io_uring, libaio performance across Linux kernels and an unexpected IOMMU trap

Re:io_uring, libaio performance across Linux kernels and an ... by eivanov89 on Tuesday March 24, 2026 @09:28AM (#47502288)

Dear folks, I'm the author of that post.

A short summary below.

We ran fio benchmarks comparing libaio and io_uring across kernels (5.4 -> 7.0-rc3). The most surprising part wasn’t io_uring gains (~2x), but a ~30% regression caused by IOMMU being enabled by default between releases.

Happy to share more details about setup or reproduce results.

[ Parent | on HN ]

Re:io_uring, libaio performance across Linux kernels and an ... by jcalvinowens on Tuesday March 24, 2026 @12:48PM (#47505564)

Thanks for sharing this.

Was the iommu using strict or lazy invalidation? I think lazy is the default but I'm not sure how long that's been true.

[ Parent | on HN ]

Re:io_uring, libaio performance across Linux kernels and an ... by eivanov89 on Tuesday March 24, 2026 @01:00PM (#47505772)

We compared IOMMU fully disabled vs enabled. When it is enabled, I expect it to be lazy (should be the default for IOMMU). Note, that we recommend to use passthrough to completely bypass translation for most devices independent on strict/lazy mode.

[ Parent | on HN ]