libwebrtc does not like it currently. Setting MaxBundle will then remove the possibility to set other fields.
Created attachment 303855 [details] Patch
Comment on attachment 303855 [details] Patch Clearing flags on attachment: 303855 Committed r213613: <http://trac.webkit.org/changeset/213613>
All reviewed patches have been landed. Closing bug.
We should find a way to support this option.
rdar://problem/33006091
Created attachment 313929 [details] Patch
Comment on attachment 313929 [details] Patch Attachment 313929 [details] did not pass mac-wk2-ews (mac-wk2): Output: http://webkit-queues.webkit.org/results/4007096 New failing tests: fast/mediastream/RTCSessionDescription.html imported/w3c/web-platform-tests/webrtc/RTCDataChannel-id.html imported/w3c/web-platform-tests/webrtc/rtcpeerconnection/rtcpeerconnection-constructor.html
Created attachment 313932 [details] Archive of layout-test-results from ews105 for mac-elcapitan-wk2 The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews. Bot: ews105 Port: mac-elcapitan-wk2 Platform: Mac OS X 10.11.6
Comment on attachment 313929 [details] Patch Attachment 313929 [details] did not pass ios-sim-ews (ios-simulator-wk2): Output: http://webkit-queues.webkit.org/results/4007114 New failing tests: imported/w3c/web-platform-tests/webrtc/RTCDataChannel-id.html imported/w3c/web-platform-tests/webrtc/RTCPeerConnection-createAnswer.html imported/w3c/web-platform-tests/webrtc/rtcpeerconnection/rtcpeerconnection-constructor.html
Created attachment 313934 [details] Archive of layout-test-results from ews124 for ios-simulator-wk2 The attached test failures were seen while running run-webkit-tests on the ios-sim-ews. Bot: ews124 Port: ios-simulator-wk2 Platform: Mac OS X 10.12.5
Created attachment 314038 [details] Patch
Comment on attachment 314038 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=314038&action=review > LayoutTests/fast/mediastream/RTCPeerConnection-getConfiguration-expected.txt:56 > -PASS successfullyParsed is true > +FAIL iceServers.length should be 1. Was 2. This seems like a test regression.
(In reply to Alex Christensen from comment #12) > Comment on attachment 314038 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=314038&action=review > > > LayoutTests/fast/mediastream/RTCPeerConnection-getConfiguration-expected.txt:56 > > -PASS successfullyParsed is true > > +FAIL iceServers.length should be 1. Was 2. > > This seems like a test regression. Right. Previously, we were trying to set the configuration, failing to do so at libwebrtc backend but reporting that this was done at RTCPeerConnection level. With the changes done to RTCPeerConnection::setConfiguration, we will report more accurately the actual configuration.
Comment on attachment 314038 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=314038&action=review > Source/WebCore/Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.cpp:83 > + else if (configuration.bundlePolicy == RTCBundlePolicy::MaxBundle) This should probably be a switch with no default so we will see compiler failures if another one is added.
Created attachment 314043 [details] Patch
(In reply to youenn fablet from comment #15) > Created attachment 314043 [details] > Patch Reduced patch without the setConfiguration behavior change.
(In reply to Alex Christensen from comment #14) > Comment on attachment 314038 [details] > Patch > > View in context: > https://bugs.webkit.org/attachment.cgi?id=314038&action=review > > > Source/WebCore/Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.cpp:83 > > + else if (configuration.bundlePolicy == RTCBundlePolicy::MaxBundle) > > This should probably be a switch with no default so we will see compiler > failures if another one is added. Will do that as a follow-up or at landing time. I probably also need to update the fixes about the setConfiguration issue with a bugzilla entry.
(In reply to youenn fablet from comment #16) > (In reply to youenn fablet from comment #15) > > Created attachment 314043 [details] > > Patch > > Reduced patch without the setConfiguration behavior change. Filed bug 173938 for setConfiguration issue
Created attachment 314056 [details] Correcting style and adding bug link
Comment on attachment 314056 [details] Correcting style and adding bug link Clearing flags on attachment: 314056 Committed r218903: <http://trac.webkit.org/changeset/218903>
This patch makes more tests flaky in the bots. The only change I see is that we are now really disabling CPU overuse detection with this patch. This might cause issues with the bots. If that is the case, we might actually want to activate CPU overuse detection, at least on the bots.
Reopening to attach new patch.
Created attachment 314163 [details] Patch
(In reply to youenn fablet from comment #24) > Created attachment 314163 [details] > Patch Patch to disable cpu overuse detection. This will allow verifying whether this is what is causing issues with bots.
The commit-queue encountered the following flaky tests while processing attachment 314163 [details]: workers/bomb.html bug 171985 (author: fpizlo@apple.com) The commit-queue is continuing to process your patch.
Comment on attachment 314163 [details] Patch Clearing flags on attachment: 314163 Committed r218963: <http://trac.webkit.org/changeset/218963>
I looks like the patch did not fix the issue with the test it is still failing on all WK2 testers. https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=imported%2Fw3c%2Fweb-platform-tests%2Fwebrtc%2Fsimplecall.html https://build.webkit.org/builders/Apple%20Sierra%20Release%20WK2%20(Tests)/builds/2628 Talked with Youenn, rolling out both patches.
Reverted r218963 for reason: This patch and its fix cause immediate flakiness on all WK2 testers Committed r218972: <http://trac.webkit.org/changeset/218972>
Reverted r218903 for reason: This patch and its fix cause immediate flakiness on all WK2 testers Committed r218973: <http://trac.webkit.org/changeset/218973>
Created attachment 314265 [details] Disabling ice candidate pool size
(In reply to youenn fablet from comment #32) > Created attachment 314265 [details] > Disabling ice candidate pool size WK2 testers might have issues with ice candidate pool size
Comment on attachment 314265 [details] Disabling ice candidate pool size Clearing flags on attachment: 314265 Committed r218994: <http://trac.webkit.org/changeset/218994>