*BSD News Article 16242


Return to BSD News archive

Xref: sserve comp.windows.x.i386unix:1583 comp.os.386bsd.questions:2539
Newsgroups: comp.windows.x.i386unix,comp.os.386bsd.questions
Path: sserve!newshost.anu.edu.au!munnari.oz.au!constellation!convex!convex!convex!darwin.sura.net!wupost!csus.edu!netcom.com!hasty
From: hasty@netcom.com (Amancio Hasty Jr)
Subject: Re: XFree1-2 + 386BSD performance
Message-ID: <hastyC78F50.364@netcom.com>
Organization: Netcom Online Communications Services (408-241-9760 login: guest)
References: <1t9mg3INNoep@srvr1.engin.umich.edu> <1ta5ro$ip5@zaphod.axion.bt.co.uk> <1tarqr$jmv@binkley.cs.mcgill.ca>
Date: Tue, 18 May 1993 16:58:11 GMT
Lines: 30

In article <1tarqr$jmv@binkley.cs.mcgill.ca> storm@binkley.cs.mcgill.ca (Marc Wandschneider) writes:
>In article <1ta5ro$ip5@zaphod.axion.bt.co.uk> lessen@axion.bt.co.uk (Lee Essen) writes:
>>
>>I think this unquestionably a big argument in favour of shared libraries, some
>>of use can afford to have 32 Meg systems, and even those that do probably dont
>>like the idea of 28 Meg of it filled with several copies of the static libraries.
>>
>>Why is there so much 'anti-shared-library' feeling?
>
I am beta testing shared library support and  we are running
into minor problems. When they are ready they will be posted :-)

About bsd performance, one side of the equation is the support
for shared libraries, the other side is that drivers execute
at ipl level thus locking out servers like X. What people have
done in the past is to split the driver into at least two parts,
one basically, that handles interrupts and writing to the devices
and the other at an interruptable priority.

So if we get shareable library support we will still run into
performance problems while compiling.

Cheers,
Amancio

-- 
This message brought to you by the letters X and S and the number 3
Amancio Hasty           |  
Home: (415) 495-3046    |  ftp-site depository of all my work:
e-mail hasty@netcom.com	|  sunvis.rtpnc.epa.gov:/pub/386bsd/incoming