[Test] port flash attention from sglang #3011
Closed
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
| b_seq_len_extend_test = b_seq_len_extend_test.to(device) | ||
| b_start_loc_extend_test = torch.arange(0, batch_size, dtype=torch.int32) * seq_len | ||
| b_start_loc_extend_test = b_start_loc_extend_test.to(device) | ||
| extend_attention_fwd(q_test, k_test, v_test, o_tensor_ptr, k_buffer_test, v_buffer_test, req_to_tokens_test, |
There was a problem hiding this comment.
can we add timing mechanism and result checking to ensure functional correctness?
There was a problem hiding this comment.
I don't know how to compare the result...
it's originated from end2end test
| b_seq_len_extend_test = b_seq_len_extend_test.to(device) | ||
| b_start_loc_extend_test = torch.arange(0, batch_size, dtype=torch.int32) * seq_len | ||
| b_start_loc_extend_test = b_start_loc_extend_test.to(device) | ||
| extend_attention_fwd(q_test, k_test, v_test, o_tensor_ptr, k_buffer_test, v_buffer_test, req_to_tokens_test, |
There was a problem hiding this comment.
What are the differences between this implementation of flash attention and https://github.com/intel/intel-xpu-backend-for-triton/blob/main/benchmarks/triton_kernels_benchmark/flash_attention_fwd_benchmark.py?
There was a problem hiding this comment.
seems many condition control, but the main difference is not using block pointer.
Dewei-Wang-sh
requested review from
sommerlukas,
alexbaden,
etiotto and
quintinwang5
|
make it draft for now, need further discussion about how to support end2end kernel. |
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Outdated suggestions cannot be applied.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.
No description provided.