summaryrefslogtreecommitdiffstats
path: root/base
Commit message (Collapse)AuthorAgeFilesLines
* FindBugs: Rename org.chromium.base.test.util.ScalableTimeout.ScaleTimeout to ↵thestig@chromium.org2014-02-061-1/+1
| | | | | | | | | | scaleTimeout. NOTRY=true Review URL: https://codereview.chromium.org/143803017 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@249177 0039d316-1c4b-4281-b951-d872f2087c98
* Comment typo: TASK_POWER_INFO only supported on OSX >= 10.8.4jeremy@chromium.org2014-02-051-1/+1
| | | | | | | | BUG=None Review URL: https://codereview.chromium.org/148253007 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@249170 0039d316-1c4b-4281-b951-d872f2087c98
* Make net_unittests_run work with component=shared_library.maruel@chromium.org2014-02-053-3/+48
| | | | | | | | | | | | Specifically, this adds support to run net_unittests isolated via 'isolate.py run' when built with component build. R=vadimsh@chromium.org,brettw@chromium.org,rsleevi@chromium.org BUG=336439 Review URL: https://codereview.chromium.org/134003004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@249150 0039d316-1c4b-4281-b951-d872f2087c98
* Remove unnecessary uses of aura::Env::GetDispatchersadrul@chromium.org2014-02-052-8/+1
| | | | | | | | | | | | | Avoid using aura::Env::GetDispatcher() as the dispatcher for new instances of base::RunLoop()s, since for both Windows and X11, it's the same as the message-pump. This patch also removes some non-aura views code, and some unnecessary USE_AURA ifdefs. R=darin@chromium.org, sky@chromium.org Review URL: https://codereview.chromium.org/154203002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@249105 0039d316-1c4b-4281-b951-d872f2087c98
* libaddressinput - reduce number of copies in storage class by 2estade@chromium.org2014-02-052-0/+24
| | | | | | | | | | | 1. on the way in (Put) 2. on the way out (Get) BUG=337679 Review URL: https://codereview.chromium.org/148463003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@249102 0039d316-1c4b-4281-b951-d872f2087c98
* Make CopyDirectory() not copy the read only bit on Windows by reimplementing it.maruel@chromium.org2014-02-054-92/+162
| | | | | | | | | | | | | | | Add CopyDirectoryACL test similar to CopyFileACL. Now both CopyDirectory and CopyFile exhibits the same kind of behavior. CopyDirectory is used in unit tests, on Windows installer and on OSX web_application. R=thakis@chromium.org,grt@chromium.org,rvargas@chromium.org BUG=116251 Review URL: https://codereview.chromium.org/141273010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@249088 0039d316-1c4b-4281-b951-d872f2087c98
* Note the overhead of no-parameter TraceEvent methodswangxianzhu@chromium.org2014-02-051-11/+8
| | | | | | | | | | | | | | The overhead of no-parameter TraceEvent methods is at ms or sub-ms order. Note the fact in javadoc. BTW, removed unnecessary references to #begin(). BUG=340872 NOTRY=true Review URL: https://codereview.chromium.org/155903004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@249067 0039d316-1c4b-4281-b951-d872f2087c98
* Remove third_party/jemallochans@chromium.org2014-02-052-60/+5
| | | | | | | | | | | | | We weren't using it, weren't updating it, and don't want to do the maintenance to keep it building. BUG=82385 TBR=cpu NOTRY=true Review URL: https://codereview.chromium.org/154543003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@249025 0039d316-1c4b-4281-b951-d872f2087c98
* Add more owners to base for Androidnyquist@chromium.org2014-02-042-0/+2
| | | | | | | | | | | | Adding nyquist@chromium.org to OWNERS for Android in base/android and base/test/android. BUG=None NOTRY=true Review URL: https://codereview.chromium.org/149803003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@248797 0039d316-1c4b-4281-b951-d872f2087c98
* Include the MSan interface header in base/compiler_specific.h.earthdok@chromium.org2014-02-031-3/+1
| | | | | | | | | | | | | | | Since that header is now available in Chrome, we can include it instead of re-defining the interface. Also, the definition of __msan_unpoison() in base/compiler_specific.h conflicted with the definition in sanitizer/msan_interface.h, preventing us from including the interface header anywhere (e.g. during debugging). BUG=178409 R=eugenis@chromium.org, mark@chromium.org Review URL: https://codereview.chromium.org/132913011 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@248555 0039d316-1c4b-4281-b951-d872f2087c98
