From: Helge Hafting <helgehaf@idb.hist.no>
To: linux-kernel@vger.kernel.org
Subject: 2.4.10-pre12 is impressive
Date: Wed, 19 Sep 2001 13:16:46 +0200 [thread overview]
Message-ID: <3BA87E9E.215A1E3@idb.hist.no> (raw)
2.4.10-pre12 is impressive compared to pre10.
128M, 300MHz and a 3G ide-drive is usually enough for office work
and compiling, but there are times I use some swap. Reading the
impressive reviews for pre11,
I decided to put it through a worst-case test. Well, worst-case
for my kind of use anyway.
I started staroffice with 40 documents. This is slow and tends to
ruin interactive performance for a long time. I used
to get lots of swapping, and plenty of swapping in long after
finishing staroffice.
This time I also started a kernel compile, a updatedb run, and a
rgrep in /usr/src while waiting for staroffice to load.
I also had netscape up.
It came up, with only 10M total in swap. 20-30M just for starting
staroffice used to be the case. And interactive performance
was fine, even with all that going on.
The normal under such circumstances is several seconds just to
raise a xterm, with typing oocationally lagging and delays when
pressing enter. But only commands needing disk-io got delayed and
only slightly. Typing and window moving were just like
a normal X session with nothing else going on.
The rgrep seemed to
proceed at great speed, the compile ran very slowly. I guess it
lost the bandwidth competition.
Then I went away for lunch, and when I came back there were
no more than 20M in swap. Everything had finished except the
compile - it was still hampered by netscape doing one of its
100% cpu bugs.
Staroffice and netscape _still_ performed fine, all that
io didn't destroy their working sets! I quit staroffice,
the compile finished, and swap usage dropped do 15M.
The machine have recovered perfectly - there seems to be none of
that long-term sluggishness that used to happen after
updatedb or after swapping 20M. The machine used to
recover to a usable state before too, but never this good.
Before, I had to reboot or use it a few hours.
Good work! There's noticeable improvement for desktop use!
Helge Hafting
reply other threads:[~2001-09-19 11:17 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=3BA87E9E.215A1E3@idb.hist.no \
--to=helgehaf@idb.hist.no \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®