On Tue, Mar 05, 2013 at 04:40:12PM +0800, Peter Chen wrote:
> On Fri, Mar 01, 2013 at 03:42:26PM +0100, Michael Grzeschik wrote:
> > Instead of having a limited number of usable tds in the udc we use a
> > linked list to support dynamic amount of needed tds for all special
> > gadget types. This improves throughput.
> >
> > This patch also adresses a possible momory leak in _ep_nuke found
> > while porting the request handling to an linked list.
> >
> > - The call of _ep_nuke can lead to an memory leak, if it is called on an
> > endpoint with currently dynamic allocated tds. This was aswell a problem
> > in the special case before the dynamic td handling was implemented and
> > an zero length packet was added to mark the end of the transfer.
>
> It is another problem, better have another commit, besides, can you
> describe more detail?
Its possible that we add one extra request to mReq->zptr with
dma_pool_alloc in hardware_enqueu, but that one will never be released
in ep_nuke. It makes not much sense to move that fix into an extra patch
in the first place, as the codepath will be changed by this patch
anyway. So i will fix that comment above to be more clear.
> > +
> > +static int add_td_to_list(struct ci13xxx_ep *mEp, struct ci13xxx_req *mReq, unsigned length)
> > +{
> > + struct td_node *lastnode, *node = kzalloc(sizeof(struct td_node), GFP_ATOMIC);
>
> Better split above one line to two lines like below "dma_pool_alloc"
I will do that.
--
Pengutronix e.K. | |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
