Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

[Bug] Unable to Setup STA #193

Open
TemporalLabsLLC-SOL opened this issue Feb 19, 2025 · 7 comments
Open

[Bug] Unable to Setup STA #193

TemporalLabsLLC-SOL opened this issue Feb 19, 2025 · 7 comments

Comments

@TemporalLabsLLC-SOL
Copy link

Environment

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Tue_Feb_27_16:19:38_PST_2024
Cuda compilation tools, release 12.4, V12.4.99
Build cuda_12.4.r12.4/compiler.33961263_0

Python 3.10.16

Describe the bug

I think the link to TK may not be working? When I download the ZIP it does not contain the proper files. It may be some local bug but if anyone else is getting this I could use the help getting it going on an H100.

Reproduction

I'm trying to do the STA setup

@jzhang38
Copy link
Collaborator

I just pushed a commit: f9482d1 Can you try again

@TemporalLabsLLC-SOL
Copy link
Author

I just pushed a commit: f9482d1 Can you try again

I just did a new zip download and clone and neither have a populated tk folder.

@jzhang38
Copy link
Collaborator

You need to git pull --recurse-submodules

@TemporalLabsLLC-SOL
Copy link
Author

You need to git pull --recurse-submodules

(fastvideo2) azureuser@TemporalProcessorVideoCore:~/FastVideo/csrc/sliding_tile_attention$ python setup.py install
st_attn root: /home/azureuser/FastVideo/csrc/sliding_tile_attention/tk
Python include: /home/azureuser/anaconda3/envs/fastvideo2/include/python3.10
Torch include directories: -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/TH -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/THC
running install
/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/cmd.py:79: SetuptoolsDeprecationWarning: setup.py install is deprecated.
!!

    ********************************************************************************
    Please avoid running ``setup.py`` directly.
    Instead, use pypa/build, pypa/installer or other
    standards-based tools.

    See https://blog.ganssle.io/articles/2021/10/setup-py-deprecated.html for details.
    ********************************************************************************

!!
self.initialize_options()
/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/cmd.py:79: EasyInstallDeprecationWarning: easy_install command is deprecated.
!!

    ********************************************************************************
    Please avoid running ``setup.py`` and ``easy_install``.
    Instead, use pypa/build, pypa/installer or other
    standards-based tools.

    See https://github.com/pypa/setuptools/issues/917 for details.
    ********************************************************************************

