Return to BSD News archive
Xref: sserve comp.unix.bsd:12550 comp.os.linux:54382 comp.os.386bsd.misc:868
Newsgroups: comp.unix.bsd,comp.os.linux,comp.os.386bsd.misc
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!agate!doc.ic.ac.uk!uknet!mcsun!chsun!mw
From: mw@eunet.ch (Markus Wild)
Subject: Re: BSD UNIX
Message-ID: <CCuHI1.M97@eunet.ch>
Organization: CHUUG/EUnet Switzerland
References: <michaelv.747084422@ponderous.cc.iastate.edu> <CGD.93Sep3160517@erewhon.CS.Berkeley.EDU> <michaelv.747102277@ponderous.cc.iastate.edu> <CCu0s1.29o@ssesco.com>
Date: Sat, 4 Sep 1993 19:44:24 GMT
Lines: 83
In article <CCu0s1.29o@ssesco.com> rhealey@gorp.ssesco.com (Rob Healey) writes:
>In article <michaelv.747102277@ponderous.cc.iastate.edu> michaelv@iastate.edu (Michael L. VanLoon) writes:
>>In <CGD.93Sep3160517@erewhon.CS.Berkeley.EDU> cgd@erewhon.CS.Berkeley.EDU (Chris G. Demetriou) writes:
>>
>>> amiga "production"
>>
>>>re: amiga: just because binaries aren't available on -lamp doesn't
>>>mean it's not in production use... there are many places carrying
>>>binaries for it in europe, but their names all escape me right now...
>>
>>Finally, a reason to buy an Amiga... :-) I'm almost tempted.
>>
> Hey now! B^). I'll pit the 3000's 0.9 I/O performance against any
> 386 or 486 based system any day!
>
> I hear the screen/console code doesn't take full advantage of
> the hardware yet tho. And there is the issue of shared, or
> better yet dynamic, librarys.
>
> I run SVR4 on my Amiga tho so I'm not 100% up to date on 0.9's
> status.
Huh.. take care Rob.. The A3000 does indeed have nice I/O performance,
but you won't exploit it with Amiga SVR4.. this one contains one of the
most dreadful implementations of a scsi driver I've seen on the amiga...
I did a quick iozone test on a 30M file (as was suggested in an
unrelated thread here):
bash# a.out 30 8192
IOZONE: Performance Test of Sequential File I/O -- V1.16 (10/28/92)
By Bill Norcott
Operating System: POSIX 1003.1-1988
Send comments to: norcott_bill@tandem.com
IOZONE writes a 30 Megabyte sequential file consisting of
3840 records which are each 8192 bytes in length.
It then reads the file. It prints the bytes-per-second
rate at which the computer can read and write files.
Writing the 30 Megabyte file, 'iozone.tmp'...53.816667 seconds
Reading the file...29.700000 seconds
IOZONE performance measurements:
584526 bytes/second for writing the file
1059167 bytes/second for reading the file
This was on a totally "untuned" fs, I'll try and see if tunefs can
change the values significantly, I doubt it though. If I could hack
disksort() to not only sort blocks, but really merge them, that would
make for a whole lot more improvement (I could more or less double
performance when reading from the raw disk device just by doubling
block sizes...)
The screen/console code has been improved in the meantime, the console
part that runs on the native chip set (I'm myself using the Retina
console, which is based on the NCR 77C22E+ VGA chip) now uses the
Copper for scrolling, which should give it similar performance as the
Amiga SVR4 console, you don't really notice you're writing into a
bitmapped framebuffer.
As for dynamic libraries, there will be a NetBSD solution, not an
Amiga solution.
For those interested into more details about the Amiga port, subscribe
to the netbsd-amiga mailing list, just write to
netbsd-admin@cbmuucp.commodore.com (note that commodore just kindly
allowed to run the list on one of their machines, they're not involved
into NetBSD at all).
-Markus
BTW: binaries for NetBSD for the Amiga are available on ftp.eunet.ch,
software/os/bsd/NetBSD/NetBSD-Amiga, however, the current ones are
really old, if you can, either recompile yourself or wait a week
or so, I'll upload a new set by then.
--
CHUUG/EUnet Switzerland Markus Wild
Zweierstrasse 35 Tel: +41 1 291 45 80 mw@eunet.ch
CH-8004 Zuerich Fax: +41 1 291 46 42 S=mw;P=EUnet;A=EUnet;C=CH