problem with attaching device to LPC bus

Sepherosa Ziehau sepherosa at gmail.com
Tue Aug 23 18:44:04 PDT 2016


Code itself looks correct, can you post the full output of devinfo -r?

Thanks,
sephe

On Tue, Aug 23, 2016 at 9:23 PM, karu.pruun <karu.pruun at gmail.com> wrote:
> Hello
>
> This is a newbie question: I am trying to write a simple DragonFly kernel
> module for the gmux device that is attached to the LPC bus. Probing works
> fine but I am stuck with the attachment routine: I can't allocate IO ports
> for the device.
>
> The device shows up in the device tree but is not attached:
>
> ---
> # devinfo -rv | grep GMUX:
>
> unknown pnpinfo _HID=APP000B _UID=0 at handle=\_SB_.PCI0.LPCB.GMUX
> ---
>
> I looked up acpi tables (acpidump etc), the IO resources for GMUX seem to be
> there:
>
> --- ---
>     Scope (\_SB.PCI0.LPCB)
>     {
>         Device (GMUX)
>         {
>             Name (_HID, EisaId ("APP000B"))  // _HID: Hardware ID
>             Name (_CID, "gmux")  // _CID: Compatible ID
>             Name (_STA, 0x0B)  // _STA: Status
>             Name (_CRS, ResourceTemplate ()  // _CRS: Current Resource
> Settings
>             {
>                 IO (Decode16,
>                     0x0700,             // Range Minimum
>                     0x07FF,             // Range Maximum
>                     0x01,               // Alignment
>                     0xFF,               // Length
>                     )
>             })
> --- ---
>
> Also, looking devinfo -rv shows that the range 0x0700 to 0x07FF has been set
> aside by the kernel:
>
> ---
> # devinfo -u
>
> . . .
> I/O ports:
>     . . .
>     0x400 - 0x1fff (root0)
>     . . .
> ---
>
> So am I right supposing that bus_alloc_resource_any should be able to claim
> the range 0x700 to 0x7FF for the gmux device? At the moment, trying to
> allocate IO ports with bus_alloc_resource returns NULL.
>
> I wonder if anyone has suggestions about what has gone wrong or where to dig
> further? I attach the code.
>
> Thanks
>
> Peeter
>
> --
>



-- 
Tomorrow Will Never Die



More information about the Kernel mailing list