Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!munnari.oz.au!brolga!uqvax.cc.uq.oz.au!wattle!zcpanews
From: rnews@qut.edu.au
Newsgroups: comp.os.minix
Subject: RE: Confusion with 1.3 to 1.5 upgrade
Message-ID: <9010301816.AA03569@stehekin.boeing.com>
Date: 30 Oct 90 18:16:37 GMT
Organization: Boeing Computer Services AI Center, Seattle
Lines: 32
To:	rnews@qut.edu.au
CC:
To: rnews@qut.edu.au
In-Reply-To: <19058.2722ac99@qut.edu.au>
X-Envelope-to: rnews

In article <19058.2722ac99@qut.edu.au> you write:
>Surprise! Surprise! Another problem with building the library as part of
>upgrading from 1.3 to 1.5.
>Are the lists of object files in Glen Overby's Upgrade Tutorial (Glen.O.tute)
>Appendix C and D difinitive ? I ask this because my libc.a file does not
>contain an object called 'perf.s'; but it does have one called 'error.s'.
>
>There is a source file 'perf.c' that came with the string.00.Z upgrade
>source file, but it doesn't seem to have a suitable format for a library
>function.
>
>If anyone could enlighten me on this, I would really appreciate it.

I'll try.  :-)  

The official 1.5.10 crc list has the file "perf.c" in lib/string.  Since it's
in lib, I presume that it goes in libc.a.  (My sources and my libc.a are at
home, so I can't tell you if perf.s is actually in my library.)  I find
no file called either "error.c" or "error.s" in the 1.5.10 crc list.

The most important question is: "What happens when you link stuff with
your library the way it is?"  If things link without unresolved externals,
then you've got what you absolutely need and can worry about this detail
later.

Hope this helps!

Paul Allen
-- 
------------------------------------------------------------------------
Paul L. Allen                       | pallen@atc.boeing.com
Boeing Advanced Technology Center   | ...!uw-beaver!bcsaic!pallen