A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 13:30 UTC
[BugFix] Revert ROCm Custom Paged Attention Env Flag Check (#17022)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
vllmellm committed
Apr 23, 2025 at 10:17 UTC
7e081ba7cad2cf5c98376135de781fc76bfc103c
1 file changed
+1
vllm/platforms/rocm.py
+1
@@ -118,6 +118,7 @@ def use_rocm_custom_paged_attention(qtype: torch.dtype, head_size: int,
118
and (head_size == 64 or head_size == 128)
119
and (block_size == 16 or block_size == 32)
120
and (gqa_ratio >= 1 and gqa_ratio <= 16) and max_seq_len <= 32768
121
+ and (envs.VLLM_ROCM_CUSTOM_PAGED_ATTN)
122
and not (envs.VLLM_ROCM_USE_AITER_PAGED_ATTN
123
and envs.VLLM_ROCM_USE_AITER))
124