Unverified Commit 3bffe112 authored by 633WHU's avatar 633WHU Committed by GitHub
Browse files

Fix chunked prefill size validation for disabled state (#8973)


Co-authored-by: default avatargemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
parent 44426e54
...@@ -1999,9 +1999,11 @@ class ServerArgs: ...@@ -1999,9 +1999,11 @@ class ServerArgs:
), "enable_mixed_chunk is required for speculative decoding" ), "enable_mixed_chunk is required for speculative decoding"
# Check chunked prefill # Check chunked prefill
assert ( # Skip validation if chunked prefill is disabled (i.e., size <= 0).
self.chunked_prefill_size % self.page_size == 0 if self.chunked_prefill_size > 0:
), "chunked_prefill_size must be divisible by page_size" assert (
self.chunked_prefill_size % self.page_size == 0
), "chunked_prefill_size must be divisible by page_size"
def check_lora_server_args(self): def check_lora_server_args(self):
assert self.max_loras_per_batch > 0, "max_loras_per_batch must be positive" assert self.max_loras_per_batch > 0, "max_loras_per_batch must be positive"
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment