hexagon: add support for CONCAT op (#23648)
* hexagon: add support for CONCAT with optimized concat_2d_transposed qwen3.5 models are quite heavy on the CONCAT with large and transposed src1. * hex-concat: use fastdiv in generic version * hex-concat: make checks for transposed a bit more readable * hex-concat: reoder dma ops for better pipelining * hex-cont/cpy: optimize CPY and CONT ops The primary change is to avoid scalar divs in the inner loops. We were calling hvx_copy_uu(... type_size) where type_size is non a constexpr. This causes runtime divs by that value which is normally just 4 or 2 (f32/f16). * hex-get-rows: optimize GET_ROWS for large rows We now use DMA for larger rows and also split them into chunks to improve perf for Qwen3.5 and other models that do lots of GET_ROWS with huge (2MB+ rows). Also bump the DMA queue depth now that we can take advantage of it. * hex-concat: unroll the inner loops of concat_2d * hex-concat: more updates to concat_2d to improve perf a bit further * hex-cpy: fixed n_rows per thread checks in the copy ops * hmx-fa: fix alignment issues while computing dma sizes * hex-set-rows: add early returns for idle threads * hvx-rope: minor optimization to replace loops with fastdiv logic * hex-rope: replace scalar tail processing with HVX * hex-rope: optimize rope cache init with HVX Add hvx-utils sin/cos helpers that use an aprox method (similar to rsqrt, inverse, etc) Use the helpers to optimize ROPE.
This commit is contained in:
@@ -104,6 +104,7 @@ int op_argsort(struct htp_ops_context * octx);
|
||||
int op_ssm_conv(struct htp_ops_context * octx);
|
||||
int op_cumsum(struct htp_ops_context * octx);
|
||||
int op_fill(struct htp_ops_context * octx);
|
||||
int op_concat(struct htp_ops_context * octx);
|
||||
int op_diag(struct htp_ops_context * octx);
|
||||
int op_solve_tri(struct htp_ops_context * octx);
|
||||
int op_gated_delta_net(struct htp_ops_context * octx);
|
||||
|
||||
Reference in New Issue
Block a user