public inbox for linuxppc-dev@ozlabs.org 
 help / color / mirror / Atom feed
From: Scott Wood <scottwood@freescale•com>
To: Kumar Gala <kumar.gala@freescale•com>
Cc: linuxppc-dev <linuxppc-dev@ozlabs•org>,
	Li Li <r64360@freescale•com>, Li Tony <Tony.Li@freescale•com>
Subject: Re: [PATCH] Add MPC837xEMDS PCIE RC mode support
Date: Fri, 30 Nov 2007 09:19:28 -0600	[thread overview]
Message-ID: <20071130151928.GB29484@loki.buserror.net> (raw)
In-Reply-To: <6FF36777-D91C-4B80-968B-CAEF11BC98ED@freescale.com>

On Fri, Nov 30, 2007 at 08:57:33AM -0600, Kumar Gala wrote:
> On Nov 30, 2007, at 3:37 AM, Li Li wrote:
> > If do it standard, a 256M config space, at least 256M mem space and  
> > 16M
> > io space are needed for each PCIE controller.
> > To allocate PCIE window, the window size only can be 512M or 1G.
> > If we choose 1G space, two PCIE controller needs 2G space.
> > We do not have 2G free physical space now. Usually, we use upper 128M
> > configure space. So, we have to cut down the config space.

That decision should be made in the kernel, not the dts.

> Is it possible to make the outbound window for cfg space 4k and change  
> the region of config space its looking at?

Yes, that'd be best.

> > Here is a little complex. The MPC837xE board needs a carrier board to
> > extend PCIE slot. If user does not populate carrier board onto MPC837xE
> > board and do PCIE scan, the system will halt. I just want to provide a
> > easy way to disable the PCIe other than modify and recompile the dts.
> 
> So I have to recompile the kernel for this case, that seems even more  
> painful to the user.  Can we not detect if this board isn't there and  
> not hang?

Assuming it's similar to PCI on other 83xxMDS, yes -- there's a bit in the
reset control words that can be checked.

We don't currently do it on any 83xx that I'm aware of, though. :-P

-Scott

  reply	other threads:[~2007-11-30 15:19 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-30  3:45 [PATCH] Add MPC837xEMDS PCIE RC mode support Li Li
2007-11-30  4:14 ` Olof Johansson
2007-11-30  5:33   ` Li Li
2007-11-30 15:12   ` Scott Wood
2007-11-30  7:37 ` Kumar Gala
2007-11-30  8:48   ` Li Li
2007-11-30  9:05     ` Kumar Gala
2007-11-30  9:37       ` Li Li
2007-11-30 14:57         ` Kumar Gala
2007-11-30 15:19           ` Scott Wood [this message]
2007-12-03  3:43           ` Li Li

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=20071130151928.GB29484@loki.buserror.net \
    --to=scottwood@freescale$(echo .)com \
    --cc=Tony.Li@freescale$(echo .)com \
    --cc=kumar.gala@freescale$(echo .)com \
    --cc=linuxppc-dev@ozlabs$(echo .)org \
    --cc=r64360@freescale$(echo .)com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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