168150 2017-02-10 15:56:18 -0800 run-webkit-tests for iOS Simulator always complains that stopping WebKitTestRunnerApp.app times out 2017-02-14 08:35:09 -0800 1 1 1 Unclassified WebKit Tools / Tests WebKit Nightly Build Unspecified Unspecified RESOLVED FIXED InRadar P2 Normal --- 1 simon.fraser jbedard aakash_jain ap commit-queue dbates glenn jbedard lforschler simon.fraser webkit-bug-importer oldest_to_newest 1275599 0 simon.fraser 2017-02-10 15:56:18 -0800 563$ $ wktests --debug --ios-simulator LayoutTests/compositing/masks/ This machine could support 4 simulators, but is only configured for 3. Please see <https://trac.webkit.org/wiki/IncreasingKernelLimits>. Using port 'ios-simulator' Test configuration: <, x86_64, debug> Placing test results in /Volumes/Data/Development/OSX/webkit/OpenSource/WebKitBuild/Debug-iphonesimulator/layout-test-results Baseline search path: platform/ios-simulator-wk2 -> platform/ios-simulator -> platform/wk2 -> generic Using Debug build Pixel tests disabled Regular timeout: 30000, slow test timeout: 150000 Command line: /Volumes/Data/Development/OSX/webkit/OpenSource/WebKitBuild/Debug-iphonesimulator/WebKitTestRunnerApp.app - Found 21 tests; running 19, skipping 2. Running 19 tests Creating app:/tmp/WebKitTestingSimulators/Simulator0.app /tmp/WebKitTestingSimulators/Simulator0.app: replacing existing signature Waiting for all iOS Simulators to finish booting. Running 1 WebKitTestRunnerApp.app. [16/19] compositing/masks/solid-color-masked.html passed unexpectedly stopping WebKitTestRunnerApp.app(pid 84992) timed out, killing it <------------------- here 18 tests ran as expected, 1 didn't: Expected to fail, but passed: (1) compositing/masks/solid-color-masked.html I also this this when running DRT with -1. 1275631 1 webkit-bug-importer 2017-02-10 16:28:08 -0800 <rdar://problem/30475467> 1275633 2 jbedard 2017-02-10 16:34:40 -0800 I can fix this, and I will shortly upload a patch which does. That being said, I'm not sure this is a bug. This happens because simctl terminate does not, despite it's name, terminate the app it is given. It simply sends it to a background state. As a result, the process is still running. We could send the process a SIGTERM after using simctl's api to terminate the app and it's process. 1275634 3 301213 jbedard 2017-02-10 16:37:00 -0800 Created attachment 301213 Patch 1275635 4 301213 ap 2017-02-10 16:39:03 -0800 Comment on attachment 301213 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=301213&action=review > Tools/Scripts/webkitpy/port/simulator_process.py:134 > + os.kill(self._pid, signal.SIGTERM) Do we need to give it some time to terminate? Overall, this sequence seems confusing. 