Re: Stability (2.2.14/15/16/17pre1)

From: Marcelo Tosatti (marcelo@conectiva.com.br)
Date: Wed Jun 14 2000 - 11:53:15 EST


On Wed, 14 Jun 2000, Andrea Arcangeli wrote:

> On Wed, 14 Jun 2000, Marcelo Tosatti wrote:
>
> >
> >
> >On Wed, 14 Jun 2000, Andrea Arcangeli wrote:
> >
> >> On Wed, 14 Jun 2000, Marcelo Tosatti wrote:
> >>
> >> >2.2.15, mm/vmscan.c, kswapd thread:
> >> >
> >> > while (1) {
> >> > interruptible_sleep_on(&kswapd_wait);
> >> > while (nr_free_pages < freepages.high)
> >> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> >> > {
> >> > if (do_try_to_free_pages(GFP_KSWAPD))
> >> > {
> >> > if (tsk->need_resched)
> >> > schedule();
> >> > continue;
> >> > }
> >> > tsk->state = TASK_INTERRUPTIBLE;
> >> > schedule_timeout(10*HZ);
> >> > }
> >> > }
> >> >
> >> >What i'm missing here ?
> >>
> >> linux-2.2.15/mm/vmscan.c line 383 or linux-2.2.16/mm/vmscan.c line 390.
> >>
> >> >mmap002 on 2.2.15 gets killed.
> >>> mmap002 on 2.2.15 + 2.2.16's thrasing heuristic runs fine. Try it.
> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> >
> >Andrea, _please_ read what i wrote correctly. Get 2.2.15, apply
> >thrasing heuristic changes, and run mmap002.
>
> Mind to send me the patch (I only have whole 2.2.16 global)? If with
> removing trashing_mem you also mean inserting the atomic_t
> free_before_allocate then it's another issue
With 'trashing heuristic' changes i also mean free_before_allocate.
Sorry about not being clear about that.

> and you shouldn't really blame trashing_mem as you're doing.

*I'm not blaming trashing_mem because of the instability issue*, as I
wrote down
(kernelnotes.org/lnxlists/linux-kernel/lk_0006_02/msg01247.html),
and you seem to be agreeing on this part.
What i'm blaming is 2.2.15 VM.

> I can't see how trashing_mem can be related to any stability issue. Please
> explain that if you can.
>
> Could you also try again 2.2.15 and increase freepages.min? The
> free_before_allocate addition just takes the nr_free_pages higher.

-
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/



This archive was generated by hypermail 2b29 : Thu Jun 15 2000 - 21:00:34 EST