Skip to content

perf(rope): fix token/head-parallel dispatch threshold on CDNA3 #264

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

Merged
demandal25 merged 1 commit into from
Jun 24, 2026
Merged

perf(rope): fix token/head-parallel dispatch threshold on CDNA3 #264

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
23 changes: 12 additions & 11 deletions include/flashinfer/attention/generic/pos_enc.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -606,12 +606,14 @@ gpuError_t BatchQKApplyRotaryPosIdsCosSinCache(
auto kernel_0 = BatchQKApplyRotaryPosIdsCosSinCacheKernel<INTERLEAVE, HEAD_DIM, vec_size, bdx,
DType, IdType>;

int num_blocks_per_sm_0 = 0;
FI_GPU_CALL(gpuOccupancyMaxActiveBlocksPerMultiprocessor(&num_blocks_per_sm_0, kernel_0,
num_threads, /*smem_size=*/0));
uint32_t num_ctas_0 = num_blocks_per_sm_0 * num_sms;

if ((nnz + bdy - 1) / bdy >= num_ctas_0) {
// Pick token-parallel (kernel_0) once it launches enough blocks to fill the
// device. kernel_0 gives each block bdy contiguous tokens across all heads
// (contiguous DRAM), while kernel_1 scatters one head across tokens (strided).
// The token-parallel layout is markedly faster (measured ~1.4x at seq_len 8192
// on gfx942) whenever it has enough blocks; below that it under-fills the GPU
// and the head-parallel form (num_heads x more blocks) wins. Crossover measured
// at nblks_x ~= 2 * num_sms on CDNA3 (independent of head count).
if (nblks_x >= 2u * static_cast<uint32_t>(num_sms)) {
dim3 nblks(nblks_x);
dim3 nthrs(bdx, bdy);
FI_GPU_CALL(gpuLaunchKernel((void*)kernel_0, nblks, nthrs, args, 0, stream));
Expand Down Expand Up @@ -677,11 +679,10 @@ gpuError_t BatchQKApplyRotaryPosIds(
auto kernel_0 =
BatchQKApplyRotaryPosIdsKernel<INTERLEAVE, HEAD_DIM, vec_size, bdx, DType, IdType>;

int num_blocks_per_sm_0 = 0;
FI_GPU_CALL(gpuOccupancyMaxActiveBlocksPerMultiprocessor(&num_blocks_per_sm_0, kernel_0,
num_threads, /*smem_size=*/0));
uint32_t num_ctas_0 = num_blocks_per_sm_0 * num_sms;
if (nblks_x >= num_ctas_0) {
// Same token- vs head-parallel tradeoff as the cos/sin-cache path: token-parallel
// (kernel_0) keeps each block's memory accesses contiguous across heads and wins
// once it has enough blocks to fill the device (~2 * num_sms on CDNA3).
if (nblks_x >= 2u * static_cast<uint32_t>(num_sms)) {
dim3 nblks(nblks_x);
dim3 nthrs(bdx, bdy);

Expand Down
Loading