Re: [PATCH v4 0/7] firmware: smccc: Add basic SMCCC v1.2 + ARCH_SOC_ID support

From: Will Deacon
Date: Thu May 21 2020 - 03:58:03 EST


On Thu, May 21, 2020 at 09:34:10AM +0200, Arnd Bergmann wrote:
> On Thu, May 21, 2020 at 9:07 AM Sudeep Holla <sudeep.holla@xxxxxxx> wrote:
> > On Wed, May 20, 2020 at 11:54:16PM +0200, Arnd Bergmann wrote:
> > > On Wed, May 20, 2020 at 11:29 PM Will Deacon <will@xxxxxxxxxx> wrote:
> > > > Applied to arm64 (for-next/smccc), thanks!
> > > >
> > > > Arnd -- Sudeep's reply to you about the sysfs groups seemed reasonable to me,
> > > > but please shout if you'd rather I dropped this in order to pursue an
> > > > alternative approach.
> > >
> > > I missed the reply earlier, thanks for pointing me to it again.

D'oh, I took your silence as "no objections". Oh well!

> > > I'm not entirely convinced, but don't revert it for now because of that,
> > > I assume we can find a solution.

Ok, cheers. It's on a separate branch so it's easy enough to drop if
necessary (i.e. no reverts needed). Sudeep -- please send any extra patches
on top of the branch.

> > I liked your idea of making this generic and hardcode values if required
> > for other drivers. I will take a look at that/
> >
> > > However, please have a look at the build failure report for patch 5
> > > and fix it if you can see what went wrong.
> > >
> >
> > Any pointers for that failure ? I seem to have missed them. I pushed
> > branch couple of times to my tree but got build success both times.
> > Any specific config or compilers ?
>
> See below for the reply from the 0day build bot to your email. It seems it
> was not sent to the mailing list, but you were on Cc. Looking at it now,
> the fix should be trivial.

[...]

> >> drivers/firmware/smccc/smccc.c:14:13: warning: no previous prototype for function 'arm_smccc_version_init' [-Wmissing-prototypes]
> void __init arm_smccc_version_init(u32 version, enum arm_smccc_conduit conduit)
> ^
> drivers/firmware/smccc/smccc.c:14:1: note: declare 'static' if the
> function is not intended to be used outside of this translation unit
> void __init arm_smccc_version_init(u32 version, enum arm_smccc_conduit conduit)

I saw that when I applied the patches, but since the function is called from
another compilation unit (psci/psci.o), I just ignored it as we have loads
of these already and it only screams if you build with W=1.

Will