search

Dao-AILab / flash-attention

Fast and memory-efficient exact attention
BSD 3-Clause "New" or "Revised" License
14.18k stars 1.32k forks source link

[Flashv2] Windows support #345

Open danthe3rd opened 1 year ago

danthe3rd commented 1 year ago

Flashv2 is based on CUTLASS v3, which does not support windows at the moment. I have a bunch of errors - I'll try to dig a bit further, might be related to MSVC version, will report if I manage to get it to work.

Would be curious if anyone got it to work there (I don't have a windows machine to test things out)

mnicely commented 1 year ago

CUTLASS is planning to have official Windows support later this year

drisspg commented 1 year ago

Quick FYI: Cutlass 3.2 has a bullet point on support for windows build, I haven't had a chance to try building flash with 3.2 but at least for 3.1 still seeing errors around some of the copy utils.

tridao commented 1 year ago

Quick FYI: Cutlass 3.2 has a bullet point on support for windows build, I haven't had a chance to try building flash with 3.2 but at least for 3.1 still seeing errors around some of the copy utils.

Awesome! I'm hoping to find time to try Cutlass 3.2 soon!

Panchovix commented 1 year ago

I have tried with CUDA 12.1 on Windows 11, VS 2022 and no luck so far to build FAv2.

@danthe3rd have you found something?

mnicely commented 1 year ago

@Panchovix do you know what version of CUTLASS you used?

Panchovix commented 1 year ago

@Panchovix do you know what version of CUTLASS you used?

Sadly I don't know how to check. I just installed CUDA 11.8/12.1 independently, assigned each env variable for each torch version, and when trying to install, it fail with this, based if compiling from source or installing.

From source:

