On Tue, Aug 08, 2006 at 06:19:16PM -0400, Daniel Veillard wrote: > On Tue, Aug 08, 2006 at 11:04:44PM +0100, Daniel P. Berrange wrote: > > On Mon, Aug 07, 2006 at 12:11:17PM -0400, Daniel Veillard wrote: > > diff -c -r1.10 driver.h > > *** src/driver.h 8 Aug 2006 22:22:55 -0000 1.10 > > --- src/driver.h 8 Aug 2006 22:53:55 -0000 > > *************** > > *** 104,109 **** > > --- 104,112 ---- > > typedef int > > (*virDrvDomainRestore) (virConnectPtr conn, > > const char *from); > > + typedef char * > > + (*virDrvDomainDumpXML) (virDomainPtr dom, > > + int flags); > > > > typedef int > > (*virDrvDomainSetVcpus) (virDomainPtr domain, > > *************** > > *** 164,169 **** > > --- 167,173 ---- > > virDrvDomainSetVcpus domainSetVcpus; > > virDrvDomainPinVcpu domainPinVcpu; > > virDrvDomainGetVcpus domainGetVcpus; > > + virDrvDomainDumpXML domainDumpXML; > > }; > > > > that and the updates of the driver tables will conflict with the > change I just commited to migrate vCPU and affinity functions to the > driver system too. Sorry :-) Actually it should't - I updated to CVS HEAD and merged wrt to your changes just before submitting this patch, so I think its OK. > > ! NULL, /* domainGetVcpus */ > > ! xenProxyDomainDumpXML, /* domainDumpXML */ > > }; > > usually I drop the coma after the last field of a structure. I force gcc > in a very pedantic mode and it complains about this :-) I prefer to keep the trailing comma because it means one-line less diff next time we add more ffunctions to the driver API. I'll chop them out though to match your existing coding style in libvirt sources. > Your patch also includes other changes already commited about TTY and VNC > port extraction, maybe you were not working from a CVS fully updated ? > > but that looks good, the merge may be a bit annoying though :-) I needed to modify some of the signatures on the methods I added for TTY and VNC port extraction because the proxy daemon doesn't have access to the full virDomainPtr object, only the virConnectPtr and domain ID. So that's why these functiosn appear in the diff - the patch was taken from 'cvs diff -c' against a checkout updated to HEAD so I don;t expect any merge issues. Regards, Dan. -- |=- Red Hat, Engineering, Emerging Technologies, Boston. +1 978 392 2496 -=| |=- Perl modules: http://search.cpan.org/~danberr/ -=| |=- Projects: http://freshmeat.net/~danielpb/ -=| |=- GnuPG: 7D3B9505 F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 -=|