diff options
| author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2022-09-16 14:14:21 +0100 |
|---|---|---|
| committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2022-11-01 09:24:45 +0000 |
| commit | 3394f3e3df7fd2d924c41822a8564493fc06473a (patch) | |
| tree | 8859ab95e39a237b204031a2aa68cde752003dde /src/core/CL/cl_kernels/tile_helpers.h | |
| parent | 910e3f9b686d16657e37d4c18f234b566c8deec2 (diff) | |
| download | ComputeLibrary-3394f3e3df7fd2d924c41822a8564493fc06473a.tar.gz | |
Rework direct convolution heuristic on OpenCL
Resolves COMPMID-5634
Change-Id: I075de70d509d0c4430b4bcf3f218384e237a3a56
Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/453708
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Comments-Addressed: bsgcomp <bsgcomp@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8473
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Diffstat (limited to 'src/core/CL/cl_kernels/tile_helpers.h')
| -rw-r--r-- | src/core/CL/cl_kernels/tile_helpers.h | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/src/core/CL/cl_kernels/tile_helpers.h b/src/core/CL/cl_kernels/tile_helpers.h index 4b6144a22d..6279fb4fb6 100644 --- a/src/core/CL/cl_kernels/tile_helpers.h +++ b/src/core/CL/cl_kernels/tile_helpers.h @@ -653,6 +653,17 @@ }) \ }) +#define T_LOAD2D_INDIRECT(DATA_TYPE, TILE_AREA, TILE_CHANNELS, TENSOR_TYPE, TENSOR, B, Y, X, C, TENSOR_WIDTH, TENSOR_HEIGHT, STRIDE_Y, yi, dst) \ + ({ \ + LOOP_UNROLLING(int, _i, 0, 1, TILE_AREA, \ + { \ + if(yi[_i].v >= 0) \ + { \ + dst[_i].v = V_LOAD(DATA_TYPE, TILE_CHANNELS, TENSOR_TYPE, TENSOR, C, yi[_i].v, STRIDE_Y); \ + } \ + }) \ + }) + /** Load a tile from global memory (tensor) when the tensor is stored using a NDHWC layout using indirect X, Y and Z coordinates * * @param[in] DATA_TYPE Data type |