linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Bjorn Helgaas <helgaas@kernel.org>
To: Alex Williamson <alex.williamson@redhat.com>
Cc: bhelgaas@google.com, linux-pci@vger.kernel.org, geoff@hostfission.com
Subject: Re: [PATCH] PCI: Release unused bridge resources during resize
Date: Mon, 10 Jun 2024 10:20:34 -0500	[thread overview]
Message-ID: <20240610152034.GA943788@bhelgaas> (raw)
In-Reply-To: <20240607170156.3e59f580.alex.williamson@redhat.com>

On Fri, Jun 07, 2024 at 05:01:56PM -0600, Alex Williamson wrote:
> On Fri, 7 Jun 2024 17:33:20 -0500
> Bjorn Helgaas <helgaas@kernel.org> wrote:
> 
> > On Tue, May 07, 2024 at 03:31:23PM -0600, Alex Williamson wrote:
> > > Resizing BARs can be blocked when a device in the bridge hierarchy
> > > itself consumes resources from the resized range.  This scenario is
> > > common with Intel Arc DG2 GPUs where the following is a typical
> > > topology:
> > > 
> > >  +-[0000:5d]-+-00.0-[5e-61]----00.0-[5f-61]--+-01.0-[60]----00.0  Intel Corporation DG2 [Arc A380]
> > >                                              \-04.0-[61]----00.0  Intel Corporation DG2 Audio Controller
> ...

> > > pci 0000:60:00.0: BAR 2 [mem 0xbfe0000000-0xbfefffffff 64bit pref]: releasing
> > > pcieport 0000:5f:01.0: bridge window [mem 0xbfe0000000-0xbfefffffff 64bit pref]: releasing
> > > pci 0000:5e:00.0: bridge window [mem 0xbfe0000000-0xbfefffffff 64bit pref]: releasing
> > > pci 0000:5e:00.0: BAR 0 [mem 0xbff0000000-0xbff07fffff 64bit pref]: releasing
> > > pcieport 0000:5d:00.0: bridge window [mem 0xbfe0000000-0xbff07fffff 64bit pref]: releasing
> > > pcieport 0000:5d:00.0: bridge window [mem 0xb000000000-0xb2ffffffff 64bit pref]: assigned
> > > pci 0000:5e:00.0: bridge window [mem 0xb000000000-0xb1ffffffff 64bit pref]: assigned
> > > pci 0000:5e:00.0: BAR 0 [mem 0xb200000000-0xb2007fffff 64bit pref]: assigned

It didn't occur to me before, but don't we have a potential issue here
because we relocated 5e:00.0 BAR 0 from 0xbff0000000 to 0xb200000000
without notifying any code that might be using it?

I'm worried about vendor-specific perf/EDAC/etc drivers that can't
claim the the bridge using pci_register_driver() because the pcieport
driver is in the way.  I think some of those drivers go behind the
back of the driver model to locate their device and ioremap the BAR
direction.

Bjorn

      reply	other threads:[~2024-06-10 15:20 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-05-07 21:31 [PATCH] PCI: Release unused bridge resources during resize Alex Williamson
2024-05-13 13:46 ` Ilpo Järvinen
2024-05-16 13:49   ` Alex Williamson
2024-05-20 14:59     ` Ilpo Järvinen
2024-06-07 20:44 ` Alex Williamson
2024-06-07 22:33 ` Bjorn Helgaas
2024-06-07 23:01   ` Alex Williamson
2024-06-10 15:20     ` Bjorn Helgaas [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=20240610152034.GA943788@bhelgaas \
    --to=helgaas@kernel.org \
    --cc=alex.williamson@redhat.com \
    --cc=bhelgaas@google.com \
    --cc=geoff@hostfission.com \
    --cc=linux-pci@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;
as well as URLs for NNTP newsgroup(s).