Skip to content
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 dmmv cols requirement to 2*GGML_CUDA_DMMV_X #8800

Merged
merged 4 commits into from
Aug 1, 2024

Conversation

slaren
Copy link
Collaborator

@slaren slaren commented Jul 31, 2024

May fix #8798

Copy link
Collaborator

@cebtenzzre cebtenzzre left a comment

Choose a reason for hiding this comment

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

I can't reproduce #8798 with this change applied. I think it would be nice if we also added a regression test to test-backend-ops, since you could get it to fail that way.

Nexesenex added a commit to Nexesenex/croco.cpp that referenced this pull request Jul 31, 2024
Nexesenex added a commit to Nexesenex/croco.cpp that referenced this pull request Jul 31, 2024
Nexesenex added a commit to Nexesenex/croco.cpp that referenced this pull request Jul 31, 2024
Copy link
Collaborator

@JohannesGaessler JohannesGaessler left a comment

Choose a reason for hiding this comment

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

I think it would make sense to also adapt the asserts in dmmv.cu which currently only assert a multiple of GGML_CUDA_DMMV_X.

@github-actions github-actions bot added the Nvidia GPU Issues specific to Nvidia GPUs label Aug 1, 2024
@mofosyne mofosyne added bugfix fixes an issue or bug Review Complexity : Low Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix labels Aug 1, 2024
@github-actions github-actions bot added the testing Everything test related label Aug 1, 2024
@slaren slaren merged commit 7a11eb3 into master Aug 1, 2024
54 checks passed
@slaren slaren deleted the sl/fix-dmmv-req branch August 1, 2024 13:26
arthw pushed a commit to arthw/llama.cpp that referenced this pull request Aug 2, 2024
* cuda : fix dmmv cols requirement to 2*GGML_CUDA_DMMV_X

* update asserts

* only use dmmv for supported types

* add test
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bugfix fixes an issue or bug Nvidia GPU Issues specific to Nvidia GPUs Review Complexity : Low Trivial changes to code that most beginner devs (or those who want a break) can tackle. e.g. UI fix testing Everything test related
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Bug: CUDA illegal memory access related to KV/n_ctx padding and F16 DMMV
4 participants