From: <ams AT ludd DOT luth DOT se>
Message-Id: <200305101733.h4AHXKtd020020@speedy.ludd.luth.se>
Subject: float_t and double_t
To: DJGPP-WORKERS <djgpp-workers AT delorie DOT com>
Date: Sat, 10 May 2003 19:33:20 +0200 (CEST)
X-Mailer: ELM [version 2.4ME+ PL78 (25)]
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII
X-MailScanner: Found to be clean
Reply-To: djgpp-workers AT delorie DOT com

Hello.

C99 says we should define the types float_t and double_t to "the
implementation's most efficient types at least as wide as float and
double respectively".

1. Most efficient? As in speed I suppose.

2. Which is the most efficient? As all fp operations are done as long
double within the CPU (IIRC), is my guess that long double is as
efficient as any of the other two correct? Does this apply to the 387
FPU as well?

3. Should we define them to long double then?

4. We need to rename our already present types with those names (in
libc/ieee.h).

Comments?


Right,

						MartinS