Date: Fri, 22 Jan 1999 11:53:25 -0800 (PST) From: Matthew Dillon <dillon@apollo.backplane.com> To: "John S. Dyson" <dyson@iquest.net> Cc: hackers@FreeBSD.ORG Subject: Re: Error in vm_fault change Message-ID: <199901221953.LAA56414@apollo.backplane.com> References: <199901221749.MAA03968@y.dyson.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Basically what it comes down to is that I do not think it is appropriate
for there to be hacks all around the kernel to arbitrarily block processes
in low memory situations. At the very worst, those same 'blockages' could
be implemented in one place - the memory allocator, and nowhere else. But
we can do much better.
I like your RLIMIT_RSS code, but it isn't enough and I think it is
implemented in the wrong place.
I believe the correct place to implement the 'memory scheduler' is
in vm_page_alloc() and vm_wait(). vm_page_alloc() uses some algorithm X
( which could simply be the RSS check ) to determine whether to return
NULL or not for VM_ALLOC_NORMAL or VM_ALLOC_ZERO requests.
vm_wait() ould implement the RSS enforcement as well as other
sophistication. For example, it could determine that the enforcement
is insufficient and start scaling it, or it could determine that
the system is thrashing and implement a more sophisticated swap/sleep
strategy.
We can play with the scheduling and enforcement algorithms much more
easily this way, too. What do you say ?
-Matt
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199901221953.LAA56414>