| /aosp_15_r20/external/pytorch/aten/src/ATen/native/transformers/cuda/mem_eff_attention/iterators/ |
| H A D | epilogue_predicated_tile_iterator.h | 77 using ThreadMap = ThreadMap_;
78 using Shape = typename ThreadMap::Shape;
90 static int const kElementsPerAccess = ThreadMap::kElementsPerAccess;
91 static int const kThreads = ThreadMap::kThreads;
92 static int const kIterations = ThreadMap::Count::kTile;
95 ThreadMap::Iterations::kRow > 0,
98 ThreadMap::Iterations::kGroup > 0,
101 ThreadMap::Iterations::kCluster > 0,
104 ThreadMap::Iterations::kColumn > 0,
110 ThreadMap::Iterations::kColumn * ThreadMap::Iterations::kRow *
[all …]
|
| H A D | predicated_tile_iterator_residual_last.h | 148 typename ThreadMap,
149 int AccessSize = ThreadMap::kElementsPerAccess,
187 using ThreadMap = ThreadMap_; variable
211 ThreadMap,
220 ThreadMap::Iterations::kCount * ThreadMap::kElementsPerAccess>;
376 for (int s = 0; s < ThreadMap::Iterations::kStrided; ++s) { in load_with_byte_offset()
378 for (int c = 0; c < ThreadMap::Iterations::kContiguous; ++c) { in load_with_byte_offset()
382 kAccessesPerVector * (c + s * ThreadMap::Iterations::kContiguous); in load_with_byte_offset()
421 for (int s = 0; s < ThreadMap::Iterations::kStrided; ++s) { in store_with_byte_offset()
423 for (int c = 0; c < ThreadMap::Iterations::kContiguous; ++c) { in store_with_byte_offset()
[all …]
|
| H A D | predicated_tile_access_iterator_residual_last.h | 76 typename ThreadMap,
111 using ThreadMap = ThreadMap_; variable
129 ThreadMap,
133 ThreadMap::kElementsPerAccess / AccessType::kElements;
136 !(ThreadMap::kElementsPerAccess % AccessType::kElements),
159 ThreadMap>()()) {} in Params()
326 (ThreadMap::Delta::kContiguous * sizeof_bits<Element>::value / in get()
330 the_predicates.iteration_strided_ * ThreadMap::Delta::kStrided; in get()
342 (ThreadMap::Delta::kContiguous * in get()
362 ThreadMap::Iterations::kContiguous) {
[all …]
|
| H A D | make_residual_last.h | 26 typename ThreadMap,
34 ThreadMap,
42 ThreadMap,
52 typename ThreadMap,
60 ThreadMap,
68 ThreadMap,
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/transformers/cuda/mem_eff_attention/epilogue/ |
| H A D | epilogue_pipelined.h | 594 using ThreadMap = typename OutputTileIterator::ThreadMap;
597 for (int cluster = 0; cluster < ThreadMap::Iterations::kCluster;
600 for (int group = 0; group < ThreadMap::Iterations::kGroup; ++group) {
602 for (int row = 0; row < ThreadMap::Iterations::kRow; ++row) {
603 int row_offset = row * ThreadMap::Delta::kRow +
604 group * ThreadMap::Delta::kGroup +
605 cluster * ThreadMap::Delta::kCluster;
608 ThreadMap::Iterations::kRow *
609 (group + ThreadMap::Iterations::kGroup * cluster));
611 for (int column = 0; column < ThreadMap::Iterations::kColumn;
[all …]
|
| /aosp_15_r20/external/webrtc/modules/remote_bitrate_estimator/test/ |
| H A D | bwe_test_logging.cc | 85 ThreadMap::iterator it = thread_map_.find(rtc::CurrentThreadId()); in Log()
122 ThreadMap::iterator it = thread_map_.find(rtc::CurrentThreadId()); in Plot()
136 ThreadMap::iterator it = thread_map_.find(rtc::CurrentThreadId()); in PlotBar()
149 ThreadMap::iterator it = thread_map_.find(rtc::CurrentThreadId()); in PlotBaselineBar()
165 ThreadMap::iterator it = thread_map_.find(rtc::CurrentThreadId()); in PlotErrorBar()
184 ThreadMap::iterator it = thread_map_.find(rtc::CurrentThreadId()); in PlotLimitErrorBar()
199 ThreadMap::iterator it = thread_map_.find(rtc::CurrentThreadId()); in PlotLabel()
246 ThreadMap::iterator it = thread_map_.find(rtc::CurrentThreadId()); in PopState()
|
| H A D | bwe_test_logging.h | 339 typedef std::map<uint32_t, ThreadState> ThreadMap; typedef
353 ThreadMap thread_map_;
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/transformers/cuda/mem_eff_attention/transform/ |
| H A D | tile_smem_loader.h | 30 using ThreadMap = cutlass::transform::PitchLinearStripminedThreadMap< variable
43 ThreadMap>; // ThreadMap
50 ThreadMap>; // ThreadMap
