36908 2010-03-31 16:27:19 -0700 Several tests in fast/canvas/webgl/ failed randomly on Leopard Commit Bot 2010-06-18 12:02:14 -0700 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) PC OS X 10.5 RESOLVED FIXED P2 Normal --- 33416 38560 1 eric zmo abarth cmarrin commit-queue kbr oliver zmo oldest_to_newest 206799 0 eric 2010-03-31 16:27:19 -0700 fast/canvas/webgl/index-validation.html failed on Leopard Commit Bot Twice actually. Somehow lightning struck twice. https://bugs.webkit.org/show_bug.cgi?id=36876#c3 /tmp/layout-test-results/fast/canvas/webgl/index-validation-actual.txt 44 55 PASS gl.getError() is 0 66 PASS gl.drawElements(gl.TRIANGLES, 3, gl.UNSIGNED_SHORT, 0) is undefined. 7 PASS gl.getError() is 0 7 FAIL gl.getError() should be 0. Was 1286. 88 PASS successfullyParsed is true 99 1010 TEST COMPLETE I suspect it may be related to the commit-bot's hardware? 206801 1 eric 2010-03-31 16:28:41 -0700 http://trac.webkit.org/browser/trunk/LayoutTests/fast/canvas/webgl/index-validation.html 206802 2 eric 2010-03-31 16:29:01 -0700 I'm happy to send a full hardware report to anyone who might need one. 206805 3 eric 2010-03-31 16:31:14 -0700 Grepping the commit-queue logs, I see the commit-queue having failed this test 9 times. We just happen to hit two failures in a row while trying to land Adam's patch https://bugs.webkit.org/show_bug.cgi?id=36876#c3 206928 4 zmo 2010-03-31 20:32:34 -0700 This happened once on my machine before, but not recently. I would suggest to use "antialias:false" in this test to resolve this random failure. However, we need to dig into the reason why this happens and find a fix for it. 207011 5 eric 2010-04-01 00:21:16 -0700 Failed again: https://bugs.webkit.org/show_bug.cgi?id=36932#c4 207030 6 eric 2010-04-01 00:50:05 -0700 Something happened today to make this failure much more common than previously. 207043 7 eric 2010-04-01 01:09:44 -0700 Looks like the first time the Leopard Commit Bot ever saw this test fail was around 3:30PM this afternoon. 207044 8 eric 2010-04-01 01:10:23 -0700 It was shortly after http://trac.webkit.org/changeset/56872, which is probably related. :) 207045 9 eric 2010-04-01 01:11:36 -0700 This test has failed 39 times on the Leopard Commit Bot since it first failed this afternoon! 207195 10 abarth 2010-04-01 09:14:18 -0700 This failure rate is unacceptable. Can we rollback that rev and see if it fixes the problem? 207203 11 kbr 2010-04-01 09:35:42 -0700 *Please* don't roll back that revision. It is a large one that we've been trying to land for weeks, and the failure is ultimately caused by driver bugs and not the code. Can we temporarily disable this test on this bot? We may have another workaround we can land in a couple of hours. 