Hi Eric,
On Wed, Sep 19, 2018 at 2:36 PM Auger Eric <eric.auger@xxxxxxxxxx> wrote:
> On 9/17/18 6:39 PM, Geert Uytterhoeven wrote:
> > Vfio-platform requires dedicated reset support, provided either by ACPI,
> > or, on DT platforms, by a device-specific reset driver matching against
> > the device's compatible value.
> >
> > On many SoCs, devices are connected to an SoC-internal reset controller.
> > If the reset hierarchy is described in DT using "resets" properties, or
> > in lookup tables in platform code, such devices can be reset in a
> > generic way through the reset controller subsystem. Hence add support
> > for this, avoiding the need to write device-specific reset drivers for
> > each single device on affected SoCs.
> >
> > Devices that do require a more complex reset procedure can still provide
> > a device-specific reset driver, as that takes precedence.
> >
> > Note that this functionality depends on CONFIG_RESET_CONTROLLER=y, and
> > becomes a no-op (as in: "No reset function found for device") if reset
> > controller support is disabled.
> >
> > Signed-off-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
> > Reviewed-by: Philipp Zabel <p.zabel@xxxxxxxxxxxxxx>
> > Reviewed-by: Simon Horman <horms+renesas@xxxxxxxxxxxx>
> > --- a/drivers/vfio/platform/vfio_platform_common.c
> > +++ b/drivers/vfio/platform/vfio_platform_common.c
> > @@ -128,8 +131,16 @@ static int vfio_platform_get_reset(struct vfio_platform_device *vdev)
> > vdev->of_reset = vfio_platform_lookup_reset(vdev->compat,
> > &vdev->reset_module);
> > }
> > + if (vdev->of_reset)
> > + return 0;
> > +
> > + rstc = reset_control_get_dedicated(vdev->device, NULL);
> > + if (!IS_ERR(rstc)) {
> > + vdev->reset_control = rstc;
> > + return 0;
> > + }
> >
> > - return vdev->of_reset ? 0 : -ENOENT;
> > + return PTR_ERR(rstc);
> This changes the returned value as seen by the user (probe returned
> valud). Can we keep -ENOENT in case of no reset controller found?
On success, it still returns 0.
On failure, it forwards the error from reset_control_get_dedicated(), which
is IMHO better than replacing it by -ENOENT: we try to propagate error
codes as much as possible. It could e.g. return -EPROBE_DEFER.
Is there anything that relies on the function returning -ENOENT?
> Otherwise looks good to me with the new "dedicated" reset semantics.
Thanks!
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
