From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <fthain@telegraphics.com.au>
Received: from vps4.telegraphics.com.au (vm4.telegraphics.com.au
 [98.124.60.149]) by ozlabs.org (Postfix) with ESMTP id 264491007D1
 for <linuxppc-dev@lists.ozlabs.org>; Fri, 25 Nov 2011 14:15:18 +1100 (EST)
Date: Fri, 25 Nov 2011 14:15:10 +1100 (EST)
From: Finn Thain <fthain@telegraphics.com.au>
To: Alan Cox <alan@lxorguk.ukuu.org.uk>
Subject: Re: [PATCH 01/16] pmac_zilog: fix unexpected irq
In-Reply-To: <20111124145624.24438832@lxorguk.ukuu.org.uk>
Message-ID: <alpine.LNX.2.00.1111250207040.6314@nippy.intranet>
References: <20111023141108.856998818@telegraphics.com.au>
 <20111023141115.208699274@telegraphics.com.au>
 <alpine.LNX.2.00.1111250133321.6314@nippy.intranet>
 <20111124145624.24438832@lxorguk.ukuu.org.uk>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Cc: linux-m68k@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
 linux-serial@vger.kernel.org, Geert Uytterhoeven <geert@linux-m68k.org>
List-Id: Linux on PowerPC Developers Mail List <linuxppc-dev.lists.ozlabs.org>
List-Unsubscribe: <https://lists.ozlabs.org/options/linuxppc-dev>,
 <mailto:linuxppc-dev-request@lists.ozlabs.org?subject=unsubscribe>
List-Archive: <http://lists.ozlabs.org/pipermail/linuxppc-dev>
List-Post: <mailto:linuxppc-dev@lists.ozlabs.org>
List-Help: <mailto:linuxppc-dev-request@lists.ozlabs.org?subject=help>
List-Subscribe: <https://lists.ozlabs.org/listinfo/linuxppc-dev>,
 <mailto:linuxppc-dev-request@lists.ozlabs.org?subject=subscribe>


On Thu, 24 Nov 2011, Alan Cox wrote:

> Given the change should work for all hardware do we really need the 
> ifdefs. Far better I would have thought to just change it so we don't 
> have to maintain what is effectively two versions of the code between 
> now and 2038.

I agree.

> 
> So no ack from me yet - I'd like to understand the ifdef decision first.

Removing ifdefs makes the changes more invasive and the suspend/resume 
code then has to be addressed, which I've avoided.

The suspend/resume code path can't be tested on m68k macs and the common 
code paths I can't easily test on a powermac.

This patch should not be needed because the chip reset shouldn't leave the 
tx and rx interrupts enabled. Those interrupts are explicitly enabled only 
after request_irq(), so patching the master interrupt enable behaviour 
should be redundant. But that's not the case in practice.

The chip reset code is already messy. I was inclined towards ifdefs and 
reluctant to share more code after practical experience suggested possible 
differences in the SCC/ESCC devices.

I guess I was hoping that the powermac maintainers might prefer ifdefs to 
increased risk of destabilising the driver on powermacs...

But a more invasive patch would make for better code. I will see if I can 
borrow a suitable PCI PowerMac.

Finn

> Otherwise it looks sensible.
> 
> Alan