>>>>> On Sun, 6 Mar 2005 22:34:52 -0500,
	Thor Lancelot Simon <tls@rek.tjls.com> said:


>> BTW, I think it is better to decrease current lower bound of
>> vm.bufcache from 5% to something like 1% (Note that I'm not talking
>> about default value but lower bound value here). Because it's often
>> desirable to use such small value on machines like pc532 which have
>> smaller RAM.

> I don't like this idea.  There are common use cases already (e.g.
> building the entire system on a large-memory machine) where the file
> cache steals too many pages from the buffer cache as it is: things can
> be measurably sped up by _increasing_ the lower bound of vm.bufcache.
> If we're going to decrease vm.bufcache we need to work harder to make
> sure it doesn't just get pinned at that new lower bound inappropriately
> as it currently does.

Hm, I guess my description was ambiguous.

What I meant in "lower bound" is not vm.bufmem_lowater, but to change
the following code fragment in kern/vfs_bio.c:
		if (bufcache < 5) {
			printf("forcing bufcache %d -> 5", bufcache);
			bufcache = 5;
		}
to:
		if (bufcache < 1) {
			printf("forcing bufcache %d -> 1", bufcache);
			bufcache = 1;
		}
i.e. I don't intend to change vm.bufmem_lowater.

Do you really dislike this change?
--
soda