Skip to content

add mask to matmul kernel in lecture 14 #52

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.

Sign up for GitHub

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

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

add mask to matmul kernel in lecture 14 #52

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 5 additions & 2 deletions lecture_014/A_Practitioners_Guide_to_Triton.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1221,8 +1221,11 @@
" acc = tl.zeros((bm, bn), dtype=tl.float32)\n",
" for _ in range(0, k, bk):\n",
" # todo umer: don't we need mask when loading a & b?\n",
" a = tl.load(offs_a)\n",
" b = tl.load(offs_b)\n",
" mask_a = get_2d_mask(rm, rn, m, k)
" mask_b = get_2d_mask(rm, rn, k, n)

" a = tl.load(offs_a, mask_a)\n",
" b = tl.load(offs_b, mask_b)\n",
" acc += tl.dot(a, b, allow_tf32=False) # matmul in block ; Weirdness: allow_tf32 must be set to False for older GPUs, otherwise won't compile\n",
" # increase offets, so next iteration loads next chunks\n",
" offs_a += bk * stride_ak\n",
Expand Down