!!
self.initialize_options()
running bdist_egg
running egg_info
creating st_attn.egg-info
writing st_attn.egg-info/PKG-INFO
writing dependency_links to st_attn.egg-info/dependency_links.txt
writing top-level names to st_attn.egg-info/top_level.txt
writing manifest file 'st_attn.egg-info/SOURCES.txt'
reading manifest file 'st_attn.egg-info/SOURCES.txt'
writing manifest file 'st_attn.egg-info/SOURCES.txt'
installing library code to build/bdist.linux-x86_64/egg
running install_lib
running build_py
creating build/lib.linux-x86_64-cpython-310/st_attn
copying st_attn/init.py -> build/lib.linux-x86_64-cpython-310/st_attn
running build_ext
/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/utils/cpp_extension.py:426: UserWarning: There are no g++ version bounds defined for CUDA version 12.4
warnings.warn(f'There are no {compiler_name} version bounds defined for CUDA version {cuda_str_version}')
building 'st_attn_cuda' extension
creating /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310
creating /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/st_attn
Emitting ninja build file /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/build.ninja...
Compiling objects...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/2] /usr/local/cuda-12.4/bin/nvcc --generate-dependencies-with-compile --dependency-output /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/st_attn/st_attn_h100.o.d -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/TH -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/THC -I/usr/local/cuda-12.4/include -I/home/azureuser/anaconda3/envs/fastvideo2/include/python3.10 -c -c /home/azureuser/FastVideo/csrc/sliding_tile_attention/st_attn/st_attn_h100.cu -o /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/st_attn/st_attn_h100.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options ''"'"'-fPIC'"'"'' -DNDEBUG -Xcompiler=-Wno-psabi -Xcompiler=-fno-strict-aliasing --expt-extended-lambda --expt-relaxed-constexpr -forward-unknown-to-host-compiler --use_fast_math -std=c++20 -O3 -Xnvlink=--verbose -Xptxas=--verbose -Xptxas=--warn-on-spills -I/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include -I/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/prototype -I/home/azureuser/anaconda3/envs/fastvideo2/include/python3.10 -DTORCH_COMPILE -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/TH -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/THC -DKITTENS_HOPPER -arch=sm_90a -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="gcc"' '-DPYBIND11_STDLIB="libstdcpp"' '-DPYBIND11_BUILD_ABI="cxxabi1011"' -DTORCH_EXTENSION_NAME=st_attn_cuda -D_GLIBCXX_USE_CXX11_ABI=0
FAILED: /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/st_attn/st_attn_h100.o
/usr/local/cuda-12.4/bin/nvcc --generate-dependencies-with-compile --dependency-output /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/st_attn/st_attn_h100.o.d -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/TH -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/THC -I/usr/local/cuda-12.4/include -I/home/azureuser/anaconda3/envs/fastvideo2/include/python3.10 -c -c /home/azureuser/FastVideo/csrc/sliding_tile_attention/st_attn/st_attn_h100.cu -o /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/st_attn/st_attn_h100.o -D__CUDA_NO_HALF_OPERATORS
-D__CUDA_NO_HALF_CONVERSIONS
_ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options ''"'"'-fPIC'"'"'' -DNDEBUG -Xcompiler=-Wno-psabi -Xcompiler=-fno-strict-aliasing --expt-extended-lambda --expt-relaxed-constexpr -forward-unknown-to-host-compiler --use_fast_math -std=c++20 -O3 -Xnvlink=--verbose -Xptxas=--verbose -Xptxas=--warn-on-spills -I/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include -I/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/prototype -I/home/azureuser/anaconda3/envs/fastvideo2/include/python3.10 -DTORCH_COMPILE -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/TH -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/THC -DKITTENS_HOPPER -arch=sm_90a -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=st_attn_cuda -D_GLIBCXX_USE_CXX11_ABI=0
/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(123): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr bf16 zero() { return std::bit_cast<__nv_bfloat16>(uint16_t(0x0000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(123): error: type name is not allowed
static attribute((device)) inline constexpr bf16 zero() { return std::bit_cast<__nv_bfloat16>(uint16_t(0x0000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(124): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr bf16 one() { return std::bit_cast<__nv_bfloat16>(uint16_t(0x3F80)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(124): error: type name is not allowed
static attribute((device)) inline constexpr bf16 one() { return std::bit_cast<__nv_bfloat16>(uint16_t(0x3F80)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(125): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr bf16 pos_infty() { return std::bit_cast<__nv_bfloat16>(uint16_t(0x7F80)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(125): error: type name is not allowed
static attribute((device)) inline constexpr bf16 pos_infty() { return std::bit_cast<__nv_bfloat16>(uint16_t(0x7F80)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(126): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr bf16 neg_infty() { return std::bit_cast<__nv_bfloat16>(uint16_t(0xFF80)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(126): error: type name is not allowed
static attribute((device)) inline constexpr bf16 neg_infty() { return std::bit_cast<__nv_bfloat16>(uint16_t(0xFF80)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(135): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr half zero() { return std::bit_cast<__half>(uint16_t(0x0000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(135): error: type name is not allowed
static attribute((device)) inline constexpr half zero() { return std::bit_cast<__half>(uint16_t(0x0000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(136): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr half one() { return std::bit_cast<__half>(uint16_t(0x3C00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(136): error: type name is not allowed
