fix(gpu): enable larger number of samples in the keyswitch #2067
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
|
I ran the benchmark integer tests, and performance is not affected |
pdroalves
requested changes
Feb 14, 2025
| @@ -114,7 +114,7 @@ __host__ void host_keyswitch_lwe_ciphertext_vector( | |||
| num_blocks, num_threads_x); | |||
|
|
|||
| int shared_mem = sizeof(Torus) * num_threads_y * num_threads_x; | |||
| dim3 grid(num_blocks, num_samples, 1); | |||
| dim3 grid(num_samples, num_blocks, 1); | |||
There was a problem hiding this comment.
This is the way to go. We should not have to support more than a few thousands blocks, and the y-axis supports up to 65k.
Can you add maybe a comment justifying this line change and add an assert checking if the number of blocks is greater than the maximum number of y-blocks?
guillermo-oyarzun
force-pushed
the
go/fix/allow-large-number-of-samples-in-ks
branch
from
0a328f1 to
6d0f664
Compare
pdroalves
approved these changes
Feb 17, 2025
|
Thanks @guillermo-oyarzun . Let's merge when we have a green CI. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
closes: please link all relevant issues
PR content/description
Check-list: