Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!usc!snorkelwacker!spdcc!esegue!compilers-sender
From: pardo@cs.washington.edu (David Keppel)
Newsgroups: comp.compilers
Subject: Re: '.stabs' info requested
Keywords: C, debug
Message-ID: <1990Jun7.040810.3665@esegue.segue.boston.ma.us>
Date: 7 Jun 90 04:08:10 GMT
References: <1990Jun5.173819.2970@esegue.segue.boston.ma.us> <1990Jun6.150120.1572@esegue.segue.boston.ma.us>
Sender: compilers-sender@esegue.segue.boston.ma.us
Reply-To: pardo@cs.washington.edu (David Keppel)
Organization: University of Washington, Computer Science, Seattle
Lines: 25
Approved: compilers@esegue.segue.boston.ma.us

title@Think.COM (Richard Title) writes:
>[dbx .stabs documentation: `stabstring.c']

One piece of information that you won't find in the dbx documentation:
some perfectly legal .stabs entries cause popular (e.g., BSD VAX)
assemblers to barf with a message like ``illegal .stabs value''.

The problem is severe enough that at least for a while, GNU C++ had a
compiler compile-time flag ``FASCIST_ASSEMBLER''.  The first time that I
built a version of GNU C++ that had this flag, I missed the fact that I
needed to set the flag, and assembly failed as a result.  I took the
problem to one of the authors of BSD `as', and he agreed that it was a
bug in `as'.

One moral of the story is ``do it the right way, have a flag for broken
tools.''

		;-D on  ( Undebugger )  Pardo
-- 
		    pardo@cs.washington.edu
    {rutgers,cornell,ucsd,ubc-cs,tektronix}!uw-beaver!june!pardo
-- 
Send compilers articles to compilers@esegue.segue.boston.ma.us
{spdcc | ima | lotus}!esegue.  Meta-mail to compilers-request@esegue.
Please send responses to the author of the message, not the poster.