mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
To: Rob Herring <rob.herring@calxeda.com>
Cc: barebox@lists.infradead.org
Subject: Re: [PATCH 1/1] highbank: use the provided dtb by the firmware to probe barebox device and mem size
Date: Wed, 13 Feb 2013 10:13:51 +0100	[thread overview]
Message-ID: <20130213091351.GH19322@game.jcrosoft.org> (raw)
In-Reply-To: <511B057A.2090607@calxeda.com>

On 21:16 Tue 12 Feb     , Rob Herring wrote:
> On 02/12/2013 03:11 PM, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > the dtb is at 0x1000
> > 
> > if no dtb present use C code device
> > 
> > kepp in C the timer/gpio/uart
> > 
> > Cc: Rob Herring <rob.herring@calxeda.com>
> > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> > ---
> >  arch/arm/boards/highbank/env/bin/init_board |    7 +++++
> >  arch/arm/boards/highbank/init.c             |   44 ++++++++++++++++++++++++---
> >  2 files changed, 47 insertions(+), 4 deletions(-)
> >  create mode 100644 arch/arm/boards/highbank/env/bin/init_board
> > 
> > diff --git a/arch/arm/boards/highbank/env/bin/init_board b/arch/arm/boards/highbank/env/bin/init_board
> > new file mode 100644
> > index 0000000..610db15
> > --- /dev/null
> > +++ b/arch/arm/boards/highbank/env/bin/init_board
> > @@ -0,0 +1,7 @@
> > +#!/bin/sh
> > +
> > +if [ -e /dev/dtb ]
> > +then
> > +	oftree -l /dev/dtb
> > +	oftree -p
> > +fi
> > diff --git a/arch/arm/boards/highbank/init.c b/arch/arm/boards/highbank/init.c
> > index 9fb7986..a01db5b 100644
> > --- a/arch/arm/boards/highbank/init.c
> > +++ b/arch/arm/boards/highbank/init.c
> > @@ -15,20 +15,56 @@
> >  #include <sizes.h>
> >  #include <io.h>
> >  
> > +struct fdt_header *fdt = NULL;
> > +
> >  static int highbank_mem_init(void)
> >  {
> > +	struct device_node *np;
> > +	u32 reg[2];
> > +	int ret;
> > +
> > +	/* load by the firmware at 0x1000 */
> > +	fdt = IOMEM(0x1000);
> > +
> > +	ret = of_unflatten_dtb(fdt);
> 
> full blown DT seems a bit much. Can't you do what's needed with libfdt?
> I guess if you've already pulled it all in, it doesn't matter.
yes
> 
> > +	if (ret) {
> > +		pr_warn("no dtb found at 0x1000 use default configuration\n");
> > +		fdt = NULL;
> > +		goto not_found;
> > +	}
> > +
> > +	np = of_find_node_by_path("/memory");
> > +	if (!np) {
> > +		pr_warn("no memory node use default configuration\n");
> > +		goto not_found;
> > +	}
> > +
> > +	ret = of_property_read_u32_array(np, "reg", reg, 2); 
> 
> You assume the memory address and sizes are 1 word which soon will not
> be true. See ecx-2000.dts. Note that I'm planning to merge the 2 memory
> nodes to 1 node with 2 addresses and sizes in reg property.
I'll check
> 
> > +	if (ret) {
> > +		pr_warn("memory node: no reg property use default configuration\n");
> > +		goto not_found;
> > +	}
> > +
> > +	pr_info("highbank: dtb probed mem_size = 0x%x\n", reg[1]);
> > +
> > +	highbank_add_ddram(reg[1]);
> > +
> > +not_found:
> >  	//highbank_add_ddram(4089 << 20);
> >  	highbank_add_ddram(SZ_1G);
> > -
> >  	return 0;
> >  }
> >  mem_initcall(highbank_mem_init);
> >  
> >  static int highbank_devices_init(void)
> >  {
> > -	highbank_register_ahci();
> > -	highbank_register_xgmac(0);
> > -	highbank_register_xgmac(1);
> > +	if (!fdt) {
> > +		highbank_register_ahci();
> > +		highbank_register_xgmac(0);
> > +		highbank_register_xgmac(1);
> > +	} else {
> > +		devfs_add_partition("ram1", 0x1000, SZ_64K, DEVFS_PARTITION_FIXED, "dtb");
> > +	}
> >  
> >  	armlinux_set_bootparams((void *)(0x00000100));
> >  
> > 
> 

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

  reply	other threads:[~2013-02-13  9:15 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-12 21:11 Jean-Christophe PLAGNIOL-VILLARD
2013-02-12 21:11 ` Jean-Christophe PLAGNIOL-VILLARD
2013-02-13  3:16 ` Rob Herring
2013-02-13  9:13   ` Jean-Christophe PLAGNIOL-VILLARD [this message]
2013-02-13  8:18 ` Sascha Hauer
2013-02-13  9:53   ` Jean-Christophe PLAGNIOL-VILLARD

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=20130213091351.GH19322@game.jcrosoft.org \
    --to=plagnioj@jcrosoft.com \
    --cc=barebox@lists.infradead.org \
    --cc=rob.herring@calxeda.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