Re: [PATCH 40/41] mm: separate vma->lock from vm_area_struct

2023-01-17 Thread Suren Baghdasaryan
On Tue, Jan 17, 2023 at 10:34 AM Jann Horn wrote: > > On Mon, Jan 9, 2023 at 9:55 PM Suren Baghdasaryan wrote: > > vma->lock being part of the vm_area_struct causes performance regression > > during page faults because during contention its count and owner fields > > are constantly updated and ha

Re: [PATCH 40/41] mm: separate vma->lock from vm_area_struct

2023-01-17 Thread Jann Horn
On Mon, Jan 9, 2023 at 9:55 PM Suren Baghdasaryan wrote: > vma->lock being part of the vm_area_struct causes performance regression > during page faults because during contention its count and owner fields > are constantly updated and having other parts of vm_area_struct used > during page fault h

[PATCH 40/41] mm: separate vma->lock from vm_area_struct

2023-01-09 Thread Suren Baghdasaryan
vma->lock being part of the vm_area_struct causes performance regression during page faults because during contention its count and owner fields are constantly updated and having other parts of vm_area_struct used during page fault handling next to them causes constant cache line bouncing. Fix that