Re: [PATCH] USB: Driver for Freescale QUICC Engine USB Host Controller

[Anton Vorontsov <avorontsov@ru.mvista.com>]

On Tue, Dec 23, 2008 at 01:28:40PM -0800, Andrew Morton wrote:
> On Wed, 24 Dec 2008 00:03:22 +0300
> Anton Vorontsov <avorontsov@ru.mvista.com> wrote:
> 
> > This patch adds support for the FHCI USB controller, as found
> > in the Freescale MPC836x and MPC832x processors. It can support
> > Full or Low speed modes.
> > 
> > Quite a lot the hardware is doing by itself (SOF generation, CRC
> > generation and checking), though scheduling and retransmission is on
> > software's shoulders.
> > 
> > This controller does not integrate the root hub, so this driver also
> > fakes one-port hub. External hub is required to support more than
> > one device.
> > 
> 
> <quick scan>
> 
> Nice-looking driver.  But the namespace pollution is tremendous!

Thanks for the review, Andrew.

Name-space pollution fixed.

[..]
> > +static int fhci_mem_init(struct fhci_hcd *fhci)
> > +{
> > +	int i;
> > +	int error = 0;
> > +
> > +	fhci->hc_list = kzalloc(sizeof(*fhci->hc_list), GFP_KERNEL);
> > +	if (!fhci->hc_list)
> > +		return -ENOMEM;
> > +
> > +	INIT_LIST_HEAD(&fhci->hc_list->ctrl_list);
> > +	INIT_LIST_HEAD(&fhci->hc_list->bulk_list);
> > +	INIT_LIST_HEAD(&fhci->hc_list->iso_list);
> > +	INIT_LIST_HEAD(&fhci->hc_list->intr_list);
> > +	INIT_LIST_HEAD(&fhci->hc_list->done_list);
> > +
> > +	fhci->vroot_hub = kzalloc(sizeof(*fhci->vroot_hub), GFP_KERNEL);
> > +	if (!fhci->vroot_hub)
> > +		return -ENOMEM;
> 
> Did we leak fhci->hc_list?

Apparently. Fixed.

[...]
> > +struct ed *get_empty_ed(struct fhci_hcd *fhci)
> > +{
> > +	struct ed *ed;
> > +
> > +	if (!list_empty(&fhci->empty_eds)) {
> > +		ed = list_entry(fhci->empty_eds.next, struct ed, node);
> > +		list_del(fhci->empty_eds.next);
> > +	} else {
> > +		ed = kmalloc(sizeof(*ed), GFP_ATOMIC);
> > +		if (!ed)
> > +			fhci_err(fhci, "No memory to allocate to ED\n");
> > +		else
> > +			init_ed(ed);
> > +	}
> > +
> > +	return ed;
> > +}
> 
> The GFP_ATOMICs here are regrettable.  Are these functions ever called
> from a context in which a more reliable allocation mode can be used?
> If so, the caller should pass in the gfp_t.

No, these are called with a spinlock held. But we don't normally
allocate things via this function, instead we pre-allocate the eds
and tds in the fhci_mem_init, so the kmalloc above is the last resort
when no empty tds/eds left in the appropriate lists (normally should
not happen).

Thanks,

-- 
Anton Vorontsov
email: cbouatmailru@gmail.com
irc://irc.freenode.net/bd2