On 3/10/23 09:28, David Woodhouse wrote:
On Fri, 2023-03-10 at 11:15 +0800, Xiaoyao Li wrote:
On 3/1/2023 9:51 PM, David Woodhouse wrote:
From: David Woodhouse <d...@amazon.co.uk>
The xen_overlay device (and later similar devices for event channels and
grant tables) need to be instantiated. Do this from a kvm_type method on
the PC machine derivatives, since KVM is only way to support Xen emulation
for now.
Just curious, isn't there any more reasonable place to add machine
specific initialization?
abusing the mc->kvm_type() looks bad to me.
Hm, good question. Off the top of my head I have no better answer than
"Paolo made me do it":
https://lore.kernel.org/qemu-devel/8495140d-3301-7693-b804-055416680...@redhat.com/
But I have gained a bit more clue since December, and reading that
message again I'll put a lot more focus on the fact that he said
"during mc->kvm_type OR AFTERWARDS".
I don't think this is abusing mc->kvm_type; that is the point where
startup code tells the machine "now you have your accelerator
configuration, do what you want with that info". In fact I find using
xen_enabled() in mc->kvm_type to be less ugly than having a MachineClass
entry just for KVM. :)
But, if you want to move it to pc_basic_device_init() that is certainly
okay too. It's not like people are going to add TCG/Xen support
tomorrow but it is a tiny step in the direction of less
accelerator-specific code for Xen emulation.
Paolo
