Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!rutgers!apple!oliveb!pyramid!prls!philabs!ttidca!mb
From: mb@ttidca.TTI.COM (Michael Bloom)
Newsgroups: comp.windows.x
Subject: Re: Purdue 2.0 2.1 on VAX with gcc 1.34
Keywords: Purdue VAX gcc
Message-ID: <4040@ttidca.TTI.COM>
Date: 13 Mar 89 20:22:01 GMT
References: <3541@sdsu.UUCP> <4485@pt.cs.cmu.edu>
Reply-To: mb@ttidca.tti.com (Michael Bloom)
Distribution: usa
Organization: Citicorp/TTI, Santa Monica
Lines: 25

This is what I'm using:
-------------
#ifdef vax
#define FASTGETBITS(psrc,x,w,dst) \
    asm ("extzv %1,%2,%3,%0" \
	 : "=g" (dst) \
	 : "g" (x), "g" (w), "m" (*(char *)(psrc)))
#define getbits(psrc,x,w,dst) FASTGETBITS(psrc,x,w,dst)

#define FASTPUTBITS(src, x, w, pdst) \
    asm ("insv %3,%1,%2,%0" \
	 : "=m" (*(char *)(pdst)) \
	 : "g" (x), "g" (w), "g" (src))
#define putbits(src, x, w, pdst) FASTPUTBITS(src, x, w, pdst)
#endif vax
-------------

By the way, if you're using a qdss, when you run your working server,
be sure to use the "-bs" option. The man page's mention of poor
performance without this option is an understatement!

And if you're running Ultrix and compiling on an NFS mounted file system,
compile everything twice and compare each object.  As of Ultrix 3.0
Digital ***STILL*** hasn't fixed the NFS client data corruption bug
that every other vendor fixed years ago.