207204 12 abarth 2010-04-01 09:40:03 -0700 > Can we temporarily disable this test on this bot? We don't have a way to disable tests on the commit-queue specifically. We can disable it more generally however. > We may have another workaround we can land in a couple of hours. Ok. By way of context, this test is flaky enough that it often fails twice in a row. We have something like 13k tests. If an appreciable fraction of them failed this often, the entire test suite would be useless. 207205 13 kbr 2010-04-01 09:43:07 -0700 (In reply to comment #12) > > Can we temporarily disable this test on this bot? > > We don't have a way to disable tests on the commit-queue specifically. We can > disable it more generally however. Can we add it to the Skipped list for mac-leopard for the moment? > > We may have another workaround we can land in a couple of hours. > > Ok. > > By way of context, this test is flaky enough that it often fails twice in a > row. We have something like 13k tests. If an appreciable fraction of them > failed this often, the entire test suite would be useless. Understood. I apologize for the flakiness introduced by this checkin. 207235 14 52310 zmo 2010-04-01 10:30:47 -0700 Created attachment 52310 patch This patch should stabilize the test on the bot. However, we need to dig deeper into the cause of this flaky behavior - very likely hardware/driver bugs - and find a better way to solve it. Will create another bug to track this down. 207243 15 zmo 2010-04-01 10:43:46 -0700 Just created a new bug to track this issue, but the patch should solve it for now. https://bugs.webkit.org/show_bug.cgi?id=36972 207244 16 52310 kbr 2010-04-01 10:44:15 -0700 Comment on attachment 52310 patch Looks good to me. 207256 17 52310 eric 2010-04-01 11:12:24 -0700 Comment on attachment 52310 patch So is this work-around we intend to roll out eventually? Or is this a permanent change to this test? The ChangeLog doesn't really say. I'm fine with either of course. 207258 18 eric 2010-04-01 11:13:58 -0700 (In reply to comment #12) > By way of context, this test is flaky enough that it often fails twice in a > row. We have something like 13k tests. If an appreciable fraction of them > failed this often, the entire test suite would be useless. Actually, the statement is stronger. After Adam's changes to speed up the commit-queue yesterday (which happened right after this regression landed) it now requires a flaky test to fail *three* times in a row for it to cause the commit queue to wrongly reject a patch. :) 207261 19 kbr 2010-04-01 11:15:42 -0700 (In reply to comment #17) > (From update of attachment 52310 [details]) > So is this work-around we intend to roll out eventually? Or is this a > permanent change to this test? The ChangeLog doesn't really say. I'm fine with > either of course. Right now we don't know why this OpenGL error is reported on this test, so the change is basically permanent. 