[PATCH] arm64: enable ARCH_HAS_BANDGAP config

[catalin.marinas@arm.com (Catalin Marinas)]

On Fri, Nov 28, 2014 at 02:39:09PM +0000, Abhilash Kesavan wrote:
> On Fri, Nov 28, 2014 at 7:42 PM, Catalin Marinas
> <catalin.marinas@arm.com> wrote:
> > On Sun, Nov 23, 2014 at 08:05:05AM +0000, Alim Akhtar wrote:
> >> On Fri, Nov 14, 2014 at 6:00 PM, Abhilash Kesavan <a.kesavan@samsung.com> wrote:
> >> > Some of the existing thermal drivers (ti and exynos) have a dependency on
> >> > ARCH_HAS_BANDGAP. Activate ARCH_HAS_BANDGAP config entry for ARM64 so that
> >> > these drivers may be re-used.
> >> >
> >> > Signed-off-by: Abhilash Kesavan <a.kesavan@samsung.com>
> >> > ---
> >> HAS_BANDGAP was added to ARM to indicate that arch supports a bandgap
> >> device like a thermal management unit, so this looks good to me.
> >
> > But it doesn't look good to me. ARCH_HAS_* kind of implies that the
> > architecture has some feature or the port provides some feature. But in
> > the current use, it simply allows the enabling of the menu config
> > entries for two drivers (EXYNOS_THERMAL and TI_SOC_THERMAL).
> >
> > If you want to specify that the SoC has a device, just use DT and not
> > some random config option that only enables the menu entry for the
> > corresponding driver.
> 
> So, you are suggesting that the ARCH_HAS_BANDGAP dependency for these
> two drivers be removed ? I am not aware of the history behind adding
> this symbol, but as it is only used  by the 2 platforms it should be
> easy to remove.

Yes. You have defconfig for those platforms anyway. If you only want to
enable a driver for a certain SoC, put a dependency on that SoC config
(it doesn't look like some generic driver that is used by multiple
SoCs).

-- 
Catalin