summaryrefslogtreecommitdiffstats
path: root/base
Commit message (Collapse)AuthorAgeFilesLines
* Sampling profiler for chrome://tracingscottmg@chromium.org2013-02-265-5/+244
| | | | | | | | | | | | | | | | | | | Allows using TRACE_EVENT_STATE to set a current state. This state is encoded as a static const string pointer. Using the macro stores the static string pointer to a global. While tracing is active, a background thread is spun up and samples this global at a regular interval. The motivation (as compared with other TRACE_EVENT_* macros) is that the runtime cost is lower (in both the disabled and enabled state), at a cost of lower accuracy and less information gathered. A "normal" sampling profiler would try to sample the program counter instead, however, this induces the cost of suspending the target thread(s), whereas this approach only adds the constant overhead of storing to a global. Currently, in chrome://tracing the data shows up as a fake thread containing only one event which contains the sampling data as an "instant" event. This UI will be improved in future patches. BUG= Review URL: https://chromiumcodereview.appspot.com/12224011 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184542 0039d316-1c4b-4281-b951-d872f2087c98
* Fix instant extended grey autocompletion for RTL languages in CrOS and Win.beaudoin@chromium.org2013-02-263-14/+134
| | | | | | | | | | BUG=176572 TEST=Run Chrome with --enable-instant-extended-api and --lang=he. Type "flow" in the omnibox and get "flower" with grey autocompletion. Review URL: https://chromiumcodereview.appspot.com/12328010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184522 0039d316-1c4b-4281-b951-d872f2087c98
* Linux: apply a different hyphen patch.phajdan.jr@chromium.org2013-02-253-0/+17
| | | | | | | | | | This will make it possible to use system hyphen. BUG=176285 Review URL: https://codereview.chromium.org/12259025 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184515 0039d316-1c4b-4281-b951-d872f2087c98
* Make Ash unittests clean up zombie viewer processes between test runs. These ↵robertshield@chromium.org2013-02-253-0/+186
| | | | | | | | | | | | | | zombie viewer processes seem to be a side effect of attempting to quit a metro app while the same app is activating. *while here can mean within a certain number of seconds due to metro life cycle management. Refactor KillAllNamedProcessesWithArgument into base/test. BUG=154081 TEST=Run "ash_unittests.exe --ash-metro-tests" on Win8. Observe that 417 chrome.exe processes do not get created :) Review URL: https://chromiumcodereview.appspot.com/12314043 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184413 0039d316-1c4b-4281-b951-d872f2087c98
* Remove forwarding include for base/files/file_path except for MTPD.brettw@chromium.org2013-02-241-1/+4
| | | | | | | | | | | Fix up remaining/new uses of base/file_path.h MTPD must be updated in a separate pass due to a circular dependency. BUG= Review URL: https://codereview.chromium.org/12316090 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184358 0039d316-1c4b-4281-b951-d872f2087c98
* base: Move MemoryMappedFile out of file_util.h and into its own header file.tfarina@chromium.org2013-02-2411-221/+284
| | | | | | | | | | | BUG=175002 TEST=base_unittests R=brettw@chromium.org TBR=erikwright@chromium.org,tommi@chromium.org,fischman@chromium.org,ben@chromium.org Review URL: https://chromiumcodereview.appspot.com/12321062 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184356 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 184352brettw@chromium.org2013-02-243-30/+10
| | | | | | | | | | | | | | | | | > Add utf_string_conversions to base namespace. > > This adds "using"s for all functions so those can be fixed in a separate pass. > > This converts the "Wide" versions of the functions in the Chrome directory as a first pass on the changeover. > > BUG= > > Review URL: https://codereview.chromium.org/12314090 TBR=brettw@chromium.org Review URL: https://codereview.chromium.org/12315071 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184355 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 184353brettw@chromium.org2013-02-242-12/+2
| | | | | | | | | | | | | | | | | > Move WideToASCII to the base namespace. > > I accidentally changed some callers to use "base::" for this call. Since the > plan is to move this to the base namespace eventually, I went ahead and added > it with a "using" (to keep the non-converted code compiling) rather than revert > my other changes. > > TBR=sky > Review URL: https://codereview.chromium.org/12330103 TBR=brettw@chromium.org Review URL: https://codereview.chromium.org/12326094 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184354 0039d316-1c4b-4281-b951-d872f2087c98
* Move WideToASCII to the base namespace.brettw@chromium.org2013-02-242-2/+12
