Between these two nightly trunk builds, something regressed so that many pages now have terrible scrolling performance along with 100% CPU usage (rendering time is also affected). 2002-04-16-06/ ID: 2002041603 Good 2002-04-17-17/ ID: 2002041710 Bad Another example page: http://www.aintitcool.com/ As far as I've seen, the RC1 branch has not regressed.
Yes, I see this, and it is terrible. Someone destroyed my webpage! Bastards!
Status: UNCONFIRMED → NEW
Ever confirmed: true
If we could figure out a recipe for this slowdown, that would make it easier to track down. I'm CC'ing people who checked in changes to gecko during this time period -- if you know it's not you, feel free to un-CC. :-)
Keywords: perf, regression
Here's the link I used to check what changed to gecko during this time period: http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=SeaMonkeyAll&branch=HEAD&branchtype=match&dir=mozilla%2Flayout+mozilla%2Fcontent+mozilla%2Fview+mozilla%2Fdocshell&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=04%2F16%2F2002&maxdate=04%2F17%2F2002&cvsroot=%2Fcvsroot
looks like a dup of bug 138034?
BTW bug 138119 looks also related.
Note, this is a regression from only days ago. This needs to be restored back to it's previous performance before extra work can be done on perf in those other bugs.
I don't see this problem on Linux... are people seeing this on Win9x? Or just Win2k/NT? Ccing dcone, since he's been handling most of the "scrolling perf on NT sucks" bugs...
*** This bug has been marked as a duplicate of 102321 ***
Status: NEW → RESOLVED
Last Resolved: 16 years ago
Resolution: --- → DUPLICATE
You need to log in before you can comment on or make changes to this bug.