minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[Kernel] Migrate generic attention kernels to sglang.kernels (RFC #29630, Phase 2.5, 4/7) (#30789)

Browse Source 
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Xiaoyu Zhang
2026-07-14 16:53:46 +08:00
committed by GitHub
co-authored by Claude Fable 5
parent a5a71c6c26
commit 1a35440c4a
35 changed files with 173 additions and 143 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/manual/attention/test_trtllm_mla_backend.py
+1 -1
View File
@@ -10,13 +10,13 @@ from sglang.srt.runtime_context import get_parallel, get_server_args
_parallel_override = get_parallel().override(attn_tp_size=1)
_parallel_override.__enter__()
from sglang.kernels.ops.attention.utils import get_num_page_per_block_flashmla
from sglang.srt.configs.model_config import AttentionArch
from sglang.srt.layers.attention.flashinfer_mla_backend import FlashInferMLAAttnBackend
from sglang.srt.layers.attention.trtllm_mla_backend import (
TRTLLMMLABackend,
TRTLLMMLADecodeMetadata,
)
from sglang.srt.layers.attention.utils import get_num_page_per_block_flashmla
from sglang.srt.layers.radix_attention import RadixAttention
from sglang.srt.mem_cache.memory_pool import MLATokenToKVPool
from sglang.srt.model_executor.forward_batch_info import ForwardBatch, ForwardMode