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-patches@sources.redhat.com
Subject: Re: [rfc] Swap out current when creating a new architecture
Date: Mon, 01 Oct 2001 04:34:00 -0000
Message-id: <3BB854B3.208DC17A@axis.com>
References: <3BB16441.30805@cygnus.com> <3BB771A1.4070201@cygnus.com>
X-SW-Source: 2001-10/msg00001.html

Andrew Cagney wrote:
> 
> The CRIS target also dumps core with it applied.  Like rs6000 was, it is
> refering to the previous (current_gdbarch) architecture.
> 
> I'll sit on this for a bit longer.  Could I encourage target maintainers
> to check their XXX_gdbarch_init() function for references to
> current_gdbarch.

Just a quick recap: what the old code was doing by referring to
current_gdbarch was to avoid changing the ABI (which I infer from the
bfd) if one of the other target specific commands were being used.  The
patch below sets the abfd field in the info struct to exec_bfd before
calling gdbarch_update_p, so that information on the current bfd is
passed along, just as it is when gdbarch_update_p is called from
set_gdbarch_from_file.

Is this an acceptable way of doing it?  I know I'm in the risky business
of dealing with global pointers.  (I'm assuming a commit could go on
both trunk and branch.)


2001-10-01  Orjan Friberg  <orjanf@axis.com>

	* cris-tdep.c (cris_gdbarch_init): Remove reference to current_gdbarch.
	(cris_version_update, cris_mode_update, cris_abi_update): Set abfd in
	info struct to exec_bfd.

Index: cris-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/cris-tdep.c,v
retrieving revision 1.4
diff -c -3 -p -r1.4 cris-tdep.c
*** cris-tdep.c 2001/09/24 08:04:10     1.4
--- cris-tdep.c 2001/10/01 10:58:11
*************** cris_version_update (char *ignore_args, 
*** 3596,3601 ****
--- 3596,3605 ----
    
        /* Update the current architecture, if needed.  */
        memset (&info, 0, sizeof info);
+ 
+       /* Supply the current bfd (if any).  */
+       info.abfd = exec_bfd;
+ 
        if (!gdbarch_update_p (info))
          internal_error (__FILE__, __LINE__, "cris_gdbarch_update:
failed to update architecture.");
      }  
*************** cris_mode_update (char *ignore_args, int
*** 3614,3619 ****
--- 3618,3627 ----
    
        /* Update the current architecture, if needed.  */
        memset (&info, 0, sizeof info);
+ 
+       /* Supply the current bfd (if any).  */
+       info.abfd = exec_bfd;
+ 
        if (!gdbarch_update_p (info))
          internal_error (__FILE__, __LINE__, "cris_gdbarch_update:
failed to update architecture.");
      }
*************** cris_abi_update (char *ignore_args, int 
*** 3632,3637 ****
--- 3640,3649 ----
    
        /* Update the current architecture, if needed.  */
        memset (&info, 0, sizeof info);
+ 
+       /* Supply the current bfd (if any).  */
+       info.abfd = exec_bfd;
+ 
        if (!gdbarch_update_p (info))
          internal_error (__FILE__, __LINE__, "cris_gdbarch_update:
failed to update architecture.");
      }
*************** cris_gdbarch_init (struct gdbarch_info i
*** 3736,3748 ****
            /* Unknown bfd flavour.  Assume it's the new ABI.  */
            cris_abi = CRIS_ABI_V2;
          }
-     }
-   else if (gdbarch_tdep (current_gdbarch))
-     {
-       /* No bfd available.  Stick with whatever ABI we're currently
using.  
-          (This is to avoid changing the ABI when the user updates the 
-          architecture with the 'set cris-version' command.)  */
-       cris_abi = gdbarch_tdep (current_gdbarch)->cris_abi;
      }
    else
      {
--- 3748,3753 ----


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