Assert Function is non-null in RunLoop::dispatch
Created attachment 368452 [details] Patch
Comment on attachment 368452 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=368452&action=review > Source/WTF/wtf/RunLoop.cpp:129 > + RELEASE_ASSERT(function); I am not convinced this will help us debug rdar://problem/48679972 since in rdar://problem/48679972 we're crashing when trying to call a CompletionHandler, not a Function. I think what ends up happening is that somebody calls dispatch() with a Function that wraps a CompletionHandler whose m_function is null. This case would not get caught by your proposed assertion.
Reopening to try a different approach.
Created attachment 369194 [details] Patch
Comment on attachment 369194 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=369194&action=review > Source/WebKit/ChangeLog:10 > + * NetworkProcess/cache/CacheStorageEngine.cpp: Why only these 4? It seems there are other calls to RunLoop::main().dispatch() in the networkProcess which pass a CompletionHandler. E.g. saveCookies() in NetworkProcessCocoa.
I've decided to try https://bugs.webkit.org/show_bug.cgi?id=197641 instead
Comment on attachment 369194 [details] Patch Attachment 369194 [details] did not pass win-ews (win): Output: https://webkit-queues.webkit.org/results/12119790 New failing tests: http/tests/css/filters-on-iframes.html
Created attachment 369236 [details] Archive of layout-test-results from ews215 for win-future The attached test failures were seen while running run-webkit-tests on the win-ews. Bot: ews215 Port: win-future Platform: CYGWIN_NT-10.0-17763-3.0.5-338.x86_64-x86_64-64bit