From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric BENARD <ebenard@free.fr>
Date: Thu, 02 Apr 2009 21:47:29 +0200
Subject: [U-Boot] [PATCH v2] at91: change CONFIG_SYS_HZ to 1000
In-Reply-To: <20090402190616.GJ2651@game.jcrosoft.org>
References: <1235527085-23900-1-git-send-email-dv@vollmann.ch>	<49D50DDC.2000001@free.fr>
	<20090402190616.GJ2651@game.jcrosoft.org>
Message-ID: <49D51651.3060702@free.fr>
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

Hi Jean Christophe,

Jean-Christophe PLAGNIOL-VILLARD a ?crit :
> On 21:11 Thu 02 Apr     , Eric BENARD wrote:
>> I confirm that this patch is fixing timeout problems with tftp downloads 
>> on a SAM9260.
>> What is blocking from pushing it mainline ?
> see response e-mail and please note that the current patch is not supposed to
> work because AT91C_MAIN_CLOCK does not exist and it assusme that the
it's using MASTER_CLOCK and it seems to work, at least on my board, and 
the improvement is clearly seen when using tftp (no more T and no more 
retry when downloading a large file).

> main_clock will be at 100000000 which is not the case everytime
> as example on 9G20 it's not
true, I think G20 support was not mainlined when the patch was created 
which can explain this.

> I've send a new patch that introduce an improve of the clock on at91
> which give you the calculated main_clock based on the main crystal and the
> lowlevel settings
> 
I've just seen this patch. I will give a try asap.
Is it interesting to loose time calculating PLL values at each boot when 
these value can be fixed in the board config file ?

Eric