From: pavenis AT lanet DOT lv
To: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Date: Thu, 6 Apr 2000 11:52:41 +0200
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject: Re: Gdb and floating point
CC: djgpp-workers AT delorie DOT com
Message-ID: <38EC7A89.6065.2077F9@localhost>
References: <Pine DOT A41 DOT 4 DOT 05 DOT 10004031011540 DOT 47140-100000 AT ieva05 DOT lanet DOT lv>
In-reply-to: <Pine.SUN.3.91.1000403105513.26246H-100000@is>
X-mailer: Pegasus Mail for Win32 (v3.12c)
Reply-To: djgpp-workers AT delorie DOT com

On 3 Apr 2000, at 10:57, Eli Zaretskii wrote:

> 
> On Mon, 3 Apr 2000, Andris Pavenis wrote:
> 
> > > > Also. Had to hack RHIDE rather much to be able to build it using this
> > > > snapshot.
> > > 
> > > Could you summarize the problems you had with building RHIDE?  Is
> > > there something in GDB 5.0 that should be changed to make it easier?
> > 
> > I don't think so. RHIDE should be modified in this case.
> 

Earlier RHIDE provided it's own copy of fputc_unfiltered() (which is 
ugly hack of course). Now I changed it to use mem_fileopen() for both 
gdb_stdout and gdb_stderr.

One nuisance:

to get actual positon in ui_file opened with mem_fileopen currently 
only possibility is to retrieve entire buffer with ui_file_xstrdup()
(RHIDE needs to get such position currently). I think it would be nice 
to add something like

	ui_file_xstrlen()

to get length of buffer only (really it's current pointer) like (untested):

long ui_file_xstrlen (struct ui_file *file)
{
  if (file->magic==&mem_file_magic)
    {
       struct mem_file * mem = file->to_data;
       if (mem->magic==&mem_file_magic)
         return mem->length_buffer;
    }
  return 0;
}


> Perhaps in the long run RHIDE should switch to the GDB-mi interface 
> instead of the annotations.  It seems like the GDB developers want to 
> deprecate annotations, they are kept mainly because Emacs uses that 
> interface.
> 
> Note that currently, the DJGPP version of GDB is built without GDB-mi.  I 
> need time (which I don't have ;-) to work on adding that.

Andris