* compiler_specific.h: define FINAL as final when using Clang on Windowshans@chromium.org2014-02-031-3/+3
| | | | | | | | | | | | | | Clang on Windows sneakily identifies itself as both Clang and MSVC :) Previously, we would end up defining FINAL as "sealed" here, but in Blink, we'd define it as "final", causing lots of warnings. This change means we define it the same in both places. BUG=82385 NOTRY=true Review URL: https://codereview.chromium.org/134243005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@248499 0039d316-1c4b-4281-b951-d872f2087c98
* Revert r156303 as non-Aura Metro mode is goneyukawa@chromium.org2014-02-032-40/+1
| | | | | | | | | | | | | | | We no longer need to support TSF in the browser code. Now TSF-related code is completely consolidated into the metro_driver process and the browser process doesn't need to know about TSF any more. Note: TSF-related code remaining in base/win/metro.{cc, h} and content/browser/browser_main_runner.{h, cc} will be reverted by subsequent CLs. BUG=319122, 330735 Review URL: https://codereview.chromium.org/137493006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@248475 0039d316-1c4b-4281-b951-d872f2087c98
* Use an alternate mechanism for CreateFile calls in Chromecaitkp@chromium.org2014-02-021-2/+13
| | | | | | | | | | | | BUG=334379 Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=245464 Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=246313 Review URL: https://codereview.chromium.org/138593004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@248380 0039d316-1c4b-4281-b951-d872f2087c98
* Declare the explicit specialization of ScopedGDIObject::Close 'inline'hans@chromium.org2014-01-311-1/+1
| | | | | | | | | | | | Defining it without inline in the header file violates the ODR rule. MSVC doesn't seem to care, but it causes link failures with Clang. BUG=82385 NOTRY=true Review URL: https://codereview.chromium.org/150833002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@248296 0039d316-1c4b-4281-b951-d872f2087c98
* Only report timings for PrefHashFilter::FilterSerializeData() if work is done.gab@chromium.org2014-01-311-18/+2
| | | | | | | | | | Move timing histograms for PrefHashFilter into PrefHashFilter itself. BUG=330908 Review URL: https://codereview.chromium.org/137893028 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@248260 0039d316-1c4b-4281-b951-d872f2087c98
* Get rid of some uses of base::Create*Valueestade@chromium.org2014-01-305-42/+39
| | | | | | | | | BUG=160586 TBR=finnur, ajuma, thakis, atwilson Review URL: https://codereview.chromium.org/131503015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247922 0039d316-1c4b-4281-b951-d872f2087c98
* Add a unit test for JsonPrefStore::Remove.dgrogan@chromium.org2014-01-301-0/+21
| | | | | | Review URL: https://codereview.chromium.org/144633002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247915 0039d316-1c4b-4281-b951-d872f2087c98
* Add a Settings.FilterOnLoadTime histogram to observe time spent on this ↵gab@chromium.org2014-01-301-2/+18
| | | | | | | | | | | | critical path in the wild. This call is on the startup path; we believe all regressions have been addressed, but would like data from the wild to see how this truly behaves in various user configurations. BUG=330908 Review URL: https://codereview.chromium.org/149073003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247893 0039d316-1c4b-4281-b951-d872f2087c98
* JSONWriter cleanups falling out from optimization attempts.gab@chromium.org2014-01-292-55/+40
| | | | | | | | | | As part of issue 330908, I've scrutinized JSONWriter on all angles. Despite failing to optimize it much; here are some micro optimizations and style cleanups I came up with, might as well put them in (although their impact appears to be minimal from local profiling). BUG=330908 Review URL: https://codereview.chromium.org/139903013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247765 0039d316-1c4b-4281-b951-d872f2087c98
* Implement additional set operators: STLSetUnion, STLSetIntersection and ↵rpaquay@chromium.org2014-01-292-3/+200