1275864 5 301213 jbedard 2017-02-11 15:02:35 -0800 Comment on attachment 301213 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=301213&action=review >> Tools/Scripts/webkitpy/port/simulator_process.py:134 >> + os.kill(self._pid, signal.SIGTERM) > > Do we need to give it some time to terminate? Overall, this sequence seems confusing. I don't believe so. Basically what's going on here is that self._device.terminate_app will not stop the app process. SIGTERM should stop the process, if it does not, we will pass the pid to our executive class in server_process which does a more involved kill sequence which includes SIGKILL as well as SIGTERM. This sequence will prevent the lines that Simon is seeing from being logged unless the processes failed to respond to the SIGTERM as expected. As I mentioned, I don't really think this is broken at the moment, but if we wanted to suppress the logging line Simon is seeing, this would be the correct way to do so. 1276217 6 ap 2017-02-13 09:47:57 -0800 If we don't need to give it time, why do we need to call self._device.terminate_app right before sending the signal? > As I mentioned, I don't really think this is broken at the moment, but if we wanted to suppress the logging line Simon is seeing, this would be the correct way to do so. We do not want spurious error messages in run-webkit-tests output, as that's super misleading. I suspect that it wastes time waiting for the timeout too. 1276248 7 301361 jbedard 2017-02-13 10:58:44 -0800 Created attachment 301361 Patch 1276251 8 jbedard 2017-02-13 11:00:32 -0800 (In reply to comment #6) Ok, this is actually an issue. I've updated the patch. The timeout is real, and is a regression caused by <http://trac.webkit.org/changeset/211370>. It only would have effected things when the app being closed at the end of a test suite. What I said in comment #1 is also true, hence the need to both terminate the app and terminate the process. This is done to give the app a chance to gracefully exit before terminating the process. 1276280 9 ap 2017-02-13 12:00:12 -0800 > This is done to give the app a chance to gracefully exit before terminating the process. To give it a fair chance, how long should we wait before terminating? 1276317 10 jbedard 2017-02-13 12:53:45 -0800 (In reply to comment #9) > > This is done to give the app a chance to gracefully exit before terminating the process. > > To give it a fair chance, how long should we wait before terminating? Unfortunately, this is going to depend on the app. DumpRenderTree and WebKitTestRunner should be closed almost instantly. Does a quarter second pause sound reasonable? 