From mboxrd@z Thu Jan  1 00:00:00 1970
From: Scott Wood <scottwood@freescale.com>
Date: Mon, 06 Nov 2006 14:48:50 -0600
Subject: [U-Boot-Users] Why are some global vars part of the image,
 and some not?
In-Reply-To: <20061106202611.973EC353A60@atlas.denx.de>
References: <20061106202611.973EC353A60@atlas.denx.de>
Message-ID: <454F9FB2.8090009@freescale.com>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Wolfgang Denk wrote:
> The concept of BSS for using uninitialized data (meaning: implicitely
> initialized to zero)

However, with the current u-boot linker scripts (some of them, at least; 
I'm not going to make a sweeping statement about all 270), data 
*explicitly* initialized to zero is also going in the bss.  Combined 
with the fact that some C code is executed before the bss is cleared, 
that is simply evil.

> is as old as C and Unix, actualy even older than
> that (from IBM 7090  times,  IIRC).  It  is  simple,  efficient,  and
> powerful.

Your point being?  I'm not attacking the concept of a bss, just the use 
of it in a codebase where C code is executed prior to being able to 
clear the bss.

> I see zero reason to drop it.

I have stated the reason.  If you prefer an environment that invites 
bugs and ugly workarounds thereto, then suit yourself.

-Scott