[inductor] fix bandwidth extimation for StarDep #120266
Closed
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
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/120266
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (3 Unrelated Failures)As of commit b2fca72 with merge base cccacf6 ( BROKEN TRUNK - The following jobs failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
UNSTABLE - The following job failed but was likely due to flakiness present on trunk and has been marked as unstable:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
eellison
approved these changes
Feb 21, 2024
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
jansel
approved these changes
Feb 21, 2024
pytorchmergebot
pushed a commit
that referenced
this pull request
Feb 22, 2024
Log a few more fields - num_atomic_add: perf of kernels using atomic_add are usually data dependent. Our benchmarking code generate all indices to be 0 which will result in worse perf than reality. - kernel_args_num_gb: estimate the amount of read/writes for kernel args. In-place args will be double counted. If we have a good estimation, this should be the lower bound of memory access that the GPU performs. Sometimes GPU will do more memory access since a single buffer may be access multiple times (e.g. for softmax when input tensor is quite large. cache only help a bit here). With this logged, and if we augment the metadata with amount of memory the GPU actually accessed, then it would be nice to dig into kernels that GPU access more memory. Pull Request resolved: #120274 Approved by: https://github.com/jansel ghstack dependencies: #120266
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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):
A lot of HF models fail when inductor_config.bechmark_kernel is enabled. The reason is the bandwidth estimation code assumes every dependencies has an index but StarDep does not. An exception is raised when StarDep.index is being accessed.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @aakhundov @ColinPeppler @amjames