This is the mail archive of the mailing list for the GDB project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: [RFA]dwarf reader: Avoid complaint on const type

Hi Pedro,

  I checked this a little, but it is still unclear to me:
gcc-3 only emits a DW_AT_type attribute, no DW_AT_name,
so there is no problem in that case.
  On gcc16, I do see the failure that you see,
but I rechecked my tests when I submitted, and there 
is no sigaltstack failure.
  I think that wehen I submitted the patch, I was supposing that
if the DW_TAG_volatile_type would have a name attribute,
it would only be to give the name of the newly defined type,
i.e. for a typedef.
  Apparently, GCC changed a reports the old "unmodified" type name.

  The patch below should fix this.

Pierre Muller

gcc does not emit typedefs for volatile types:

>>>>Start of test-vol.c<<<<<<

/* Levels completed flag.  */
typedef volatile enum level level2;
volatile enum level level = NR_LEVELS;

typedef volatile int  volint;

volint vol;

level2 test = OUTER;

int local ()
  typedef volatile enum level level;
  level loc;
  loc = INNER;
  return 0;

main ()
  vol = 5;
  level = MAIN;
  test = LEAF;
  local ();
  return 0;
>>>>Start of test-vol.c<<<<<<
gcc -gdwarf2 test-vol.c
results in an  executable with no information
about volint type.
(stabs info gives correct information)...
I have to add here that I hate the C way of describing typedefs:
(gdb) inf type myint
type int;
This is almost completely useless, especially when you do a 
(gdb) inf type
myint does not appear anywhere :(

Back to the patch itself:

ChangeLog entry:

2010-06-21  Pierre Muller  <>

	* dwarf2read.c (process_die): Do not call new_symbol
	for DW_TAG_volatile_type and DW_TAG_const_type.
	(new_symbol): Do not add the name of DW_TAG_volatile_type 
	and DW_TAG_const_type to the symbol list.

Index: dwarf2read.c
RCS file: /cvs/src/src/gdb/dwarf2read.c,v
retrieving revision 1.400
diff -u -p -r1.400 dwarf2read.c
--- dwarf2read.c        17 Jun 2010 22:36:41 -0000      1.400
+++ dwarf2read.c        20 Jun 2010 21:57:23 -0000
@@ -3210,12 +3210,14 @@ process_die (struct die_info *die, struc
     case DW_TAG_base_type:
     case DW_TAG_subrange_type:
     case DW_TAG_typedef:
-    case DW_TAG_const_type:
-    case DW_TAG_volatile_type:
       /* Add a typedef symbol for the type definition, if it has a
          DW_AT_name.  */
       new_symbol (die, read_type_die (die, cu), cu);
+    case DW_TAG_const_type:
+    case DW_TAG_volatile_type:
+      read_type_die (die, cu);
+      break;
     case DW_TAG_common_block:
       read_common_block (die, cu);
@@ -8912,12 +8914,16 @@ new_symbol (struct die_info *die, struct
        case DW_TAG_base_type:
         case DW_TAG_subrange_type:
-        case DW_TAG_const_type:
-        case DW_TAG_volatile_type:
          SYMBOL_CLASS (sym) = LOC_TYPEDEF;
          SYMBOL_DOMAIN (sym) = VAR_DOMAIN;
          add_symbol_to_list (sym, cu->list_in_scope);
+        case DW_TAG_const_type:
+        case DW_TAG_volatile_type:
+         /* The name of the type given in the dwarf name atribute is the
+            name of the `normal' type and not a new type name, so
+            do not register this as a new type name.  */
+         break;
        case DW_TAG_enumerator:
          attr = dwarf2_attr (die, DW_AT_const_value, cu);
          if (attr)

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]