Re: Problems with merlot* AMD Opteron 6376 systems (Was Re: stable trees

[Boris Ostrovsky <boris.ostrovsky@oracle.com>]

On 06/29/2015 06:23 AM, Ian Campbell wrote:
> On Fri, 2015-06-26 at 21:07 +0100, Ian Campbell wrote:
>> On Fri, 2015-06-26 at 15:36 -0400, Boris Ostrovsky wrote:
>>> On 06/26/2015 10:52 AM, Jan Beulich wrote:
>>>>>>> On 26.06.15 at 16:34, <ian.campbell@citrix.com> wrote:
>>>>> I did this using rdmsr from mst-tools instead, running on a native
>>>>> kernel gave:
>>>>>
>>>>> # for i in $(seq 0 31) ;do rdmsr -p $i MSR_K8_TOP_MEM2; done
>>>>> 0
>>>>> [...]
>>>>> 0
>>> Is MSR_K8_TOP_MEM2 defined somewhere in the shell?
>> There is no $ there, so it wouldn't make any difference...
>>
>> I had foolishly assumed that rdmsr would either know the names of the
>> MSRs or it would complain about a string it didn't understand which
>> wasn't a number.
>>
>> Instead it just reads some random register which happens to be
>> strtoul("MSR_K8_TOP_MEM2"), how helpful.
> => https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=790075
>
>>> Just to make sure, could you use explicit address, i.e.
>>>
>>>       for i in $(seq 0 31) ;do rdmsr -p $i 0xc001001d; done
> It reported 43f000000 on all processors on native (and only the first 8
> on Xen due to limited dom0 vcpus).
>
>>> (and if they are still all zeroes, can you read 0xc0010010 (SYSCFG) as
>>> well?)
> It wasn't all zeroes, but anyway, it reported 740000 on all processors
> on native (I forgot to run under Xen).

Thanks, so this means that we do have WB memory above 4G.

(And I am not sure I understand why Jan said MTRRs show that memory 
above 4G is UC in 
http://lists.xenproject.org/archives/html/xen-devel/2015-06/msg04397.html . 
The log also seems to suggest that it is WB, doesn't it?)


-boris