mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Robin Murphy <robin.murphy@arm.com>
To: Lorenzo Pieralisi <lpieralisi@kernel.org>,
	Hanjun Guo <guohanjun@huawei.com>,
	Sudeep Holla <sudeep.holla@arm.com>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	Len Brown <lenb@kernel.org>, Russell King <linux@armlinux.org.uk>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Danilo Krummrich <dakr@kernel.org>,
	Stuart Yoder <stuyoder@gmail.com>,
	Laurentiu Tudor <laurentiu.tudor@nxp.com>,
	Nipun Gupta <nipun.gupta@amd.com>,
	Nikhil Agarwal <nikhil.agarwal@amd.com>,
	Joerg Roedel <joro@8bytes.org>, Will Deacon <will@kernel.org>,
	Rob Herring <robh@kernel.org>,
	Saravana Kannan <saravanak@google.com>,
	Bjorn Helgaas <bhelgaas@google.com>
Cc: linux-acpi@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, iommu@lists.linux.dev,
	devicetree@vger.kernel.org, linux-pci@vger.kernel.org,
	Charan Teja Kalla <quic_charante@quicinc.com>
Subject: [PATCH v2 3/4] iommu: Keep dev->iommu state consistent
Date: Fri, 28 Feb 2025 15:46:32 +0000	[thread overview]
Message-ID: <d219663a3f23001f23d520a883ac622d70b4e642.1740753261.git.robin.murphy@arm.com> (raw)
In-Reply-To: <cover.1740753261.git.robin.murphy@arm.com>

At the moment, if of_iommu_configure() allocates dev->iommu itself via
iommu_fwspec_init(), then suffers a DT parsing failure, it cleans up the
fwspec but leaves the empty dev_iommu hanging around. So far this is
benign (if a tiny bit wasteful), but we'd like to be able to reason
about dev->iommu having a consistent and unambiguous lifecycle. Thus
make sure that the of_iommu cleanup undoes precisely whatever it did.

Signed-off-by: Robin Murphy <robin.murphy@arm.com>
---

v2: New

 drivers/iommu/iommu-priv.h | 2 ++
 drivers/iommu/iommu.c      | 2 +-
 drivers/iommu/of_iommu.c   | 6 +++++-
 3 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/iommu/iommu-priv.h b/drivers/iommu/iommu-priv.h
index 17e245cf17bb..6574c1f97e14 100644
--- a/drivers/iommu/iommu-priv.h
+++ b/drivers/iommu/iommu-priv.h
@@ -17,6 +17,8 @@ static inline const struct iommu_ops *dev_iommu_ops(struct device *dev)
 	return dev->iommu->iommu_dev->ops;
 }
 
+void dev_iommu_free(struct device *dev);
+
 const struct iommu_ops *iommu_ops_from_fwnode(const struct fwnode_handle *fwnode);
 
 static inline const struct iommu_ops *iommu_fwspec_ops(struct iommu_fwspec *fwspec)
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 37b7088e129a..a3b45b84f42b 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -352,7 +352,7 @@ static struct dev_iommu *dev_iommu_get(struct device *dev)
 	return param;
 }
 
-static void dev_iommu_free(struct device *dev)
+void dev_iommu_free(struct device *dev)
 {
 	struct dev_iommu *param = dev->iommu;
 
diff --git a/drivers/iommu/of_iommu.c b/drivers/iommu/of_iommu.c
index 97987cd78da9..e10a68b5ffde 100644
--- a/drivers/iommu/of_iommu.c
+++ b/drivers/iommu/of_iommu.c
@@ -116,6 +116,7 @@ static void of_pci_check_device_ats(struct device *dev, struct device_node *np)
 int of_iommu_configure(struct device *dev, struct device_node *master_np,
 		       const u32 *id)
 {
+	bool dev_iommu_present;
 	int err;
 
 	if (!master_np)
@@ -127,6 +128,7 @@ int of_iommu_configure(struct device *dev, struct device_node *master_np,
 		mutex_unlock(&iommu_probe_device_lock);
 		return 0;
 	}
+	dev_iommu_present = dev->iommu;
 
 	/*
 	 * We don't currently walk up the tree looking for a parent IOMMU.
@@ -147,8 +149,10 @@ int of_iommu_configure(struct device *dev, struct device_node *master_np,
 		err = of_iommu_configure_device(master_np, dev, id);
 	}
 
