Paul's startup testing shows a consistent ~1/2 sec. increase (from the last stable time of ~19) in Windows startup time beginning around 4/18. In general, these tests seem a little fluky; numbers of have hit 20 and 21 before in the past, but since they've hovered in 21.xx sec. for a week now, I think it's time to take a look. See results in n.p.m.perf.
Er, to clarify, that's "1 or 2" seconds. Not half a second. Need to find an owner for this...
taking the bug for now. investigating startup improvement in conjuction with bug 76705
PSM2 bug 75947 is on our list of starup improvements bugs. this is definitely one aspect that's affecting starup timing, ever since it becomes part of client build.
Windows startup times are now--for the second day in a row-- over 22 seconds (to surpass 6.01 startup :-/). That's up about 3 seconds from what the 'average' time was. Anyone looking into this?
this increase was due to PSM landing in mozilla tree. See bug 75947, it's fixed, and improved 5% on startup. marking FIXED.