Skip to content

[Relax][KV Cache] Refactor _attention_sequence_prefill function to … #17362

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Sep 11, 2024
Merged

[Relax][KV Cache] Refactor _attention_sequence_prefill function to … #17362

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion python/tvm/relax/frontend/nn/llm/kv_cache.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1237,7 +1237,7 @@ def merge_state_inplace(


def _attention_sequence_prefill(
batch_size, h_kv, h_q, d, dtype, target: Target, causal=0, attn_score_scaling_factor=1.0
h_kv, h_q, d, dtype, target: Target, causal=0, attn_score_scaling_factor=1.0
): # pylint: disable=line-too-long
LOAD_VEC = 8 // ((DataType(dtype).bits + 7) // 8) # 8 bytes
group_size = h_q // h_kv
Expand All @@ -1264,6 +1264,7 @@ def batch_sequence_prefill_kv( # pylint: disable=too-many-branches
var_output: T.handle, # [total_len, h_q, d]
var_lse: T.handle # [total_len, h_q]
):
batch_size = T.int32(is_size_var=True)
qo_len = T.int32(is_size_var=True)
kv_len = T.int32(is_size_var=True)
q = T.match_buffer(var_q, (batch_size, qo_len, h_q, d), dtype)
Expand Down