"usb device not accepting address" caused by IFP-driver (IRiver) interrupt?

Mitchua · 09-24-2004, 09:14 PM

I have an IRiver iFP-790 (on USB 2-1) that I'm trying to use in Linux kernel 2.6.7. I'm connecting the device on a USB card that is also being used by my USB Microsoft Intellimouse (on USB 2-2).

I'm using the IFP-driver from http://ifp-driver.sourceforge.net/ . Whenever I run an "ifp" command to read/write from the IRiver device, the operation works correctly (i.e. I can upload files successfully to the device) but as the command finishes, my mouse freezes up and the usb "device not accepting address X". From /var/log/messages:

Code:

Sep 24 21:19:18 athlon usb 2-1: new full speed USB device using address 3
Sep 24 21:19:18 athlon usb.agent[8720]: ... no modules for USB product 4102/1007/1
Sep 24 21:21:09 athlon usb 2-1: usbfs: USBDEVFS_CONTROL failed cmd ifp rqt 192 rq 2 len 4 ret -75
Sep 24 21:21:14 athlon usb 2-1: control timeout on ep0out
Sep 24 21:21:19 athlon usb 2-1: USB disconnect, address 3
Sep 24 21:24:06 athlon usb 2-2: USB disconnect, address 2
Sep 24 21:24:14 athlon usb 2-2: new low speed USB device using address 4
Sep 24 21:24:19 athlon usb 2-2: control timeout on ep0out
Sep 24 21:24:25 athlon usb 2-2: control timeout on ep0out
Sep 24 21:24:25 athlon usb 2-2: device not accepting address 4, error -110
Sep 24 21:24:25 athlon usb 2-2: new low speed USB device using address 5
Sep 24 21:24:30 athlon usb 2-2: control timeout on ep0out
Sep 24 21:24:35 athlon usb 2-2: control timeout on ep0out
Sep 24 21:24:36 athlon usb 2-2: device not accepting address 5, error -110
Sep 24 21:29:03 athlon usbcore: registered new driver usb-storage
Sep 24 21:39:51 athlon usbcore: registered new driver usbfs
Sep 24 21:39:51 athlon usbcore: registered new driver hub
Sep 24 21:39:52 athlon usb 2-2: new low speed USB device using address 2
Sep 24 21:39:53 athlon usbcore: registered new driver hiddev
Sep 24 21:39:53 athlon usbcore: registered new driver usbhid

I saw in http://www.linuxquestions.org/questi...ghlight=iriver that people were patching their 2.6.* kernels so they could mount the Iriver, but I'm content to use the IFP-driver. Is this strange behaviour being caused by an interrupt that's being sent to the IRiver and USB Mouse?

Thank you, thank you, thank you for any help you can give me.

--Mitchua