On Thu, Apr 16, 2015 at 05:27:27PM +0100, Zeeshan Ali (Khattak) wrote: > On Thu, Apr 16, 2015 at 4:44 PM, Christophe Fergeau <cfergeau@xxxxxxxxxx> wrote: > > On Thu, Apr 16, 2015 at 04:23:56PM +0100, Zeeshan Ali (Khattak) wrote: > >> Sure, how do I do that? > > > > New <device-class> node? > > So we go through the pain of adding new API to honour a > differentiation that we know is unlikely to be significant? This is the right way of expressing the information you need. Maybe it's overkill, maybe it would avoid some problems in the future... (some applications start explicitly checking for that USB mouse model, then we realize it really is a special VMWare device and this causes issues in other applications, and we'll have to break one or the other). You want a quick and dirty solution, that's fine, I can be convinced this way, but don't try to convince me this is the best solution ;) > > If you want me to add the exact devices that qemu uses, I'm willing to > do that or we can go with these patches. Just tell me which you > prefer. I said in a previous email that the tablet device already did not correspond to the one QEMU would add to guests. > > > The next issue here is that saying that Windows XP does not support USB > > mouses/keyboard is incorrect. The magic knowledge encoding that XP automatic > > installation should not use USB keyboard/mouse would make more sense in > > libvirt-designer/libvirt-builder (or Boxes in the mean time). > > Sure, I will update commit log. The patch needs changing as currently it only adds the devices to win7 and later while WinXP also supports USB keyboards. Christophe
Attachment:
pgp4NhvteUAZf.pgp
Description: PGP signature
_______________________________________________ Libosinfo mailing list Libosinfo@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libosinfo
