Re: Strange device init

[Pierre Ossman <drzeus-list@drzeus.cx>]

Adam Belay wrote:

>On Sat, Feb 05, 2005 at 06:10:02PM +0100, Pierre Ossman wrote:
>  
>
>>I'm having problem with a card reader on a laptop (Acer Aspire 1501). 
>>The device doesn't get its resources configured properly.
>>
>>The reader is connected to the LPC bus so there is no standardised way 
>>to configure the device. On other laptops the configuration is done via 
>>ACPI (_STA & co. in the DSDT). On this laptop these functions don't do a 
>>damn thing.
>>In Windows this device gets configured through some other means. It's 
>>not in the driver (I've disected it to confirm this). But under Linux 
>>the device is left unconfigured.
>>
>>So my question is if anyone has any ideas on how this device gets 
>>configured by Windows and possibly how we can get this to work on Linux.
>>
>>The reason this is an issue is that one cannot detect all the quirks of 
>>the hardware so a PNP solution is prefered. In those cases the 
>>manufacturer has chosen resources that work ok.
>>
>>For some context: I am the maintainer of the driver for this hardware. I 
>>have a laptop where the DSDT properly sets up the hardware. The Acer 
>>belongs to some of my users but they are not familiar with the kernel so 
>>I'm trying to fix this for them.
>>
>>Rgds
>>Pierre
>>    
>>
>
>So the device is not listed in the DSDT, or _SRS doesn't work?  Does _STA
>succeed?  Finally have you checked if PnPBIOS detects the device?  Any
>additional information you could provide would be appreciated.
>
>  
>
The device is listed in the DSDT but the functions are empty. E.g. _SRS:

Method (_SRS, 1, NotSerialized)
{
}

_PRS and _CRS at least returns the proper resources that should be used.

PNPBIOS cannot be compiled in. This is a x86_64 machine and PNPBIOS is 
dependent on ISA. Are there any user-space tools that can do the same 
thing? Just to do some testing.

Rgds
Pierre