[amd-mounikk]

Summary

• Merge a shape rule with the type already stored on the result: keep a known size, and fill ? only when the rule has a number. An unranked tensor<*xT> still takes the rule type.
• Read Slice, Pad, Unsqueeze, reduction, and Split bounds from a constant input when the attribute is absent. A runtime tensor returns no type, so the stored shape stays.
• Rank AveragePool (1x64x512x512, kernel 2, stride 2 becomes 1x64x256x256) and opset 7-8 Upsample from its scales attribute. An equal Split of a static axis writes the chunk size into extract_slice when that axis is still ?.

Related issue or design

Follow-up to the Gemma-4 shape regression reported on #1306. None filed separately.

Why

#1306 runs the shape rules on any ranked result that still contains a ?, then replaces the stored type. Slice and Pad, when starts/ends/pads are inputs rather than attributes, answered with a same-rank all-? type, so a stored size such as tensor<1x?x12x32xf16> became tensor<?x?x?x?xf16>. The Topaz UNet also left AveragePool and attribute-form Upsample unranked, so conversion never saw a rank.

What

• mergeInferredType in forwardInferUnrankedResults combines the stored type with the rule result.
• AveragePool uses the ONNX spatial formula. Upsample reads the scales attribute when there is no scales input.
• Slice, Pad, Unsqueeze, reductions, and Split accept a constant input vector and return no type for a runtime vector.
• SplitToStdTensor puts a constant chunk size on the extract_slice result and casts back to the declared type.

Test plan

• pre-commit run --files on the four changed files (passed after lintrunner formatting)
• hip-mlir-opt --hip-infer-loop-body-shapes test/lit/Dialect/hip-infer-loop-body-shapes.mlir | FileCheck
• hip-mlir-opt test/lit/Conversion/onnx-to-hip/test_split.mlir --hip-add-context-arg --convert-onnx-to-hip | FileCheck
• Gemma-4 and the Topaz UNet on a full session

Notes for reviewers

Drafted with Cursor assistance. I reviewed the merge rule, the constant-input fallback, and the lit checks.

The Gemma Slice case is locked as tensor<1x?x12x32xf16> when starts, ends, and axes are runtime tensors. The pool and attribute Upsample cases are in hip-infer-loop-body-shapes.mlir.

Checklist

• The change is focused, or links a design/series explaining its scope.
• Relevant tests were added or updated and the results are documented.
• User-facing or design documentation was updated when needed.
• Substantial AI assistance is disclosed, and I reviewed and understand the result.

[github-actions]

Thanks for opening a PR!

This project follows LLVM's incremental-development and AI-tool-use
guidance. See CONTRIBUTING.md
for the project workflow.

Before requesting review, please check that:

1. The change is focused. Substantial work links the relevant issue
   or design discussion.
2. The PR documents relevant test results and updates affected
   documentation.
3. If AI tools provided substantial assistance, the description
   explains what was assisted and how it was validated, and commit
   trailers identify the tool. The contributor has reviewed and
   understands the result.

Reviewers are assigned through
CODEOWNERS where ownership
is configured.

[Copilot]

🟡 Changes recommended

Empty reduction axes, ceil-mode pooling boundaries, and integer-bound decoding can produce incorrect shapes or undefined behavior.

3 open issues

• Empty axes incorrectly infer no-op instead of all input axes · New
• Ceil-mode pooling retains window entirely in right padding · New
• Integer constant evaluation loses precision and overflows · New

What changed in this PR

This PR follows up on #1306 by preserving known result dimensions during incomplete ONNX shape inference and expanding inference for additional operator forms.

Changes:

• Merge inferred shapes with stored types without replacing known dimensions.
• Read constant shape inputs and infer AveragePool and attribute-form Upsample shapes.
• Refine static equal-Split slice sizes and add targeted regression tests.

