Re: [PATCH -tip] Documentation/accounting/getdelays.c intialize the variable before using it

[Andrew Morton <akpm@linux-foundation.org>]

On Wed, 20 May 2009 10:44:44 +0530 Jaswinder Singh Rajput <jaswinder@kernel.org> wrote:

> 
> May be in some cases cmd_type will be used uninitialized.
> 
> Also fixes compilation warning:
> 
>    CC [M]  Documentation/DocBook/procfs_example.o
>   Documentation/accounting/getdelays.c: In function ___main___:
>   Documentation/accounting/getdelays.c:249: warning: ___cmd_type___ may be used uninitialized in this function
> 
> Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
> ---
>  Documentation/accounting/getdelays.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/Documentation/accounting/getdelays.c b/Documentation/accounting/getdelays.c
> index 7ea2311..36aae3b 100644
> --- a/Documentation/accounting/getdelays.c
> +++ b/Documentation/accounting/getdelays.c
> @@ -246,7 +246,7 @@ void print_ioacct(struct taskstats *t)
>  
>  int main(int argc, char *argv[])
>  {
> -	int c, rc, rep_len, aggr_len, len2, cmd_type;
> +	int c, rc, rep_len, aggr_len, len2, cmd_type = TASKSTATS_CMD_ATTR_UNSPEC;
>  	__u16 id;
>  	__u32 mypid;
>  

>From my reading of the code there is no bug here - we'll only use
cmd_type if tid != 0, and cmd_type is always initialised when tid != 0.

But I suppose we should squish the warning anyway.

btw, whichever version of gcc you're using has apparently regressed -
my gcc-4.3.0 manages to work out what's happening.  Hence please prefer
to include the gcc version in the changelog.  And send a bug report at
the gcc guys if appropriate ;)