| | | | | | | | | | | | STLIncludes. BUG=336824 TBR=willchan@chromium.org, jyasskin@chromium.org (lgtm'ed in https://codereview.chromium.org/145663004/) Review URL: https://codereview.chromium.org/148183012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247711 0039d316-1c4b-4281-b951-d872f2087c98
* Increase default stack size on Android with ASan to 2Mb.eugenis@chromium.org2014-01-291-0/+6
| | | | | | | | | | BUG=263749,325685 TEST=cc_unittests on ASan/Android bot R=glider@chromium.org, thakis@chromium.org, bulach@chromium.org, navabi@chromium.org Review URL: https://codereview.chromium.org/136113006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247670 0039d316-1c4b-4281-b951-d872f2087c98
* Add support for two arguments to base::BindBlock.joaodasilva@chromium.org2014-01-282-0/+32
| | | | | | | | BUG=None Review URL: https://codereview.chromium.org/148263010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247502 0039d316-1c4b-4281-b951-d872f2087c98
* Pull new GN binary @ r247206brettw@chromium.org2014-01-281-1/+1
| | | | | | | | | | | | | | Additionally, the Windows binary is a debug build to help find a rare assertion failure. Updates build files to use a list which this new build requires. Update win-tool link commands to match GYP TBR=scottmg Review URL: https://codereview.chromium.org/139283005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247408 0039d316-1c4b-4281-b951-d872f2087c98
* Use icudat*.dat file on Macjshin@chromium.org2014-01-281-0/+11
| | | | | | | | | | | | | | | | | | | | | | 1. Set icu_use_data_file_flag to 1 in common.gypi 2. Add icudatl.dat to the resournce bundle list in chrome_dll_bundle.gypi and content_shell.gypi 3. Move ICU_UTIL_DATA_IMPL to base.gyp This CL has to be landed after https://codereview.chromium.org/111723007/ is landed and rolled in. (done in https://codereview.chromium.org/118313004/ ) Note to perf-sheriff: This CL adds ~10MB to the resource bundle while cutting down the same amount from the Chrome executable/binary. BUG=72633 TEST=All the Mac builds (static, shared) go through and tests (e.g. layout tests, base_unittests --gtest_filter=*String*, net_unittests --gtest_filter=*IDN*) loading the ICU data pass. Review URL: https://codereview.chromium.org/109013004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247404 0039d316-1c4b-4281-b951-d872f2087c98
* Threading: Remove TaskRunnerHandlesNonZeroDelays from TaskRunnerTestDelegatenhiroki@chromium.org2014-01-283-57/+0
| | | | | | | | | | | | This is no longer necessary (always returns true). BUG=149144 TEST=n/a (no behavioral changes) Review URL: https://codereview.chromium.org/146833008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247393 0039d316-1c4b-4281-b951-d872f2087c98
* Fix duplicate condition in target base that was added in r233804.maruel@chromium.org2014-01-281-5/+3
| | | | | | | | | | | That's where you know the gyp file is too large. :) R=spang@chromium.org, thakis@chromium.org BUG= Review URL: https://codereview.chromium.org/130513005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247370 0039d316-1c4b-4281-b951-d872f2087c98
* Convert Media Galleries to use base::Filervargas@chromium.org2014-01-273-61/+69
| | | | | | | | | | | | | | | Unfortunately, this brings in changes to webkit/browser/fileapi, and once that changes, a lot of files have to be updated. The bright side is that most of the collateral changes are just trivial renaming of PlatformFileError -> File::Error and PlatformFileInfo -> File::Info BUG=322664 Review URL: https://codereview.chromium.org/145303002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247301 0039d316-1c4b-4281-b951-d872f2087c98
* Remove GetMonitorInfoWrapper, which is no longer needed.brettw@chromium.org2014-01-272-31/+0
| | | | | | | | | | | We always use ASH on Windows, so the code in the wrapper does not have any effect. This removes the wrapper and updates the callers to use the unwrapped Windows API verison. R=shrikant@chromium.org BUG=247430 Review URL: https://codereview.chromium.org/137353002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@247204 0039d316-1c4b-4281-b951-d872f2087c98
* GTTF: Clearly tag unreliable results in the new test launcherphajdan.jr@chromium.org2014-01-241-0/+9
