A problem of umoven and umovestr in util.c

[Cai Fei]

Hi,

How about the following problem. Is it necessary to fix it?

Cai Fei wrote:
> Hi,
>
> I used strace to trace the following program test_accept.c:
> #include <sys/types.h>
> #include <sys/socket.h>
> int main(int argc, char *argv[])
> {
> accept(0, (struct sockaddr *)1, (socklen_t *)1);
> return 1;
> }
>
> $ strace ./test_accept
> ......
> accept(0, 0x1, ptrace: umoven: Input/output error
> 0x1) = -1 ENOTSOCK (Socket operation on non-socket)
> ......
>
> If the third param of accept() is invalid, the message "ptrace: umoven:
> Input/output error" will outputted. I think users will be confused to
> see the message. Users may be more confused even with -o option.
>
> $ strace -o log ./test_accept
> ptrace: umoven: Input/output error
>
> Also, the same problem exists in umovestr(). I found it with the following
> program test_access.c:
>
> #include <unistd.h>
> int main(int argc, char *argv[])
> {
> access((char *)1, R_OK);
> return 1;
> }
>
> $ strace ./test_access
> ......
> access(umovestr: Input/output error
> 0x1, R_OK) = -1 EFAULT (Bad address)
> ......
>
> I think it will be better if the error messages are not outputted.
> If it is right, here is the patch:
>
> Signed-off-by: "Cai Fei" <caifei at cn.fujitsu.com>
>
> --- util.c	2007-07-16 09:21:47.000000000 +0800
> +++ util.c.new	2007-07-16 09:23:08.000000000 +0800
> @@ -690,7 +690,6 @@ char *laddr;
>  				return 0;
>  			}
>  			/* But if not started, we had a bogus address. */
> -			perror("ptrace: umoven");
>  			return -1;
>  		}
>  		started = 1;
> @@ -705,8 +704,6 @@ char *laddr;
>  				/* Ran into 'end of memory' - stupid "printpath" */
>  				return 0;
>  			}
> -			if (addr != 0)
> -				perror("ptrace: umoven");
>  			return -1;
>  		}
>  		started = 1;
> @@ -840,7 +837,6 @@ char *laddr;
>  				/* Ran into 'end of memory' - stupid "printpath" */
>  				return 0;
>  			}
> -			perror("umovestr");
>  			return -1;
>  		}
>  		started = 1;
> @@ -858,7 +854,6 @@ char *laddr;
>  				/* Ran into 'end of memory' - stupid "printpath" */
>  				return 0;
>  			}
> -			perror("umovestr");
>  			return -1;
>  		}
>  		started = 1;
>
>
> Regards
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>   


-- 
A new email address of FJWAN is launched from Apr.1 2007.
The updated address is: caifei at cn.fujitsu.com
--------------------------------------------------
Cai Fei
Development Dept.I
Nanjing Fujitsu Nanda Software Tech. Co., Ltd.(FNST)
caifei at cn.fujitsu.com  (内7551-5875)
--------------------------------------------------