Date: Wed, 14 Jun 2000 13:38:12 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Nimrod Abing <n_abing AT hotmail DOT com>
cc: djgpp AT delorie DOT com
Subject: Re: Using PMODE/DJ
In-Reply-To: <20000614071645.95605.qmail@hotmail.com>
Message-ID: <Pine.SUN.3.91.1000614133249.29611M-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Wed, 14 Jun 2000, Nimrod Abing wrote:

> I would like to know whether PMODE/DJ
> uses the floating point coprocessor to
> function properly and thus requires one
> to be present.

I don't know whether PMODE/DJ issues FP instruction (I would expect it 
not to do that).  However, the sources for PMODE are available, so you 
should be able to look there and know for sure.

It's possible that some function in your program emits FP instructions.  
You should be able to use "objdump --disassembly", and then grep the 
result for FP instructions.

> So, who's responsible for the SIGNOFP?

When i386 sees an FP instruction, and there's no 80387, the CPU generates 
a special exception ("Coprocessor Not Present", exception no. 7).  
The startup code of the DJGPP library installs an exception handler for 
this exception, which catches it and converts it into the signal SIGNOFP.