Added support for normal MLA kernel #17624
Merged
+147
−13
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
|
@MasterJH5574 can you take a look? |
annanyapr
force-pushed
the
generic-attention
branch
from
3eac670 to
e3ac7b5
Compare
annanyapr
changed the title
annanyapr
force-pushed
the
generic-attention
branch
from
506386d to
7569674
Compare
annanyapr
force-pushed
the
generic-attention
branch
from
7569674 to
acd9fa0
Compare
|
@MasterJH5574 TVM seems to building correctly and tvm/tests/python/relax/test_runtime_builtin_paged_attention_kv_cache_tir.py seems to be working fine |
MasterJH5574
approved these changes
Feb 20, 2025
MasterJH5574
force-pushed
the
generic-attention
branch
from
1e4b697 to
a533a11
Compare
MasterJH5574
force-pushed
the
generic-attention
branch
from
a533a11 to
bd88313
Compare
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Outdated suggestions cannot be applied.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.
Have refactored _attention_prefill_ragged to allow for different v dimension from q/k dimension. This can be used for MLA attention in deepseek models.