Repository navigation
Optimize FP32 MobileNet on Hexagon HVX by dispatching 7x64__hvx_broadcast for f32_gemm/f32_igemm, adding 1x32/8x32__hvx_broadcast IGEMM to f32_gemm_nr2_config, and using 64c__hvx_acc2 for f32_dwconv. - #11717
Merged
Conversation
…dcast` for `f32_gemm`/`f32_igemm`, adding `1x32`/`8x32__hvx_broadcast` IGEMM to `f32_gemm_nr2_config`, and using `64c__hvx_acc2` for `f32_dwconv`. Benchmark on Qualcomm Snapdragon 8 Elite Gen 5 (Hexagon V81 DSP): ```bash bazel build -c opt --config=hexagon_device_v81 --define=xnn_enable_hvx=true //bench/subgraph:mobilenet ``` | Benchmark | Before (us) | After (us) | Speedup | | :--- | ---: | ---: | ---: | | `FP32MobileNetV1` | 81,547 | 72,922 | 1.12x (-10.6%) | | `FP32MobileNetV2` | 48,450 | 43,851 | 1.10x (-9.5%) | | `FP32MobileNetV3Large` | 34,877 | 32,424 | 1.08x (-7.0%) | | `FP32MobileNetV3Small` | 11,809 | 10,537 | 1.12x (-10.8%) | PiperOrigin-RevId: 1002243336
copybara-service
Bot
force-pushed
the
test_1001390095
branch
from
October 19, 2026 11:31
82aed63 to
878f1d8
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Optimize FP32 MobileNet on Hexagon HVX by dispatching
7x64__hvx_broadcastforf32_gemm/f32_igemm, adding1x32/8x32__hvx_broadcastIGEMM tof32_gemm_nr2_config, and using64c__hvx_acc2forf32_dwconv.Benchmark on Qualcomm Snapdragon 8 Elite Gen 5 (Hexagon V81 DSP):
FP32MobileNetV1FP32MobileNetV2FP32MobileNetV3LargeFP32MobileNetV3Small