unittest: add MLA test cases where kv_len is evenly divided by page_size. #861
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
yzh119
reviewed
Feb 17, 2025
tests/test_deepseek_mla.py
Outdated
| head_dim_ckv = 512 | ||
| head_dim_kpe = 64 | ||
| q_nope = torch.randn( | ||
| batch_size * qo_len, num_heads, head_dim_ckv, dtype=torch.half, device="cuda" | ||
| ) | ||
| q_pe = torch.randn( | ||
| batch_size * qo_len, num_heads, head_dim_kpe, dtype=torch.half, device="cuda" | ||
| ) | ||
| ) | ||
| pages_num = math.ceil(kv_len // page_size) |
There was a problem hiding this comment.
Suggested change
| pages_num = math.ceil(kv_len // page_size) | |
| pages_num = math.ceil(kv_len / page_size) |
There was a problem hiding this comment.
After this fix, all unittests passed.
yzh119
force-pushed
the
add_test_for_mla
branch
from
b5bc72a to
03c7916
Compare
yzh119
approved these changes
Feb 17, 2025
yzh119
changed the title
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Outdated suggestions cannot be applied.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.
@yzh119 PLZ take a look for this test. It will fail when
kv_lenis not multiple ofpage_size. I checked the kernel but had no clue. If I was not wrong, some bound check in mla kernel was wrong?