mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Andrey Smirnov <andrew.smirnov@gmail.com>
To: Ian Abbott <abbotti@mev.co.uk>
Cc: Barebox List <barebox@lists.infradead.org>
Subject: Re: barebox state alias with same name as node doesn't work
Date: Thu, 17 Jan 2019 13:10:46 -0800	[thread overview]
Message-ID: <CAHQ1cqGmLGkHQXDOQgasp0naeX5NomG9iODJrx0veAb_ByNGzQ@mail.gmail.com> (raw)
In-Reply-To: <65aaade8-4825-e310-5006-b86df5f72f81@mev.co.uk>

On Thu, Jan 17, 2019 at 8:16 AM Ian Abbott <abbotti@mev.co.uk> wrote:
>
> Hi,
>
> All the examples of "barebox,state" compatible nodes I can find in
> Barebox seem to follow this basic pattern:
>
> / {
>    /* ... */
>    aliases {
>      /* ... */
>      state = &state;
>    };
>    /* ... */
>    state: state {
>      magic = <0xdeadbeef>; /* or whatever */
>      compatible = "barebox,state";
>      /* ... */
>    };
>    /* ... */
> };
>
> Note that the alias name is the same as the node name in the above example.
>
> However, when I try to follow that pattern for my own board, I get an
> error (-EINVAL):
>
> ERROR: register_device: already registered state
>
> This is because the node "state" has been already registered as a device
> called "state" by of_platform_device_create() (via of_probe() and
> of_arm_init()).  Later, the state device driver is initialized and
> state_probe() is called on the device that was previously registered.
> state_probe() calls state_new_from_node() which calls of_alias_get() to
> get the name of the alias for this device (the alias is also called
> "state" in this example), and calls state_new() with this alias name.
> state_new() allocates a struct state with an embedded struct device_d,
> sets the device name to the alias name it was called with ("state"), and
> calls register_device() to register this embedded struct device_d.  The
> call to register_device() fails with return value -EINVAL because the
> name is not unique.
>
> I'm wondering if this has ever worked or whether there is a regression
> bug.  I fixed it on my board by renaming node to "state_nor", keeping
> the alias name as "state".
>
> I'm currently using Barebox 2018.12.0, but I don't see any relevant
> changes in Barebox 2019.01.0 or barebox-next.
>

This should be fixed by:
http://lists.infradead.org/pipermail/barebox/2018-November/035432.html
I don't think it is a part of 2018.12.0, though

Thanks,
Andrey Smirnov

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

  reply	other threads:[~2019-01-17 21:11 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-17 16:16 Ian Abbott
2019-01-17 21:10 ` Andrey Smirnov [this message]
2019-01-18  7:02   ` Sascha Hauer
2019-01-18 15:58   ` Ian Abbott

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=CAHQ1cqGmLGkHQXDOQgasp0naeX5NomG9iODJrx0veAb_ByNGzQ@mail.gmail.com \
    --to=andrew.smirnov@gmail.com \
    --cc=abbotti@mev.co.uk \
    --cc=barebox@lists.infradead.org \
    /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