mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
To: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>,
	barebox@lists.infradead.org
Subject: Re: [PATCH 0/5] Early DT and MVEBU DT fixups
Date: Fri, 25 Jul 2014 09:24:40 +0200	[thread overview]
Message-ID: <53D20638.9090106@gmail.com> (raw)
In-Reply-To: <20140725070610.GN23235@pengutronix.de>

On 07/25/2014 09:06 AM, Sascha Hauer wrote:
> On Wed, Jul 23, 2014 at 11:28:05AM +0200, Sebastian Hesselbarth wrote:
>> Historically, early Armada 370 and XP boards where shipped with u-boot
>> not remapping internal register base therefore the upstream dts files
>> reflect that. On barebox we always remap the internal register base to
>> 0xf1000000 and use DT to parse devices, causing the upstream dts to
>> diverge from what we actually see. Currently, we work around the issue
>> by applying a barebox specific dts overlay to reflect the different
>> internal register base.
>>
>> This patch set provides a way to call DT fixups early and apply them
>> even on pbl or appended DTs. This is required for Marvell MVEBU SoCs
>> which can (and do) remap their internal register base address.
>>
>> First, a call to of_fix_tree() is added to of_arm_init() right before
>> of_probe() starts. This allows to apply early registered DT fixups.
>>
>> Patch 2 extends mvebu-mbus to fixup mbus ranges identified by their
>> target ID and target attribute. Patch 3 adds the required internal
>> register ranges to be fixed up for each of the 4 supported MVEBU SoCs.
>> Patch 4 then removes the now redundant DT overlay workarounds.
>>
>> Patch 5 finally installs another DT fixup to properly set directly
>> attached RAM sizes which can read from internal SoC registers. This
>> also removes a runtime warning caused by double-registration of RAM
>> resources from arm_add_memory_device() and of_add_memory() later.
>>
>> Patches have been tested on Armada 370, Dove, and Kirkwood.
>>
>> Sebastian Hesselbarth (5):
>>   ARM: execute OF fixups early
>>   ARM: mvebu: allow to fixup mbus ranges
>>   ARM: mvebu: add register remap for mbus ids
>>   ARM: dts: mvebu: remove mbus ranges overwrite
>>   ARM: mvebu: add fixup for directly attached memory
> 
> I had to think about executing the fixups early a bit. This used
> to fixup the device tree for Linux, now it's used to fixup the
> tree for barebox aswell. I hope this does not have unwanted side
> effects.
> Anyway, it solves your remapping problem nicely, so:
> 
> Applied, thanks

It should cause no harm to other boards that set up fixup later,
you'll have to register the fixup in pure_initcall to make it
modify the early dtb.

Thanks for applying!

Sebastian



_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

      reply	other threads:[~2014-07-25  7:25 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-23  9:28 Sebastian Hesselbarth
2014-07-23  9:28 ` [PATCH 1/5] ARM: execute OF fixups early Sebastian Hesselbarth
2014-07-23  9:28 ` [PATCH 2/5] ARM: mvebu: allow to fixup mbus ranges Sebastian Hesselbarth
2014-07-23  9:28 ` [PATCH 3/5] ARM: mvebu: add register remap for mbus ids Sebastian Hesselbarth
2014-07-23  9:28 ` [PATCH 4/5] ARM: dts: mvebu: remove mbus ranges overwrite Sebastian Hesselbarth
2014-08-02  3:07   ` Ezequiel Garcia
2014-07-23  9:28 ` [PATCH 5/5] ARM: mvebu: add fixup for directly attached memory Sebastian Hesselbarth
2014-07-25  7:06 ` [PATCH 0/5] Early DT and MVEBU DT fixups Sascha Hauer
2014-07-25  7:24   ` Sebastian Hesselbarth [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=53D20638.9090106@gmail.com \
    --to=sebastian.hesselbarth@gmail.com \
    --cc=barebox@lists.infradead.org \
    --cc=s.hauer@pengutronix.de \
    --cc=thomas.petazzoni@free-electrons.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox