Messages in this thread | | | Date | Tue, 24 May 2011 10:20:37 -0400 (EDT) | From | Alan Stern <> | Subject | RE: [PATCH v12 7/8] usb: Adding SuperSpeed support to dummy_hcd |
| |
On Tue, 24 May 2011, Tanya Brokhman wrote:
> Hi Felipe > > > > Yes :) The driver->speed is updated in usb_composite_probe() if > > > CONFIG_USB_GADGET_SUPERSPEED is defined. > > > > > > So, are we ok with this solution? The module parameter I mean? > > > Are you going to try the v13 in your branch? Please let me know how > > it > > > goes and of course if you have any comments. > > > > I think it still gives the possibility for failure. I would rather not > > take that until all gadget drivers are fixed. We can help you doing > > that and we only change driver->speed after all gadget drivers have > > their "sensible defaults" SuperSpeed descriptors. > > By "until all gadget drivers are fixed" you mean until all gadget drivers > provide SS descriptors? This will take for ever... > I wasn't about to modify all gadget drivers and to add SS descriptors for > them. I can add default values (as generate_ss_descriptors() did if you > remember) but I don't think this is the right approach because as you said - > different gadget drivers might have different SS descriptors and I don't > feel confident enough to set these values. Nor do I have the ability to test > each of the gadget drivers the way I would like to after this change. > The only gadget driver I felt confident adding SS descriptors for is UASP, > which I tested properly. > > Actually if the CONFIG_USB_GADGET_SUPERSPEED is turned off, which is the > default of it, the speed won't be updated and all these series won't be > functional so I don't see any possibilities for failure in such > configuration. Or am I missing something?
dummy_hcd should work when CONFIG_USB_GADGET_SUPERSPEED is enabled, even if the usb_gadget_driver structure is initialized with the speed field set to USB_SPEED_HIGH. This will be true for all the standalone gadget drivers until they are updated.
Which leaves a question about the composite gadget framework. Should it be updated with SS support? Probably not until the various function drivers have all been updated.
Alan Stern
| |