| | | | | | | | | | | | | | | | | | The new chromium_trybot recipe retries failing tests after deapplying developer's patch. If the tests still fail without the patch, the run is considered a success. To avoid the above masking real failures (e.g. when the launcher is exiting early and skipping some tests), have a single global tag covering all similar conditions in which chromium_trybot or similar recipe should not rely on retries without the patch and return failure. BUG=317931, 236893 R=sky@chromium.org Review URL: https://codereview.chromium.org/135963008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246946 0039d316-1c4b-4281-b951-d872f2087c98
* Update trace file format for NaN/Infinit values to use stringsrbyers@chromium.org2014-01-242-9/+21
| | | | | | | | | | | | Turns out my previous fix was inadequate - the trace viewer doesn't expect null values for arguments. Instead just use hard-coded strings (better anyway since it's clearer to the user what happened). BUG=336787 Review URL: https://codereview.chromium.org/135573004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246758 0039d316-1c4b-4281-b951-d872f2087c98
* Enable icu_use_data_file on Windowsjshin@chromium.org2014-01-242-12/+19
| | | | | | | | | | | | | | | | | | | | | Set icu_use_data_file_flag to 1 on Windows in common.gypi and make the dependency on icudata conditional on 'icu_use_data_file_flag != 0' in common.gypi, base.gyp and net.gyp Add icudtl.dat to the Windows build/archive/install file lists. Load icudtl.dat from DIR_MODULE instead of DIR_EXE on Windows. (to fix bug 337116) This also requires a change in third_party/icu/icu.gyp ( https://codereview.chromium.org/111723007/ ), which was rolled in by https://codereview.chromium.org/118313004/ BUG=72633,337116 TEST=All windows builds work fine and there's no perf regression (start-up etc). Chrome installed via mini_installer runs fine (no issue with icu data file). Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=246387 Review URL: https://codereview.chromium.org/99473012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246751 0039d316-1c4b-4281-b951-d872f2087c98
* Use RtlGenRandom() directly instead of going through rand_s().dalecurtis@chromium.org2014-01-241-20/+22
| | | | | | | | | | | | | | | | | Results in a massive 44x improvement to RandBytes() for larger buffer sizes. Using the RandBytesPerf test, the results are: Original: 2961196.6us Modified: 63656.8us Improvement: 44x BUG=none TEST=randbytes unittests. NOTRY=true Review URL: https://codereview.chromium.org/141753009 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246736 0039d316-1c4b-4281-b951-d872f2087c98
* Fix incorrect assumption in IDMap and SmallMap tests.earthdok@chromium.org2014-01-232-34/+45
| | | | | | | | | | | | | Some of the tests assumed that base::hash_map is ordered, which it is not. They worked under libstdc++, but broke in libc++ builds. BUG=327334 R=thakis@chromium.org, brettw@chromium.org TEST=base_unittests Review URL: https://codereview.chromium.org/130783003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246644 0039d316-1c4b-4281-b951-d872f2087c98
* Revert of Enable icu_use_data_file on Windows ↵jshin@chromium.org2014-01-232-13/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | (https://codereview.chromium.org/99473012/) Reason for revert: mini_installer somehow does not install icudtl.dat in the top product directory (alongside chrome.exe) leading to a failure in InitializeICU(). icudtl.dat is listed in both mini_installer.gypi and FILES.cfg for windows. Original issue's description: > Enable icu_use_data_file on Windows > > > Set icu_use_data_file_flag to 1 on Windows in common.gypi and make the dependency on icudata conditional on 'icu_use_data_file_flag != 0' in common.gypi, base.gyp and net.gyp > > This also requires a change in third_party/icu/icu.gyp > ( https://codereview.chromium.org/111723007/ ), which was rolled in by https://codereview.chromium.org/118313004/ > > BUG=72633 > TEST=All windows builds work fine and there's no perf regression (start-up etc). > > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=246387 TBR=mark@chromium.org,scottmg@chromium.org,laforge@chromium.org,maruel@chromium.org,grt@chromium.org,wtc@chromium.org NOTREECHECKS=true NOTRY=true BUG=72633 Review URL: https://codereview.chromium.org/139403006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246632 0039d316-1c4b-4281-b951-d872f2087c98
