summaryrefslogtreecommitdiffstats
path: root/skia/ext
Commit message (Collapse)AuthorAgeFilesLines
* Remove redundant memory clears when constructing BitmapPlatformDevice ↵twiz@chromium.org2012-03-1615-78/+121
| | | | | | | | | | | | | | | instances on Mac and Windows. PlatformCanvas construction is showing up as a performance bottleneck due to unnecessary initialization. The change moves the clear to the call sites where it is necessary. Note: On Linux, cairo always allocates an initialized surface, so there is no way to bypass the performance penalty. BUG=112009 TEST=All of them Review URL: http://codereview.chromium.org/9416017 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@127196 0039d316-1c4b-4281-b951-d872f2087c98
* skia/ext: Add SK_API to a function used from a testthakis@chromium.org2012-03-161-2/+2
| | | | | | | | | BUG=90078 TEST=none Review URL: http://codereview.chromium.org/9699078 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@127068 0039d316-1c4b-4281-b951-d872f2087c98
* remoting: Move CGRectToSkIRect() to the only file where it's calledthakis@chromium.org2012-03-152-11/+0
| | | | | | | | | | | The alternative would be to add SK_API to that function. BUG=90078 TEST=none Review URL: http://codereview.chromium.org/9705064 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@127036 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 126889 - Implement input type=color UImarkusheintz@chromium.org2012-03-152-25/+0
| | | | | | | | | | | | BUG=92608 TEST= Review URL: http://codereview.chromium.org/9203001 TBR=keishi@chromium.org git-svn-id: svn://svn.chromium.org/chrome/trunk/src@126899 0039d316-1c4b-4281-b951-d872f2087c98
* Implement input type=color UIkeishi@chromium.org2012-03-152-0/+25
| | | | | | | | | | BUG=92608 TEST= Review URL: http://codereview.chromium.org/9203001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@126889 0039d316-1c4b-4281-b951-d872f2087c98
* Remove static initializers in Skia.pliard@chromium.org2012-03-132-21/+25
| | | | | | | | | | | | | | | | | | This patch does the following things: - Add missing SK_USE_POSIX_THREADS in skia.gyp. This enables the patch sent by David Turner making SkMutex POD in order to remove static initializers (on Android, Linux and Mac). - Remove static initializers in SkFontHost_fontconfig.cpp and SkMemory_new_handler.cpp. Note that the global_fc_typefaces map was populated but never read therefore I removed it. Let me know if I'm wrong. BUG=94925 Review URL: http://codereview.chromium.org/9668002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@126385 0039d316-1c4b-4281-b951-d872f2087c98
* Add _EXPORT macros to skia/ext.thakis@chromium.org2012-03-091-3/+3
| | | | | | | | | | | SkMatrixToCGAffineTransform and SkiaBitLocker are both used by WebKit. BUG=90078 TEST=none Review URL: http://codereview.chromium.org/9648024 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@125892 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 125473 - skia: Change MakeOpaque() function to take a gfx::Rect ↵akalin@chromium.org2012-03-073-23/+19
| | | | | | | | | | | | | | | parameter. BUG=100898 R=pkasting@chromium.org TBR=sky@chromium.org Review URL: https://chromiumcodereview.appspot.com/9581040 TBR=tfarina@chromium.org Review URL: https://chromiumcodereview.appspot.com/9621019 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@125478 0039d316-1c4b-4281-b951-d872f2087c98
* skia: Change MakeOpaque() function to take a gfx::Rect parameter.tfarina@chromium.org2012-03-073-19/+23
| | | | | | | | | | BUG=100898 R=pkasting@chromium.org TBR=sky@chromium.org Review URL: https://chromiumcodereview.appspot.com/9581040 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@125473 0039d316-1c4b-4281-b951-d872f2087c98
* Remove implicit dependency from 'skia' on 'ui'.thakis@chromium.org2012-03-051-2/+6
| | | | | | | | | | | This is needed for the components build. Just adding an explicit dependency from skia to ui doesn't work because ui already depends on skia. BUG=90078 TEST=No linker errors when linking libskia.dylib Review URL: http://codereview.chromium.org/9578010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@125009 0039d316-1c4b-4281-b951-d872f2087c98
* Remove dependency on Chromium base libraries and libicu for ↵reed@google.com2012-02-242-16/+7
| | | | | | | | SkFontHost_fontconfig in preparation for inclusion into Skia Review URL: https://chromiumcodereview.appspot.com/9456020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@123469 0039d316-1c4b-4281-b951-d872f2087c98
