[2.1.131-ac5] MM Experience with X in 6,8 Megs RAM (page cache too large
Benjamin Redelings I (bredelin@ucsd.edu)
Wed, 09 Dec 1998 12:09:50 -0800
Well, I booted my 64Mb, UP PPro with mem=6M and then mem=8M to see how
it worked. My IDE hard disk gets about 3.8 Mb/s max (as recorded by
hdparm) and I have a 34M swap partition on it.
In 6M, I actually got to an xterm, but it was paging every keystroke,
so I quickly (actually, it was NOT very quick :) rebooted into mem=8M,
which was a lot better. I could us xterm in 8Mb, and I could use emacs
at the same time. The write cache remained quite small... I didn't
really do much writing though.
The main problem was that about 1/2 the RAM was filled with page
cache. Now, since "1/2 the RAM" was only 3.0 - 3.5 Mb, the stuff in the
page cache may have been worth swapping out large parts of emacs, xterm,
and the X-server. Perhaps there are some statistics in /proc that tell
the hit rate in the page cache. I'll go take a look in the
Documentation/....
HOWEVER.
I then did 'echo "1 10 11" > /proc/sys/vm/pagecache'. THIS HAD NO
EFFECT. The page cache did NOT go back down to 11% of memory. I can't
think of a good reason why this would be so, but perhaps the page cache
grew as rapidly as it shrank, showing that it was very valuable
indeed... I tried running ddd instead of emacs, because I thought emacs
might use a lot of files for its bytecodes. The same thing happened
with ddd.
I would think that this wouldn't happen, because state should always be
set to 0 in do_try_to_free_pages.
-BenRI
--
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/