| | | | | | | | | | | | I accidentally changed some callers to use "base::" for this call. Since the plan is to move this to the base namespace eventually, I went ahead and added it with a "using" (to keep the non-converted code compiling) rather than revert my other changes. TBR=sky Review URL: https://codereview.chromium.org/12330103 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184353 0039d316-1c4b-4281-b951-d872f2087c98
* Add utf_string_conversions to base namespace.brettw@chromium.org2013-02-243-10/+30
| | | | | | | | | | | | This adds "using"s for all functions so those can be fixed in a separate pass. This converts the "Wide" versions of the functions in the Chrome directory as a first pass on the changeover. BUG= Review URL: https://codereview.chromium.org/12314090 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184352 0039d316-1c4b-4281-b951-d872f2087c98
* Move file_path.h to base/files.brettw@chromium.org2013-02-2481-525/+531
| | | | | | TBR=sky git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184344 0039d316-1c4b-4281-b951-d872f2087c98
* Enable browser_tests on linux_aura.erg@chromium.org2013-02-231-0/+7
| | | | | | | | | | | | | This patch is two parts: - Disables some of the remaining failing/flaky tests so we can turn the rest of browser_tests on. - Fix several places where we try to pass empty sizes to X11. BUG=167114, 163931, 173621 Review URL: https://chromiumcodereview.appspot.com/11818065 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184284 0039d316-1c4b-4281-b951-d872f2087c98
* Add a mode flag to the tracing framework.dsinclair@chromium.org2013-02-234-41/+111
| | | | | | | | | | | | | | The mode is provided when SetEnabled is called. Currently there is only one mode which is the trace until buffer is full mode. We will be adding a continuous tracing mode which this will support. BUG=156025 Review URL: https://chromiumcodereview.appspot.com/12302036 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184258 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 184179dcheng@chromium.org2013-02-231-22/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is trigger a latent issue in views which results in a memory leak each time a Widget is destroyed. > Update scoped_ptr<T>::reset() to more closely match std::unique_ptr<T>. > > Remove the no-op behavior of a scoped_ptr "self-reset", and update the > reset() method to detect problematic dependencies on the sequence of > events. Eventually, reset() will be updated to exactly match the > implementation of std::unique_ptr<T> by setting data_.ptr to the new > value before deleting the old value. > > However, this will expose latent bugs where a destructor invoked > transitively by reset() attempts to dereference the same scoped_ptr. > Relying on the value of get() in this instance will dispatch calls to > the incorrect object. As a temporary measure to detect this class of > bugs, set data_.ptr to NULL during deletion so that it results in a > crash. > > BUG=162971,176091 > > Review URL: https://codereview.chromium.org/12223113 TBR=dcheng@chromium.org Review URL: https://codereview.chromium.org/12317084 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184255 0039d316-1c4b-4281-b951-d872f2087c98
* Update scoped_ptr<T>::reset() to more closely match std::unique_ptr<T>.dcheng@chromium.org2013-02-221-19/+22
| | | | | | | | | | | | | | | | | | | | | Remove the no-op behavior of a scoped_ptr "self-reset", and update the reset() method to detect problematic dependencies on the sequence of events. Eventually, reset() will be updated to exactly match the implementation of std::unique_ptr<T> by setting data_.ptr to the new value before deleting the old value. However, this will expose latent bugs where a destructor invoked transitively by reset() attempts to dereference the same scoped_ptr. Relying on the value of get() in this instance will dispatch calls to the incorrect object. As a temporary measure to detect this class of bugs, set data_.ptr to NULL during deletion so that it results in a crash. BUG=162971,176091 Review URL: https://codereview.chromium.org/12223113 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184179 0039d316-1c4b-4281-b951-d872f2087c98
* [Android] Make content_shell_apk and chromium_testshell components friendly.yfriedman@chromium.org2013-02-223-3/+4
| | | | | | | | | | | | | | | | | | | | | | This includes the remaining changes to atleast _compile_ the above targets. More work is needed for starting up the apps and running. Inclues: - Misc EXPORT additions (I assume I can TBR these) - Introduce a ShellDialogs JNI registrar since it's now a separate component and ui can't depend on it. - Random gyp tweaks that are needed to fully specify dependencies - Change from using NPN_MemAlloc to just plain malloc for the java bridge. For some reason the NPN_MemAlloc symbol cannot be found in the component build, but regardless all our implementations end up calling straight throught to malloc. BUG=158821 TBR=wjia@chromium.org,willchan@chromium.org,jcivelli@chromium.org Review URL: https://chromiumcodereview.appspot.com/12281017 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184173 0039d316-1c4b-4281-b951-d872f2087c98