* Fix some gcc 4.6 compiler warnings in skia/ext.tony@chromium.org2012-02-242-3/+6
| | | | | | | | | BUG=None Review URL: http://codereview.chromium.org/9456022 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@123431 0039d316-1c4b-4281-b951-d872f2087c98
* Remove ValidFontID.bungeman@google.com2012-02-231-13/+0
| | | | | | Review URL: http://codereview.chromium.org/9347043 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@123282 0039d316-1c4b-4281-b951-d872f2087c98
* Skia DEPS roll to 3147epoger@chromium.org2012-02-104-4/+6
| | | | | | Review URL: https://chromiumcodereview.appspot.com/9346023 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@121451 0039d316-1c4b-4281-b951-d872f2087c98
* Intoroducing Motoya font for ChromeOSkochi@chromium.org2012-02-091-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | This is a retry of r120926, which was reverted. The error was due to duplicate entry in the translation resource file, which only happens on 'Official' build. We will introduce Motoya fonts (MotoyaG04Gothic, MotoyaG04Mincho) as the default Japanese font for ChromeOS. This CL is for preparing resource to define default font and Skia to check the font family is compatible with their counterparts of MS Windows fonts. Note that GTK font specification in the resource does not handle multiple fonts (e.g. "MotoyaG04Gothic 10, IPAPGothic 10") and I left it as is (see crosbug.com/17382 for details). Fontconfig should cover the compatibility check (IPAPGothic => MotoyaG04Gothic) and display the proper font. BUG=chromium-os:16690 TEST=build, run and see Japanese fonts are properly displayed in both web contents and Chrome UI. Review URL: http://codereview.chromium.org/9365012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@121170 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 120926 - Intoroducing Motoya font for ChromeOSkochi@google.com2012-02-081-4/+0
| | | | | | | | | | | | | | | | | | | | | | | | We will introduce Motoya fonts (MotoyaG04Gothic, MotoyaG04Mincho) as the default Japanese font for ChromeOS. This CL is for preparing resource to define default font and Skia to check the font family is compatible with their counterparts of MS Windows fonts. Note that GTK font specification in the resource does not handle multiple fonts (e.g. "MotoyaG04Gothic 10, IPAPGothic 10") and I left it as is (see crosbug.com/17382 for details). Fontconfig should cover the compatibility check (IPAPGothic => MotoyaG04Gothic) and display the proper font. BUG=chromium-os:16690 TEST=build, run and see Japanese fonts are properly displayed in both web contents and Chrome UI. Review URL: http://codereview.chromium.org/9309048 TBR=kochi@chromium.org Review URL: https://chromiumcodereview.appspot.com/9363017 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120949 0039d316-1c4b-4281-b951-d872f2087c98
* Intoroducing Motoya font for ChromeOSkochi@chromium.org2012-02-081-0/+4
| | | | | | | | | | | | | | | | | | | | | We will introduce Motoya fonts (MotoyaG04Gothic, MotoyaG04Mincho) as the default Japanese font for ChromeOS. This CL is for preparing resource to define default font and Skia to check the font family is compatible with their counterparts of MS Windows fonts. Note that GTK font specification in the resource does not handle multiple fonts (e.g. "MotoyaG04Gothic 10, IPAPGothic 10") and I left it as is (see crosbug.com/17382 for details). Fontconfig should cover the compatibility check (IPAPGothic => MotoyaG04Gothic) and display the proper font. BUG=chromium-os:16690 TEST=build, run and see Japanese fonts are properly displayed in both web contents and Chrome UI. Review URL: http://codereview.chromium.org/9309048 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@120926 0039d316-1c4b-4281-b951-d872f2087c98
* Set CG clip to empty explicitlycaryclark@chromium.org2012-01-311-1/+9
| | | | | | | | | | | | | | | Skia draws Mac user elements by creating a CoreGraphics context that mirrors Skia's SkCanvas' context. CoreGraphics does not consider a newly created path to be empty. Explicitly set it to empty so the subsequent drawing is clipped out. BUG: 111642 TEST: http://jsfiddle.net/kenjibaheux/UN3Dt/3/ Review URL: http://codereview.chromium.org/9104010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@119962 0039d316-1c4b-4281-b951-d872f2087c98
* roll skia to 3035reed@google.com2012-01-131-1/+1
