io-pgtable lock removal
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest0 of 8 patches 0 repliesNew
Patches in v1 No applies recorded yet
None of the 8 patches the cover letter announces are in this archive.
Reviews trailers per patch
No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.
Cover letter [PATCH 0/8] io-pgtable lock removal
Quoted 7 lines · Ray Jui
With the NVMf target test with 4 SSDs, fio based test, random read, 4k, 8 jobs:
Without IOMMU:
IOPS = 1080K
Quoted 7 lines
With IOMMU, but without your latest patch:
IOPS = 520K
With IOMMU and your latest patch:
IOPS = 500K ~ 850K (a lot of variation observed during the same test run)
That variation might come from the RB-Tree used in the IOVA allocator. For block-device workloads the allocation size of iova ranges might be bigger than what is cached in the magazines, so that the fall-back to the old (locked) allocator is used.
Joerg