Re: PCI expansion card working?

Hi! Tim,


From: Tim Rightnour <[email protected]>
Date: Mon, 21 Jan 2008 20:41:08 -0700 (MST)

> Some bridge cards get *really* pissed when you probe past device 8 on them.  In
> this case yours starts returning made up values.  There is code in prep to deal
> with this on certain known cards, I think we need to add your card to that list.
> 
> look at prep/pci/pci_machdep.c:prep_pci_bus_maxdevs().  see how it looks for
> prep-pcibus-maxdevices?  Now look at the bottom of that file, you will see
> where we set that for some crazy dec card.  Do the same for yours.

I added it as follows. And, it enabled the start with the vga console. 
# Because CVS became a connection timeout, I was not able to obtain diff. 

prep/pci/pci_machdep.c::prep_pci_conf_hook() ---

                /* now look for bus quirks */
 
                if ((PCI_VENDOR(id) == PCI_VENDOR_DEC &&
                    PCI_PRODUCT(id) == PCI_PRODUCT_DEC_21154) ||
                    (PCI_VENDOR(id) == PCI_VENDOR_HINT &&
                    PCI_PRODUCT(id) == PCI_PRODUCT_HINT_HB1)) {
                        bmax = prop_number_create_integer(8); 
                        KASSERT(bmax != NULL);

--- prep/pci/pci_machdep.c::prep_pci_conf_hook()


However, I saw a stranger message.  This correction disabled MEM/IO of
ppb.


ppb0 at pci0 dev 22 function 0: PCI configuration registers:
  Common header:
    0x00: 0x00213388 0x02900144 0x06040013 0x0001ff08

    Vendor Name: HiNT (0x3388)
    Device Name: HB1 PCI-PCI Bridge (0x0021)
    Command register: 0x0144
      I/O space accesses: off			<-- off ?
      Memory space accesses: off		<-- off ?
      Bus mastering: on
      Special cycles: off
      MWI transactions: off
      Palette snooping: off
      Parity error checking: on
      Address/data stepping: off
      System error (SERR): on
      Fast back-to-back transactions: off
      Interrupt disable: off

 ... snip ...


HiNT HB1 PCI-PCI Bridge (PCI bridge, revision 0x13) at ? dev 22 function 0 (intr
swiz 0, intrpin 0, i/o off, mem off, no quirks): HiNT HB1 PCI-PCI Bridge (rev. 0x13)
pci1 at ppb0 bus 1
pci1: no spaces enabled!			<--- Oops...



However, I saw IO/MEM of ppb enable in yesterday's log. 

yesterday's log ---

ppb0 at pci0 dev 22 function 0: PCI configuration registers:
  Common header:
    0x00: 0x00213388 0x02900147 0x06040013 0x0001ff08

    Vendor Name: HiNT (0x3388)
    Device Name: HB1 PCI-PCI Bridge (0x0021)
    Command register: 0x0147
      I/O space accesses: on
      Memory space accesses: on

--- yesterday's log


Should I compulsorily make MEM/IO of this ppb enable?

Thanks,
--
kiyohara