mozilla/testing/performance/talos PerfConfigurator.py bcontroller.py devicemanager.py ffprocess.py ffprocess_linux.py ffprocess_mac.py ffprocess_win32.py ffprocess_winmo.py ffsetup.py run_tests.py tte

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=PerfConfigurator.py:bcontroller.py:devicemanager.py:ffprocess.py:ffprocess_linux.py:ffprocess_mac.py:ffprocess_win32.py:ffprocess_winmo.py:ffsetup.py:run_tests.py:ttest.py:winmo.config&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv14467 Modified Files: PerfConfigurator.py bcontroller.py devicemanager.py ffprocess.py ffprocess_linux.py ffprocess_mac.py ffprocess_win32.p

mozilla/testing/performance/talos run_tests.py post_file.py

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=run_tests.py:post_file.py&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv13225 Modified Files: run_tests.py post_file.py Log Message: bug 548320 (winnt 5.1/6.1 have trouble pinging the graph server) use HTTP instead of HTTPConnection p=anodelman r=bhearsum

Current tree status

The current tree status is complicated: * The Windows bustage in nsAnnotationService has been solved by booting the Windows build slaves with /3GB. We got nightlies on all platforms this morning. * Soon after, I reopened the tree for metered checkins and the first major checkin broke the tree because of missing files and other errors. That checkin has been backed out. * Then, tinderbox.mozilla.org died (disk controller is failing). This is being tracked in bug 546470, assigned to justdave. The current tinderbox.mozilla.org is actually tinderbox-stage, and it apparently has

mozilla/testing/performance/talos sample.config

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=sample.config&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv19741 Modified Files: sample.config Log Message: bug 539135 (Tracking bug for scheduled downtime (14-jan-2010)) bustage fix, disable collection of Xres counter, p/r=anodelman

mozilla/testing/performance/talos cmanager_linux.py cmanager_win32.py ffprocess_linux.py

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=cmanager_linux.py:cmanager_win32.py:ffprocess_linux.py&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv27929 Modified Files: cmanager_linux.py cmanager_win32.py ffprocess_linux.py Log Message: bug 519568 (Create new Talos suite: metrics for plugins) re-checkin backed out patch p/r=anodelman

mozilla/testing/performance/talos bcontroller.py

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=bcontroller.py&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv30202 Modified Files: bcontroller.py Log Message: bug 537526 (clearer talos message for non-zero browser return code error) p=anodelman r=catlee

mozilla/testing/performance/talos bcontroller.py ffprocess.py ffprocess_linux.py ffprocess_mac.py ffprocess_win32.py ffsetup.py run_tests.py ttest.py

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=bcontroller.py:ffprocess.py:ffprocess_linux.py:ffprocess_mac.py:ffprocess_win32.py:ffsetup.py:run_tests.py:ttest.py&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv27837 Modified Files: bcontroller.py ffprocess.py ffprocess_linux.py ffprocess_mac.py ffprocess_win32.py ffsetup.py run_tests.py ttest.py Log Message: bug 474478 (set up talos on winmo) Migrate python code

everythingelse

If there are no objections, I'm going to be splitting up the 'everythingelse' set of unit tests into the following: mochitest-other will consist of: mochitest-plain-1..5, mochitest-chrome, mochitest-browser-chrome, mochitest-a11y, mochitest-ipcplugins reftest, crashtest, xpcshell, jsreftest will each run as a separate job. This should result in a more even distribution of time spent per test job. Currently the 'everythingelse' job takes the longest of all the unit test jobs. See https://bugzilla.mozilla.org/show_bug.cgi?id=452861 for details. Cheers, Chris

Recalibrating Talos performance results on newer hardware

If you don't care about Talos performance results, or Talos hardware, stop reading now! In the last dev platform meeting, I mentioned we'd be doing a lot of Talos recalibration work in early January 2010, and promised updates as we figured out the plan. We're still working out some details, but here's what we know so far. Talos runs performance tests on mac mini hardware. The 1.83GHz minis we use today were all spec'd 2+ years ago, when we migrated from the previous 1.66GHz minis. During 2009, we kept buying and powering up more 1.83GHz minis to keep up with increasing demand of

mozilla/testing/performance/talos cmanager_linux.py

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=cmanager_linux.py&root=/cvsroot Changes committed by lsblakk%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv6572 Modified Files: cmanager_linux.py Log Message: Bug 531690 - fix for invalid literal for float in tp4_xres p=lsblakk r=anodelman

mozilla/testing/performance/talos bcontroller.py ffprocess.py ffprocess_linux.py ffprocess_mac.py ffprocess_win32.py ffsetup.py run_tests.py ttest.py

