alt.hn

3/18/2026 at 12:28:44 PM

Linux Page Faults, MMAP, and userfaultfd for faster VM boots

https://www.shayon.dev/post/2026/65/linux-page-faults-mmap-and-userfaultfd/

by shayonj

3/20/2026 at 7:59:11 AM

Oh neat, a post I actually know something about! I worked a lot on userfaultfd performance for GCE's live migration post-copy a couple years ago. Or more specifically, I worked on mechanisms to avoid it entirely- due to lock contention in the kennel, faults become veeeerry slow as the number of vcpus scales, and as it happens VMs these days can have a lot of vcpus

by anlsh

3/20/2026 at 11:29:14 AM

that's very interesting! I was noticing page vault storm on live migrations as well and I wonder if that's what you were running into / mentioning here regarding the lock contention

by shayonj

3/20/2026 at 11:42:23 AM

[dead]

by samsudin

3/20/2026 at 8:23:13 AM

> Userfaultfd is a Linux mechanism, available since kernel 4.3, with additional event features like non-cooperative mode and fork/remap/remove tracking added in 4.11, that lets a userspace thread intercept and handle page faults.

Is this the same feature Windows has had forever, or is there more to it?

by dataflow