Open
Bug 1310321
Opened 8 years ago
Updated 2 years ago
Remove unnecessary uses of Expose*ToActiveJS from CodeGen.py
Categories
(Core :: DOM: Core & HTML, defect, P3)
Core
DOM: Core & HTML
Tracking
()
REOPENED
People
(Reporter: mccr8, Unassigned)
References
Details
No description provided.
Reporter | ||
Comment 1•8 years ago
|
||
Although, the uses here seem to be reading out of a Rooted<> and not a Heap<> so maybe there's no read barrier?
Reporter | ||
Comment 2•8 years ago
|
||
Ok, I guess we can't really do this. I suppose in the future ideally we'd never store a gray object into a Rooted<>.
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → WONTFIX
Comment 3•8 years ago
|
||
Hmm. Which uses are you looking at? In order: 1) The one in the isAny() case in getWrapTemplateForType() is there because for any types we just had callees hand out Value and not worry about unmarking. But now that Heap unmarks, I think this can become an assert. 2) The isObject() case in the same method, same thing, both branches. 3) The isObject() case in CGEventGetter, is returning a property stored in a HeapObject, so is already getting barriered. 4) The isAny() case in CGEventGetter is returning from a HeapValue. 4)
Updated•8 years ago
|
Flags: needinfo?(continuation)
Reporter | ||
Comment 4•8 years ago
|
||
Oh, sorry, I was only looking at the invariants local to the method. It makes sense that all getters read from a Heap<> etc.
Status: RESOLVED → REOPENED
Flags: needinfo?(continuation)
Resolution: WONTFIX → ---
Comment 5•7 years ago
|
||
I'm assuming P3 here due to the lack of recent activity but feel free to correct me :)
Priority: -- → P3
Reporter | ||
Updated•7 years ago
|
Assignee: continuation → nobody
Assignee | ||
Updated•5 years ago
|
Component: DOM → DOM: Core & HTML
Updated•2 years ago
|
Severity: normal → S3
You need to log in
before you can comment on or make changes to this bug.
Description
•