Expose ExecutionPlan in prep for function calls #21273
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
Summary: This reorganizes GraphExecutor into a stage that looks up the execution plan getPlanFor, and a stage that runs the plan. This separation is necessary for first-class functions since the interpreter will call getPlanFor to look up the Code for the function it will be calling rather than recursively calling run on the graph executor. Test Plan: test_jit.py
This was referenced Jun 3, 2019
Expose ExecutionPlan in prep for function calls Summary: This reorganizes GraphExecutor into a stage that looks up the execution plan getPlanFor, and a stage that runs the plan. This separation is necessary for first-class functions since the interpreter will call getPlanFor to look up the Code for the function it will be calling rather than recursively calling run on the graph executor. Test Plan: test_jit.py gh-metadata: pytorch pytorch 21273 gh/zdevito/47/head
Krovatkin
approved these changes
Jun 4, 2019
|
|
||
| // entry point where execution begins | ||
| virtual void run(Stack& stack) = 0; | ||
| void run(Stack& stack); |
There was a problem hiding this comment.
i think we might need virtual GraphExecutorImpl::run eventually, but I guess, we can make it virtual when we need it
There was a problem hiding this comment.
It would be possible. Keep in mind that run will only get called on entry from C++ into TorchScript. Function calls go directly through the getPlan pathway. So anything that has to happen for every function call will need to be expressed as part of getPlan.
Expose ExecutionPlan in prep for function calls Summary: This reorganizes GraphExecutor into a stage that looks up the execution plan getPlanFor, and a stage that runs the plan. This separation is necessary for first-class functions since the interpreter will call getPlanFor to look up the Code for the function it will be calling rather than recursively calling run on the graph executor. Test Plan: test_jit.py gh-metadata: pytorch pytorch 21273 gh/zdevito/47/head
Expose ExecutionPlan in prep for function calls Summary: This reorganizes GraphExecutor into a stage that looks up the execution plan getPlanFor, and a stage that runs the plan. This separation is necessary for first-class functions since the interpreter will call getPlanFor to look up the Code for the function it will be calling rather than recursively calling run on the graph executor. Test Plan: test_jit.py gh-metadata: pytorch pytorch 21273 gh/zdevito/47/head
This was referenced Jun 7, 2019
Expose ExecutionPlan in prep for function calls Summary: This reorganizes GraphExecutor into a stage that looks up the execution plan getPlanFor, and a stage that runs the plan. This separation is necessary for first-class functions since the interpreter will call getPlanFor to look up the Code for the function it will be calling rather than recursively calling run on the graph executor. Test Plan: test_jit.py gh-metadata: pytorch pytorch 21273 gh/zdevito/47/head
Expose ExecutionPlan in prep for function calls Summary: This reorganizes GraphExecutor into a stage that looks up the execution plan getPlanFor, and a stage that runs the plan. This separation is necessary for first-class functions since the interpreter will call getPlanFor to look up the Code for the function it will be calling rather than recursively calling run on the graph executor. Test Plan: test_jit.py gh-metadata: pytorch pytorch 21273 gh/zdevito/47/head
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 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:
Summary: This reorganizes GraphExecutor into a stage that looks up
the execution plan getPlanFor, and a stage that runs the plan. This
separation is necessary for first-class functions since the interpreter
will call getPlanFor to look up the Code for the function it will be
calling rather than recursively calling run on the graph executor.
Test Plan: test_jit.py
Differential Revision: D15600067