Be more liberal in ignoring ARM mapping symbols

Be more liberal in ignoring ARM mapping symbols

[Daniel Jacobowitz]

I noticed today that the ARM compiler also generates $b mapping symbols.
Rather than continuing to play catch-up, this makes $<character> and
$<character>.<string> be generally ignored.  Is this OK?

-- 
Daniel Jacobowitz
CodeSourcery

2006-04-11  Daniel Jacobowitz  <dan@codesourcery.com>

	* bfd/cpu-arm.c (bfd_is_arm_mapping_symbol_name): Accept more
	mapping symbols.

Index: bfd/cpu-arm.c
===================================================================
RCS file: /cvs/src/src/bfd/cpu-arm.c,v
retrieving revision 1.19.12.1
diff -u -p -r1.19.12.1 cpu-arm.c
--- bfd/cpu-arm.c	11 Apr 2006 15:54:53 -0000	1.19.12.1
+++ bfd/cpu-arm.c	11 Apr 2006 16:25:25 -0000
@@ -405,11 +405,11 @@ bfd_boolean
 bfd_is_arm_mapping_symbol_name (const char * name)
 {
   /* The ARM compiler outputs several obsolete forms.  Recognize them
-     in addition to the standard $a, $t and $d.  */
+     in addition to the standard $a, $t and $d.  We are somewhat loose
+     in what we accept here, since the full set is not documented.  */
   return (name != NULL)
     && (name[0] == '$')
-    && ((name[1] == 'a') || (name[1] == 't') || (name[1] == 'd')
-	|| (name[1] == 'm') || (name[1] == 'f') || (name[1] == 'p'))
+    && (name[1] >= 'a' && name[1] <= 'z')
     && (name[2] == 0 || name[2] == '.');
 }
 

Re: Be more liberal in ignoring ARM mapping symbols

[Nick Clifton]

Hi Daniel,

> 2006-04-11  Daniel Jacobowitz  <dan@codesourcery.com>
> 
> 	* bfd/cpu-arm.c (bfd_is_arm_mapping_symbol_name): Accept more
> 	mapping symbols.

Approved - please apply.

Cheers
   Nick