File	Description
test/​lit/​Dialect/​hip-infer-loop-body-shapes.mlir	Adds pooling, Upsample, and Slice shape-inference regressions.
test/​lit/​Conversion/​onnx-to-hip/​test_split.mlir	Tests constant slice sizes for dynamically typed Split results.
lib/​Dialect/​Transforms/​InferLoopBodyShapesPass.cpp	Preserves stored dimensions and extends operator shape rules.
lib/​Conversion/​OnnxToHip/​ReshapeConversion.cpp	Refines Split slice types and casts back to declared types.

_{Review effort: Balanced}

[github-actions]

L2 Accuracy Results (EP vs CPU)

Model	Combined L2	Total Elems	Skipped NaN/Inf
conv_test_hybrid	1.12256E-07	64	0
GroupQueryAttention_seq256	59.2561	2621440	0
MatMulNBits_o_seq128	259.905	368640	0
QMoE_seq128	24.2583	368640	0

Threshold: 0.01 | Run: 7942 - Commit: afa8642

[github-actions]

Binary sizes

Baseline: run 40440689818 - 7f2f3ce (dll-sizes)

File	This (MB)	Main (MB)	Delta (MB)
hipgpu.dll	59.32	59.31	+0.01
custom_kernels_gfx1170.dll	5.55	5.55	0.00
custom_kernels_gfx11-generic.dll	5.79	5.79	0.00

MorphiZen EP Performance Results

Model	QPS	Session (s)	1st Infer (ms)	CPU%	Mem (MB)
full_model_seq128	8.77	7.45	246	3	1158
GroupQueryAttention_seq128	11345.00	2.55156	1	5	229
matmul_down_seq128	508.90	3.28	116	3	274

EPContext Export Performance

Model	QPS	Session (s)	1st Infer (ms)	CPU%	Mem (MB)
full_model_seq128	8.82	48.31	251	3	15505

EPContext Import Performance

Model	QPS	Session (s)	1st Infer (ms)	CPU%	Mem (MB)
full_model_seq128	8.83	9.97	255	3	15639

OGA Benchmark Results

Model	Warmup	Reps	Prompt Len	Gen Tokens	TTFT (ms)	TPS	Peak Mem (GB)	GPU Mem (GB)
gpt-oss-20b-webgpu-int4-rtn-block-32	1	5	128	128	91.7	74.7	1.25	12.24
Llama-3.1-8B-awq-g128-int4-asym-fp16-onnx-dml	1	5	128	128	118.6	47.7	1.15	5.23

OGA Wheel Smoke (Python benchmark_e2e.py)

Model	TTFT (ms)	TPS
Llama-3.1-8B-awq-g128-int4-asym-fp16-onnx-dml	117	48.0

VLM Benchmark Results

Model	TTFT (ms)	TPS	Output
gemma-4-E4B-it-fp16-int4-kquant-prune-logits	4005.3	7.7	This is a high-quality, outdoor photograph featuring a beautiful, fluffy white dog, likely a West Highland White Terrier or a si
Qwen3.8-27B-fp16-ve-fp16-int4-k_quant-gs128-text-dml-onnx	4922.1	14.0	The user wants a detailed description of the image. Let me carefully observe all elements before writing. Subject: A white do

TTFT / TPS vs main

Baseline: run 40440689818 - 7f2f3ce | Gate: TTFT +20% / TPS -20%

Suite	Model	TTFT (ms)	Main	Delta	TPS	Main	Delta
oga	gpt-oss-20b-webgpu-int4-rtn-block-32	91.7	90.4	+1.5%	74.7	74.9	-0.2%
oga	Llama-3.1-8B-awq-g128-int4-asym-fp16-onnx-dml	118.6	116.7	+1.6%	47.7	47.8	-0.3%
vlm	gemma-4-E4B-it-fp16-int4-kquant-prune-logits	4005.3	4009.7	-0.1%	7.7	7.2	+7.1%
vlm	Qwen3.8-27B-fp16-ve-fp16-int4-k_quant-gs128-text-dml-onnx	4922.1	14742.4	-66.6%	14.0	14.0	-0.1%

Run: 7942 - Commit: afa8642