On Sat, 2009-05-23 at 18:59 +0200, Rudolf Marek wrote: > Hi, > > Yes it works for most CPUs, but for unlucky users we might get complains from > them - and false alarms. > > I'm talking about: > > http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/41322.pdf > > Errata #319 > > So far all CPUs have the issue. >From manual : >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 319 Inaccurate Temperature Measurement Description The internal thermal sensor used for CurTmp (F3xA4[31:21]), hardware thermal control (HTC), software thermal control (STC) thermal zone, and the sideband temperature sensor interface (SB-TSI) may report inconsistent values. Potential Effect on System HTC, STC thermal zone, and SB-TSI do not provide reliable thermal protection. This does not affect THERMTRIP or the use of the STC-active state through StcPstateLimit or StcPstateEn (F3x68[30:28, 5]). Suggested Workaround None. Systems should be designed with conventional thermal control and throttling methods or utilize PROCHOT_L functionality based on temperature measurements from an analog thermal diode (THERMDA/THERMDC). Systems should not rely on the HTC features, STC thermal zone features, or use SB-TSI. Software should not modify HtcTmpLmt (F3x64[22:16]) or enable any of the STC thermal zone features by setting StcApcTmpLoEn, StcApcTmpHiEn, StcSbcTmpLoEn, or StcSpcTmpHiEn (F3x68[3:0]). <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< It seems that this problem is for some K10 processors. And K11 is OK, right ? I think our patch is safe enough and we will fix the problem for some K10 processors or in the worst case we will remove support for that particular cpu. Ingo: do you mind to applying this patch on -tip and lets see its feedback and fix the problem. Thanks -- JSR
