This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC 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: Wformat-security and string literal

From: Hagen Paul Pfeifer <hppgccml at 0xdef dot net>
To: gcc at gcc dot gnu dot org, Hagen Paul Pfeifer <hppgccml at 0xdef dot net>
Date: Mon, 17 Oct 2005 18:51:51 +0200
Subject: Re: Wformat-security and string literal
References: <20051017094905.GA6538@0xdef.net> <20051017162850.GA25653@synopsys.com>

* Joe Buck | 2005-10-17 09:28:51 [-0700]:

>If we go down that route, the option would only work correctly if
>optimization is enabled, because you expect gcc to trace the source
>of pointers, and see what the pointers point to, and check the original
>strings for format characters.

ACK - gcc behave correctly if optimization is enabled. To my mind: a
pointer who points to a string literal is also a string literal. The
generated string is stored in the .rodata section, on either side (-O0,
-O2). The generated code is the same, but the warning messages differ!


HGN

References:
- Wformat-security and string literal
  - From: Hagen Paul Pfeifer
- Re: Wformat-security and string literal
  - From: Joe Buck

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