From mboxrd@z Thu Jan  1 00:00:00 1970
From: gregory.clement@free-electrons.com (Gregory CLEMENT)
Date: Fri, 17 Feb 2017 10:52:31 +0100
Subject: [PATCH 0/3] Extend rtc-armada38x support for Armada 7K/8K
In-Reply-To: <20170216173748.GU27312@n2100.armlinux.org.uk> (Russell King's
 message of "Thu, 16 Feb 2017 17:37:48 +0000")
References: <20170216153136.27009-1-gregory.clement@free-electrons.com>
 <20170216173348.GT27312@n2100.armlinux.org.uk>
 <20170216173748.GU27312@n2100.armlinux.org.uk>
Message-ID: <87zihlb0v4.fsf@free-electrons.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Hi Russell King,
 
 On jeu., f?vr. 16 2017, Russell King - ARM Linux <linux@armlinux.org.uk> wrote:

> On Thu, Feb 16, 2017 at 05:33:48PM +0000, Russell King - ARM Linux wrote:
>> On Thu, Feb 16, 2017 at 04:31:33PM +0100, Gregory CLEMENT wrote:
>> > The Armada 7K/8K SoCs use the same RTC IP than the Armada 38x. However
>> > the SOC integration differs in 2 points:
>> >      - MBUS bridge timing initialization
>> >      - IRQ configuration at SoC level
>> > 
>> > This patch set extends the driver support to these SoCs family.
>> 
>> I've just tried merging this into my kernel, and...
>> 
>> root at mcbin:~# date
>> Thu Feb 16 17:33:00 GMT 2017
>> root at mcbin:~# hwclock -uw
>> root at mcbin:~# hwclock
>> Sun 07 Feb 2106 01:49:51 GMT  -10.010475 seconds
>> root at mcbin:~# hwclock
>> Sun 07 Feb 2106 01:49:51 GMT  -10.010458 seconds
>> 
>> I guess it needs the same hack that Armada38x needs - "date reset" twice
>> in uboot?
>
> Hmm, even with that sorted:
>
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:35:50 GMT  -0.786571 seconds
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:35:51 GMT  -1.347541 seconds
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:35:52 GMT  -1.889295 seconds
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:35:53 GMT  -1.787338 seconds
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:35:54 GMT  -1.444397 seconds
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:35:55 GMT  -1.419772 seconds
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:35:58 GMT  -2.555282 seconds
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:36:09 GMT  -0.515289 seconds
> root at mcbin:~# hwclock
> Thu 16 Feb 2017 17:36:11 GMT  -1.773030 seconds
> root at mcbin:~#
>
> I think this RTC is a random number generator.  It's certainly not
> a reliable indication of time.

I found the root of this issue. I tested on Aramda 7040 whihc have onle
the CP master. But actually the RTC is present in each CPM (master and
slave). And it occurs that the RTC requires an external
oscillator. However on the Armada 80x0, the RTC clock in CP master is
not connected (by package) to the oscillator. But the one in CP slave is
connected.

So I am sending a new version of the series with an updated device tree
allowing using the RTC from CP slave in 80x0 instead of the one from CP
master.

Gregory

>
> -- 
> RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
> FTTC broadband for 0.8mile line: currently at 9.6Mbps down 400kbps up
> according to speedtest.net.

-- 
Gregory Clement, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com