| | | | | | Review URL: http://codereview.chromium.org/9199012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@117659 0039d316-1c4b-4281-b951-d872f2087c98
* Adjust the CoreGraphics context for offscreen layers.caryclark@google.com2012-01-111-5/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | UI elements like buttons are drawn on Mac by pointing the CoreGraphics context at the Skia bitmap associated with the current device. If the elements are drawn in a transparent context, Skia creates an offscreen layer to capture the draw, then applies the transparency. CoreGraphics does not know that it is drawing to an offscreen, so the clip and matrix need to be adjusted by the offscreen's origin, which is its position on the device. Also, use getTopDevice() instead of getDevice() to obtain the bitmap. Normally, they will return the same SkDevice (and in that case, getOrigin() will return (0,0)) but if there's a layer present, getDevice() will return the actual device instead of the tempoary offscreen. BUG:108767 TEST:http://jsfiddle.net/casaschi/JWkfA/ Review URL: http://codereview.chromium.org/9129001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@117217 0039d316-1c4b-4281-b951-d872f2087c98
* Beef up and reenable memory diagnostics for skia canvascpu@chromium.org2012-01-061-39/+38
| | | | | | | | | | We got a big crash spike in canary. See bug below. BUG=7913 TEST=see bug Review URL: http://codereview.chromium.org/9112026 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@116646 0039d316-1c4b-4281-b951-d872f2087c98
* Fix Coverity defects in skia/ext:vandebo@chromium.org2012-01-037-25/+20
| | | | | | | | | | | | | - Three uninit ctors, two dead codes, and a misuse of var args. CID=100083, 101537, 17746, 9369, 9279, 9278 BUG=none TEST=none Review URL: http://codereview.chromium.org/8965060 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@116189 0039d316-1c4b-4281-b951-d872f2087c98
* Optimize |SkTypeface| creation on Linux.asvitkine@chromium.org2011-12-222-16/+67
| | | | | | | | | | | | | | | | | | Add a map to cache the results of FontConfig lookup to avoid making excessive FontConfig queries. This improves the speed of |gfx::Font| creation under Linux, which is backed by a |SkTypeface|, as well as |RenderTextLinux| drawing code. BUG=107394 TEST=In Linux Aura, launch about:tracing and notice that "RenderTextLinux::DrawVisualText SetFont" times have gone way down. Review URL: http://codereview.chromium.org/9028001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@115619 0039d316-1c4b-4281-b951-d872f2087c98
* Refactor some parts of |FontConfigDirect::Match()| into helper functions.asvitkine@chromium.org2011-12-221-107/+114
| | | | | | | | | | | | | | | No functional changes. I'm looking to make some changes to this class to optimize SkTypeface creation on Linux. This refactoring makes the code a little bit less unwieldy to work with. BUG=none TEST=none Review URL: http://codereview.chromium.org/9021035 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@115454 0039d316-1c4b-4281-b951-d872f2087c98
* Add a new ctor to CanvasSkia that takes a const SkBitmap&.tfarina@chromium.org2011-12-211-3/+3
| | | | | | | | | | BUG=100898 R=pkasting@chromium.org TBR=ben@chromium.org Review URL: http://codereview.chromium.org/9006001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@115384 0039d316-1c4b-4281-b951-d872f2087c98
* Change SkFileDescriptorStream to handle failure to open the file descriptorreed@google.com2011-12-061-11/+8
| | | | | | | | (by initializing its length_ to 0), and to return the number of bytes skipped if buffer==null, rather than returning the offset. Review URL: http://codereview.chromium.org/8800017 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@113197 0039d316-1c4b-4281-b951-d872f2087c98
* Add OVERRIDE to skia/.avi@chromium.org2011-11-232-11/+20
| | | | | | | | | BUG=104314 TEST=no change Review URL: http://codereview.chromium.org/8681014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@111413 0039d316-1c4b-4281-b951-d872f2087c98
* Implement skia sandbox callbackarthurhsu@chromium.org2011-11-232-0/+41
| | | | | | | | | | BUG=103032 TEST=none Review URL: http://codereview.chromium.org/8566026 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@111400 0039d316-1c4b-4281-b951-d872f2087c98
* Add OVERRIDE to skia/.avi@chromium.org2011-11-211-3/+4
| | | | | | | | | BUG=104314 TEST=no change Review URL: http://codereview.chromium.org/8595024 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@110915 0039d316-1c4b-4281-b951-d872f2087c98
