llama.cpp
1613ef8d - CUDA: CUDART < 11.7 workaround for hmax, hmax2 (#7019)

Commit

1 year ago

CUDA: CUDART < 11.7 workaround for __hmax, __hmax2 (#7019)

References

#7019 - CUDA: implement __hmax and __hmax2 for CUDA < 11.7

Author

JohannesGaessler

JohannesGaessler

Parents

Loading