This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc 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: [PATCH] vfprintf: Fix memory with large width and precision [BZ #19931]

From: Florian Weimer <fweimer at redhat dot com>
To: Paul Eggert <eggert at cs dot ucla dot edu>
Cc: GNU C Library <libc-alpha at sourceware dot org>
Date: Mon, 11 Apr 2016 18:02:27 +0200
Subject: Re: [PATCH] vfprintf: Fix memory with large width and precision [BZ #19931]
Authentication-results: sourceware.org; auth=none
References: <570BC02D dot 2040901 at redhat dot com> <570BC840 dot 6060807 at cs dot ucla dot edu>

On 04/11/2016 05:52 PM, Paul Eggert wrote:

On 04/11/2016 08:18 AM, Florian Weimer wrote:

+      /* Deallocate any previously allocated buffer because it is
+         too small.  */
+      if (workstart != NULL)
+        {
+          free (workstart);
+          workstart = NULL;
+        }


Other places in that function use this code instead:

     if (__glibc_unlikely (workstart != NULL))
       free (workstart);
     workstart = NULL;

Is there some reason to do things differently here?

I can add the __glibc_unlikely, it's one of the few cases where it isunlikely to hurt. (The branch is only taken if the memory leak is aboutto happen, so I assume it's quite rare indeed, but I don't know for sure.)

Regarding the braces, both variants already exist in this file. I don'thave a firm presence. I find the one with the braces more logical.


Florian

References:
- [PATCH] vfprintf: Fix memory with large width and precision [BZ #19931]
  - From: Florian Weimer
- Re: [PATCH] vfprintf: Fix memory with large width and precision [BZ #19931]
  - From: Paul Eggert

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