[compiler][bugfix] expand StoreContext to const / let / function variants #32747
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
mofeiZ
commented
Mar 25, 2025
Comment on lines
+904
to
+917
| /** | ||
| * StoreContext kinds: | ||
| * Reassign: context variable reassignment in source | ||
| * Const: const declaration + assignment in source | ||
| * ('const' context vars are ones whose declarations are hoisted) | ||
| * Let: let declaration + assignment in source | ||
| * Function: function declaration in source (similar to `const`) | ||
| */ | ||
| lvalue: { | ||
| kind: InstructionKind.Reassign; | ||
| kind: | ||
| | InstructionKind.Reassign | ||
| | InstructionKind.Const | ||
| | InstructionKind.Let | ||
| | InstructionKind.Function; |
There was a problem hiding this comment.
This is the key change. Instead of always lowering context variables to a DeclareContext followed by a StoreContext reassign, we can keep the kind
mofeiZ
added a commit
that referenced
this pull request
Mar 25, 2025
…tions (note: also see alternative implementation in #32747) `PruneHoistedContexts` currently strips hoisted declarations and rewrites the first `StoreContext` reassignment to a declaration. For example, in the following example, instruction 0 is removed while a synthetic `DeclareContext let` is inserted before instruction 1. ```js // source const cb = () => x; // reference that causes x to be hoisted let x = 4; x = 5; // React Compiler IR [0] DeclareContext HoistedLet 'x' ... [1] StoreContext reassign 'x' = 4 [2] StoreContext reassign 'x' = 5 ``` Currently, we don't account for `DeclareContext let`. As a result, we're rewriting to insert duplicate declarations. For the below example, we should only remove instruction 0 (no need to insert a DeclareContext `let` since one is already present). ```js // source const cb = () => x; // reference that causes x to be hoisted let x; x = 5; // React Compiler IR [0] DeclareContext HoistedLet 'x' ... [1] DeclareContext Let 'x' [2] StoreContext reassign 'x' = 5 ```
josephsavona
approved these changes
Apr 30, 2025
josephsavona
approved these changes
Apr 30, 2025
| @@ -30,8 +30,7 @@ function Component(props) { | |||
| const $ = _c(1); | |||
| let t0; | |||
| if ($[0] === Symbol.for("react.memo_cache_sentinel")) { | |||
| let x; | |||
| x = null; | |||
| let x = null; | |||
mofeiZ
added a commit
that referenced
this pull request
Apr 30, 2025
Fixes an edge case in React Compiler's effects inference model. Returned values should only be typed as 'frozen' if they are (1) local and (2) not a function expression which may capture and mutate this function's outer context. See test fixtures for details --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/facebook/react/pull/33047). * #32765 * #32747 * __->__ #33047
github-actions bot
pushed a commit
that referenced
this pull request
Apr 30, 2025
Fixes an edge case in React Compiler's effects inference model. Returned values should only be typed as 'frozen' if they are (1) local and (2) not a function expression which may capture and mutate this function's outer context. See test fixtures for details --- [//]: # (BEGIN SAPLING FOOTER) Stack created with [Sapling](https://sapling-scm.com). Best reviewed with [ReviewStack](https://reviewstack.dev/facebook/react/pull/33047). * #32765 * #32747 * __->__ #33047 DiffTrain build for [12f4cb8](12f4cb8)
…ants
```js
function Component() {
useEffect(() => {
let hasCleanedUp = false;
document.addEventListener(..., () => hasCleanedUp ? foo() : bar());
// effect return values shouldn't be typed as frozen
return () => {
hasCleanedUp = true;
}
};
}
```
### Problem
`PruneHoistedContexts` currently strips hoisted declarations and rewrites the first `StoreContext` reassignment to a declaration. For example, in the following example, instruction 0 is removed while a synthetic `DeclareContext let` is inserted before instruction 1.
```js
// source
const cb = () => x; // reference that causes x to be hoisted
let x = 4;
x = 5;
// React Compiler IR
[0] DeclareContext HoistedLet 'x'
...
[1] StoreContext reassign 'x' = 4
[2] StoreContext reassign 'x' = 5
```
Currently, we don't account for `DeclareContext let`. As a result, we're rewriting to insert duplicate declarations.
```js
// source
const cb = () => x; // reference that causes x to be hoisted
let x;
x = 5;
// React Compiler IR
[0] DeclareContext HoistedLet 'x'
...
[1] DeclareContext Let 'x'
[2] StoreContext reassign 'x' = 5
```
### Solution
Instead of always lowering context variables to a DeclareContext followed by a StoreContext reassign, we can keep `kind: 'Const' | 'Let' | 'Reassign' | etc` on StoreContext.
Pros:
- retain more information in HIR, so we can codegen easily `const` and `let` context variable declarations back
- pruning hoisted `DeclareContext` instructions is simple.
Cons:
- passes are more verbose as we need to check for both `DeclareContext` and `StoreContext` declarations
~(note: also see alternative implementation in #32745
### Testing
Context variables are tricky. I synced and diffed changes in a large meta codebase and feel pretty confident about landing this. About 0.01% of compiled files changed. Among these changes, ~25% were [direct bugfixes](https://www.internalfb.com/phabricator/paste/view/P1800029094). The [other changes](https://www.internalfb.com/phabricator/paste/view/P1800028575) were primarily due to changed (corrected) mutable ranges from #33047. I tried to represent most interesting changes in new test fixtures
`
github-actions bot
pushed a commit
that referenced
this pull request
Apr 30, 2025
…ants (#32747) ```js function Component() { useEffect(() => { let hasCleanedUp = false; document.addEventListener(..., () => hasCleanedUp ? foo() : bar()); // effect return values shouldn't be typed as frozen return () => { hasCleanedUp = true; } }; } ``` ### Problem `PruneHoistedContexts` currently strips hoisted declarations and rewrites the first `StoreContext` reassignment to a declaration. For example, in the following example, instruction 0 is removed while a synthetic `DeclareContext let` is inserted before instruction 1. ```js // source const cb = () => x; // reference that causes x to be hoisted let x = 4; x = 5; // React Compiler IR [0] DeclareContext HoistedLet 'x' ... [1] StoreContext reassign 'x' = 4 [2] StoreContext reassign 'x' = 5 ``` Currently, we don't account for `DeclareContext let`. As a result, we're rewriting to insert duplicate declarations. ```js // source const cb = () => x; // reference that causes x to be hoisted let x; x = 5; // React Compiler IR [0] DeclareContext HoistedLet 'x' ... [1] DeclareContext Let 'x' [2] StoreContext reassign 'x' = 5 ``` ### Solution Instead of always lowering context variables to a DeclareContext followed by a StoreContext reassign, we can keep `kind: 'Const' | 'Let' | 'Reassign' | etc` on StoreContext. Pros: - retain more information in HIR, so we can codegen easily `const` and `let` context variable declarations back - pruning hoisted `DeclareContext` instructions is simple. Cons: - passes are more verbose as we need to check for both `DeclareContext` and `StoreContext` declarations ~(note: also see alternative implementation in #32745 ### Testing Context variables are tricky. I synced and diffed changes in a large meta codebase and feel pretty confident about landing this. About 0.01% of compiled files changed. Among these changes, ~25% were [direct bugfixes](https://www.internalfb.com/phabricator/paste/view/P1800029094). The [other changes](https://www.internalfb.com/phabricator/paste/view/P1800028575) were primarily due to changed (corrected) mutable ranges from #33047. I tried to represent most interesting changes in new test fixtures ` DiffTrain build for [9d795d3](9d795d3)
This was referenced May 1, 2025
This was referenced May 9, 2025
This was referenced May 19, 2025
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.
Problem
PruneHoistedContextscurrently strips hoisted declarations and rewrites the firstStoreContextreassignment to a declaration. For example, in the following example, instruction 0 is removed while a syntheticDeclareContext letis inserted before instruction 1.Currently, we don't account for
DeclareContext let. As a result, we're rewriting to insert duplicate declarations.Solution
Instead of always lowering context variables to a DeclareContext followed by a StoreContext reassign, we can keep
kind: 'Const' | 'Let' | 'Reassign' | etcon StoreContext.Pros:
constandletcontext variable declarations backDeclareContextinstructions is simple.Cons:
DeclareContextandStoreContextdeclarations(note: also see alternative implementation in #32745)Testing
Context variables are tricky. I synced and diffed changes in a large meta codebase and feel pretty confident about landing this. About 0.01% of compiled files changed. Among these changes, ~25% were direct bugfixes. The other changes were primarily due to changed (corrected) mutable ranges from #33047. I tried to represent most interesting changes in new test fixtures
`