Open
Bug 805769
Opened 12 years ago
Updated 2 years ago
Unify the weak maps for pause and thread-lifetime actors
Categories
(DevTools :: Debugger, task, P5)
DevTools
Debugger
Tracking
(Not tracked)
NEW
People
(Reporter: past, Unassigned)
References
(Blocks 1 open bug)
Details
Attachments
(1 file)
3.37 KB,
patch
|
Details | Diff | Splinter Review |
Now that upgrading a pause-lifetime actor to a thread-lifetime one does not alter its ID, we can unify the weak maps we use to retrieve the actors.
Reporter | ||
Comment 1•12 years ago
|
||
WIP. Breaks a few tests.
Updated•10 years ago
|
Blocks: dbg-server
Reporter | ||
Comment 2•9 years ago
|
||
I no longer have the time to work on this.
Assignee: past → nobody
Status: ASSIGNED → NEW
Updated•6 years ago
|
Product: Firefox → DevTools
Updated•5 years ago
|
Priority: P2 → P3
Comment 3•5 years ago
|
||
Logan this looks like a nice win, what do you think?
Flags: needinfo?(lsmyth)
Comment 4•5 years ago
|
||
the code has shifted, but this still looks relevant. https://searchfox.org/mozilla-central/source/devtools/server/actors/thread.js
Flags: needinfo?(poirot.alex)
Comment 5•5 years ago
|
||
TBH, I don't really know this precise code.
But generally, note that manually managed ActorPool might be simplified if we convert the actor to protocol.js Actor.
Actor are pools and you can register the child actors to the base actor, if the child actor lifetime is the same than the base actor. Otherwise, you still need a dedicated ActorPool, so that you can destroy the pool individually and keep the base actor alive.
Flags: needinfo?(poirot.alex)
Comment 6•5 years ago
|
||
Also not that familiar, but this does seem reasonable from what I've seen.
Flags: needinfo?(lsmyth)
Updated•5 years ago
|
Type: defect → task
Priority: P3 → P5
Updated•2 years ago
|
Severity: normal → S3
You need to log in
before you can comment on or make changes to this bug.
Description
•