hexagon: add neg, exp, sigmoid, softplus ops, cont, repeat ops (#20701)

Add element-wise unary ops needed by Qwen 3.5's DeltaNet linear
attention layers. These ops follow the existing unary-ops pattern
with VTCM DMA double-buffering.

- neg: negate via scale by -1.0
- exp: uses existing hvx_exp_f32 HVX intrinsics
- sigmoid: uses existing hvx_sigmoid_f32_aa HVX intrinsics
- softplus: log(1 + exp(x)) scalar fallback
- CONT reuses the existing CPY infrastructure since making a tensor
  contiguous is equivalent to a same-type copy.
- REPEAT implements tiled memory copy with multi-threaded execution via
  the worker pool, supporting f32 and f16 types. The kernel parallelizes
  across output rows and uses memcpy for each tile.

Co-authored-by: Max Krasnyansky <maxk@qti.qualcomm.com>

This commit is contained in:

Krishna Sridhar

2026-03-17 15:34:36 -07:00

committed by

GitHub

co-authored by

Max Krasnyansky

parent 892e3c333a

commit cf23ee2447

11 changed files with 441 additions and 28 deletions

									
										ggml/src/ggml-hexagon/htp/hvx-sigmoid.h
									
		+1
		
												View File
												
				@@ -2,6 +2,7 @@

				#define HVX_SIGMOID_H

				#include "hvx-base.h"

				#include "hvx-inverse.h"

				#define FAST_SIGMOID_LOG2F (0x3fb8aa3b)  // 1.442695022

				#define FAST_SIGMOID_C1    (0x3d009076)  // 0.03138777