Re: The gdb x86 function prologue parser

[Eli Zaretskii <eliz@gnu.org>]

> Cc: Mark Kettenis <kettenis@jive.nl>
> From: Jason Molenda <jmolenda@apple.com>
> Date: Tue, 7 Jun 2005 22:51:36 -0700
> 
> I can't even begin to imagine how annoyed developers using the FSF
> gdb on x86 must be.

Well, I'm one of the annoyed, although I have no idea whether the
problems that annoyed me would be solved by your patches.  In any
case, thanks.

> --- i386-tdep.c	28 May 2005 16:44:28 -0000	1.213
> +++ i386-tdep.c	8 Jun 2005 05:24:14 -0000
> @@ -21,6 +21,7 @@
>     Foundation, Inc., 59 Temple Place - Suite 330,
>     Boston, MA 02111-1307, USA.  */
>  
> +#include <stdint.h>

I don't think we can use stdint.h freely, as we still don't require a
C9x compiler.  I see you needed it for things like uint8_t and
uint32_t, which should be easily replaceable by suitable standard x86
types.

I also agree with Daniel: it would be nice to find a more graceful way
of storing knowledge about so many instructions.

Last, but not least, I'd surely appreciate some write-up, even in
plain ASCII, about how prologue analyzers in general and the x86 one
in particular work: that stuff is sorely needed in gdbint.texinfo.

TIA