On Sat, 16 Jul 2016, Bruce Korb wrote: > On Sat, Jul 16, 2016 at 4:18 PM, Greg KH <greg@xxxxxxxxx> wrote: > >> [13043.528023] usb 1-1.3: Product: Gaming Mouse G400 > >> [13043.528027] usb 1-1.3: Manufacturer: Logitech > >> [13043.528309] usb 1-1.3: Not enough bandwidth for new device state. > >> [13043.528319] usb 1-1.3: can't set config #1, error -28 > > > > That's your issue, odds are OS-X doesn't check the bandwidth needed for > > the devices and tries to use them anyway. We used to do that on Linux > > many years ago, but had numerous problems with devices so we now follow > > the specification. It's better to have your device not work at all, > > than to just randomly not work, don't you think? > > That sounds right, but here we have a situation where a keyboard and a > mouse will overwhelm the bandwidth, but a Class-10/UDH-1 SSD will not. > I really don't type that fast. I should be able to tell the driver > that I don't type several billion words per minute or move my hand > faster than sound, yes? OK, I'm being a little facetious, but it does > seem to me that if a HID device is being used, there should be a way > to downgrade the claim that it is a "full speed USB device". > > In any case, "better to" question, in my case, it's "no". I do not > like having three keyboards stacked one on top of another. So I > really need some way to fix this. :( That error message is clearly bogus. The problem is that it comes from the xHCI hardware in your computer, not from the Linux kernel. In the past, there was a report of a similar problem. An xHCI controller reported insufficient bandwidth when Link Power Management was enabled, but not when it wasn't. I don't recall if this was ever resolved in a useful way. What kernel version are you using? Alan Stern -- 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
