Re: [Xen-devel] some thoughts about merlot{0|1} issues [was: Re: [xen-unstable test] 102522: tolerable FAIL - PUSHED]

[Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>]

On 11/28/2016 12:19 PM, Dario Faggioli wrote:
> On Mon, 2016-11-28 at 08:48 -0500, Boris Ostrovsky wrote:
>> On 11/24/2016 10:31 AM, Jan Beulich wrote:
>>>  
>>> This indeed looks surprising for a half way modern system - is the
>>> BIOS perhaps limiting C-states (maybe instructed to via BIOS
>>> setup)?
>> IIRC some BIOSes indeed provided an option to disable C2. Dumping
>> SSDT
>> would tell us whether C2 is there is BIOS is not accessible.
>>
> I will try to extract that.
>
> In any case, what I'm wondering is, could this ACPI / PM thing be
> linked to the behavior we are seeing?

I find it somewhat unlikely.

BTW, I wouldn't worry about C2 specifically --- it's pretty clear that
no C-state stats are collected at all:

Nov 28 08:07:42.146057 (XEN) active state:              C-1
Nov 28 08:07:42.153968 (XEN) max_cstate:                C7
Nov 28 08:07:42.153998 (XEN) states:
Nov 28 08:07:42.154021 (XEN)     C1:    type[C1] latency[000] usage[00000000] 
method[ HALT] duration[0]
Nov 28 08:07:42.161992 (XEN)     C0:    usage[00000000] duration[3907882618433]


duration is the only reported value and most likely it's a NOW(). And
the reason C1 is still mentioned is because it is a required C-state,
whether or not it's in SSDT.

It is strange but I don't see anything in the serial log that would
indicate a problem.


>
> I'm not sure I see how... perhaps the system gets too hot and its
> throttled down? (Yes, shooting in the dark, I know.)

I'd expect an event to be triggered (SMI?) and a warning of some sort to
be printed.

-boris

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel