Subject: Re: C vs. prototypes (was: Problems with gdb under NetBSD 1.0 )
To: Ronald Khoo <ronald@cpm.COM.MY>
From: Greg A. Woods <woods@kuma.web.net>
List: current-users
Date: 12/19/1994 10:54:38
[ On Mon, December 19, 1994 at 11:55:38 (+0000), Ronald Khoo wrote: ]
> Subject: Re: C vs. prototypes (was: Problems with gdb under NetBSD 1.0 )
>
> It's actually quite a sensible principle.  After all, it's not for
> nothing that plan 9 compiles in 10 minutes flat -- I'd like to see
> NetBSD do that!  If you're developing or debugging software, you end
> up compiling the same code over and over again, so if your compiler
> has to keep all the context around to give you all those warnings,
> you're paying over and over again for checks that you'd only want to
> run every week or so.

Thanks!  My sentiments exactly!

It is quite contrary to the philosophy of tool use (i.e. what's made
UNIX famous!) to have a monolithic code checking, verification, and
compilation tool.

> When evaluating compilers for use in a commercial environment,
> lost programmer hours waiting for compilation figure reasonably high in
> my list of criteria!

Especially when you're often running that gigantic monolithic tool
simultaneously for several programmers (maybe GCC should be a shared
library! ;-).  I know the cost of memory and fast CPU's is getting
smaller, but still the inefficiency of this extravagance is sickening.

-- 
						Greg A. Woods

+1 416 443-1734			VE3TCP		robohack!woods
Planix, Inc. <woods@planix.com>; UniForum Canada <woods@uniforum.ca>