Open
Conversation
gunes-arm
reviewed
Feb 6, 2026
filelist.json
Outdated
| "src/cpu/kernels/topkv/generic/neon/qasymm8_signed.cpp" | ||
| ] | ||
| } | ||
| "files": { |
Contributor
There was a problem hiding this comment.
why the indentation change?
Contributor
Author
There was a problem hiding this comment.
fixed in next patch
morgolock
force-pushed
the
pr/topkv_sve_kernels
branch
from
1deb127 to
7944860
Compare
morgolock
force-pushed
the
pr/topkv_sve_kernels
branch
from
7944860 to
21ba8e9
Compare
gunes-arm
reviewed
Mar 6, 2026
gunes-arm
requested changes
Mar 9, 2026
| inline uint32_t count_gt_block<float16_t>(const float16_t *ptr, float16_t thr, uint32_t block_elems) | ||
| { | ||
| const svbool_t pg = svwhilelt_b16(static_cast<uint64_t>(0), static_cast<uint64_t>(block_elems)); | ||
| const svfloat16_t v = svld1_f16(pg, ptr); |
Contributor
There was a problem hiding this comment.
I think I have two questions to ask:
- Why do we convert to Fp32 in the Neon(TM) implementation?
- We should incorporate epsilon in both implementations.
Contributor
Author
There was a problem hiding this comment.
Addressed in next patch.
Contributor
Author
There was a problem hiding this comment.
Why do we convert to Fp32 in the Neon(TM) implementation
We could do the computation in f16 but the epsilon must be f32 to align with ref. I can address this in a different patch. It's not a serious problem, if anything there is room to optimize the neon kernel even further.
morgolock
force-pushed
the
pr/topkv_sve_kernels
branch
from
21ba8e9 to
36f1ac1
Compare
morgolock
force-pushed
the
pr/topkv_sve_kernels
branch
from
36f1ac1 to
c6b4071
Compare
Resolves MLCE-1719 Change-Id: I7a0c7bd1154b9cb7f35c7fd1c3b8ad54698f8799 Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com>
morgolock
force-pushed
the
pr/topkv_sve_kernels
branch
from
c6b4071 to
98ab9ec
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
2 participants
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.
Change-Id: I7a0c7bd1154b9cb7f35c7fd1c3b8ad54698f8799