> On 11 Dec 2015, at 10:52, Dan Carpenter wrote:
>
>> On Fri, Dec 11, 2015 at 10:33:13AM +, Okash Khawaja wrote:
>> Thanks, that's a cleaner approach. Apologies for delayed reply as I'm in
>> middle of a house move. I'll send an updated patch soon.
>
> Someone already fixed it. Thanks.
>
On Fri, Dec 11, 2015 at 10:33:13AM +, Okash Khawaja wrote:
> Thanks, that's a cleaner approach. Apologies for delayed reply as I'm in
> middle of a house move. I'll send an updated patch soon.
>
Someone already fixed it. Thanks.
regards,
dan carpenter
_
Thanks, that's a cleaner approach. Apologies for delayed reply as I'm in middle
of a house move. I'll send an updated patch soon.
Okash
> On 2 Dec 2015, at 13:17, Dan Carpenter wrote:
>
> Probably it's better to create a different struct:
>
> struct udata_s {
>unsigned int size;
>void
Probably it's better to create a different struct:
struct udata_s {
unsigned int size;
void __user *buf;
};
regards,
dan carpenter
___
devel mailing list
de...@linuxdriverproject.org
http://driverdev.linuxdriverproject.org/mailman/list
copy_to_user() and copy_from_user() functions expect the user space
pointers to be marked with __user. Sparse throws following warnings.
For copy_to_user():
warning: incorrect type in argument 1 (different address spaces)
expected void [noderef] *to
got void *buf
For copy_from_user():
warning:
