Skip to content

Commit eb58f59

Browse files
authored
[TPU][Bugfix] fix test_pallas (#20666)
Signed-off-by: Chengji Yao <chengjiyao@google.com>
1 parent 4ac9c33 commit eb58f59

File tree

1 file changed

+2
-0
lines changed

1 file changed

+2
-0
lines changed

tests/v1/tpu/test_pallas.py

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -50,6 +50,7 @@ class FakeAttentionLayer:
5050
slot_mapping = torch.zeros((3, num_tokens), dtype=torch.int64)
5151
max_num_reqs = 8
5252
max_num_blocks_per_req = 8
53+
num_kv_update_slices = torch.tensor([num_tokens], dtype=torch.int32)
5354
block_tables = torch.zeros((max_num_reqs, max_num_blocks_per_req),
5455
dtype=torch.int32)
5556
context_lens = torch.ones((max_num_reqs, ), dtype=torch.int32)
@@ -65,6 +66,7 @@ class FakeAttentionLayer:
6566
context_lens=context_lens,
6667
query_start_loc=query_start_loc,
6768
num_seqs=num_seqs,
69+
num_kv_update_slices=num_kv_update_slices,
6870
num_slices_per_kv_cache_update_block=8,
6971
)
7072

0 commit comments

Comments
 (0)