2017-11-19 04:10:01

by Kershner, David A

[permalink] [raw]
Subject: RE: [PATCH 2/2] drivers: visorbus: move driver out of staging

> -----Original Message-----
> From: Greg KH [mailto:[email protected]]
> Sent: Saturday, November 18, 2017 5:26 AM
> To: Kershner, David A <[email protected]>
> Cc: [email protected]; [email protected]; driverdev-
> [email protected]; *S-Par-Maintainer
> <[email protected]>; [email protected];
> [email protected]
> Subject: Re: [PATCH 2/2] drivers: visorbus: move driver out of staging
>
> On Fri, Nov 17, 2017 at 12:27:39PM -0500, David Kershner wrote:
> > {drivers/staging/unisys/include => include/linux/visorbus}/visorbus.h |
0
> > .../staging/unisys/include => include/linux/visorbus}/visorchannel.h |
0
>
> Do we really need two different include/linux .h files for this bus
> subsystem? Can we merge them both together?
>

I will merge them together, they were separate so the other channel
definitions
didn't have to include all of visorbus.h for clarity.

> > diff --git a/drivers/staging/unisys/visorbus/Kconfig
> b/drivers/visorbus/Kconfig
> > similarity index 90%
> > rename from drivers/staging/unisys/visorbus/Kconfig
> > rename to drivers/visorbus/Kconfig
> > index 5113880..a99285a 100644
> > --- a/drivers/staging/unisys/visorbus/Kconfig
> > +++ b/drivers/visorbus/Kconfig
> > @@ -4,7 +4,9 @@
> >
> > config UNISYS_VISORBUS
> > tristate "Unisys visorbus driver"
> > - depends on UNISYSSPAR
> > + depends on X86_64 && !UML
> > + select PCI
> > + select ACPI
>
> Wait, what? Why are you messing with the dependancies now? And never
> do a select if at all possible, do a 'depends on' instead.
>
> And why not UML? What about other arches, why are you now restricting
> it? That's an odd change to be doing here in a "move this out of
> staging" patch...
>

When I moved them out of staging, I noticed that it had "depends on
UNISYSSPAR"
flag. The dependencies I added were copied from the UNISYSSPAR dependency
list.

I will make a separate patch to switch them over to depends and determine if
the
!UML flag is still needed.

Thanks,
David Kershner

> thanks,
>
> greg k-h


Attachments:
smime.p7s (7.68 kB)