On 19.10.2008, at 12:27, Avi Kivity wrote:
Alexander Graf wrote:
So far we're talking about reissuing pio. The only case where the
guest and host pio are likely to have the same addresses is the
vga ports, and these aren't heavily used.
So it's better to leave it disabled, I guess. The IOPM is only
relevant when passing through IO ports 1:1. And parsing the IOPM
takes a lot of effort (read: time). It's also only relevant if the
same device, the guest is exposed to, is passed through to the l2
guest. IMHO speed is more important atm.
Okay. Please drop those bits then.
Can do. I only implemented them for functional completeness in the
first place, but didn't want to affect performance because of them -
that's why I ifdef'ed them out.
Any more requests for v5?
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html