Re: [2.1.130-3] Page cache DEFINATELY too persistant... feature?

Benjamin Redelings I (bredelin@ucsd.edu)
Fri, 27 Nov 1998 10:37:30 -0800

"Stephen C. Tweedie" wrote:
>
> Hi,
>
> Looks like I have a handle on what's wrong with the 2.1.130 vm (in
> particular, its tendency to cache too much at the expense of
> swapping).
>
Looks like you do :) I ran the command
"find /usr/src/linux -name '*.c' | xargs grep shrink_mmap" in 2
different scenarios, so I have some 'before'/'after' results from the
free command.

The BEFORE is when I applied your patch by hand, and accidentally did
'count_max = limit<<1'. In this case, after I ran the command, free
looked like THIS:

telomere:~> cat freesample
total used free shared buffers
cached
Mem: 63472 61780 1692 2428 3336
54484
-/+ buffers/cache: 3960 59512
Swap: 34236 18136 16100

For the AFTER scenario, I then fixed my mistake, and voila!, we get
this:
telomere:~> cat freesample2
total used free shared buffers
cached
Mem: 63472 61508 1964 31228 3336
39308
-/+ buffers/cache: 18864 44608
Swap: 34236 128 34108

Nice work :)
-benRI

>
> ----------------------------------------------------------------
> --- mm/filemap.c.~1~ Thu Nov 26 18:48:52 1998
> +++ mm/filemap.c Fri Nov 27 12:45:03 1998
> @@ -200,8 +200,8 @@
> struct page * page;
> int count_max, count_min;
>
> - count_max = (limit<<1) >> (priority>>1);
> - count_min = (limit<<1) >> (priority);
> + count_max = limit;
> + count_min = (limit<<2) >> (priority);
>
[snip]

-- 
If E. Coli was a computer:
 "OPERON.DLL: Mutation in LACTOSE.DLL.  Entire system halted."


Benjamin Redelings I       <><      http://sdcc13.ucsd.edu/~bredelin


-
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/