From: "Mark E." <snowball3 AT bigfoot DOT com>
To: djgpp-workers AT delorie DOT com
Date: Sun, 7 Jan 2001 13:00:05 -0500
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject: Re: Backslashes in debug info
Message-ID: <3A586855.22721.66BDCF@localhost>
In-reply-to: <7458-Sun07Jan2001192945+0200-eliz@is.elta.co.il>
References: <3A585ADA DOT 250 DOT 321507 AT localhost> (snowball3 AT bigfoot DOT com)
X-mailer: Pegasus Mail for Win32 (v3.12c)
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> Mark, could you, or someone else, say a few words about the reasons of
> the switch to the BFD assembler, and its advantages and disadvantages?
> Should we consider having both in the binary distributions?

One reason is the BFD assembler can handle as many sections as memory allows 
while the non-BFD can only handle a very small number of sections (about 40 I 
think). Unique sections generated by gcc (like .gnu.linkonce.* in C++) can 
easily reach this limit in one source file. Another reason is the generic 
dwarf2 support in gas now can be used only by the BFD assembler.

Mark