Home
last modified time | relevance | path

Searched refs:compute (Results 1 – 25 of 125) sorted by relevance

12345

/hardware/interfaces/neuralnetworks/utils/common/test/
DResilientExecution.cpp100 TEST(ResilientExecutionTest, compute) { in TEST() argument
103 EXPECT_CALL(*mockExecution, compute(_)).Times(1).WillOnce(Return(kNoExecutionError)); in TEST()
106 const auto result = execution->compute({}); in TEST()
116 EXPECT_CALL(*mockExecution, compute(_)).Times(1).WillOnce(kReturnGeneralFailure); in TEST()
119 const auto result = execution->compute({}); in TEST()
129 EXPECT_CALL(*mockExecution, compute(_)).Times(1).WillOnce(kReturnDeadObject); in TEST()
133 const auto result = execution->compute({}); in TEST()
143 EXPECT_CALL(*mockExecution, compute(_)).Times(1).WillOnce(kReturnDeadObject); in TEST()
145 EXPECT_CALL(*recoveredMockExecution, compute(_)).Times(1).WillOnce(Return(kNoExecutionError)); in TEST()
149 const auto result = execution->compute({}); in TEST()
DMockExecution.h28 MOCK_METHOD((ExecutionResult<std::pair<std::vector<OutputShape>, Timing>>), compute,
/hardware/google/gfxstream/codegen/vulkan/vulkan-docs-next/appendices/
DVK_NV_device_generated_commands_compute.adoc23 This extension allows the device to generate commands for binding compute
24 pipelines, setting push constants and launching compute dispatches.
DVK_NV_compute_shader_derivatives.adoc28 compute shaders to use built-ins that evaluate compute derivatives
51 derivatives in a compute shader are the same groups of four invocations that
DVK_AMDX_shader_enqueue.adoc36 This extension adds the ability for developers to enqueue compute shader
37 workgroups from other compute shaders.
DVK_EXT_shader_subgroup_vote.adoc34 to compute the composite of a set of boolean conditions across a group of
50 In particular, a subgroup has no necessary relationship with the compute
51 shader _local workgroup_ -- any pair of shader invocations in a compute
56 workgroup), but many implementations will also group non-compute shader
DVK_HUAWEI_cluster_culling_shader.adoc29 Cluster Culling Shaders (CCS) are similar to the existing compute shaders.
34 The traditional 2-pass GPU culling solution using a compute shader sometimes
35 needs a pipeline barrier between compute and graphics pipeline to optimize
38 This extension addresses these shortcomings, allowing compute shaders to
51 directly outputting transformed vertices from the compute shader.
/hardware/google/gfxstream/codegen/vulkan/vulkan-docs-next/chapters/
Ddispatch.adoc9 work in a compute pipeline.
11 a queue, will produce work which executes according to the bound compute
13 A compute pipeline must: be bound to a command buffer before any dispatching
16 [open,refpage='vkCmdDispatch',desc='Dispatch compute work items',type='protos']
55 [open,refpage='vkCmdDispatchIndirect',desc='Dispatch compute work items with indirect parameters',t…
120 [open,refpage='vkCmdDispatchBase',desc='Dispatch compute work items with non-zero base values for t…
187 not zero, then the bound compute pipeline must: have been created with
196 [open,refpage='vkCmdSubpassShadingHUAWEI',desc='Dispatch compute work items',type='protos']
200 A subpass shading dispatches a compute pipeline work with the work dimension
/hardware/interfaces/neuralnetworks/1.2/utils/test/
DPreparedModelTest.cpp357 const auto computeResult = createResult.value()->compute({}); in TEST()
380 const auto computeResult = createResult.value()->compute({}); in TEST()
401 const auto computeResult = createResult.value()->compute({}); in TEST()
422 const auto computeResult = createResult.value()->compute({}); in TEST()
446 const auto computeResult = createResult.value()->compute({}); in TEST()
470 const auto computeResult = createResult.value()->compute({}); in TEST()
492 const auto computeResult = createResult.value()->compute({}); in TEST()
513 const auto computeResult = createResult.value()->compute({}); in TEST()
534 const auto computeResult = createResult.value()->compute({}); in TEST()
557 const auto computeResult = createResult.value()->compute({}); in TEST()
/hardware/interfaces/neuralnetworks/aidl/utils/test/
DExecutionTest.cpp92 const auto result = execution->compute({}); in TEST()
108 const auto result = execution->compute({}); in TEST()
124 const auto result = execution->compute({}); in TEST()
140 const auto result = execution->compute({}); in TEST()
/hardware/interfaces/neuralnetworks/1.0/utils/test/
DPreparedModelTest.cpp245 const auto computeResult = createResult.value()->compute({}); in TEST()
267 const auto computeResult = createResult.value()->compute({}); in TEST()
288 const auto computeResult = createResult.value()->compute({}); in TEST()
308 const auto computeResult = createResult.value()->compute({}); in TEST()
328 const auto computeResult = createResult.value()->compute({}); in TEST()
