Skip to content

Align causal GQA threadgroup memory length to 16 bytes #44

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

Align causal GQA threadgroup memory length to 16 bytes #44

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
4 changes: 3 additions & 1 deletion h3_gpu.m
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4286,7 +4286,9 @@ int h3_gpu_gqa_causal_bf16(h3_gpu *opaque, h3_gpu_tensor *output,
if (getenv("H3_MPS_GQA") && h3_gpu_gqa_mps(
gpu, output, query, key, value, sequence, query_heads,
kv_heads, head_dim, scale)) return 1;
size_t score_bytes = (size_t)sequence * sizeof(float);
/* Metal requires threadgroup memory lengths to be 16-byte multiples;
* the API validation layer aborts on unaligned lengths. */
size_t score_bytes = ((size_t)sequence * sizeof(float) + 15) & ~(size_t)15;
id<MTLComputePipelineState> pipeline = h3_gpu_pipeline(gpu,
@"h3_gqa_causal_bf16");
if (!pipeline) return 0;
Expand Down