* Implement WebStorage API methodsboliu@chromium.org2013-02-223-0/+55
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Most of the methods involves calling methods to QuotaManager on the IO thread and translating the arguments between Java and native code. Introduce AwQuotaManagerBridge to facilitate this logic. The Java AwQuotaManagerBridge is currently a singleton but should be owned by AwBrowserContext when we have one. The native one is owned by native AwBrowserContext. Java calls the corresponding native AwBrowserContext to obtain the pointer. Introduced JniDependencyFactory interface used to create native objects under native but is used or passed in BrowserContext or ContentsBrowserClient. Also added base::android::ToJavaLongArray to convert to Java long arrays. BUG= Android only change. Ran through android bots. NOTRY=true Review URL: https://chromiumcodereview.appspot.com/12253057 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@184139 0039d316-1c4b-4281-b951-d872f2087c98
* enable -Wstring-conversion when compiling with clangscottmg@chromium.org2013-02-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Summary of try run with flag enabled as warning: On linux_clang, there's false positives from a bunch of assert usage that does !"string" or "string" for a true or false value. It's only in third_party (Mesa, skia, libjingle, etc.) so it probably wouldn't be too hard to disable the warnings in those subtrees. It diagnoses 4 useful locations: 1. ../../base/debug/trace_event_unittest.cc:70:41: warning: implicit conversion turns string literal into bool: 'const char [2]' to 'bool' [-Wstring-conversion] TraceLog::GetInstance()->SetEnabled("*"); ~~~~~~~~ ^~~ 2. ../../sync/syncable/entry_kernel.cc:77:38: warning: implicit conversion turns string literal into bool: 'const char [5]' to 'bool' [-Wstring-conversion] value->SetBoolean("encrypted", "true"); ~~~~~ ^~~~~~ 3. ../../chrome/browser/webdata/web_data_service.cc:334:17: warning: implicit conversion turns string literal into bool: 'const char [44]' to 'bool' [-Wstring-conversion] DLOG_ASSERT("WebDataService dtor called without Shutdown"); ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ../../base/logging.h:595:43: note: expanded from macro 'DLOG_ASSERT' 4. ../../chrome/browser/renderer_host/web_cache_manager_unittest.cc:259:20: warning: implicit conversion turns string literal into bool: 'const char [29]' to 'bool' [-Wstring-conversion] EXPECT_FALSE("Unexpected entry in strategy"); ~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Finding #1 is what made me look into this, #3 and #4 are not doing what they're supposed to, and #2 is kind of funny so not all is lost. So, 3 bugs and 1 very-close-to-a-bug: seems worth enabling. R=thakis@chromium.org TBR=darin@chromium.org, dhollowa@chromium.org, brettw@chromium.org, zea@chromium.org BUG=177306 Review URL: https://chromiumcodereview.appspot.com/12258039 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183998 0039d316-1c4b-4281-b951-d872f2087c98
* Add gyp flag to specify whether we should optimize JNI generation.tedchoc@chromium.org2013-02-221-2/+12
| | | | | | | | | | | | | | | | | | Do not regenerate JNI files (and subsequently the .so file) if they have not changed. This will happen if you edit a java file that has a native counterpart, but the native signatures were not touched. We do not enable this all the time as some build systems require that when you modify an input, the output should also be modified. This also will result in more command line output, so we'll keep it behind a developer flag for now. BUG= Review URL: https://chromiumcodereview.appspot.com/12314025 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183936 0039d316-1c4b-4281-b951-d872f2087c98
* fixes to pre_libc.py script to correctly strip x64wfh@chromium.org2013-02-212-9/+18
| | | | | | | | | | | | object files from libcmt during 64bit build BUG=166496 Review URL: https://chromiumcodereview.appspot.com/12303010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183908 0039d316-1c4b-4281-b951-d872f2087c98
* Add a Clock and TickClock interface for mocking out timeakalin@chromium.org2013-02-2115-0/+325
| | | | | | | | | | | | | | Add DefaultClock and DefaultTickClock implementations. Add SimpleTestClock and SimpleTestTickClock implementations for test. Port some classes in sync/ and media/ to use SimpleTest{,Tick}Clock. BUG=166469 Review URL: https://codereview.chromium.org/11607003 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183865 0039d316-1c4b-4281-b951-d872f2087c98
* Get rid of the ability to unregister preferences.joi@chromium.org2013-02-214-45/+7