* Add OVERRIDE to skia/.avi@chromium.org2011-11-161-1/+1
| | | | | | | | | BUG=104314 TEST=no change Review URL: http://codereview.chromium.org/8539046 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@110318 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 109461 - twiz@chromium.org2011-11-142-58/+4
| | | | | | | | | | | | | | | | | | | Re-introduction of release mode checks to determine the cause of CreateDIBSection failures. The crash is happening outside of the expected conditions in which CreateDIBSection would fail (large bitmap size, GDI and virtual memory pressure). To track the cause, this CL attempts to capture the last system error on bitmap allocation failure. See previous checks here: http://codereview.chromium.org/8341090 BUG=101934, 104063 TEST=NONE Review URL: http://codereview.chromium.org/8509030 TBR=twiz@chromium.org Review URL: http://codereview.chromium.org/8565027 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@109978 0039d316-1c4b-4281-b951-d872f2087c98
* Cleanup: Remove unneeded forward declarations in printing, sandbox, and skia.thestig@chromium.org2011-11-142-9/+4
| | | | | | | | | BUG=none TEST=none Review URL: http://codereview.chromium.org/8538001 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@109961 0039d316-1c4b-4281-b951-d872f2087c98
* Re-introduction of release mode checks to determine the cause of ↵twiz@chromium.org2011-11-102-4/+58
| | | | | | | | | | | | | | | CreateDIBSection failures. The crash is happening outside of the expected conditions in which CreateDIBSection would fail (large bitmap size, GDI and virtual memory pressure). To track the cause, this CL attempts to capture the last system error on bitmap allocation failure. See previous checks here: http://codereview.chromium.org/8341090 BUG=101934 TEST=NONE Review URL: http://codereview.chromium.org/8509030 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@109461 0039d316-1c4b-4281-b951-d872f2087c98
* roll skia to 2633reed@google.com2011-11-091-8/+0
| | | | | | Review URL: http://codereview.chromium.org/8499024 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@109227 0039d316-1c4b-4281-b951-d872f2087c98
* OpenBSD patches for skia, split from CR #8275005robert.nagy@gmail.com2011-11-073-2/+6
| | | | | | | | | | | | | | | | | bitmap_platform_device_linux.cc: - according to pkg-config cairo include dir is /usr/local/include/cairo (even on linux), so we have to include cairo.h and not cairo/cairo.h - Define DrawToNativeContext on OpenBSD too BUG= TEST= Review URL: http://codereview.chromium.org/8339020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@108930 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 108023 - Debug checks to make it easier to find reason for ↵groby@chromium.org2011-11-032-35/+4
| | | | | | | | | | | | | | | CanvasT<>::initialize crash BUG=101934,102740 TEST=none Review URL: http://codereview.chromium.org/8341090 TBR=groby@chromium.org, vandebo@chromium.org Review URL: http://codereview.chromium.org/8437105 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@108557 0039d316-1c4b-4281-b951-d872f2087c98
* Add traces in compositor, views and skiapiman@chromium.org2011-11-013-0/+17
| | | | | | | | | | BUG=None TEST=None Review URL: http://codereview.chromium.org/8440008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@108174 0039d316-1c4b-4281-b951-d872f2087c98
* Debug checks to make it easier to find reason for CanvasT<>::initialize crashgroby@chromium.org2011-10-312-4/+35
| | | | | | | | | | BUG=101934 TEST=none Review URL: http://codereview.chromium.org/8341090 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@108023 0039d316-1c4b-4281-b951-d872f2087c98
* Check for null contextcaryclark@google.com2011-10-061-2/+4
| | | | | | | | | | | | | | | | | | | | https://bugs.webkit.org/show_bug.cgi?id=69149 describes a situation where running webkit tests sporadically crashes in the Debug build. The context returned by CGContextForData is NULL, so an early return prevents the BitmapPlatformDevice constructor from firing an assert in this case. The callers already anticipate a NULL result, and will propogate the error correctly. BUG=WK69149 TEST=run webkit tests in a Debug build repeatedly Review URL: http://codereview.chromium.org/8118014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@104374 0039d316-1c4b-4281-b951-d872f2087c98
