On 27/02/15(Fri) 12:07, Stefan Sperling wrote: > On Thu, Feb 26, 2015 at 10:45:01PM +0100, Stefan Sperling wrote: > > There are other issues which are not fixed by this diff and which > > my diff did not fix either. > > > But these are bugs we can investigate another day. Happy travels! > > I suppose there might well be power supply issues when chaining too many > passive hubs and devices off a single USB port on the host controller. > I've now put an active USB hub (separate power source) at the end of > my long-cable hub and so far things are looking good. I'll report back > if my problems appear in this configuration, too. > > My setup changed from > > [ehci host]<--[usb extension cable hub]<--[usb keyboard (+hub)]<--[usb mouse] > > to this: > [power adapter] > | > V > [ehci host]<--[usb extension cable hub]<--[active hub]<--[usb keyboard (+hub)] > ^ > | > +--[usb mouse]
Well there's an obvious "bug" to fix in our stack regarding HUB power sources, I don't know if that's the one you're hitting but we can try :) When a device is attached the stack gets its power usage from the configuration descriptor. But the logic to check if the hub this device is attached to has enough power budget is dumb and does not work well with multiple devices below the same hub. Instead of making a guess about how much power can be drained by a port we should really keep track of the power available per hub. If you're curious grep for 'powersrc' in sys/dev/usb. I have this on my TODO list because SuperSpeed devices do things slightly differently. I'll try to cook a diff, but it will be for after 5.7.
