summaryrefslogtreecommitdiffstats
path: root/tools
Commit message (Collapse)AuthorAgeFilesLines
* Suppress yet another fontconfig leakdkegel@google.com2009-08-061-0/+15
| | | | | | | | | | TBR=agl BUG=18590 TEST=watch ui valgrind bots for a week Review URL: http://codereview.chromium.org/165027 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22572 0039d316-1c4b-4281-b951-d872f2087c98
* Fishing expedition in mac ui tests. dank@chromium.org2009-08-052-13/+118
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adds suppressions for lots of bugs. The only problems left I know of result in valgrind internal error messages. Widen suppression for 16128 to match on Mac. Remove suppression for 16210, which is a dup of 16128, and was out of date. The reference builds don't have symbols on the mac, so suppressions don't work. And they have old known valgrind issues. Disable SunSpiderReferenceTest.Perf and V8BenchmarkReferenceTest.Perf under valgrind until those problems are addressed (and possibly longer -- reference builds tend to have old valgrind errors since fixed and no longer of interest). Also, disable DownloadTest.UnknownSize under valgrind, it is known to send garbage. None of these errors are showing up on the bots for some reason; they only show up when you do a local run of shard_all_tests.sh, which does one test at a time instead of the normal batches of 30. It seems valgrind on the mac isn't stable enough to do large batches of ui tests in a single run. BUG=6522,16128,16210,17113,17451,18189,18223,18236,18252,18253,18268,18540 TEST=run tools/valgrind/shard_all_tests.sh for ui_tests on mac, verify it finds no normal valgrind warnings Review URL: http://codereview.chromium.org/159770 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22533 0039d316-1c4b-4281-b951-d872f2087c98
* GTK: implement bookmark manager copy/paste.estade@chromium.org2009-08-041-0/+5
| | | | | | | | | | | | - Refactor bookmark util clipboard code to be cross platform. - Add a CBF_DATA format for writing arbitrary data to the clipboard. - Change MenuGtk so that showing a menu only makes the menu items update asynchronously. This way we can show the menu immediately and we won't get in trouble if the delegate's IsCommandEnabled() runs a nested message loop. I did not do any mac-side implementation of the new clipboard stuff. The implementation should be easy when they get around to writing the bookmark manager. Review URL: http://codereview.chromium.org/159815 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22419 0039d316-1c4b-4281-b951-d872f2087c98
* Fix memory leak introduced by r22043.aa@chromium.org2009-08-041-23/+0
| | | | | | | | | BUG=18070 TEST=none Review URL: http://codereview.chromium.org/160509 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22367 0039d316-1c4b-4281-b951-d872f2087c98
* Reverting hunspell upgrade 22263,22257,22252,22243. Causing many valgrind ↵mhm@chromium.org2009-08-031-40/+1
| | | | | | | | | warnings. TBR: brettw, dank, jshin Review URL: http://codereview.chromium.org/159797 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22287 0039d316-1c4b-4281-b951-d872f2087c98
* valgrind: Add a suppression for a hunspell memory leak during the ↵mhm@chromium.org2009-08-031-0/+19
| | | | | | | | | | | SpellCheckStrings_EN_US test. TBR=dank BUG=18254 TEST=Watch the linux valgrind bot turn green Review URL: http://codereview.chromium.org/159764 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22263 0039d316-1c4b-4281-b951-d872f2087c98
* valgrind: Add a suppression for a hunspell memory leak during the ↵jhawkins@chromium.org2009-08-031-1/+21
| | | | | | | | | | | SpellCheckTest_GetAutoCorrectionWord_EN_US test. TBR=dank BUG=18254 TEST=Watch the valgrind bot turn green. Review URL: http://codereview.chromium.org/160515 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22257 0039d316-1c4b-4281-b951-d872f2087c98
* Suppress four really common valgrind warnings in mac ui tests.dank@chromium.org2009-08-011-0/+44
| | | | | | | | | BUG=16661,18215,18218,18219 TEST=watch mac valgrind ui tests go greener Review URL: http://codereview.chromium.org/159757 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22239 0039d316-1c4b-4281-b951-d872f2087c98
