[DCU] Llama a8w8 inference performance optimization #8800
Merged
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
|
Thanks for your contribution! |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## develop #8800 +/- ##
===========================================
+ Coverage 55.37% 55.58% +0.21%
===========================================
Files 631 630 -1
Lines 99707 98382 -1325
===========================================
- Hits 55211 54687 -524
+ Misses 44496 43695 -801 ☔ View full report in Codecov by Sentry. |
YanhuiDua
reviewed
Jul 24, 2024
| @@ -1384,7 +1392,10 @@ def compute_mmha(self, qkv_out, caches, attn_mask, seq_lens, rotary_embs, rotary | |||
| )[0] | |||
|
|
|||
| def compute_out_linear(self, fmha_out, i): | |||
| out_linear_out = paddle.matmul(fmha_out, self.linear_weights[i], False, True) | |||
| if paddle.is_compiled_with_rocm(): | |||
YanhuiDua
approved these changes
Jul 24, 2024
6 tasks
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Outdated suggestions cannot be applied.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.
PR types
Performance optimization
PR changes
Models
Description
Optimize the inference performance in Llama a8w8 case.
On the DCU platform, performance of rocblas gemm under different transpositions is NT > NN > TN. Due to the matmul of paddle, NT cannot be triggered in this scenario, so a suboptimal solution is chosen, which is NN.