Compiling from source error ``` [2/49] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim160_fp16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim160_fp16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim160_fp16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 FAILED: F:/ChatIAs/oobabooga/flash-attention/build/temp.win-amd64-cpython-310/Release/csrc/flash_attn/src/flash_bwd_hdim160_fp16_sm80.obj C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim160_fp16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim160_fp16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim160_fp16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 flash_bwd_hdim160_fp16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim160_fp16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim160_fp16_sm80.cu F:/ChatIAs/oobabooga/flash-attention/csrc/cutlass/include\cute/arch/mma_sm90_desc.hpp(143): warning #226-D: invalid format string conversion printf("GmmaDescriptor: 0x%016 %lli\n", static_cast(t.desc_)); ^ Remark: The warnings can be suppressed with "-diag-suppress " tmpxft_00003160_00000000-7_flash_bwd_hdim160_fp16_sm80.cudafe1.cpp F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(270): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim160_fp16_sm80.cu(9): note: Vea la referencia a la creación de una instancia de la función plantilla "void run_mha_bwd_hdim160(Flash_bwd_params &,cudaStream_t,const bool)" que se está compilando F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(270): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(270): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(270): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' [3/49] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim160_bf16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim160_bf16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim160_bf16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 FAILED: F:/ChatIAs/oobabooga/flash-attention/build/temp.win-amd64-cpython-310/Release/csrc/flash_attn/src/flash_bwd_hdim160_bf16_sm80.obj C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim160_bf16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim160_bf16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim160_bf16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 flash_bwd_hdim160_bf16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim160_bf16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim160_bf16_sm80.cu F:/ChatIAs/oobabooga/flash-attention/csrc/cutlass/include\cute/arch/mma_sm90_desc.hpp(143): warning #226-D: invalid format string conversion printf("GmmaDescriptor: 0x%016 %lli\n", static_cast(t.desc_)); ^ Remark: The warnings can be suppressed with "-diag-suppress " tmpxft_00005ccc_00000000-7_flash_bwd_hdim160_bf16_sm80.cudafe1.cpp F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(270): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim160_bf16_sm80.cu(9): note: Vea la referencia a la creación de una instancia de la función plantilla "void run_mha_bwd_hdim160(Flash_bwd_params &,cudaStream_t,const bool)" que se está compilando F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(270): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(270): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(270): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' [4/49] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim192_bf16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim192_bf16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim192_bf16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 FAILED: F:/ChatIAs/oobabooga/flash-attention/build/temp.win-amd64-cpython-310/Release/csrc/flash_attn/src/flash_bwd_hdim192_bf16_sm80.obj C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim192_bf16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim192_bf16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim192_bf16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 flash_bwd_hdim192_bf16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim192_bf16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim192_bf16_sm80.cu F:/ChatIAs/oobabooga/flash-attention/csrc/cutlass/include\cute/arch/mma_sm90_desc.hpp(143): warning #226-D: invalid format string conversion printf("GmmaDescriptor: 0x%016 %lli\n", static_cast(t.desc_)); ^ Remark: The warnings can be suppressed with "-diag-suppress " tmpxft_000038c0_00000000-7_flash_bwd_hdim192_bf16_sm80.cudafe1.cpp F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(287): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim192_bf16_sm80.cu(9): note: Vea la referencia a la creación de una instancia de la función plantilla "void run_mha_bwd_hdim192(Flash_bwd_params &,cudaStream_t,const bool)" que se está compilando F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(287): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(287): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(287): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' [5/49] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim192_fp16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim192_fp16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim192_fp16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 FAILED: F:/ChatIAs/oobabooga/flash-attention/build/temp.win-amd64-cpython-310/Release/csrc/flash_attn/src/flash_bwd_hdim192_fp16_sm80.obj C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim192_fp16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim192_fp16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim192_fp16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 flash_bwd_hdim192_fp16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim192_fp16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim192_fp16_sm80.cu F:/ChatIAs/oobabooga/flash-attention/csrc/cutlass/include\cute/arch/mma_sm90_desc.hpp(143): warning #226-D: invalid format string conversion printf("GmmaDescriptor: 0x%016 %lli\n", static_cast(t.desc_)); ^ Remark: The warnings can be suppressed with "-diag-suppress " tmpxft_00002c68_00000000-7_flash_bwd_hdim192_fp16_sm80.cudafe1.cpp F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(287): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim192_fp16_sm80.cu(9): note: Vea la referencia a la creación de una instancia de la función plantilla "void run_mha_bwd_hdim192(Flash_bwd_params &,cudaStream_t,const bool)" que se está compilando F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(287): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(287): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(287): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' [6/49] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim128_fp16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim128_fp16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim128_fp16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 FAILED: F:/ChatIAs/oobabooga/flash-attention/build/temp.win-amd64-cpython-310/Release/csrc/flash_attn/src/flash_bwd_hdim128_fp16_sm80.obj C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim128_fp16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim128_fp16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim128_fp16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 flash_bwd_hdim128_fp16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim128_fp16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim128_fp16_sm80.cu