Date: Thu, 15 May 1997 20:02:01 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: "A.V. Kondratiev" <akondra AT mmf DOT univ DOT simbirsk DOT su>
cc: djgpp AT delorie DOT com
Subject: Re: problem with inline functions in templates
In-Reply-To: <AA8rshY0pst@mmf.univ.simbirsk.su>
Message-ID: <Pine.SUN.3.91.970515200003.10056Y-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On Thu, 15 May 1997, A.V. Kondratiev wrote:

> The compiler crashes when optimization is enabled (is says
> Abort! and nothing more). Everything is OK if -fno-inline
> option is turned on (But I'd like my inline members to be
> inline - they are small and frequently called).
> Are there any known dangerous cases of using inlines which
> GCC programmer should avoid?
> Or may it be possible to invoke some compiler diagnostic to
> locate the code dangerous for the compiler ?

Did you try the advice of section 6.3 in the DJGPP FAQ list?  There are 
some suggestions there, including an undocumented gcc switch that might 
help you.