* Add Windows utillity function to verify if a computer is part of a domain.pastarmovj@chromium.org2014-01-232-0/+14
| | | | | | | | | BUG=none TEST=none Review URL: https://codereview.chromium.org/140553005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246600 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 246313 "Use an alternate mechanism for CreateFile calls i..."kochi@google.com2014-01-231-13/+2
| | | | | | | | | | | | | | | | | | This caused reading profile error on Windows8 platform. > Use an alternate mechanism for CreateFile calls in Chrome > > BUG=334379 > > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=245464 > > Review URL: https://codereview.chromium.org/138593004 TBR=caitkp@chromium.org Review URL: https://codereview.chromium.org/144333003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246541 0039d316-1c4b-4281-b951-d872f2087c98
* Improve base::RandBytes() performance by 1.75x-2.10x on POSIX.dalecurtis@chromium.org2014-01-235-30/+55
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | No real changes, just avoids doling out Uint64 sized chunks by using the underlying method to hand out correctly sized blocks. Windows is the only platform which doesn't have a byte stream based generator, so I've moved the generic RandBytes() method there and added native methods for other platforms which reuse each platforms internal generator. Performance measured by the new benchmark test over 5 runs, each representing 10 generations of 1mb of random data: Linux x64: Original: 1199625.4 Modified: 686480.2 Improvement: 1.75x On OSX (10.9.1): Original: 1532669.8 Modified: 734808.0 Improvement: 2.10x BUG=none TEST=new benchmark unittest. Review URL: https://codereview.chromium.org/140773006 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246486 0039d316-1c4b-4281-b951-d872f2087c98
* LogMessage::Init must check untrusted severity_ value before using it as an ↵tsepez@chromium.org2014-01-231-1/+8
| | | | | | | | | | | array index. Another common case hit with the IPC fuzzer, and easy to handle. BUG=329723 Review URL: https://codereview.chromium.org/138953003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246474 0039d316-1c4b-4281-b951-d872f2087c98
* Fix trace file format for NaN/infinite valuesrbyers@chromium.org2014-01-223-21/+147
| | | | | | | | | | | | | JSON doesn't support NaN/Infinity. When we get a non-finite float value we should just emit null to be consistent with JavaScript. Also removes a orphaned TraceEvent function declaration. BUG=336787 Review URL: https://codereview.chromium.org/145013003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246419 0039d316-1c4b-4281-b951-d872f2087c98
* Enable icu_use_data_file on Windowsjshin@chromium.org2014-01-222-11/+13
| | | | | | | | | | | | | | Set icu_use_data_file_flag to 1 on Windows in common.gypi and make the dependency on icudata conditional on 'icu_use_data_file_flag != 0' in common.gypi, base.gyp and net.gyp This also requires a change in third_party/icu/icu.gyp ( https://codereview.chromium.org/111723007/ ), which was rolled in by https://codereview.chromium.org/118313004/ BUG=72633 TEST=All windows builds work fine and there's no perf regression (start-up etc). Review URL: https://codereview.chromium.org/99473012 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246387 0039d316-1c4b-4281-b951-d872f2087c98
* Use an alternate mechanism for CreateFile calls in Chromecaitkp@chromium.org2014-01-221-2/+13
| | | | | | | | | | BUG=334379 Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=245464 Review URL: https://codereview.chromium.org/138593004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246313 0039d316-1c4b-4281-b951-d872f2087c98
* Make GetDefaultThreadStackSize() return 16M under ThreadSanitizerglider@chromium.org2014-01-221-0/+6
| | | | | | | | | | | This should help some browser tests that are currently hitting the roof under TSan. BUG=335516 R=epenner@chromium.org,dvyukov@chromium.org Review URL: https://codereview.chromium.org/143933002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246280 0039d316-1c4b-4281-b951-d872f2087c98
