[ET-VK][BE] Remove usage of vTensorPtr and get_tensor

[SS-JIA]

Stack from ghstack (oldest at bottom):

• [ET-VK][ez] Small optimization to DynamicDispatchNode #13159
• [ET-VK][BE] Move all ops to use DynamicDispatchNode #13158
• [ET-VK][ez] Make get_tensor() API protected #13157
• -> [ET-VK][BE] Remove usage of vTensorPtr and get_tensor #13156

Note that although the volume of changes in this diff are very high, the changes themselves are extremely mechanical. This diff was written almost entirely with a LLM, but I have looked through each file and validated the changes.

Changes

This diff updates callsites using graph->get_tensor(value_ref) in favor of just using the ValueRef directly.

A simple example (and the vast majority of changes in this diff) is a change such as:

vTensorPtr tensor = graph->get_tensor(tensor_ref);

some_fn(tensor->sizes());

To instead be

std::vector<int64_t> tensor_sizes = graph->sizes_of(tensor_ref);
some_fn(tensor_sizes);

or

some_fn(graph->sizes_of(tensor_ref));

Motivation

Overall, the goal is to make the get_tensor() API protected so that it can only be used in specific situations.

In addition to the primary motivation of improving the consistency of API usage throughout the codebase, there is a practical benefit as well. get_tensor has a limitation that no values can be added to the graph while the vTensorPtr is in scope. Also, forcing tensor modifications via functions like virtual_resize() to go through the ComputeGraph will allow the graph to track changes for the purposes of determining when a command buffer re-encode or resize propagation is necessary, which will result in performance benefits.

Differential Revision: D79564594

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/13156

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• ghstack-mergeability-check and Check labels failing with 'Resource not accessible by integration'

❌ 1 New Failure, 1 Unrelated Failure

As of commit 7c32276 with merge base 6bc312a ():

NEW FAILURE - The following job has failed:

• pull / test-moshi-linux / linux-job (gh)
  RuntimeError: Command docker exec -t b000a6ba803dafb04dbce5723942a562d88d195ecf49cf3f9848908e103a9603 /exec failed with exit code 1

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• pull / test-openvino-linux / linux-job (gh) (trunk failure)
  AttributeError: '_OpNamespace' 'quantized_decomposed' object has no attribute 'convert_element_type'

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

[facebook-github-bot]

This pull request was exported from Phabricator. Differential Revision: D79564594

[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.