207269 20 eric 2010-04-01 11:25:14 -0700 Another victim: https://bugs.webkit.org/show_bug.cgi?id=36918#c3 207272 21 eric 2010-04-01 11:29:17 -0700 Another victim: https://bugs.webkit.org/show_bug.cgi?id=36932#c6 207276 22 52310 commit-queue 2010-04-01 11:37:24 -0700 Comment on attachment 52310 patch Clearing flags on attachment: 52310 Committed r56925: <http://trac.webkit.org/changeset/56925> 207277 23 commit-queue 2010-04-01 11:37:36 -0700 All reviewed patches have been landed. Closing bug. 207630 24 eric 2010-04-01 21:23:28 -0700 This test has still failed at least twice on the commit-queue since this fix went in. I can get you the failure diff. 207631 25 eric 2010-04-01 21:25:13 -0700 The failure: /tmp/layout-test-results/fast/canvas/webgl/index-validation-actual.txt 44 55 PASS gl.getError() is 0 66 PASS gl.drawElements(gl.TRIANGLES, 3, gl.UNSIGNED_SHORT, 0) is undefined. 7 PASS gl.getError() is 0 7 FAIL gl.getError() should be 0. Was 1286. 88 PASS successfullyParsed is true 99 1010 TEST COMPLETE 207853 26 zmo 2010-04-02 07:56:34 -0700 After the previous patch, the only difference in this test before and after the multisample patch is the stencil buffer is on by default (before by default we only have color and depth buffer). I'll get another patch ready to turn off the stencil buffer for this test shortly. 207855 27 zmo 2010-04-02 08:02:32 -0700 Also, I am curious, after the previous patch in which I turned off antialias for this test, does the frequency of this random failure reduced on the bot or still the same? 207901 28 eric 2010-04-02 10:02:45 -0700 It appears to be greatly reduced. There have been 9 failure on the bot since last night. Yesterday afternoon we had 66. The bot may simply be doing less now... but I think the failure incidence was reduced, but not eliminated by your previous change. 207915 29 52421 zmo 2010-04-02 10:20:51 -0700 Created attachment 52421 patch: gather extra information about failure This patch is an attempt to gather further information about the cause of this failure (sorry, but seems like the bot is the only machine that we get this failure at the moment). With this patch, the failure behavior won't be better or worse. 207918 30 zmo 2010-04-02 10:22:42 -0700 Eric, when this patch lands, and the bot failed this test again, could you send the output to me? Then I will upload another patch to fix the failure problem (hopefully). 207920 31 52421 kbr 2010-04-02 10:24:11 -0700 Comment on attachment 52421 patch: gather extra information about failure Looks good to me. 207924 32 52421 eric 2010-04-02 10:27:37 -0700 Comment on attachment 52421 patch: gather extra information about failure OK. 207996 33 52421 commit-queue 2010-04-02 12:22:00 -0700 Comment on attachment 52421 patch: gather extra information about failure Clearing flags on attachment: 52421 Committed r57015: <http://trac.webkit.org/changeset/57015> 207997 34 commit-queue 2010-04-02 12:22:07 -0700 All reviewed patches have been landed. Closing bug. 212895 35 zmo 2010-04-15 09:59:03 -0700 These tests are still failing on the commit-bot. 212897 36 53444 zmo 2010-04-15 10:04:07 -0700 Created attachment 53444 patch 212899 37 kbr 2010-04-15 10:07:29 -0700 The patch still has an OOPS about no new tests. Also, I think it's worth separately mentioning in the ChangeLog that this fixes an uninitialized variable bug and restores a glFinish call that used to be present. 