From: "Andrew Crabtree" <andrewc AT rosemail DOT rose DOT hp DOT com>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: sizeof(long double) = 12 ???
Date: Mon, 3 Aug 1998 10:14:20 -0700
Organization: Hewlett-Packard, Roseville
Lines: 11
Message-ID: <6q4r56$cgn$1@rosenews.rose.hp.com>
References: <199808020838 DOT EAA10324 AT delorie DOT com> <35C4540B DOT 75DEE699 AT alcyone DOT com> <35C5797D DOT 48365D65 AT taniwha DOT tssc DOT co DOT nz>
NNTP-Posting-Host: ros51675cra.rose.hp.com
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk


>(586+).  This is the reason gcc will *always, no matter what* allocate
>12 bytes to a long double (ok, 16 bytes would be better for 586+ (and
>pgcc might), but this is even more of a waste of memory, so you have a
>compramise).
no.  This would have consequences like breaking code compiled with regular
gcc and pgcc.  There are -m options for the i386 that can control alignment
of doubles and long doubles but they break the ABI so you have to recompile
everything (libraries, start up code ....)