* valgrind: Remove a suppression for a leak that doesn't exist anymore. The ↵jhawkins@chromium.org2009-07-311-7/+0
| | | | | | | | | | v8 bindings have been upstreamed to WebKit, and this particular code has changed in the meantime. V8Proxy no longer has a method RegisterGlobalHandle. BUG=9450 TEST=Watch the valgrind bot stay green. Review URL: http://codereview.chromium.org/160472 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22186 0039d316-1c4b-4281-b951-d872f2087c98
* Move a valgrind suppression from "pls fix" to "not our problem"estade@chromium.org2009-07-301-9/+9
| | | | | | | | BUG=16161 Review URL: http://codereview.chromium.org/160429 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22123 0039d316-1c4b-4281-b951-d872f2087c98
* Remove the valgrind suppression in crbug.com/15762nirnimesh@chromium.org2009-07-301-7/+0
| | | | | | | | | | | The corresponding valgrind error has been fixed. TEST=test_shell_tests under Valgrind should pass BUG=15762 Review URL: http://codereview.chromium.org/159646 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22096 0039d316-1c4b-4281-b951-d872f2087c98
* Add suppression for yet another SVGNames::init() -related leakdank@chromium.org2009-07-301-1/+11
| | | | | | | | | BUG=17996 TEST=watch valgrind ui builder 2 for 30 builds, verify no SVGNames::init() -related leaks reported Review URL: http://codereview.chromium.org/159626 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22052 0039d316-1c4b-4281-b951-d872f2087c98
* Temp suppression for mem leak with Version::GetVersionFromString()nirnimesh@chromium.org2009-07-301-0/+24
| | | | | | | | | | | This should be reverted when the leak is fixed. BUG=18070 TBR=aa Review URL: http://codereview.chromium.org/160390 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22050 0039d316-1c4b-4281-b951-d872f2087c98
* Fix typo that prevented invalid reads from matchingdkegel@google.com2009-07-291-7/+12
| | | | | | | | | Update to support make We now generate suppressions by default, so remove obsolete --generate_suppressions flag Review URL: http://codereview.chromium.org/155794 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@22013 0039d316-1c4b-4281-b951-d872f2087c98
* Update smoketests.py so we can run ui_tests in parallel, withhuanr@chromium.org2009-07-291-0/+86
| | | | | | | | the number of shards equal to NUMBER_OF_PROCESSORS. Review URL: http://codereview.chromium.org/159568 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21996 0039d316-1c4b-4281-b951-d872f2087c98
* Widen bug 17385's suppression to fit today's variant (which seems to have ↵dank@chromium.org2009-07-291-3/+2
| | | | | | | | | | | | | | | one less function inlined?), http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Mac%20(valgrind)/builds/403/steps/valgrind%20test:%20unit/logs/stdio and also match bug 17979 http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Mac%20(valgrind)/builds/413/steps/valgrind%20test:%20unit/logs/stdio BUG=17385,17979 TEST=sh src/tools/valgrind/chrome_tests.sh --test unit --gtest_filter=RenderThread.* | grep IPC::ChannelProxy::Context::CreateChannel should be empty in 20 out of 20 runs, say Review URL: http://codereview.chromium.org/160326 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21956 0039d316-1c4b-4281-b951-d872f2087c98
* Update suppression to adapt to function renamed as it was upstreamed; see ↵dkegel@google.com2009-07-291-3/+3
| | | | | | | | | | | | | | | https://bugs.webkit.org/show_bug.cgi?id=27103 Also be flexible about new vs. malloc in that suppression. Should make http://chrome-buildbot:8016/builders/Webkit%20Linux%20(valgrind%20layout)/builds/176/steps/valgrind%20test:%20layout/logs/stdio happy. BUG=9458 TEST=sh tools/valgrind/valgrind_webkit_tests.sh LayoutTests/http/tests/misc/timer-vs-loading.html | grep NPN_RegisterObject should be empty Review URL: http://codereview.chromium.org/159560 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21952 0039d316-1c4b-4281-b951-d872f2087c98
* Widen a suppression to match different operator new to takedkegel@google.com2009-07-281-1/+1
