Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 10 May 2002 15:35:46 +1000 (EST)
From:      Bruce Evans <bde@zeta.org.au>
To:        John Baldwin <jhb@FreeBSD.org>
Cc:        cvs-all@FreeBSD.org, <cvs-committers@FreeBSD.org>
Subject:   RE: cvs commit: src/sys/boot/i386/boot2 Makefile
Message-ID:  <20020510151516.Q7524-100000@gamplex.bde.org>
In-Reply-To: <XFMail.20020510000808.jhb@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 10 May 2002, John Baldwin wrote:

> AFAICT, the problem is that gcc 3.1 is not properly respecting the
> -fno-align-foo arguments.  My guess is that this is a bug in gcc, but
> I have no idea if it is a bug in general or if it is i386-specific.  Using
> alignment values greater than 4 noticably bloated the binary further, but
> values <= 4 make no difference.  (no-align-foo is supposedly equivalent to
> align-foo=1, and align-foo=0 is some MD default value).

Alignment to 4-byte boundaries only accounts for half of the bloat.  I get
the following amounts free with gcc-old and various alignment options:

-malign-foo=0 (default)                    148
-malign-foo=1                              116 (bloat =  32)
-malign-foo=2                               20 (bloat = 128)
-malign-foo=3                             -108 (bloat = 256)
-malign-foo=4                             -428 (bloat = 576)

Unbreaking the Makefile to include <bsd.kern.mk> so that my
-mno-align-long-strings flag is put in CFLAGS and fixing the syntax errors
exposed by -ansi gives:

-malign-foo=0 -mno-align-long-strings      212 (bloat = -64)

Bruce


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020510151516.Q7524-100000>