| | | | | | | | | | | | | | | | | This was being used only in a couple of places that were clearing or migrating old prefs, and as far as I understand, it is not necessary, especially if we hide the names of those old prefs from users outside of the class that clears them. Also, do registrations of the prefs being migrated at up-front registration time rather than in the migration functions. TBR=ivankr@chromium.org BUG=155525 Review URL: https://codereview.chromium.org/12330008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183816 0039d316-1c4b-4281-b951-d872f2087c98
* Remove a TODO that we don't plan to address (at least not for now). Add ↵joi@chromium.org2013-02-211-1/+3
| | | | | | | | | | comment discouraging use. BUG=155525 Review URL: https://codereview.chromium.org/12321008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183810 0039d316-1c4b-4281-b951-d872f2087c98
* Fix crash in Android NCN due to the use of ActivityStatus.CREATED event.pliard@chromium.org2013-02-211-6/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | NetworkChangeNotifierAutoDetect was subscribing to the ActivityStatus.CREATED event to perform its initialization which seems reasonable. Unfortunately this event can't be used reliably since its notification is deferred (i.e. observers are notified a long time after Main.onCreate() is called). This means that the CREATED event can follow any event including STOPPED which is problematic. Here is what was happening in this specific case: Main activity: CREATED -> RESUMED -> PAUSED -> STOPPED -------------> DESTROYED NetworkChangeNotifierAutoDetect: STOPPED -> CREATED This is problematic since NetworkChangeNotifierAutoDetect must unregister its intent receiver before the main activity is destroyed which was impossible in this case since the last event NCNAutoDetect was seeing before the DESTROYED event (which it doesn't handle) was CREATED. One could argue that the correct fix would be to have NCNAutoDetect handle the DESTROYED event. However clients of ActivityStatus should not have to deal manually with the DESTROYED event in theory since this event is always preceded by STOPPED (according to the standard Android activity life cycle graph). This is only true of course if ActivityStatus follows this life cycle graph (which it should do) but this is not the case unfortunately for now since the CREATED event is deferred. TBR=digit BUG=176292,176837 Review URL: https://codereview.chromium.org/12321044 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183790 0039d316-1c4b-4281-b951-d872f2087c98
* Truncate the download file name if it exceeds the filesystem limit.kinaba@chromium.org2013-02-213-0/+34
| | | | | | | | | | BUG=162734 TEST=Manually tested (Rename a file on Google Drive to >300 characters, and download). Review URL: https://chromiumcodereview.appspot.com/12212010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183727 0039d316-1c4b-4281-b951-d872f2087c98
* Use _target_name instead of package_name for java*.gypicjhopman@chromium.org2013-02-201-2/+0
| | | | | | | | | | | | | | | We were using package_name as a unique name for naming output files and directories. package_name was typically the same as _target_name or a variation of it (like dropping _apk). Using _target_name instead means we need to specify one less thing and it is (maybe?) guaranteed to be unique. TBR=brettw,joi,jar,fischman,zea,sky BUG= Review URL: https://chromiumcodereview.appspot.com/11308030 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183639 0039d316-1c4b-4281-b951-d872f2087c98
* Roll DEPS for mtpd to use base::FilePath.thestig@chromium.org2013-02-201-6/+0
| | | | | | Review URL: https://chromiumcodereview.appspot.com/12301038 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183486 0039d316-1c4b-4281-b951-d872f2087c98
* Add a missing base:: in the tracing macros.dsinclair@chromium.org2013-02-191-1/+1
| | | | | | | | | | | | | When the macro gets expanded outside of the base directory there are compile errors. Add the needed base:: to make everything happy. BUG=111509 Review URL: https://chromiumcodereview.appspot.com/12289010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183274 0039d316-1c4b-4281-b951-d872f2087c98
* Fix crash on Android when the Main activity is resumed after it was destroyed.pliard@chromium.org2013-02-191-1/+7
| | | | | | | | | | | | | | NetworkChangeNotifierAutoDetect was storing and using a pointer to the main activity's Context that can be stale in certain circumstances. This is problematic in case the Main activity is destroyed and then recreated which happens for example when the user changes the system language in the Android settings. BUG=176292 Review URL: https://codereview.chromium.org/12294021 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183217 0039d316-1c4b-4281-b951-d872f2087c98
* Convert FilePath to base::FilePath.brettw@chromium.org2013-02-192-7/+7
| | | | | | | | | | This removes the using for Windows since it is no longer necessary. BUG= Review URL: https://codereview.chromium.org/12303016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183170 0039d316-1c4b-4281-b951-d872f2087c98
