On Wed, Mar 5, 2014 at 12:18 PM, Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote:
> On Wed, 5 Mar 2014, Dan Williams wrote:
>
>> Please have a look at the following. I'm sure Greg will be happy that
>> we are killing these bugs before they become fodder for -stable. Thanks
>
> Wow, do you really think all these changes will ever go into -stable?
> I'm doubtful; they're rather large.
Heck no :-). I meant the later fixes for these bugs that we're
squashing at review time.
>> for the review! The other change I made was to convert
>> find_default_peer to find_and_link_peer per your other comments about
>> centralizing peering on patch 6.
>
> Good idea, since it now finds all peers and not just the default ones.
>
>> 8<---------
>> Subject: usb: assign default peer ports for root hubs
>>
>> From: Dan Williams <dan.j.williams@xxxxxxxxx>
>>
>> Assume that the peer of a superspeed port is the port with the same id
>> on the shared_hcd root hub. This identification scheme is required of
>> external hubs by the USB3 spec [1]. However, for root hubs, tier mismatch
>> may be in effect [2]. Tier mismatch can only be enumerated via platform
>> firmware. For now, simply perform the nominal association.
>>
>> A new lock 'usb_port_peer_mutex' is introduced to synchronize port
>> device add/remove with peer lookups. It protects peering against
>> changes to hcd->shared_hcd, hcd->self.root_hub, hdev->maxchild, and
>> port_dev->child pointers.
>>
>> [1]: usb 3.1 section 10.3.3
>> [2]: xhci 1.1 appendix D
>>
>> Cc: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>
>> [alan: usb_port_peer_mutex locking scheme]
>> Signed-off-by: Dan Williams <dan.j.williams@xxxxxxxxx>
>
>> diff --git a/drivers/usb/core/port.c b/drivers/usb/core/port.c
>> index 531a591a7b1e..0d36610af156 100644
>> --- a/drivers/usb/core/port.c
>> +++ b/drivers/usb/core/port.c
>> @@ -151,9 +151,66 @@ static struct device_driver usb_port_driver = {
>> .owner = THIS_MODULE,
>> };
>>
>> +static void link_peers(struct usb_port *left, struct usb_port *right)
>> +{
>> + if (left->peer == right && right->peer == left)
>> + return;
>> +
>> + if (left->peer || right->peer) {
>> + struct usb_port *lpeer = left->peer;
>> + struct usb_port *rpeer = right->peer;
>> +
>> + WARN(1, "failed to peer %s and %s (%s -> %p) (%s -> %p)\n",
>> + dev_name(&left->dev), dev_name(&right->dev),
>> + dev_name(&left->dev), lpeer,
>> + dev_name(&right->dev), rpeer);
>> + return;
>> + }
>> +
>> + left->peer = right;
>> + right->peer = left;
>> +}
>> +
>> +static void unlink_peers(struct usb_port *left, struct usb_port *right)
>> +{
>> + WARN(right->peer != left || left->peer != right,
>> + "%s and %s are not peers?\n",
>> + dev_name(&left->dev), dev_name(&right->dev));
>
> At some point, maybe we'll have enough confidence in the locking to
> remove all these checks. :-)
>
> Acked-by: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>
Cheers.
--
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