static attribute((device)) inline constexpr half one() { return std::bit_cast<__half>(uint16_t(0x3C00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(137): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr half pos_infty() { return std::bit_cast<__half>(uint16_t(0x7C00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(137): error: type name is not allowed
static attribute((device)) inline constexpr half pos_infty() { return std::bit_cast<__half>(uint16_t(0x7C00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(138): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr half neg_infty() { return std::bit_cast<__half>(uint16_t(0xFC00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(138): error: type name is not allowed
static attribute((device)) inline constexpr half neg_infty() { return std::bit_cast<__half>(uint16_t(0xFC00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(148): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e4m3 zero() { return std::bit_cast<__nv_fp8_e4m3>(uint8_t(0x00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(148): error: type name is not allowed
static attribute((device)) inline constexpr fp8e4m3 zero() { return std::bit_cast<__nv_fp8_e4m3>(uint8_t(0x00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(149): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e4m3 one() { return std::bit_cast<__nv_fp8_e4m3>(uint8_t(0x38)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(149): error: type name is not allowed
static attribute((device)) inline constexpr fp8e4m3 one() { return std::bit_cast<__nv_fp8_e4m3>(uint8_t(0x38)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(152): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e4m3_2 zero() { return std::bit_cast<fp8e4m3_2>(uint16_t(0x0000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(152): error: type name is not allowed
static attribute((device)) inline constexpr fp8e4m3_2 zero() { return std::bit_cast<fp8e4m3_2>(uint16_t(0x0000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(153): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e4m3_2 one() { return std::bit_cast<fp8e4m3_2>(uint16_t(0x3838)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(153): error: type name is not allowed
static attribute((device)) inline constexpr fp8e4m3_2 one() { return std::bit_cast<fp8e4m3_2>(uint16_t(0x3838)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(156): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e4m3_4 zero() { return std::bit_cast<fp8e4m3_4>(uint32_t(0x00000000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(156): error: type name is not allowed
static attribute((device)) inline constexpr fp8e4m3_4 zero() { return std::bit_cast<fp8e4m3_4>(uint32_t(0x00000000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(157): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e4m3_4 one() { return std::bit_cast<fp8e4m3_4>(uint32_t(0x38383838)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(157): error: type name is not allowed
static attribute((device)) inline constexpr fp8e4m3_4 one() { return std::bit_cast<fp8e4m3_4>(uint32_t(0x38383838)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(160): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e5m2 zero() { return std::bit_cast<__nv_fp8_e5m2>(uint8_t(0x00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(160): error: type name is not allowed
static attribute((device)) inline constexpr fp8e5m2 zero() { return std::bit_cast<__nv_fp8_e5m2>(uint8_t(0x00)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(161): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e5m2 one() { return std::bit_cast<__nv_fp8_e5m2>(uint8_t(0x3C)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(161): error: type name is not allowed
static attribute((device)) inline constexpr fp8e5m2 one() { return std::bit_cast<__nv_fp8_e5m2>(uint8_t(0x3C)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(164): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e5m2_2 zero() { return std::bit_cast<fp8e5m2_2>(uint16_t(0x0000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(164): error: type name is not allowed
static attribute((device)) inline constexpr fp8e5m2_2 zero() { return std::bit_cast<fp8e5m2_2>(uint16_t(0x0000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(165): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e5m2_2 one() { return std::bit_cast<fp8e5m2_2>(uint16_t(0x3C3C)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(165): error: type name is not allowed
static attribute((device)) inline constexpr fp8e5m2_2 one() { return std::bit_cast<fp8e5m2_2>(uint16_t(0x3C3C)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(168): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e5m2_4 zero() { return std::bit_cast<fp8e5m2_4>(uint32_t(0x00000000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(168): error: type name is not allowed
static attribute((device)) inline constexpr fp8e5m2_4 zero() { return std::bit_cast<fp8e5m2_4>(uint32_t(0x00000000)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(169): error: namespace "std" has no member "bit_cast"
static attribute((device)) inline constexpr fp8e5m2_4 one() { return std::bit_cast<fp8e5m2_4>(uint32_t(0x3C3C3C3C)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/common/base_types.cuh(169): error: type name is not allowed
static attribute((device)) inline constexpr fp8e5m2_4 one() { return std::bit_cast<fp8e5m2_4>(uint32_t(0x3C3C3C3C)); }
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/ops/warp/register/tile/complex/complex_mma.cuh(42): error: namespace "std" has no member "bit_cast"
constexpr half factor = std::bit_cast<__half>(uint16_t(0xFB80));
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/ops/warp/register/tile/complex/complex_mma.cuh(42): error: type name is not allowed
constexpr half factor = std::bit_cast<__half>(uint16_t(0xFB80));
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/ops/warp/register/tile/complex/complex_mma.cuh(77): error: namespace "std" has no member "bit_cast"
constexpr bf16 factor = std::bit_cast<__nv_bfloat16>(uint16_t(0xBF80));
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/tk/include/ops/warp/register/tile/complex/complex_mma.cuh(77): error: type name is not allowed
constexpr bf16 factor = std::bit_cast<__nv_bfloat16>(uint16_t(0xBF80));
^

