iommu/vt-d: Optimize the use of locks
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v4 Latest11 patches 2 repliesIn review
- v311 patches 12 repliesSuperseded
- v212 patches 18 repliesSuperseded
Earlier versions · 1
| Version | Posted | Patches | Replies | Status |
|---|---|---|---|---|
| v1 | 12 patches | 42 replies | Superseded |
Changes in v4 from the cover letter
- Kevin pointed out that the copy table failure path needs some improvements. We agreed that it worth a separated series to handle all improvements on copy table.
- Kevin suggested that we could put all spinlock replacement changes in a single patch. Updated in this series.
- Others look good to Kevin. (Thanks and very appreciated for all the review comments.)
Patches in v4 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/11 | iommu/vt-d: debugfs: Remove device_domain_lock usagedrivers/ drivers/ drivers/ | +32 | 1 reply | |
| 2/11 | iommu/vt-d: Remove clearing translation data in disable_dmar_iommu()drivers/ drivers/ | +8 | 1 reply | |
| 3/11 | iommu/vt-d: Use pci_get_domain_bus_and_slot() in pgtable_walk()drivers/ drivers/ | +6 | 0 replies | |
| 4/11 | iommu/vt-d: Unnecessary spinlock for root table alloc and freedrivers/ | +5 | 0 replies | |
| 5/11 | iommu/vt-d: Replace spin_lock_irqsave() with spin_lock()drivers/ drivers/ drivers/ | +32 | 0 replies | |
| 6/11 | iommu/vt-d: Acquiring lock in domain ID allocation helpersdrivers/ | +9 | 0 replies | |
| 7/11 | iommu/vt-d: Acquiring lock in pasid manipulation helpersdrivers/ drivers/ drivers/ | +55 | 0 replies | |
| 8/11 | iommu/vt-d: Check device list of domain in domain free pathdrivers/ | +3 | 0 replies | |
| 9/11 | iommu/vt-d: Fold __dmar_remove_one_dev_info() into its callerdrivers/ | +7 | 0 replies | |
| 10/11 | iommu/vt-d: Use device_domain_lock accuratelydrivers/ | +19 | 0 replies | |
| 11/11 | iommu/vt-d: Convert global spinlock into per domain lockdrivers/ drivers/ | +22 | 0 replies |
Reviews trailers per patch
All 11 patches have a Reviewed-by or Acked-by.
| Patch | Kevin Tian | Jason Gunthorpe |
|---|---|---|
| 1/11 | R-b | |
| 2/11 | R-b | |
| 3/11 | R-b | R-b |
| 4/11 | R-b | |
| 5/11 | R-b | |
| 6/11 | R-b | |
| 7/11 | R-b | |
| 8/11 | R-b | R-b |
| 9/11 | R-b | |
| 10/11 | R-b | |
| 11/11 | R-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH v4 00/11] iommu/vt-d: Optimize the use of locks
Hi folks,
This series tries to optimize the uses of two locks in the Intel IOMMU driver:
- The intel_iommu::lock is used to protect the IOMMU resources shared by devices. They include the IOMMU root and context tables, the pasid tables and the domain IDs.
- The global device_domain_lock is used to protect the global and the per-domain device tracking lists.
The optimization includes:
- Remove the unnecessary global device tracking list;
- Remove unnecessary locking;
- Reduce the scope of the lock as much as possible, that is, use the lock only where necessary;
- The global lock is transformed into a local lock to improve the efficiency.
This series is also available on github:
https://github.com/LuBaolu/intel-iommu/commits/intel-iommu-lock-optimization-v4
Your comments and suggestions are very appreciated.
Best regards,
baolu
Change log:
v4:
- Kevin pointed out that the copy table failure path needs some improvements. We agreed that it worth a separated series to handle all improvements on copy table.
- Kevin suggested that we could put all spinlock replacement changes in a single patch. Updated in this series.
- Others look good to Kevin. (Thanks and very appreciated for all the review comments.)
v3:
- https://lore.kernel.org/linux-iommu/[email protected]/
- Split reduction of lock ranges from changing irqsave.
https://lore.kernel.org/linux-iommu/BN9PR11MB52760A3D7C6BF1AF9C9D34658CAA9@BN9PR11MB5276.namprd11.prod.outlook.com/
- Fully initialize the dev_info before adding it to the list.
https://lore.kernel.org/linux-iommu/BN9PR11MB52764D7CD86448C5E4EB46668CAA9@BN9PR11MB5276.namprd11.prod.outlook.com/
- Various code and comments refinement.
v2:
- https://lore.kernel.org/linux-iommu/[email protected]/
- Split the lock-free page walk issue into a new patch:
https://lore.kernel.org/linux-iommu/[email protected]/
- Drop the conversion from spinlock to mutex and make this series
cleanup purpose only.
- Address several comments received during v1 review.
v1:
- https://lore.kernel.org/linux-iommu/[email protected]/
- Initial post.
Lu Baolu (11):
iommu/vt-d: debugfs: Remove device_domain_lock usage iommu/vt-d: Remove clearing translation data in disable_dmar_iommu() iommu/vt-d: Use pci_get_domain_bus_and_slot() in pgtable_walk() iommu/vt-d: Unnecessary spinlock for root table alloc and free iommu/vt-d: Replace spin_lock_irqsave() with spin_lock() iommu/vt-d: Acquiring lock in domain ID allocation helpers iommu/vt-d: Acquiring lock in pasid manipulation helpers iommu/vt-d: Check device list of domain in domain free path iommu/vt-d: Fold __dmar_remove_one_dev_info() into its caller iommu/vt-d: Use device_domain_lock accurately
iommu/vt-d: Convert global spinlock into per domain lock
drivers/iommu/intel/iommu.h | 3 +- drivers/iommu/intel/pasid.h | 1 + drivers/iommu/intel/debugfs.c | 49 ++++--- drivers/iommu/intel/iommu.c | 249 ++++++++++------------------------ drivers/iommu/intel/pasid.c | 103 +++++++------- drivers/iommu/intel/svm.c | 5 +- 6 files changed, 163 insertions(+), 247 deletions(-)
Signature · 1 line
2.25.1