212901 38 53445 zmo 2010-04-15 10:11:23 -0700 Created attachment 53445 revised patch : responding to Ken Russell's review 212903 39 kbr 2010-04-15 10:13:02 -0700 Looks good to me. 212936 40 53445 abarth 2010-04-15 11:06:15 -0700 Comment on attachment 53445 revised patch : responding to Ken Russell's review Me too. 212952 41 53445 commit-queue 2010-04-15 11:37:14 -0700 Comment on attachment 53445 revised patch : responding to Ken Russell's review Rejecting patch 53445 from commit-queue. Failed to run "['WebKitTools/Scripts/build-webkit', '--debug']" exit_code: 1 Last 500 characters of output: n/yacc /bin/sh -c /Users/eseidel/Projects/CommitQueue/WebKitBuild/WebCore.build/Debug/WebCore.build/Script-5DF50887116F3077005202AB.sh ** BUILD FAILED ** The following build commands failed: WebCore: Distributed-CompileC /Users/eseidel/Projects/CommitQueue/WebKitBuild/WebCore.build/Debug/WebCore.build/Objects-normal/i386/GraphicsContext3DMac.o /Users/eseidel/Projects/CommitQueue/WebCore/platform/graphics/mac/GraphicsContext3DMac.cpp normal i386 c++ com.apple.compilers.gcc.4_2 (1 failure) Full output: http://webkit-commit-queue.appspot.com/results/1711103 212959 42 53458 zmo 2010-04-15 11:53:23 -0700 Created attachment 53458 revised patch: initialize members in the same order as they are declared 212960 43 kbr 2010-04-15 11:56:16 -0700 LGTM 212965 44 53458 abarth 2010-04-15 12:01:19 -0700 Comment on attachment 53458 revised patch: initialize members in the same order as they are declared Ah, the old "order of initialization" check. :) 212990 45 53458 commit-queue 2010-04-15 13:08:00 -0700 Comment on attachment 53458 revised patch: initialize members in the same order as they are declared Clearing flags on attachment: 53458 Committed r57664: <http://trac.webkit.org/changeset/57664> 212991 46 commit-queue 2010-04-15 13:08:08 -0700 All reviewed patches have been landed. Closing bug. 52310 2010-04-01 10:30:47 -0700 2010-04-15 10:04:07 -0700 patch test.patch text/plain 1173 zmo SW5kZXg6IExheW91dFRlc3RzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9D aGFuZ2VMb2cJKHJldmlzaW9uIDU2OTIwKQorKysgTGF5b3V0VGVzdHMvQ2hhbmdlTG9nCSh3b3Jr aW5nIGNvcHkpCkBAIC0xLDMgKzEsMTIgQEAKKzIwMTAtMDQtMDEgIFpoZW55YW8gTW8gIDx6bW9A Z29vZ2xlLmNvbT4KKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAg ICAgICBmYXN0L2NhbnZhcy93ZWJnbC9pbmRleC12YWxpZGF0aW9uLmh0bWwgZmFpbGVkIG9uIExl b3BhcmQgQ29tbWl0IEJvdAorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1 