gcc HEAD changes stab function names for static functions

gcc HEAD changes stab function names for static functions

[Michael Elizabeth Chastain]

I've got another regression in gcc HEAD and I just want to check whether
it's a gcc bug or whether we should handle it in gdb.

Here is the source code:

  static void f1 (int i);
  static void f1 (int i)
  {
    return;
  }

Here is the output from gcc HEAD 20030716:

	  .stabs  "f1:f(0,21)=(0,21)",36,0,3,f1                                           .stabs  "i:p(0,1)",160,0,2,8
	  .type   f1, @function                                                   f1:
	  .stabn 68,0,3,.LM1-f1

And here is the output from gcc HEAD 20030722:

	  .stabs  "f1.0:f(0,21)=(0,21)",36,0,3,f1.0
	  .stabs  "i:p(0,1)",160,0,2,8                                                    .type   f1.0, @function
  f1.0:                                                                                   .stabn 68,0,3,.LM1-f1.0

The function name change from "f1" to "f1.0", even though "f1"
is declared and defined at file scope.  This happens with -gstabs+.
It does not happen with -gdwarf-2 (or if it does, gdb is handling
it okay).

This looks like a bug in gcc.  Or should gdb handle this, even at
file scope?

gcc has new code here and it will be easy to isolate the regression.
gcc tries to test whether "f1" is at file scope or a local scope.  I
think it's getting mislead by the "static void f1 (int i);" declaration.
But my real question is: should I whinge at gcc to change the compiler
back, or whinge at gdb to accommodate the new compiler behavior.

Michael C

Re: gcc HEAD changes stab function names for static functions

[Daniel Jacobowitz]

On Fri, Jul 25, 2003 at 02:29:02PM -0400, Michael Elizabeth Chastain wrote:
> I've got another regression in gcc HEAD and I just want to check whether
> it's a gcc bug or whether we should handle it in gdb.
> 
> Here is the source code:
> 
>   static void f1 (int i);
>   static void f1 (int i)
>   {
>     return;
>   }
> 
> Here is the output from gcc HEAD 20030716:
> 
> 	  .stabs  "f1:f(0,21)=(0,21)",36,0,3,f1                                           .stabs  "i:p(0,1)",160,0,2,8
> 	  .type   f1, @function                                                   f1:
> 	  .stabn 68,0,3,.LM1-f1
> 
> And here is the output from gcc HEAD 20030722:
> 
> 	  .stabs  "f1.0:f(0,21)=(0,21)",36,0,3,f1.0
> 	  .stabs  "i:p(0,1)",160,0,2,8                                                    .type   f1.0, @function
>   f1.0:                                                                                   .stabn 68,0,3,.LM1-f1.0
> 
> The function name change from "f1" to "f1.0", even though "f1"
> is declared and defined at file scope.  This happens with -gstabs+.
> It does not happen with -gdwarf-2 (or if it does, gdb is handling
> it okay).
> 
> This looks like a bug in gcc.  Or should gdb handle this, even at
> file scope?
> 
> gcc has new code here and it will be easy to isolate the regression.
> gcc tries to test whether "f1" is at file scope or a local scope.  I
> think it's getting mislead by the "static void f1 (int i);" declaration.
> But my real question is: should I whinge at gcc to change the compiler
> back, or whinge at gdb to accommodate the new compiler behavior.

This is fallback from Geoff's program-at-a-time patch.  I thought he'd
been persuaded to change it back for file-at-a-time, but maybe he
didn't do it right - see the message from Bob Wilson yesterday.  Don't
remember which gcc list.

It will probably be reverted.  In any case, it should have been
"f1:f(0,21)" and f1.0:, the f1.0 bit should not appear in a stabs
string.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer