LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
To: Rob Herring <robh@kernel.org>
Cc: linuxarm@huawei.com, mauro.chehab@huawei.com,
	Mauro Carvalho Chehab <mchehab+huawei@kernel.org>,
	Bjorn Helgaas <bhelgaas@google.com>,
	linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org
Subject: [PATCH 1/3] PCI: of: Fix handling of multi-level PCI devices
Date: Thu,  5 Aug 2021 10:28:58 +0200	[thread overview]
Message-ID: <3365235394fe4e7f35694c95af95fce96da7c9bb.1628151761.git.mchehab+huawei@kernel.org> (raw)
In-Reply-To: <cover.1628151760.git.mchehab+huawei@kernel.org>

When the DT schema is like:

	pcie@f4000000 {
		pcie@0,0 {
			pcie@1,0 {
			};
			pcie@5,0 {
			};
			pcie@7,0 {
			};
		};
	};

The logic under pci_set_bus_of_node() will try to setup some
buses with a NULL node, causing it to register just a small
set of devices:

	$ find /sys/devices/platform/soc/f4000000.pcie/ -name of_node
	/sys/devices/platform/soc/f4000000.pcie/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/pci_bus/0000:01/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/pci_bus/0000:00/of_node

On such case, it needs to go to the parent node, in order to register
everything:

	$ find /sys/devices/platform/soc/f4000000.pcie/ -name of_node
	/sys/devices/platform/soc/f4000000.pcie/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/0000:02:07.0/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/0000:02:07.0/pci_bus/0000:06/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/pci_bus/0000:02/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/0000:02:05.0/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/0000:02:05.0/pci_bus/0000:05/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/0000:02:01.0/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/0000:02:01.0/pci_bus/0000:03/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/0000:00:00.0/pci_bus/0000:01/of_node
	/sys/devices/platform/soc/f4000000.pcie/pci0000:00/pci_bus/0000:00/of_node

Adding some debug prints with such change will produce the following
output:

	$ dmesg|grep of_node
	[    4.932405]  (null): pci_set_bus_of_node: of_node: /soc/pcie@f4000000
	[    4.985916] pci 0000:00:00.0: pci_set_of_node: of_node: /soc/pcie@f4000000
	[    5.014190] pci_bus 0000:01: pci_set_bus_of_node: of_node: /soc/pcie@f4000000/pcie@0,0
	[    5.065680] pci 0000:01:00.0: pci_set_of_node: of_node: /soc/pcie@f4000000/pcie@0,0
	[    5.118754] pci_bus 0000:02: pci_set_bus_of_node: use of_node of the parent
	[    5.135279] pci_bus 0000:02: pci_set_bus_of_node: of_node: /soc/pcie@f4000000/pcie@0,0
	[    5.158921] pci 0000:02:01.0: pci_set_of_node: of_node: /soc/pcie@f4000000/pcie@0,0
	[    5.187393] pci 0000:02:04.0: pci_set_of_node: of_node: /soc/pcie@f4000000/pcie@0,0
	[    5.215982] pci 0000:02:05.0: pci_set_of_node: of_node: /soc/pcie@f4000000/pcie@0,0
	[    5.244607] pci 0000:02:07.0: pci_set_of_node: of_node: /soc/pcie@f4000000/pcie@0,0
	[    5.272825] pci 0000:02:09.0: pci_set_of_node: of_node: /soc/pcie@f4000000/pcie@0,0
	[    5.335258] pci_bus 0000:03: pci_set_bus_of_node: of_node: /soc/pcie@f4000000/pcie@0,0/pcie@1,0
	[    5.367538] pci 0000:03:00.0: pci_set_of_node: of_node: /soc/pcie@f4000000/pcie@0,0/pcie@1,0
	[    5.415959] pci_bus 0000:04: pci_set_bus_of_node: use of_node of the parent
	[    5.424190] pci_bus 0000:04: pci_set_bus_of_node: of_node: (null)
	[    5.438727] pci_bus 0000:05: pci_set_bus_of_node: of_node: /soc/pcie@f4000000/pcie@0,0/pcie@5,0
	[    5.455691] pci_bus 0000:06: pci_set_bus_of_node: of_node: /soc/pcie@f4000000/pcie@0,0/pcie@7,0
	[    5.491643] pci 0000:06:00.0: pci_set_of_node: of_node: /soc/pcie@f4000000/pcie@0,0/pcie@7,0
	[    5.526157] pci_bus 0000:07: pci_set_bus_of_node: use of_node of the parent
	[    5.534361] pci_bus 0000:07: pci_set_bus_of_node: of_node: (null)

Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
---
 drivers/pci/of.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/pci/of.c b/drivers/pci/of.c
index a143b02b2dcd..b6fa3bd46ae6 100644
--- a/drivers/pci/of.c
+++ b/drivers/pci/of.c
@@ -41,6 +41,8 @@ void pci_set_bus_of_node(struct pci_bus *bus)
 		node = pcibios_get_phb_of_node(bus);
 	} else {
 		node = of_node_get(bus->self->dev.of_node);
+		if (!node)
+			node = of_node_get(bus->self->dev.parent->of_node);
 		if (node && of_property_read_bool(node, "external-facing"))
 			bus->self->external_facing = true;
 	}
-- 
2.31.1


  reply	other threads:[~2021-08-05  8:29 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-05  8:28 [PATCH 0/3] Fix handling of multi-level OF nodes Mauro Carvalho Chehab
2021-08-05  8:28 ` Mauro Carvalho Chehab [this message]
2021-08-06 22:52   ` [PATCH 1/3] PCI: of: Fix handling of multi-level PCI devices Bjorn Helgaas
2021-08-05  8:28 ` [PATCH 2/3] PCI: of: Setup PCI before setting the OF node Mauro Carvalho Chehab
2021-08-05  8:29 ` [PATCH 3/3] PCI: of: Add some debug printks to track problems with of_node setup Mauro Carvalho Chehab

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=3365235394fe4e7f35694c95af95fce96da7c9bb.1628151761.git.mchehab+huawei@kernel.org \
    --to=mchehab+huawei@kernel.org \
    --cc=bhelgaas@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=mauro.chehab@huawei.com \
    --cc=robh@kernel.org \
    --subject='Re: [PATCH 1/3] PCI: of: Fix handling of multi-level PCI devices' \
    /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

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).