linux-kselftest.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jason Gunthorpe <jgg@nvidia.com>
To: iommu@lists.linux.dev, Joerg Roedel <joro@8bytes.org>,
	Kevin Tian <kevin.tian@intel.com>,
	linux-kselftest@vger.kernel.org,
	Robin Murphy <robin.murphy@arm.com>,
	Shuah Khan <shuah@kernel.org>, Will Deacon <will@kernel.org>
Cc: Lixiao Yang <lixiao.yang@intel.com>,
	Matthew Rosato <mjrosato@linux.ibm.com>,
	Nicolin Chen <nicolinc@nvidia.com>,
	patches@lists.linux.dev, stable@vger.kernel.org,
	syzbot+c2f65e2801743ca64e08@syzkaller.appspotmail.com,
	Yi Liu <yi.l.liu@intel.com>
Subject: [PATCH 1/2] iommufd: Prevent ALIGN() overflow
Date: Thu, 17 Jul 2025 16:15:08 -0300	[thread overview]
Message-ID: <1-v1-7b4a16fc390b+10f4-iommufd_alloc_overflow_jgg@nvidia.com> (raw)
In-Reply-To: <0-v1-7b4a16fc390b+10f4-iommufd_alloc_overflow_jgg@nvidia.com>

When allocating IOVA the candidate range gets aligned to the target
alignment. If the range is close to ULONG_MAX then the ALIGN() can
wrap resulting in a corrupted iova.

Open code the ALIGN() using get_add_overflow() to prevent this.
This simplifies the checks as we don't need to check for length earlier
either.

Consolidate the two copies of this code under a single helper.

This bug would allow userspace to create a mapping that overlaps with some
other mapping or a reserved range.

Cc: stable@vger.kernel.org
Fixes: 51fe6141f0f6 ("iommufd: Data structure to provide IOVA to PFN mapping")
Reported-by: syzbot+c2f65e2801743ca64e08@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/r/685af644.a00a0220.2e5631.0094.GAE@google.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
---
 drivers/iommu/iommufd/io_pagetable.c | 41 +++++++++++++++++-----------
 1 file changed, 25 insertions(+), 16 deletions(-)

diff --git a/drivers/iommu/iommufd/io_pagetable.c b/drivers/iommu/iommufd/io_pagetable.c
index abf4aadca96c0b..c0360c450880b8 100644
--- a/drivers/iommu/iommufd/io_pagetable.c
+++ b/drivers/iommu/iommufd/io_pagetable.c
@@ -70,20 +70,34 @@ struct iopt_area *iopt_area_contig_next(struct iopt_area_contig_iter *iter)
 	return iter->area;
 }
 
+static bool __alloc_iova_check_range(unsigned long *start, unsigned long last,
+				     unsigned long length,
+				     unsigned long iova_alignment,
+				     unsigned long page_offset)
+{
+	unsigned long aligned_start;
+
+	/* ALIGN_UP() */
+	if (check_add_overflow(*start, iova_alignment - 1, &aligned_start))
+		return false;
+	aligned_start &= ~(iova_alignment - 1);
+	aligned_start |= page_offset;
+
+	if (aligned_start >= last || last - aligned_start < length - 1)
+		return false;
+	*start = aligned_start;
+	return true;
+}
+
 static bool __alloc_iova_check_hole(struct interval_tree_double_span_iter *span,
 				    unsigned long length,
 				    unsigned long iova_alignment,
 				    unsigned long page_offset)
 {
-	if (span->is_used || span->last_hole - span->start_hole < length - 1)
+	if (span->is_used)
 		return false;
-
-	span->start_hole = ALIGN(span->start_hole, iova_alignment) |
-			   page_offset;
-	if (span->start_hole > span->last_hole ||
-	    span->last_hole - span->start_hole < length - 1)
-		return false;
-	return true;
+	return __alloc_iova_check_range(&span->start_hole, span->last_hole,
+					length, iova_alignment, page_offset);
 }
 
 static bool __alloc_iova_check_used(struct interval_tree_span_iter *span,
@@ -91,15 +105,10 @@ static bool __alloc_iova_check_used(struct interval_tree_span_iter *span,
 				    unsigned long iova_alignment,
 				    unsigned long page_offset)
 {
-	if (span->is_hole || span->last_used - span->start_used < length - 1)
+	if (span->is_hole)
 		return false;
-
-	span->start_used = ALIGN(span->start_used, iova_alignment) |
-			   page_offset;
-	if (span->start_used > span->last_used ||
-	    span->last_used - span->start_used < length - 1)
-		return false;
-	return true;
+	return __alloc_iova_check_range(&span->start_used, span->last_used,
+					length, iova_alignment, page_offset);
 }
 
 /*
-- 
2.43.0


  reply	other threads:[~2025-07-17 19:15 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-17 19:15 [PATCH 0/2] Fix undetected overflow when allocating IOVA Jason Gunthorpe
2025-07-17 19:15 ` Jason Gunthorpe [this message]
2025-07-18  8:16   ` [PATCH 1/2] iommufd: Prevent ALIGN() overflow Nicolin Chen
2025-07-18 13:03   ` Yi Liu
2025-07-17 19:15 ` [PATCH 2/2] iommufd/selftest: Test reserved regions near ULONG_MAX Jason Gunthorpe
2025-07-18  2:45   ` Nicolin Chen
2025-07-18  8:13     ` Nicolin Chen
2025-07-18 16:21       ` Jason Gunthorpe
2025-07-18 18:23       ` Robin Murphy
2025-07-18 18:50         ` Nicolin Chen
2025-07-18 20:16           ` Jason Gunthorpe
2025-07-18 19:56         ` Jason Gunthorpe
2025-07-18 13:03   ` Yi Liu
2025-07-18 18:10 ` [PATCH 0/2] Fix undetected overflow when allocating IOVA Nicolin Chen

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=1-v1-7b4a16fc390b+10f4-iommufd_alloc_overflow_jgg@nvidia.com \
    --to=jgg@nvidia.com \
    --cc=iommu@lists.linux.dev \
    --cc=joro@8bytes.org \
    --cc=kevin.tian@intel.com \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=lixiao.yang@intel.com \
    --cc=mjrosato@linux.ibm.com \
    --cc=nicolinc@nvidia.com \
    --cc=patches@lists.linux.dev \
    --cc=robin.murphy@arm.com \
    --cc=shuah@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=syzbot+c2f65e2801743ca64e08@syzkaller.appspotmail.com \
    --cc=will@kernel.org \
    --cc=yi.l.liu@intel.com \
    /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).