Re: [PATCH] mfd: cros_ec: Use proper protocol transfer function

From: Brian Norris
Date: Mon Nov 28 2016 - 17:13:35 EST


Hi Lee,

On Wed, Nov 23, 2016 at 08:37:49AM +0000, Lee Jones wrote:
> On Tue, 22 Nov 2016, Brian Norris wrote:
> > On Wed, Aug 10, 2016 at 01:45:12PM -0700, Brian Norris wrote:
> > > From: Shawn Nematbakhsh <shawnn@xxxxxxxxxxxx>
> > >
> > > pkt_xfer should be used for protocol v3, and cmd_xfer otherwise. We had
> > > one instance of these functions correct, but not the second, fall-back
> > > case. We use the fall-back only when the first command returns an
> > > IN_PROGRESS status, which is only used on some EC firmwares where we
> > > don't want to constantly poll the bus, but instead back off and
> > > sleep/retry for a little while.
> > >
> > > Fixes: 2c7589af3c4d ("mfd: cros_ec: add proto v3 skeleton")
>
> You need to Cc stable with a note saying which kernel version it
> fixes. Grep the `git log`s for stable.*#.

I understand how the stable process works. This was intentional. Fixes
!= stable. It's easy to tag what commit it's supposed to fix. It's less
easy to backport and verify that it's a useful for all affected kernels.

(If somebody else determines it should be backported -- e.g., Benson or
Olof -- then I won't stop them, of course.)

Side note: isn't 'Fixes' a more accurate way of describing which version
it applies to than just putting the kernel version with a '#' comment?
That helps anyone who might have backported the buggy commit -- e.g., as
part of the stable process.

> > > Signed-off-by: Shawn Nematbakhsh <shawnn@xxxxxxxxxxxx>
> > > Signed-off-by: Brian Norris <briannorris@xxxxxxxxxxxx>
> > > ---
> > > MAINTAINERS tells me this goes through Olof, but many things have gone through
> > > Lee.
> >
> > I believe this was supposed to go through Olof. Olof, are you out there?
>
> Yes, I concur.
>
> There maybe some confusion since the $SUBJECT line is
> misleading/incorrect.

Thanks. I'll resend with a 'platform/chrome:' prefix, and make sure it
shows up in Benson's inbox too. I think I didn't look deeply enough into
history -- the last few commits touching drivers/platform/chrome/ had an
'mfd' tag, but they also happened to touch mfd-prefixed files too.

Brian

> > Or Benson? I see this:
> >
> > http://www.mail-archive.com/linux-kernel@xxxxxxxxxxxxxxx/msg1273587.html
> > [PATCH] platform/chrome : Add myself as Maintainer
> >
> > but it's not merged anywhere AFAICT.
> >
> > I can resend if that helps.
> >
> > Brian
[...]