Re: broken non-BFD assembler

Re: broken non-BFD assembler

[Nathan Sidwell]

Jan Beulich wrote:
>>	* as.h (seg_not_empty_p): Declare.
>>	* subsegs.c (seg_not_empty_p): New predicate.
> 
> 
> The use of bfd_boolean here breaks non-BFD assemblers. If only used in
> such, declaration and definition should be bounded by #if
> BFD_ASSEMBLER.

ok, it should be change to a plain int.  I'll patch it when sourceware
comes back to life

nathan

-- 
Nathan Sidwell    ::   http://www.codesourcery.com   ::     CodeSourcery LLC
nathan@codesourcery.com    ::     http://www.planetfall.pwp.blueyonder.co.uk

Re: broken non-BFD assembler

[Nathan Sidwell]

[-- Attachment #1: Type: text/plain, Size: 474 bytes --]

Jan Beulich wrote:
>>	* as.h (seg_not_empty_p): Declare.
>>	* subsegs.c (seg_not_empty_p): New predicate.
> 
> 
> The use of bfd_boolean here breaks non-BFD assemblers. If only used in
> such, declaration and definition should be bounded by #if
> BFD_ASSEMBLER.

I've installed the attached obvious patch.

nathan

-- 
Nathan Sidwell    ::   http://www.codesourcery.com   ::     CodeSourcery LLC
nathan@codesourcery.com    ::     http://www.planetfall.pwp.blueyonder.co.uk


[-- Attachment #2: bool.patch --]
[-- Type: text/plain, Size: 1776 bytes --]

2005-02-07  Nathan Sidwell  <nathan@codesourcery.com>

	* as.h (seg_not_empty_p): Return int, not bfd_boolean.
	* subsegs.c (seg_not_empty_p): Likewise.

Index: gas/as.h
===================================================================
RCS file: /cvs/src/src/gas/as.h,v
retrieving revision 1.41
diff -c -3 -p -r1.41 as.h
*** gas/as.h	27 Jan 2005 16:33:46 -0000	1.41
--- gas/as.h	7 Feb 2005 17:47:58 -0000
*************** segT   subseg_new (const char *, subsegT
*** 577,583 ****
  segT   subseg_force_new (const char *, subsegT);
  void   subseg_set (segT, subsegT);
  int    subseg_text_p (segT);
! bfd_boolean seg_not_empty_p (segT);
  void   start_dependencies (char *);
  void   register_dependency (char *);
  void   print_dependencies (void);
--- 577,583 ----
  segT   subseg_force_new (const char *, subsegT);
  void   subseg_set (segT, subsegT);
  int    subseg_text_p (segT);
! int    seg_not_empty_p (segT);
  void   start_dependencies (char *);
  void   register_dependency (char *);
  void   print_dependencies (void);
Index: gas/subsegs.c
===================================================================
RCS file: /cvs/src/src/gas/subsegs.c,v
retrieving revision 1.19
diff -c -3 -p -r1.19 subsegs.c
*** gas/subsegs.c	27 Jan 2005 16:33:47 -0000	1.19
--- gas/subsegs.c	7 Feb 2005 17:47:58 -0000
*************** subseg_text_p (segT sec)
*** 601,607 ****
     fr_fix == 0 one still contributes data.  Think of this as
     seg_definitely_not_empty_p.  */
  
! bfd_boolean
  seg_not_empty_p (segT sec)
  {
    segment_info_type *seginfo = seg_info (sec);
--- 601,607 ----
     fr_fix == 0 one still contributes data.  Think of this as
     seg_definitely_not_empty_p.  */
  
! int
  seg_not_empty_p (segT sec)
  {
    segment_info_type *seginfo = seg_info (sec);

Re: broken non-BFD assembler

[Alan Modra]

On Mon, Feb 07, 2005 at 05:54:19PM +0000, Nathan Sidwell wrote:
> Jan Beulich wrote:
> >>	* as.h (seg_not_empty_p): Declare.
> >>	* subsegs.c (seg_not_empty_p): New predicate.
> >
> >
> >The use of bfd_boolean here breaks non-BFD assemblers. If only used in
> >such, declaration and definition should be bounded by #if
> >BFD_ASSEMBLER.

Sorry.  My fault for telling Nathan to use bfd_boolean rather than int.

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre