This is a follow-up to the discussion thread here:

https://lore.kernel.org/linux-iommu/20240201193427.GQ50608@ziepe.ca/

This proposes a per-IOMMU red-black tree for iommu probed devices. It
benefits fault handling paths, where the VT-d hardware reports the
Source ID of the related device, and software needs to retrieve the
corresponding device pointer based on this ID.

Lu Baolu (2):
  iommu/vt-d: Use rbtree to track iommu probed devices
  iommu/vt-d: Use device rbtree in iopf reporting path

 drivers/iommu/intel/iommu.h |  8 ++++
 drivers/iommu/intel/dmar.c  |  3 +-
 drivers/iommu/intel/iommu.c | 91 ++++++++++++++++++++++++++++++++++++-
 drivers/iommu/intel/svm.c   | 14 +++---
 4 files changed, 105 insertions(+), 11 deletions(-)

-- 
2.34.1