/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include/torch/nn/modules/container/any_module_holder.h(89): warning #3189-D: "module" is parsed as an identifier rather than a keyword because the tokens that follow it do not match those of a preprocessor directive
module->_forward_populate_default_args(std::move(arguments)));
^

Remark: The warnings can be suppressed with "-diag-suppress "

/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include/torch/python.h(93): warning #3189-D: "module" is parsed as an identifier rather than a keyword because the tokens that follow it do not match those of a preprocessor directive
module.attr(name) = wrapper_class;
^

/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include/torch/python.h(155): warning #3189-D: "module" is parsed as an identifier rather than a keyword because the tokens that follow it do not match those of a preprocessor directive
module.to(
^

/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include/torch/python.h(159): warning #3189-D: "module" is parsed as an identifier rather than a keyword because the tokens that follow it do not match those of a preprocessor directive
module.to(detail::py_object_to_dtype(object), non_blocking);
^

/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include/torch/python.h(170): warning #3189-D: "module" is parsed as an identifier rather than a keyword because the tokens that follow it do not match those of a preprocessor directive
module.to(detail::py_object_to_dtype(dtype), non_blocking);
^

/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include/torch/python.h(172): warning #3189-D: "module" is parsed as an identifier rather than a keyword because the tokens that follow it do not match those of a preprocessor directive
module.to(detail::py_object_to_device(device), non_blocking);
^

/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include/torch/python.h(174): warning #3189-D: "module" is parsed as an identifier rather than a keyword because the tokens that follow it do not match those of a preprocessor directive
module.to(
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/st_attn/st_attn_h100.cu(398): warning #191-D: type qualifier is meaningless on cast type
torch::Tensor l_vec = torch::empty({static_cast(batch),
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/st_attn/st_attn_h100.cu(399): warning #191-D: type qualifier is meaningless on cast type
static_cast(qo_heads),
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/st_attn/st_attn_h100.cu(400): warning #191-D: type qualifier is meaningless on cast type
static_cast(seq_len),
^

/home/azureuser/FastVideo/csrc/sliding_tile_attention/st_attn/st_attn_h100.cu(401): warning #191-D: type qualifier is meaningless on cast type
static_cast(1)},
^