F:/ChatIAs/oobabooga/flash-attention/csrc/cutlass/include\cute/arch/mma_sm90_desc.hpp(143): warning #226-D: invalid format string conversion printf("GmmaDescriptor: 0x%016 %lli\n", static_cast(t.desc_)); ^ Remark: The warnings can be suppressed with "-diag-suppress " tmpxft_000030a8_00000000-7_flash_bwd_hdim128_fp16_sm80.cudafe1.cpp F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(235): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim128_fp16_sm80.cu(9): note: Vea la referencia a la creación de una instancia de la función plantilla "void run_mha_bwd_hdim128(Flash_bwd_params &,cudaStream_t,const bool)" que se está compilando F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(235): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(235): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(235): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' [7/49] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim128_bf16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim128_bf16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim128_bf16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 FAILED: F:/ChatIAs/oobabooga/flash-attention/build/temp.win-amd64-cpython-310/Release/csrc/flash_attn/src/flash_bwd_hdim128_bf16_sm80.obj C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\bin\nvcc --generate-dependencies-with-compile --dependency-output F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim128_bf16_sm80.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn -IF:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src -IF:\ChatIAs\oobabooga\flash-attention\csrc\cutlass\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\torch\csrc\api\include -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\TH -IF:\ChatIAs\oobabooga\venv\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include" -IF:\ChatIAs\oobabooga\venv\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\include -IC:\Users\Pancho\AppData\Local\Programs\Python\Python310\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.36.32532\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22000.0\\cppwinrt" -c F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim128_bf16_sm80.cu -o F:\ChatIAs\oobabooga\flash-attention\build\temp.win-amd64-cpython-310\Release\csrc/flash_attn/src/flash_bwd_hdim128_bf16_sm80.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -O3 -std=c++17 -U__CUDA_NO_HALF_OPERATORS__ -U__CUDA_NO_HALF_CONVERSIONS__ -U__CUDA_NO_HALF2_OPERATORS__ -U__CUDA_NO_BFLOAT16_CONVERSIONS__ --expt-relaxed-constexpr --expt-extended-lambda --use_fast_math -lineinfo -gencode arch=compute_80,code=sm_80 --threads 4 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=flash_attn_2_cuda -D_GLIBCXX_USE_CXX11_ABI=0 flash_bwd_hdim128_bf16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim128_bf16_sm80.cu cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_OPERATORS__' con '/U__CUDA_NO_HALF_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF_CONVERSIONS__' con '/U__CUDA_NO_HALF_CONVERSIONS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_HALF2_OPERATORS__' con '/U__CUDA_NO_HALF2_OPERATORS__' cl : Línea de comandos warning D9025 : invalidando '/D__CUDA_NO_BFLOAT16_CONVERSIONS__' con '/U__CUDA_NO_BFLOAT16_CONVERSIONS__' flash_bwd_hdim128_bf16_sm80.cu F:/ChatIAs/oobabooga/flash-attention/csrc/cutlass/include\cute/arch/mma_sm90_desc.hpp(143): warning #226-D: invalid format string conversion printf("GmmaDescriptor: 0x%016 %lli\n", static_cast(t.desc_)); ^ Remark: The warnings can be suppressed with "-diag-suppress " tmpxft_0000556c_00000000-7_flash_bwd_hdim128_bf16_sm80.cudafe1.cpp F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(235): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_hdim128_bf16_sm80.cu(9): note: Vea la referencia a la creación de una instancia de la función plantilla "void run_mha_bwd_hdim128(Flash_bwd_params &,cudaStream_t,const bool)" que se está compilando F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(235): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(235): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\flash_bwd_launch_template.h(235): error C2975: "kHeadDim_": argumento de plantilla no válido para "Flash_bwd_kernel_traits"; se esperaba una expresión constante en tiempo de compilación F:\ChatIAs\oobabooga\flash-attention\csrc\flash_attn\src\kernel_traits.h(186): note: vea la declaración de 'kHeadDim_' ninja: build stopped: subcommand failed. Traceback (most recent call last): File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 2100, in _run_ninja_build subprocess.run( File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\subprocess.py", line 526, in run raise CalledProcessError(retcode, process.args, subprocess.CalledProcessError: Command '['ninja', '-v', '-j', '6']' returned non-zero exit status 1. The above exception was the direct cause of the following exception: Traceback (most recent call last): File "F:\ChatIAs\oobabooga\flash-attention\setup.py", line 287, in setup( File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\__init__.py", line 87, in setup return distutils.core.setup(**attrs) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\core.py", line 185, in setup return run_commands(dist) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\core.py", line 201, in run_commands dist.run_commands() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 968, in run_commands self.run_command(cmd) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\dist.py", line 1217, in run_command super().run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command cmd_obj.run() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\install.py", line 74, in run self.do_egg_install() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\install.py", line 123, in do_egg_install self.run_command('bdist_egg') File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\cmd.py", line 319, in run_command self.distribution.run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\dist.py", line 1217, in run_command super().run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command cmd_obj.run() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\bdist_egg.py", line 165, in run cmd = self.call_command('install_lib', warn_dir=0) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\bdist_egg.py", line 151, in call_command self.run_command(cmdname) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\cmd.py", line 319, in run_command self.distribution.run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\dist.py", line 1217, in run_command super().run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command cmd_obj.run() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\install_lib.py", line 11, in run self.build() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\install_lib.py", line 112, in build self.run_command('build_ext') File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\cmd.py", line 319, in run_command self.distribution.run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\dist.py", line 1217, in run_command super().run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command cmd_obj.run() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\build_ext.py", line 84, in run _build_ext.run(self) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 346, in run self.build_extensions() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 873, in build_extensions build_ext.build_extensions(self) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 466, in build_extensions self._build_extensions_serial() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 492, in _build_extensions_serial self.build_extension(ext) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\build_ext.py", line 246, in build_extension _build_ext.build_extension(self, ext) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 547, in build_extension objects = self.compiler.compile( File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 845, in win_wrap_ninja_compile _write_ninja_file_and_compile_objects( File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 1774, in _write_ninja_file_and_compile_objects _run_ninja_build( File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 2116, in _run_ninja_build raise RuntimeError(message) from e RuntimeError: Error compiling objects for extension ```