* History: Pass min/max timestamps as query parameters to history server.dubroy@chromium.org2013-02-182-3/+5
| | | | | | | | | | | | | | | Queries to the history server should match the same time range as the query to the history backend. Also de-dupe and sort results before handing them to the JS frontend. BUG=None Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=183116 Review URL: https://codereview.chromium.org/12217125 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183129 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 183116grt@chromium.org2013-02-182-5/+3
| | | | | | | | | | | | | | | | | | | | This change appears to have broken HistoryApiTest.DeleteProhibited on Win7 Tests (2). > History: Pass min/max timestamps as query parameters to history server. > > Queries to the history server should match the same time range as the > query to the history backend. > > Also de-dupe and sort results before handing them to the JS frontend. > > BUG=None > > Review URL: https://codereview.chromium.org/12217125 TBR=dubroy@chromium.org Review URL: https://codereview.chromium.org/12304010 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183124 0039d316-1c4b-4281-b951-d872f2087c98
* History: Pass min/max timestamps as query parameters to history server.dubroy@chromium.org2013-02-182-3/+5
| | | | | | | | | | | | | Queries to the history server should match the same time range as the query to the history backend. Also de-dupe and sort results before handing them to the JS frontend. BUG=None Review URL: https://codereview.chromium.org/12217125 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183116 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 183089jschuh@chromium.org2013-02-181-498/+438
| | | | | | | | | | | | | | | | | | | | Needs owner review > Linux: add option to use system tcmalloc (off by default) > > This changes things only for Linux distros, Google Chrome > will continue to use bundled tcmalloc. > > TBR=darin > > BUG=174634 > > Review URL: https://codereview.chromium.org/12224030 TBR=phajdan.jr@chromium.org Review URL: https://codereview.chromium.org/12298019 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183104 0039d316-1c4b-4281-b951-d872f2087c98
* Linux: add option to use system tcmalloc (off by default)phajdan.jr@chromium.org2013-02-181-438/+498
| | | | | | | | | | | | | This changes things only for Linux distros, Google Chrome will continue to use bundled tcmalloc. TBR=darin BUG=174634 Review URL: https://codereview.chromium.org/12224030 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183089 0039d316-1c4b-4281-b951-d872f2087c98
* Fix more remaining FilePath -> base::FilePath.brettw@chromium.org2013-02-1810-280/+283
| | | | | | | | This removes the "using" in file_path.h for Mac since Mac now passes trybots with no using. It still leaves the using for Windows. Review URL: https://codereview.chromium.org/12294008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183066 0039d316-1c4b-4281-b951-d872f2087c98
* cros: Use /var/run/debugfs_gpu for graphics memory infojamescook@chromium.org2013-02-181-5/+6
| | | | | | | | | | | | | | The /sys/kernel/debug file system was recently made non-world-readable, but we've added a bind mount in /var/run for this information. Switch to the new path. BUG=175828 TEST=manual, on a device about:discards shows a "Graphics" line at the bottom Review URL: https://chromiumcodereview.appspot.com/12285016 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183065 0039d316-1c4b-4281-b951-d872f2087c98
* Disable "using base::FilePath" on Linux since it now compiles.brettw@chromium.org2013-02-178-22/+31
| | | | | | | | | | | | | This keeps the using for other platforms temporarily. This patch is to "stop the bleeding" since new conflicts are added every day. The other platforms will be fixed in followup patches. This also adds an exception for the MTPD code which has a circular dependency on FilePath. Review URL: https://codereview.chromium.org/12282019 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183043 0039d316-1c4b-4281-b951-d872f2087c98
* Replace FilePath with base::FilePath.brettw@chromium.org2013-02-1721-106/+129
| | | | | | | | This is im preparation for removing the 'using" in file_path.h Review URL: https://codereview.chromium.org/12286020 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@183021 0039d316-1c4b-4281-b951-d872f2087c98
* Make a new TRACE macro that is slower but uses less memorygman@chromium.org2013-02-162-0/+74
| | | | | | | | | | | | | | | | | | | | | | | The current TRACE macros are designed for speed but it means they generate lots of code. I added a TRACE_EVENT_BINARY_EFFICENTx for those cases where memory is more important the speed. The issue here is I'm auto generating traces. The normal TRACE_EVENT0 ended up adding 90k+ to the release build. This new macro does not. This won't effect perf because these macros used through a wrapper that only gets used when certain command line flags are passed in. In other words the code that uses these macros is not called without the command line flags. BUG=176383 Review URL: https://chromiumcodereview.appspot.com/12278018 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182993 0039d316-1c4b-4281-b951-d872f2087c98
