mirror of
https://github.com/saymrwulf/onnxruntime.git
synced 2026-07-10 17:37:14 +00:00
891fba3b9c
### Description This PR optimizes the gather op, which is improved ~6ms in segment anything model in ADL. The problem in original algorithm is that it includes a for loop to calculate a block size of data. However, the block size may be very large, like `65536`. In GPU shader, we should try to avoid large loop in shader and try to use more threads to do it parallelly. Before: ``` [profiling] kernel "41771992|[Gather] 41771992" input[0]: [4,65536] | float32, input[1]: [1] | int64, output[0]: [1,65536] | float32, execution time: 6886207 ns ``` After: ``` [profiling] kernel "41771992|[Gather] 41771992" input[0]: [4,65536] | float32, input[1]: [1] | int64, output[0]: [1,65536] | float32, execution time: 11719 ns |
||
|---|---|---|
| .. | ||
| 3rd-party | ||
| argminmax.ts | ||
| binary-op.ts | ||
| common.ts | ||
| concat.ts | ||
| conv-grouped.ts | ||
| conv-transpose.ts | ||
| conv.ts | ||
| conv2d-mm.ts | ||
| einsum.ts | ||
| expand.ts | ||
| fuse-utils.ts | ||
| gather-elements.ts | ||
| gather.ts | ||
| gemm.ts | ||
| instance-norm.ts | ||
| layer-norm.ts | ||
| matmul.ts | ||
| pad.ts | ||
| pool.ts | ||
| reduce.ts | ||
| resize.ts | ||
| skip-layer-norm.ts | ||
| slice.ts | ||
| softmax.ts | ||
| split.ts | ||
| tile.ts | ||
| transpose.ts | ||
| unary-op.ts | ||