On Fri, Dec 04, 2015 at 08:12:33PM +0530, Saurabh Sengar wrote:
> Inorder to notify the user that value is not successfuly set in sys
> entry, error should be returned from store function instead of count
>
> Signed-off-by: Saurabh Sengar <saurabh.truth@xxxxxxxxx>
> ---
> drivers/staging/speakup/kobjects.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/staging/speakup/kobjects.c b/drivers/staging/speakup/kobjects.c
> index fdfeb42..b3a83fb 100644
> --- a/drivers/staging/speakup/kobjects.c
> +++ b/drivers/staging/speakup/kobjects.c
> @@ -640,7 +640,8 @@ ssize_t spk_var_store(struct kobject *kobj, struct kobj_attribute *attr,
> len = E_INC;
> else
> len = E_SET;
> - if (kstrtol(cp, 10, &value) == 0)
> + ret = kstrtol(cp, 10, &value);
> + if (!ret)
> ret = spk_set_num_var(value, param, len);
Both kstrtol() and spk_set_num_var() return -ERANGE. The next lines
expect that if we got -ERANGE, then it came from spk_set_num_var() so
they print a wrong message.
> else
> pr_warn("overflow or parsing error has occurred");
> @@ -688,6 +689,8 @@ ssize_t spk_var_store(struct kobject *kobj, struct kobj_attribute *attr,
>
> if (ret == -ERESTART)
> pr_info("%s reset to default value\n", param->name);
Is this really true?
This function is so weird and broken. Please look at it some more and
fix it harder with a mallet.
regards,
dan carpenter
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
