Messages in this thread | | | Date | Thu, 28 Apr 2011 08:59:17 +0900 | Subject | Re: [RFC 6/8] In order putback lru core | From | Minchan Kim <> |
| |
Hi Rik,
On Thu, Apr 28, 2011 at 8:46 AM, Rik van Riel <riel@redhat.com> wrote: > On 04/26/2011 12:25 PM, Minchan Kim wrote: > >> But this approach has a problem on contiguous pages. >> In this case, my idea can not work since friend pages are isolated, too. >> It means prev_page->next == next_page always is false and both pages are >> not >> LRU any more at that time. It's pointed out by Rik at LSF/MM summit. >> So for solving the problem, I can change the idea. >> I think we don't need both friend(prev, next) pages relation but >> just consider either prev or next page that it is still same LRU. > >> Any comment? > > If the friend pages are isolated too, then your condition > "either prev or next page that it is still same LRU" is > likely to be false, no?
H - P1 - P2 - P3 - P4 - P5 - P6 - P7 - P8 - P9 - P10 - T
assume : we isolate pages P3~P7 and we consider only next pointer.
H - P1 - P2 - P8 - P9 - P10 - T
If we start to putback P7 as starting point, next P8 is valid so,
H - P1 - P2 - P7 - P8 - P9 - P10 - T Then, if we consider P6, next P7 is valid, too. So,
H - P1 - P2 - P6 - P7 - P8 - P9 - P10 - T
continue until P3.
-- Kind regards, Minchan Kim
| |