Adjust KV cache shape for compatibility with updated APIs for graph mode #657
+23
−8
Conversation
This file contains hidden or 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
linfeng-yuan
force-pushed
the
modify_kv_caches_for_graph_mode
branch
from
0d92f3a to
ae5c578
Compare
Bumps [actions/setup-python](https://github.com/actions/setup-python) from 5.5.0 to 5.6.0. Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
linfeng-yuan
force-pushed
the
modify_kv_caches_for_graph_mode
branch
from
ae5c578 to
6beab94
Compare
|
need rebase first |
| @@ -920,7 +920,7 @@ def exec_kv( | |||
| # npu_kv_rmsnorm_rope_cache needs [B, N, S, D] | |||
| kv = kv.view(B, N, S, self.kv_lora_rank + self.qk_rope_head_dim) | |||
|
|
|||
| k_pe, k_nope = torch.ops.npu_inference.npu_kv_rmsnorm_rope_cache( | |||
| k_pe, k_nope, _, _ = torch.ops.npu_inference.npu_kv_rmsnorm_rope_cache( | |||
There was a problem hiding this comment.
Why can we run this before, is it caused by torchair's update?
There was a problem hiding this comment.
Right. It results from torchair's update in new version of CANN. I'll submit a new PR containing some new modifications and #683.
|
|
need rebase first |
|
This pull request has conflicts, please resolve those before we can evaluate the pull request. |
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Outdated suggestions cannot be applied.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.
adapt: modify kv caches' shape to adapt to the updated kv_rmsnorm_rope_cache and npu_fused_infer_attention_score APIs.