Zy5jZ2k/aWQ9MzY5MDgKKworICAgICAgICAqIGZhc3QvY2FudmFzL3dlYmdsL3NjcmlwdC10ZXN0 cy9pbmRleC12YWxpZGF0aW9uLmpzOiBEaXNhYmxlIGFudGlhbGlhcyBmb3IgdGhpcyB0ZXN0Lgor CiAyMDEwLTA0LTAxICBDaHJpcyBGbGVpemFjaCAgPGNmbGVpemFjaEBhcHBsZS5jb20+CiAKICAg ICAgICAgUmV2aWV3ZWQgYnkgQmV0aCBEYWtpbi4KSW5kZXg6IExheW91dFRlc3RzL2Zhc3QvY2Fu dmFzL3dlYmdsL3NjcmlwdC10ZXN0cy9pbmRleC12YWxpZGF0aW9uLmpzCj09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0t IExheW91dFRlc3RzL2Zhc3QvY2FudmFzL3dlYmdsL3NjcmlwdC10ZXN0cy9pbmRleC12YWxpZGF0 aW9uLmpzCShyZXZpc2lvbiA1Njg4OCkKKysrIExheW91dFRlc3RzL2Zhc3QvY2FudmFzL3dlYmds L3NjcmlwdC10ZXN0cy9pbmRleC12YWxpZGF0aW9uLmpzCSh3b3JraW5nIGNvcHkpCkBAIC0xLDYg KzEsNiBAQAogZGVzY3JpcHRpb24oIlRlc3Qgb2YgZ2V0IGNhbGxzIGFnYWluc3QgR0wgb2JqZWN0 cyBsaWtlIGdldEJ1ZmZlclBhcmFtZXRlciwgZXRjLiIpOwogCi12YXIgZ2wgPSBjcmVhdGUzRENv bnRleHQoKTsKK3ZhciBnbCA9IGNyZWF0ZTNEQ29udGV4dCh7IGFudGlhbGlhczogZmFsc2UgfSk7 CiB2YXIgcHJvZ3JhbSA9IGxvYWRTdGFuZGFyZFByb2dyYW0oZ2wpOwogCiAvLyAzIHZlcnRpY2Vz ID0+IDEgdHJpYW5nbGUsIGludGVybGVhdmVkIGRhdGEK 52421 2010-04-02 10:20:51 -0700 2010-04-15 10:04:07 -0700 patch: gather extra information about failure info.patch text/plain 2230 zmo SW5kZXg6IExheW91dFRlc3RzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0cy9D aGFuZ2VMb2cJKHJldmlzaW9uIDU3MDA2KQorKysgTGF5b3V0VGVzdHMvQ2hhbmdlTG9nCSh3b3Jr aW5nIGNvcHkpCkBAIC0xLDMgKzEsMTMgQEAKKzIwMTAtMDQtMDIgIFpoZW55YW8gTW8gIDx6bW9A Z29vZ2xlLmNvbT4KKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAg ICAgICBmYXN0L2NhbnZhcy93ZWJnbC9pbmRleC12YWxpZGF0aW9uLmh0bWwgZmFpbGVkIG9uIExl b3BhcmQgQ29tbWl0IEJvdAorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1 Zy5jZ2k/aWQ9MzY5MDgKKworICAgICAgICAqIGZhc3QvY2FudmFzL3dlYmdsL2luZGV4LXZhbGlk YXRpb24tZXhwZWN0ZWQudHh0OiBHYXRoZXJpbmcgbW9yZSBpbmZvcm1hdGlvbiBhYm91dCB0aGUg Y2F1c2Ugb2YgdGhlIGZhaWx1cmUuICBJdCdzIG5vdCBhIGZpeCwgYW5kIGl0IHdvbid0IHdvcnNl IGNoYW5nZSB0aGUgY3VycmVudCB0ZXN0IGJlaGF2aW9yIGVpdGhlciwgaS5lLiwgaXQgd29uJ3Qg bWFrZSBpdCBiZXR0ZXIgb3Igd29yc2UuCisgICAgICAgICogZmFzdC9jYW52YXMvd2ViZ2wvc2Ny aXB0LXRlc3RzL2luZGV4LXZhbGlkYXRpb24uanM6IERpdHRvLgorCiAyMDEwLTA0LTAyICBFcmlj IFNlaWRlbCAgPGVyaWNAd2Via2l0Lm9yZz4KIAogICAgICAgICBSZXZpZXdlZCBieSBEYXJpbiBB ZGxlci4KSW5kZXg6IExheW91dFRlc3RzL2Zhc3QvY2FudmFzL3dlYmdsL2luZGV4LXZhbGlkYXRp b24tZXhwZWN0ZWQudHh0Cj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIExheW91dFRlc3RzL2Zhc3QvY2FudmFzL3dl YmdsL2luZGV4LXZhbGlkYXRpb24tZXhwZWN0ZWQudHh0CShyZXZpc2lvbiA1NjkzNikKKysrIExh eW91dFRlc3RzL2Zhc3QvY2FudmFzL3dlYmdsL2luZGV4LXZhbGlkYXRpb24tZXhwZWN0ZWQudHh0 CSh3b3JraW5nIGNvcHkpCkBAIC0yLDYgKzIsNyBAQCBUZXN0IG9mIGdldCBjYWxscyBhZ2FpbnN0 IEdMIG9iamVjdHMgbGlrCiAKIE9uIHN1Y2Nlc3MsIHlvdSB3aWxsIHNlZSBhIHNlcmllcyBvZiAi UEFTUyIgbWVzc2FnZXMsIGZvbGxvd2VkIGJ5ICJURVNUIENPTVBMRVRFIi4KIAorUEFTUyBnbC5j aGVja0ZyYW1lYnVmZmVyU3RhdHVzKGdsLkZSQU1FQlVGRkVSKSBpcyBnbC5GUkFNRUJVRkZFUl9D T01QTEVURQogUEFTUyBnbC5nZXRFcnJvcigpIGlzIDAKIFBBU1MgZ2wuZHJhd0VsZW1lbnRzKGds