fe937bac6e7511827d503a529bdd198f7b32d6f6: Bug 710103: Update XPCShell tests for GfxInfo blocklisting r=joe a=mbrubeck
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 21:04:35 -0800 - rev 82623
Push 21688 by dsherk@mozilla.com at Thu, 15 Dec 2011 05:04:49 +0000
Bug 710103: Update XPCShell tests for GfxInfo blocklisting r=joe a=mbrubeck
dd2e6ce53715c7467022eff5be640b0a8d9564f1: Bug 710432: fix Linux GfxInfo to do feature checks r=joe
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 21:03:13 -0800 - rev 82622
Push 21688 by dsherk@mozilla.com at Thu, 15 Dec 2011 05:04:49 +0000
Bug 710432: fix Linux GfxInfo to do feature checks r=joe Also includes changes to allow Linux GfxInfo to provide a vendor and device id.
838e8168ea50ac1403f4ccbb4fd88dc7ff25c15b: Bug 689598: fix crash reporter on Android r=BenWa
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 21:03:11 -0800 - rev 82621
Push 21688 by dsherk@mozilla.com at Thu, 15 Dec 2011 05:04:49 +0000
Bug 689598: fix crash reporter on Android r=BenWa
49b8bec6d175ac0256502d4e1aed40665c1fcaf3: Bug 706739: don't evaluate static blocklist or special cases when evaluating downloaded blocklist r=joe
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 21:03:08 -0800 - rev 82620
Push 21688 by dsherk@mozilla.com at Thu, 15 Dec 2011 05:04:49 +0000
Bug 706739: don't evaluate static blocklist or special cases when evaluating downloaded blocklist r=joe A bug was introduced in 668004 which caused features to be preffed off while evaluating the downloaded blocklist, even if they weren't in the list. This shouldn't actually have any impact on the end user experience, but is definitely not optimal.
8f7893e1c20c865f51f2f2688a91e4d2c0690b34: Bug 705959: move some special cases into global blocklists r=joe
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 21:03:06 -0800 - rev 82619
Push 21688 by dsherk@mozilla.com at Thu, 15 Dec 2011 05:04:49 +0000
Bug 705959: move some special cases into global blocklists r=joe Some of the special cases in GfxInfo classes were only necessary because there was no generalized blocklist. These have been moved into the generalized blocklist for each GfxInfo.
b9c1b8afb35a534edefafb3a04eb8222cf72a789: Bug 689598: implement Android gfx blocklisting r=joe
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 21:03:03 -0800 - rev 82618
Push 21688 by dsherk@mozilla.com at Thu, 15 Dec 2011 05:04:49 +0000
Bug 689598: implement Android gfx blocklisting r=joe Also includes changes to support using the "driver version" field as the Android SDK version.
8c075fee9be499590b5ee1d4deac3d50c8deada0: Bug 704710: refactor GfxDriverInfo/GfxInfo(Base) to support string-based vendor and device id r=joe
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 21:03:01 -0800 - rev 82617
Push 21688 by dsherk@mozilla.com at Thu, 15 Dec 2011 05:04:49 +0000
Bug 704710: refactor GfxDriverInfo/GfxInfo(Base) to support string-based vendor and device id r=joe
f9203039a956038bb6341042398c21c8426d570e: Bug 699482: refactored GfxDriverInfo init to happen after global init r=bjacob
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 21:02:59 -0800 - rev 82616
Push 21688 by dsherk@mozilla.com at Thu, 15 Dec 2011 05:04:49 +0000
Bug 699482: refactored GfxDriverInfo init to happen after global init r=bjacob This patch moves all static initialization of GfxDriverInfo and DeviceFamily classes to the point that they're actually used. It also converts all static GfxDriverInfo arrays into nsTArray<GfxDriverInfo> so that they can be used interchangeably with the downloadable blocklist. This patch also introduces a new phase of blocklist checking called BEING_PROCESSED, which is the status set when a blocklist check is currently being processed. NO_INFO now only means that we have confirmed that a device is not blocked.
