Re: [PATCH for-6.2] hw/misc/sifive_u_otp: Use IF_PFLASH for the OTP device instead of IF_NONE

[Thomas Huth]

On 19/11/2021 11.40, Philippe Mathieu-Daudé wrote:
> On 11/19/21 11:25, Thomas Huth wrote:
> > Configuring a drive with "if=none" is meant for creation of a backend
> > only, it should not get automatically assigned to a device frontend.
> > Use "if=pflash" for the One-Time-Programmable device instead (like
> > it is e.g. also done for the efuse device in hw/arm/xlnx-zcu102.c).
> >
> > Since the old way of configuring the device has already been published
> > with the previous QEMU versions, we cannot remove this immediately, but
> > have to deprecate it and support it for at least two more releases.
> >
> > Signed-off-by: Thomas Huth <thuth@redhat.com>
> > ---
> >   docs/about/deprecated.rst | 6 ++++++
> >   hw/misc/sifive_u_otp.c    | 9 ++++++++-
> >   2 files changed, 14 insertions(+), 1 deletion(-)
>
> > diff --git a/hw/misc/sifive_u_otp.c b/hw/misc/sifive_u_otp.c
> > index 18aa0bd55d..cf6098ff2c 100644
> > --- a/hw/misc/sifive_u_otp.c
> > +++ b/hw/misc/sifive_u_otp.c
> > @@ -209,7 +209,14 @@ static void sifive_u_otp_realize(DeviceState *dev, Error 
> > **errp)
> >                             TYPE_SIFIVE_U_OTP, SIFIVE_U_OTP_REG_SIZE);
> >       sysbus_init_mmio(SYS_BUS_DEVICE(dev), &s->mmio);
> >   
> > -    dinfo = drive_get_next(IF_NONE);
> > +    dinfo = drive_get_next(IF_PFLASH);
> > +    if (!dinfo) {
> > +        dinfo = drive_get_next(IF_NONE);
>
> Isn't it a bug to call drive_get_next() from DeviceRealize()?
>
> Shouldn't drive_get_next() be restricted to the MachineClass?

Yes, that would certainly be better - but considering that we are already 
past RC1 of the 6.2 release, I'd rather prefer to keep this patch rather as 
small as possible and do such refactorings during the next development cycle 
instead.

 Thomas