Install from pip error ``` ninja: build stopped: subcommand failed. Traceback (most recent call last): File "C:\Users\Pancho\AppData\Local\Temp\pip-install-0q3amvk2\flash-attn_1ac95a7d9f7749dd90e6733135f93c62\setup.py", line 257, in run urllib.request.urlretrieve(wheel_url, wheel_filename) File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\urllib\request.py", line 241, in urlretrieve with contextlib.closing(urlopen(url, data)) as fp: File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\urllib\request.py", line 216, in urlopen return opener.open(url, data, timeout) File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\urllib\request.py", line 525, in open response = meth(req, response) File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\urllib\request.py", line 634, in http_response response = self.parent.error( File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\urllib\request.py", line 563, in error return self._call_chain(*args) File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\urllib\request.py", line 496, in _call_chain result = func(*args) File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\urllib\request.py", line 643, in http_error_default raise HTTPError(req.full_url, code, msg, hdrs, fp) urllib.error.HTTPError: HTTP Error 404: Not Found During handling of the above exception, another exception occurred: Traceback (most recent call last): File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 2100, in _run_ninja_build subprocess.run( File "C:\Users\Pancho\AppData\Local\Programs\Python\Python310\lib\subprocess.py", line 526, in run raise CalledProcessError(retcode, process.args, subprocess.CalledProcessError: Command '['ninja', '-v', '-j', '6']' returned non-zero exit status 1. The above exception was the direct cause of the following exception: Traceback (most recent call last): File "", line 2, in File "", line 34, in File "C:\Users\Pancho\AppData\Local\Temp\pip-install-0q3amvk2\flash-attn_1ac95a7d9f7749dd90e6733135f93c62\setup.py", line 277, in setup( File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\__init__.py", line 87, in setup return distutils.core.setup(**attrs) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\core.py", line 185, in setup return run_commands(dist) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\core.py", line 201, in run_commands dist.run_commands() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 968, in run_commands self.run_command(cmd) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\dist.py", line 1217, in run_command super().run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command cmd_obj.run() File "C:\Users\Pancho\AppData\Local\Temp\pip-install-0q3amvk2\flash-attn_1ac95a7d9f7749dd90e6733135f93c62\setup.py", line 274, in run super().run() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\wheel\bdist_wheel.py", line 343, in run self.run_command("build") File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\cmd.py", line 319, in run_command self.distribution.run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\dist.py", line 1217, in run_command super().run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command cmd_obj.run() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build.py", line 132, in run self.run_command(cmd_name) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\cmd.py", line 319, in run_command self.distribution.run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\dist.py", line 1217, in run_command super().run_command(command) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\dist.py", line 987, in run_command cmd_obj.run() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\build_ext.py", line 84, in run _build_ext.run(self) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 346, in run self.build_extensions() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 873, in build_extensions build_ext.build_extensions(self) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 466, in build_extensions self._build_extensions_serial() File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 492, in _build_extensions_serial self.build_extension(ext) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\command\build_ext.py", line 246, in build_extension _build_ext.build_extension(self, ext) File "F:\ChatIAs\oobabooga\venv\lib\site-packages\setuptools\_distutils\command\build_ext.py", line 547, in build_extension objects = self.compiler.compile( File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 845, in win_wrap_ninja_compile _write_ninja_file_and_compile_objects( File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 1774, in _write_ninja_file_and_compile_objects _run_ninja_build( File "F:\ChatIAs\oobabooga\venv\lib\site-packages\torch\utils\cpp_extension.py", line 2116, in _run_ninja_build raise RuntimeError(message) from e RuntimeError: Error compiling objects for extension [end of output] note: This error originates from a subprocess, and is likely not a problem with pip. ERROR: Failed building wheel for flash-attn Running setup.py clean for flash-attn Failed to build flash-attn ERROR: Could not build wheels for flash-attn, which is required to install pyproject.toml-based projects ```
grimulkan commented 1 year ago

Sadly I don't know how to check. I just installed CUDA 11.8/12.1 independently, assigned each env variable for each torch version, and when trying to install, it fail with this, based if compiling from source or installing.

Technically this repo should pull cutlass @ 34fd980 if you cloned it recently, which I think should have been 3.2

Panchovix commented 1 year ago

Sadly I don't know how to check. I just installed CUDA 11.8/12.1 independently, assigned each env variable for each torch version, and when trying to install, it fail with this, based if compiling from source or installing.

Technically this repo should pull cutlass @ 34fd980 if you cloned it recently, which I think should have been 3.2

I have cloned it every day I think, but the issue persists, so I'm not sure what would be happening. The error log is not that conclusive for me.

BadisG commented 1 year ago

If someone found the fix, please tell us how to make it work :(

grimulkan commented 1 year ago

Interestingly flash-attn v1 does build on Windows, as pointed out on Reddit

fenglui commented 1 year ago

https://github.com/NVIDIA/cutlass/blob/ff02da266713bd3365aed65c552412e126c040cb/media/docs/build/building_in_windows_with_visual_studio.md?plain=1#L4