Home
last modified time | relevance | path

Searched defs:lane_offset (Results 1 – 7 of 7) sorted by relevance

/external/pytorch/aten/src/ATen/native/transformers/cuda/mem_eff_attention/
Ddebug_utils.h181 LaneOffsetT lane_offset, in print_warp_accum()
Dkernel_forward.h823 auto lane_offset = MM0::AccumLambdaIterator::get_lane_offset( in attention_kernel() local
847 auto lane_offset = MM0::AccumLambdaIterator::get_lane_offset( in attention_kernel() local
882 auto lane_offset = MM0::AccumLambdaIterator::get_lane_offset( in attention_kernel() local
1223 auto lane_offset = in iterative_softmax() local
Dkernel_backward.h1617 auto lane_offset = MatmulQK::AccumLambdaIterator::get_lane_offset( in processBlockIJ() local
1632 auto lane_offset = MatmulQK::AccumLambdaIterator::get_lane_offset( in processBlockIJ() local
1653 auto lane_offset = MatmulQK::AccumLambdaIterator::get_lane_offset( in processBlockIJ() local
1759 auto lane_offset = MatmulDOIVJ::AccumLambdaIterator::get_lane_offset( in processBlockIJ() local
1916 auto lane_offset = LambdaIterator::get_lane_offset( in processBlockIJ() local
/external/pytorch/aten/src/ATen/native/transformers/cuda/mem_eff_attention/gemm/
Dmma_accum_lambda_iterator.h274 cutlass::MatrixCoord lane_offset = lane_layout.inverse(lane_id) * in get_lane_offset() local
/external/mesa3d/src/intel/compiler/
Dbrw_nir_lower_ray_queries.c154 nir_def *lane_offset = in get_ray_query_shadow_addr() local
/external/tensorflow/tensorflow/core/util/
Dgpu_device_functions.h243 int lane_offset = src_lane & width - 1; in GpuShuffleGetSrcLane() local
/external/coreboot/src/northbridge/intel/gm45/
Draminit_read_write_training.c93 const int lane_offset = lane & 4; in read_training_test() local