build: reduce build time for bquant tests by splitting into multiple cpp & support on other gfx10 case#3395
Merged
ThomasNing merged 8 commits intodevelopfrom Dec 15, 2025
Conversation
8c8a3d0 to
f52ee0a
Compare
Collaborator
Author
amd-khushbu
previously approved these changes
Dec 12, 2025
Contributor
There was a problem hiding this comment.
Pull request overview
This PR significantly reduces build time for bquant tests by splitting large test files into smaller, more focused compilation units. The build time improvement is approximately 65.5% (from 4m 24s to 1m 31s). Additionally, the PR extends architecture support to include additional gfx10 variants (gfx1011, gfx1012, gfx1013, and gfx10_1_generic).
Key Changes
- Split monolithic test files into multiple smaller test files organized by configuration type
- Added support for gfx10.1 architecture variants (gfx1011, gfx1012, gfx1013, gfx10_1_generic)
- Created umbrella build target for all gemm quant tests
Reviewed changes
Copilot reviewed 23 out of 23 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Description |
|---|---|
| test/ck_tile/grouped_gemm_quant/CMakeLists.txt | Commented out previously split test executables |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant.cpp | Removed monolithic BQuant test file |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_1d_128.cpp | New file for 1D GroupSize 128 BQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_1d_64.cpp | New file for 1D GroupSize 64 BQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_2d_small_n.cpp | New file for 2D small N (8N, 16N) BQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_2d_medium_n.cpp | New file for 2D medium N (32N, 64N) BQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_2d_large_n.cpp | New file for 2D large N (128N) BQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_transpose.cpp | New file for transpose layout BQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_preshuffle.cpp | Removed monolithic preshuffle BQuant test file |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_preshuffle_decode_1d.cpp | New file for 1D decode preshuffle tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_preshuffle_prefill_1d.cpp | New file for 1D prefill preshuffle tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_preshuffle_tiled_permute.cpp | New file for tiled permute preshuffle tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_preshuffle_decode_2d.cpp | New file for 2D decode preshuffle tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_bquant_preshuffle_prefill_2d.cpp | New file for 2D prefill preshuffle tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_aquant.cpp | Removed monolithic AQuant test file |
| test/ck_tile/gemm_block_scale/test_gemm_quant_aquant_base_rcr.cpp | New file for RCR layout AQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_aquant_base_rrr_crr.cpp | New file for RRR/CRR layout AQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_aquant_base_ccr.cpp | New file for CCR layout AQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_aquant_prefill.cpp | New file for prefill configuration AQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_aquant_transpose_c.cpp | New file for transpose C configuration AQuant tests |
| test/ck_tile/gemm_block_scale/test_gemm_quant_aquant_preshuffle.cpp | New file for preshuffle configuration AQuant tests |
| test/ck_tile/gemm_block_scale/CMakeLists.txt | Updated to build new test executables and added umbrella target |
| include/ck_tile/core/config.hpp | Added architecture support for gfx1011, gfx1012, gfx1013, and gfx10_1_generic |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
ThomasNing
approved these changes
Dec 15, 2025
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.

Proposed changes
Described as the PR title.
We reduce the build time from 4m 24s to 1m 31s.
The build time was reduced by approximately 65.5%.
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