Merge group related improvements for convolution operations #3439
+2,097
−525
Conversation
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
zsotakal
requested review from
a team,
ThomasNing,
afagaj,
andriy-ca,
aosewski,
asleepzzz,
bartekxk,
carlushuang,
cgmillette,
coderfeli,
geyyer,
illsilin,
poyenc,
qianfengz,
shumway,
tenpercent and
vidyasagar-amd
as code owners
zsotakal
force-pushed
the
streamhpc/conv_mergegroups_updates
branch
from
18ead08 to
cf8a4d4
Compare
zsotakal
force-pushed
the
streamhpc/conv_mergegroups_updates
branch
from
5fa418d to
706ca76
Compare
zsotakal
force-pushed
the
streamhpc/conv_mergegroups_updates
branch
from
706ca76 to
6fe2eaa
Compare
EnricoDeg
reviewed
Dec 19, 2025
Comment on lines
+594
to
+596
| // implement on demand | ||
| static_assert(NumGroupsToMerge == 1); | ||
|
|
Contributor
There was a problem hiding this comment.
Can you add a message that merge group doesn't support GNHWK?
Comment on lines
+611
to
+613
| // implement on demand | ||
| static_assert(NumGroupsToMerge == 1); | ||
|
|
Contributor
There was a problem hiding this comment.
Same here and in the other parts
| Sequence<3>{}, | ||
| Sequence<4>{})); | ||
|
|
||
| #if CK_USE_CUSTOM_TENSOR_TRANSFORM_FOR_BWD_DATA_OUT == 0 |
Contributor
There was a problem hiding this comment.
what's the purpose of this?
| index_t Ho, | ||
| index_t Wo, | ||
| index_t K, | ||
| [[maybe_unused]] index_t YDot, |
kabrahamAMD
approved these changes
Dec 22, 2025
| } | ||
| }; | ||
|
|
||
| /** |
Contributor
There was a problem hiding this comment.
Is there a way to avoid duplication with the struct above? Afaik, these functions are only used in two places, so I would recommend to always call the MG variant and to have a defalut value (and ignore the GStep return value) to avoid duplicating the rest of the logic
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.
A collection of several depthwise convolution related improvements.
Proposed changes
The list of todos came from the investigation of 2d convolution performance on fp32 data input. It turns out CK has limited support for merged group convolutions. The purpose of this PR is to add some of the missing functionality.
Checklist
Please put an
xinto the boxes that apply. You can also fill these out after creating the PR. If you're not sure, please don't hesitate to ask.clang-formaton all changed filesDiscussion
If this is a relatively large or complex change, feel free to start a discussion by explaining why you chose the solution you did and what alternatives you considered