-
Notifications
You must be signed in to change notification settings - Fork 26.3k
Get rid of DefaultNumBailouts #73689
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
[ghstack-poisoned]
CI Flow Status⚛️ CI FlowRuleset - Version:
|
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 715b8b6 (more details on the Dr. CI page):
🕵️ 1 new failure recognized by patternsThe following CI failures do not appear to be due to upstream breakages:
|
| Job | Step | Action |
|---|---|---|
| Run mypy | 🔁 rerun |
This comment was automatically generated by Dr. CI (expand for details).
Please report bugs/suggestions to the (internal) Dr. CI Users group.
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor. [ghstack-poisoned]
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor. [ghstack-poisoned]
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor. [ghstack-poisoned]
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor. [ghstack-poisoned]
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor. [ghstack-poisoned]
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor. [ghstack-poisoned]
| # we haven't tested every single test in this file | ||
| # but we enable FULL_PROFILER for a large subset | ||
| # of the tests with "with enable_profiling_mode_for_profiling_tests" | ||
| torch._C._jit_set_profiling_mode(False) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
also updates this finally, needed to fix a couple tests @Krovatkin
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor. [ghstack-poisoned]
davidberard98
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Awesome! Looks good to me
nit: is there a reason we can't get rid of all the bailOut arguments, e.g. in graph_executor.cpp? It doesn't seem like they are used anywhere
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor. [ghstack-poisoned]
i think some of it is still being used for passing in the number of remaining specializations.. but yea some cleanup could probably still be done |
ghstack-source-id: 2cb01bd Pull Request resolved: pytorch/pytorch#73875 Get rid of DefaultNumBailouts Pull Request resolved: pytorch/pytorch#73689
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
Stack from ghstack:
The number of bailouts of each executor should just be instantiated from whatever the current fusion strategy is, which is then stored on the executor.