Re: [patch v8 2/3] platform/mellanox: mlxreg-hotplug: allow driver for ARM architecture

[Darren Hart]

On Tue, Jan 16, 2018 at 05:50:44PM -0800, Darren Hart wrote:
> On Tue, Jan 16, 2018 at 08:06:33AM +0000, Vadim Pasternak wrote:
> > 
> > 
> > > -----Original Message-----
> > > From: Darren Hart [mailto:dvhart@xxxxxxxxxxxxx]
> > > Sent: Tuesday, January 16, 2018 5:43 AM
> > > To: Vadim Pasternak <vadimp@xxxxxxxxxxxx>
> > > Cc: andy.shevchenko@xxxxxxxxx; gregkh@xxxxxxxxxxxxxxxxxxx; linux-
> > > kernel@xxxxxxxxxxxxxxx; platform-driver-x86@xxxxxxxxxxxxxxx; jiri@xxxxxxxxxxx
> > > Subject: Re: [patch v8 2/3] platform/mellanox: mlxreg-hotplug: allow driver for
> > > ARM architecture
> > > 
> > > On Thu, Jan 11, 2018 at 08:58:42AM +0000, Vadim Pasternak wrote:
> > > > It allows driver to run on x86 and ARM architecture based systems.
> > > > Also some unnecessary includes are removed.
> > > 
> > > Hi Vadim,
> > > 
> > > These changes are not all readily mapped to this description, nor to the
> > > changelog below.
> > > 
> > > Only the kconfig change is obviously linked to building on ARM and not just x86.
> > > 
> > > Things like the removal of the dev_err messages and the relocation of the create
> > > and destroy methods seem entirely unrelated to building on ARM.
> > > 
> > > I'd like to get this merged sooner rather than later as it has been lingering for
> > > quite some time now - can you provide some more detail about how the
> > > changes included in this patch are needed to make this driver build on ARM?
> > 
> > Yes. Right.
> > Should I split it to two separate patches, one only with Kconfig changes and the
> > second with text like:
> > Remove some unnecessary includes;
> > Remove dev_err message;
> > Change filed bus to nr for consistency with kernel definitions;
> 
> This would be a better split, yes. Note that we need to document *why* it is an
> unnecessary dev_err message. When you respin, please pull the X86_64 Kconfig
> change from 1/3 into the Kconfig changes for ARM support as well.

Scratch the bit about the x86_64 config - just do that as a separate patch for
the x86/mlx-platform driver as that is independent from everything else here.

-- 
Darren Hart
VMware Open Source Technology Center