Re: [PATCH] drivers/usb/gadget: add missing kfree calls

[Connor H <cmdkhh@gmail.com>]

On Sat, Jun 4, 2011 at 2:54 PM, Andre Bartke
<andre.bartke@googlemail.com> wrote:
> On Sat, 04 Jun 2011 23:35:16 +0200
> "Michal Nazarewicz" <mina86@mina86.com> wrote:
>
>> On Sat, 04 Jun 2011 21:34:54 +0200, Andre Bartke
>> <andre.bartke@googlemail.com> wrote:
>> > diff --git a/drivers/usb/gadget/rndis.c b/drivers/usb/gadget/rndis.c
>> > @@ -700,6 +701,7 @@ static int rndis_reset_response(int configNr,
>> > rndis_reset_msg_type *buf)
>> >     resp->AddressingReset = cpu_to_le32(1);
>> >     params->resp_avail(params->v);
>> > +   kfree(r);
>> >     return 0;
>> >  }
>>
>> Have you tested this?  It does not look right to me.
>> rndis_add_response() allocates memory and puts the request on a
>> list.  It is later freed in rndis_free_response().
>>
>
> Nope, the only way I can test this is a build - which it does.
> You might be right though.

building != correct

r is being added to a list in rndis_add_response
Andre is right, rndis_free_response handles list removal

982 if (r && r->buf == buf) {
983                         list_del(&r->list);
984                         kfree(r);
985                 }

in fact all of the kfree calls I believe are incorrect.

Connor

> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>