From: Toivo Pedaste (toivo@ucs.uwa.OZ.AU)
Date: 08/03/92


From: toivo@ucs.uwa.OZ.AU (Toivo Pedaste)
Subject: Re: .97 (new buffer allocation code)
Date: 4 Aug 1992 01:54:52 GMT

tdunbar@vttcf.cc.vt.edu (Thomas Dunbar) writes:

> v .97 is faster in X on my 386sx/387 with 8M ram according to my very
>unscientific test:
> date
> xv -root -rmode 5 -quit /root/lake.marron
> date
>(which uses @3 Meg) took 77 sec in v .96cp2 but only takes 70 sec now
>(both having just booted up and with olvwm and one xterm)

Well I did a test by compiling the 0.97 kernal (after a make clean) under both
0.96 and 0.97. I have a 386-40 with 8meg memory and ran the test in an xterm
on an otherwise idle machine (xclock was the only other X application running).

0.96 kernal
        730s real 448s user 127s system

0.97 kernal
       1521s real 449s user 166s system

It took twice as long, I assume this was increased disk traffic (hence the
higher system time). Is there a problem with dirty disk blocks filing
up memory?