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=bcontroller.py:ffprocess.py:ffprocess_linux.py:ffprocess_mac.py:ffprocess_win32.py:ffsetup.py:run_tests.py:ttest.py&root=/cvsroot Changes committed by asasaki%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv28005 Modified Files: bcontroller.py ffprocess.py ffprocess_linux.py ffprocess_mac.py ffprocess_win32.py ffsetup.py run_tests.py ttest.py Log Message: talos -> classes, for eventual winmo integration. bug 474

Making unexpected assertions turn the tree orange (for reftest/crashtest/jsreftest)

[ Cross-posted to dev-planning and dev-tree-management. Please follow up to dev-tree-management. ] A while back, I added code to the reftest harness to report test failures for unexpected assertions. I'm planning to turn this on on mozilla-central in the next week or two. This will significantly expand our test coverage for assertion failures, which are currently tested for only on the very simple leak test that the "L" builds do, and, I think, in C++ unit tests and xpcshell tests. This will help us catch regressions that cause new assertions, which are both annoying in thems

merging *-Unittest trees back into main trees?

The split of unit test boxes onto other tinderbox trees (Firefox vs. Firefox-Unittest, etc.) was originally done on the assumption that people were using the tinderboxpushlog UI, and therefore that it didn't matter whether there was one tree or many for a given branch. This assumption has proven to be false in a number of cases; a bunch of people still use the tinderbox UI, and therefore don't notice unittest orange. I think we should merge the *-Unittest trees back into the main trees so that everybody is aware of all the machines running for that branch. I think having the UI b

consistent crash in mochitest-browser-chrome on debug boxes, after tracemonkey merge

It looks like the debug everythingelse tinderboxes have been perma-orange with a crash in mochitest-browser-chrome, in every cycle since today's tracemonkey merge. I filed this bug on the issue: https://bugzilla.mozilla.org/show_bug.cgi?id=532341 The #developers channel is pretty quiet & unresponsive at the moment, so I figured I'd send this message to m.d.tree-management, to make sure this is on our radar... I imagine we need either a bustage fix or a backout, ASAP. (though the backout could be more trouble than usual, given that the guilty changeset is a huge tracemonkey

about all the Leopard improvements.

If you have been following along with * https://bugzilla.mozilla.org/show_bug.cgi?id=519893 (CGContextFillPath sometimes takes over 100ms on Talos, Talos machines should behave as if there was a screen connected ) you'd know that resistors were added to all the leopard talos boxes to simulate having a real monitor attached this afternoon. That explains the shift in the numbers. This probably should have occurred during a downtime and I hope that it hasn't caused too much confusion. alice.

RelEng Downtime Thursday morning

We're going to have a quick downtime tomorrow morning to restart one of our buildbot masters, and land some changes. This will close the mozilla-1.9.1, tracemonkey and electrolysis trees, as well as mobile builds on mozilla-1.9.2 and mozilla-central.

mozilla-central to be closed to Firefox 3.6/mozilla-1.9.2 blockers until code freeze

Hey everyone, As discussed at yesterday's development meeting, we'll be restricting the mozilla-central tree to: - fixes for Firefox 3.6 or mozilla-1.9.2 blockers [1], - patches marked with approval1.9.2+ [2], - patches that are not part of the Firefox build (including tests) This is to keep the tree as quiet as possible so that our final blockers can land for the Firefox 3.6 Release Candidate code freeze which was originally scheduled for end of day today. Based on estimates from yesterday, we expect the freeze will slip to the end of day tomorrow, November 19th. If

