X-Authentication-Warning:	ieva01.lanet.lv: pavenis owned process doing -bs
Date:	Fri, 27 Aug 1999 10:23:22 +0300 (WET)
From:	Andris Pavenis <pavenis AT lanet DOT lv>
To:	Shawn Hargreaves <ShawnH AT Probe DOT co DOT uk>
cc:	djgpp AT delorie DOT com
Subject:	Re: Scratch that, installing allegro
In-Reply-To:	<8D53104ECD0CD211AF4000A0C9D60AE3015CAC4F@probe-2.acclaim-euro.net>
Message-ID:	<Pine.A41.4.05.9908271018010.50552-100000@ieva01.lanet.lv>
MIME-Version:	1.0
Reply-To:	djgpp AT delorie DOT com
X-Mailing-List:	djgpp AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Thu, 26 Aug 1999, Shawn Hargreaves wrote:

> Johan Venter writes:
> >> Seems that You're trying to compile Allegro with GCC-2.95. There are
> >> some changes I had to apply to build it (first patch is to satisfy
> >> binutils snapshot I'm using). There are some more warnings from AS
> >> (990817 binutils snapshot) about wrong length suffix in assembler
> >> instructions (eg. suffix 'w', register %esi). I didn't fix them all.
> >
> > Why not just download the latest WIP? The DOS version is extrememly 
> > stable because it is basically still allegro 3.11
> 
> Actually, the current WIP Allegro codebase is quite different to
> 3.11: there have been a lot of basic changes to the way things are
> organised. It does seem to be fairly stable at the moment, though.
> 
> It would, however, be nice to fix up the 3.11 version that is currently 
> on Simtel to make it work with gcc 2.95. I don't have this installed
> myself (lack of time to download it right now), but if someone could
> prepare a patch that would make Allegro 3.11 work correctly with it,
> that would be very much appreciated...
> 

I'm including my patch below. It cowers not only problems I met with
gcc-2.95 but also other one with recent development snapshot of binutils
(should not break earlier versions). There are still some warnings
from assembler (I used 990817 snapshot) as some instructions contained
wrong operand length suffix (eg. suffix 'w', register %esi). As I didn't
want to study  sources much then I left all as it was (that give warnings
only)

Andris

*** allegro/src/asmdefs.inc~1	Sat Feb 20 16:51:22 1999
--- allegro/src/asmdefs.inc	Tue Aug 24 12:55:50 1999
***************
*** 41,48 ****
   * %eax. Registers will be unchanged, except %eax will return a pointer 
   * to the start of the selected scanline.
   */
! #define WRITE_BANK()    call BMP_WBANK(%edx)
! #define READ_BANK()     call BMP_RBANK(%edx)
  
  
  /* Helper macro for looking up a position in the pattern bitmap. Passed
--- 41,48 ----
   * %eax. Registers will be unchanged, except %eax will return a pointer 
   * to the start of the selected scanline.
   */
! #define WRITE_BANK()    call *BMP_WBANK(%edx)
! #define READ_BANK()     call *BMP_RBANK(%edx)
  
  
  /* Helper macro for looking up a position in the pattern bitmap. Passed
*** allegro/src/cpu.c~1	Thu Feb 25 20:38:56 1999
--- allegro/src/cpu.c	Tue Aug 24 12:58:00 1999
***************
*** 60,66 ****
  
     : "=a" (result)
     :
!    : "eax", "ecx", "memory"
     );
  
     return result;
--- 60,66 ----
  
     : "=a" (result)
     :
!    : "ecx", "memory"
     );
  
     return result;
***************
*** 120,126 ****
  
     : "=a" (result)
     :
!    : "%eax", "memory"
     );
  
     return result;
--- 120,126 ----
  
     : "=a" (result)
     :
!    : "memory"
     );
  
     return result;
***************
*** 152,158 ****
  
     : "=a" (result)
     :
!    : "%eax", "%ebx", "memory" );
  
     return result;
  }
--- 152,158 ----
  
     : "=a" (result)
     :
!    : "%ebx", "memory" );
  
     return result;
  }
***************
*** 272,278 ****
  
     : "=a" (result)
     :
!    : "%eax", "%ebx", "memory"
     );
  
     return result;
--- 272,278 ----
  
     : "=a" (result)
     :
!    : "%ebx", "memory"
     );
  
     return result;
  

- Raw text -