Closed Bug 311550 Opened 19 years ago Closed 19 years ago

[FIX]EndUpdateViewBatch sometimes does lots of extra work

Categories

(Core :: Web Painting, defect, P2)

Product:
Core
Component:
Web Painting
Platform:
x86
Linux
Type:
defect

Tracking

()

Status:
RESOLVED FIXED
Milestone:
mozilla1.9alpha1

People

(Reporter: bzbarsky, Assigned: bzbarsky)

References

Details

(Keywords: perf)

Attachments

(1 file)

Like so
992 bytes, patch
roc
: review+
roc
: superreview+
Details | Diff | Splinter Review
If an update view batch is ended and it was a DEFERRED batch, we'll post an invalidate event. We'll even do this if mHasPendingUpdates is false. This seems unfortunate -- posting an invalidate event is pretty expensive. I think we should just bail out of nsViewManager::EnableRefresh early if there are no pending updates. Failing that, perhaps we should bail out of PostInvalidateEvent early.
Priority: -- → P2
Target Milestone: --- → mozilla1.9alpha
Attached patch Like so — — Splinter Review
Attachment #198822 - Flags: superreview?(roc)
Attachment #198822 - Flags: review?(roc)
Attachment #198822 - Flags: superreview?(roc)
Attachment #198822 - Flags: superreview+
Attachment #198822 - Flags: review?(roc)
Attachment #198822 - Flags: review+
Fixed on trunk.
Status: NEW → RESOLVED
Closed: 19 years ago
Resolution: --- → FIXED
There's no way this could have caused bug 325455, is there? I may be reaching, but there aren't many trunk-only patches checked in between 20051007 and 20051008 that seemed to be likely suspects.
Possibly, I suppose, but unlikely. Is bug 325455 reproducible on non-Mac? That is, would it be possible for me to reproduce it on Linux and thus test whether this patch is the issue?
I know the bug doesn't occur on Windows, but I don't have any way of checking it on Linux.
Depends on: 325455
Depends on: 336878
Component: Layout: View Rendering → Layout: Web Painting
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: