Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!uwm.edu!zaphod.mps.ohio-state.edu!wuarchive!udel!mmdf
From: eo@ansa.co.uk (Ed Oskiewicz)
Newsgroups: comp.os.minix
Subject: (ST) Has anyone else run the tests?
Message-ID: <10326@nigel.udel.EDU>
Date: 6 Feb 90 21:33:13 GMT
Sender: mmdf@udel.EDU
Lines: 33

I've performed most of the upgrade on my 1040/TCT winchester and have run
across the same problems as others have reported.  However, after applying the
posted fixes - especially patch #24 it seems to run all but one of the tests
OK (on a floppy only system of course).  The exception is test6 which babbles

	"Error 6 : No Error"

repeatedly then says ok!

There are three problems here:

1)	Saying "Error n : no error" is just confusing (I assume the latter
	means that errno was 0).

2)	Saying ok when errors have been reported is sloppy programming!

3)	The problem with test6 seems to be that two calls to sbrk() have
	arguments cast to (char *) when sbrk wants an int.  I'm not typing
	this at home so can't look at the problem until I get back but sbrk()
	is certainly being called with inconsistently typed arguments.  Either
	these two calls to sbrk() should be calls to brk() or the cast should
	be removed.

Has anyone else run test6 on an ST?
	
Incidentally the recently posted 'missing' files for the ST in part #25
contains the broken lib.h (it needs patching) and a broken lib/atari/makefile
(change the .c to .s).

These minor gripes apart, Minix 1.5.0 looks to be a significant step forward,
thanks to all concerned.

Ed Oskiewicz (eo@ansa.co.uk)