168647 2017-02-21 00:53:20 -0800 [GTK] Several WebRTC tests failing since r212812 2017-02-23 08:00:52 -0800 1 1 1 Unclassified WebKit WebKitGTK WebKit Local Build Unspecified Unspecified RESOLVED DUPLICATE 168736 https://bugs.webkit.org/show_bug.cgi?id=168736 Gtk, LayoutTestFailure P2 Normal --- 167293 1 cgarcia webkit-unassigned achristensen alex aperez bugs-noreply clopez mcatanzaro youennf oldest_to_newest 1279274 0 cgarcia 2017-02-21 00:53:20 -0800 fast/mediastream/RTCPeerConnection-more-media-to-negotiate.html imported/w3c/web-platform-tests/webrtc/datachannel-emptystring.html imported/w3c/web-platform-tests/webrtc/no-media-call.html imported/w3c/web-platform-tests/webrtc/promises-call.html 1279279 1 cgarcia 2017-02-21 00:57:31 -0800 r212699 has just been rolled out in r212704. Closing then, sorry for the noise. 1279834 2 cgarcia 2017-02-22 03:05:47 -0800 Patch landed again with the same failures. 1279852 3 clopez 2017-02-22 04:31:41 -0800 Please, when reporting a bug about new failures, CC the authors of the commit that introduced that to let them know about the issue. 1279904 4 youennf 2017-02-22 08:53:36 -0800 Sorry about breaking these tests here. Mac and GTK ports no longer use the same WebRTC mock backends, hence the test differences. I tried to move GTK test expectations to the GTK platform folder but failed apparently to do so for some tests. I guess the simplest approach might be to do the same for these 4 tests. 1279991 5 youennf 2017-02-22 12:24:15 -0800 I will land https://bugs.webkit.org/show_bug.cgi?id=167293 which should fix some of the failing tests. 1280229 6 alex 2017-02-23 01:29:43 -0800 That patch fixed the tests https://trac.webkit.org/changeset/212817 Thanks 1280248 7 alex 2017-02-23 05:48:59 -0800 My mistake, some of this tests are inside the general TestExpectations and some in the platform, they are not really passing. We have to remove some from the GTK expectations and just check those in this bug. The other ones were added here to the TestExpectations: https://trac.webkit.org/changeset/212848 1280261 8 clopez 2017-02-23 07:37:05 -0800 The 3 tests from imported/w3c/web-platform-tests/webrtc were failing also before r212812, but the expectation for them was just wrong. And fast/mediastream/RTCPeerConnection-more-media-to-negotiate.html is passing with the general expectation but not with the one for GTK+ that got copied from the general at r212338 <https://trac.webkit.org/r212338> $ diff -u LayoutTests/fast/mediastream/RTCPeerConnection-more-media-to-negotiate-expected.txt LayoutTests/platform/gtk/fast/mediastream/RTCPeerConnection-more-media-to-negotiate-expected.txt --- LayoutTests/fast/mediastream/RTCPeerConnection-more-media-to-negotiate-expected.txt 2017-02-16 03:33:41.255592879 +0100 +++ LayoutTests/platform/gtk/fast/mediastream/RTCPeerConnection-more-media-to-negotiate-expected.txt 2017-02-23 16:32:14.768622317 +0100 @@ -18,6 +18,7 @@ PASS pc.getTransceivers().length is 2 Create an answer for the audio only offer Set answer as local description +PASS negotiationneededevent fired: There is local media not negotiated that needs to be negotiated in a follow-up offer PASS Answer set PASS successfullyParsed is true I'm not 100% sure... but I'm going to assume that the general is the right one, remove the specific GTK+ expectation and mark this test as passing. 1280271 9 clopez 2017-02-23 08:00:52 -0800 I have landed https://trac.webkit.org/changeset/212884 unskipping many mediastream/webrtc tests for GTK that were skipped on the general and also marking the expected failures as expected. fast/mediastream/RTCPeerConnection-more-media-to-negotiate.html has been rebaselined to the general expectation And the bug tracking the failures on imported/w3c/web-platform-tests/webrtc/datachannel-emptystring.html imported/w3c/web-platform-tests/webrtc/no-media-call.html and imported/w3c/web-platform-tests/webrtc/promises-call.html is 168736 We can close this *** This bug has been marked as a duplicate of bug 168736 ***