Closed
Bug 1213464
Opened 9 years ago
Closed 9 years ago
ImageBridgeChild and CompositorChild leak mTransport
Categories
(Core :: Graphics: Layers, defect)
Core
Graphics: Layers
Tracking
()
RESOLVED
FIXED
mozilla45
People
(Reporter: mccr8, Assigned: mccr8)
References
Details
(Whiteboard: [gfx-noted])
Attachments
(1 file)
1.66 KB,
patch
|
billm
:
review+
roc
:
feedback+
|
Details | Diff | Splinter Review |
~ImageBridgeParent does some boilerplate to destroy mTransport. ImageBridgeChild probably needs to do the same thing. Of course, this is all fairly academic at the moment because we leak ImageBridgeChild itself. A similar issue likely exists for CompositorChild, or whatever the child process equivalent is of CrossProcessCompositorParent, if any.
Whiteboard: [gfx-noted]
Assignee | ||
Comment 1•9 years ago
|
||
I confirmed that this happens, and that putting the DeleteTask boilerplate in the dtor fixes the leak (in a hacked up version that runs the dtor).
Summary: ImageBridgeChild probably leaks mTransport → ImageBridgeChild leaks mTransport
Assignee | ||
Comment 2•9 years ago
|
||
Assignee | ||
Comment 3•9 years ago
|
||
Comment on attachment 8691581 [details] [diff] [review] ImageBridgeChild and CompositorChild should delete their Transport. Does this fix the remaining leaks you were seeing in bug 1227398? I hadn't bothered landing these because I thought it was somewhat questionable to land code that never runs (because these data structures both leak).
Attachment #8691581 -
Flags: feedback?(roc)
Assignee | ||
Updated•9 years ago
|
Assignee: nobody → continuation
Comment on attachment 8691581 [details] [diff] [review] ImageBridgeChild and CompositorChild should delete their Transport. Review of attachment 8691581 [details] [diff] [review]: ----------------------------------------------------------------- This looks good yes
Attachment #8691581 -
Flags: feedback?(roc) → feedback+
Assignee | ||
Updated•9 years ago
|
Attachment #8691581 -
Flags: review?(wmccloskey)
Attachment #8691581 -
Flags: review?(wmccloskey) → review+
(In reply to Andrew McCreight [:mccr8] from comment #3) > Does this fix the remaining leaks you were seeing in bug 1227398? I hadn't > bothered landing these because I thought it was somewhat questionable to > land code that never runs (because these data structures both leak). It does fix the leaks, by the way. Thanks!
Assignee | ||
Updated•9 years ago
|
No longer depends on: 1117203
Summary: ImageBridgeChild leaks mTransport → ImageBridgeChild and CompositorChild leak mTransport
Comment 8•9 years ago
|
||
bugherder |
https://hg.mozilla.org/mozilla-central/rev/22812d820754
Status: NEW → RESOLVED
Closed: 9 years ago
status-firefox45:
--- → fixed
Resolution: --- → FIXED
Target Milestone: --- → mozilla45
You need to log in
before you can comment on or make changes to this bug.
Description
•