[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH 3/4] usb/ohci-pci: deprecate, don't build by default
|
From: |
Paolo Bonzini |
|
Subject: |
Re: [PATCH 3/4] usb/ohci-pci: deprecate, don't build by default |
|
Date: |
Tue, 28 May 2024 15:54:05 +0200 |
On Tue, May 28, 2024 at 12:35 PM Thomas Huth <thuth@redhat.com> wrote:
> > diff --git a/hw/usb/Kconfig b/hw/usb/Kconfig
> > index 84bc7fbe36cd..c4a6ea5a687f 100644
> > --- a/hw/usb/Kconfig
> > +++ b/hw/usb/Kconfig
> > @@ -17,7 +17,6 @@ config USB_OHCI_SYSBUS
> >
> > config USB_OHCI_PCI
> > bool
> > - default y if PCI_DEVICES
> > depends on PCI
> > select USB_OHCI
>
> Not sure whether we should disable it by default just because it is
> deprecated. We don't do that for any other devices as far as I know.
> Anyway, you should add the device to docs/about/deprecated.rst to really
> mark it as deprecated, since that's our official list (AFAIK).
That would mean removing it, but that's a bad idea. It's not like the
device is blocking improvements elsewhere (in fact it's not even
removing any code because the sysbus OHCI is still there).
> Also, there are still some machines that use this device:
>
> $ grep -r USB_OHCI_PCI *
> hw/hppa/Kconfig: imply USB_OHCI_PCI
> hw/mips/Kconfig: imply USB_OHCI_PCI
> hw/ppc/Kconfig: imply USB_OHCI_PCI
> hw/ppc/Kconfig: imply USB_OHCI_PCI
>
> pseries could certainly continue without OHCI AFAICT, but the others?
Yeah, this needs to be a per-machine type choice to warn about
discouraged devices. Some, such as Cirrus, can probably be
unconditional, but still I wouldn't remove them.
Paolo