ggml-org/llama.cpp b11963

on GitHub

sycl: work-group argmax for short TOP_K rows (#31110)

• sycl: work-group argmax for short TOP_K rows

Used angelX to optimize SYCL/drama 3 coordination, results are as follows

Rows with ncols <= max work-group size: one work-item per column, k rounds of work-group argmax. Ties take the lower column, NaN is never picked. Longer rows keep the existing paths.

Arc Pro B70, oneAPI 2026.1, test-backend-ops perf (existing short-row cases), base 9b56ccf vs patch, median ratio of 6 rotating A/B/A rounds:

The 1024 x 8192, k=2 case is 3.7% slower.

Correctness: test-backend-ops -o TOP_K 533/533 and -o ARGSORT 98/98 on the patch, again on the rebase onto 3f03b17.

Full test-backend-ops aborts on base and patch alike at the non-F32 SWIGLU_OAI assert (element_wise.cpp:1187, see #30857). With only those cases excluded: 20145/20173 on base and patch. One CPY f16->q2_0 [256,2,3,4] failure was seen once on the patch, and 3/3 re-runs pass on base and patch. All other patch-run failures also fail on base.

Assisted-by: Claude Opus 5.5

• made changes as suggested

Assisted-by: Claude Opus 5.5