[ET-VK][benchmarking][ez] Don't perform copies when benchmarking #9468
Conversation
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
## Context The bencmarks generated by the generated operator benchmarks currently have a high amount of copy overhead: 1. Copy from CPU to staging 2. Copy from staging to GPU Buffer/Image And this is done for both inputs and outputs. Since benchmarks are not correctness tests, copying data in/out is not really necessary especially if the compute shader does not have behaviour dependent on the contents of the input/output tensor. Make it so that by default, the benchmark will only execute the op without adding copy overhead. However, test cases can optionally specify that the copy overhead should be included in the benchmark. Differential Revision: [D71570143](https://our.internmc.facebook.com/intern/diff/D71570143/) [ghstack-poisoned]
SS-JIA
added a commit
that referenced
this pull request
Mar 20, 2025
## Context The bencmarks generated by the generated operator benchmarks currently have a high amount of copy overhead: 1. Copy from CPU to staging 2. Copy from staging to GPU Buffer/Image And this is done for both inputs and outputs. Since benchmarks are not correctness tests, copying data in/out is not really necessary especially if the compute shader does not have behaviour dependent on the contents of the input/output tensor. Make it so that by default, the benchmark will only execute the op without adding copy overhead. However, test cases can optionally specify that the copy overhead should be included in the benchmark. Differential Revision: [D71570143](https://our.internmc.facebook.com/intern/diff/D71570143/) ghstack-source-id: 273059929 Pull Request resolved: #9468
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/9468
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 1 PendingAs of commit f4c4585 with merge base 7159650 ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D71570143 |
This PR needs a
|
…arking" ## Context The bencmarks generated by the generated operator benchmarks currently have a high amount of copy overhead: 1. Copy from CPU to staging 2. Copy from staging to GPU Buffer/Image And this is done for both inputs and outputs. Since benchmarks are not correctness tests, copying data in/out is not really necessary especially if the compute shader does not have behaviour dependent on the contents of the input/output tensor. Make it so that by default, the benchmark will only execute the op without adding copy overhead. However, test cases can optionally specify that the copy overhead should be included in the benchmark. Differential Revision: [D71570143](https://our.internmc.facebook.com/intern/diff/D71570143/) [ghstack-poisoned]
SS-JIA
added a commit
that referenced
this pull request
Mar 26, 2025
Pull Request resolved: #9468 ## Context The bencmarks generated by the generated operator benchmarks currently have a high amount of copy overhead: 1. Copy from CPU to staging 2. Copy from staging to GPU Buffer/Image And this is done for both inputs and outputs. Since benchmarks are not correctness tests, copying data in/out is not really necessary especially if the compute shader does not have behaviour dependent on the contents of the input/output tensor. Make it so that by default, the benchmark will only execute the op without adding copy overhead. However, test cases can optionally specify that the copy overhead should be included in the benchmark. Differential Revision: [D71570143](https://our.internmc.facebook.com/intern/diff/D71570143/) ghstack-source-id: 274197244
|
This pull request was exported from Phabricator. Differential Revision: D71570143 |
yipjustin
approved these changes
Mar 26, 2025
facebook-github-bot
merged commit e68552d
into
gh/SS-JIA/199/base
82 of 83 checks passed
facebook-github-bot
temporarily deployed
to
cherry-pick-bot
GitHub Actions
Inactive
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.
Stack from ghstack (oldest at bottom):
Context
The bencmarks generated by the generated operator benchmarks currently have a high amount of copy overhead:
And this is done for both inputs and outputs.
Since benchmarks are not correctness tests, copying data in/out is not really necessary especially if the compute shader does not have behaviour dependent on the contents of the input/output tensor.
Make it so that by default, the benchmark will only execute the op without adding copy overhead. However, test cases can optionally specify that the copy overhead should be included in the benchmark.
Differential Revision: D71570143