LlRSSUFOR0xFUywgMywgZ2wuVU5TSUdORURfU0hPUlQsIDApIGlzIHVuZGVmaW5lZC4KIFBBU1Mg Z2wuZ2V0RXJyb3IoKSBpcyAwCkluZGV4OiBMYXlvdXRUZXN0cy9mYXN0L2NhbnZhcy93ZWJnbC9z Y3JpcHQtdGVzdHMvaW5kZXgtdmFsaWRhdGlvbi5qcwo9PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBMYXlvdXRUZXN0 cy9mYXN0L2NhbnZhcy93ZWJnbC9zY3JpcHQtdGVzdHMvaW5kZXgtdmFsaWRhdGlvbi5qcwkocmV2 aXNpb24gNTY5MzYpCisrKyBMYXlvdXRUZXN0cy9mYXN0L2NhbnZhcy93ZWJnbC9zY3JpcHQtdGVz dHMvaW5kZXgtdmFsaWRhdGlvbi5qcwkod29ya2luZyBjb3B5KQpAQCAtMjUsNiArMjUsNyBAQCBn bC52ZXJ0ZXhBdHRyaWJQb2ludGVyKHZlcnRleExvYywgNCwgZ2wuCiBnbC5lbmFibGVWZXJ0ZXhB dHRyaWJBcnJheSh2ZXJ0ZXhMb2MpOwogZ2wudmVydGV4QXR0cmliUG9pbnRlcihub3JtYWxMb2Ms IDMsIGdsLkZMT0FULCBmYWxzZSwgNyAqIGdsLnNpemVJbkJ5dGVzKGdsLkZMT0FUKSwgMyAqIGds LnNpemVJbkJ5dGVzKGdsLkZMT0FUKSk7CiBnbC5lbmFibGVWZXJ0ZXhBdHRyaWJBcnJheShub3Jt YWxMb2MpOworc2hvdWxkQmUoJ2dsLmNoZWNrRnJhbWVidWZmZXJTdGF0dXMoZ2wuRlJBTUVCVUZG RVIpJywgJ2dsLkZSQU1FQlVGRkVSX0NPTVBMRVRFJyk7CiBzaG91bGRCZSgnZ2wuZ2V0RXJyb3Io KScsICcwJyk7CiBzaG91bGRCZVVuZGVmaW5lZCgnZ2wuZHJhd0VsZW1lbnRzKGdsLlRSSUFOR0xF UywgMywgZ2wuVU5TSUdORURfU0hPUlQsIDApJyk7CiBzaG91bGRCZSgnZ2wuZ2V0RXJyb3IoKScs ICcwJyk7Cg== 53444 2010-04-15 10:04:07 -0700 2010-04-15 10:11:23 -0700 patch fix.patch text/plain 1789 zmo SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIFdlYkNvcmUvQ2hhbmdlTG9n CShyZXZpc2lvbiA1NzY1MCkKKysrIFdlYkNvcmUvQ2hhbmdlTG9nCSh3b3JraW5nIGNvcHkpCkBA IC0xLDMgKzEsMTYgQEAKKzIwMTAtMDQtMTUgIFpoZW55YW8gTW8gIDx6bW9AZ29vZ2xlLmNvbT4K KworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBTZXZlcmFs IHRlc3RzIGluIGZhc3QvY2FudmFzL3dlYmdsLyBmYWlsZWQgcmFuZG9tbHkgb24gTGVvcGFyZCBD b21taXQgQm90CisgICAgICAgIGh0dHBzOi8vYnVncy53ZWJraXQub3JnL3Nob3dfYnVnLmNnaT9p ZD0zNjkwOAorCisgICAgICAgIE5vIG5ldyB0ZXN0cy4gKE9PUFMhKQorCisgICAgICAgICogcGxh dGZvcm0vZ3JhcGhpY3MvbWFjL0dyYXBoaWNzQ29udGV4dDNETWFjLmNwcDoKKyAgICAgICAgKFdl YkNvcmU6OkdyYXBoaWNzQ29udGV4dDNEOjpHcmFwaGljc0NvbnRleHQzRCk6IEluaXRpYWxpemUg d2lkdGgvaGVpZ2h0IHRvIDAvMC4KKyAgICAgICAgKFdlYkNvcmU6OkdyYXBoaWNzQ29udGV4dDNE OjpwcmVwYXJlVGV4dHVyZSk6IGdsRmluaXNoKCkgYWx3YXlzLgorCiAyMDEwLTA0LTE1ICBTdGV2 ZSBGYWxrZW5idXJnICA8c2ZhbGtlbkBhcHBsZS5jb20+CiAKICAgICAgICAgUmV2aWV3ZWQgYnkg QWRhbSBSb2Jlbi4KSW5kZXg6IFdlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3MvbWFjL0dyYXBoaWNz Q29udGV4dDNETWFjLmNwcAo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBXZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNz L21hYy9HcmFwaGljc0NvbnRleHQzRE1hYy5jcHAJKHJldmlzaW9uIDU3NjQ5KQorKysgV2ViQ29y ZS9wbGF0Zm9ybS9ncmFwaGljcy9tYWMvR3JhcGhpY3NDb250ZXh0M0RNYWMuY3BwCSh3b3JraW5n IGNvcHkpCkBAIC05MSw2ICs5MSw4IEBAIEdyYXBoaWNzQ29udGV4dDNEOjpHcmFwaGljc0NvbnRl eHQzRChHcmEKICAgICAsIG1fbXVsdGlzYW1wbGVGQk8oMCkKICAgICAsIG1fbXVsdGlzYW1wbGVE ZXB0aFN0ZW5jaWxCdWZmZXIoMCkKICAgICAsIG1fbXVsdGlzYW1wbGVDb2xvckJ1ZmZlcigwKQor ICAgICwgbV9jdXJyZW50V2lkdGgoMCkKKyAgICAsIG1fY3VycmVudEhlaWdodCgwKQogewogICAg IFZlY3RvcjxDR0xQaXhlbEZvcm1hdEF0dHJpYnV0ZT4gYXR0cmliczsKICAgICBDR0xQaXhlbEZv