Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ulowell!masscomp!ftw
From: ftw@masscomp.UUCP (Farrell Woods)
Newsgroups: comp.lang.c
Subject: Re: bitfields considered harmful?
Message-ID: <786@masscomp.UUCP>
Date: 29 Apr 89 16:42:57 GMT
References: <1473@uwbull.uwbln.UUCP>
Reply-To: ftw@quasar.UUCP (Farrell Woods)
Distribution: comp
Organization: MASSCOMP - Westford, Ma
Lines: 25
Approved: news

In article <1473@uwbull.uwbln.UUCP> ckl@uwbln.UUCP (Christoph Kuenkel) writes:
>Now, are bitfields generally considered harmfull?

If they're harmful, then why would we keep them around or use them?

It's tempting to use them for things like mapping hardware device registers
to the bit level, in some kind of device driver.  This is non-portable if I'm
not mistaken, but not explicitly harmful.

>Are there other good reasons to avoid them?

If they're not implemented correctly in your compiler...

>what does ANSI C say about it?

The pANS will tell you what behavior to expect from bitfields, and it will
tell you about some gray areas (implementation-defined) in their definition
and use.


-- 
Farrell T. Woods				Voice:  (508) 392-2471
Concurrent Computer Corporation			Domain: ftw@masscomp.com
1 Technology Way				uucp:   {backbones}!masscomp!ftw
Westford, MA 01886				OS/2:   Half an operating system