Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Dev][AMD] Implement conditional async load for AMD HIP Backend #250

Merged
merged 5 commits into from
Nov 28, 2024

Conversation

LeiWang1999
Copy link
Contributor

Usage:

  #pragma unroll
  for (int i_3 = 0; i_3 < 2; ++i_3) {
    tl::cp_async_gs_conditional<16>(buf_dyn_shmem+((((i_3 * 4096) + ((((int)threadIdx.x) >> 2) * 64)) + (((((int)threadIdx.x) & 3) ^ (((((int)threadIdx.x) >> 2) / 2) & 3)) * 16)) + 8192), data+(((((((((int)blockIdx.y) * 16384) + (((k_iter + 1) / 12) * 8192)) + (i_3 * 8192)) + ((((int)threadIdx.x) >> 2) * 128)) + (((k_iter + 1) % 12) * 32)) + ((((int)threadIdx.x) & 3) * 8)) - 8320), ((((1 <= ((((((int)blockIdx.y) & 31) * 2) + ((k_iter + 1) / 12)) + i_3)) && (1 <= ((((int)threadIdx.x) >> 2) + (((k_iter + 1) % 12) >> 2)))) && (((((((int)blockIdx.y) & 31) * 2) + ((k_iter + 1) / 12)) + i_3) < 65)) && (((((int)threadIdx.x) >> 2) + (((k_iter + 1) % 12) >> 2)) < 65)));
  }

@LeiWang1999 LeiWang1999 merged commit 645ccd7 into microsoft:main Nov 28, 2024
6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant