FasterTransformer
FasterTransformer copied to clipboard
NVIDIA
maybe some advices for cmake
1，I think in FindNCCL.cmake, the "set(NCCL_INCLUDE_DIR $ENV{NCCL_INCLUDE_DIR} CACHE ...)" should better be surounded by a "if (DEFINED ENV{...} " thing, to avoid the variable in the cache to be set to "" when the env var is not set. in such cases when afterwards the variable is set, the "null" variable in the cache still works 2，when building a pytorch version ,setting BUILD_GPT=OFF doesn't work, maybe because the gpt.h still has to be compiled. 3, line 63 of fused_multihead_attention_op.cc, The rank of from tensor should be 2, not 3
and it seems some cmake files lack -lmpi_cxx, so there are some compile errors
We will solve the issue 2 and 3 first. And for issue 1, we will take some time to check. Do you encounter any problem due to lacking -lmpi_cxx?
We will solve the issue 2 and 3 first. And for issue 1, we will take some time to check. Do you encounter any problem due to lacking -lmpi_cxx?
yes, I encountered a "undefined reference to ompi_mpi_cxx_op_intercept "etc. , just like: https://stackoverflow.com/questions/17548436/openmpi-error-during-compilation. with -DBUILD_GPT=ON -DBUILD_TF=ON，and set the LD_LIBRARY_PATH, sample/cpp/gpt_sample.cc throws that error
We will solve the issue 2 and 3 first. And for issue 1, we will take some time to check. Do you encounter any problem due to lacking -lmpi_cxx?
yes, I encountered a "undefined reference to ompi_mpi_cxx_op_intercept "etc. , just like: https://stackoverflow.com/questions/17548436/openmpi-error-during-compilation. with -DBUILD_GPT=ON -DBUILD_TF=ON，and set the LD_LIBRARY_PATH, sample/cpp/gpt_sample.cc throws that error
Do you run on different environment? You can propose a pull request to fix the bug since we can reproduce it in the docker we recommend.
We will solve the issue 2 and 3 first. And for issue 1, we will take some time to check. Do you encounter any problem due to lacking -lmpi_cxx?
yes, I encountered a "undefined reference to ompi_mpi_cxx_op_intercept "etc. , just like: https://stackoverflow.com/questions/17548436/openmpi-error-during-compilation. with -DBUILD_GPT=ON -DBUILD_TF=ON，and set the LD_LIBRARY_PATH, sample/cpp/gpt_sample.cc throws that error
Do you run on different environment? You can propose a pull request to fix the bug since we can reproduce it in the docker we recommend.
Yes, I run on a different environment, my mpi is openmpi 1.8.5. Alright, I get it~
and it seems some cmake files lack -lmpi_cxx, so there are some compile errors
Thanks for your advise! I met this error and fixed it by adding -lmpi_cxx -lmpi to target_link_libraries in CMakeLists.txt and sample/cpp/CMakeLists.txt
Close this bug because it is inactivated. Feel free to re-open this issue if you still have any problem.
I made the changes suggested by @zhanzy178 at multiple places and I was able to build without any erros. Thanks @zhanzy178 for the inputs.
