Re: [RFC v2 0/2] vmevent: A bit reworked pressure attribute + docs + man page

2012-10-26 Thread Anton Vorontsov
On Fri, Oct 26, 2012 at 11:37:20AM +0900, Minchan Kim wrote: [...] > > > Of course, it's very flexible and potential to add new VM knob easily but > > > the thing we is about to use now is only VMEVENT_ATTR_PRESSURE. > > > Is there any other use cases for swap or free? or potential user? > > > > N

Re: [RFC v2 0/2] vmevent: A bit reworked pressure attribute + docs + man page

2012-10-25 Thread Minchan Kim
On Thu, Oct 25, 2012 at 02:08:14AM -0700, Anton Vorontsov wrote: > Hello Minchan, > > Thanks a lot for the email! > > On Thu, Oct 25, 2012 at 03:40:09PM +0900, Minchan Kim wrote: > [...] > > > What applications (well, activity managers) are really interested in is > > > this: > > > > > > 1. Do w

Re: [RFC v2 0/2] vmevent: A bit reworked pressure attribute + docs + man page

2012-10-25 Thread Anton Vorontsov
On Thu, Oct 25, 2012 at 02:08:14AM -0700, Anton Vorontsov wrote: [...] > Maybe it makes sense to implement something like PRESSURE_MILD with an > additional nr_pages threshold, which basically hits the kernel about how > many easily reclaimable pages userland has (that would be a part of our > defi

Re: [RFC v2 0/2] vmevent: A bit reworked pressure attribute + docs + man page

2012-10-25 Thread Anton Vorontsov
Hello Minchan, Thanks a lot for the email! On Thu, Oct 25, 2012 at 03:40:09PM +0900, Minchan Kim wrote: [...] > > What applications (well, activity managers) are really interested in is > > this: > > > > 1. Do we we sacrifice resources for new memory allocations (e.g. files > >cache)? > > 2.

Re: [RFC v2 0/2] vmevent: A bit reworked pressure attribute + docs + man page

2012-10-25 Thread Minchan Kim
Hi Pekka, On Thu, Oct 25, 2012 at 09:44:52AM +0300, Pekka Enberg wrote: > On Thu, Oct 25, 2012 at 9:40 AM, Minchan Kim wrote: > > Your description doesn't include why we need new vmevent_fd(2). > > Of course, it's very flexible and potential to add new VM knob easily but > > the thing we is about

Re: [RFC v2 0/2] vmevent: A bit reworked pressure attribute + docs + man page

2012-10-24 Thread Pekka Enberg
On Thu, Oct 25, 2012 at 9:40 AM, Minchan Kim wrote: > Your description doesn't include why we need new vmevent_fd(2). > Of course, it's very flexible and potential to add new VM knob easily but > the thing we is about to use now is only VMEVENT_ATTR_PRESSURE. > Is there any other use cases for swa

Re: [RFC v2 0/2] vmevent: A bit reworked pressure attribute + docs + man page

2012-10-24 Thread Minchan Kim
Hi Anton, On Mon, Oct 22, 2012 at 04:19:28AM -0700, Anton Vorontsov wrote: > Hi all, > > So this is the second RFC. The main change is that I decided to go with > discrete levels of the pressure. I am very happy with that because I already have yelled it several time. > > When I started writin

[RFC v2 0/2] vmevent: A bit reworked pressure attribute + docs + man page

2012-10-22 Thread Anton Vorontsov
Hi all, So this is the second RFC. The main change is that I decided to go with discrete levels of the pressure. When I started writing the man page, I had to describe the 'reclaimer inefficiency index', and while doing this I realized that I'm describing how the kernel is doing the memory manage