Re: tlp0: unable to map device registers

["Emery Clark" <emdog4%gmail.com@localhost>]

Quentin hi,

I do not understand what you mean? Could you please explain this a
little more. I am very interested.

thanks!!
Emery

On Thu, Sep 18, 2008 at 10:56 AM, Quentin Garnier <cube%cubidou.net@localhost> 
wrote:
> On Wed, Sep 10, 2008 at 09:56:00PM -0500, Jonathan A. Kollasch wrote:
>> On Wed, Sep 10, 2008 at 09:35:59PM -0400, Emery Clark wrote:
>> > On Tue, Sep 9, 2008 at 2:51 PM, David Young <dyoung%pobox.com@localhost> 
>> > wrote:
>> > > On Tue, Sep 09, 2008 at 02:31:49PM -0400, Emery Clark wrote:
>> > >> Hi all!
>> > >>
>> > >> I am having problems with a NetBSD install on an old system of mine.
>> > >>
>> > >> First, I cannot get the operating system to configure my Ethernet
>> > >> card. The kernel recognizes it as a "Davicom DM9102A Ethernet" and
>> > >> attempts to assign the tlp (Tulip) drivers to the hardware device.
>> > >> Although the Tulip drivers seem to be very well known (and compiled
>> > >> into the GENERIC kernel?) I cannot configure the device.
>> > >>
>> > >> Dmesg gives me:
>> > >> "tlp0 at pci0 dev 5 function 0: DAvicom DM9102A Ethernet, pass 3.1
>> > >> tlp0: unable to map device registers"
>> > >
>> > > Maybe the BIOS has not correctly initialized either the BARs, the memory-
>> > > or the I/O-enable.
>> > >
>> > > What does 'pcictl pci0 dump -d 5' say?  Also, how about 'pcictl pci0
>> > > dump -d 0' ?
>> > >
>> >
>> > Sorry for the delayed response.
>> >
>> > $ pcictl pci0 dump -d 5
>> > PCI configuration registers:
>> >   Common header:
>> >     0x00: 0x91021282 0x02100000 0x02000031 0x00004200
>> >
>> >     Vendor Name: Davicom Semiconductor (0x1282)
>> >     Device Name: DM9102 10/100 Ethernet (0x9102)
>> >     Command register: 0x0000
>> >       I/O space accesses: off
>> >       Memory space accesses: off
>>
>> Both of these being off will cause the 'unable to map'.
>>
>> >       Bus mastering: off
>>
>> This should probably be on, so set the BIOS accordingly
>>
>>
>> If the BIOS has a PnP OS option, try setting it to non-PnP.
>> Hopefully that will allow it to work.
>
> We should just set the flags from the driver.  Apparently the BIOS has
> configured the spaces fine.
>
> If the BIOS hasn't configured the device, we'll try to map garbage
> instead of not even trying to map.  I think it'd take a rather evil
> BIOS to make do something incorrect.
>
> As a matter of fact, a lot of drivers explicitely set access to spaces
> and bus mastering.
>
> --
> Quentin Garnier - cube%cubidou.net@localhost - cube%NetBSD.org@localhost
> "See the look on my face from staying too long in one place
> [...] every time the morning breaks I know I'm closer to falling"
> KT Tunstall, Saving My Face, Drastic Fantastic, 2007.
>



-- 
http://emdog4.googlepages.com