Regressions

This is a multi-part message in MIME format. --------------090905070201060403040409 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit How are we supposed to be handling regressions nowadays? We've been pretty good about commenting in the bug about a regression, and saying that that was done in this forum, but then we often seem to see no other information about it. Should we just start backing things out when a regression looks real, and go from there? Or something else? Cheers, Shawn --------------090905070201060403040409

mozilla/testing/performance/talos sample.config ttest.py run_tests.py

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=sample.config:ttest.py:run_tests.py&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv4722 Modified Files: sample.config ttest.py run_tests.py Log Message: bug 419776 (Create new Talos suite: talos should measure x resources) p=anodelman r=lsblakk

Talos downtime Friday morning (9am-10am eastern time)

I'm going to be landing some changes to our Talos infrastructure tomorrow morning. It's mostly code reorganization, but it will also enable us to automate the running of Talos on release builds. All trees will be closed while this is going on, but it should be nice and quick. https://bugzilla.mozilla.org/show_bug.cgi?id=517212

mozilla/testing/performance/talos sample.config

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos&files=sample.config&root=/cvsroot Changes committed by lsblakk%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos In directory dm-cvs01:/tmp/cvs-serv971 Modified Files: sample.config Log Message: Bug 506772 - bustage fix, adds v8 to sample.config p=lsblakk r=anodelman

mozilla/testing/performance/talos/page_load_test/svg_opacity big-optimizable-group-opacity-2500.svg small-group-opacity-2500.svg svg_opacity.manifest

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos/page_load_test/svg_opacity&files=big-optimizable-group-opacity-2500.svg:small-group-opacity-2500.svg:svg_opacity.manifest&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos/page_load_test/svg_opacity In directory dm-cvs01:/tmp/cvs-serv19360/page_load_test/svg_opacity Added Files: big-optimizable-group-opacity-2500.svg small-group-opacity-2500.svg svg_opacity.manifest Log Message: Bug 524089 (Create new Talos suite

mozilla/testing/performance/talos/breakpad/win32 minidump_stackwalk.exe

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos/breakpad/win32&files=minidump_stackwalk.exe&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos/breakpad/win32 In directory dm-cvs01:/tmp/cvs-serv17828/win32 Modified Files: minidump_stackwalk.exe Log Message: bug 526026 (update copies of minidump_stackwalk used by talos) p=anodelman r=ted

mozilla/testing/performance/talos/breakpad/linux minidump_stackwalk

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos/breakpad/linux&files=minidump_stackwalk&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos/breakpad/linux In directory dm-cvs01:/tmp/cvs-serv17828/linux Modified Files: minidump_stackwalk Log Message: bug 526026 (update copies of minidump_stackwalk used by talos) p=anodelman r=ted

Lots of windows build slaves and XP talos slaves went missing overnight

Hi folks, I noticed that we were missing nearly all of our Windows build slaves this morning, and most of the XP Talos slaves as well. For the past 12 hours or so it's possible that checkins to all branches (except try) were subject to reduced coverage - slow results, more checkins clustered together. Tracking for the acute issue was in https://bugzilla.mozilla.org/show_bug.cgi?id=527229. The overall problem is https://bugzilla.mozilla.org/show_bug.cgi?id=522078. - Ben

mozilla/testing/performance/talos/page_load_test/svg_opacity - New directory

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos/page_load_test/svg_opacity&files=-:New:directory&root=/cvsroot Changes committed by anodelman%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos/page_load_test/svg_opacity In directory dm-cvs01:/tmp/cvs-serv16405/svg_opacity Log Message: Directory /cvsroot/mozilla/testing/performance/talos/page_load_test/svg_opacity added to the repository

mozilla/testing/performance/talos/startup_test fennecmark.html

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos/startup_test&files=fennecmark.html&root=/cvsroot Changes committed by asasaki%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos/startup_test In directory dm-cvs01:/tmp/cvs-serv15513/startup_test Added Files: fennecmark.html Log Message: update fennecmark (tpan, tzoom) in talos. bug 510325, r=alice

mozilla/testing/performance/talos/tpan - New directory

Diffs: http://bonsai.mozilla.org/cvsview2.cgi?command=DIRECTORY&subdir=mozilla/testing/performance/talos/tpan&files=-:New:directory&root=/cvsroot Changes committed by asasaki%mozilla.com: Update of /cvsroot/mozilla/testing/performance/talos/tpan In directory dm-cvs01:/tmp/cvs-serv21692/tpan Log Message: Directory /cvsroot/mozilla/testing/performance/talos/tpan added to the repository

