x Sep 24, 2024
zhzhcookie added a commit to FlagTreeZhouyi/flagtree-zhouyi that referenced this pull request Mar 31, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants