On Fri, Nov 19, 2010 at 04:52:40PM -0800, Greg KH wrote: > On Fri, Nov 19, 2010 at 04:49:40PM -0800, Sarah Sharp wrote: > > Hi Greg, > > > > Here's three more bug fixes that should be queued for 2.6.37. > > > > Paul's patch fixes an issue he had with the xHCI driver while unplugging > > and replugging in USB devices a bunch of times. > > Is this a regression? Yes, Paul's patch fixes a regression introduced in 2.6.35 (noted if you read the bit about which stable trees to apply it to). It should definitely go into 2.6.37. > > The last two patches partially solve Don Zickus' issues with a broken USB > > 3.0 device not surviving system suspend. With these two patches, it > > comes back as a SuperSpeed device, and that's an improvement over being > > stuck as a High Speed device. However, USB persist doesn't work because > > the device disconnects from SS on system resume and reconnects on the SS > > port after it's reset as a HS device. I'm going to try to work up > > another patch to fix the USB persist issue next week, but please take > > these two patches for now. > > Are these really needed now for .37? Or could they wait? The last two patches solve a particularly nasty bug that people will run into when they try to test the newly added xHCI suspend support. No oopses, but they'll loose the ability to have *any* USB 3.0 devices connect at SuperSpeed for any port this happens to. They'll only regain the ability for devices to connect at SuperSpeed once they reboot. I think this will cause great confusion for users. But if you think it's too big, it's probably ok to wait for 2.6.38. Let me know if you need me to shuffle patches around. Sarah -- 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