| | | | | | | | | | | care of warning seen today at http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Mac%20(valgrind)/builds/379/steps/valgrind%20test:%20unit/logs/stdio BUG=17540 TEST=throw rotten tomatoes at chromium mac valgrind bot if it goes red Review URL: http://codereview.chromium.org/159531 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21890 0039d316-1c4b-4281-b951-d872f2087c98
* Cherrypick valgrind patch to fix thinko that keptdkegel@google.com2009-07-282-0/+22
| | | | | | | | | | leak suppressions from being tallied. BUG=17890 TEST=run sh tools/valgrind/chrome_tests.sh -t base --tool_flags="--nocleanup_on_exit" & look at <suppcounts> in valgrind.tmp/* before and after sh tools/valgrind/build-valgrind-for-chromium.sh, verify leak suppressions show up after. Review URL: http://codereview.chromium.org/160275 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21868 0039d316-1c4b-4281-b951-d872f2087c98
* Add suppression for intermittent leak in IPC::Channel::Channeldkegel@google.com2009-07-281-0/+14
| | | | | | | | | | that's been showing up a lot lately BUG=17888 TEST=watch mac valgrind chromium bot for a day, verify no leaks involving IPC::Channel::Channel() Review URL: http://codereview.chromium.org/159507 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21858 0039d316-1c4b-4281-b951-d872f2087c98
* Add suppression for occasional leak in IPC::Channel::Connect().dkegel@google.com2009-07-281-0/+10
| | | | | | | | BUG=17540 Review URL: http://codereview.chromium.org/159439 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21764 0039d316-1c4b-4281-b951-d872f2087c98
* While we're trying to diagnose the incomplete valgrind logfile problem,dank@chromium.org2009-07-281-0/+3
| | | | | | | | | | output the last 100 lines of the incomplete log(s). We can remove this later. BUG=17453 TEST=modify valgrind to output </XXXvalgrindoutput> at end, verify last 100 lines show up in log Review URL: http://codereview.chromium.org/159437 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21763 0039d316-1c4b-4281-b951-d872f2087c98
* LD_LIBRARY_PATH needs to be set for media tests that uses FFmpeg to work ↵hclam@google.com2009-07-281-0/+4
| | | | | | | | | | | properly. BUG=16945 TEST=test_shell_tests --gtest_filter=MediaLeakTest.* Review URL: http://codereview.chromium.org/159386 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21756 0039d316-1c4b-4281-b951-d872f2087c98
* Widen suppression for bug 17385 again. Should suppress warningdkegel@google.com2009-07-271-4/+2
| | | | | | | | | | | seen this morning in http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Linux%20(valgrind)/builds/1364/steps/valgrind%20test:%20unit/logs/stdio BUG=17385 TEST=watch valgrind unit tests go green Review URL: http://codereview.chromium.org/160168 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21677 0039d316-1c4b-4281-b951-d872f2087c98
* Add suppression for occasional leak in SVGNames::init, seen on dank@chromium.org2009-07-271-0/+9
| | | | | | | | | | | http://build.chromium.org/buildbot/waterfall/builders/Linux%20UI%201%20of%203%20(valgrind)/builds/239/steps/valgrind%20test:%20ui/logs/stdio this morning (and fairly often before that). BUG=17795 TEST=watch valgrind UI bots for a day or two and verify no leak in SVGNames::init Review URL: http://codereview.chromium.org/159420 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21674 0039d316-1c4b-4281-b951-d872f2087c98
* Add experimental TSAN support (off by default)dkegel@google.com2009-07-271-0/+13
| | | | | | Review URL: http://codereview.chromium.org/160042 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21659 0039d316-1c4b-4281-b951-d872f2087c98
* re-do CL 2136 - move image_diff to tools/imagediff, add deps to test_shelldpranke@google.com2009-07-242-0/+388
| | | | | | | | | | | | | | | | | moving image_diff from chrome/tools/test/image_diff to tools/imagediff (not moving to tools/image_diff to avoid some svn unpleasantness with that directory having been created in CL 21366). Also change test_shell.gyp to depend on everything needed to run the layout tests. R=darin@chromium.org TEST=none BUG=none Review URL: http://codereview.chromium.org/159361 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21584 0039d316-1c4b-4281-b951-d872f2087c98
* Fix memory leak in VisitSegmentDatabase::QuerySegmentUsage and ↵meelapshah@chromium.org2009-07-231-7/+0
