From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marek Vasut <marex@denx.de>
Date: Mon, 4 Jan 2016 14:23:56 +0100
Subject: [U-Boot] [PATCH V2 1/2] imx: spl: enable
	CONFIG_SPL_WATCHDOG_SUPPORT
In-Reply-To: <568A6C57.3020408@denx.de>
References: <1451910448-28192-1-git-send-email-peng.fan@nxp.com>
 <201601041345.49201.marex@denx.de> <568A6C57.3020408@denx.de>
Message-ID: <201601041423.56227.marex@denx.de>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

On Monday, January 04, 2016 at 01:57:59 PM, Stefano Babic wrote:
> Hi Marek, Peng,
> 
> On 04/01/2016 13:45, Marek Vasut wrote:
> > On Monday, January 04, 2016 at 01:40:10 PM, Peng Fan wrote:
> >> Hi Marek,
> >> 
> >> On Mon, Jan 04, 2016 at 01:38:23PM +0100, Marek Vasut wrote:
> >>> On Monday, January 04, 2016 at 01:27:27 PM, Peng Fan wrote:
> >>>> This commit 4bdcbe60a142b08eefccb0e326a37ba81d3389e8 removes reset_cpu
> >>>> which breaks SPL build when DEBUG macro defined.
> >>>> 
> >>>> "
> >>>> arch/arm/lib/built-in.o: In function `do_reset':
> >>>> ~/uboot/arch/arm/lib/reset.c:45: undefined reference to `reset_cpu'
> >>>> scripts/Makefile.spl:244: recipe for target 'spl/u-boot-spl' failed
> >>>> "
> >>>> Enable CONFIG_SPL_WATCHDOG_SUPPORT to fix this issue.
> >>> 
> >>> Why don't you implement dummy reset_cpu() {} instead ?
> >> 
> >> Do you mean this, https://patchwork.ozlabs.org/patch/562232/?
> > 
> > Yes, I'd prefer that, since I do not want to see watchdog support
> > becoming mandatory part of the SPL build. Watchdog support should be
> > optional.
> 
> Yes, enabling watchdog should be not be mandatory. Anyway, this happens
> only if CONFIG_IMX_WATCHDOG is set, else CONFIG_SPL_WATCHDOG_SUPPORT
> lets only to link reset_cpu() to the build.

This is completely counter-intuitive and stupid. I suspect this behavior
is required on iMX because iMXes reset through watchdog, yes ? In that
case, add a comment explaining why we always select watchdog in SPL on
i.MXes.

> This is also what happens
> for all MX6 boards that do not enable the watchdog but need reset_cpu().
> 
> Adding another dummy function looks nasty to me, and Fabio sent some
> time ago patches to clean up this and drop empty reset_cpu() inside
> board files. We should not create this mess again.

Fine

Best regards,
Marek Vasut