-	if (err)
+	if (err && dev_iommu_present)
 		iommu_fwspec_free(dev);
+	else if (err && dev->iommu)
+		dev_iommu_free(dev);
 	mutex_unlock(&iommu_probe_device_lock);
 
 	if (!err && dev->bus)
-- 
2.39.2.101.g768bb238c484.dirty


  parent reply	other threads:[~2025-02-28 15:46 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-28 15:46 [PATCH v2 0/4] iommu: Fix the longstanding probe issues Robin Murphy
2025-02-28 15:46 ` [PATCH v2 1/4] iommu: Handle race with default domain setup Robin Murphy
2025-02-28 15:46 ` [PATCH v2 2/4] iommu: Resolve ops in iommu_init_device() Robin Murphy
2025-03-05 17:55   ` Jason Gunthorpe
2025-02-28 15:46 ` Robin Murphy [this message]
2025-03-05 18:14   ` [PATCH v2 3/4] iommu: Keep dev->iommu state consistent Jason Gunthorpe
2025-02-28 15:46 ` [PATCH v2 4/4] iommu: Get DT/ACPI parsing into the proper probe path Robin Murphy
2025-03-05 18:28   ` Jason Gunthorpe
2025-03-07 14:24   ` Lorenzo Pieralisi
2025-03-07 20:20     ` Robin Murphy
2025-03-11 18:42   ` Joerg Roedel
2025-03-12  7:07     ` Baolu Lu
2025-03-12 10:10     ` Robin Murphy
2025-03-12 14:34       ` Baolu Lu
2025-03-12 15:21       ` Joerg Roedel
     [not found]   ` <CGME20250313095633eucas1p29cb55f2504b4bcf67c16b3bd3fa9b8cd@eucas1p2.samsung.com>
2025-03-13  9:56     ` Marek Szyprowski
2025-03-13 11:01       ` Robin Murphy
2025-03-13 12:23         ` Marek Szyprowski
2025-03-13 13:06           ` Robin Murphy
2025-03-13 14:12             ` Robin Murphy
2025-03-17  7:37               ` Marek Szyprowski
2025-03-17 18:22                 ` Robin Murphy
2025-03-21 12:15                   ` Marek Szyprowski
2025-03-21 16:48                     ` Robin Murphy
2025-04-01 20:34                       ` Marek Szyprowski
2025-03-13 16:30         ` Anders Roxell
2025-03-18 16:37   ` Geert Uytterhoeven
2025-03-18 17:24     ` Robin Murphy
2025-03-25 15:32       ` Geert Uytterhoeven
2025-03-27  9:47   ` Chen-Yu Tsai
2025-03-27 11:00     ` Louis-Alexis Eyraud
2025-04-11  8:02   ` Johan Hovold
2025-04-14 15:37     ` Robin Murphy
2025-04-15 15:08       ` Johan Hovold
2025-04-24 13:58         ` Robin Murphy
2025-04-21 21:19   ` William McVicker
2025-04-22 19:00     ` Jason Gunthorpe
2025-04-22 21:55       ` William McVicker
2025-04-22 23:41         ` Jason Gunthorpe
2025-04-23 17:31           ` William McVicker
2025-04-23 18:18             ` Jason Gunthorpe
2025-08-11 16:44   ` Eric Auger
2025-08-11 17:01     ` Bjorn Helgaas
2026-03-23 17:18   ` Tudor Ambarus
2026-03-23 20:49     ` Robin Murphy
2026-04-01 11:49       ` Tudor Ambarus
2025-03-10  8:29 ` [PATCH v2 0/4] iommu: Fix the longstanding probe issues Joerg Roedel

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=d219663a3f23001f23d520a883ac622d70b4e642.1740753261.git.robin.murphy@arm.com \
    --to=robin.murphy@arm.com \
    --cc=bhelgaas@google.com \
    --cc=dakr@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=guohanjun@huawei.com \
    --cc=iommu@lists.linux.dev \
    --cc=joro@8bytes.org \
    --cc=laurentiu.tudor@nxp.com \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux@armlinux.org.uk \
    --cc=lpieralisi@kernel.org \
    --cc=nikhil.agarwal@amd.com \
    --cc=nipun.gupta@amd.com \
    --cc=quic_charante@quicinc.com \
    --cc=rafael@kernel.org \
    --cc=robh@kernel.org \
    --cc=saravanak@google.com \
    --cc=stuyoder@gmail.com \
    --cc=sudeep.holla@arm.com \
    --cc=will@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

all inboxes | Powered by JetHome®