1276362 11 ap 2017-02-13 14:17:25 -0800 I don't know what number is OK, which is part of what makes me question giving the app an opportunity to close nicely. What will break if we always send the signal? 1276386 12 jbedard 2017-02-13 15:00:26 -0800 (In reply to comment #11) > I don't know what number is OK, which is part of what makes me question > giving the app an opportunity to close nicely. What will break if we always > send the signal? With the current implementation of DumpRenderTree and WebKitTestRunner, nothing will break. Some other app may fail to save assets, gracefully close sockets, that sort of thing. But in the case of DumpRenderTree and WebKitTestRunner, sending a SIGTERM shouldn't cause an problems. 1276399 13 ap 2017-02-13 15:16:13 -0800 In this case, I suggest just always sending the signal, as we don't give the process any time to orderly terminate anyway. 1276405 14 301402 jbedard 2017-02-13 15:30:53 -0800 Created attachment 301402 Patch 1276415 15 301402 dbates 2017-02-13 15:53:50 -0800 Comment on attachment 301402 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=301402&action=review > Tools/Scripts/webkitpy/port/simulator_process.py:-133 > - self._device.terminate_app(self._bundle_id) Last I recall this was the only function that called Device.terminate_app(). Is there reason to keep this function around? 1276419 16 301402 jbedard 2017-02-13 16:03:03 -0800 Comment on attachment 301402 Patch View in context: https://bugs.webkit.org/attachment.cgi?id=301402&action=review >> Tools/Scripts/webkitpy/port/simulator_process.py:-133 >> - self._device.terminate_app(self._bundle_id) > > Last I recall this was the only function that called Device.terminate_app(). Is there reason to keep this function around? It is not used anywhere else. Removing. 1276426 17 301411 jbedard 2017-02-13 16:15:04 -0800 Created attachment 301411 Patch 1276759 18 301411 commit-queue 2017-02-14 08:35:03 -0800 Comment on attachment 301411 Patch Clearing flags on attachment: 301411 Committed r212297: <http://trac.webkit.org/changeset/212297> 1276760 19 commit-queue 2017-02-14 08:35:09 -0800 All reviewed patches have been landed. Closing bug. 301213 2017-02-10 16:37:00 -0800 2017-02-13 10:58:40 -0800 Patch bug-168150-20170210163433.patch text/plain 1163 jbedard SW5kZXg6IFRvb2xzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBUb29scy9DaGFuZ2VMb2cJKHJl dmlzaW9uIDIxMjE2NSkKKysrIFRvb2xzL0NoYW5nZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwz ICsxLDEzIEBACisyMDE3LTAyLTEwICBKb25hdGhhbiBCZWRhcmQgIDxqYmVkYXJkQGFwcGxlLmNv bT4KKworICAgICAgICB4Y3J1biBzaW1jdGwgdGVybWluYXRlIGRvZXMgbm90IHRlcm1pbmF0ZSB0 aGUgYXBwIHByb2Nlc3NlcworICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1 