All of lore.kernel.org
 help / color / mirror / Atom feed
From: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
To: Roger Quadros <rogerq@ti.com>
Cc: Javier Martinez Canillas <martinez.javier@gmail.com>,
	Benoit Cousson <bcousson@baylibre.com>,
	Tony Lindgren <tony@atomide.com>,
	Enric Balletbo i Serra <eballetbo@gmail.com>,
	"linux-omap@vger.kernel.org" <linux-omap@vger.kernel.org>,
	"devicetree@vger.kernel.org" <devicetree@vger.kernel.org>
Subject: Re: [PATCH 2/3] ARM: dts: omap3-igep0020: Add HS USB Host support
Date: Mon, 07 Oct 2013 13:53:12 +0200	[thread overview]
Message-ID: <5252A0A8.3050607@collabora.co.uk> (raw)
In-Reply-To: <52529011.5080209@ti.com>

On 10/07/2013 12:42 PM, Roger Quadros wrote:
> On 10/07/2013 01:22 PM, Javier Martinez Canillas wrote:
>> On Mon, Oct 7, 2013 at 11:13 AM, Javier Martinez Canillas
>> <javier.martinez@collabora.co.uk> wrote:
>>> On 10/07/2013 11:06 AM, Roger Quadros wrote:
>>>>>
>>>>> Well that's a very good question indeed.
>>>>>
>>>>> The thing is that the IGEP0030 is a Computer-on-Module [1] that is used in
>>>>> conjunction with expansion boards and some of them have USB HOST support such as
>>>>> IGEP Paris [2] and IGEP Berlin [3].
>>>>>
>>>>> Support for this expansion boards is still not in mainline but there are plans
>>>>> to add them using Device Tree overlays [4] once/if this land on mainline.
>>>>>
>>>>
>>>> Why would your boards need Device Tree overlays? From the looks of it, the configuration
>>>> of SOM + base board don't seem to change at runtime.
>>>>
>>>
>>> Indeed, a static configuration (DTS) would be enough now that I think about it.
>>>
>> 
>> Hi Roger,
>> 
>> Now that I had coffee I remember why I think that even when Device
>> Tree overlays are not strictly required for a SOM + base board it
>> could be handy to use. If we use a static configuration (DTB) then the
>> same firmware can't be used by any IGEP COM Module user. She would
>> have to choose a DTB to pass to the kernel on boot.
>> 
>> While using DT overlays the same firmware that provides a minimal DTB
>> can be used regardless of the base board used (as long as there are
>> all the needed fragment/overlays hooks in the DTS).
>> 
>> After all the SOM has a NAND flash memory and a uSD/MMC slot so a
>> minimal DTB is needed to boot and the support for all the peripherals
>> present on the base board can be added by triggering a device tree
>> overlay load from user-space.
> 
> Consider this example. You need to boot your board using NFS over ethernet
> dongle connected to USB host. If you need user space to get that to work,
> it will be a unnecessary challenge, whereas you can easily do that if you
> have a static DT blob.
> 
>> 
>> Or maybe I'm misunderstanding the use case for DT overlays since I had
>> just read about it but I don't have practical experience with it.
>> 
> 
> DT overlays is a solution to the problem faced by the beagle bone community.
> There they have a relatively large number of accessories (called capes).
> Since the capes don't use a dynamically probed interface like USB/MMC, 
> the hardware information needs to be hard coded somewhere and loaded
> into the DT at runtime whenever a new cape is connected.
> 
> Using overlays for a SOM + base board architecture is an overkill IMO. A base board
> is not an accessory, but the platform board itself, hence qualifies for it's own
> dts file.
> 
> It is much easier to use a base dts for the SOM which contains all the information
> for the SOM and leaves the base board details to the base board specific dts.
> Each base board can be considered to be a extended variant of the SOM.
> 
> cheers,
> -roger
> 

Hi Roger,

Thanks a lot for the explanation, is very clear for me now.

Best regards,
Javier

  reply	other threads:[~2013-10-07 11:53 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-05  0:04 [PATCH 0/3] ARM: dts: omap3-igep0020: improvements for v3.13 Javier Martinez Canillas
2013-10-05  0:04 ` [PATCH 1/3] ARM: dts: omap3-igep0020: Add USB OTG support Javier Martinez Canillas
2013-10-06 10:26   ` Enric Balletbo Serra
2013-10-07 10:43   ` Roger Quadros
2013-10-07 11:54     ` Javier Martinez Canillas
2013-10-07 11:58       ` Benoit Cousson
2013-10-07 12:02         ` Javier Martinez Canillas
2013-10-07 12:09           ` Benoit Cousson
2013-10-07 16:05             ` Javier Martinez Canillas
2013-10-05  0:04 ` [PATCH 2/3] ARM: dts: omap3-igep0020: Add HS USB Host support Javier Martinez Canillas
2013-10-06 10:27   ` Enric Balletbo Serra
2013-10-07  8:33   ` Roger Quadros
2013-10-07  8:50     ` Javier Martinez Canillas
     [not found]       ` <525275C8.2050208-ZGY8ohtN/8pPYcu2f3hruQ@public.gmane.org>
2013-10-07  9:06         ` Roger Quadros
2013-10-07  9:13           ` Javier Martinez Canillas
2013-10-07 10:22             ` Javier Martinez Canillas
2013-10-07 10:42               ` Roger Quadros
2013-10-07 11:53                 ` Javier Martinez Canillas [this message]
2013-10-05  0:04 ` [PATCH 3/3] ARM: dts: omap3-igep0020: use standard constant for IRQ flags Javier Martinez Canillas

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=5252A0A8.3050607@collabora.co.uk \
    --to=javier.martinez@collabora.co.uk \
    --cc=bcousson@baylibre.com \
    --cc=devicetree@vger.kernel.org \
    --cc=eballetbo@gmail.com \
    --cc=linux-omap@vger.kernel.org \
    --cc=martinez.javier@gmail.com \
    --cc=rogerq@ti.com \
    --cc=tony@atomide.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.