Re: [Xenomai-help] Xenomai on i7-870

[Jakub Nowacki <j.s.nowacki@domain.hid>]

On 17/06/2011 20:00, Gilles Chanteperdrix wrote:
> On 06/17/2011 04:05 PM, Jakub Nowacki wrote:
>>>
>>>
>>> I've added my ID to the table in smi.c file as:
>>
>> {PCI_DEVICE(PCI_VENDOR_ID_INTEL,
>> PCI_DEVICE_ID_INTEL_5_3400_SERIES_LPC_MIN+0xa)}
>>
>> Everything complies OK but during the startup I get message:
>>
>> [    2.137733] Xenomai: SMI-enabled chipset found
>> [    2.137744] Xenomai: SMI workaround failed!
>>
>> Should I do something extra apart from adding it to the table?
>
> There are two reasons why it may not work:
> - the way to globally disable SMIs has changed with the chipset version
> you are using;
> - the BIOS of your PC locks the SMI disabling bit.
>
> So, the next step is to check the chipset datasheet.
>
Following the datasheet available at:

http://www.intel.com/Assets/PDF/datasheet/322169.pdf

SMI_EN—SMI Control and Enable Register has address: PMBASE + 30 (see p. 
531), which seems to be the same as in 'smi.c' (PMBASE is 0x40-0x43; see 
p. 460). Structure seems to be the same, namely, GBL_SMI_EN_BIT is the 
first bit. However, SMI enable can be locked using SMI_LOC bit in 
GEN_PMCON_1—General PM Configuration 1 Register
(PM—D31:F0) with Offset Address: A0h; see p. 513. Hence, this one can 
potentially stop 'smi.c' from applying the workaround. Sorry for 
slightly descriptive approach but I don't know 'smi.c' code that well to 
fiddle with it.

Hope it helps slightly.

Best wishes,

Jakub