mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: David Brownell <david-b@pacbell.net>
To: "Frank Buss" <fb@frank-buss.de>
Cc: linux-kernel@vger.kernel.org
Subject: Re: SL811 problem on mach-pxa
Date: Thu, 17 Feb 2005 12:10:56 -0800	[thread overview]
Message-ID: <200502171210.56627.david-b@pacbell.net> (raw)
In-Reply-To: <20050217191130.B750B5B874@frankbuss.de>

On Thursday 17 February 2005 11:11 am, Frank Buss wrote:
> > Some of that looks reasonable, not all.  In particular, don't
> > change the convention on resources (memory to i/o), or expect
> > that the two regions involve more than one byte each ... the
> > hardware only has two single-byte registers!
> 
> ok, perhaps I've misunderstood the meaning of IORESOURCE_IO and
> IORESOURCE_MEM. Is IORESOURCE_IO for "outb" and "inb" (Intel assembler,
> don't know the Arm aquivalent)?

Yes, and there _is_ no ARM equivalent.  Modern CPUs generaly won't
bother with a separate physical I/O space, and special instructions
to access it.  So Linux drivers use macros that always boil down to
normal memory-mapped I/O accessors ... that's what ARM does, and
oddly enough all current users of this driver are on PXA hardware.

I'm not sure what the CFU1U(*) CF/PCMCIA support will need; if that
card maps those registers to I/O space, this driver will need some
minor tweaks (in addition to the PCMCIA framework glue that registers
a new platform_device).  That is, the registers would normally be
memory-mapped, but on some boards they might need to be in I/O space.

All your platform should need to do is initialize a platform_device
with the three resources (two memory, one IRQ), and platform_data
initialized to match the structure defined for that purpose.


> > That looks like the sort of thing that
> > should be done in the reset() routine rather than start(); 
> > and it should
> > certainly use a symbolic constant not 0x08.
> 
> do you mean sl811->board->reset? 

If your board is wired to support a separate chip reset (maybe
through a GPIO or something), yes; otherwise, I suppose in this
case I meant the port_power() routine should be handling that.

This driver doesn't have a separate hc_driver.reset() entry, it
does the analagous stuff as part of powering up the chip ... but
as you've noted, that approach is a bit troublesome on boards
that don't have that level of control over the hardware.

It's best to ask questions about USB drivers on linux-usb-devel,
not all USB developers make time to swim through LKML.

- Dave

(*) http://www.ratocsystems.com/english/products/subpages/cfu1u.html
    CompactFlash single-port USB host adapter, for PDAs, using
    the SL811HS chip.


      reply	other threads:[~2005-02-17 20:17 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-02-14 19:03 Frank Buss
2005-02-17  3:51 ` Frank Buss
2005-02-17 18:09   ` David Brownell
2005-02-17 19:11     ` Frank Buss
2005-02-17 20:10       ` David Brownell [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=200502171210.56627.david-b@pacbell.net \
    --to=david-b@pacbell.net \
    --cc=fb@frank-buss.de \
    --cc=linux-kernel@vger.kernel.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

all inboxes | Powered by JetHome®