Andy Lutomirski <l...@amacapital.net> writes: > On Mon, Apr 14, 2014 at 1:15 AM, Markus Armbruster <arm...@redhat.com> wrote: >> Peter Crosthwaite <peter.crosthwa...@xilinx.com> writes: >> >>> Hi Andy, >>> >>> On Thu, Apr 10, 2014 at 5:55 AM, Andy Lutomirski <l...@amacapital.net> >>> wrote: >>>> Currently, -M q35 boots linux quite a bit slower than the default >>>> machine type. This seems to be because it takes a few hundred ms to >>>> determine that there's nothing attached to the AHCI controller. >>>> >>>> In virtio setups, there will probably never be anything attached to >>>> the AHCI controller. Would it be possible to add something like >>>> -machine default_storage=off to turn off default storage devices? >>>> This could include the AHCI on q35 and the cdrom and such on pc. >>>> >>>> There's precedent: -machine usb=off turns off the default USB >>>> controllers, which is great for setups that use xhci. >>>> >>> >>> Is there a more generic solution to your problem? Can you implement >>> command line device removal in a non specific way and avoid having to >>> invent AHCI or even "storage" specific arguments. You could >>> considering bringing the xhci use case you mentioned under the same >>> umbrella. >> >> USB has always been off by default, at least for the boards I'm familiar >> with, due to the USB emulation's non-trivial CPU use. >> >> There's no such thing as a Q35 board without USB in the physical world. >> Can't stop us from making a virtual one, of course. >> >> Likewise, there's no such thing as a Q35 board without AHCI in the >> physical world, and again that can't stop us from making a virtual one. >> >> The difference to USB is that our q35 machines have always had AHCI even >> with -nodefaults. You seem to propose adding a switch to disable AHCI, >> yet leave it enabled with -nodefaults. >> >> -nodefaults should give you a board with all the optional components >> suppressed. > > Will this break libvirt, which may expect -nodefaults to still come > with an IDE bus?
Possibly. >> On the one hand, I'd rather not add exceptions to -nodefaults "give me >> the board with all its optional components suppressed" semantics. >> >> On the other hand, a few hundred ms are a long time. > > That's why I proposed a new option. Yes, it's ugly :/ Does real hardware take that long, too? Or is the delay some virtualization artifact?