Mozilla Scheduled Downtime October 30, 2009 7am EDT

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hey all, RelEng needs to do a quick stop/start on production-master tomorrow morning in order to resurrect the l10n repack-on-change schedulers: https://bugzilla.mozilla.org/show_bug.cgi?id=524798 Downtime details: Friday, October 30, 2009 7am EDT (4am PDT) It should be a quick op, but any builds in progress on the following branches will be lost: * mozilla-central * mozilla-1.9.2 * places If any nightlies get lost as a result of this, I will re-trigger a new set. Please let me know ASAP if there is any reason we can

Talos Improvement: DHTML decrease 0.35% on Vista Firefox

--===============0749089497== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Improvement: DHTML decrease 0.35% on Vista Firefox Previous results: 694.176 from build 20091026200307 of revision c0a5b63c3f3d at 2009-10-26 21:38:00 on talos-rev2-vista04 run # 0 New results: 691.765 from build 20091026202704 of revision 556728346e52 at 2009-10-26 22:18:00 on talos-rev2-vista07 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":67,"test":18,"branch":1},{"machine":68,"test":18,"branch":1},{"machine":69,

Talos Regression: Tp4 (Private Bytes) increase 11.21% on XP Firefox3.5

--===============1643174771== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Tp4 (Private Bytes) increase 11.21% on XP Firefox3.5 Previous results: 76119500.0 from build 20091023113436 of revision 2b9f6f5f5d18 at 2009-10-23 13:24:00 on talos-rev2-xp04 run # 0 New results: 84655500.0 from build 20091023160335 of revision 0be7c0bfe443 at 2009-10-23 17:45:00 on talos-rev2-xp18 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":108,"test":39,"branch":3},{"machine":109,"test":39,"branch":3

Talos Regression: Tp4 (RSS) increase 4.14% on Tiger Firefox3.5

--===============0142218372== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Tp4 (RSS) increase 4.14% on Tiger Firefox3.5 Previous results: 107256000.0 from build 20091023104909 of revision 2b9f6f5f5d18 at 2009-10-23 11:00:00 on talos-rev2-tiger05 run # 0 New results: 111695000.0 from build 20091023152309 of revision 0be7c0bfe443 at 2009-10-23 15:32:00 on talos-rev2-tiger12 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":148,"test":41,"branch":3},{"machine":149,"test":41,"branch":3

Talos Improvement: Tp4 decrease 1.24% on Linux Firefox3.5

--===============0903447569== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Improvement: Tp4 decrease 1.24% on Linux Firefox3.5 Previous results: 642.475 from build 20091023105242 of revision 2b9f6f5f5d18 at 2009-10-23 11:41:00 on talos-rev2-linux10 run # 0 New results: 634.495 from build 20091023153055 of revision 0be7c0bfe443 at 2009-10-23 16:28:00 on talos-rev2-linux16 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":188,"test":38,"branch":3},{"machine":189,"test":38,"branch":3},{"machine":

Talos Regression: Tp4 (RSS) increase 7.75% on Linux Firefox3.5

--===============0177196270== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Tp4 (RSS) increase 7.75% on Linux Firefox3.5 Previous results: 89161400.0 from build 20091023105242 of revision 2b9f6f5f5d18 at 2009-10-23 11:41:00 on talos-rev2-linux10 run # 0 New results: 96069900.0 from build 20091023153055 of revision 0be7c0bfe443 at 2009-10-23 16:28:00 on talos-rev2-linux16 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":188,"test":41,"branch":3},{"machine":189,"test":41,"branch":3},

Talos Improvement: Ts, MIN Dirty Profile decrease 0.40% on XP Firefox

--===============0654961481== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Improvement: Ts, MIN Dirty Profile decrease 0.40% on XP Firefox Previous results: 444.053 from build 20091023145351 of revision be5eeaf56443 at 2009-10-23 16:50:00 on talos-rev2-xp09 run # 0 New results: 442.263 from build 20091023153135 of revision a75d7681fa2f at 2009-10-23 17:27:00 on talos-rev2-xp11 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":108,"test":52,"branch":1},{"machine":109,"test":52,"branch":1},{"mac

state of the tree (after mochitest splitting)

A few notes on the state of the tree after today's infrastructure changes: The major change that happened today was that mochitest-plain has been split into multiple runs. Rather than running all the mochitests at once, they're run in five separate groups. This change uncovered a number of bugs (one of which, the last one described below, is still causing some orange). A bunch of fixes went in in this interval: http://hg.mozilla.org/mozilla-central/pushloghtml?startdate=2009-10-22+00%3A00%3A00&enddate=2009-10-22+13%3A30 The one of these fixed problems that I know something abo

Determining the null distribution of talos numbers

Some time ago I filed a bug asking for more structure in the way talos is run on the try tree, so that the numeric results would have more statistical power to them, and one could actually draw conclusions about the performance impact of a patch. That bug never went anywhere, but I've been thinking about it since, and it occurs to me that to make coherent recommendations about improvements to talos - both on try and on real branches - we ought to take a good hard look at the null distribution. The problem as is, after all, is that it *appears* there is so much variation due to chanc

Talos Regression: Txul increase 0.52% on Linux Firefox

--===============0020551154== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Txul increase 0.52% on Linux Firefox Previous results: 112.211 from build 20091021173252 of revision 2b9f7c28c481 at 2009-10-21 18:09:00 on talos-rev2-linux14 run # 0 New results: 112.789 from build 20091021183511 of revision e4a200db5497 at 2009-10-21 18:51:00 on talos-rev2-linux05 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":51,"test":17,"branch":1},{"machine":52,"test":17,"branch":1},{"machine":53,"t

Talos Regression: DHTML increase 38.75% on Leopard Firefox

--===============0441269697== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: DHTML increase 38.75% on Leopard Firefox Previous results: 790.382 from build 20091020203630 of revision 9d4bef551670 at 2009-10-20 22:13:00 on talos-rev2-leopard14 run # 0 New results: 1096.68 from build 20091021001423 of revision 766d68bdf639 at 2009-10-21 01:55:00 on talos-rev2-leopard05 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":63,"test":18,"branch":1},{"machine":64,"test":18,"branch":1},{"machin

Talos Regression: DHTML increase 0.24% on XP Firefox

--===============1444637306== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: DHTML increase 0.24% on XP Firefox Previous results: 1027.62 from build 20091020134532 of revision 9fa60d07b156 at 2009-10-20 15:30:00 on talos-rev2-xp13 run # 0 New results: 1030.09 from build 20091020162058 of revision cdebeb90dbd8 at 2009-10-20 18:05:00 on talos-rev2-xp14 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":55,"test":18,"branch":1},{"machine":56,"test":18,"branch":1},{"machine":57,"test":18,

Talos Regression: Txul increase 2.45% on XP Firefox

--===============1864014794== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Txul increase 2.45% on XP Firefox Previous results: 94.6316 from build 20091020134532 of revision 9fa60d07b156 at 2009-10-20 15:30:00 on talos-rev2-xp20 run # 0 New results: 96.9474 from build 20091020162058 of revision cdebeb90dbd8 at 2009-10-20 18:05:00 on talos-rev2-xp08 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":55,"test":17,"branch":1},{"machine":56,"test":17,"branch":1},{"machine":57,"test":17,"

split mochitests landing on Thursday

We're going to be trying to land the split mochitests patch on trunk again this Thursday morning. This change splits up the mochitest test suite into 5 pieces, based on which directory the test is in. Each of the 5 pieces can be run in parallel on different slaves. The aim is to bring the wallclock time for mochitests down from around 50 minutes to somewhere around 15 minutes. These results will be reported to the Firefox-Unittest Tinderbox tree, and will appear as, e.g. WINNT 5.2 mozilla-central test opt mochitests-1/5 Last time this was landed, several tests started faili

Please ignore the GECKO192b1_20091015_RELBRANCH in mozilla-1.9.2

Hi folks, Due to a screw-up, I accidentally created a second release branch for 3.6b1/1.9.2b1. I've moved all the tags off of it, back to their rightful place, but there's no way to get rid of it completely, so please ignore it. - Ben

Talos Regression: Ts, MIN Dirty Profile increase 6.34% on XP Firefox

--===============1169305160== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Ts, MIN Dirty Profile increase 6.34% on XP Firefox Previous results: 439.211 from build 20091017072608 of revision 5ae7b09f730f at 2009-10-17 09:13:00 on talos-rev2-xp08 run # 0 New results: 467.053 from build 20091017065100 of revision d2ac46221368 at 2009-10-17 08:25:00 on talos-rev2-xp20 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":108,"test":52,"branch":1},{"machine":109,"test":52,"branch":1},{"mach

Talos Regression: Ts, MED Dirty Profile increase 6.00% on XP Firefox

--===============1272883331== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Ts, MED Dirty Profile increase 6.00% on XP Firefox Previous results: 439.526 from build 20091017072608 of revision 5ae7b09f730f at 2009-10-17 09:13:00 on talos-rev2-xp08 run # 0 New results: 465.895 from build 20091017065100 of revision d2ac46221368 at 2009-10-17 08:25:00 on talos-rev2-xp20 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":108,"test":53,"branch":1},{"machine":109,"test":53,"branch":1},{"mach

Talos Regression: Ts, MIN Dirty Profile increase 6.34% on XP Firefox

--===============1410548832== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Ts, MIN Dirty Profile increase 6.34% on XP Firefox Previous results: 439.211 from build 20091017072608 of revision 5ae7b09f730f at 2009-10-17 09:13:00 on talos-rev2-xp08 run # 0 New results: 467.053 from build 20091017065100 of revision d2ac46221368 at 2009-10-17 08:25:00 on talos-rev2-xp20 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":108,"test":52,"branch":1},{"machine":109,"test":52,"branch":1},{"mach

Talos Regression: Ts, MED Dirty Profile increase 6.00% on XP Firefox

--===============1344106183== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Ts, MED Dirty Profile increase 6.00% on XP Firefox Previous results: 439.526 from build 20091017072608 of revision 5ae7b09f730f at 2009-10-17 09:13:00 on talos-rev2-xp08 run # 0 New results: 465.895 from build 20091017065100 of revision d2ac46221368 at 2009-10-17 08:25:00 on talos-rev2-xp20 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":108,"test":53,"branch":1},{"machine":109,"test":53,"branch":1},{"mach

Talos Regression: Ts, MED Dirty Profile increase 6.00% on XP Firefox

--===============1128396964== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Ts, MED Dirty Profile increase 6.00% on XP Firefox Previous results: 439.526 from build 20091017072608 of revision 5ae7b09f730f at 2009-10-17 09:13:00 on talos-rev2-xp08 run # 0 New results: 465.895 from build 20091017065100 of revision d2ac46221368 at 2009-10-17 08:25:00 on talos-rev2-xp20 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":108,"test":53,"branch":1},{"machine":109,"test":53,"branch":1},{"mach

Talos Regression: Ts increase 5.36% on XP Firefox

--===============0385727593== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Regression: Ts increase 5.36% on XP Firefox Previous results: 447.895 from build 20091017072608 of revision 5ae7b09f730f at 2009-10-17 09:13:00 on talos-rev2-xp14 run # 0 New results: 471.895 from build 20091017065100 of revision d2ac46221368 at 2009-10-17 08:25:00 on talos-rev2-xp04 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":55,"test":16,"branch":1},{"machine":56,"test":16,"branch":1},{"machine":57,"test":16,"br

Talos Improvement: Ts, MIN Dirty Profile decrease 2.07% on Vista Firefox

--===============1575037869== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Improvement: Ts, MIN Dirty Profile decrease 2.07% on Vista Firefox Previous results: 510.316 from build 20091016081938 of revision c045d251e69f at 2009-10-16 10:17:00 on talos-rev2-vista02 run # 0 New results: 499.737 from build 20091016092528 of revision c19711e6eef2 at 2009-10-16 11:42:00 on talos-rev2-vista14 run # 0 http://graphs.mozilla.org/graph.html#tests=[{"machine":128,"test":52,"branch":1},{"machine":129,"test":52,"branch"