[PATCH] stack.c, memory leak (no-effect code?)

[PATCH] stack.c, memory leak (no-effect code?)

[msnyder]

[-- Attachment #1: Type: text/plain, Size: 194 bytes --]

This patch is to fix a memory leak, but while you're looking at it...
does it seem to you that these two blocks have any effect?

[Sorry if I posted this before, I've been searching for it...]


[-- Attachment #2: 160.txt --]
[-- Type: text/plain, Size: 926 bytes --]

2007-08-04  Michael Snyder  <msnyder@access-company.com>

	* stack.c (print_frame): Memory leak.

Index: stack.c
===================================================================
RCS file: /cvs/src/src/gdb/stack.c,v
retrieving revision 1.150
diff -p -r1.150 stack.c
*** stack.c	25 Jul 2007 00:28:25 -0000	1.150
--- stack.c	5 Aug 2007 02:16:10 -0000
*************** print_frame (struct frame_info *frame, i
*** 582,587 ****
--- 582,589 ----
  		   the symbol table. That'll have parameters, but
  		   that's preferable to displaying a mangled name.  */
  		funname = SYMBOL_PRINT_NAME (func);
+ 	      else
+ 		xfree (demangled);
  	    }
  	}
      }
*************** frame_info (char *addr_exp, int from_tty
*** 882,887 ****
--- 884,891 ----
  	     preferable to displaying a mangled name.  */
  	  if (demangled == NULL)
  	    funname = SYMBOL_PRINT_NAME (func);
+ 	  else
+ 	    xfree (demangled);
  	}
      }
    else

Re: [PATCH] stack.c, memory leak (no-effect code?)

[Jim Blandy]

msnyder@sonic.net writes:
> This patch is to fix a memory leak, but while you're looking at it...
> does it seem to you that these two blocks have any effect?
>
> [Sorry if I posted this before, I've been searching for it...]

You did post something very much like this, but I recall that the
blocks in question in the older post really didn't have any effect.
These at least set funname.

I'm kind of surprised that they don't set funname to demangled when
it's non-NULL.

Re: [PATCH] stack.c, memory leak (no-effect code?)

[msnyder]

>
> msnyder@sonic.net writes:
>> This patch is to fix a memory leak, but while you're looking at it...
>> does it seem to you that these two blocks have any effect?
>>
>> [Sorry if I posted this before, I've been searching for it...]
>
> You did post something very much like this, but I recall that the
> blocks in question in the older post really didn't have any effect.
> These at least set funname.
>
> I'm kind of surprised that they don't set funname to demangled when
> it's non-NULL.

OK, how about the patch itself?

Re: [PATCH] stack.c, memory leak (no-effect code?)

[msnyder]

>>
>> msnyder@sonic.net writes:
>>> This patch is to fix a memory leak, but while you're looking at it...
>>> does it seem to you that these two blocks have any effect?
>>>
>>> [Sorry if I posted this before, I've been searching for it...]
>>
>> You did post something very much like this, but I recall that the
>> blocks in question in the older post really didn't have any effect.
>> These at least set funname.
>>
>> I'm kind of surprised that they don't set funname to demangled when
>> it's non-NULL.
>
> OK, how about the patch itself?

So, checking in the patch, ignoring the possibility of dead code.