cm1hdE9iaiBwaXhlbEZvcm1hdE9iaiA9IDA7CkBAIC0zNDAsOCArMzQyLDggQEAgdm9pZCBHcmFw aGljc0NvbnRleHQzRDo6cHJlcGFyZVRleHR1cmUoKQogICAgICAgICA6OmdsQmluZEZyYW1lYnVm ZmVyRVhUKEdMX0RSQVdfRlJBTUVCVUZGRVJfRVhULCBtX2Zibyk7CiAgICAgICAgIDo6Z2xCbGl0 RnJhbWVidWZmZXJFWFQoMCwgMCwgbV9jdXJyZW50V2lkdGgsIG1fY3VycmVudEhlaWdodCwgMCwg MCwgbV9jdXJyZW50V2lkdGgsIG1fY3VycmVudEhlaWdodCwgR0xfQ09MT1JfQlVGRkVSX0JJVCwg R0xfTElORUFSKTsKICAgICAgICAgOjpnbEJpbmRGcmFtZWJ1ZmZlckVYVChHTF9GUkFNRUJVRkZF Ul9FWFQsIG1fYm91bmRGQk8pOwotICAgICAgICA6OmdsRmluaXNoKCk7CiAgICAgfQorICAgIDo6 Z2xGaW5pc2goKTsKIH0KIAogdm9pZCBHcmFwaGljc0NvbnRleHQzRDo6YWN0aXZlVGV4dHVyZSh1 bnNpZ25lZCBsb25nIHRleHR1cmUpCg== 53445 2010-04-15 10:11:23 -0700 2010-04-15 11:53:23 -0700 revised patch : responding to Ken Russell's review fix.patch text/plain 1875 zmo SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIFdlYkNvcmUvQ2hhbmdlTG9n CShyZXZpc2lvbiA1NzY1MCkKKysrIFdlYkNvcmUvQ2hhbmdlTG9nCSh3b3JraW5nIGNvcHkpCkBA IC0xLDMgKzEsMTUgQEAKKzIwMTAtMDQtMTUgIFpoZW55YW8gTW8gIDx6bW9AZ29vZ2xlLmNvbT4K KworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBTZXZlcmFs IHRlc3RzIGluIGZhc3QvY2FudmFzL3dlYmdsLyBmYWlsZWQgcmFuZG9tbHkgb24gTGVvcGFyZCBD b21taXQgQm90CisgICAgICAgIFRoaXMgZml4ZXMgYW4gdW5pbml0aWFsaXplZCB2YXJpYWJsZSBi dWcgYW5kIHJlc3RvcmVzIGEgZ2xGaW5pc2ggY2FsbCB0aGF0IHVzZWQgdG8gYmUgcHJlc2VudC4K KyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTM2OTA4CisK KyAgICAgICAgKiBwbGF0Zm9ybS9ncmFwaGljcy9tYWMvR3JhcGhpY3NDb250ZXh0M0RNYWMuY3Bw OgorICAgICAgICAoV2ViQ29yZTo6R3JhcGhpY3NDb250ZXh0M0Q6OkdyYXBoaWNzQ29udGV4dDNE KTogSW5pdGlhbGl6ZSB3aWR0aC9oZWlnaHQgdG8gMC8wLgorICAgICAgICAoV2ViQ29yZTo6R3Jh cGhpY3NDb250ZXh0M0Q6OnByZXBhcmVUZXh0dXJlKTogUmVzdG9yZSBnbEZpbmlzaCgpIGluIGV2 ZXJ5IHBhdGguCisKIDIwMTAtMDQtMTUgIFN0ZXZlIEZhbGtlbmJ1cmcgIDxzZmFsa2VuQGFwcGxl LmNvbT4KIAogICAgICAgICBSZXZpZXdlZCBieSBBZGFtIFJvYmVuLgpJbmRleDogV2ViQ29yZS9w bGF0Zm9ybS9ncmFwaGljcy9tYWMvR3JhcGhpY3NDb250ZXh0M0RNYWMuY3BwCj09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0K LS0tIFdlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3MvbWFjL0dyYXBoaWNzQ29udGV4dDNETWFjLmNw cAkocmV2aXNpb24gNTc2NDkpCisrKyBXZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNzL21hYy9HcmFw aGljc0NvbnRleHQzRE1hYy5jcHAJKHdvcmtpbmcgY29weSkKQEAgLTkxLDYgKzkxLDggQEAgR3Jh cGhpY3NDb250ZXh0M0Q6OkdyYXBoaWNzQ29udGV4dDNEKEdyYQogICAgICwgbV9tdWx0aXNhbXBs ZUZCTygwKQogICAgICwgbV9tdWx0aXNhbXBsZURlcHRoU3RlbmNpbEJ1ZmZlcigwKQogICAgICwg bV9tdWx0aXNhbXBsZUNvbG9yQnVmZmVyKDApCisgICAgLCBtX2N1cnJlbnRXaWR0aCgwKQorICAg ICwgbV9jdXJyZW50SGVpZ2h0KDApCiB7CiAgICAgVmVjdG9yPENHTFBpeGVsRm9ybWF0QXR0cmli dXRlPiBhdHRyaWJzOwogICAgIENHTFBpeGVsRm9ybWF0T2JqIHBpeGVsRm9ybWF0T2JqID0gMDsK QEAgLTM0MCw4ICszNDIsOCBAQCB2b2lkIEdyYXBoaWNzQ29udGV4dDNEOjpwcmVwYXJlVGV4dHVy ZSgpCiAgICAgICAgIDo6Z2xCaW5kRnJhbWVidWZmZXJFWFQoR0xfRFJBV19GUkFNRUJVRkZFUl9F WFQsIG1fZmJvKTsKICAgICAgICAgOjpnbEJsaXRGcmFtZWJ1ZmZlckVYVCgwLCAwLCBtX2N1cnJl