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

[refactor] Move trtllm_fp8_kv_kernel to triton_ops directory (#15044)

Browse Source
This commit is contained in:
Hudson Xing
2025-12-14 15:22:37 -08:00
committed by GitHub
parent 0612175ce1
commit 99cb2ed988
3 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/srt/layers/attention/trtllm_fp8_kv_kernel.py → python/sglang/srt/layers/attention/triton_ops/trtllm_fp8_kv_kernel.py
View File
python/sglang/srt/layers/attention/trtllm_mha_backend.py
+3 -1
View File
@@ -15,7 +15,9 @@ from sglang.srt.layers.attention.flashinfer_backend import (
FlashInferAttnBackend, FlashInferAttnBackend,
FlashInferMultiStepDraftBackend, FlashInferMultiStepDraftBackend,
) )
from sglang.srt.layers.attention.trtllm_fp8_kv_kernel import fused_fp8_set_kv_buffer from sglang.srt.layers.attention.triton_ops.trtllm_fp8_kv_kernel import (
fused_fp8_set_kv_buffer,
)
from sglang.srt.model_executor.forward_batch_info import ForwardBatch, ForwardMode from sglang.srt.model_executor.forward_batch_info import ForwardBatch, ForwardMode
from sglang.srt.utils import is_flashinfer_available from sglang.srt.utils import is_flashinfer_available
test/manual/test_trtllm_fp8_kv_kernel.py
+3 -1
View File
@@ -6,7 +6,9 @@ import unittest
import torch import torch
from sglang.srt.layers.attention.trtllm_fp8_kv_kernel import fused_fp8_set_kv_buffer from sglang.srt.layers.attention.triton_ops.trtllm_fp8_kv_kernel import (
fused_fp8_set_kv_buffer,
)
from sglang.test.test_utils import CustomTestCase from sglang.test.test_utils import CustomTestCase