* Allow CanvasSkia to bind to an existing SkCanvas.piman@chromium.org2011-10-056-25/+58
| | | | | | | | | | BUG=None TEST=None Review URL: http://codereview.chromium.org/8122013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@104146 0039d316-1c4b-4281-b951-d872f2087c98
* 1 Add ScopedGetDC and ScopedSelectObjectyosin@chromium.org2011-10-041-5/+6
| | | | | | | | | | | | | | | | | | | | | | * To mark@, sky@ ** base/win/scoped_hdc.h ** base/win/scoped_select_object.h 2 Fix leak in printing * To kmadhusu@ ** Play object creation command in metafile once ** Fix trivial leaks 3 Fix leak in SKIA * To twiz@ ** Fix font leak BUG=98523 TEST=Manual on Win7 Review URL: http://codereview.chromium.org/8084018 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@103852 0039d316-1c4b-4281-b951-d872f2087c98
* Reland 103021: Printing preview of a PDF on Mac with Skia only previews the ↵kmadhusu@chromium.org2011-09-282-0/+57
| | | | | | | | | | | | | | | | | last page of the PDF BUG=93145 TEST=Please refer to bug report. Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=103021 Original review URL: codereview.chromium.org/7719014 Please refer to trybot results in the last patch of original review url. Review URL: http://codereview.chromium.org/8059034 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@103082 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 103021 - PrintPreview: Printing preview of a PDF on Mac with Skia ↵kmadhusu@chromium.org2011-09-272-57/+0
| | | | | | | | | | | | | | only previews the last page of the PDF BUG=93145 TEST=Please refer to bug report. Review URL: http://codereview.chromium.org/7719014 TBR=kmadhusu@chromium.org Review URL: http://codereview.chromium.org/8052020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@103026 0039d316-1c4b-4281-b951-d872f2087c98
* PrintPreview: Printing preview of a PDF on Mac with Skia only previews the ↵kmadhusu@chromium.org2011-09-272-0/+57
| | | | | | | | | | | last page of the PDF BUG=93145 TEST=Please refer to bug report. Review URL: http://codereview.chromium.org/7719014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@103021 0039d316-1c4b-4281-b951-d872f2087c98
* Fixed build error for skia/ext/canvas_paint_x.hsaintlou@chromium.org2011-09-271-1/+1
| | | | | | | | | | | | There is no conversion from Double in Skia. BUG=97131 TEST=none Review URL: http://codereview.chromium.org/8038033 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@102850 0039d316-1c4b-4281-b951-d872f2087c98
* Fixed some chrome aura build problemsdavemoore@chromium.org2011-09-233-1/+114
| | | | | | | | | | | | - Removed default_plugin from aura build - Added new canvas_paint_x and renamed canvas_paint_linux to canvas_paint_gtk BUG=97131 TEST=none Review URL: http://codereview.chromium.org/8015012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@102555 0039d316-1c4b-4281-b951-d872f2087c98
* Build skia for Android.zhenghao@google.com2011-09-1710-18/+138
| | | | | | | | | | | | Add some files and fix some code to make skia build on Android. BUG=none TEST=none Review URL: http://codereview.chromium.org/7920020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@101644 0039d316-1c4b-4281-b951-d872f2087c98
* Remove dead cairo code.abodenha@chromium.org2011-09-082-826/+0
| | | | | | | | | | BUG= TEST=No visible changes. Review URL: http://codereview.chromium.org/7847002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@100281 0039d316-1c4b-4281-b951-d872f2087c98
* implement MakeOpaque at the canvas level, respecting matrix and clipreed@google.com2011-08-316-72/+16
| | | | | | Review URL: http://codereview.chromium.org/7799023 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@99009 0039d316-1c4b-4281-b951-d872f2087c98
* CL removing inheritance of SkDevice from PlatformDevice. twiz@google.com2011-08-2820-519/+233
| | | | | | | | | | | | | PlatformDevice is now a base interface, which is implemented by the various flavours of BitmapPlatformDevice, and VectorPlatformDevice. The BitmapPlatformDevice and VectorPlatformDevice classes now inherit directly from SkDevice, or SkPDFDevice, as appropriate. PlatformDevice helper functions access the PlatformDevice interface attached to a SkDevice via meta-data on the SkDevice. BUG=none TEST=none Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=98230 Review URL: http://codereview.chromium.org/7633040 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@98585 0039d316-1c4b-4281-b951-d872f2087c98