I am planning to add CUDA compilation support in LibAV with a CUDA sample. The challenge here is that NVCC Cuda Compiler supports GCC compiler under Linux and MSVC compiler on Windows. So GCC compilation under Windows becomes an issue.
The Clang compiler is still in infancy stage and doesn't support the latest SDK and architectures. I am planning to have the CUDA files compiled for the Linux/GCC and Windows/MSVC (MingW) platforms, and for Windows/GCC (MingW) we can store the PTX files that have been generated on Linux or Windows using MSVC compiler. We can improve this on a later date once we have more support on MinGW/GCC either through nvcc or Clang. Let me know if there are more ideas to explore. Thanks, Yogender ----------------------------------------------------------------------------------- This email message is for the sole use of the intended recipient(s) and may contain confidential information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. ----------------------------------------------------------------------------------- _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
