From mboxrd@z Thu Jan  1 00:00:00 1970
From: Orjan Friberg <orjan.friberg@axis.com>
To: Andrew Cagney <ac131313@cygnus.com>
Cc: gdb@sources.redhat.com
Subject: Re: Uninitialized variable in CRIS port
Date: Fri, 21 Sep 2001 11:12:00 -0000
Message-id: <3BAB82FA.A043CDCE@axis.com>
References: <3BA9143D.1000502@cygnus.com>
X-SW-Source: 2001-09/msg00193.html

Andrew Cagney wrote:
> 
>   /home/scratch/GDB/src/gdb/cris-tdep.c: In function
> `reg_mode_add_sub_cmp_and_or_move_op':
> /home/scratch/GDB/src/gdb/cris-tdep.c:3054: warning: `operand1' might be
> used uninitialized in this function
> 
> checking the code, the problem looks real.

Thanks for the reminder.  Patch below, confirmed with the weirdest test
case I've ever written.  Ok to commit on trunk and branch?


2001-09-21  Orjan Friberg  <orjanf@axis.com>

        * cris-tdep.c (reg_mode_add_sub_cmp_and_or_move_op): Fetch
operand1
        from correct register.


Index: cris-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/cris-tdep.c,v
retrieving revision 1.3
diff -c -3 -p -r1.3 cris-tdep.c
*** cris-tdep.c 2001/09/19 21:59:41     1.3
--- cris-tdep.c 2001/09/21 17:54:03
*************** reg_mode_add_sub_cmp_and_or_move_op (uns
*** 3070,3076 ****
            return;
          }
        /* The instruction has the PC as its target register.  */
!       operand1 = inst_env->reg[operand1]; 
        operand2 = inst_env->reg[REG_PC];
  
        /* Check if it's a extend, signed or zero instruction.  */
--- 3070,3076 ----
            return;
          }
        /* The instruction has the PC as its target register.  */
!       operand1 = inst_env->reg[cris_get_operand1 (inst)]; 
        operand2 = inst_env->reg[REG_PC];
  
        /* Check if it's a extend, signed or zero instruction.  */

-- 
Orjan Friberg              E-mail: orjan.friberg@axis.com
Axis Communications AB     Phone:  +46 46 272 17 68