44 errors detected in the compilation of "/home/azureuser/FastVideo/csrc/sliding_tile_attention/st_attn/st_attn_h100.cu".
[2/2] c++ -MMD -MF /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/st_attn.o.d -pthread -B /home/azureuser/anaconda3/envs/fastvideo2/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -fwrapv -O2 -Wall -fPIC -O2 -isystem /home/azureuser/anaconda3/envs/fastvideo2/include -fPIC -O2 -isystem /home/azureuser/anaconda3/envs/fastvideo2/include -fPIC -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/TH -I/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/include/THC -I/usr/local/cuda-12.4/include -I/home/azureuser/anaconda3/envs/fastvideo2/include/python3.10 -c -c /home/azureuser/FastVideo/csrc/sliding_tile_attention/st_attn.cpp -o /home/azureuser/FastVideo/csrc/sliding_tile_attention/build/temp.linux-x86_64-cpython-310/st_attn.o -std=c++20 -O3 -DTK_COMPILE_ATTN -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1011"' -DTORCH_EXTENSION_NAME=st_attn_cuda -D_GLIBCXX_USE_CXX11_ABI=0
ninja: build stopped: subcommand failed.
Traceback (most recent call last):
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 2104, in _run_ninja_build
subprocess.run(
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/subprocess.py", line 526, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/home/azureuser/FastVideo/csrc/sliding_tile_attention/setup.py", line 52, in
setup(name='st_attn',
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/init.py", line 117, in setup
return distutils.core.setup(**attrs)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/core.py", line 186, in setup
return run_commands(dist)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/core.py", line 202, in run_commands
dist.run_commands()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/dist.py", line 983, in run_commands
self.run_command(cmd)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/dist.py", line 999, in run_command
super().run_command(command)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/dist.py", line 1002, in run_command
cmd_obj.run()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/command/install.py", line 109, in run
self.do_egg_install()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/command/install.py", line 167, in do_egg_install
self.run_command('bdist_egg')
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/cmd.py", line 339, in run_command
self.distribution.run_command(command)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/dist.py", line 999, in run_command
super().run_command(command)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/dist.py", line 1002, in run_command
cmd_obj.run()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/command/bdist_egg.py", line 177, in run
cmd = self.call_command('install_lib', warn_dir=False)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/command/bdist_egg.py", line 163, in call_command
self.run_command(cmdname)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/cmd.py", line 339, in run_command
self.distribution.run_command(command)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/dist.py", line 999, in run_command
super().run_command(command)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/dist.py", line 1002, in run_command
cmd_obj.run()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/command/install_lib.py", line 19, in run
self.build()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/command/install_lib.py", line 110, in build
self.run_command('build_ext')
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/cmd.py", line 339, in run_command
self.distribution.run_command(command)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/dist.py", line 999, in run_command
super().run_command(command)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/dist.py", line 1002, in run_command
cmd_obj.run()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/command/build_ext.py", line 99, in run
_build_ext.run(self)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/command/build_ext.py", line 365, in run
self.build_extensions()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 868, in build_extensions
build_ext.build_extensions(self)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/command/build_ext.py", line 481, in build_extensions
self._build_extensions_serial()
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/command/build_ext.py", line 507, in _build_extensions_serial
self.build_extension(ext)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/command/build_ext.py", line 264, in build_extension
_build_ext.build_extension(self, ext)
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/setuptools/_distutils/command/build_ext.py", line 562, in build_extension
objects = self.compiler.compile(
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 681, in unix_wrap_ninja_compile
_write_ninja_file_and_compile_objects(
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 1784, in _write_ninja_file_and_compile_objects
_run_ninja_build(
File "/home/azureuser/anaconda3/envs/fastvideo2/lib/python3.10/site-packages/torch/utils/cpp_extension.py", line 2120, in _run_ninja_build
raise RuntimeError(message) from e
RuntimeError: Error compiling objects for extension
(fastvideo2) azureuser@TemporalProcessorVideoCore:/FastVideo/csrc/sliding_tile_attention$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Tue_Feb_27_16:19:38_PST_2024
Cuda compilation tools, release 12.4, V12.4.99
Build cuda_12.4.r12.4/compiler.33961263_0
(fastvideo2) azureuser@TemporalProcessorVideoCore:
/FastVideo/csrc/sliding_tile_attention$ python --version
Python 3.10.16
(fastvideo2) azureuser@TemporalProcessorVideoCore:~/FastVideo/csrc/sliding_tile_attention$ nvidia-smi
Thu Feb 20 07:40:08 2025
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.183.01 Driver Version: 535.183.01 CUDA Version: 12.2 |
|-----------------------------------------+----------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 NVIDIA H100 NVL Off | 00000001:00:00.0 Off | 0 |
| N/A 30C P0 62W / 400W | 85MiB / 95830MiB | 0% Default |
| | | Disabled |
+-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| 0 N/A N/A 1462 G /usr/lib/xorg/Xorg 72MiB |
+---------------------------------------------------------------------------------------+
(fastvideo2) azureuser@TemporalProcessorVideoCore:~/FastVideo/csrc/sliding_tile_attention$

@TemporalLabsLLC-SOL
Copy link
Author

TemporalLabsLLC-SOL commented Feb 20, 2025

Does it force STA when on a H100? This is very annoying. I've had exactly what I'm trying to do working on an A100 that is dedicated to other tasks. I'm just trying to transfer. This has caused days of delays and I have no idea why it's doing this at this point. Mostly because I can't get the files to download again.

The only thing stopping it from starting is it's missing st_attn module. This happens even when running the normal inference script. That's the frustrating part. I've tried commenting out lines and all sorts of combinations.

I've restarted fresh like 6 times on this H100 machine.

I have the files it is asking for, I don't have them in the right structure it expects.

Long story there.

@jzhang38
Copy link
Collaborator

@TemporalLabsLLC-SOL Sorry for the trouble! Now you should be able to use other features without st_attn

@TemporalLabsLLC-SOL
Copy link
Author

@TemporalLabsLLC-SOL Sorry for the trouble! Now you should be able to use other features without st_attn

I'll spin up another H100 soon and test it out.

Thank you for the update!

# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants