Saad Rahim (AMD)
Saad Rahim (AMD)
@ex-rzr Can you point out what line in Thrust has __host__ __device__ with a kernel launch?
@feizheng10 @malcolmroberts Can I get a review for this PR?
@evetsso and @cgmb can you please take a look at this again?
@malcolmroberts shall I merge the PR into this repository?
@gargrahul Can you please share this feedback with the windows team?
@MathiasMagnus please update the instructions?
Have you seen our BLAS libraries? https://rocm.docs.amd.com/en/latest/reference/gpu_libraries/linear_algebra.html They are already tuned for many GPUs.
Please take a look at the redone HIP documentation. https://rocm.docs.amd.com/projects/HIP/en/latest/index.html
> I have the feeling there is still the point "6 Document assumes familiarity with CUDA, but why assume that?" which could be explained. Feedback accepted. I have added it...
@baryluk Thank you for your thorough feedback. I will prioritize this ticket and #1760. HIP documentation will be completely rewritten.