On Tue, Oct 23, 2012 at 12:20:08PM +0200, Alexander Graf wrote:
>
> On 23.10.2012, at 12:09, Avik Sil wrote:
>
> > This patch makes default boot order machine specific instead of
> > set globally. The default boot order can be set per machine in
> > QEMUMachine default_machine_opts, or by the command line using
> > -machine <machine_name>,boot=<boot_devices>, or by standard -boot
> > option. This allows a machine to receive a NULL boot order when
> > -boot isn't used and take an appropriate action accordingly. This
> > helps machine boots from the devices as set in guest's non-volatile
> > memory location in case no boot order is provided by the user.
> >
> > Signed-off-by: Avik Sil <avik...@linux.vnet.ibm.com>
> > ---
> > hw/nseries.c | 2 ++
> > hw/pc_piix.c | 13 ++++++++++++-
> > hw/ppc_newworld.c | 1 +
> > hw/ppc_oldworld.c | 1 +
> > hw/ppc_prep.c | 1 +
> > hw/sun4m.c | 10 ++++++++++
> > hw/sun4u.c | 3 +++
>
> That feels like too much magic for my taste. Can't we have some
> default macro (or even the explicit "cad" thing) that we push to
> _every_ machine description out there? The next patch to implement
> boot= for pseries would then remove this default value for the
> pseries machine.
I'm not 100% clear on what you mean here. Do you just mean:
#define DEFAULT_BOOT_ORDER "boot=cad"
and then we use
.default_machine_opts = DEFAULT_BOOT_ORDER,
or something else?
> The way the patch works right now, the "make machine specific" patch
> indirectly also changes the behavior for a bunch of machines.
A bunch? I thought the only machine whose behaviour changed was
pseries. I agree that moving the behaviour change into a separate
patch would be a better idea, but if there are machines other than
pseries with changed behaviour by this patch, then that's a plain old
bug, rather than a mere question of patch elegance.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
