Re: Out Of Memory in v. 2.1
david parsons (o.r.c@p.e.l.l.p.o.r.t.l.a.n.d.o.r.u.s)
4 Oct 1998 22:00:19 -0700
In article <linux.kernel.199810050305.XAA27278@moisil.cs.columbia.edu>,
Ion Badulescu <linux-kernel@vger.rutgers.edu> wrote:
>On 4 Oct 1998 15:05:32 -0700, david parsons wrote:
>> In article <linux.kernel.m0zPwOX-000aNdC@the-village.bc.nu>,
>> Alan Cox <linux-kernel@vger.rutgers.edu> wrote:
>> [On overcommit]
>>
>> [someone else wrote]
>>
>>>> So, for a solution, let's do what all programs know how to deal with. Return
>>>> nulls instead of overallocating. No processes need to get killed, and the
>>>> system won't get deadlocked by overallocation.
>>
>>>And to boot a machine with 512Mb of RAM you'll need at least 600Mb or
>>>so of swap file.
>>
>> That's one thing I've always wondered about. Why not have the system
>> keep track of total memory instead of wanting to mirror memory onto
>> swap? It seems to me that if you have 512mb of core and 128mb of
>> swap that you've got 640mb of memory and you should be able to do
>> allocations against that, since either a page will be in core or
>> on the swap device.
>
>Or on both.
That's a performance optimisation, though, which you can blow away
as the system fills up. Yes, things will run slower when memory
gets congested, but they already do, albeit in a fairly steep "oh,
no more memory so we'll just lock the machine up as we scramble for
free pages" manner.
____
david parsons \bi/ Though I've not played around with recent 2.1.1xx
\/ kernels, so I'm not confident of recent performance.
-
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/