Zy5jZ2k/aWQ9MTY4MTUwCisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisK KyAgICAgICAgKiBTY3JpcHRzL3dlYmtpdHB5L3BvcnQvc2ltdWxhdG9yX3Byb2Nlc3MucHk6Cisg ICAgICAgIChTaW11bGF0b3JQcm9jZXNzLl9raWxsKTogU2VuZCBTSUdURVJNIHRvIHNpbXVsYXRv ciBwcm9jZXNzLgorCiAyMDE3LTAyLTEwICBTaW1vbiBGcmFzZXIgIDxzaW1vbi5mcmFzZXJAYXBw bGUuY29tPgogCiAgICAgICAgIE1ha2Ugc3VyZSB0aGUgImlud2luZG93IiBmbGFnIHByb3BhZ2F0 ZXMgdG8gVGlsZWRCYWNraW5ncyBmb3IgbWFza3MgYW5kIHJlZmxlY3Rpb25zCkluZGV4OiBUb29s cy9TY3JpcHRzL3dlYmtpdHB5L3BvcnQvc2ltdWxhdG9yX3Byb2Nlc3MucHkKPT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PQot LS0gVG9vbHMvU2NyaXB0cy93ZWJraXRweS9wb3J0L3NpbXVsYXRvcl9wcm9jZXNzLnB5CShyZXZp c2lvbiAyMTIxMzUpCisrKyBUb29scy9TY3JpcHRzL3dlYmtpdHB5L3BvcnQvc2ltdWxhdG9yX3By b2Nlc3MucHkJKHdvcmtpbmcgY29weSkKQEAgLTEzMSw0ICsxMzEsNSBAQCBjbGFzcyBTaW11bGF0 b3JQcm9jZXNzKFNlcnZlclByb2Nlc3MpOgogCiAgICAgZGVmIF9raWxsKHNlbGYpOgogICAgICAg ICBzZWxmLl9kZXZpY2UudGVybWluYXRlX2FwcChzZWxmLl9idW5kbGVfaWQpCisgICAgICAgIG9z LmtpbGwoc2VsZi5fcGlkLCBzaWduYWwuU0lHVEVSTSkKICAgICAgICAgc3VwZXIoU2ltdWxhdG9y UHJvY2Vzcywgc2VsZikuX2tpbGwoKQo= 301361 2017-02-13 10:58:44 -0800 2017-02-13 15:30:48 -0800 Patch bug-168150-20170213105612.patch text/plain 1428 jbedard SW5kZXg6IFRvb2xzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBUb29scy9DaGFuZ2VMb2cJKHJl dmlzaW9uIDIxMjIzMykKKysrIFRvb2xzL0NoYW5nZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwz ICsxLDE0IEBACisyMDE3LTAyLTEzICBKb25hdGhhbiBCZWRhcmQgIDxqYmVkYXJkQGFwcGxlLmNv bT4KKworICAgICAgICBydW4td2Via2l0LXRlc3RzIGZvciBpT1MgU2ltdWxhdG9yIGFsd2F5cyBj b21wbGFpbnMgdGhhdCBzdG9wcGluZyBXZWJLaXRUZXN0UnVubmVyQXBwLmFwcCB0aW1lcyBvdXQK KyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTE2ODE1MAor CisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgICogU2NyaXB0 cy93ZWJraXRweS9wb3J0L3NpbXVsYXRvcl9wcm9jZXNzLnB5OgorICAgICAgICAoU2ltdWxhdG9y UHJvY2Vzcy5zdG9wKTogQ2xvc2UgYXBwIGJlZm9yZSBjYWxsaW5nIFNlcnZlclByb2Nlc3PigJlz IHN0b3AuCisgICAgICAgIChTaW11bGF0b3JQcm9jZXNzLl9raWxsKTogTW92ZSBjbG9zaW5nIG9m IGFwcCBpbnRvIHN0b3AoKS4KKwogMjAxNy0wMi0xMiAgTWljaGFlbCBDYXRhbnphcm8gIDxtY2F0 YW56YXJvQGlnYWxpYS5jb20+CiAKICAgICAgICAgW0dUS10gU2V2ZXJhbCBmYWlsaW5nIFdlYlZp ZXdFZGl0b3IgQVBJIHRlc3RzCkluZGV4OiBUb29scy9TY3JpcHRzL3dlYmtpdHB5L3BvcnQvc2lt dWxhdG9yX3Byb2Nlc3MucHkKPT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gVG9vbHMvU2NyaXB0cy93ZWJraXRweS9w b3J0L3NpbXVsYXRvcl9wcm9jZXNzLnB5CShyZXZpc2lvbiAyMTIxMzUpCisrKyBUb29scy9TY3Jp cHRzL3dlYmtpdHB5L3BvcnQvc2ltdWxhdG9yX3Byb2Nlc3MucHkJKHdvcmtpbmcgY29weSkKQEAg LTEyOSw2ICsxMjksNyBAQCBjbGFzcyBTaW11bGF0b3JQcm9jZXNzKFNlcnZlclByb2Nlc3MpOgog CiAgICAgICAgIHNlbGYuX3Byb2MgPSBTaW11bGF0b3JQcm9jZXNzLlBvcGVuKHNlbGYuX3BpZCwg