* Add support for component=shared_library for base_unittests_run.maruel@chromium.org2014-01-222-1/+39
| | | | | | | | | | | | | | Specifically, this adds support to run base_unittests isolated via 'isolate.py run' when built with component build. This requires rolling icu @ 246118 to get the new icu.isolate. R=thakis@chromium.org,jshin@chromium.org,frankf@chromium.org,rsleevi@chromium.org BUG=336439 Review URL: https://codereview.chromium.org/132233030 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246247 0039d316-1c4b-4281-b951-d872f2087c98
* Documentation fix - Delete meaningless paragraph.girard@chromium.org2014-01-211-4/+0
| | | | | | | | BUG=none Review URL: https://codereview.chromium.org/135903010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246106 0039d316-1c4b-4281-b951-d872f2087c98
* Only compute pref hashes when about to serialize preferences to disk.gab@chromium.org2014-01-212-7/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | This is expected to help a perf regression reported in http://crbug.com/331273. Previously, a new hash would be computed and updated in memory every time a pref's value changes. Computing the hash is expensive as it requires serializing the pref's value to a string. This is particularly bad for the extensions dictionary which is big and can update often in a short time span. The hash is only really needed when writing the file to disk, this change makes sure to only compute the hash at that point. The hashes are still stored in local_state though, so for now this means we're going from a world in which a crash after updating pref1 in memory could result in either: 1) New pref1 in Preferences, new hash for pref1 in local_state 2) New pref1 in Preferences, old hash for pref1 in local_state 3) Old pref1 in Preferences, new hash for pref1 in local_state 4) Old pref1 in Preferences, old hash for pref1 in local_state Depending on which file was flushed to disk. We only want (1) and (4) to be possible; with this CL we remove (3). We still always get (1) on clean shutdowns as local_state is always last flushed after all profiles have been flushed/shutdown. We will later need to move the hashes to the same file as the prefs' values to get rid of (2) as well. BUG=331273 Review URL: https://codereview.chromium.org/143723002 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@246052 0039d316-1c4b-4281-b951-d872f2087c98
* Change the operations touching counts_[] and redundant_count_ to be atomic.glider@chromium.org2014-01-215-18/+34
| | | | | | | | | | | This will make it explicit that the counters are accessed without synchronization and will also suppress TSan v2 warnings. BUG=46840 R=jar@chromium.org,kaiwang@chromium.org Review URL: https://codereview.chromium.org/116983004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245997 0039d316-1c4b-4281-b951-d872f2087c98
* Remove all exclude rules that filename_rules.gypi already handles.dongseong.hwang@intel.com2014-01-202-24/+0
| | | | | | | | BUG= Review URL: https://codereview.chromium.org/141703005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245910 0039d316-1c4b-4281-b951-d872f2087c98
* Removes MessageLoop::Type checks in favor of IsCurrent on MessageLoops.sky@chromium.org2014-01-202-1/+11
| | | | | | | | | | | | This is part of removing the MessageLoop::Type enum. BUG=none TEST=none R=darin@chromium.org Review URL: https://codereview.chromium.org/141683005 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245882 0039d316-1c4b-4281-b951-d872f2087c98
* Fix SecurityTest.CallocOverflow when it's SyzyASan instrumented.sebmarchand@chromium.org2014-01-181-2/+3
| | | | | | | | | | SyzyASan doesn't change the behavior of calloc (the others sanitizers do). BUG= Review URL: https://codereview.chromium.org/138143018 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245715 0039d316-1c4b-4281-b951-d872f2087c98
* Cleanup: Remove a reference to a heapchecker define that no longer exists.thestig@chromium.org2014-01-181-1/+1
| | | | | | | | BUG=325380 Review URL: https://codereview.chromium.org/137063015 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245683 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-03-04 01:17:44 +0000