iq2000 warning fix

iq2000 warning fix

[Alan Modra]

This is an obvious patch to fix a warning, and of course needs applying
to the /src/cpu/ dir as well.  It seems like /src/cpu/ is now supposed
to be the master, but I see the /src/cgen/cpu/ "copy" differs quite a
bit..  What's the story?

	* cpu/iq2000.opc (parse_jtargq10): Change type of valuep to bfd_vma *.

Index: cgen/cpu/iq2000.opc
===================================================================
RCS file: /cvs/src/src/cgen/cpu/iq2000.opc,v
retrieving revision 1.2
diff -u -p -r1.2 iq2000.opc
--- cgen/cpu/iq2000.opc	14 Apr 2004 21:29:33 -0000	1.2
+++ cgen/cpu/iq2000.opc	19 Nov 2004 11:03:23 -0000
@@ -39,7 +39,7 @@ extern int iq2000_cgen_insn_supported (C
 #include "safe-ctype.h"
 
 static int iq2000_cgen_isa_register PARAMS ((const char **));
-static const char * parse_jtargq10 PARAMS ((CGEN_CPU_DESC, const char **, int, int, enum cgen_parse_operand_result *, unsigned long *));
+static const char * parse_jtargq10 PARAMS ((CGEN_CPU_DESC, const char **, int, int, enum cgen_parse_operand_result *, bfd_vma *));
 static const char * parse_mimm PARAMS ((CGEN_CPU_DESC, const char **, int, long *));
 static const char * parse_imm  PARAMS ((CGEN_CPU_DESC, const char **, int, unsigned long *));
 static const char * parse_hi16 PARAMS ((CGEN_CPU_DESC, const char **, int, unsigned long *));
@@ -156,7 +156,7 @@ parse_jtargq10 (cd, strp, opindex, reloc
      int opindex;
      int reloc ATTRIBUTE_UNUSED;
      enum cgen_parse_operand_result *type_addr ATTRIBUTE_UNUSED;
-     unsigned long *valuep;
+     bfd_vma *valuep;
 {
   const char *errmsg;
   bfd_vma value;

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre

Re: iq2000 warning fix

[Nick Clifton]

Hi Alan,

> This is an obvious patch to fix a warning, and of course needs applying
> to the /src/cpu/ dir as well.  It seems like /src/cpu/ is now supposed
> to be the master, but I see the /src/cgen/cpu/ "copy" differs quite a
> bit..  What's the story?

I suspect that what has happened is that patches have been checked in to 
/src/cpu without being checked in to /src/cgen/cpu.

I think that what we need is a way to mark files in /src/cgen/cpu as 
obsolete and/or replace them with links to the master copies in /src/cpu.

Cheers
   Nick