Le Tue, 26 Jul 2011 13:03:16 +0300, Felipe Balbi <balbi@xxxxxx> a écrit : > > These requirements are identical in consequence. > > So you have a budget per hub of 700mA, which translates > > into 500mA per port (limit from the spec) > > But as soon as you configure a device to actually use > > 500mA the limit per port would have to drop to 200mA, correct? > > from what I understood, 700mA is for the whole thing, so if he > configures a 500mA on port A, port B only has 200mA to spare. If he > configures a 400mA on port B, port A has only 300mA to spare. Thomas ? Exact. What Olivier proposes has the same final effect but is implemented differently: he proposes to dynamically adjust the allowed per-port limit as devices gets connected/disconnected. For things to be really clear, I think it'd be better to distinguish : * The per-hub maximum current available * The per-port maximum current available By default, the per-port maximum current available is 500 mA and the per-hub maximum current available is nb_ports * 500 mA. In my case however, the per-port maximum current available is 500 mA but the per-hub maximum current is 700 mA. Then, when a device gets connected, both the following conditions are checked : * Does the device requests an amount of current that exceeds the per-port maximum current ? * Does the device requests an amount of current that exceeds the remaining amount of power available at the level of the hub, considering all devices already connected to this hub ? If one of those conditions are true, then the device is rejected. Would such a thing make sense ? Regards, Thomas -- Thomas Petazzoni, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com -- 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
