iOS A11y memory leak #7244
iOS A11y memory leak #7244
Conversation
|
/cc @jason-simmons since you recently touched this as well |
goderbauer
left a comment
goderbauer
left a comment
There was a problem hiding this comment.
Looks good to me, but can you have one of the ObjC experts take a look at well, e.g. @cbracken?
chinmaygarde
left a comment
chinmaygarde
left a comment
There was a problem hiding this comment.
Semantics objects can be accessed from both the platform and UI threads. On the other hand all fml::WeakPtr managed objects must be created, used and destroyed on one thread. "Use" of the object includes the get() call. This makes the current usage of WeakPtr's thread unsafe.
I suggest using the std::weak_ptr instead.
Otherwise, lgtm (though I still have to convince myself of this in a debugger).
dnfield
force-pushed
the
a11y_memory_leak
branch
from
9248f90 to
4ea610d
Compare
| _parent = nil; | ||
| [_container release]; | ||
| _container = nil; | ||
| _container.get().semanticsObject = nil; |
There was a problem hiding this comment.
this really worked? maybe better:
_container.release();
There was a problem hiding this comment.
I can't remember the details now, but this was part of breaking the circular references.
There was a problem hiding this comment.
I can't remember the details now, but this was part of breaking the circular references.
5 participants
This has been split out from #6879
The current implementation of the Accessibility Bridge on iOS leads to retain cycles. This patch can be applied independently of the othe retain cycle related work, and hopefully makes this easier to review/track.