Mozilla 2/StatusMeetings/2008-05-28: Difference between revisions

From MozillaWiki
Jump to navigation Jump to search
 
(One intermediate revision by one other user not shown)
Line 14: Line 14:
* dmandelin
* dmandelin
* taras
* taras
** [https://bugzilla.mozilla.org/show_bug.cgi?id=432915 bug 432915] push/pop stuff
** [https://bugzilla.mozilla.org/show_bug.cgi?id=430321 bug 430321] setting up buildbot in spare cycles
* bsmedberg
* bsmedberg
** XPCOMGC stuck on a mozilla-central merge... jorendorff do you need help with that?
** XPCOMGC stuck on a mozilla-central merge... jorendorff do you need help with that?
Line 23: Line 25:
* ignore incremental for the moment, since we can't use it with the cycle collector anyway?
* ignore incremental for the moment, since we can't use it with the cycle collector anyway?
* stats on PR_Lock taking a long time?
* stats on PR_Lock taking a long time?
Wins:
* tune GC-from-alloc
* per-thread freelists
* mochitest crash
* general profiling
* MMgc debug performance
* 64bit
* maybe valgrind


Getting jimb unstuck on TC-MMGC to work on other things...
Getting jimb unstuck on TC-MMGC to work on other things...

Latest revision as of 18:28, 28 May 2008

« previous week | index | next week »

Followups

  • opening mozilla-central for general checkins is tracking bug 422754
    • daily meetings - unit-test failures are possibly VM loading issues or some difference between different VMs - for the moment, we're moving some known-good machines from Firefox
    • bhearsum working on codesize test... some assumptions have changed
    • performance regression was a fluke... a clobber fixed it :-(
  • Consolidated post-1.9 planning wiki pages are pending. Damons is merging the work week spreadsheets into a single location tbd.

Status Reports

  • jimb
  • dmandelin
  • taras
  • bsmedberg
    • XPCOMGC stuck on a mozilla-central merge... jorendorff do you need help with that?
  • jorendorff

Discussion Topics

Actionmonkey performance measurements and MMgc fixup: status and plan of attack.

  • ignore incremental for the moment, since we can't use it with the cycle collector anyway?
  • stats on PR_Lock taking a long time?

Wins:

  • tune GC-from-alloc
  • per-thread freelists
  • mochitest crash
  • general profiling
  • MMgc debug performance
  • 64bit
  • maybe valgrind

Getting jimb unstuck on TC-MMGC to work on other things...

Purpose of this meeting (general, or just tamarin, MMgc, code rewriting, etc.)

Getting mobile into hg?