Summary: | Web Inspector: REGRESSION(r266885): dyld: Symbol not found: __ZN9Inspector17BackendDispatcher12sendResponseElON3WTF6RefPtrINS1_8JSONImpl6ObjectENS1_13DumbPtrTraitsIS4_EEEEb | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Devin Rousso <hi> | ||||
Component: | Web Inspector | Assignee: | Devin Rousso <hi> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | bburg, ebrahim, ews-watchlist, hi, inspector-bugzilla-changes, joepeck, keith_miller, m13590491852, marcos, mark.lam, msaboff, saam, tzagallo, webkit-bug-importer | ||||
Priority: | P2 | Keywords: | InRadar | ||||
Version: | WebKit Nightly Build | ||||||
Hardware: | All | ||||||
OS: | All | ||||||
See Also: |
https://bugs.webkit.org/show_bug.cgi?id=179847 https://bugs.webkit.org/show_bug.cgi?id=216675 |
||||||
Bug Depends on: | 216302 | ||||||
Bug Blocks: | |||||||
Attachments: |
|
Description
Devin Rousso
2020-09-14 11:26:41 PDT
Created attachment 408729 [details]
Patch
Comment on attachment 408729 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=408729&action=review rs=me > Source/JavaScriptCore/inspector/InspectorBackendDispatcher.cpp:201 > + sendResponse(requestId, result.releaseNonNull(), false); Should it send the `bool` passed into this function, even though it is "unused"? Comment on attachment 408729 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=408729&action=review >> Source/JavaScriptCore/inspector/InspectorBackendDispatcher.cpp:201 >> + sendResponse(requestId, result.releaseNonNull(), false); > > Should it send the `bool` passed into this function, even though it is "unused"? IIRC this `bool` only exists to differentiate this `sendResponse` from the `sendResponse` in <https://trac.webkit.org/browser/trunk/Source/JavaScriptCore/inspector/cocoa/DeprecatedInspectorValues.cpp#L195>. This also matches the behavior before r266885. Committed r267028: <https://trac.webkit.org/changeset/267028> All reviewed patches have been landed. Closing bug and clearing flags on attachment 408729 [details]. This seems to have regressed again... STR: 1. build-webkit --debug 2. run-safari --debug Expected: Safari to start up Actual: ``` Starting SafariForWebKitDevelopment with DYLD_FRAMEWORK_PATH set to point to built WebKit in /Users/mcaceres/dev/WebKit/WebKitBuild/Debug. dyld: Symbol not found: __ZN9Inspector17BackendDispatcher12sendResponseElON3WTF6RefPtrINS1_8JSONImpl6ObjectENS1_13DumbPtrTraitsIS4_EEEEb Referenced from: /System/Library/PrivateFrameworks/WebInspector.framework/Versions/A/WebInspector Expected in: /Users/mcaceres/dev/WebKit/WebKitBuild/Debug/JavaScriptCore.framework/Versions/A/JavaScriptCore in /System/Library/PrivateFrameworks/WebInspector.framework/Versions/A/WebInspector ``` Marcos, this is fallout from JavaScriptCore breaking ABI compatibility. For this to not be hit will require the changes in WebInspector.framework to be in your system library. This won't happen until you update Safari. The fix for this should be in Safari 14.1 beta, which is available via here: https://developer.apple.com/safari/download/ Thanks, BJ. Appreciate the update! I'll give the beta a go. *** Bug 221528 has been marked as a duplicate of this bug. *** |