On Sat, 14 Apr 2012, Martin Mokrejs wrote: > First: > > Why is kernel trying to reset devices connected to the hub. In my eyes, kernel > was trying hard to do the reset. Otherwise, what these lines mean? Why would the external > hub ask kernel to do its own reset? The hub didn't ask the kernel to do anything. When the hub disabled the connection to the keyboard, the usbhid driver was no longer able to communicate with the keyboard hardware. Consequently the driver asked the kernel to reset the keyboard. I don't know why the resets failed; for that I'd have to see a usbmon trace with the keyboard plugged in. Maybe the hub wasn't responding to the reset requests. > And why is the "EMI?" line in the middle of the looping cycle? Doesn't this show > kernel is at fault? The kernel polls each hub about four times per second for port status changes. The first poll after the port was disabled (which was after a whole bunch of failed resets) told the kernel that the hub had disabled the keyboard's port. That's when the "EMI" line got logged. > No, I don't have any other hub, but could buy some, sure. But is always a gamble > which one is really a good one. :( True enough. I don't have any particular recommendations. > generic-usb 0003:04F2:0116.0007: can't reset device, 0000:00:1d.0-1.2.1/input0, status -71 > generic-usb 0003:046D:C050.0008: can't reset device, 0000:00:1d.0-1.2.4/input0, status -71 ... If you enable CONFIG_PRINTK_TIME then these lines will have timestamps. You'll be able to see how often the resets are attempted, and how much time passes between the start of the resets and the "EMI" message. > hub 2-1.2:1.0: port 1 disabled by hub (EMI?), re-enabling... > usb 2-1.2.1: USB disconnect, device number 13 > generic-usb 0003:046D:C050.0008: can't reset device, 0000:00:1d.0-1.2.4/input0, status -71 > generic-usb 0003:046D:C050.0008: can't reset device, 0000:00:1d.0-1.2.4/input0, status -71 ... The usbhid driver keeps trying to reset the keyboard until the USB core tells it that the keyboard has been disconnected. > Second: > > How can I enable my eSATA port? I did not reboot yet, so how can I dump current status of > the eSATA port which does not recognize any new devices. I hope this is a driver issue > and not a hardware issue. So if I can provide some debug info for anybody I would be glad > if that helps to find out why the eSATA port claims to be dead completely. Sorry, I can't help with that -- I don't know anything about it. You should try posting on a different mailing list, such as linux-ide@xxxxxxxxxxxxxxx or LKML. 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
