Re: [Libvncserver-common] rfbMakeRichCursorFromXCursor crash

> -----Ursprüngliche Nachricht-----
> Von: Christian Beier <in...@ch...>
> Gesendet: Mittwoch, 28. September 2022 15:22
> An: Johann Obermayr via LibVNCServer-common <libvncserver-
> co...@li...>
> Cc: Johann Obermayr <joh...@si...>
> Betreff: Re: [Libvncserver-common] rfbMakeRichCursorFromXCursor crash
> 
> Am Wed, 28 Sep 2022 12:53:03 +0000
> schrieb Johann Obermayr via LibVNCServer-common
> <lib...@li...>:
> 
> > Hello,
> >
> > i get a crash in rfbMakeRichCursorFromXCursor
> >
> > #0  0xf65fbef0 in rfbMakeRichCursorFromXCursor () from
> > /usr/lib/libvncserver.so.1 #1  0xf65fc370 in rfbSendCursorShape ()
> > from
> > /usr/lib/libvncserver.so.1 #2  0xf65ee034 in rfbSendFramebufferUpdate
> > () from
> > /usr/lib/libvncserver.so.1 #3  0xf65e8440 in rfbUpdateClient () from
> > /usr/lib/libvncserver.so.1 #4  0xf65e84d4 in rfbProcessEvents () from
> > /usr/lib/libvncserver.so.1 #5  0x00147f00 in processClientEvents ()
> >
> > After add some logging I see, that "cursor->source" is NULL.
> > So it crash at line
> > if(cursor->source[ j * w + i / 8] & bit)
> >
> > Is this a knowing bug ?
> > Exists there a bug fix ?
> >
> > Thanks & best regards
> >   Johann
> >
> 
> Hi Johann,
> 
> I don't know which version you are using, but you might want to scan
> through the list of known issues at
> https://github.com/LibVNC/libvncserver/issues
> 
> If the bug is not there, please create an issue.
> 
> Best,
> --
> Christian Beier
> 
> what is, is; what is not is possible.

Hello,

After more test, I found 
	SHA-1: d138cf90130b0e8d5062f136ecdbcaa85e734d5d

	* libvncserver: don't NULL out internal of the default cursor

	...otherwise an rfbScreen created after rfbScreenCleanup() was called
	gets assigned an invalid cursor struct.

This solve my problem.

Are there any plans to create a "LibVNCServer-0.9.14" release ?

Best regards
  Johann