[Development] CI broken again?

Thiago Macieira thiago.macieira at intel.com
Fri Feb 28 23:46:06 CET 2014


Em qui 20 fev 2014, às 06:21:44, Sarajärvi Tony escreveu:
> QDEBUG : tst_QTcpSocket::blockingIMAP(WithSocks5Proxy SSL)
> QSocks5SocketEngine(0x99e730) _q_controlSocketReadNotification more bytes
> available, calling _q_controlSocketReadNotification() again , read
> notifications enabled? false

This debug statement is nowhere in the Qt sources.

It's exactly where the problem occurred, but I can't locate where that might 
have been.

The debug message would indicate that it's in _q_controlSocketReadNotification, 
but that function never recurses back into itself. So that can't be.

That function is only called by the readyRead() signal of the control socket 
and by QSocks5SocketEngine::initialize (the overload taking a socket 
descriptor). But the rest of the debug log suggests that this was a regular 
connection, not an accepted socket via QTcpServer.

My conclusion is that the test output is bogus, out of running a patched 
version of Qt. Therefore, my recommendation is to remove the patches.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center




More information about the Development mailing list