[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20230322064956.263419-1-baolu.lu@linux.intel.com>
Date: Wed, 22 Mar 2023 14:49:50 +0800
From: Lu Baolu <baolu.lu@...ux.intel.com>
To: Joerg Roedel <joro@...tes.org>
Cc: Jason Gunthorpe <jgg@...dia.com>,
Robin Murphy <robin.murphy@....com>,
Christoph Hellwig <hch@...radead.org>,
Kevin Tian <kevin.tian@...el.com>,
Will Deacon <will@...nel.org>, iommu@...ts.linux.dev,
linux-kernel@...r.kernel.org, Lu Baolu <baolu.lu@...ux.intel.com>
Subject: [PATCH v4 0/6] iommu: Extend changing default domain to normal group
The IOMMU group sysfs interface allows users to change the default
domain of a group. The current implementation uses device_lock() to make
sure that the devices in the group are not bound to any driver and won't
be bound during the process of changing the default domain. In order to
avoid a possible deadlock caused by lock order of device_lock and
group->mutex, it limits the functionality to singleton groups only.
The recently implemented DMA ownership framework can be applied here to
replace device_lock(). In addition, use group->mutex to ensure that the
iommu ops of the device are always valid during the process of changing
default domain.
With above replacement and enhancement, the device_lock() could be
removed and the singleton-group-only limitation could be removed.
This series is based on v6.3-rc3, and I also have it on github:
https://github.com/LuBaolu/intel-iommu/commits/iommu-sysfs-default-domain-extension-v4
Change log:
v4:
- Whole series reviewed by Jason.
- Robin suggested further simplification of using
arm_iommu_release_mapping() directly.
https://lore.kernel.org/linux-iommu/7b248ba1-3967-5cd8-82e9-0268c706d320@arm.com/
- Rebase the whole series to v6.3-rc3.
v3:
- https://lore.kernel.org/linux-iommu/20230306025804.13912-1-baolu.lu@linux.intel.com/
- "arm_iommu_detach_device() is a noop" is not entirely right. It is
used to make the iommu driver stop using the domain that it is
about to free. It cannot be a NOP or it is a UAF. [Jason]
- Use Jason's new arm_iommu_release_device() proposal instead.
v2:
- https://lore.kernel.org/linux-iommu/20230217094736.159005-1-baolu.lu@linux.intel.com/
- Use group->mutex instead of an additional rw lock.
v1: initial post
- https://lore.kernel.org/linux-iommu/20230213074941.919324-1-baolu.lu@linux.intel.com/
Lu Baolu (6):
iommu/ipmmu-vmsa: Call arm_iommu_release_mapping() in release path
iommu: Split iommu_group_remove_device() into helpers
iommu: Same critical region for device release and removal
iommu: Move lock from iommu_change_dev_def_domain() to its caller
iommu: Replace device_lock() with group->mutex
iommu: Cleanup iommu_change_dev_def_domain()
drivers/iommu/iommu.c | 273 ++++++++----------
drivers/iommu/ipmmu-vmsa.c | 14 +-
.../ABI/testing/sysfs-kernel-iommu_groups | 1 -
3 files changed, 130 insertions(+), 158 deletions(-)
--
2.34.1
Powered by blists - more mailing lists