[Refactor][MRV2] Unify uniform decode token count helper (#51917)

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
This commit is contained in:
Lucas Wilkinson
2026-08-12 05:58:43 +00:00
committed by GitHub
co-authored by OpenAI Codex
parent f2906b492e
commit 02ac17851d
3 changed files with 12 additions and 24 deletions
+8 -5
View File
@@ -124,10 +124,11 @@ def test_dynamic_sd_full_cudagraph_covers_all_uniform_decode_shapes(monkeypatch)
# Uniform decode means every request contributes the same number of
# tokens, so the total token count is exactly num_reqs * query_len.
num_tokens = num_reqs * max_query_len
uniform_tok_count = gpu_cudagraph_utils.get_uniform_token_count(
uniform_tok_count = get_uniform_decode_token_count(
num_reqs,
num_tokens,
max_query_len,
has_prefill=False,
)
# The scheduler should mark every one of these shapes as a uniform
@@ -233,8 +234,8 @@ def test_prompt_chunks_shaped_like_spec_decode_miss_the_full_graph(monkeypatch):
# The batch shape is indistinguishable from a full batch of spec decodes.
assert (
gpu_cudagraph_utils.get_uniform_token_count(
num_reqs, num_tokens, decode_query_len
get_uniform_decode_token_count(
num_reqs, num_tokens, decode_query_len, has_prefill=False
)
== decode_query_len
)
@@ -305,10 +306,11 @@ def test_basic_sd_does_not_capture_shorter_full_decode_shapes(monkeypatch):
# These are still uniform decode batches, but basic SD should only
# have FULL graphs for query_len == max_decode_query_len.
num_tokens = num_reqs * max_query_len
uniform_tok_count = gpu_cudagraph_utils.get_uniform_token_count(
uniform_tok_count = get_uniform_decode_token_count(
num_reqs,
num_tokens,
max_query_len,
has_prefill=False,
)
assert uniform_tok_count == max_query_len
@@ -370,10 +372,11 @@ def test_dynamic_sd_only_captures_scheduled_query_lengths(monkeypatch):
for num_reqs in range(1, max_num_seqs + 1):
for max_query_len in range(1, max_decode_query_len + 1):
num_tokens = num_reqs * max_query_len
uniform_tok_count = gpu_cudagraph_utils.get_uniform_token_count(
uniform_tok_count = get_uniform_decode_token_count(
num_reqs,
num_tokens,
max_query_len,
has_prefill=False,
)
assert uniform_tok_count == max_query_len