| | | | | | | | | | | HistoryBackend::QueryTopURLsAndRedirects. BUG=16104 TEST=sh tools/valgrind/chrome_test.sh -t ui --gtest_filter=TabRestoreUITest.RestoreIntoSameWindow Review URL: http://codereview.chromium.org/160060 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21472 0039d316-1c4b-4281-b951-d872f2087c98
* Widen suppression for bug 16128 to also match unit testsdkegel@google.com2009-07-231-2/+0
| | | | | | | | | BUG=16128 TEST=watch "Builder Chromium Linux (valgrind)" go green again Review URL: http://codereview.chromium.org/159300 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21443 0039d316-1c4b-4281-b951-d872f2087c98
* Widen/add suppression for leak in sqlite.dkegel@google.com2009-07-231-0/+10
| | | | | | | | BUG=17576 TEST=watch valgrind ui bots go green(er) Review URL: http://codereview.chromium.org/160017 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21416 0039d316-1c4b-4281-b951-d872f2087c98
* Update valgrind suprression file again.jam@chromium.org2009-07-231-3/+2
| | | | | | Review URL: http://codereview.chromium.org/159279 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21413 0039d316-1c4b-4281-b951-d872f2087c98
* Unrevert 21355 because the revert might be causing thensylvain@chromium.org2009-07-231-5/+6
| | | | | | | | worker tests failures. Review URL: http://codereview.chromium.org/159276 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21412 0039d316-1c4b-4281-b951-d872f2087c98
* Revert "Remove suppressions that don't reproduce."erg@google.com2009-07-231-0/+18
| | | | | | | | | | | | | | It turns out while I couldn't reproduce these valgrind errors locally, they reproduce on the buildbot. This reverts commit dd24218ae59e4c9665c6ebf9fa4ebb5784bf3d88 (r21396). http://crbug.com/16089 TBR=dank Review URL: http://codereview.chromium.org/160023 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21403 0039d316-1c4b-4281-b951-d872f2087c98
* Revert 21355 because it might be causing all the newnsylvain@chromium.org2009-07-231-6/+5
| | | | | | | | | | | | crashes on reliability. It also seems to be causing valgrind error. Original change: Switch the first thread in a child process to be the main thread, and make theIO thread be the second thread. The change is needed for plugins on mac. Review URL: http://codereview.chromium.org/159274 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21398 0039d316-1c4b-4281-b951-d872f2087c98
* Remove suppressions that don't reproduce.erg@google.com2009-07-231-18/+0
| | | | | | | | http://crbug.com/16089 Review URL: http://codereview.chromium.org/159207 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21395 0039d316-1c4b-4281-b951-d872f2087c98
* change smoke test to run true rather than /bin/true to make macosx happydkegel@google.com2009-07-231-1/+1
| | | | | | Review URL: http://codereview.chromium.org/159258 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21387 0039d316-1c4b-4281-b951-d872f2087c98
* Add export_tarball tool.phajdan.jr@chromium.org2009-07-231-0/+90
| | | | | | | | | | It creates a .bz2 tarball with all the sources, but without .svn directories, and has an option to remove files which are not required for build, and take a lot of space. Review URL: http://codereview.chromium.org/155863 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21384 0039d316-1c4b-4281-b951-d872f2087c98
* Part 2 of secret plan to speed up valgrind log processing.dank@chromium.org2009-07-231-0/+2
| | | | | | | | | | | Commit after previous change to build-valgrind-for-chromium.sh is committed and run. BUG=17484 TEST=sh tools/valgrind/chrome_tests.sh -t base --tool_flags="--nocleanup_on_exit"; grep PossiblyLost valgrind.tmp/* should find no matches. Review URL: http://codereview.chromium.org/159252 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21374 0039d316-1c4b-4281-b951-d872f2087c98
* back out CLs 21367, 21366 - no idea why these keep breakingdpranke@google.com2009-07-232-388/+0
| | | | | | Review URL: http://codereview.chromium.org/155983 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21371 0039d316-1c4b-4281-b951-d872f2087c98
* re-do CL 21252 - move image_diff to tools, add deps to test_shelldpranke@google.com2009-07-232-0/+388