* This fixes the overflows identified in the parsingepenner@chromium.org2013-02-162-15/+51
| | | | | | | | | | | | | function, and adds the dalvik-heap-limit. BUG=176201 NOTRY=true No try since it has passed many times and is taking days in the CQ. Review URL: https://chromiumcodereview.appspot.com/12223064 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182986 0039d316-1c4b-4281-b951-d872f2087c98
* Remove PrefServiceBase, since everyone can now depend directly on PrefService.joi@chromium.org2013-02-1611-262/+184
| | | | | | | | | | TBR=ben@chromium.org BUG=155525 Review URL: https://chromiumcodereview.appspot.com/12252008 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182918 0039d316-1c4b-4281-b951-d872f2087c98
* Make base compile with no "using base::FilePath".brettw@chromium.org2013-02-169-16/+30
| | | | | | | | | | | | For base .cc files not using the base namespace, I added a using since theses files should be moved to the base namespace, and then explicit qualification will no longer be necessary. Original review URL: https://codereview.chromium.org/12226121 (reland of r182040). Review URL: https://codereview.chromium.org/12278014 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182916 0039d316-1c4b-4281-b951-d872f2087c98
* Change call_new_handler to be non extern C; not called outside this code and ↵wfh@chromium.org2013-02-151-4/+5
| | | | | | | | | | | it's causing issues with the linker on Win-64 BUG=166496 Review URL: https://chromiumcodereview.appspot.com/12256047 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182788 0039d316-1c4b-4281-b951-d872f2087c98
* Adding raw tracing to trace framework.dsinclair@chromium.org2013-02-155-78/+295
| | | | | | | | | | | | With some types of tracing you do not want to use the system time as the recorded event time. In those cases, we want to provide a timestamp that will be used to display the event. In a similar vein, the thread which is outputting the event may not be the thread where the event happened so we also allow an explicit thread_id to be provided. BUG=111509 Review URL: https://chromiumcodereview.appspot.com/11366109 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182768 0039d316-1c4b-4281-b951-d872f2087c98
* linux: Both gtk and the aura port should set the icon theme name.erg@chromium.org2013-02-152-34/+10
| | | | | | | | BUG=176103 Review URL: https://codereview.chromium.org/12230026 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182748 0039d316-1c4b-4281-b951-d872f2087c98
* Support JellyBean MR2 style of systracewangxianzhu@chromium.org2013-02-152-7/+19
| | | | | | | | | | | | | | | | | | JB MR2 systrace.py combined the steps of trace tags settings and trace collection into one step, so we need to react to the change of trace tags settings to enable/disable atrace in chrome. The change also benefit pre-JB-MR2 systems that we don't need to restart chrome after changing the trace tags settings in Developer options. However, we still need "adb shell stop; adb shell start" if we set the trace tags using pre-JB-MR2 systrace.py. BUG=173954 Review URL: https://chromiumcodereview.appspot.com/12221047 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182739 0039d316-1c4b-4281-b951-d872f2087c98
* Base: account for calloc aborting in security unittestsjln@chromium.org2013-02-151-32/+47
| | | | | | | | | | | | | | On Linux, when not using tcmalloc, we still have a small wrapper that aborts when allocation functions OOM. In that configuration, we now We consider it a successful detection of an overflow condition if the process aborts. BUG=175500 Review URL: https://chromiumcodereview.appspot.com/12220107 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182628 0039d316-1c4b-4281-b951-d872f2087c98
* Add unit tests for base::GetSystemMemoryInfojamescook@chromium.org2013-02-141-0/+34
| | | | | | | | | BUG=176044 TEST=base_unittests ProcessUtilTest Review URL: https://codereview.chromium.org/12257013 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182540 0039d316-1c4b-4281-b951-d872f2087c98
* Moving last generic Prefs implementation files to base/prefs/.joi@chromium.org2013-02-147-2/+803
| | | | | | | | | | | | | | This completes componentization of Prefs, although there are some remaining TODOs that were created during componentization and interface improvement that still need to be tackled. TBR=ben@chromium.org BUG=155525 Review URL: https://chromiumcodereview.appspot.com/12253004 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@182485 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-03-30 17:44:34 +0000