Allow overriding gfx version per-node#104
Open
AdamNiederer wants to merge 2 commits intoROCm:masterfrom
Open
Conversation
AdamNiederer
force-pushed
the
per-node-overrides
branch
from
fa07f70 to
7d5f503
Compare
This lets you run two unsupported-but-really-supported cards of different architecture together in the same program. Works great w/ llama.cpp on my 7900XT + 6600. Example usage (device 0 is RDNA3, device 1 is RDNA2): HSA_OVERRIDE_GFX_VERSION_1="11.0.0" HSA_OVERRIDE_GFX_VERSION_2="10.3.0" ollama serve
AdamNiederer
force-pushed
the
per-node-overrides
branch
from
7d5f503 to
8135a1c
Compare
kentrussell
reviewed
Jun 5, 2024
Contributor
There was a problem hiding this comment.
This is cleaner than the strcat, I think:
snprintf(per_device_override_name, sizeof(per_device_override_name), "HSA_OVERRIDE_GFX_VERSION_%d", node_id);
I'll test this out internally and if it works (and doesn't break any other flows), I'll get it pushed out. Hopefully for 6.2
Contributor
Author
There was a problem hiding this comment.
That's much better, thank you! Changed in da8055d. And thanks for giving it a spin internally!
Contributor
|
Internal testing looks good, we'll try to get this released in ROCm 6.2. Thanks for your contribution! |
kentrussell
force-pushed
the
master
branch
2 times, most recently
from
56a16c2 to
06a9b94
Compare
This was referenced Mar 29, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This lets you run two unsupported-but-really-supported cards of different architecture together in the same program. Works great w/ llama.cpp on my 7900XT + 6600; I'm seeing a 72% perf uplift running LLaMA3-70B Q2 across the two cards (7.4 vs 4.3 tok/s).
Example usage (device 0 is RDNA3, device 1 is RDNA2):