[digantdesai]

Stack from ghstack (oldest at bottom):

• [executorch][native] Record dynamic-shape linears as chained bundled samples #24405
• [executorch][native] Run the ggml CPU act_quant sample test in CI #24373
• -> [executorch][vulkan] Round f16 image stores to nearest even #24372
• [executorch][native-vk] Note the int4 weight-only N % 4 and rank-4 linear gaps #24371
• [executorch][vulkan] Fix dq8ca q4gsw weight sums stride when N % 8 is 1 to 4 #24370
• [executorch][vulkan] Accept 1-D activations in 8da4w linear #24369
• [executorch][native] Cover negative, rounded and subnormal GGUF super-block scales in the op samples #24368
• [executorch][native] Decode quantized weights in float32 for precise samples #24367
• [executorch][native] Add GGUF q4_k and q6_k packages to the op samples #24366
• [executorch][native] Add FACTO-drawn linear cases to the op samples #24365
• [executorch][native] Suggest rounding to nearest when engines narrow values #24364
• [executorch][native-ggml] Compute cuBLAS GEMMs in f32 in the CUDA sample tests #24363
• [executorch][native] Judge activation-quantized engines against an int8-activation golden #24362
• [executorch][native-ggml] Reject weight-only quantized linears unless activation quant is allowed #24361
• [executorch][native] Run engines on bundled op samples with gtest #24360
• [executorch][native] Judge engine outputs against golden and precise samples #24359
• [executorch][native] Pin eager kernels when generating op samples #24358
• [executorch][native] Record op packages with bundled golden and precise samples #24357
• [executorch][native] Split engine host factories out of the PTN runners #24356
• [executorch][test-suite] Feed the Linear tests inputs with the model's 67 input features #24355

Vulkan lets the device choose how stores to rgba16f images round, and devices that round toward zero bias f16 outputs; shader float controls don't govern that conversion. gen_vulkan_spv.py now wraps imageStore on writable rgba16f images in a round-to-nearest-even helper, and six shaders that declared such images by hand, two of them test shaders, now use the codegen.

On devices that round toward zero, stores of 65520 and above now give inf, as the CPU conversion does, instead of 65504. The four texture shaders that already call round_to_half_rte on their outputs (binary_scalar_texture, full_texture, reduce, scalar_tensor) now round twice, with identical results; removing those calls is a follow-up. aten._to_copy from f32 to f16 textures is a vkCmdBlitImage, which the codegen doesn't reach, so test_to_copy keeps its one-ulp tolerance.

Authored with Claude Code.

@bypass-github-export-checks
@bypass-github-pytorch-ci-checks
@bypass-github-executorch-ci-checks
@exported-using-ghexport

Differential Revision: D128254886

Differential Revision: D128254886

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/24372

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure

As of commit 21f064b with merge base 0f8e096 ():

NEW FAILURE - The following job has failed:

• pull / nxp-mcuxpresso-test / linux-job (gh)
  undefined reference to SYS_INFO_PrintAll'`

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[github-actions]

This PR needs a release notes: label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[SS-JIA]

Review automatically exported from Phabricator review in Meta.