[CK_TILE] Grouped Convolution Backward Data Direct Load - #4270
Closed
assistant-librarian[bot] wants to merge 38 commits into
Closed
[CK_TILE] Grouped Convolution Backward Data Direct Load#4270assistant-librarian[bot] wants to merge 38 commits into
assistant-librarian[bot] wants to merge 38 commits into
Conversation
Co-authored-by: Jakub Piasecki <jakpia21@gmail.com>
…data_direct_loads
…evelop/ROCm_composable_kernel/pr-3685
assistant-librarian
Bot
requested review from
afagaj,
andriy-ca,
aosewski,
asleepzzz,
bartekxk,
carlushuang,
geyyer,
illsilin,
poyenc and
qianfengz
as code owners
February 3, 2026 21:46
assistant-librarian
Bot
requested review from
ThomasNing,
coderfeli and
tenpercent
as code owners
February 3, 2026 21:46
…evelop/ROCm_composable_kernel/pr-3670
…ROCm_composable_kernel/pr-3685
jakpiase
changed the title
Contributor
|
Reoepend as: #6624 due to CI bugs |
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
4 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Proposed changes
Add Grouped Convolution Backward Data with Direct Load into DeviceGroupedConvBwdDataMultipleD_Xdl_CShuffleV3 device implementation. This enables direct global memory loading (bypassing LDS) for the backward data convolution path on gfx950, following the same pattern used in both backward weight and forward convolution.
Direct load convolution backward data improves performance by avoiding LDS round-trips for certain configurations on gfx950, which supports a wider range of instructions. Currently correctness is checked only at usage point, but should be extended to a standalone UT in the future.
🔁 Imported from ROCm/composable_kernel#3685
🧑💻 Originally authored by @jakpiase