Messages in this thread | | | Date | Tue, 10 Nov 1998 16:45:38 +0100 | From | Andi Kleen <> | Subject | Re: 2.1.127 BUGS was Re: Linux-2.1.127 |
| |
On Tue, Nov 10, 1998 at 04:20:57PM +0100, Stephen C. Tweedie wrote: > Hi, > > On Tue, 10 Nov 1998 15:37:19 +0100, Andi Kleen <ak@muc.de> said: > > > I think the reason why Linus' fix doesn't help completely is that swap_out() > > has its own loop that frees a lot of pages according to the priority. Maybe > > the nr_free_pages > 2*freepages.high exit condition should be added there > > too. > > No: swap_out() exits as soon as it has found one page to free. The loop > only runs while it is searching for that one page.
Ok, never mind then. Sorry for spreading misinformation.
But I wonder then why Linus' patch doesn't prevent the killing of the complete page cache then.
-Andi
P.S.: while reading vmscan.c: would it make sense to add the following into the inner loop of swap_out (assuming maxover is 0 initially)?
over = (p->rlim[RLIMIT_RSS].rlim_cur>>PAGE_SHIFT) - p->mm->rss; if (over < maxover) { pbest = p; maxover = over; continue; }
With that RLIMIT_RSS would at least make some sense.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |