linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sven Peter <sven@svenpeter.dev>
To: iommu@lists.linux-foundation.org
Cc: Sven Peter <sven@svenpeter.dev>, Joerg Roedel <joro@8bytes.org>,
	Will Deacon <will@kernel.org>,
	Robin Murphy <robin.murphy@arm.com>,
	Arnd Bergmann <arnd@kernel.org>,
	Mohamed Mediouni <mohamed.mediouni@caramail.com>,
	Alexander Graf <graf@amazon.com>,
	Hector Martin <marcan@marcan.st>,
	linux-kernel@vger.kernel.org
Subject: [RFC PATCH 1/3] iommu: Move IOMMU pagesize check to attach_device
Date: Fri,  6 Aug 2021 17:55:21 +0200	[thread overview]
Message-ID: <20210806155523.50429-2-sven@svenpeter.dev> (raw)
In-Reply-To: <20210806155523.50429-1-sven@svenpeter.dev>

The iova allocator is capable of handling any granularity which is a power
of two. Remove the much stronger condition that the granularity must be
smaller or equal to the CPU page size from a BUG_ON there.
Instead, check this condition during __iommu_attach_device and fail
gracefully.

Signed-off-by: Sven Peter <sven@svenpeter.dev>
---
 drivers/iommu/iommu.c | 34 +++++++++++++++++++++++++++++++---
 drivers/iommu/iova.c  |  7 ++++---
 2 files changed, 35 insertions(+), 6 deletions(-)

diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 1de503ddb343..5854a4ef5681 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -78,6 +78,8 @@ static struct iommu_domain *__iommu_domain_alloc(struct bus_type *bus,
 						 unsigned type);
 static int __iommu_attach_device(struct iommu_domain *domain,
 				 struct device *dev);
+static void __iommu_detach_device(struct iommu_domain *domain,
+				  struct device *dev);
 static int __iommu_attach_group(struct iommu_domain *domain,
 				struct iommu_group *group);
 static void __iommu_detach_group(struct iommu_domain *domain,
@@ -1964,6 +1966,18 @@ void iommu_domain_free(struct iommu_domain *domain)
 }
 EXPORT_SYMBOL_GPL(iommu_domain_free);
 
+static int iommu_check_page_size(struct iommu_domain *domain)
+{
+	if (!(domain->type & __IOMMU_DOMAIN_PAGING))
+		return 0;
+
+	if ((1 << __ffs(domain->pgsize_bitmap)) > PAGE_SIZE) {
+		pr_warn("IOMMU page size cannot represent CPU pages.\n");
+		return -EFAULT;
+	}
+
+	return 0;
+}
 static int __iommu_attach_device(struct iommu_domain *domain,
 				 struct device *dev)
 {
@@ -1973,9 +1987,23 @@ static int __iommu_attach_device(struct iommu_domain *domain,
 		return -ENODEV;
 
 	ret = domain->ops->attach_dev(domain, dev);
-	if (!ret)
-		trace_attach_device_to_domain(dev);
-	return ret;
+	if (ret)
+		return ret;
+
+	/*
+	 * Check that CPU pages can be represented by the IOVA granularity.
+	 * This has to be done after ops->attach_dev since many IOMMU drivers
+	 * only limit domain->pgsize_bitmap after having attached the first
+	 * device.
+	 */
+	ret = iommu_check_page_size(domain);
+	if (ret) {
+		__iommu_detach_device(domain, dev);
+		return ret;
+	}
+
+	trace_attach_device_to_domain(dev);
+	return 0;
 }
 
 int iommu_attach_device(struct iommu_domain *domain, struct device *dev)
diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index b6cf5f16123b..e0f8adde0f1b 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -50,10 +50,11 @@ init_iova_domain(struct iova_domain *iovad, unsigned long granule,
 {
 	/*
 	 * IOVA granularity will normally be equal to the smallest
-	 * supported IOMMU page size; both *must* be capable of
-	 * representing individual CPU pages exactly.
+	 * supported IOMMU page size; while both usually are capable of
+	 * representing individual CPU pages exactly the IOVA allocator
+	 * supports any granularities that are an exact power of two.
 	 */
-	BUG_ON((granule > PAGE_SIZE) || !is_power_of_2(granule));
+	BUG_ON(!is_power_of_2(granule));
 
 	spin_lock_init(&iovad->iova_rbtree_lock);
 	iovad->rbroot = RB_ROOT;
-- 
2.25.1


  reply	other threads:[~2021-08-06 15:56 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-06 15:55 [RFC PATCH 0/3] iommu/dma-iommu: Support IOMMU page size larger than the CPU page size Sven Peter
2021-08-06 15:55 ` Sven Peter [this message]
2021-08-06 15:55 ` [RFC PATCH 2/3] iommu/dma-iommu: Support iovad->granule > PAGE_SIZE Sven Peter
2021-08-06 18:04   ` Robin Murphy
2021-08-07  8:41     ` Sven Peter
2021-08-09 18:37       ` Robin Murphy
2021-08-09 19:57         ` Sven Peter
2021-08-07 11:47     ` Sven Peter
2021-08-09 17:41       ` Robin Murphy
2021-08-09 20:45         ` Sven Peter
2021-08-10  9:51           ` Robin Murphy
2021-08-11 20:18             ` Sven Peter
2021-08-12 12:43               ` Robin Murphy
2021-08-06 15:55 ` [RFC PATCH 3/3] iommu: Introduce __IOMMU_DOMAIN_LARGE_PAGES Sven Peter

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=20210806155523.50429-2-sven@svenpeter.dev \
    --to=sven@svenpeter.dev \
    --cc=arnd@kernel.org \
    --cc=graf@amazon.com \
    --cc=iommu@lists.linux-foundation.org \
    --cc=joro@8bytes.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=marcan@marcan.st \
    --cc=mohamed.mediouni@caramail.com \
    --cc=robin.murphy@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;
as well as URLs for NNTP newsgroup(s).