public inbox for linuxppc-dev@ozlabs.org
 help / color / mirror / Atom feed
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: "Hommel, Thomas (GE Indust, GE Fanuc)" <Thomas.Hommel@gefanuc.com>
Cc: linuxppc-dev@ozlabs.org
Subject: Re: Set up device tree for PCI bus
Date: Fri, 12 Oct 2007 21:27:14 +1000	[thread overview]
Message-ID: <1192188434.8476.112.camel@pasglop> (raw)
In-Reply-To: <62DDBB9E5E23CC4A929EE46F9427CEAF3682BB@BUDMLVEM04.e2k.ad.ge.com>


On Fri, 2007-10-12 at 12:48 +0200, Hommel, Thomas (GE Indust, GE Fanuc)
wrote:
> Hi all,
> I'm wondering how to set up a proper device tree for a PCI bus. The bus
> has a tree-like structure with several bridges and can be extended
> dynamically (by adding PMC/XMC modules).

  .../...

> A problem is that the modules in the Expansion/PMC slot can contain more
> bridges and therefore the bus numbering isn't fixed. For example, if the
> PMC adds one more bus, #5 becomes #6 and #6 becomes #7.
> Can I assign fixed resources for all the bridge parts of the system?

ppc32 should cope with renumbering... though ppc64 will probably not.

What you can do perhaps is to assign wide bus ranges to your slots. For
example, slot 0 would provide bus 0x10...0x1f, slot 1 bus 0x20 to 0x2f
etc... You would thus only put the switch busses in the device-tree.

There is still a potential issue if you connect enough sub busses to
blow that range of 16 away but that's unlikely.

Such a solution would work for the time being. In the long run, I do
intend to make the kernel more flexible overall with bus renumbering vs.
OF tree.

      reply	other threads:[~2007-10-12 11:27 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-10-12 10:48 Set up device tree for PCI bus Hommel, Thomas (GE Indust, GE Fanuc)
2007-10-12 11:27 ` Benjamin Herrenschmidt [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=1192188434.8476.112.camel@pasglop \
    --to=benh@kernel.crashing.org \
    --cc=Thomas.Hommel@gefanuc.com \
    --cc=linuxppc-dev@ozlabs.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