c3RkaW4sIHN0ZG91dCwgc3RkZXJyKQogCi0gICAgZGVmIF9raWxsKHNlbGYpOgorICAgIGRlZiBz dG9wKHNlbGYsIHRpbWVvdXRfc2Vjcz0zLjApOgogICAgICAgICBzZWxmLl9kZXZpY2UudGVybWlu YXRlX2FwcChzZWxmLl9idW5kbGVfaWQpCi0gICAgICAgIHN1cGVyKFNpbXVsYXRvclByb2Nlc3Ms IHNlbGYpLl9raWxsKCkKKyAgICAgICAgb3Mua2lsbChzZWxmLl9waWQsIHNpZ25hbC5TSUdURVJN KQorICAgICAgICBzdXBlcihTaW11bGF0b3JQcm9jZXNzLCBzZWxmKS5zdG9wKHRpbWVvdXRfc2Vj cykK 301402 2017-02-13 15:30:53 -0800 2017-02-13 16:15:00 -0800 Patch bug-168150-20170213152821.patch text/plain 1428 jbedard SW5kZXg6IFRvb2xzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBUb29scy9DaGFuZ2VMb2cJKHJl dmlzaW9uIDIxMjIzMykKKysrIFRvb2xzL0NoYW5nZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwz ICsxLDE0IEBACisyMDE3LTAyLTEzICBKb25hdGhhbiBCZWRhcmQgIDxqYmVkYXJkQGFwcGxlLmNv bT4KKworICAgICAgICBydW4td2Via2l0LXRlc3RzIGZvciBpT1MgU2ltdWxhdG9yIGFsd2F5cyBj b21wbGFpbnMgdGhhdCBzdG9wcGluZyBXZWJLaXRUZXN0UnVubmVyQXBwLmFwcCB0aW1lcyBvdXQK KyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTE2ODE1MAor CisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgICogU2NyaXB0 cy93ZWJraXRweS9wb3J0L3NpbXVsYXRvcl9wcm9jZXNzLnB5OgorICAgICAgICAoU2ltdWxhdG9y UHJvY2Vzcy5zdG9wKTogQ2xvc2UgYXBwIGJlZm9yZSBjYWxsaW5nIFNlcnZlclByb2Nlc3PigJlz IHN0b3AuCisgICAgICAgIChTaW11bGF0b3JQcm9jZXNzLl9raWxsKTogTW92ZSBjbG9zaW5nIG9m IGFwcCBpbnRvIHN0b3AoKS4KKwogMjAxNy0wMi0xMiAgTWljaGFlbCBDYXRhbnphcm8gIDxtY2F0 YW56YXJvQGlnYWxpYS5jb20+CiAKICAgICAgICAgW0dUS10gU2V2ZXJhbCBmYWlsaW5nIFdlYlZp ZXdFZGl0b3IgQVBJIHRlc3RzCkluZGV4OiBUb29scy9TY3JpcHRzL3dlYmtpdHB5L3BvcnQvc2lt dWxhdG9yX3Byb2Nlc3MucHkKPT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PQotLS0gVG9vbHMvU2NyaXB0cy93ZWJraXRweS9w b3J0L3NpbXVsYXRvcl9wcm9jZXNzLnB5CShyZXZpc2lvbiAyMTIxMzUpCisrKyBUb29scy9TY3Jp cHRzL3dlYmtpdHB5L3BvcnQvc2ltdWxhdG9yX3Byb2Nlc3MucHkJKHdvcmtpbmcgY29weSkKQEAg LTEyOSw2ICsxMjksNiBAQCBjbGFzcyBTaW11bGF0b3JQcm9jZXNzKFNlcnZlclByb2Nlc3MpOgog CiAgICAgICAgIHNlbGYuX3Byb2MgPSBTaW11bGF0b3JQcm9jZXNzLlBvcGVuKHNlbGYuX3BpZCwg c3RkaW4sIHN0ZG91dCwgc3RkZXJyKQogCi0gICAgZGVmIF9raWxsKHNlbGYpOgotICAgICAgICBz ZWxmLl9kZXZpY2UudGVybWluYXRlX2FwcChzZWxmLl9idW5kbGVfaWQpCi0gICAgICAgIHN1cGVy KFNpbXVsYXRvclByb2Nlc3MsIHNlbGYpLl9raWxsKCkKKyAgICBkZWYgc3RvcChzZWxmLCB0aW1l b3V0X3NlY3M9My4wKToKKyAgICAgICAgb3Mua2lsbChzZWxmLl9waWQsIHNpZ25hbC5TSUdURVJN KQorICAgICAgICBzdXBlcihTaW11bGF0b3JQcm9jZXNzLCBzZWxmKS5zdG9wKHRpbWVvdXRfc2Vj cykK 301411 2017-02-13 16:15:04 -0800 2017-02-14 08:35:03 -0800 Patch bug-168150-20170213161232.patch text/plain 2238 jbedard SW5kZXg6IFRvb2xzL0NoYW5nZUxvZwo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBUb29scy9DaGFuZ2VMb2cJKHJl dmlzaW9uIDIxMjI1NikKKysrIFRvb2xzL0NoYW5nZUxvZwkod29ya2luZyBjb3B5KQpAQCAtMSwz