linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Geert Uytterhoeven <geert@linux-m68k.org>
To: Deepak Saxena <dsaxena@plexity.net>
Cc: Linus Torvalds <torvalds@osdl.org>,
	Linux Kernel Development <linux-kernel@vger.kernel.org>
Subject: Re: Being more anal about iospace accesses..
Date: Thu, 16 Sep 2004 14:51:38 +0200 (MEST)	[thread overview]
Message-ID: <Pine.GSO.4.58.0409161410080.23693@waterleaf.sonytel.be> (raw)
In-Reply-To: <20040915230904.GA19450@plexity.net>

On Wed, 15 Sep 2004, Deepak Saxena wrote:
> On Sep 15 2004, at 15:46, Linus Torvalds was caught saying:
> > Quite frankly, of your two suggested interfaces, I would select neither.
> > I'd just say that if your bus is special enough, just write your own
> > drivers, and use
> >
> > 	cookie = ixp4xx_iomap(dev, xx);
> > 	...
> > 	ixp4xx_iowrite(val, cookie + offset);
> >
> > which is perfectly valid. You don't have to make these devices even _look_
> > like a PCI device. Why should you?
>
> Problem is that some of those devices are not that special. For example,
> the on-board 16550 is accessed using readb/writeb in the 8250.c driver.

Just what I was thinking...

> I don't think we want to add that level of low-level detail to that
> driver and instead should just hide it in the platform code. I look
> at it from the point of view that the driver should not care about how
> the access actually occurs on the bus. It just says, write data foo at
> location bar regardless of whether bar is ISA, PCI, on-chip, RapidIO,
> etc and that writing of the data is hidden in the implementation of
> the accessor API.

While 16550 serial is a bad example since it does byte accesses only (and thus
doesn't suffer from endianness), I have no problem to imagine there exist
platforms where you have multiple instances of a `standard' (cfr. 16550 serial)
device block, while each of them must be accessed differently:
  - one of them is in PCI I/O space (little endian)
  - one of them is in PCI MMIO space (little endian)
  - one of them is on-chip MMIO (big endian)
  - one of them is somewhere else, but sparsely addressed (some bytes of
    padding between each register)
and we can for sure come up with a few more examples of weird addressing.

How to solve this, without cluttering each ioread*() with many if clauses?

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

  reply	other threads:[~2004-09-16 12:51 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <Pine.LNX.4.58.0409081543320.5912@ppc970.osdl.org>
     [not found] ` <Pine.LNX.4.58.0409150737260.2333@ppc970.osdl.org>
2004-09-15 16:30   ` Being more anal about iospace accesses Linus Torvalds
2004-09-15 16:54     ` Jörn Engel
2004-09-15 17:07       ` Linus Torvalds
2004-09-15 17:32         ` Jörn Engel
2004-09-15 17:57           ` Linus Torvalds
2004-09-15 18:06             ` Linus Torvalds
2004-09-15 19:34             ` Greg KH
2004-09-15 19:53               ` Linus Torvalds
2004-09-15 20:06                 ` Greg KH
2004-09-16 14:58             ` Horst von Brand
2004-09-15 17:45         ` Nikita Danilov
2004-09-15 18:09           ` Linus Torvalds
2004-09-15 18:40         ` Chris Wedgwood
2004-09-15 17:07       ` Jeff Garzik
2004-09-15 17:16       ` Roland Dreier
2004-09-15 17:39         ` Linus Torvalds
2004-09-15 20:07           ` Russell King
2004-09-15 17:36       ` Horst von Brand
2004-09-15 17:40       ` Brian Gerst
2004-09-15 16:56     ` Dave Jones
2004-09-15 17:19     ` Roger Luethi
2004-09-15 17:23     ` Richard B. Johnson
2004-09-15 22:21     ` Deepak Saxena
2004-09-15 22:46       ` Linus Torvalds
2004-09-15 23:09         ` Deepak Saxena
2004-09-16 12:51           ` Geert Uytterhoeven [this message]
2004-09-16 22:10             ` Deepak Saxena
2004-09-15 22:29     ` Roland Dreier
2004-09-15 23:26       ` Being more careful " Linus Torvalds
2004-09-16  0:10         ` viro
2004-09-16 11:40           ` David Woodhouse
2004-09-16 12:25         ` David Woodhouse
2004-09-16 14:01           ` Linus Torvalds
2004-09-18  9:46             ` Kai Henningsen
2004-09-15 18:25 Being more anal " linux

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=Pine.GSO.4.58.0409161410080.23693@waterleaf.sonytel.be \
    --to=geert@linux-m68k.org \
    --cc=dsaxena@plexity.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@osdl.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;
as well as URLs for NNTP newsgroup(s).