ae42e4497ff2129bd73a33d4316637a0b55961a0: Silence unused-value warnings with various Valgrind macros in clang, and add a silencing instance to jsgc.cpp. No bug, rs=jorendorff, a=philor
Jeff Walden <jwalden@mit.edu> - Tue, 13 Dec 2011 17:31:08 -0500 - rev 82615
Push 21687 by bzbarsky@mozilla.com at Thu, 15 Dec 2011 04:44:47 +0000
Silence unused-value warnings with various Valgrind macros in clang, and add a silencing instance to jsgc.cpp. No bug, rs=jorendorff, a=philor
ba1d8b3a53e493de189974d450fb5c543d7fb27d: Bug 511909. Allow @-rules to nest when parsing CSS. In particular, allow them inside @media and @-moz-document. r=dbaron
Jet Villegas <jet@mozilla.com> - Wed, 14 Dec 2011 23:42:15 -0500 - rev 82614
Push 21687 by bzbarsky@mozilla.com at Thu, 15 Dec 2011 04:44:47 +0000
Bug 511909. Allow @-rules to nest when parsing CSS. In particular, allow them inside @media and @-moz-document. r=dbaron
a3f62505cd16c5f4633bb3a0388ba1ac33e6aa06: Bug 709483. Fix off-by-one error in the call to memmove that could cause us to copy memory we didn't own. r=bzbarsky
Christian Holler <choller@mozilla.com> - Wed, 14 Dec 2011 23:42:15 -0500 - rev 82613
Push 21687 by bzbarsky@mozilla.com at Thu, 15 Dec 2011 04:44:47 +0000
Bug 709483. Fix off-by-one error in the call to memmove that could cause us to copy memory we didn't own. r=bzbarsky
a41a3d8816000594e36b2098a9cba316ea51998c: Bug 709256 part 5. Fast-path nsAnimationManager::DispatchEvents when there are no events. r=dbaron
Boris Zbarsky <bzbarsky@mit.edu> - Wed, 14 Dec 2011 23:42:15 -0500 - rev 82612
Push 21687 by bzbarsky@mozilla.com at Thu, 15 Dec 2011 04:44:47 +0000
Bug 709256 part 5. Fast-path nsAnimationManager::DispatchEvents when there are no events. r=dbaron
43c603bd01637c8d138466aaa0ec5898cdff97ed: Bug 709256 part 4. Don't flush the sink for HTML if we've already started layout. r=smaug
Boris Zbarsky <bzbarsky@mit.edu> - Wed, 14 Dec 2011 23:42:15 -0500 - rev 82611
Push 21687 by bzbarsky@mozilla.com at Thu, 15 Dec 2011 04:44:47 +0000
Bug 709256 part 4. Don't flush the sink for HTML if we've already started layout. r=smaug
cdc587a042ae1342b9e472133e62552e942bb530: Bug 709256 part 3. Skip calling PresShell::FlushPendingNotifications altogether if there might not be anything to flush. r=roc
Boris Zbarsky <bzbarsky@mit.edu> - Wed, 14 Dec 2011 23:42:15 -0500 - rev 82610
Push 21687 by bzbarsky@mozilla.com at Thu, 15 Dec 2011 04:44:47 +0000
Bug 709256 part 3. Skip calling PresShell::FlushPendingNotifications altogether if there might not be anything to flush. r=roc
d79786b469512453e00b41b99a5ad9763d02c62b: Bug 709256 part 2. Short-circuit PresShell::ProcessReflowCommands when there aren't any. r=roc
Boris Zbarsky <bzbarsky@mit.edu> - Wed, 14 Dec 2011 23:42:14 -0500 - rev 82609
Push 21687 by bzbarsky@mozilla.com at Thu, 15 Dec 2011 04:44:47 +0000
Bug 709256 part 2. Short-circuit PresShell::ProcessReflowCommands when there aren't any. r=roc
f586cb3fa70d5662cf58e482041632eaec2d40d4: Bug 709256 part 1. Fast-path RestyleTracker::ProcessRestyles when there aren't any restyles to process. r=roc
Boris Zbarsky <bzbarsky@mit.edu> - Wed, 14 Dec 2011 23:42:14 -0500 - rev 82608
Push 21687 by bzbarsky@mozilla.com at Thu, 15 Dec 2011 04:44:47 +0000
Bug 709256 part 1. Fast-path RestyleTracker::ProcessRestyles when there aren't any restyles to process. r=roc
0afa7a488c75ead57a4fdb79ade397b06da1a2d5: Bug 710956 - Disable mobile/xul/chrome/tests/browser_addons.js. a=me
Matt Brubeck <mbrubeck@mozilla.com> - Wed, 14 Dec 2011 20:33:19 -0800 - rev 82607
Push 21686 by mbrubeck@mozilla.com at Thu, 15 Dec 2011 04:34:01 +0000
Bug 710956 - Disable mobile/xul/chrome/tests/browser_addons.js. a=me
243ff1422033a5cac674f83c4bdd7718692f1d7a: Bug 709531 - Addons Manager tests should have AMO-related prefs reset after each test to disallow network access. r=dtownsend a=mbrubeck
Blair McBride <bmcbride@mozilla.com> - Thu, 15 Dec 2011 14:00:19 +1300 - rev 82606
Push 21685 by bmcbride@mozilla.com at Thu, 15 Dec 2011 01:02:23 +0000
Bug 709531 - Addons Manager tests should have AMO-related prefs reset after each test to disallow network access. r=dtownsend a=mbrubeck
3404df37a347ce0eea40d1b161ec71e3b564214d: Bug 707861: implement EXT_robustness for ANGLE r=bjacob a=mbrubeck
Doug Sherk <dsherk@mozilla.com> - Wed, 14 Dec 2011 16:57:09 -0800 - rev 82605
Push 21684 by dsherk@mozilla.com at Thu, 15 Dec 2011 00:57:28 +0000
Bug 707861: implement EXT_robustness for ANGLE r=bjacob a=mbrubeck Similar to ARB_robustness, uses our current robustness framework. There are some problems preventing this from being completely usable, but we can't do anything about them yet. See the bug for more info.
472b4a4ebea7b1c73fd8c9ec15d07a805acc068d: Bug 709162 - crash test for cycle collector dumping. rs=bholley a=mbrubeck
Andrew McCreight <amccreight@mozilla.com> - Wed, 14 Dec 2011 16:02:12 -0800 - rev 82604
Push 21683 by amccreight@mozilla.com at Thu, 15 Dec 2011 00:03:24 +0000
Bug 709162 - crash test for cycle collector dumping. rs=bholley a=mbrubeck
(0) -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 +300000 tip