On Tue, May 31, 2011 at 9:38 PM, DJ Delorie <dj@xxxxxxxxxx> wrote: > >> This is confusing me now... I can only find >> linux-2.6/drivers/net/usb/smsc95xx.c, which definitely is not >> providing a usb-hub (like linux-2.6/drivers/usb/core/hub.c) and also >> not a USB-storage interface. > > http://www.smsc.com/media/Downloads_Public/Data_Sheets/9514.pdf > > See page 6 for a block diagram of the SMSC LAN9514 chip, if that helps > understand the hardware. So, that seems to be one chip, which provides at least presents two USB-devices (one hub with 4-ports for other devices and one ethernet-port). The information can be found in Chapters 1.1.2, 1.1.3 and Chapter 3 which describes the descriptors that can be programmed in the EEPROM. It is possible to unbind the smsc95xx from the ethernet-device (via sysfs). Obviously you could unload the module already (it was does not seem to be compiled in your kernel). Unloading a driver is similar to unbinding the port/device/driver. >From my understanding in an other reply you sent to the list, unloading the smsc95xx module does not make a difference, you seem to be able to reproduce the behavior with different devices (serial-port and usb-mouse). It seems that the smsc95xx module can not be kept responsible for the bad performance of your drive. Other points that might investigation, include the USB-controller on your PandaBoard (ehci, ohci, ...) or possibly the USB-storage-stack. It might well be that there is an issue around the scheduling of the interrupt-handling done by the USB-controller... Note that USB-mouses tend to be USB-1.x and USB-storage is usually USB-2.x. Different USB-versions are normally handled by different USB-controllers and therefore have their own drivers (for example, ehci = USB-2, uhci = USB-1). /proc/interrupts might point you to some more ideas. Possibly the USB2-controller uses a different interrupt than the USB1-controller and the serial-port. Depending the driver used for your USB2-controller, you might have some module-parameters that influence interrupt-handling: $ modinfo ehci-hcd filename: /lib/modules/2.6.38.3-kw/kernel/drivers/usb/host/ehci-hcd.ko alias: platform:orion-ehci license: GPL author: David Brownell description: USB 2.0 'Enhanced' Host Controller (EHCI) Driver alias: pci:v*d*sv*sd*bc0Csc03i20* depends: usbcore vermagic: 2.6.38.3-kw preempt mod_unload ARMv5 parm: log2_irq_thresh:log2 IRQ latency, 1-64 microframes (int) parm: park:park setting; 1-3 back-to-back async packets (uint) parm: ignore_oc:ignore bogus hardware overcurrent indications (bool) parm: hird:host initiated resume duration, +1 for each 75us (int) HTH, Niels _______________________________________________ arm mailing list arm@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/arm