350 const auto computeResult = createResult.value()->compute({}); in TEST()
/hardware/interfaces/neuralnetworks/aidl/utils/include/nnapi/hal/aidl/
DExecution.h55 nn::ExecutionResult<std::pair<std::vector<nn::OutputShape>, nn::Timing>> compute(
82 nn::ExecutionResult<std::pair<std::vector<nn::OutputShape>, nn::Timing>> compute(
/hardware/interfaces/neuralnetworks/utils/common/src/
DResilientExecution.cpp104 ResilientExecution::compute(const nn::OptionalTimePoint& deadline) const { in compute() function in android::hardware::neuralnetworks::utils::ResilientExecution
106 return execution.compute(deadline); in compute()
DInvalidExecution.cpp28 nn::ExecutionResult<std::pair<std::vector<nn::OutputShape>, nn::Timing>> InvalidExecution::compute( in compute() function in android::hardware::neuralnetworks::utils::InvalidExecution
/hardware/google/gfxstream/codegen/vulkan/vulkan-docs-next/chapters/VK_NV_device_generated_commands/
Dgeneratedcommands.adoc18 * Retrieve device addresses of compute pipelines by
32 pipeline from either graphics or compute.
/hardware/qcom/display/msm8960/liboverlay/
DoverlayUtils.cpp223 static inline int compute(const uint32_t& x, const uint32_t& y, in compute() function
233 srcCrop.x = compute(whf.w, srcCrop.x, srcCrop.w); in preRotateSource()
236 srcCrop.y = compute(whf.h, srcCrop.y, srcCrop.h); in preRotateSource()
240 srcCrop.x = compute(whf.h, in preRotateSource()
/hardware/qcom/display/msm8226/liboverlay/
DoverlayUtils.cpp314 static inline int compute(const uint32_t& x, const uint32_t& y, in compute() function
321 srcCrop.x = compute(whf.w, srcCrop.x, srcCrop.w); in preRotateSource()
324 srcCrop.y = compute(whf.h, srcCrop.y, srcCrop.h); in preRotateSource()
328 srcCrop.x = compute(whf.h, in preRotateSource()
/hardware/qcom/display/msm8994/liboverlay/
DoverlayUtils.cpp307 static inline int compute(const uint32_t& x, const uint32_t& y, in compute() function
314 srcCrop.x = compute(whf.w, srcCrop.x, srcCrop.w); in preRotateSource()
317 srcCrop.y = compute(whf.h, srcCrop.y, srcCrop.h); in preRotateSource()
321 srcCrop.x = compute(whf.h, in preRotateSource()
/hardware/interfaces/neuralnetworks/utils/common/include/nnapi/hal/
DInvalidExecution.h31 nn::ExecutionResult<std::pair<std::vector<nn::OutputShape>, nn::Timing>> compute(
/hardware/qcom/display/msm8084/liboverlay/
DoverlayUtils.cpp303 static inline int compute(const uint32_t& x, const uint32_t& y, in compute() function
310 srcCrop.x = compute(whf.w, srcCrop.x, srcCrop.w); in preRotateSource()
313 srcCrop.y = compute(whf.h, srcCrop.y, srcCrop.h); in preRotateSource()
317 srcCrop.x = compute(whf.h, in preRotateSource()
/hardware/google/gfxstream/codegen/vulkan/vulkan-docs-next/proposals/
DVK_HUAWEI_cluster_culling_shader.adoc15 …g many clusters than the CPU. Many developers use compute shader for GPU culling tasks. Because co…
18 …on to connect the output of the compute shader to the existing rednering pipeline, developers who …
24 …culling on GPU. This mechanism does not require pipeline barrier between compute shader and other …
26 This new shader type have execution environments similar to that of compute shaders, where a collec…
DVK_EXT_mesh_shader.adoc17 …eneck; many developers are augmenting their rasterization pipelines with compute shaders to make t…
18 Using compute shaders comes at a cost though - data has to be piped via global memory, and once the…
19 …compute shaders processing geometry, the role of the vertex shader is also somewhat redundant - tr…
26 …the set of geometry as a whole in the way developers are currently using compute shaders, and how …
30 . Use compute shaders in place of pre-rasterization shader stages
39 …o effectively use compute shaders in place of existing rasterization stages, enabling applications…
40 …efining the interface between the shader and the rasterizer, as existing compute shaders simply wr…
42 …n shaders in the graphics pipeline with two new stages that operate like compute shaders but with …
54 Mesh shaders are a new compute-like shader stage that has a primary aim of generating a set of prim…
55 For the most part, these map well to compute shaders - they are dispatched in workgroups and have a…
[all …]
/hardware/interfaces/neuralnetworks/1.0/utils/src/
DExecution.cpp60 nn::ExecutionResult<std::pair<std::vector<nn::OutputShape>, nn::Timing>> Execution::compute( in compute() function in android::hardware::neuralnetworks::V1_0::utils::Execution
/hardware/interfaces/neuralnetworks/1.0/utils/include/nnapi/hal/1.0/
DExecution.h49 nn::ExecutionResult<std::pair<std::vector<nn::OutputShape>, nn::Timing>> compute(
/hardware/interfaces/neuralnetworks/1.2/utils/include/nnapi/hal/1.2/
DExecution.h50 nn::ExecutionResult<std::pair<std::vector<nn::OutputShape>, nn::Timing>> compute(

12345