On Thu, 2009-07-16 at 10:57 -0700, Jordan Justen wrote: > On Wed, Jul 15, 2009 at 10:37 PM, Sheng Yang<sheng@xxxxxxxxxxxxxxx> wrote: > > Make sense to me. So what's mattered here is not bios, but qemu-kvm and kvm > > code. The user can replace bios binary by UEFI binary easily, but not with kvm > > related part. > > > > I realized you still need separate the qemu-kvm patch into two: one for bios > > and another for qemu. > > Okay, I will split this change apart. > > > And I just hope this modification won't break some old > > OS(any OS got assumption about bios size? I think Tiano team should have idea > > on it)... > > I don't think the OS should care about the bios size, but it is > important to update the INT15-E820 memory ranges to help > make sure the OS knows which regions are in use. > > Even without the E820 change, I think it would be unlikely > for an OS to try utilize these memory regions, but it is > definitely better to properly describe it. Hello, I don't know if I get this right, but is it enough to apply this patch to KVM-userspace, or do I also need to have a patched kernel in order to get large images (my image is 4MB) to work in KVM? Thanks, Cristi -- Ing. Cristi Măgherușan, System/Network Engineer Technical University of Cluj-Napoca, Romania http://cc.utcluj.ro +40264 401247
Attachment:
signature.asc
Description: This is a digitally signed message part
