Skip to content

Add conv winograd for gfx12#4936

Draft
pfultz2 wants to merge 34 commits into
developfrom
winograd4
Draft

Add conv winograd for gfx12#4936
pfultz2 wants to merge 34 commits into
developfrom
winograd4

Conversation

@pfultz2

@pfultz2 pfultz2 commented Jun 3, 2026

Copy link
Copy Markdown
Collaborator

Motivation

Technical Details

Changelog Category

Add a CHANGELOG.md entry for any option other than Not Applicable

    • Added: New functionality.
    • Changed: Changes to existing functionality.
    • Removed: Functionality or support that has been removed. (Compared to a previous release)
    • Optimized: Component performance that has been optimized or improved.
    • Resolved Issues: Known issues from a previous version that have been resolved.
    • Not Applicable: This PR is not to be included in the changelog.

@codecov

codecov Bot commented Jun 3, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.

Additional details and impacted files
@@             Coverage Diff             @@
##           develop    #4936      +/-   ##
===========================================
- Coverage    92.66%   92.62%   -0.04%     
===========================================
  Files          588      587       -1     
  Lines        30412    30456      +44     
===========================================
+ Hits         28180    28208      +28     
- Misses        2232     2248      +16     
Files with missing lines Coverage Δ
src/include/migraphx/op/leaky_relu.hpp 100.00% <100.00%> (ø)

... and 28 files with indirect coverage changes

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

repeat_c<4>([&](auto j) {
const int32_t e_off = (hi[i] and wj[j]) ? off + static_cast<int>(i) * sh_b +
static_cast<int>(j) * sw_b
: oob_byte;

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[format.py] reported by reviewdog 🐶

Suggested change
: oob_byte;
: oob_byte;

match::find_matches(mpm, find_gemm_softmax_gemm{enable_attention});
if(is_navi)
match::find_matches(mpm.get_module(), find_channelwise_convolution{});
if(is_gfx12)

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we have plans to support gfx11?

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes I do. It might already work on gfx11 but I havent tested it yet.

@TedThemistokleous TedThemistokleous self-requested a review June 15, 2026 19:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants