Concurrent page-fault handling with per-VMA locks
Concurrent page-fault handling with per-VMA locks
Posted Sep 16, 2022 2:15 UTC (Fri) by kiryl (subscriber, #41516)Parent article: Concurrent page-fault handling with per-VMA locks
Do Emacs and gnome-shell suffer from scalability issues? If so, I believe the problem is on the other side of syscall.
Workloads that require extreme scaling (databases, VMMs, etc) tend to have one or a few very hot VMAs and per-VMA locking would not bring any help.