ICsxLDE3IEBACisyMDE3LTAyLTEzICBKb25hdGhhbiBCZWRhcmQgIDxqYmVkYXJkQGFwcGxlLmNv bT4KKworICAgICAgICBydW4td2Via2l0LXRlc3RzIGZvciBpT1MgU2ltdWxhdG9yIGFsd2F5cyBj b21wbGFpbnMgdGhhdCBzdG9wcGluZyBXZWJLaXRUZXN0UnVubmVyQXBwLmFwcCB0aW1lcyBvdXQK KyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTE2ODE1MAor CisgICAgICAgIFJldmlld2VkIGJ5IEFsZXhleSBQcm9za3VyeWFrb3YuCisKKyAgICAgICAgKiBT Y3JpcHRzL3dlYmtpdHB5L3BvcnQvc2ltdWxhdG9yX3Byb2Nlc3MucHk6CisgICAgICAgIChTaW11 bGF0b3JQcm9jZXNzLnN0b3ApOiBDbG9zZSBhcHAgYmVmb3JlIGNhbGxpbmcgU2VydmVyUHJvY2Vz cydzIHN0b3AuCisgICAgICAgIChTaW11bGF0b3JQcm9jZXNzLl9raWxsKTogTW92ZSBjbG9zaW5n IG9mIGFwcCBpbnRvIHN0b3AoKS4KKyAgICAgICAgKiBTY3JpcHRzL3dlYmtpdHB5L3hjb2RlL3Np bXVsYXRvci5weToKKyAgICAgICAgKERldmljZS5sYXVuY2hfYXBwKToKKyAgICAgICAgKERldmlj ZS50ZXJtaW5hdGVfYXBwKTogRGVsZXRlZC4KKwogMjAxNy0wMi0xMyAgV2Vuc29uIEhzaWVoICA8 d2Vuc29uX2hzaWVoQGFwcGxlLmNvbT4KIAogICAgICAgICBbV0syXSBBZGQgdGVzdCBpbmZyYXN0 cnVjdHVyZSBhbmQgdW5pdCB0ZXN0cyBmb3IgZGF0YSBpbnRlcmFjdGlvbgpJbmRleDogVG9vbHMv U2NyaXB0cy93ZWJraXRweS9wb3J0L3NpbXVsYXRvcl9wcm9jZXNzLnB5Cj09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0t IFRvb2xzL1NjcmlwdHMvd2Via2l0cHkvcG9ydC9zaW11bGF0b3JfcHJvY2Vzcy5weQkocmV2aXNp b24gMjEyMjU2KQorKysgVG9vbHMvU2NyaXB0cy93ZWJraXRweS9wb3J0L3NpbXVsYXRvcl9wcm9j ZXNzLnB5CSh3b3JraW5nIGNvcHkpCkBAIC0xMjksNiArMTI5LDYgQEAgY2xhc3MgU2ltdWxhdG9y UHJvY2VzcyhTZXJ2ZXJQcm9jZXNzKToKIAogICAgICAgICBzZWxmLl9wcm9jID0gU2ltdWxhdG9y UHJvY2Vzcy5Qb3BlbihzZWxmLl9waWQsIHN0ZGluLCBzdGRvdXQsIHN0ZGVycikKIAotICAgIGRl ZiBfa2lsbChzZWxmKToKLSAgICAgICAgc2VsZi5fZGV2aWNlLnRlcm1pbmF0ZV9hcHAoc2VsZi5f YnVuZGxlX2lkKQotICAgICAgICBzdXBlcihTaW11bGF0b3JQcm9jZXNzLCBzZWxmKS5fa2lsbCgp CisgICAgZGVmIHN0b3Aoc2VsZiwgdGltZW91dF9zZWNzPTMuMCk6CisgICAgICAgIG9zLmtpbGwo c2VsZi5fcGlkLCBzaWduYWwuU0lHVEVSTSkKKyAgICAgICAgc3VwZXIoU2ltdWxhdG9yUHJvY2Vz cywgc2VsZikuc3RvcCh0aW1lb3V0X3NlY3MpCkluZGV4OiBUb29scy9TY3JpcHRzL3dlYmtpdHB5 L3hjb2RlL3NpbXVsYXRvci5weQo9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09Ci0tLSBUb29scy9TY3JpcHRzL3dlYmtpdHB5 L3hjb2RlL3NpbXVsYXRvci5weQkocmV2aXNpb24gMjEyMjU2KQorKysgVG9vbHMvU2NyaXB0cy93 ZWJraXRweS94Y29kZS9zaW11bGF0b3IucHkJKHdvcmtpbmcgY29weSkKQEAgLTMwOSw5ICszMDks NiBAQCBjbGFzcyBEZXZpY2Uob2JqZWN0KToKICAgICAgICAgICAgIHJhaXNlIFJ1bnRpbWVFcnJv cignRmFpbGVkIHRvIGZpbmQgcHJvY2VzcyBpZCBmb3Ige306IHt9Jy5mb3JtYXQoYnVuZGxlX2lk LCBvdXRwdXQpKQogICAgICAgICByZXR1cm4gaW50KG1hdGNoLmdyb3VwKCdwaWQnKSkKIAotICAg IGRlZiB0ZXJtaW5hdGVfYXBwKHNlbGYsIGJ1bmRsZV9pZCk6Ci0gICAgICAgIHJldHVybiBub3Qg c2VsZi5faG9zdC5leGVjdXRpdmUucnVuX2NvbW1hbmQoWyd4Y3J1bicsICdzaW1jdGwnLCAndGVy bWluYXRlJywgc2VsZi51ZGlkLCBidW5kbGVfaWRdLCByZXR1cm5fZXhpdF9jb2RlPVRydWUpCi0K ICAgICBkZWYgX19lcV9fKHNlbGYsIG90aGVyKToKICAgICAgICAgcmV0dXJuIHNlbGYudWRpZCA9 PSBvdGhlci51ZGlkCiAK