[commit/multiarch/hppa] push_dummy_frame method taken care of

[commit/multiarch/hppa] push_dummy_frame method taken care of

[Joel Brobecker]

[-- Attachment #1: Type: text/plain, Size: 723 bytes --]

Hello,

boring stuff, I just committed the following patch, using the multiarch
pass. Next is the fix_call_dummy method. After that, unless my notes are
incomplete, we should be able to go to multiarch partial for hppa
(hppa64 will come after).

2003-03-19  J. Brobecker  <brobecker@gnat.com>

        Continuing work to convert the hppa targets to multiarch partial.

        * hppa-tdep.c (hppa_gdbarch_init): Set the push_dummy_frame gdbarch
        method, now that hppa_push_dummy_frame has a conformant prototype.
        * config/pa/tm-hppa.h (DEPRECATED_PUSH_DUMMY_FRAME): Wrap macro
        inside "#if !GDB_MULTI_ARCH ... #endif" conditional, in preparation
        for the switch to multiarch partial.

-- 
Joel

[-- Attachment #2: hpux.diff --]
[-- Type: text/plain, Size: 2311 bytes --]

Index: hppa-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/hppa-tdep.c,v
retrieving revision 1.58
diff -c -3 -p -r1.58 hppa-tdep.c
*** hppa-tdep.c	18 Mar 2003 18:57:47 -0000	1.58
--- hppa-tdep.c	19 Mar 2003 20:02:57 -0000
*************** hppa_gdbarch_init (struct gdbarch_info i
*** 5000,5006 ****
    set_gdbarch_frame_locals_address (gdbarch, hppa_frame_locals_address);
    set_gdbarch_frame_num_args (gdbarch, hppa_frame_num_args);
    set_gdbarch_frame_args_skip (gdbarch, 0);
!   /* set_gdbarch_deprecated_push_dummy_frame (gdbarch, hppa_push_dummy_frame);  */
    set_gdbarch_deprecated_pop_frame (gdbarch, hppa_pop_frame);
    set_gdbarch_call_dummy_length (gdbarch, INSTRUCTION_SIZE * 28);
    set_gdbarch_call_dummy_start_offset (gdbarch, 0);
--- 5000,5006 ----
    set_gdbarch_frame_locals_address (gdbarch, hppa_frame_locals_address);
    set_gdbarch_frame_num_args (gdbarch, hppa_frame_num_args);
    set_gdbarch_frame_args_skip (gdbarch, 0);
!   set_gdbarch_deprecated_push_dummy_frame (gdbarch, hppa_push_dummy_frame);
    set_gdbarch_deprecated_pop_frame (gdbarch, hppa_pop_frame);
    set_gdbarch_call_dummy_length (gdbarch, INSTRUCTION_SIZE * 28);
    set_gdbarch_call_dummy_start_offset (gdbarch, 0);
Index: config/pa/tm-hppa.h
===================================================================
RCS file: /cvs/src/src/gdb/config/pa/tm-hppa.h,v
retrieving revision 1.37
diff -c -3 -p -r1.37 tm-hppa.h
*** config/pa/tm-hppa.h	18 Mar 2003 18:57:48 -0000	1.37
--- config/pa/tm-hppa.h	19 Mar 2003 20:02:57 -0000
*************** extern void hppa_frame_find_saved_regs (
*** 450,459 ****
  
  /* Things needed for making the inferior call functions.  */
  
! /* Push an empty stack frame, to record the current PC, etc. */
! 
  #define DEPRECATED_PUSH_DUMMY_FRAME hppa_push_dummy_frame ()
  extern void hppa_push_dummy_frame (void);
  
  /* Discard from the stack the innermost frame, 
     restoring all saved registers.  */
--- 450,459 ----
  
  /* Things needed for making the inferior call functions.  */
  
! #if !GDB_MULTI_ARCH
  #define DEPRECATED_PUSH_DUMMY_FRAME hppa_push_dummy_frame ()
  extern void hppa_push_dummy_frame (void);
+ #endif
  
  /* Discard from the stack the innermost frame, 
     restoring all saved registers.  */

Re: [commit/multiarch/hppa] push_dummy_frame method taken care of

[Andrew Cagney]

> Hello,
> 
> boring stuff, I just committed the following patch, using the multiarch
> pass. Next is the fix_call_dummy method. After that, unless my notes are
> incomplete, we should be able to go to multiarch partial for hppa
> (hppa64 will come after).
> 
> 2003-03-19  J. Brobecker  <brobecker@gnat.com>
> 
>         Continuing work to convert the hppa targets to multiarch partial.
> 
>         * hppa-tdep.c (hppa_gdbarch_init): Set the push_dummy_frame gdbarch
>         method, now that hppa_push_dummy_frame has a conformant prototype.
>         * config/pa/tm-hppa.h (DEPRECATED_PUSH_DUMMY_FRAME): Wrap macro
>         inside "#if !GDB_MULTI_ARCH ... #endif" conditional, in preparation
>         for the switch to multiarch partial.

Ya!  I think the FRAME_EXTRA_INFO needs to be converted first - that's 
the big one.

Andrew