Fix non layout #309
Merged
Fix non layout #309
Conversation
This file contains 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
The existing design is pretty fraught with error. We should probably rethink this but in the meantime, this fixes a bug where calling setNeedsLayout can start failing for nodes. Essentially the method ASDisplayNodeShouldApplyBridgedWriteToView has a side effect of registering a node to apply it's pending state *if* it doesn't currently need the pending state applied. My guess is this was to avoid continually registering the node and this behavior actually helped expose this bug. The bug: after the node is registered for flushing it's state, several code paths released the lock before applying that state to the pending state object. Before it could re-obtain the lock to apply it to the pending state, the pending state controller flushed it on the main thread. On subsequent calls to setNeedsLayout, the pending state had pending state already (from previous calls which missed the flush) and thus wasn't registered for future flushing.
Adlai-Holler
approved these changes
May 26, 2017
|
Huzzah! |
|
YES! This fixes #75 for me. :) I'll close that one. |
|
@christianselig sweet! |
bernieperez
pushed a commit
to AtomTickets/Texture
that referenced
this pull request
Apr 25, 2018
* Lock released between add to pend controller and modifying pend state The existing design is pretty fraught with error. We should probably rethink this but in the meantime, this fixes a bug where calling setNeedsLayout can start failing for nodes. Essentially the method ASDisplayNodeShouldApplyBridgedWriteToView has a side effect of registering a node to apply it's pending state *if* it doesn't currently need the pending state applied. My guess is this was to avoid continually registering the node and this behavior actually helped expose this bug. The bug: after the node is registered for flushing it's state, several code paths released the lock before applying that state to the pending state object. Before it could re-obtain the lock to apply it to the pending state, the pending state controller flushed it on the main thread. On subsequent calls to setNeedsLayout, the pending state had pending state already (from previous calls which missed the flush) and thus wasn't registered for future flushing. * Add changelog
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.
Lock released between add to pend controller and modifying pend state
The existing design is pretty fraught with error. We should probably
rethink this but in the meantime, this fixes a bug where calling
setNeedsLayout can start failing for nodes.
Essentially the method ASDisplayNodeShouldApplyBridgedWriteToView has
a side effect of registering a node to apply it's pending state if
it doesn't currently need the pending state applied. My guess is this
was to avoid continually registering the node and this behavior actually
helped expose this bug.
The bug: after the node is registered for flushing it's state, several
code paths released the lock before applying that state to the pending
state object. Before it could re-obtain the lock to apply it to the pending
state, the pending state controller flushed it on the main thread.
On subsequent calls to setNeedsLayout, the pending state had pending state
already (from previous calls which missed the flush) and thus wasn't
registered for future flushing.