From mboxrd@z Thu Jan  1 00:00:00 1970
From: Marek Vasut <marex@denx.de>
Date: Thu, 27 Sep 2012 19:27:37 +0200
Subject: [U-Boot] [PATCH 1/6] am33xx: Enable UART{1,2,4,5} clocks
In-Reply-To: <50648B3C.3080703@ti.com>
References: <1348689859-21803-1-git-send-email-andrew@bradfordembedded.com>
	<201209271911.31263.marex@denx.de> <50648B3C.3080703@ti.com>
Message-ID: <201209271927.37776.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

Dear Tom Rini,

> On 09/27/12 10:11, Marek Vasut wrote:
> > Dear Tom Rini,
> > 
> >> On 09/27/12 09:45, Marek Vasut wrote:
> >>> Dear Tom Rini,
> >>> 
> >>>> On Thu, Sep 27, 2012 at 06:13:36PM +0200, Marek Vasut wrote:
> >>>>> Dear Andrew Bradford,
> >>>>> 
> >>>>>> If configured to use UART{1,2,4,5}, such as on the Beaglebone
> >>>>>> RS232 cape, enable the required clocks for the UART in use.
> >>>>>> 
> >>>>>> Signed-off-by: Andrew Bradford <andrew@bradfordembedded.com>
> >>>>>> ---
> >>>>>> 
> >>>>>> arch/arm/cpu/armv7/am33xx/clock.c |   28
> >>>>>> ++++++++++++++++++++++++++++ 1 file changed, 28
> >>>>>> insertions(+)
> >>>>>> 
> >>>>>> diff --git a/arch/arm/cpu/armv7/am33xx/clock.c
> >>>>>> b/arch/arm/cpu/armv7/am33xx/clock.c index 2b19506..4eb9226
> >>>>>> 100644 --- a/arch/arm/cpu/armv7/am33xx/clock.c +++
> >>>>>> b/arch/arm/cpu/armv7/am33xx/clock.c @@ -114,6 +114,34 @@
> >>>>>> static void enable_per_clocks(void)
> >>>>>> 
> >>>>>> while (readl(&cmwkup->wkup_uart0ctrl) != PRCM_MOD_EN)  ;
> >>>>>> 
> >>>>>> +	/* UART1 */ +#ifdef CONFIG_SERIAL2 +	writel(PRCM_MOD_EN,
> >>>>>> &cmper->uart1clkctrl); +	while (readl(&cmper->uart1clkctrl)
> >>>>>> != PRCM_MOD_EN) +		;
> >>>>> 
> >>>>> Call WATCHDOG_RESET() here, fix glboally
> >>>> 
> >>>> We don't have WATCHDOG_RESET...
> >>> 
> >>> You do, and it opts-out to udelay(1) is most cases.
> >> 
> >> It looks like it opts-out to {} in most cases, in <watchdog.h>
> > 
> > Correct, we use it to retrigger watchdog timer if implemented.
> 
> Which the SoC support isn't doing and the rest of the code also isn't
> trying to use.  Arguably the whole file should be doing udelay(1) in
> each of these instances and a clean up patch which this series depends
> on might be useful.

So we're changing the practice from doing WATCHDOG_RESET() to udelay(1) ? And 
we're doing so in generic code?

Best regards,
Marek Vasut