search

daadaada / turingas

Assembler for NVIDIA Volta and Turing GPUs
MIT License
196 stars 41 forks source link

Is Winograd code available? #2

Open Jokeren opened 4 years ago

Jokeren commented 4 years ago

I just attended your presentation. Is there anywhere I can find the Winograd code as described in your paper? Much appreciate it.

daadaada commented 4 years ago

I just attended your presentation. Is there anywhere I can find the Winograd code as described in your paper? Much appreciate it.

Thank you for listening!

We plan to open source the Winograd implementation later.

Before that, we will add more examples including saxpy and GEMM.

jesselu-hp commented 4 years ago

Hi! Any update on the saxpy and GEMM examples? Turingas looks super useful, thank you!

Elon-Wang commented 3 years ago

Hi, I have read your paper and very interested in the program. Is there any update of the CUDA program of the Winograd Algorithm? I've also read the code of Xuqiantong's CUDA-Winograd repositories (https://github.com/daadaada/CUDA-Winograd) and think the implement of the matrix multiplication need more optimization. I'd like to know how you implement this algorithm, thx!

Shaquille-Wu commented 1 year ago

same question, where to find the code which the paper described

hgl71964 commented 9 months ago

hey,

Is it possible to see the example where you reorder the sass instruction for better latency hiding? Many thanks

Best,