[sgl-kernel] update deepgemm (#6942)

python/sglang/srt/models/deepseek_v2.py

@@ -1719,7 +1719,10 @@ class DeepseekV2ForCausalLM(nn.Module):
                     logger,
                     "Only Deepseek V3/R1 on NV-platform can use shared experts fusion optimization. Shared experts fusion optimization is disabled.",
                 )
-            elif (global_server_args_dict["enable_deepep_moe"] or global_server_args_dict["enable_ep_moe"]):
+            elif (
+                global_server_args_dict["enable_deepep_moe"]
+                or global_server_args_dict["enable_ep_moe"]
+            ):
                 self.num_fused_shared_experts = 0
                 global_server_args_dict["disable_shared_experts_fusion"] = True
                 log_info_on_rank0(

sgl-kernel/CMakeLists.txt

@@ -53,7 +53,7 @@ FetchContent_Populate(repo-cutlass)
 FetchContent_Declare(
     repo-deepgemm
     GIT_REPOSITORY https://github.com/deepseek-ai/DeepGEMM
-    GIT_TAG        d75b218b7b8f4a5dd5406ac87905039ead3ae42f
+    GIT_TAG        8dfa3298274bfe6b242f6f8a3e6f3eff2707dd9f
     GIT_SHALLOW    OFF
 )
 FetchContent_Populate(repo-deepgemm)