-
Notifications
You must be signed in to change notification settings - Fork 11k
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
CUDA: fix FA out-of-bounds writes #7465
Merged
JohannesGaessler
merged 1 commit into
ggml-org:master
from
JohannesGaessler:cuda-fix-fa-oob
May 22, 2024
Merged
CUDA: fix FA out-of-bounds writes #7465
JohannesGaessler
merged 1 commit into
ggml-org:master
from
JohannesGaessler:cuda-fix-fa-oob
May 22, 2024
+18
−2
Conversation
This file contains 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
This resolved my issues regarding FA. Tested on 1x p40 and 2x p40 using llama3-8B @ Q8_0. |
slaren
approved these changes
May 22, 2024
teleprint-me
pushed a commit
to teleprint-me/llama.cpp
that referenced
this pull request
May 23, 2024
This was referenced Jun 3, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
ggml
changes relating to the ggml tensor library for machine learning
Nvidia GPU
Issues specific to Nvidia GPUs
Review Complexity : Medium
Generally require more time to grok but manageable by beginner to medium expertise level
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Outdated suggestions cannot be applied.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.
Fixes #7421 .
The issue seems to be that all FlashAttention kernels except for the one using tensor cores are missing a check to avoid out-of-bounds writes for batch sizes that are not powers of 2.