| /aosp_15_r20/external/pytorch/aten/src/ATen/native/metal/mpscnn/ |
| H A D | MPSImageUtils.mm | 60 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
83 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
145 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
165 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
197 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
229 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MPSCNNUtils.mm | 22 const auto threadsPerThreadgroup = MTLSizeMake(
27 divRoundUp(width, threadsPerThreadgroup.width),
28 divRoundUp(height, threadsPerThreadgroup.height),
32 return {threadsPerThreadgroup, threadgroupsPerGrid, threadsPerGrid};
|
| H A D | MPSCNNClampOp.mm | 41 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MPSCNNUtils.h | 26 MTLSize threadsPerThreadgroup; member
|
| /aosp_15_r20/external/skia/src/gpu/graphite/mtl/ |
| H A D | MtlComputeCommandEncoder.h | 106 MTLSize threadsPerThreadgroup = in dispatchThreadgroups() local
109 threadsPerThreadgroup:threadsPerThreadgroup]; in dispatchThreadgroups()
115 MTLSize threadsPerThreadgroup = in dispatchThreadgroupsWithIndirectBuffer() local
119 threadsPerThreadgroup:threadsPerThreadgroup]; in dispatchThreadgroupsWithIndirectBuffer()
|
| /aosp_15_r20/external/angle/src/libANGLE/renderer/metal/ |
| H A D | ProvokingVertexHelper.mm | 220 auto threadsPerThreadgroup = MTLSizeMake(MIN(primCount, 64u), 1, 1);
231 MTLSizeMake((primCount + threadsPerThreadgroup.width - 1) / threadsPerThreadgroup.width, 1,
233 threadsPerThreadgroup);
270 auto threadsPerThreadgroup = MTLSizeMake(MIN(primCount, 64u), 1, 1);
280 MTLSizeMake((primCount + threadsPerThreadgroup.width - 1) / threadsPerThreadgroup.width, 1,
282 threadsPerThreadgroup);
|
| H A D | mtl_render_utils.mm | 588 const MTLSize &threadsPerThreadgroup)
592 encoder->dispatchNonUniform(numThreads, threadsPerThreadgroup);
597 (numThreads.width + threadsPerThreadgroup.width - 1) / threadsPerThreadgroup.width,
598 (numThreads.height + threadsPerThreadgroup.height - 1) / threadsPerThreadgroup.height,
599 (numThreads.depth + threadsPerThreadgroup.depth - 1) / threadsPerThreadgroup.depth);
600 encoder->dispatch(groups, threadsPerThreadgroup);
611 MTLSize threadsPerThreadgroup = MTLSizeMake(w, 1, 1);
616 cmdEncoder->dispatchNonUniform(threads, threadsPerThreadgroup);
621 cmdEncoder->dispatch(groups, threadsPerThreadgroup);
1755 MTLSize threadsPerThreadgroup = MTLSizeMake(w, 1, 1);
[all …]
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/metal/ops/ |
| H A D | MetalConcat.mm | 46 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
122 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
155 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalLeakyReLU.mm | 42 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
77 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalHardshrink.mm | 44 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
79 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalHardswish.mm | 40 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
73 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalCopy.mm | 45 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalBinaryElementwise.mm | 93 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
131 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalPadding.mm | 80 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalChunk.mm | 53 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalTranspose.mm | 81 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| H A D | MetalReshape.mm | 57 threadsPerThreadgroup:launchParams.threadsPerThreadgroup];
|
| /aosp_15_r20/external/tensorflow/tensorflow/lite/delegates/gpu/metal/ |
| H A D | compute_task.cc | 289 threadsPerThreadgroup:groupsSize]; in EncodeToICB()
325 [encoder dispatchThreadgroups:groupsCount threadsPerThreadgroup:groupsSize]; in Encode()
|
| H A D | buffer_convert.mm | 120 [encoder dispatchThreadgroups:groups_count threadsPerThreadgroup:group_size];
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/mps/operations/ |
| H A D | MultiTensorApply.h | 299 … [computeEncoder dispatchThreadgroups:gridSize threadsPerThreadgroup:threadGroupSize];
334 [computeEncoder dispatchThreadgroups:gridSize threadsPerThreadgroup:threadGroupSize];
|
| H A D | Quantized.mm | 750 …[computeEncoder dispatchThreads:MTLSizeMake(N, Kdiv2 / 4, 1) threadsPerThreadgroup:MTLSizeMake(64,…
809 …[computeEncoder dispatchThreads:MTLSizeMake(N / 4 * 32, 1, M) threadsPerThreadgroup:MTLSizeMake(64…
867 threadsPerThreadgroup:MTLSizeMake(64, 1, 1)];
871 threadsPerThreadgroup:MTLSizeMake(128, 1, 1)];
|
| /aosp_15_r20/external/pytorch/test/cpp_extensions/ |
| H A D | mps_extension.mm | 66 [computeEncoder dispatchThreads: gridSize threadsPerThreadgroup: threadGroupSize];
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/test/ |
| H A D | mps_test_objc_interface.mm | 76 [computeEncoder dispatchThreads: gridSize threadsPerThreadgroup: threadGroupSize];
|
| /aosp_15_r20/external/tensorflow/tensorflow/lite/delegates/gpu/ |
| H A D | metal_delegate.mm | 133 threadsPerThreadgroup:MTLSizeMake(1, 1, 1)];
555 threadsPerThreadgroup:MTLSizeMake(1, 1, 1)];
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/mps/ |
| H A D | OperationUtils.h | 397 [encoder dispatchThreads:size threadsPerThreadgroup:threadGroupSize]; in mtl_dispatch1DJob()
|