Re: Wrong temperature with AMD and amdtemp.ko

From: Shane Ambler <FreeBSD_at_ShaneWare.Biz>
Date: Mon, 5 Oct 2015 14:54:20 +1030
On 04/10/2015 20:10, Willem Jan Withagen wrote:
> On 4-10-2015 03:26, Shane Ambler wrote:
>> On 03/10/2015 20:12, Willem Jan Withagen wrote:
>>> On 2-10-2015 23:32, Don Lewis wrote:
>>>> On  2 Oct, Willem Jan Withagen wrote:
>>>>>
>>>>> Hi
>>>>>
>>>>> 10.2-STABLE FreeBSD 10.2-STABLE #0 r287102: Mon Aug 24
>>>>>
>>>>> Processor: Opteron 6812, in Supermicro H8SGL
>>>>>
>>
>>> Thusfar I have not been able to locate the "Power and Thermal Datasheet"
>>> for the family 15h....
>>> Perhaps need to disassemble the bios, or check other tools or OSes on
>>> how they do this.
>>
>> According to Supermicro the H8SGL supports FreeBSD 8.0 but not 9.1
>
> I see this as an omission on the website.
> I've never been bothered by FreeBSD compatibility in the 15+ years I'm
> using Supermicro.
>
>> Maybe you could approach Supermicro support and ask for assistance in
>> updating FreeBSD support - maybe for other boards as well.
>>
>> Another approach is asking iXSystems, I'm pretty sure they sell
>> re-badged Supermicro machines.
>
> Turns out it has nothing to do with the Motherboard, but more on the
> temperature management hardware in the CPU.
> So asking either iXSstems or Supermicro does not really help.

But supermicro would have access to the bios source and can tell
quickly how it is done there. As a MB manufacturer they would also have
the chance to ask AMD for info not easily found in docs. If supermicro
doesn't provide info to oss developers then iXsystems could ask
supermicro for info as they sell the hardware as their own and need to
write drivers to support it and as a supporter of FreeBSD they may be
interested in writing the update themselves.

At the minimum one of them could tell you how the bios calculates the
temperature so you can match it. Would save a lot of experiments on
your part.


-- 
FreeBSD - the place to B...Software Developing

Shane Ambler
Received on Mon Oct 05 2015 - 02:24:34 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:00 UTC