From: Julien Grall <julien.grall@citrix.com>
To: "Ferger, Max" <M.Ferger@KOSTAL.COM>,
Ian Campbell <ian.campbell@citrix.com>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>
Cc: Oleksandr Tyshchenko <oleksandr.tyshchenko@globallogic.com>,
Iurii Konovalenko <iurii.konovalenko@globallogic.com>
Subject: Re: Dom0 kernel for Xen4.6 on R-Car H2 (LAGER)
Date: Tue, 20 Oct 2015 11:47:26 +0100 [thread overview]
Message-ID: <56261BBE.9090708@citrix.com> (raw)
In-Reply-To: <562611D1.5050501@citrix.com>
On 20/10/2015 11:05, Julien Grall wrote:
> Thank you for the log. The offending node is /pci@ee090000/usb@0,1 which
> is (r8a7790.dtsi):
>
> 1407 usb@0,1 {
> 1408 reg = <0x800 0 0 0 0>;
> 1409 device_type = "pci";
> 1410 phys = <&usb0 0>;
> 1411 phy-names = "usb";
> 1412 };
> 1413
>
> We assume that every node containing device_type = "pci" is a PCI bus.
> So Xen is looking for a property "ranges" which is not present.
>
> Although, here it's describing a PCI device. I wasn't able to find a
> spec telling that device_type = "pci"; should only be used by the PCI
> bus. Ian, do you remember why you implement with this assumption?
>
> For now, I would advice you to drop this node, the next one (usb@0,2)
> and doing the same in "pci@ee0d0000". We can figure out later how to fix
> it when you get a DOM0 booting.
Better solution, rather than dropping the nodes can you add "ranges;" in
each of them. Smth like
usb@0,1 {
reg = <0x800 0 0 0 0>;
device_type = "pci";
phys = <&usb0 0>;
phy-names = "usb";
ranges;
}
"ranges;" is used to tell the PCI device is using the same address space
of the PCI bus. There is DT bindings describing how look like the
subnode and they mandate the property "ranges" (see [1], [2]). Although
renesas doesn't seem to require it. Might be worth to investigate it.
Regards,
[1]
http://lxr.free-electrons.com/source/Documentation/devicetree/bindings/pci/mvebu-pci.txt
[2]
http://lxr.free-electrons.com/source/Documentation/devicetree/bindings/pci/nvidia,tegra20-pcie.txt
--
Julien Grall
next prev parent reply other threads:[~2015-10-20 10:47 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-15 14:17 Dom0 kernel for Xen4.6 on R-Car H2 (LAGER) Ferger, Max
2015-10-15 14:27 ` Ian Campbell
2015-10-19 12:09 ` Ferger, Max
2015-10-19 12:44 ` Julien Grall
2015-10-20 8:48 ` Ferger, Max
2015-10-20 10:05 ` Julien Grall
2015-10-20 10:47 ` Julien Grall [this message]
2015-10-20 11:41 ` Ferger, Max
2015-10-20 11:23 ` Ferger, Max
2015-10-20 21:44 ` Julien Grall
2015-10-21 16:05 ` Ferger, Max
2015-10-21 16:23 ` Julien Grall
2015-10-21 17:20 ` Ferger, Max
2015-10-21 18:01 ` Julien Grall
2015-10-22 15:07 ` Ferger, Max
2015-10-22 15:29 ` Julien Grall
2015-10-22 8:48 ` Ian Campbell
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=56261BBE.9090708@citrix.com \
--to=julien.grall@citrix.com \
--cc=M.Ferger@KOSTAL.COM \
--cc=ian.campbell@citrix.com \
--cc=iurii.konovalenko@globallogic.com \
--cc=oleksandr.tyshchenko@globallogic.com \
--cc=xen-devel@lists.xen.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.