public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
* patch for 7.2.1? dwarf2read.c (dwarf2_const_value_data): Never sign extend.
@ 2010-11-13  2:15 Doug Evans
  2010-11-20  5:25 ` Joel Brobecker
  0 siblings, 1 reply; 3+ messages in thread
From: Doug Evans @ 2010-11-13  2:15 UTC (permalink / raw)
  To: gdb-patches

This is rather nasty.

class test
{
  static const int k128 = 128;
};

test foo;

int
main ()
{
  return 0;
}

$ gdb a.out
(gdb) start
(gdb) p test::k128
-128

I propose adding this patch to the 7.2.1 branch.
[I haven't run a regression test.
If there are changes needed I'll check for those before checking in.]

2010-07-27  Tom Tromey  <tromey@redhat.com>

	* dwarf2read.c (dwarf2_const_value_data): Never sign extend.

Index: dwarf2read.c
===================================================================
RCS file: /cvs/src/src/gdb/dwarf2read.c,v
retrieving revision 1.431
retrieving revision 1.432
diff -u -p -r1.431 -r1.432
--- dwarf2read.c	28 Jul 2010 19:04:07 -0000	1.431
+++ dwarf2read.c	28 Jul 2010 20:05:03 -0000	1.432
@@ -10485,8 +10485,13 @@ dwarf2_const_value (struct attribute *at
 }
 
 
-/* Given an attr with a DW_FORM_dataN value in host byte order, sign-
-   or zero-extend it as appropriate for the symbol's type.  */
+/* Given an attr with a DW_FORM_dataN value in host byte order,
+   zero-extend it as appropriate for the symbol's type.  The DWARF
+   standard (v4) is not entirely clear about the meaning of using
+   DW_FORM_dataN for a constant with a signed type, where the type is
+   wider than the data.  The conclusion of a discussion on the DWARF
+   list was that this is unspecified.  We choose to always zero-extend
+   because that is the interpretation long in use by GCC.  */
 static void
 dwarf2_const_value_data (struct attribute *attr,
 			 struct symbol *sym,
@@ -10495,12 +10500,7 @@ dwarf2_const_value_data (struct attribut
   LONGEST l = DW_UNSND (attr);
 
   if (bits < sizeof (l) * 8)
-    {
-      if (TYPE_UNSIGNED (SYMBOL_TYPE (sym)))
-	l &= ((LONGEST) 1 << bits) - 1;
-      else
-	l = (l << (sizeof (l) * 8 - bits)) >> (sizeof (l) * 8 - bits);
-    }
+    l &= ((LONGEST) 1 << bits) - 1;
 
   SYMBOL_VALUE (sym) = l;
   SYMBOL_CLASS (sym) = LOC_CONST;

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: patch for 7.2.1? dwarf2read.c (dwarf2_const_value_data): Never sign extend.
  2010-11-13  2:15 patch for 7.2.1? dwarf2read.c (dwarf2_const_value_data): Never sign extend Doug Evans
@ 2010-11-20  5:25 ` Joel Brobecker
  2010-11-25 19:43   ` Doug Evans
  0 siblings, 1 reply; 3+ messages in thread
From: Joel Brobecker @ 2010-11-20  5:25 UTC (permalink / raw)
  To: Doug Evans; +Cc: gdb-patches

> 2010-07-27  Tom Tromey  <tromey@redhat.com>
> 
> 	* dwarf2read.c (dwarf2_const_value_data): Never sign extend.
>
> I propose adding this patch to the 7.2.1 branch.

Sounds like a good idea to me.

-- 
Joel

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: patch for 7.2.1? dwarf2read.c (dwarf2_const_value_data): Never sign extend.
  2010-11-20  5:25 ` Joel Brobecker
@ 2010-11-25 19:43   ` Doug Evans
  0 siblings, 0 replies; 3+ messages in thread
From: Doug Evans @ 2010-11-25 19:43 UTC (permalink / raw)
  To: Joel Brobecker; +Cc: gdb-patches

On Fri, Nov 19, 2010 at 9:25 PM, Joel Brobecker <brobecker@adacore.com> wrote:
>> 2010-07-27  Tom Tromey  <tromey@redhat.com>
>>
>>       * dwarf2read.c (dwarf2_const_value_data): Never sign extend.
>>
>> I propose adding this patch to the 7.2.1 branch.
>
> Sounds like a good idea to me.

Tested on amd64-linux, no regressions.
Checked in.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2010-11-25 19:43 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-11-13  2:15 patch for 7.2.1? dwarf2read.c (dwarf2_const_value_data): Never sign extend Doug Evans
2010-11-20  5:25 ` Joel Brobecker
2010-11-25 19:43   ` Doug Evans

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).