[Kernel] Resubmit PR 48666 - Gemma4 FP8 KV FA4 head dim 512 backend selection (#53175)
Signed-off-by: Jhao-Ting Chen <jhaotingc@nvidia.com>
This commit is contained in:
@@ -29,7 +29,7 @@ from vllm.v1.kv_cache_interface import KVCacheSpecKind
|
||||
def test_largest_kernel_block_within(supported_sizes, expected):
|
||||
class Backend:
|
||||
@staticmethod
|
||||
def get_supported_kernel_block_sizes():
|
||||
def get_supported_kernel_block_sizes(kv_cache_spec=None):
|
||||
return supported_sizes
|
||||
|
||||
assert _largest_kernel_block_within(Backend, 1024, 1024 * 1536, 2048) == expected
|
||||
|
||||
Reference in New Issue
Block a user