From: buers AT gmx DOT de (Dieter Buerssner)
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Q: operand size - FPU instructios
Date: 24 Mar 2000 21:29:56 GMT
Lines: 21
Message-ID: <8bgmoj$502uu$2@fu-berlin.de>
References: <38DA73F1 DOT FB5C375B AT mtu-net DOT ru> <200003241737 DOT WAA00845 AT midpec DOT com> <38DBC680 DOT 3B37E90F AT mtu-net DOT ru>
NNTP-Posting-Host: pec-0-114.tnt1.s2.uunet.de (149.225.0.114)
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: fu-berlin.de 953933396 5245918 149.225.0.114 (16 [17104])
X-Posting-Agent: Hamster/1.3.13.0
User-Agent: Xnews/03.02.04
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Alexei A. Frounze wrote:

>But appending these suffixes to FPU instructions, I get error messages
>like "invalid instruction" or something like that. I.e. AS doesn't
>recognize FILDW (for 16-bit integer), FLDQ (floating point double), FLDT
>(floating point extended - FPU internal 10-byte format).

From the manual of as (info as):

   * Floating point constructors are `.float' or `.single', `.double',
     and `.tfloat' for 32-, 64-, and 80-bit formats.  These correspond
     to instruction mnemonic suffixes `s', `l', and `t'. `t' stands for
     80-bit (ten byte) real. [...]
         
So to load float double and long double, use flds, fldl and fldt.
To load a 16, 32 and 64 bit signed interger use fild, fildl and fildq.

There is also some info about inline assembly and floating point
registers in the Gcc manual (info gcc).

-- Regards, Dieter