[GIT PULL] IOMMU Fixes for Linux v7.2-rc5
From: Will Deacon
Date: Fri Jul 24 2026 - 12:47:09 EST
Hi Linus,
Joerg's away at the moment so I've been looking after the IOMMU tree in
his absence. In the process of doing that, I've hoovered up a handful
of fixes for the AMD and Intel drivers which address a combination of
the usual out-of-bounds/locking/leak bugs as well as some logical issues
around SVA and command completion.
Summary in the tag; please pull.
Cheers,
Will
--->8
The following changes since commit 8cdeaa50eae8dad34885515f62559ee83e7e8dda:
Linux 7.2-rc2 (2026-07-05 14:44:06 -1000)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/iommu/linux.git tags/iommu-fixes-v7.2-rc5
for you to fetch changes up to 754f8efe45f87e3a9c6871b645b2f9d46d1b407b:
iommu/intel: Fix out-of-bounds memset in dmar_latency_disable() (2026-07-21 11:34:51 +0000)
----------------------------------------------------------------
IOMMU Fixes for Linux v7.2-rc5
- AMD:
* Fix lockdep splat from nested domain allocation
* Fix nested domain leak
* Fix broken synchronisation of command completion
* Fix OOB write in "ivrs_acpihid" command-line parsing
- VT-d:
* Prevent SVA for IOMMUs with non-coherent page-table walker
* Fix OOB write in PMU driver.
----------------------------------------------------------------
Guanghui Feng (1):
iommu/amd: Wait for completion instead of returning early in iommu_completion_wait()
Li RongQing (1):
iommu/intel: Fix out-of-bounds memset in dmar_latency_disable()
Lu Baolu (1):
iommu/vt-d: Disallow SVA if page walk is not coherent
Pengpeng Hou (1):
iommu/amd: Bound the early ACPI HID map
Tycho Andersen (AMD) (2):
iommu/amd: Fix IRQ unsafe locking in gdom allocation
iommu/amd: Fix nested domain leak
drivers/iommu/amd/init.c | 6 ++++++
drivers/iommu/amd/iommu.c | 22 ++++++++++++++++------
drivers/iommu/amd/nested.c | 34 +++++++++++++++++++---------------
drivers/iommu/intel/perf.c | 2 +-
drivers/iommu/intel/svm.c | 2 +-
5 files changed, 43 insertions(+), 23 deletions(-)