X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f
From: RayeR <glaux AT centrum DOT cz>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: problem with AS from binutils 2.21.1
Date: Mon, 2 Jan 2012 17:09:50 -0800 (PST)
Organization: http://groups.google.com
Lines: 23
Message-ID: <561636bb-1ec8-4a7b-b1d6-62e28c2f6de9@m10g2000vbc.googlegroups.com>
References: <8a32880d-526c-40ac-a9bb-e88fce4ee2aa AT z25g2000vbs DOT googlegroups DOT com>
 <21258842-845d-4b87-9c29-52d0528e2ca0 AT dp8g2000vbb DOT googlegroups DOT com>
 <76df8c48-1463-43d1-98a1-4c32fb5c7643 AT t16g2000vba DOT googlegroups DOT com>
 <3e83fb36-219b-4314-8201-7f0047eb60d2 AT z12g2000yqm DOT googlegroups DOT com>
 <9metpcFndoU1 AT mid DOT dfncis DOT de> <1b8a3a38-882b-4056-9c3d-ed23db8ab4fb AT f33g2000yqh DOT googlegroups DOT com>
NNTP-Posting-Host: 78.45.168.98
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
X-Trace: posting.google.com 1325552990 6480 127.0.0.1 (3 Jan 2012 01:09:50 GMT)
X-Complaints-To: groups-abuse AT google DOT com
NNTP-Posting-Date: Tue, 3 Jan 2012 01:09:50 +0000 (UTC)
Complaints-To: groups-abuse AT google DOT com
Injection-Info: m10g2000vbc.googlegroups.com; posting-host=78.45.168.98; posting-account=Q0wMHAoAAADjYrghh94FTf6YnbpTqZgp
User-Agent: G2/1.0
X-Google-Web-Client: true
X-Google-Header-Order: HUALESNKRC
X-HTTP-UserAgent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.19)
 Gecko/20110420 SeaMonkey/2.0.14,gzip(gfe)
Bytes: 2686
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Thanks for clarification. I didn't know there's such limit in COFF. I
expected that this 32bit system hasn't much 'stupid' 16-bit
limitations to care about...
As a mistake I wrote that object file reached about 3,4MB but it
should be size of temp. asm file (L:\WINXP\TEMP/ccOqvtFj.s) which even
growed over 11MB - hard to exactly say I was observing temp directory
and afer it was done temp file disappeared.
So I could take in mind that number of relocations depens on optimize
level and gcc version. I don't want to go back to older version.
Splitting the source file is not practical. So currently I use #pragma
for opt level as quick and easy solution but as Andris said it may not
be possible in future (mentioned gcc 4.7). I don't understand djgpp/
coff internals to make a coff patch but it seems it will be necessary.

Any idea about the waring in libc:
e:/djgpp/lib/libc.a(d0e01.o): warning: previous common is here
e:/djgpp/lib/libc.a(d0600.o): warning: multiple common of
`___dpmi_error'

I can found only definition in DPMI.H
extern unsigned short __dpmi_error;
I don't have libc sources and didn't compiled it but I use updated
version by Ozkan Sezer.