Bug 206904 – Crash under com.apple.WebKit.Networking at WebKit: WebKit::WebIDBServer::suspend

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Sihui Liu]

Thread[0] EXC_BAD_ACCESS (SIGSEGV) (KERN_INVALID_ADDRESS at 0x00000000000000a8)
   94  	    m_isSuspended = true;
-> 95  	    m_server->lock().lock();
   96  	    m_server->stopDatabaseActivitiesOnMainThread();

15 WebKit: WebKit::WebIDBServer::suspend(WebKit::WebIDBServer::ShouldForceStop) <==
  15 WebKit: WebKit::WebIDBServer::suspend(WebKit::WebIDBServer::ShouldForceStop)
    15 WebKit: WebKit::NetworkProcess::prepareToSuspend(bool, WTF::CompletionHandler<void ()>&&)
      15 WebKit: void IPC::handleMessageAsync<Messages::NetworkProcess::PrepareToSuspend, WebKit::NetworkProcess, void (WebKit::NetworkProcess::*)(bool, WTF::CompletionHandler<void ()>&&)>(IPC::Connection&, IPC::Decoder&, WebKit::NetworkProcess*, void (WebKit::NetworkProcess::*)(bool, WTF::CompletionHandler<void ()>&&))
          15 WebKit: WebKit::NetworkProcess::didReceiveNetworkProcessMessage(IPC::Connection&, IPC::Decoder&)

[Sihui Liu]

<rdar://problem/58791603>

[Sihui Liu]

Created attachment 389054 [details]
Patch

[Maciej Stachowiak]

Comment on attachment 389054 [details]
Patch

r=me (seems like this could block the main thread of the network process though, unless I misunderstand what happens on which thread.

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 389054 [details]:

editing/spelling/spellcheck-async-remove-frame.html bug 158401 (authors: [e-mail], [e-mail], and [e-mail])
The commit-queue is continuing to process your patch.

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 389054 [details]:

editing/spelling/spellcheck-attribute.html bug 206178 (authors: [e-mail], [e-mail], and [e-mail])
imported/w3c/web-platform-tests/xhr/send-send.any.worker.html bug 206345 (author: [e-mail])
The commit-queue is continuing to process your patch.

[WebKit Commit Bot]

Comment on attachment 389054 [details]
Patch

Clearing flags on attachment: 389054

Committed r255375: <https://trac.webkit.org/changeset/255375>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Geoffrey Garen]

Can we just initialize m_server directly in the constructor on the main thread, instead of posting a task? Since no tasks have been queued yet, there's nothing to synchronize with yet.

Specifically, maybe we can initialize m_server directly, and then post a task to upgradeFilesIfNecessary() (and change IDBServer::IDBServer() not to upgradeFilesIfNecessary(), to avoid I/O on the main thread).