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/cuda/cutlass_extensions/gemm/threadblock/ |
| H A D | dq_mma_multistage.h | 161 static int const AsyncCopyIterationsPerStageA = IteratorA::ThreadMap::Iterations::kCount;
164 static int const AsyncCopyIterationsPerStageB = IteratorB::ThreadMap::Iterations::kCount;
257 … * IteratorA::ThreadMap::kElementsPerAccess / IteratorA::kAccessesPerVector / 8;
288 … * IteratorB::ThreadMap::kElementsPerAccess / IteratorB::kAccessesPerVector / 8;
359 … * IteratorA::ThreadMap::kElementsPerAccess / IteratorA::kAccessesPerVector in operator()
385 … * IteratorB::ThreadMap::kElementsPerAccess / IteratorB::kAccessesPerVector in operator()
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/transformers/cuda/mem_eff_attention/gemm/ |
| H A D | custom_mma_multistage.h | 146 IteratorA::ThreadMap::Iterations::kCount;
150 IteratorB::ThreadMap::Iterations::kCount;
310 IteratorA::ThreadMap::kElementsPerAccess /
346 IteratorB::ThreadMap::kElementsPerAccess /
397 IteratorA::ThreadMap::kElementsPerAccess / in _prologue()
427 IteratorB::ThreadMap::kElementsPerAccess / in _prologue()
|
| H A D | mma_from_smem.h | 795 IteratorB1::ThreadMap::Iterations::kCount;
945 IteratorB1::ThreadMap::kElementsPerAccess /
988 IteratorB1::ThreadMap::kElementsPerAccess / in _prologue()
|
| /aosp_15_r20/external/perfetto/ui/src/plugins/dev.perfetto.Thread/ |
| H A D | index.ts | 17 import {ThreadDesc, ThreadMap} from '../dev.perfetto.Thread/threads';
58 private threads?: ThreadMap;
|
| H A D | threads.ts | 24 export type ThreadMap = ReadonlyMap<number, ThreadDesc>; alias
|
| /aosp_15_r20/external/perfetto/ui/src/plugins/org.kernel.SuspendResumeLatency/ |
| H A D | index.ts | 24 import {ThreadMap} from '../dev.perfetto.Thread/threads';
37 private readonly threads: ThreadMap,
|
| H A D | suspend_resume_details.ts | 29 import {ThreadMap} from '../dev.perfetto.Thread/threads';
48 private readonly threads: ThreadMap,
|
| /aosp_15_r20/external/perfetto/ui/src/plugins/dev.perfetto.CpuSlices/ |
| H A D | sched_details_tab.ts | 36 import {ThreadDesc, ThreadMap} from '../dev.perfetto.Thread/threads';
62 private readonly threads: ThreadMap,
|
| H A D | cpu_slice_track.ts | 41 import {ThreadMap} from '../dev.perfetto.Thread/threads';
73 private readonly threads: ThreadMap,
|
| /aosp_15_r20/external/perfetto/ui/src/plugins/dev.perfetto.ProcessSummary/ |
| H A D | process_scheduling_track.ts | 31 import {ThreadMap} from '../dev.perfetto.Thread/threads';
66 private readonly threads: ThreadMap,
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/transformers/cuda/mem_eff_attention/ |
| H A D | kernel_forward.h | 501 typename DefaultEpilogue::OutputTileIterator::ThreadMap,
505 typename DefaultEpilogue::OutputTileIterator::ThreadMap,
|
| /aosp_15_r20/out/soong/.intermediates/prebuilts/sdk/current/androidx/m2repository/androidx/compose/runtime/runtime-android/1.8.0-beta01/androidx.compose.runtime_runtime/android_common_apex30/aar/ |
| D | baseline-prof.txt | 101 HSPLandroidx/compose/runtime/internal/ThreadMap;->**(**)**
|
| D | classes.jar | META-INF/
androidx/
androidx/compose/
androidx/compose/runtime/
androidx/compose ... |
| /aosp_15_r20/out/soong/.intermediates/prebuilts/sdk/current/androidx/m2repository/androidx/compose/runtime/runtime-android/1.8.0-beta01/androidx.compose.runtime_runtime/android_common/aar/ |
| D | baseline-prof.txt | 101 HSPLandroidx/compose/runtime/internal/ThreadMap;->**(**)**
|
| D | androidx.compose.runtime_runtime.jar | META-INF/
androidx/
androidx/compose/
androidx/compose/runtime/
androidx/compose ... |
| /aosp_15_r20/out/soong/.intermediates/prebuilts/sdk/current/androidx/m2repository/androidx/compose/runtime/runtime-android/1.8.0-beta01/androidx.compose.runtime_runtime/android_common_apex34/aar/ |
| D | baseline-prof.txt | 101 HSPLandroidx/compose/runtime/internal/ThreadMap;->**(**)**
|
| D | classes.jar | META-INF/
androidx/
androidx/compose/
androidx/compose/runtime/
androidx/compose ... |