[Feat] Scale up fa3 kernel to sm8x arch (#5912)

Co-authored-by: zhyncs <me@zhyncs.com>

sgl-kernel/CMakeLists.txt

@@ -233,7 +233,7 @@ install(TARGETS common_ops LIBRARY DESTINATION sgl_kernel)
 
 # ============================ Optional Install ============================= #
 # set flash-attention sources file
-# BF16 source files
+# Now FA3 support sm80/sm86/sm90
 if (SGL_KERNEL_ENABLE_FA3)
     set(SGL_FLASH_KERNEL_CUDA_FLAGS
         "-DNDEBUG"
@@ -241,6 +241,8 @@ if (SGL_KERNEL_ENABLE_FA3)
         "-O3"
         "-Xcompiler"
         "-fPIC"
+        "-gencode=arch=compute_80,code=sm_80"
+        "-gencode=arch=compute_86,code=sm_86"
         "-gencode=arch=compute_90a,code=sm_90a"
         "-std=c++17"
         "-DCUTE_USE_PACKED_TUPLE=1"
@@ -256,6 +258,10 @@ if (SGL_KERNEL_ENABLE_FA3)
         "-Xcompiler=-fno-strict-aliasing"
     )
 
+    # SM8X Logic
+    file(GLOB FA3_SM8X_GEN_SRCS
+        "${repo-flash-attention_SOURCE_DIR}/hopper/instantiations/flash_fwd_hdim*_sm80.cu")
+
     file(GLOB FA3_BF16_GEN_SRCS
         "${repo-flash-attention_SOURCE_DIR}/hopper/instantiations/flash_fwd_hdimall_bf16*_sm90.cu")
     file(GLOB FA3_BF16_GEN_SRCS_
@@ -276,7 +282,7 @@ if (SGL_KERNEL_ENABLE_FA3)
         "${repo-flash-attention_SOURCE_DIR}/hopper/instantiations/flash_fwd_hdimdiff_e4m3*_sm90.cu")
     list(APPEND FA3_FP8_GEN_SRCS ${FA3_FP8_GEN_SRCS_})
 
-    set(FA3_GEN_SRCS ${FA3_BF16_GEN_SRCS} ${FA3_FP16_GEN_SRCS} ${FA3_FP8_GEN_SRCS})
+    set(FA3_GEN_SRCS ${FA3_BF16_GEN_SRCS} ${FA3_FP16_GEN_SRCS} ${FA3_FP8_GEN_SRCS} ${FA3_SM8X_GEN_SRCS})
 
     set(FLASH_SOURCES
         "csrc/flash_extension.cc"
@@ -297,7 +303,7 @@ if (SGL_KERNEL_ENABLE_FA3)
     install(TARGETS flash_ops LIBRARY DESTINATION "sgl_kernel")
 
     target_compile_definitions(flash_ops PRIVATE
-        FLASHATTENTION_DISABLE_SM8x
+        # FLASHATTENTION_DISABLE_SM8x
         FLASHATTENTION_DISABLE_BACKWARD
         FLASHATTENTION_DISABLE_DROPOUT
         FLASHATTENTION_DISABLE_UNEVEN_K

sgl-kernel/README.md

@@ -81,6 +81,14 @@ Third-party libraries:
 - [DeepGEMM](https://github.com/deepseek-ai/DeepGEMM)
 - [FlashAttention](https://github.com/Dao-AILab/flash-attention)
 
+### FlashAttention FYI
+
+  FA3 can fail without a enough shared memory for a some shapes, such as higher hidden_dim or some special cases. Right now, fa3 is supported for sm80/sm87 and sm86/sm89.
+
+  The main different Between sm80/sm87 and sm86/sm89 is the shared memory size. you can follow the link below for more information https://docs.nvidia.com/cuda/cuda-c-programming-guide/#shared-memory-8-x.
+
+  And for sgl-kernel right now, we can build fa3 on sm80/sm86/sm89/sm90a. Thats mean if you use **A100(tested)**/A*0/**L20(tested)**/L40/L40s/**3090(tested)** you can use fa3.
+
 ### Kernel Development
 
