linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: "Grant Likely" <grant.likely@secretlab.ca>
To: "Jean Delvare" <khali@linux-fr.org>
Cc: linuxppc-dev@ozlabs.org, Sean MacLennan <smaclennan@pikatech.com>,
	Scott Wood <scottwood@freescale.com>, Stefan Roese <sr@denx.de>,
	Timur Tabi <timur@freescale.com>
Subject: Re: "cell-index" vs. "index" vs. no index in I2C device nodes
Date: Fri, 6 Jun 2008 01:47:50 -0600	[thread overview]
Message-ID: <fa686aa40806060047n7b8b8e6flee193d46d4e3dd24@mail.gmail.com> (raw)
In-Reply-To: <20080606082150.691da6a4@hyperion.delvare>

On Fri, Jun 6, 2008 at 12:21 AM, Jean Delvare <khali@linux-fr.org> wrote:
> Hi Ben,
>
> On Fri, 06 Jun 2008 14:16:23 +1000, Benjamin Herrenschmidt wrote:
>> On Thu, 2008-06-05 at 09:48 +0200, Jean Delvare wrote:
>> > As far as I am concerned, it's really up to the maintainers and users
>> > of this platform. All I am asking for is that you do not call
>> > i2c_add_numbered_adapter() on an adapter with an automatically
>> > generated number. This function must only be used for adapter's those
>> > number is well defined. If an adapter doesn't have a well-defined
>> > number then use i2c_add_adapter() (but then you can no longer declare
>> > your I2C devices as part of the platform data.)
>>
>> Can't we just give those things -names- instead of numbers ?
>
> Good question. These i2c adapters already have names
> (i2c_adapter.name), which are set by their respective drivers with more
> or less grace depending on the driver [1]. What we don't have though is
> the possibility to instantiate i2c devices by adapter name.
>
> We could certainly add a busname field to struct i2c_devinfo and
> implement i2c_register_board_info_by_name() if it seems to be worth the
> extra code and memory. I am open to the idea if it solves a problem
> with no other clean solution.

Even this isn't necessary.  In arch/powerpc we don't really need to
register the board info separately from the bus driver so we don't
need to jump through hoops to match a board_info with an i2c bus.  The
device tree data structure already links the i2c devices with the bus
they are on, so an OF-aware i2c bus driver will match against a node
in the device tree and then use all the child nodes to register i2c
devices against itself.

This, of course, only applies to platform level i2c busses described
in the device tree.  Plugin devices like PCI cards still can create
their own i2c busses in the usual way.

Cheers,
g.

-- 
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.

  reply	other threads:[~2008-06-06  7:47 UTC|newest]

Thread overview: 74+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-06-04 15:06 "cell-index" vs. "index" vs. no index in I2C device nodes Stefan Roese
2008-06-04 15:24 ` Timur Tabi
2008-06-04 15:43   ` Scott Wood
2008-06-05  2:19     ` Josh Boyer
2008-06-05  2:41       ` David Gibson
2008-06-06  2:40         ` Segher Boessenkool
2008-06-06  3:37           ` David Gibson
2008-06-07  0:30             ` Segher Boessenkool
2008-06-06  4:07         ` Grant Likely
2008-06-06  4:29           ` Sean MacLennan
2008-06-05  2:54       ` Sean MacLennan
2008-06-05  3:05         ` Josh Boyer
2008-06-05  3:16           ` Sean MacLennan
2008-06-05  6:22             ` Stefan Roese
2008-06-05  7:48               ` Jean Delvare
2008-06-05  8:45                 ` Stefan Roese
2008-06-05 10:57                   ` David Gibson
2008-06-05 11:52                   ` Josh Boyer
2008-06-05 15:18                     ` Timur Tabi
2008-06-05 22:47                     ` David Gibson
2008-06-06  4:17                   ` Benjamin Herrenschmidt
2008-06-06  4:16                 ` Benjamin Herrenschmidt
2008-06-06  6:21                   ` Jean Delvare
2008-06-06  7:47                     ` Grant Likely [this message]
2008-06-06  8:45                     ` Benjamin Herrenschmidt
2008-06-05 15:17               ` Timur Tabi
2008-06-05 15:44                 ` Jochen Friedrich
2008-06-05 15:50                   ` Timur Tabi
2008-06-05 16:10                     ` Grant Likely
2008-06-05 16:18                       ` Timur Tabi
2008-06-05 16:22                         ` Jochen Friedrich
2008-06-05 16:30                           ` Grant Likely
2008-06-05 16:40                             ` Timur Tabi
2008-06-05 22:46                               ` David Gibson
2008-06-05 16:35                         ` Grant Likely
2008-06-05 23:59                     ` Josh Boyer
2008-06-07  0:24                       ` Segher Boessenkool
2008-06-05 21:37               ` Sean MacLennan
2008-06-05 23:48                 ` Josh Boyer
2008-06-05 15:13       ` Timur Tabi
2008-06-05 15:39         ` Grant Likely
2008-06-05 15:43           ` Timur Tabi
2008-06-05 15:52             ` Segher Boessenkool
2008-06-05 16:09               ` Timur Tabi
2008-06-05 16:27                 ` Scott Wood
2008-06-05 17:52                   ` Timur Tabi
2008-06-05 18:04                     ` Scott Wood
2008-06-05 16:00             ` Grant Likely
2008-06-05 16:13               ` Timur Tabi
2008-06-05 16:21                 ` Josh Boyer
2008-06-05 16:25                   ` Timur Tabi
2008-06-05 16:37                     ` Grant Likely
2008-06-05 18:27                     ` Josh Boyer
2008-06-05 18:35                       ` Timur Tabi
2008-06-05 18:40                         ` Josh Boyer
2008-06-05 18:46                       ` Grant Likely
2008-06-05 18:56                         ` Josh Boyer
2008-06-05 19:14                           ` Grant Likely
2008-06-05 19:15                             ` Josh Boyer
2008-06-05 19:16                             ` Timur Tabi
2008-06-05 21:31                               ` Segher Boessenkool
2008-06-05 22:56                               ` David Gibson
2008-06-06 13:09                                 ` Timur Tabi
2008-06-06 13:42                                   ` Stefan Roese
2008-06-05 22:45                             ` Sean MacLennan
2008-06-06  4:20             ` Benjamin Herrenschmidt
2008-06-25 21:46               ` Timur Tabi
2008-06-27 16:48                 ` Jochen Friedrich
2008-06-05 15:46           ` Segher Boessenkool
2008-06-05 15:52           ` Jochen Friedrich
2008-06-05 15:53             ` Grant Likely
2008-06-06  4:19         ` Benjamin Herrenschmidt
2008-06-06  4:14       ` Benjamin Herrenschmidt
2008-06-06  4:13     ` Benjamin Herrenschmidt

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=fa686aa40806060047n7b8b8e6flee193d46d4e3dd24@mail.gmail.com \
    --to=grant.likely@secretlab.ca \
    --cc=khali@linux-fr.org \
    --cc=linuxppc-dev@ozlabs.org \
    --cc=scottwood@freescale.com \
    --cc=smaclennan@pikatech.com \
    --cc=sr@denx.de \
    --cc=timur@freescale.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;
as well as URLs for NNTP newsgroup(s).