Improve error reporting with loop operator by pfultz2 · Pull Request #4949 · ROCm/AMDMIGraphX

pfultz2 · 2026-06-09T17:30:04Z

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.

Copilot

Pull request overview

Improves loop operator shape-validation and diagnostics to fail fast with clearer errors when loop inputs/submodule outputs are inconsistent.

Changes:

Tightens input validation by requiring at least 2 inputs for loop::compute_shape.
Adds error reporting when the submodule does not provide enough output shapes to cover dependency parameters.
Updates the copyright year range in the header.

    shape compute_shape(const std::vector<shape>& inputs, std::vector<module_ref> mods) const
    {
-        check_shapes{inputs, *this}.standard();
+        check_shapes{inputs, *this}.standard().has_at_least(2);


        auto dep_param_num   = inputs.size() - 2;

        // first item of the mod output shapes is condition used in loop,
        // which is not needed to compute output shape
        mod_out_shapes.erase(mod_out_shapes.begin());


codecov · 2026-06-09T17:51:22Z

Codecov Report

❌ Patch coverage is 66.66667% with 1 line in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/include/migraphx/op/loop.hpp	66.67%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #4949      +/-   ##
===========================================
- Coverage    92.63%   92.62%   -0.00%     
===========================================
  Files          587      587              
  Lines        30469    30472       +3     
===========================================
+ Hits         28222    28224       +2     
- Misses        2247     2248       +1

Files with missing lines	Coverage Δ
src/include/migraphx/op/loop.hpp	`96.61% <66.67%> (-1.60%)`	⬇️

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

kahmed10

are there any tests we can add for this one? if it's a bug with the onnx operator?

gh-app-migraphx-bot-pr-write · 2026-06-09T21:15:06Z

Test	Batch	New Rate (a533b7)	Old Rate (46e035)*	Diff	Status
torchvision-resnet50	64	1,922.38	3,149.42	-38.96%	🔴
torchvision-resnet50_fp16	64	4,057.13	6,669.17	-39.17%	🔴
torchvision-densenet121	32	522.41	2,708.19	-80.71%	🔴
torchvision-densenet121_fp16	32	3,967.19	4,558.26	-12.97%	🔴
torchvision-inceptionv3	32	1,740.64	1,800.92	-3.35%	✅
torchvision-inceptionv3_fp16	32	2,785.25	2,828.03	-1.51%	✅
cadene-inceptionv4	16	816.21	824.44	-1.00%	✅
cadene-resnext64x4	16	768.73	784.01	-1.95%	✅
slim-mobilenet	64	5,668.68	8,428.72	-32.75%	🔴
slim-nasnetalarge	64	227.49	229.43	-0.85%	✅
slim-resnet50v2	64	3,315.22	3,330.80	-0.47%	✅
bert-mrpc-onnx	8	1,162.29	1,172.93	-0.91%	✅
bert-mrpc-tf	1	488.77	487.62	0.23%	✅
pytorch-examples-wlang-gru	1	327.76	328.01	-0.08%	✅
pytorch-examples-wlang-lstm	1	477.92	452.21	5.69%	🔆
torchvision-resnet50_1	1	772.00	757.22	1.95%	✅
cadene-dpn92_1	1	455.40	441.04	3.26%	✅
cadene-resnext101_1	1	363.55	363.74	-0.05%	✅
onnx-taau-downsample	1	400.93	400.37	0.14%	✅
dlrm-criteoterabyte	1	32.41	32.72	-0.94%	✅
dlrm-criteoterabyte_fp16	1	52.65	52.58	0.13%	✅
agentmodel	1	9,914.24	10,096.04	-1.80%	✅
unet_fp16	2	57.23	57.18	0.08%	✅
resnet50v1_fp16	1	962.30	1,029.05	-6.49%	🔴
resnet50v1_int8	1	934.26	931.64	0.28%	✅
bert_base_cased_fp16	64	1,096.90	1,103.01	-0.55%	✅
bert_large_uncased_fp16	32	347.60	347.55	0.02%	✅
bert_large_fp16	1	204.94	203.61	0.65%	✅
distilgpt2_fp16	16	2,081.43	2,098.45	-0.81%	✅
yolov5s	1	566.61	568.07	-0.26%	✅
tinyllama	1	45.99	46.02	-0.05%	✅
vicuna-fastchat	1	44.06	43.98	0.20%	✅
whisper-tiny-encoder	1	419.76	420.51	-0.18%	✅
whisper-tiny-decoder	1	419.10	417.19	0.46%	✅
llama2_7b	1	20.46	20.46	0.02%	✅
qwen1.5-7b	1	18.06	23.66	-23.70%	🔴
phi3-3.8b	1	26.89	26.93	-0.15%	✅
llama3-8b	1	21.67	21.81	-0.65%	✅
whisper-large-encoder	1	9.06	10.32	-12.18%	🔴
whisper-large-decoder	1	105.90	104.08	1.75%	✅
mistral-7b	1	23.58	23.87	-1.21%	✅
FLUX.1-schnell	1	759.86	743.82	2.16%	✅

Regressions detected 🔴

* No develop baseline was found for this PR's branch point; compared against the latest available develop run instead.

gh-app-migraphx-bot-pr-write · 2026-06-09T21:15:07Z

Test	Status	Result
bert-mrpc-onnx	✅	PASSED: MIGraphX meets tolerance
bert-mrpc-tf	❌	ERROR - check error output traceback Traceback (most recent call last): File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 377, in main() File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 313, in main import tensorflow as tf File "/usr/local/lib/python3.10/dist-packages/tensorflow/init.py", line 38, in from tensorflow.python.tools import module_util as _module_util File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/init.py", line 36, in from tensorflow.python import pywrap_tensorflow as _pywrap_tensorflow File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 26, in self_check.preload_check() File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/platform/self_check.py", line 63, in preload_check from tensorflow.python.platform import _pywrap_cpu_feature_guard ImportError: libamdhip64.so.6: cannot open shared object file: No such file or directory
pytorch-examples-wlang-gru	✅	PASSED: MIGraphX meets tolerance
pytorch-examples-wlang-lstm	✅	PASSED: MIGraphX meets tolerance
dlrm-criteoterabyte	✅	PASSED: MIGraphX meets tolerance
agentmodel	✅	PASSED: MIGraphX meets tolerance
unet	✅	PASSED: MIGraphX meets tolerance
resnet50v1	✅	PASSED: MIGraphX meets tolerance
bert_base_cased_fp16	✅	PASSED: MIGraphX meets tolerance
bert_large_uncased_fp16	🔴	FAILED: MIGraphX is not within tolerance - check verbose output
bert_large	✅	PASSED: MIGraphX meets tolerance
yolov5s	✅	PASSED: MIGraphX meets tolerance
tinyllama	✅	PASSED: MIGraphX meets tolerance
vicuna-fastchat	✅	PASSED: MIGraphX meets tolerance
whisper-tiny-encoder	✅	PASSED: MIGraphX meets tolerance
whisper-tiny-decoder	✅	PASSED: MIGraphX meets tolerance
distilgpt2_fp16	🔴	FAILED: MIGraphX is not within tolerance - check verbose output
llama2_7b	✅	PASSED: MIGraphX meets tolerance
qwen1.5-7b	✅	PASSED: MIGraphX meets tolerance
phi3-3.8b	✅	PASSED: MIGraphX meets tolerance
llama3-8b	✅	PASSED: MIGraphX meets tolerance
whisper-large-encoder	❌	ERROR - check error output traceback 2026-06-09 16:09:13.153660 [WARN] [/data/src/onnx/onnx_parser.cpp:282] Model has unbound symbolic dimension(s): batch_size, encoder_sequence_length, feature_size. These default to 1 and may cause unexpected behavior. Try setting `--dim-param @<name> <value>` or `--input-dim @<input> <dims>` if program compilation fails. Traceback (most recent call last): File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 377, in main() File "/src/AMDMIGraphX/tools/accuracy/accuracy_checker.py", line 224, in main model = migraphx.parse_onnx(model_name, default_dim_value=batch) RuntimeError: /data/src/include/migraphx/op/convolution.hpp:113: normalize_compute_shape: CONVOLUTION: mismatched channel numbers: input channels (1) != weights channels (80) * group (1)
whisper-large-decoder	✅	PASSED: MIGraphX meets tolerance
mistral-7b	✅	PASSED: MIGraphX meets tolerance
FLUX.1-schnell	✅	PASSED: MIGraphX meets tolerance

Improve error reporting with loop operator

a533b73

Copilot AI review requested due to automatic review settings June 9, 2026 17:30

pfultz2 requested a review from causten as a code owner June 9, 2026 17:30

Copilot started reviewing on behalf of pfultz2 June 9, 2026 17:30 View session

Copilot AI reviewed Jun 9, 2026

View reviewed changes

kahmed10 approved these changes Jun 9, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Improve error reporting with loop operator#4949

Improve error reporting with loop operator#4949
pfultz2 wants to merge 1 commit into
developfrom
loop-error-check

pfultz2 commented Jun 9, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

codecov Bot commented Jun 9, 2026 •

edited

Loading

Uh oh!

kahmed10 left a comment •

edited

Loading

Uh oh!

gh-app-migraphx-bot-pr-write Bot commented Jun 9, 2026

Uh oh!

gh-app-migraphx-bot-pr-write Bot commented Jun 9, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

pfultz2 commented Jun 9, 2026

Motivation

Technical Details

Changelog Category

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

codecov Bot commented Jun 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

kahmed10 left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

gh-app-migraphx-bot-pr-write Bot commented Jun 9, 2026

Uh oh!

gh-app-migraphx-bot-pr-write Bot commented Jun 9, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

codecov Bot commented Jun 9, 2026 •

edited

Loading

kahmed10 left a comment •

edited

Loading