User Agent: Mozilla/5.0 (Windows NT 5.1; rv:6.0.2) Gecko/20100101 Firefox/6.0.2 Build ID: 20110902133214 Steps to reproduce: 1. Right-click on the active tab, then click in the content area 2. Repeat step 1 for several times Actual results: The active tab itself is focused, showing the dotted line, but not always. Expected results: The dotted line should not appear. Bug 563591 or 679255 may be related.
WFM - Mozilla/5.0 (Windows NT 5.1; rv:9.0a1) Gecko/20110909 Firefox/9.0a1 ID:20110909030839 You posted with 6.0.2 but marked this bug on trunk...on which does it occur? Can you find the regression window? http://harthur.github.com/mozregression/
It's on trunk. I'm not familiar with how to find a regression window.
Please attach a screenshot showing the issue. I have been unable to reproduce the issue. Please also confirm if the issue occurs if you start Firefox in Safe mode and also post the graphics section from about:support (not while in safe mode)
Created attachment 559809 [details] screenshot This is a screenshot from a brand-new profile. When switching to a new tab, right-click the active tab's favicon. If not reproduced, repeat clicking in the content area and right-clicking the active tab's favicon again.
If you hold the mouse for a while, you may find the dotted line appears when mouseup, almost at the same time as tab context menu pops up. It happens in both 8.0a2 and 9.0a1, but not in 6.0. I have no 7.0b installed.
1) You should update your graphics driver - the correct one looks to be http://support.amd.com/us/gpudownload/Pages/radeon-prer300-xp.aspx 2) Still unable to reproduce on my system 3) Your best bet at getting traction is to either find another user that can reproduce (ask in support or on IRC) or find the regression range. Instructions are at http://harthur.github.com/mozregression/ on how to find the regression range.
Thanks for your guide. I have traced it back to: Last good nightly: 2011-07-02 First bad nightly: 2011-07-03 Pushlog: http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=549e3276ed25&tochange=4e3b03de1fd3
Further in mozilla-inbound. Last good nightly: 2011-06-30 First bad nightly: 2011-07-01 Pushlog: http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=c2f48684b9f5&tochange=26ecb4d6c3d0
(In reply to ithinc from comment #8) (In reply to ithinc from comment #9) Nothing in these ranges seems to stick out. I'm afraid someone will need to bisect this further by compiling Firefox.
Can you confirm this regression, Dao?
I could reproduce this a few days ago on Windows XP, but now I can't.
It seems fixed in the trunk, probably by bug 682338. Last good nightly: 2011-09-11 First bad nightly: 2011-09-12 Pushlog: http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=06b2977afb85&tochange=569a45bfb71c