We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Hello,
We are trying to install and run this package, but kept getting low level errors that result in e.g.:
sh: line 1: 3218775 Killed cicc --c++17 --gnu_version=100400 --display_error_number --orig_src_file_name "/home/ydong/Block-Sparse-Attention/csrc/block_sparse_attn/src/flash_fwd_hdim32_fp16_sm80.cu" --orig_src_path_name "/home/ydong/Block-Sparse-Attention/csrc/block_sparse_attn/src/flash_fwd_hdim32_fp16_sm80.cu" --allow_managed --extended-lambda --relaxed_constexpr -arch compute_80 -m64 --no-version-ident -ftz=1 -prec_div=0 -prec_sqrt=0 -fmad=1 -fast-math --gen_div_approx_ftz --include_file_name "tmpxft_00311817_00000000-3_flash_fwd_hdim32_fp16_sm80.fatbin.c" -generate-line-info -tused --module_id_file_name "/tmp/tmpxft_00311817_00000000-4_flash_fwd_hdim32_fp16_sm80.module_id" --gen_c_file_name "/tmp/tmpxft_00311817_00000000-7_flash_fwd_hdim32_fp16_sm80.compute_80.cudafe1.c" --stub_file_name "/tmp/tmpxft_00311817_00000000-7_flash_fwd_hdim32_fp16_sm80.compute_80.cudafe1.stub.c" --gen_device_file_name "/tmp/tmpxft_00311817_00000000-7_flash_fwd_hdim32_fp16_sm80.compute_80.cudafe1.gpu" "/tmp/tmpxft_00311817_00000000-8_flash_fwd_hdim32_fp16_sm80.compute_80.cpp1.ii" -o "/tmp/tmpxft_00311817_00000000-7_flash_fwd_hdim32_fp16_sm80.compute_80.ptx" > /tmp/tmpxft_00311817_00000000-13_2c22450_stdout 2> /tmp/tmpxft_00311817_00000000-13_2c22450_stderr
We followed all instructions on the repo README. This is with gcc version 12.4 and CUDA version 12.6 on a Hopper GPU.
Do you know what could be causing these errors?
Thanks, Yihe
The text was updated successfully, but these errors were encountered:
No branches or pull requests
Hello,
We are trying to install and run this package, but kept getting low level errors that result in e.g.:
sh: line 1: 3218775 Killed cicc --c++17 --gnu_version=100400 --display_error_number --orig_src_file_name "/home/ydong/Block-Sparse-Attention/csrc/block_sparse_attn/src/flash_fwd_hdim32_fp16_sm80.cu" --orig_src_path_name "/home/ydong/Block-Sparse-Attention/csrc/block_sparse_attn/src/flash_fwd_hdim32_fp16_sm80.cu" --allow_managed --extended-lambda --relaxed_constexpr -arch compute_80 -m64 --no-version-ident -ftz=1 -prec_div=0 -prec_sqrt=0 -fmad=1 -fast-math --gen_div_approx_ftz --include_file_name "tmpxft_00311817_00000000-3_flash_fwd_hdim32_fp16_sm80.fatbin.c" -generate-line-info -tused --module_id_file_name "/tmp/tmpxft_00311817_00000000-4_flash_fwd_hdim32_fp16_sm80.module_id" --gen_c_file_name "/tmp/tmpxft_00311817_00000000-7_flash_fwd_hdim32_fp16_sm80.compute_80.cudafe1.c" --stub_file_name "/tmp/tmpxft_00311817_00000000-7_flash_fwd_hdim32_fp16_sm80.compute_80.cudafe1.stub.c" --gen_device_file_name "/tmp/tmpxft_00311817_00000000-7_flash_fwd_hdim32_fp16_sm80.compute_80.cudafe1.gpu" "/tmp/tmpxft_00311817_00000000-8_flash_fwd_hdim32_fp16_sm80.compute_80.cpp1.ii" -o "/tmp/tmpxft_00311817_00000000-7_flash_fwd_hdim32_fp16_sm80.compute_80.ptx" > /tmp/tmpxft_00311817_00000000-13_2c22450_stdout 2> /tmp/tmpxft_00311817_00000000-13_2c22450_stderr
We followed all instructions on the repo README. This is with gcc version 12.4 and CUDA version 12.6 on a Hopper GPU.
Do you know what could be causing these errors?
Thanks,
Yihe
The text was updated successfully, but these errors were encountered: