Re: [Openipmi-developer] ipmi_si feature request: SMBIOS-based autoloading

[Jean Delvare]

Le Tuesday 26 January 2016 Ã 09:29 -0800, Andy Lutomirski a Ãcrit :
> On Tue, Jan 26, 2016 at 5:43 AM, Corey Minyard <minyard@xxxxxxx> wrote:
> >
> > On 01/26/2016 07:32 AM, Corey Minyard wrote:
> >>
> >> On 01/24/2016 07:45 PM, Andy Lutomirski wrote:
> >>> Alternatively, maybe /sys/firmware/dmi could learn how to advertise
> >>> modaliases.  But that might be a giant mess to solve a tiny problem.
> >>
> >> This sounds like the right way, but you are probably right.  Are
> >> there any other resources that could benefit from this?  I"m
> >> guessing not.
> 
> No clue.  Jean might know.  Jean?

In principle I agree it would be the right thing to do.

Now in practice I'm afraid the interest would be limited, because 1* on
so many occasions the DMI data is wrong and 2* the level of details that
can be encoded in the DMI table data is fairly limited. The IPMI records
may be an exception there because IPMI is only found on high-end
hardware and this is where in general the DMI data is correct. But for
most other records, you want to trust the hardware itself more than the
DMI data.

So you only want to rely on DMI data for information you can't get in
any other way. That being said, advertising a modalias is something
fairly safe. The driver getting loaded is free to use or not use the DMI
data.

And even if the only modalias in /sys/firmware/dmi is going to be for
IPMI, I see no reason for not doing that, if it serves the purpose.

-- 
Jean Delvare
SUSE L3 Support