Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20191016082430.5955-1-poeschel@lemonage.de> <CANiq72=uXWpEWHixM+wwyxZfzQ41WYvQsoV8B3+JLRharDjC0w@mail.gmail.com>
 <20191017080741.GA17556@lem-wkst-02.lemonage>
In-Reply-To: <20191017080741.GA17556@lem-wkst-02.lemonage>
From:   Andy Shevchenko <andy.shevchenko@gmail.com>
Date:   Thu, 17 Oct 2019 14:26:31 +0300
Message-ID: <CAHp75Vd2WqpSMHGR6LDvePyhdDDPOOWWsStD5RWDvpw7xyM6VQ@mail.gmail.com>
Subject: Re: [PATCH 1/3] auxdisplay: Make charlcd.[ch] more general
To:     Lars Poeschel <poeschel@lemonage.de>
Cc:     Miguel Ojeda <miguel.ojeda.sandonis@gmail.com>,
        Willy Tarreau <willy@haproxy.com>,
        Ksenija Stanojevic <ksenija.stanojevic@gmail.com>,
        open list <linux-kernel@vger.kernel.org>,
        Geert Uytterhoeven <geert@linux-m68k.org>,
        Geert Uytterhoeven <geert+renesas@glider.be>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Oct 17, 2019 at 11:07 AM Lars Poeschel <poeschel@lemonage.de> wrote:
> On Wed, Oct 16, 2019 at 06:53:20PM +0200, Miguel Ojeda wrote:
> > On Wed, Oct 16, 2019 at 10:24 AM Lars Poeschel <poeschel@lemonage.de> wrote:

> > Thanks Lars, CC'ing Geert since he wrote a large portion of this, as
> > well as Andy.

I would be able to test next version if you Cc it to me (better to my
@linux.intel.com address).


> The idea with changing the return types: It seems a bit, that with this
> patch charlcd is becoming more of an universal interface and maybe more
> display backends get added - maybe with displays, that can report
> failure of operations. And I thought, it will be better to have this
> earlier and have the "interface" stable and more uniform. But you are
> the maintainer. If you don't like the changed return types I happily
> revert back to the original ones in the next version of the patch.

First rule of all, split as much as one logical change per patch. If
you wish to split, do simple split and nothing else. But I think you
also need to prepare something for the split. I would think of
introducing ops structure beforehand.

> Are you able to test the panel driver ?
I would be able to test on HD44780 connected via GPIO (4-bit mode).

-- 
With Best Regards,
Andy Shevchenko