LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Kristoffer Ericson <kristoffer.ericson@gmail.com>
To: Ben Dooks <ben@fluff.org>
Cc: Kristoffer Ericson <kristoffer.ericson@gmail.com>,
	Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: Questions regarding mfd drivers
Date: Tue, 5 Feb 2008 12:21:27 +0100	[thread overview]
Message-ID: <20080205122127.63846543.Kristoffer.ericson@gmail.com> (raw)
In-Reply-To: <20080204230155.GD27399@fluff.org.uk>

On Mon, 4 Feb 2008 23:01:55 +0000
Ben Dooks <ben@fluff.org> wrote:

> On Mon, Feb 04, 2008 at 12:37:14PM +0100, Kristoffer Ericson wrote:
> > Greetings,
> > 
> > Trying to wrap my head around sm501. From what I can tell an mfd driver is a "master" driver that takes control of all
> > memory and io areas. It then hands out areas of those to drivers. Anywhere near correct?
> 
> Yes, it is the central management for these, but also ensures that any
> of the sub-drivers have properly locked access to the clocks, gpio and
> other shared resources.

Oki

> 
> The mfd driver for the sm501 exports a number of functions for the
> sub drivers to use, you should be able to see what is exported easily
> by the fact they are exported with EXPORT_SYMBOL_GPL(). The header
> files should document basic functionality of this.
> 

Will take a look at that.
 
> > I can see some benefit but still hard for me to motivate. What am I missing? What will the mfd be able to do, that I lack now?
> > 
> > The sm501 driver seems way more advanced than I will need for hd64461/hd64465 anyhow, but still need to understand sm501 completely before
> > attempting to write one on my own. Anyone know any documentation aside from example drivers?
> 
> Are you trying to write your own SM501, or something else? It seems
> you are writing for something else.
>

Im trying to move the companion chip hd64461 to a more sensible location. Paul suggested building an mfd driver.
The hd64461 chipset supplies for example pcmcia and framebuffer support. Its not as advanced as the SM501.
 
> If the chip you are targetting has shared resources, such as clock
> gates, PLLs, or gpio that other drivers need to touch, then the best
> way to go is for an mfd driver to provide this functionality and have
> all the child drivers use the exported functionality.

Oki, sounds good. Thanks for info.

> 
> -- 
> Ben (ben@fluff.org, http://www.fluff.org/)
> 
>   'a smiley only costs 4 bytes'

      reply	other threads:[~2008-02-05 11:21 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-02-04 11:37 Kristoffer Ericson
2008-02-04 23:01 ` Ben Dooks
2008-02-05 11:21   ` Kristoffer Ericson [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=20080205122127.63846543.Kristoffer.ericson@gmail.com \
    --to=kristoffer.ericson@gmail.com \
    --cc=ben@fluff.org \
    --cc=linux-kernel@vger.kernel.org \
    --subject='Re: Questions regarding mfd drivers' \
    /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

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).