On Thu, Dec 24, 2009 at 08:38:32AM +0200, Avi Kivity wrote: > On 12/24/2009 02:17 AM, Aurelien Jarno wrote: > >On Tue, Dec 22, 2009 at 11:57:02AM +0200, Avi Kivity wrote: > >>The first such option rom will load at address 0, which isn't very nice, > >>and the second will report a conflict and abort, which is horrible. > >> > >>Signed-off-by: Avi Kivity<a...@redhat.com> > >>--- > >> > >>Changes from v1: > >>- use ->fw_file instead of ->addr to distinguish between host-loaded and > >> firmware-loaded roms > >>- add the same check in a couple more places > >This patch is a lot better and does not break the versatile platform > >anymore. I do wonder though if the same change should also be done for > >find_rom() and rom_copy(). > > Since I'm not sure what these are used for, I left them as is. IMO > the API should be improved by splitting the functions dealing with > qemu-loaded and firmware-loaded ROMs to avoid confusion. >
I have committed your patch to both stable-0.12 and HEAD. As I think these functions have to be changed too, I have committed a patch to do it, but to HEAD only. -- Aurelien Jarno GPG: 1024D/F1BCDB73 aurel...@aurel32.net http://www.aurel32.net
