Xref:	news2.mv.net comp.os.msdos.djgpp:6115
From:	Charles Sandmann <sandmann AT dmccorp DOT com>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: Near pointer improvements
Date:	Wed, 17 Jul 1996 16:26:45 CDT
Organization:	Houston Supernet& Tulsa Supernet
Lines:	15
Message-ID:	<31ed5a95.sandmann@dmccorp.com>
References:	<Pine DOT SOL DOT 3 DOT 91 DOT 960717193250 DOT 14118F-100000 AT zveris>
Reply-To:	sandmann AT dmccorp DOT com
NNTP-Posting-Host:	igate.dmccorp.com
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

> Yup. But I think setting the .._NEARPTR bit in _crt0_startup_flags would
> solve the problem, as sbrk() checks for that. Is that right?

Well not exactly.  Setting this bit does prevent setting of limits in
the sbrk() code for the current ds register and ds_alias - but it would
prevent the needed update to reflect the correct limit in the DS which
you were not using.  You should also not set ds as ds_alias (allocate
a new one...) unless you want to see some ugly crashes in some unique
situations.  You also need to review the code for places which assume
things about ds - like crt0.s and exceptn.s - for things that will break.

The only way to make the current sbrk() work right with your toggle scheme
would be to have the "normal" small limit DS loaded when you call it.
Your algorithm is a good idea, but we will need to re-write some code to
implement it fully.  

- Raw text -