Skip to content

bugfix: fix the behavior of MLA kernel when kv-length is 0 #868

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
Feb 17, 2025
Merged

bugfix: fix the behavior of MLA kernel when kv-length is 0 #868

Show file tree
Hide file tree
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
4 changes: 3 additions & 1 deletion include/flashinfer/attention/scheduler.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1134,7 +1134,8 @@ inline cudaError_t MLAPlan(void* float_buffer, size_t float_workspace_size_in_by
qo_indptr_h[i] * num_heads +
std::min((qo_tile_idx + 1) * cluster_tile_q, packed_qo_len);
}
while (remaining_len > 0) {
bool zero_kv_len = (remaining_len == 0);
while (remaining_len > 0 || zero_kv_len) {
auto [cluster_idx, accum_cost] = cluster_cost_heap.pop();
int actual_len = std::min(remaining_len, kv_len_limit);
cluster_cost_heap.insert(
Expand All @@ -1154,6 +1155,7 @@ inline cudaError_t MLAPlan(void* float_buffer, size_t float_workspace_size_in_by
cluster_kv_end[cluster_idx].push_back(kv_start + actual_len);
remaining_len -= actual_len;
kv_start += actual_len;
if (zero_kv_len) break;
}
split_kv_count += int(split_kv);
}
Expand Down
5 changes: 3 additions & 2 deletions tests/test_deepseek_mla.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -171,7 +171,7 @@ def generate_kv_from_cache(ckv, kpe, kv_len, batch_size, num_heads):


@pytest.mark.parametrize("batch_size", [1, 17, 37])
@pytest.mark.parametrize("kv_len", [17, 33, 96, 97, 114, 514, 1024])
@pytest.mark.parametrize("kv_len", [0, 17, 33, 96, 97, 114, 514, 1024])
@pytest.mark.parametrize("qo_len", [1, 17, 37, 77])
@pytest.mark.parametrize("num_heads", [4, 32, 128])
@pytest.mark.parametrize("causal", [False, True])
Expand Down Expand Up @@ -243,7 +243,8 @@ def test_batch_mla_page_attention(
o_ref, lse_ref = attention_ref(batch_size, q, k, v, causal, sm_scale)
lse_ref = lse_ref.flatten(0, 1)
torch.testing.assert_close(o, o_ref, rtol=1e-3, atol=1e-3)
torch.testing.assert_close(lse, lse_ref, rtol=1e-3, atol=1e-3)
if kv_len != 0:
torch.testing.assert_close(lse, lse_ref, rtol=1e-3, atol=1e-3)

# test with pre-allocated output
o_buffer = torch.empty_like(o)
Expand Down