226901 2021-06-10 15:06:01 -0700 [REGRESSION][Curl] Network::ResourceTiming are broken after r278391 2021-06-10 20:38:21 -0700 1 1 1 Unclassified WebKit Web Inspector WebKit Nightly Build Unspecified Unspecified RESOLVED FIXED InRadar P2 Normal --- 1 yurys yurys achristensen basuke ews-watchlist fujii galpeter inspector-bugzilla-changes webkit-bug-importer oldest_to_newest 1768760 0 yurys 2021-06-10 15:06:01 -0700 https://trac.webkit.org/changeset/278391/webkit changed how InspectorNetworkAgent::buildObjectForTiming serializes NetworkLoadMetrics to the inspector protocol. millisecondsSinceFetchStart subtracts fetchStart to get relative values for most of the metrics. In Curl implementation fetchStart is always 0 which results in incorrect values sent over the protocol. 1768814 1 431162 yurys 2021-06-10 16:51:22 -0700 Created attachment 431162 Patch 1768821 2 fujii 2021-06-10 17:02:45 -0700 Don't you need to set redirectStart? Cocoa port does. 1768823 3 fujii 2021-06-10 17:15:06 -0700 It would be great the following WinCairo test failures are fixed. fast/dom/webtiming.html [ Failure ] http/tests/misc/webtiming-cross-origin-redirect.py [ Failure ] http/tests/misc/webtiming-one-redirect.py [ Failure ] http/tests/misc/webtiming-two-redirects.py [ Failure ] 1768824 4 yurys 2021-06-10 17:24:08 -0700 (In reply to Fujii Hironori from comment #2) > Don't you need to set redirectStart? Cocoa port does. I think you are write, I noticed that redirectSart was also missing but decided not to touch it since it was not a part of the regression, can add that too. 1768864 5 ews-feeder 2021-06-10 20:37:26 -0700 Committed r278752 (238713@main): <https://commits.webkit.org/238713@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 431162. 1768865 6 webkit-bug-importer 2021-06-10 20:38:21 -0700 <rdar://problem/79176779> 431162 2021-06-10 16:51:22 -0700 2021-06-10 20:37:28 -0700 Patch bug-226901-20210610165122.patch text/plain 1545 yurys U3VidmVyc2lvbiBSZXZpc2lvbjogMjc4NzM5CmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9D aGFuZ2VMb2cgYi9Tb3VyY2UvV2ViQ29yZS9DaGFuZ2VMb2cKaW5kZXggZDlkNmFlYzMyMDY4Y2Uy MDJhYjNlZTEzNjM0NzEzOTM0Njc0NGQwNS4uZWFmOWY5YmRmOGJhZjFhYWIxMTA5ZDdhZjVhYTc0 ZjQ0NWNkZmRjZSAxMDA2NDQKLS0tIGEvU291cmNlL1dlYkNvcmUvQ2hhbmdlTG9nCisrKyBiL1Nv dXJjZS9XZWJDb3JlL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE3IEBACisyMDIxLTA2LTEwICBZdXJ5 IFNlbWlraGF0c2t5ICA8eXVyeXNAY2hyb21pdW0ub3JnPgorCisgICAgICAgIFtSRUdSRVNTSU9O XVtDdXJsXSBOZXR3b3JrOjpSZXNvdXJjZVRpbWluZyBhcmUgYnJva2VuIGFmdGVyIHIyNzgzOTEK KyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTIyNjkwMQor CisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorCisgICAgICAgIEluaXRpYWxp emUgZmV0Y2hTdGFydCB3aXRoIHN0YXJ0VGltZSBpbiBDdXJsLgorCisgICAgICAgIE5vIG5ldyB0 ZXN0cy4KKworICAgICAgICAqIHBsYXRmb3JtL25ldHdvcmsvY3VybC9DdXJsQ29udGV4dC5jcHA6 CisgICAgICAgIChXZWJDb3JlOjpDdXJsSGFuZGxlOjpnZXROZXR3b3JrTG9hZE1ldHJpY3MpOgor CiAyMDIxLTA2LTEwICBBbGV4IENocmlzdGVuc2VuICA8YWNocmlzdGVuc2VuQHdlYmtpdC5vcmc+ CiAKICAgICAgICAgTW92ZSBUaW1pbmctQWxsb3ctT3JpZ2luIGNoZWNrcyB0byB0aGUgbmV0d29y ayBwcm9jZXNzCmRpZmYgLS1naXQgYS9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9uZXR3b3JrL2N1 cmwvQ3VybENvbnRleHQuY3BwIGIvU291cmNlL1dlYkNvcmUvcGxhdGZvcm0vbmV0d29yay9jdXJs L0N1cmxDb250ZXh0LmNwcAppbmRleCA4ZWQwOWIyYWJjNWQ2NDdmYWI2ZGMwMWI5ODYxNmUxNTNm Njc4YjRlLi5lMWM2Y2M0M2E4NWVlN2JhYzM5MWNhNDk0YWRhNWY0YzA1ZDA5ZjZjIDEwMDY0NAot LS0gYS9Tb3VyY2UvV2ViQ29yZS9wbGF0Zm9ybS9uZXR3b3JrL2N1cmwvQ3VybENvbnRleHQuY3Bw CisrKyBiL1NvdXJjZS9XZWJDb3JlL3BsYXRmb3JtL25ldHdvcmsvY3VybC9DdXJsQ29udGV4dC5j cHAKQEAgLTc5Myw2ICs3OTMsNyBAQCBzdGQ6Om9wdGlvbmFsPE5ldHdvcmtMb2FkTWV0cmljcz4g Q3VybEhhbmRsZTo6Z2V0TmV0d29ya0xvYWRNZXRyaWNzKE1vbm90b25pY1RpbQogCiAgICAgTmV0 d29ya0xvYWRNZXRyaWNzIG5ldHdvcmtMb2FkTWV0cmljczsKIAorICAgIG5ldHdvcmtMb2FkTWV0 cmljcy5mZXRjaFN0YXJ0ID0gc3RhcnRUaW1lOwogICAgIG5ldHdvcmtMb2FkTWV0cmljcy5kb21h aW5Mb29rdXBTdGFydCA9IHN0YXJ0VGltZTsKICAgICBuZXR3b3JrTG9hZE1ldHJpY3MuZG9tYWlu TG9va3VwRW5kID0gc3RhcnRUaW1lICsgU2Vjb25kcyhuYW1lTG9va3VwKTsKICAgICBuZXR3b3Jr TG9hZE1ldHJpY3MuY29ubmVjdFN0YXJ0ID0gbmV0d29ya0xvYWRNZXRyaWNzLmRvbWFpbkxvb2t1 cEVuZDsK