Home
last modified time | relevance | path

Searched refs:block_x (Results 1 – 25 of 33) sorted by relevance

12

/external/llvm-project/polly/test/ScopInfo/
Dinvariant_load_complex_condition.ll26 ; CHECK-NEXT: [block_y, block_x] -> { Stmt_entry_split[] };
28 ; CHECK-NEXT: [block_y, block_x] -> { Stmt_entry_split[] -> [] };
30 …block_x] -> { Stmt_entry_split[] -> MemRef4[o0] : (-3 <= block_y < 0 and block_x <= -4 and -8 + bl…
32 ; CHECK-NEXT: [block_y, block_x] -> { Stmt_entry_split[] -> MemRef0[] };
38 define void @dct_luma(i32 %block_x, i32 %block_y) #0 {
43 %div = sdiv i32 %block_x, 4
/external/tensorflow/tensorflow/lite/delegates/gpu/common/tasks/
Dfully_connected.h71 for (int block_x = 0; 4 * block_x < padded_src_channels; block_x++) { in RearrangeFCWeightsToIOO4I4() local
74 int x = 4 * block_x + x_in_block; in RearrangeFCWeightsToIOO4I4()
77 int dst_index = block_x * padded_dst_channels * 4 + block_y * 16 + in RearrangeFCWeightsToIOO4I4()
/external/tensorflow/tensorflow/lite/delegates/gpu/common/tasks/special/
Dfc_fc_add.h49 for (int block_x = 0; 4 * block_x < padded_src_channels; block_x++) { in RearrangeFCWeightsToIOO4I4() local
52 int x = 4 * block_x + x_in_block; in RearrangeFCWeightsToIOO4I4()
53 int dst_index = block_x * padded_dst_channels * 4 + block_y * 16 + in RearrangeFCWeightsToIOO4I4()
/external/llvm-project/mlir/test/Dialect/GPU/
Doutlining.mlir28 threads(%tx, %ty, %tz) in (%block_x = %bDimX, %block_y = %bDimY,
31 "some_op"(%bx, %block_x) : (index, index) -> ()
70 threads(%tx, %ty, %tz) in (%block_x = %cst, %block_y = %cst,
101 threads(%tx, %ty, %tz) in (%block_x = %cst, %block_y = %cst,
125 threads(%tx, %ty, %tz) in (%block_x = %cst, %block_y = %cst,
153 threads(%tx, %ty, %tz) in (%block_x = %cst, %block_y = %cst,
180 threads(%tx, %ty, %tz) in (%block_x = %c1, %block_y = %c1,
206 threads(%tx, %ty, %tz) in (%block_x = %c1, %block_y = %c1,
225 threads(%tx, %ty, %tz) in (%block_x = %cst, %block_y = %cst,
Dmultiple-all-reduce.mlir11 threads(%tx, %ty, %tz) in (%block_x = %c1, %block_y = %c1, %block_z = %c1) {
Dops.mlir9 threads(%tx, %ty, %tz) in (%block_x = %sz, %block_y = %sz, %block_z = %sz) {
20 threads(%tx, %ty, %tz) in (%block_x = %thrd, %block_y = %thrd, %block_z = %thrd) {
/external/pdfium/third_party/libopenjpeg20/
Dsparse_array.c127 OPJ_UINT32 x, block_x; in opj_sparse_array_int32_read_or_write() local
134 block_x = x0 / block_width; in opj_sparse_array_int32_read_or_write()
135 for (x = x0; x < x1; block_x ++, x += x_incr) { in opj_sparse_array_int32_read_or_write()
142 src_block = sa->data_blocks[block_y * sa->block_count_hor + block_x]; in opj_sparse_array_int32_read_or_write()
242 sa->data_blocks[block_y * sa->block_count_hor + block_x] = src_block; in opj_sparse_array_int32_read_or_write()
/external/llvm-project/mlir/test/mlir-cuda-runner/
Dtwo-modules.mlir13 threads(%tx, %ty, %tz) in (%block_x = %sx, %block_y = %one, %block_z = %one) {
19 threads(%tx, %ty, %tz) in (%block_x = %sx, %block_y = %one, %block_z = %one) {
Dshuffle.mlir13 threads(%tx, %ty, %tz) in (%block_x = %sx, %block_y = %one, %block_z = %one) {
16 %width = index_cast %block_x : index to i32
Dall-reduce-op.mlir16 threads(%tx, %ty, %tz) in (%block_x = %sx, %block_y = %sy, %block_z = %sz) {
19 %t2 = muli %t1, %block_x : index
Dgpu-to-cubin.mlir8 threads(%tx, %ty, %tz) in (%block_x = %cst2, %block_y = %cst, %block_z = %cst) {
Dall-reduce-region.mlir13 threads(%tx, %ty, %tz) in (%block_x = %sx, %block_y = %one, %block_z = %one) {
Dall-reduce-max.mlir48 threads(%tx, %ty, %tz) in (%block_x = %c6, %block_y = %c1, %block_z = %c1) {
Dall-reduce-or.mlir48 threads(%tx, %ty, %tz) in (%block_x = %c6, %block_y = %c1, %block_z = %c1) {
Dall-reduce-min.mlir48 threads(%tx, %ty, %tz) in (%block_x = %c6, %block_y = %c1, %block_z = %c1) {
Dall-reduce-xor.mlir48 threads(%tx, %ty, %tz) in (%block_x = %c6, %block_y = %c1, %block_z = %c1) {
Dall-reduce-and.mlir48 threads(%tx, %ty, %tz) in (%block_x = %c6, %block_y = %c1, %block_z = %c1) {
Dmultiple-all-reduce.mlir51 threads(%tx, %ty, %tz) in (%block_x = %c6, %block_y = %c1, %block_z = %c1) {
/external/llvm-project/mlir/test/mlir-rocm-runner/
Dtwo-modules.mlir14 threads(%tx, %ty, %tz) in (%block_x = %sx, %block_y = %c1, %block_z = %c1) {
20 threads(%tx, %ty, %tz) in (%block_x = %sx, %block_y = %c1, %block_z = %c1) {
Dvector-transferops.mlir6 threads(%tx, %ty, %tz) in (%block_x = %cst, %block_y = %cst, %block_z = %cst) {
28 threads(%tx, %ty, %tz) in (%block_x = %cst, %block_y = %cst, %block_z = %cst) {
Dgpu-to-hsaco.mlir8 threads(%tx, %ty, %tz) in (%block_x = %block_dim, %block_y = %c1, %block_z = %c1) {
Dvecadd.mlir8 threads(%tx, %ty, %tz) in (%block_x = %block_dim, %block_y = %c1, %block_z = %c1) {
/external/tensorflow/tensorflow/core/profiler/internal/gpu/
Dcupti_collector.h85 uint32 block_x; member
105 kernel_info.grid_z, " block:", kernel_info.block_x, ",", in ToXStat()
/external/tensorflow/tensorflow/core/profiler/utils/
Dkernel_stats_utils_test.cc75 kernel_info.block_x = 32; in TEST()
/external/llvm/test/CodeGen/ARM/
D2007-01-19-InfiniteLoop.ll12 define fastcc i32 @dct_luma_sp(i32 %block_x, i32 %block_y, i32* %coeff_cost) {

12