Repository navigation
Conversation
veblush
force-pushed
the
fix/convolve-1xn-padding-and-s4-odd-ch
branch
from
October 29, 2026 20:48
d44b2d0 to
e730657
Compare
mansnils
requested changes
Nov 2, 2026
| quant_params, | ||
| input_dims, | ||
| input_data, | ||
| filter_dims, |
Contributor
There was a problem hiding this comment.
Can the guard be kept here and add the routing logic to the wrapper instead?
Contributor
Author
There was a problem hiding this comment.
Done. Kept the guards in arm_convolve_1_x_n_s4/s8 and moved the checks to the wrappers and buffer-size helpers.
| const int32_t bias_data[2] = {5, -4}; | ||
| const int32_t output_mult[2] = {0x40000000, 0x40000000}; | ||
| const int32_t output_shift[2] = {1, 1}; | ||
| const int8_t kernel_data[40] = {1, -1, 2, 1, -2, 1, 1, -1, 2, -1, 1, 2, 2, 3, 1, 2, -1, 2, -2, 1, |
Contributor
There was a problem hiding this comment.
Can you make datasets reproducible through existing generators and TFL/TFLM reference kernels following unit test conventions (Tests/UnitTest/README.md:147)?
Same in other unit test.
Contributor
Author
There was a problem hiding this comment.
Done. Replaced the hand-written test arrays in both unit tests with datasets from json.
veblush
force-pushed
the
fix/convolve-1xn-padding-and-s4-odd-ch
branch
from
November 3, 2026 00:27
e730657 to
d5b14e5
Compare
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fix several edge cases in
arm_convolve_1_x_n_s8,arm_convolve_1_x_n_s4, and their MVE buffer-size functions:total_padto>= 0when trailing input columns are unused (e.g. stridedVALIDconvolution).s8, computepad_size_rightfrom(pad_x + asym_pad)instead ofright_pad_numand clampnum_elem_left/num_elem_rightto the available input width soarm_memcpy_s8does not read pastinput_data.arm_convolve_s8/arm_convolve_s4when padding is non-SAME/VALID,output_dims->h != 1, left and right padding overlap ins4, orinput_chis odd ins4(wherearm_nn_mat_mul_core_1x_s4requires byte-aligned weight rows).