Hi everybody, as always i'm a little bit behind everyone (...) with my work on the ISP1160 driver :-/
I've managed to incorporate Olav's code (especially the one used in ohci-isp1xxx-emu.c) in mine, but there still is something not going well, that keeps me from reaching your status. I'll describe my problem, hoping that someone has a better understanding of my situation: In summary, my HCD never reaches USBOperational state. I have a good IO with my hardware, every R/W command reaches it's target, so I can say that the command that is issued to HcControl to put the HC in Operational Mode works fine. Here is the output of my syslog when loading the driver (part of it): Nov 23 18:45:47 naousa kernel: ohci_write_control: HCCONTROL 00000000->00000083 Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] As you can see, the masking works well, and HCFS is instructed to get the correct value for USBOperational. Then the driver keeps on as described in hc_start() function of ohci-hcd.c and sets up the Root Hub: Nov 23 18:45:47 naousa kernel: hc_start: roothub.status 00000000->00008000 Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: ohci_write_intrenable: Enabling OHCI Interrupts: 8000001a Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D<[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: hc_start: roothub.a 05000010->05000210 Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: hc_start: roothub.status 00008000->00010000 Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D<[EMAIL PROTECTED] Notice the last command, a read operation to HcControl, intended to flush the pci writes as stated in the comments of the source code. Shouldn't this return 0x80? Then the HCD keeps on going and after registering the Root Hub, it goes on to list the contents of the registers: Nov 23 18:45:47 naousa kernel: usb usb3: new device strings: Mfr=3, Product=2, SerialNumber=1 Nov 23 18:45:47 naousa kernel: usb usb3: default language 0x0409 Nov 23 18:45:47 naousa kernel: usb usb3: Manufacturer: Linux 2.6.8 ohci-isp1160_hcd Nov 23 18:45:47 naousa kernel: usb usb3: SerialNumber: 0000:00:0e.0 Nov 23 18:45:47 naousa kernel: usb usb3: hotplug Nov 23 18:45:47 naousa kernel: usb usb3: adding 3-0:1.0 (config #1, interface 0) Nov 23 18:45:47 naousa kernel: usb 3-0:1.0: hotplug Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: usb_probe_interface Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: usb_probe_interface - got id Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: USB hub found Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D<[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D<[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: 2 ports detected Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: standalone hub Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: individual port power switching Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: individual port over-current protection Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: power on to power good time: 10ms Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D<[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: local power source is good Nov 23 18:45:47 naousa kernel: hub 3-0:1.0: enabling power on all ports Nov 23 18:45:47 naousa kernel: __ohci_write_roothub_portstatus: Portstatus[0]: 00000000->00000100 Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: __ohci_write_roothub_portstatus: Portstatus[1]: 00000000->00000100 Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: ohci-isp1160_hcd 0000:00:0e.0: created debug files Nov 23 18:45:47 naousa kernel: ohci-isp1160_hcd 0000:00:0e.0: OHCI controller state Nov 23 18:45:47 naousa kernel: ohci-isp1160_hcd 0000:00:0e.0: OHCI 0.0, NO legacy support registers Nov 23 18:45:47 naousa kernel: A>[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: D<[EMAIL PROTECTED] Nov 23 18:45:47 naousa kernel: ohci-isp1160_hcd 0000:00:0e.0: control 0x003 HCFS=reset CBSR=3 So, the last lines show the status of HcControl. Don't mind the "OHCI 0.0, NO legacy support registers" thing, it was something temporary in reading the HcRevision (and I don't see how it can be connected to my problem). Now, can anybody please tell me why my HC is still in USBReset? And another thing, pretty useless to me until I get the HC to work, but is something that keeps on troubling me from the very first days of my work. Take a look at the manual of ISP1160, paragraph "9.5 HC Operational Model" on page 29. In the 3rd and 4th paragraphs, we read that SOFITLint is issued first, followed by ATLint. Now, take a look at document AN10003-01 Embedded Programming Guide, paragraph "5.4.4 Configuring Interrupts", on page 22. There is a note, just before the end of 5.4.4, that states that we cannot use both ATLint and SOFITLint together. Isn't this controversial to the manual's 9.5 paragraph? What is the correct procedure? If the AN10003-01 document is correct, why did they introduce ATLint in the first place? Anyway, thanks a lot in advance (again...) -- Dimitris Lampridis <[EMAIL PROTECTED]>
signature.asc
Description: This is a digitally signed message part
