Dan Carpenter <dan.carpenter@xxxxxxxxxx> writes:
> My static checker complains that:
>
> drivers/char/hw_random/core.c:341 hwrng_register()
> warn: we tested 'old_rng' before and it was 'false'
>
> The problem is that sometimes we test "if (!old_rng)" and sometimes we
> test "if (must_register_misc)". The static checker knows they are
> equivalent but a human being reading the code could easily be confused.
>
> I have simplified the code by removing the "must_register_misc" variable
> and I have removed the redundant check on "if (!old_rng)".
>
> Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
Yeah, clearer too.
Reviewed-by: Rusty Russell <rusty@xxxxxxxxxxxxxxx>
Thanks,
Rusty.
>
> diff --git a/drivers/char/hw_random/core.c b/drivers/char/hw_random/core.c
> index b9495a8c05c6..463382036a01 100644
> --- a/drivers/char/hw_random/core.c
> +++ b/drivers/char/hw_random/core.c
> @@ -301,7 +301,6 @@ err_misc_dereg:
>
> int hwrng_register(struct hwrng *rng)
> {
> - int must_register_misc;
> int err = -EINVAL;
> struct hwrng *old_rng, *tmp;
>
> @@ -326,7 +325,6 @@ int hwrng_register(struct hwrng *rng)
> goto out_unlock;
> }
>
> - must_register_misc = (current_rng == NULL);
> old_rng = current_rng;
> if (!old_rng) {
> err = hwrng_init(rng);
> @@ -335,13 +333,11 @@ int hwrng_register(struct hwrng *rng)
> current_rng = rng;
> }
> err = 0;
> - if (must_register_misc) {
> + if (!old_rng) {
> err = register_miscdev();
> if (err) {
> - if (!old_rng) {
> - hwrng_cleanup(rng);
> - current_rng = NULL;
> - }
> + hwrng_cleanup(rng);
> + current_rng = NULL;
> goto out_unlock;
> }
> }
--
To unsubscribe from this list: send the line "unsubscribe kernel-janitors" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
