Re: [PATCH RFC] mm/page_idle: simple idle page tracking for virtual memory

2019-07-23 Thread Konstantin Khlebnikov
On 23.07.2019 17:25, Joel Fernandes wrote: On Tue, Jul 23, 2019 at 04:59:07PM +0300, Konstantin Khlebnikov wrote: On 23.07.2019 16:46, Joel Fernandes wrote: On Tue, Jul 23, 2019 at 02:54:26PM +0300, Konstantin Khlebnikov wrote: The page_idle tracking feature currently requires looking up

Re: [PATCH RFC] mm/page_idle: simple idle page tracking for virtual memory

2019-07-23 Thread Joel Fernandes
On Tue, Jul 23, 2019 at 04:59:07PM +0300, Konstantin Khlebnikov wrote: > > > On 23.07.2019 16:46, Joel Fernandes wrote: > > On Tue, Jul 23, 2019 at 02:54:26PM +0300, Konstantin Khlebnikov wrote: > > > The page_idle tracking feature currently requires looking up the pagemap > > > for a process

Re: [PATCH RFC] mm/page_idle: simple idle page tracking for virtual memory

2019-07-23 Thread Konstantin Khlebnikov
On 23.07.2019 16:46, Joel Fernandes wrote: On Tue, Jul 23, 2019 at 02:54:26PM +0300, Konstantin Khlebnikov wrote: The page_idle tracking feature currently requires looking up the pagemap for a process followed by interacting with /sys/kernel/mm/page_idle. This is quite cumbersome and can be

Re: [PATCH RFC] mm/page_idle: simple idle page tracking for virtual memory

2019-07-23 Thread Joel Fernandes
On Tue, Jul 23, 2019 at 02:54:26PM +0300, Konstantin Khlebnikov wrote: > The page_idle tracking feature currently requires looking up the pagemap > for a process followed by interacting with /sys/kernel/mm/page_idle. > This is quite cumbersome and can be error-prone too. If between > accessing the

[PATCH RFC] mm/page_idle: simple idle page tracking for virtual memory

2019-07-23 Thread Konstantin Khlebnikov
The page_idle tracking feature currently requires looking up the pagemap for a process followed by interacting with /sys/kernel/mm/page_idle. This is quite cumbersome and can be error-prone too. If between accessing the per-PID pagemap and the global page_idle bitmap, if something changes with the