| | | | | | | | | | | | | | | CL 21252 seems to have had some svn brokenness (not sure why). I've recreated it and resubmitted it. The changes are to move image_diff from chrome/tools/test to tools/, and to make test_shell depend on image_diff and npapi_layout_test_plugin (and test_worker, where applicable) R=nsylvain@chromium.org TEST=none BUG=none Review URL: http://codereview.chromium.org/155959 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21366 0039d316-1c4b-4281-b951-d872f2087c98
* Add new option --show-possible to valgrind for later use by valgrind_test.py ↵dank@chromium.org2009-07-232-0/+61
| | | | | | | | | | | to speed up log processing. BUG=17484 TEST=run to install new valgrind; sh tools/valgrind/chrome_tests.sh -t base --tool_flags="--nocleanup_on_exit --show-possible=no"; grep PossiblyLost valgrind.tmp/* should find no matches. Review URL: http://codereview.chromium.org/159251 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21365 0039d316-1c4b-4281-b951-d872f2087c98
* Update supressions.jam@chromium.org2009-07-231-6/+6
| | | | | | Review URL: http://codereview.chromium.org/159253 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21363 0039d316-1c4b-4281-b951-d872f2087c98
* Don't kill dsymutil after it finishes.dank@chromium.org2009-07-231-3/+5
| | | | | | | | | BUG=17441 TEST=sh tools/valgrind/chrome_test.sh -t ui on mac, watch it not time out when running dsymutil Review URL: http://codereview.chromium.org/159196 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21361 0039d316-1c4b-4281-b951-d872f2087c98
* Switch the first thread in a child process to be the main thread, and make ↵jam@chromium.org2009-07-231-2/+3
| | | | | | | | theIO thread be the second thread. The change is needed for plugins on mac. Review URL: http://codereview.chromium.org/155944 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21355 0039d316-1c4b-4281-b951-d872f2087c98
* Fix a leak that only happens in layout test mode.tc@google.com2009-07-231-10/+0
| | | | | | | | | | | | | MockWebClipboardImpl is only used in layout test mode and it contains some member variables. Since the destructors weren't virtual, it was leaking these member variables. BUG=17184 Review URL: http://codereview.chromium.org/159238 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21354 0039d316-1c4b-4281-b951-d872f2087c98
* make sure to destroy main window when test shell is shutting down so we ↵estade@chromium.org2009-07-231-7/+0
| | | | | | | | | | | | don't get called back in our signal handlers and access freed memory. BUG=17188 TEST=sh tools/valgrind/valgrind_webkit_tests.sh LayoutTests/editing/pasteboard works Review URL: http://codereview.chromium.org/155893 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21345 0039d316-1c4b-4281-b951-d872f2087c98
* Use $PROGRAMFILES instead of "C:\Program Files\".pkasting@chromium.org2009-07-221-1/+3
| | | | | | Review URL: http://codereview.chromium.org/159241 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21339 0039d316-1c4b-4281-b951-d872f2087c98
* Make UtilityProcessHost destructor virtual.mattm@chromium.org2009-07-221-7/+0
| | | | | | | | BUG=17129 Review URL: http://codereview.chromium.org/155946 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21323 0039d316-1c4b-4281-b951-d872f2087c98
* Use ld.orig if present to build valgrind rather than making user switch back ↵dkegel@google.com2009-07-221-2/+13
| | | | | | | | | | | to ld.orig manually. BUG=https://bugs.kde.org/show_bug.cgi?id=193413 TEST=script should build working valgrind even if user installed gold with install-build-deps.sh Review URL: http://codereview.chromium.org/159218 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21315 0039d316-1c4b-4281-b951-d872f2087c98
* Give slightly more helpful error message if gold is present.dkegel@google.com2009-07-221-1/+1
| | | | | | | | | | | We should fix valgrind to build properly with gold; it might be too hard. BUG=https://bugs.kde.org/show_bug.cgi?id=193413 TEST=valgrind /bin/true should not fail Review URL: http://codereview.chromium.org/159217 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@21309 0039d316-1c4b-4281-b951-d872f2087c98
generated by cgit v1.1 at 2025-01-31 23:26:08 +0000