xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-14 18:38:38 +08:00
Code Issues Packages Projects Releases Wiki Activity

[TPU][Bugfix] fix test_pallas (#20666)

Browse Source 
Signed-off-by: Chengji Yao <chengjiyao@google.com>
This commit is contained in:
Chengji Yao 2025-07-09 09:32:48 -07:00 committed by GitHub
parent 4ac9c33f78
commit eb58f5953d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tests/v1/tpu/test_pallas.py
View File

@ -50,6 +50,7 @@ def test_ragged_paged_attention():
slot_mapping = torch.zeros((3, num_tokens), dtype=torch.int64) slot_mapping = torch.zeros((3, num_tokens), dtype=torch.int64)
max_num_reqs = 8 max_num_reqs = 8
max_num_blocks_per_req = 8 max_num_blocks_per_req = 8
num_kv_update_slices = torch.tensor([num_tokens], dtype=torch.int32)
block_tables = torch.zeros((max_num_reqs, max_num_blocks_per_req), block_tables = torch.zeros((max_num_reqs, max_num_blocks_per_req),
dtype=torch.int32) dtype=torch.int32)
context_lens = torch.ones((max_num_reqs, ), dtype=torch.int32) context_lens = torch.ones((max_num_reqs, ), dtype=torch.int32)
@ -65,6 +66,7 @@ def test_ragged_paged_attention():
context_lens=context_lens, context_lens=context_lens,
query_start_loc=query_start_loc, query_start_loc=query_start_loc,
num_seqs=num_seqs, num_seqs=num_seqs,
num_kv_update_slices=num_kv_update_slices,
num_slices_per_kv_cache_update_block=8, num_slices_per_kv_cache_update_block=8,
) )