If someone can look at the graph above and narrow down the timeframe of the slowdown
to a handful of CLs, that would be awesome.

The graph doesn't go all the way back to CL 15800; adjusting the history parameter doesn't help either.

The bot didn't exist further back, try http://build.chromium.org/buildbot/perf/mac-
release-10.5/new-tab-ui-cold/report.html?history=1500 for some older data.

If you look on the waterfall, you'll find 2 mac bots that offer results links on some of 
their tests.

(No comment was entered for this change.)

See also See also http://crbug.com/18361 for a startup test slowdown problem.

Ben will jrg's Mom if not fixed by Beta.

(No comment was entered for this change.)

Yes, I've been looking at this.

http://codereview.chromium.org/179069

and

http://codereview.chromium.org/184003

This is a release blocker for Mac Beta.

We are deprecating the MacBeta milestone in favor of ReleaseBlock-Beta and 
a milestone.

r25232

Fixed, and this should show an improvement on the new tab page tests on all platforms.  
I'll follow up with some numbers when they're available.

The following revision refers to this bug:
    http://src.chromium.org/viewvc/chrome?view=rev&revision=25167 

------------------------------------------------------------------------
r25167 | mark@chromium.org | 2009-09-02 07:40:47 -0700 (Wed, 02 Sep 2009) | 17 lines
Changed paths:
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/cocoa/tab_controller.h?r1=25167&r2=25166
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/cocoa/tab_controller.mm?r1=25167&r2=25166
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/cocoa/tab_controller_unittest.mm?r1=25167&r2=25166
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/cocoa/tab_strip_controller.mm?r1=25167&r2=25166

Don't show favicons or throbbers for the New Tab page on the Mac.

This change nets another 35ms (15%) improvement in the duration between the
renderer requesting the NTP and it having all of the resources for the NTP
available and being completely done with layout on my Mac laptop in release
mode.

For any page which DOMUI indicates no icon should be shown, including the New
Tab page, don't show a favicon or throbber.  When the icon is removed, the
title should expand to the left to fill the void.  Compare to the behavior on
Windows and Linux.

http://groups.google.com/group/chromium-dev/browse_thread/thread/7148074f807dc5f7

BUG=13337 20378
TEST=No throbber or favicon on the New Tab page
Review URL: http://codereview.chromium.org/184003
------------------------------------------------------------------------

The following revision refers to this bug:
    http://src.chromium.org/viewvc/chrome?view=rev&revision=25232 

------------------------------------------------------------------------
r25232 | mark@chromium.org | 2009-09-02 13:32:45 -0700 (Wed, 02 Sep 2009) | 21 lines
Changed paths:
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/dom_ui/chrome_url_data_manager.cc?r1=25232&r2=25231
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/dom_ui/chrome_url_data_manager.h?r1=25232&r2=25231
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/dom_ui/dom_ui_theme_source.cc?r1=25232&r2=25231
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/dom_ui/dom_ui_theme_source.h?r1=25232&r2=25231
   M http://src.chromium.org/viewvc/chrome/trunk/src/chrome/browser/dom_ui/new_tab_ui.cc?r1=25232&r2=25231

Improve New Tab Page load performance.

This change nets another 65ms (20%) improvement in the duration between the
renderer requesting the NTP and it having all of the resources for the NTP
available and being completely done with layout on my Mac laptop in release
mode.

Move the handling of important resource requests needed for NTP layout from the
UI thread onto the IO thread.  This is handled by caching the necessary data on
the UI thread when possible, so that no profile or theme access is needed to
service the request.  There is often a lot of contention for the UI thread when
a new tab is created, and allowing these requests to be handled more directly
by the IO thread allows the renderer to lay the page out while the browser is
still busy doing UI work.

http://groups.google.com/group/chromium-dev/browse_thread/thread/7148074f807dc5f7

TEST=Do the new tab page and new incognito tab page still work?
     Are they faster?
BUG=13337
Review URL: http://codereview.chromium.org/179069
------------------------------------------------------------------------

The new tab timings didn't move at all on any of the perf test bots, which is a huge 
disappointment to me.  I'm going to leave this as "fixed" and continue insisting that the 
changes are a massive improvement because they are, based both on the snappier "feel" 
and on my own timings.  There might be something peculiar about the test 
environment, like the automation controller doing stuff on the main thread, that 
prevents this change from having a visible impact on the bots.

4.0.212.0 SVN Revision: 26685