From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Return-path: Received: from bh-25.webhostbox.net ([208.91.199.152]:43728 "EHLO bh-25.webhostbox.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753529AbeFJFOW (ORCPT ); Sun, 10 Jun 2018 01:14:22 -0400 Date: Sat, 9 Jun 2018 22:14:20 -0700 From: Guenter Roeck To: Kai-Heng Feng Cc: Chris Ouellet , linux-hwmon@vger.kernel.org, Linux Kernel Mailing List Subject: Re: Ryzen 2700 Temperature offset? Message-ID: <20180610051420.GA22601@roeck-us.net> References: <20180607154915.GA7061@roeck-us.net> <54EAD835-C48D-4751-9BFF-71536241129A@canonical.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <54EAD835-C48D-4751-9BFF-71536241129A@canonical.com> Sender: linux-hwmon-owner@vger.kernel.org List-Id: linux-hwmon@vger.kernel.org On Sun, Jun 10, 2018 at 03:03:47AM +0800, Kai-Heng Feng wrote: > at 23:49, Guenter Roeck wrote: > > >On Thu, Jun 07, 2018 at 11:03:13AM +0800, Kai Heng Feng wrote: > >>Hi Guenter, > >> > >>A user thinks the Ryzen 2700 temperature may not be correct [1]. > >>Can you take a look when you have time? Thanks. > >> > >>Also, where can I find all these temperature offsets? I googled but with > >>no > >>result. > >Problem is that the offsets are not documented in public, making it > >difficult to find out. For Ryzen 2700, there is currently no offset > >configured; I was told that there is none. For Ryzen 2700X, the offset > >is 10 degrees C. > > I am still quite interesting where you get this information... > Primarily lots of online research. Windows tools usually support the offsets, so screendumps help in some cases. > > > >>[1] https://bugs.launchpad.net/bugs/1774817 > > > >The key here is the commit mentioned in the log (1b59788979acd). It does > >not only adjust the per-CPU offset, but also takes a configurable offset > >of 49 degrees C into account. You'll need to apply that patch in your > >kernel, or the temperature may be off by 49 degrees C. > > So unless AMD shares these offsets value, we can never know what correct > offsets are? > Unfortunately that is correct. AFAIK the information is available from AMD under NDA. I can not sign such an NDA, and even if I could and would, I could not use the information in an open source driver. Kind of a loose-loose situation. If you (Canonical) have any influence at AMD, it might be useful to let them know that their policy is not very helpful. Guenter > Kai-Heng > > > > >Hope this helps. If not, please let me know. Note that I am travelling; > >responses may be delayed. > > > >Guenter > > > -- > To unsubscribe from this list: send the line "unsubscribe linux-hwmon" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html