From: Martin Str|mberg <ams AT ludd DOT luth DOT se>
Message-Id: <200103251451.QAA18175@mother.ludd.luth.se>
Subject: Re: Bug 323
To: djgpp-workers AT delorie DOT com
Date: Sun, 25 Mar 2001 16:51:40 +0200 (MEST)
In-Reply-To: <Pine.SUN.3.91.1010325161044.25476C-100000@is> from "Eli Zaretskii" at Mar 25, 2001 04:13:49 PM
X-Mailer: ELM [version 2.5 PL2]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
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

According to Eli Zaretskii:
> 
> 
> On Sun, 25 Mar 2001, Martin Str|mberg wrote:
> 
> > +#define O_APPEND       0x0010
> 
> Not good, 0x0010 is already taken by SH_DENYRW.
> 
> In general, all of the bits of the lowest byte in the open mode variable 
> passed to open and _open are used by DOS (we do special trickery to reset 
> the O_BINARY and O_TEXT bits, because DOS doesn't know anything about 
> them).
> 
> I admit I don't remember what was wrong with the previous solution you 
> suggested, so I cannot suggest alternatives.  Could you remind what were 
> we talking about last time?

<http://www.delorie.com/djgpp/mail-archives/browse.cgi?p=djgpp-workers/2001/01/31/03:31:45>
says:

"On Tue, 30 Jan 2001, Martin Str|mberg wrote:

> -extern char *__file_handle_modes;
> +extern unsigned short *__file_handle_modes;

Please don't!  This immediately breaks all existing libraries compiled
with previous versions of this header.

Do we really need a short for each handle?  Did we use up all the 8
bits in the current definition?"


Suggestions?


Right,

						MartinS