Messages in this thread | | | Date | Thu, 8 Oct 1998 07:17:57 +0200 (CEST) | From | Rik van Riel <> | Subject | Re: [patch] oom-5 |
| |
On Wed, 7 Oct 1998, Andrea Arcangeli wrote: > On Wed, 7 Oct 1998, Rik van Riel wrote: > > _Assuming_ this is 100% right, what do you choice between these two > options? > > 1. deadlock > 2. fail allocating memory
I guess most people would choose 2, but please read on.
> >- set the settings for the {page,buffer}out-weight in > > /proc/sys/vm/swapctl low enough (something like 512 > > for your machine???) so that kswapd scans less than > > all memory on one try_to_free_page() > > That' s a completly bogus argument. The sysctl are avaible to make > the system working not to prove that the MM is buggy because doesn't > work with some malicious setting.
It's not a bogus argument. My point is that when the pageoutweight/num_physpages falls below a certain ratio, we might need to scan more often to find a freeable page.
You can get to that situation by either putting more memory in your box or lowering the weights. In both situations, the result will be the same.
That means that the stock kernel, with your patch, might fail terribly on a machine where the above is the case.
You can 'simulate' such a situation by setting the weights lower yourself -- then look at the numbers changing to something a lot different!
Rik. +-------------------------------------------------------------------+ | Linux memory management tour guide. H.H.vanRiel@phys.uu.nl | | Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ | +-------------------------------------------------------------------+
- 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/
| |