UPSTREAM PR #18934: ggml-cuda: enable cuda-graphs for n-cpu-moe#971
Open
UPSTREAM PR #18934: ggml-cuda: enable cuda-graphs for n-cpu-moe#971
n-cpu-moe#971Conversation
loci-dev
temporarily deployed
to
PROD__AL_DEMO
GitHub Actions
Inactive
|
Explore the complete analysis inside the Version Insights |
loci-dev
force-pushed
the
main
branch
26 times, most recently
from
5b137d4 to
ab9ebfa
Compare
loci-dev
force-pushed
the
main
branch
30 times, most recently
from
706d8e7 to
83ca7a9
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Mirrored from ggml-org/llama.cpp#18934
Add piece-wise cuda graph for the multiple split case. Currently cuda graphs get disabled when there are splits as we only keep 1 cuda graph per device. Multiple updates with different sized splits/shapes triggers the disable.
This PR adds a cuda graph per split (a split is keyed via the first node in the split)
Tested on 2x4090 and 1x 5090