bnRXaWR0aCwgbV9jdXJyZW50SGVpZ2h0LCAwLCAwLCBtX2N1cnJlbnRXaWR0aCwgbV9jdXJyZW50 SGVpZ2h0LCBHTF9DT0xPUl9CVUZGRVJfQklULCBHTF9MSU5FQVIpOwogICAgICAgICA6OmdsQmlu ZEZyYW1lYnVmZmVyRVhUKEdMX0ZSQU1FQlVGRkVSX0VYVCwgbV9ib3VuZEZCTyk7Ci0gICAgICAg IDo6Z2xGaW5pc2goKTsKICAgICB9CisgICAgOjpnbEZpbmlzaCgpOwogfQogCiB2b2lkIEdyYXBo aWNzQ29udGV4dDNEOjphY3RpdmVUZXh0dXJlKHVuc2lnbmVkIGxvbmcgdGV4dHVyZSkK 53458 2010-04-15 11:53:23 -0700 2010-04-15 13:08:00 -0700 revised patch: initialize members in the same order as they are declared fix.patch text/plain 1862 zmo SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIFdlYkNvcmUvQ2hhbmdlTG9n CShyZXZpc2lvbiA1NzY1MCkKKysrIFdlYkNvcmUvQ2hhbmdlTG9nCSh3b3JraW5nIGNvcHkpCkBA IC0xLDMgKzEsMTUgQEAKKzIwMTAtMDQtMTUgIFpoZW55YW8gTW8gIDx6bW9AZ29vZ2xlLmNvbT4K KworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBTZXZlcmFs IHRlc3RzIGluIGZhc3QvY2FudmFzL3dlYmdsLyBmYWlsZWQgcmFuZG9tbHkgb24gTGVvcGFyZCBD b21taXQgQm90CisgICAgICAgIFRoaXMgZml4ZXMgYW4gdW5pbml0aWFsaXplZCB2YXJpYWJsZSBi dWcgYW5kIHJlc3RvcmVzIGEgZ2xGaW5pc2ggY2FsbCB0aGF0IHVzZWQgdG8gYmUgcHJlc2VudC4K KyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTM2OTA4CisK KyAgICAgICAgKiBwbGF0Zm9ybS9ncmFwaGljcy9tYWMvR3JhcGhpY3NDb250ZXh0M0RNYWMuY3Bw OgorICAgICAgICAoV2ViQ29yZTo6R3JhcGhpY3NDb250ZXh0M0Q6OkdyYXBoaWNzQ29udGV4dDNE KTogSW5pdGlhbGl6ZSB3aWR0aC9oZWlnaHQgdG8gMC8wLgorICAgICAgICAoV2ViQ29yZTo6R3Jh cGhpY3NDb250ZXh0M0Q6OnByZXBhcmVUZXh0dXJlKTogUmVzdG9yZSBnbEZpbmlzaCgpIGluIGV2 ZXJ5IHBhdGguCisKIDIwMTAtMDQtMTUgIFN0ZXZlIEZhbGtlbmJ1cmcgIDxzZmFsa2VuQGFwcGxl LmNvbT4KIAogICAgICAgICBSZXZpZXdlZCBieSBBZGFtIFJvYmVuLgpJbmRleDogV2ViQ29yZS9w bGF0Zm9ybS9ncmFwaGljcy9tYWMvR3JhcGhpY3NDb250ZXh0M0RNYWMuY3BwCj09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0K LS0tIFdlYkNvcmUvcGxhdGZvcm0vZ3JhcGhpY3MvbWFjL0dyYXBoaWNzQ29udGV4dDNETWFjLmNw cAkocmV2aXNpb24gNTc2NDkpCisrKyBXZWJDb3JlL3BsYXRmb3JtL2dyYXBoaWNzL21hYy9HcmFw aGljc0NvbnRleHQzRE1hYy5jcHAJKHdvcmtpbmcgY29weSkKQEAgLTgyLDcgKzgyLDkgQEAgUGFz c093blB0cjxHcmFwaGljc0NvbnRleHQzRD4gR3JhcGhpY3NDbwogfQogCiBHcmFwaGljc0NvbnRl eHQzRDo6R3JhcGhpY3NDb250ZXh0M0QoR3JhcGhpY3NDb250ZXh0M0Q6OkF0dHJpYnV0ZXMgYXR0 cnMpCi0gICAgOiBtX2F0dHJzKGF0dHJzKQorICAgIDogbV9jdXJyZW50V2lkdGgoMCkKKyAgICAs IG1fY3VycmVudEhlaWdodCgwKQorICAgICwgbV9hdHRycyhhdHRycykKICAgICAsIG1fY29udGV4 dE9iaigwKQogICAgICwgbV90ZXh0dXJlKDApCiAgICAgLCBtX2ZibygwKQpAQCAtMzQwLDggKzM0 Miw4IEBAIHZvaWQgR3JhcGhpY3NDb250ZXh0M0Q6OnByZXBhcmVUZXh0dXJlKCkKICAgICAgICAg OjpnbEJpbmRGcmFtZWJ1ZmZlckVYVChHTF9EUkFXX0ZSQU1FQlVGRkVSX0VYVCwgbV9mYm8pOwog ICAgICAgICA6OmdsQmxpdEZyYW1lYnVmZmVyRVhUKDAsIDAsIG1fY3VycmVudFdpZHRoLCBtX2N1 cnJlbnRIZWlnaHQsIDAsIDAsIG1fY3VycmVudFdpZHRoLCBtX2N1cnJlbnRIZWlnaHQsIEdMX0NP TE9SX0JVRkZFUl9CSVQsIEdMX0xJTkVBUik7CiAgICAgICAgIDo6Z2xCaW5kRnJhbWVidWZmZXJF WFQoR0xfRlJBTUVCVUZGRVJfRVhULCBtX2JvdW5kRkJPKTsKLSAgICAgICAgOjpnbEZpbmlzaCgp OwogICAgIH0KKyAgICA6OmdsRmluaXNoKCk7CiB9CiAKIHZvaWQgR3JhcGhpY3NDb250ZXh0M0Q6 OmFjdGl2ZVRleHR1cmUodW5zaWduZWQgbG9uZyB0ZXh0dXJlKQo=