 Steps to add a new kernel:

sgl-kernel/python/sgl_kernel/flash_attn.py

@@ -10,10 +10,18 @@ except:
 
 
 def is_fa3_supported(device=None) -> bool:
-    #  now sgl-kernel only build fa3 for sm90a && cuda >= 12.3
+    #  There some fa3 FYI
-    return (torch.cuda.get_device_capability(device)[0] == 9) and (
+    #  FA3 can fail without a enough shared memory for a some shapes, such as higher
-        torch.version.cuda >= "12.3"
+    #  hidden_dim or some special cases.
-    )
+    #  Right now, fa3 is supported for sm80/sm87 and sm86/sm89. The main different
+    #  Between sm80/sm87 and sm86/sm89 is the shared memory size. you can follow the link below for more information
+    #  https://docs.nvidia.com/cuda/cuda-c-programming-guide/#shared-memory-8-x
+    #  And for sgl-kernel right now, we can build fa3 on sm80/sm86/sm89/sm90a.
+    #  Thats mean if you use A100/A*0/L20/L40/L40s/4090 you can use fa3.
+    return (
+        torch.cuda.get_device_capability(device)[0] == 9
+        or torch.cuda.get_device_capability(device)[0] == 8
+    ) and (torch.version.cuda >= "12.3")
 
 
 def maybe_contiguous(x):

sgl-kernel/tests/test_flash_attention.py

@@ -11,17 +11,24 @@ from einops import rearrange, repeat
 apply_rotary_emb = None
 
 
+def is_hopper():
+    #  Only Hopper supports different V headdim
+    return torch.cuda.get_device_properties(0).major >= 9
+
+
 def is_fa3_supported(device=None) -> bool:
-    # FA3 can fail without a enough shared memory for a some shapes, currently
+    #  There some fa3 FYI
-    #  only 8.0 and 8.7 have enough shared memory for all shapes
+    #  FA3 can fail without a enough shared memory for a some shapes, such as higher
+    #  hidden_dim or some special cases.
+    #  Right now, fa3 is supported for sm80/sm87 and sm86/sm89. The main different
+    #  Between sm80/sm87 and sm86/sm89 is the shared memory size. you can follow the link below for more information
     #  https://docs.nvidia.com/cuda/cuda-c-programming-guide/#shared-memory-8-x
-    #  now sgl-kernel only build fa3 for sm90a && cuda >= 12.4
+    #  And for sgl-kernel right now, we can build fa3 on sm80/sm86/sm89/sm90a.
+    #  Thats mean if you use A100/A*0/L20/L40/L40s/4090 you can use fa3.
     return (
-        (torch.cuda.get_device_capability(device)[0] == 9)
+        torch.cuda.get_device_capability(device)[0] == 9
-        and (torch.version.cuda >= "12.4")
+        or torch.cuda.get_device_capability(device)[0] == 8
-        # or torch.cuda.get_device_capability(device) == (8, 0)
+    ) and (torch.version.cuda >= "12.3")
-        # or torch.cuda.get_device_capability(device) == (8, 7)
-    )
 
 
 DISABLE_BACKWARD = True
@@ -558,7 +565,8 @@ def test_flash_attn_kvcache(
     assert nheads % nheads_k == 0
     dtype_ref = torch.bfloat16 if dtype == torch.float8_e4m3fn else dtype
     dv_vals = [128, d] if d > 128 and d <= 192 else ([256, 512, d] if d <= 64 else [d])
-    if dtype == torch.float8_e4m3fn:
+    if dtype == torch.float8_e4m3fn or not is_hopper():
+        # for fp8 and ampere arch, we not support v head dim != qk head dim
         dv_vals = [d]
     for dv in dv_vals:
         has_qv = d == 64 and dv >= 256