I'd like to help you but I'm really not an expert in hardware programming or in motherboard chipsets, I have just been able to keep this patch alive for 1 year but apparently it's not enough. Could you tell me in what this patch is 100% different from what is in lm-sensors and I could fix it? It would be a pity that these working modifications would be lost forever. Thank you. -----Message d'origine----- De : lm-sensors-bounces at lm-sensors.org [mailto:lm-sensors-bounces at lm-sensors.org]De la part de Henrique de Moraes Holschuh Envoy? : mardi 25 octobre 2005 13:20 ? : lm-sensors at lm-sensors.org Objet : Re: Patch for 'w83627thf' Chipset add SmartFan control On Tue, 25 Oct 2005, Eric Pascal wrote: > I have maintained a patch (published on a forum by Peter Oehry - > http://archives.andrew.net.au/lm-sensors/msg27345.html) for w83627thf > chipsets that enable SmartFan control on the chipset to automaticaly > control the CPU Fan speed depending on the CPU temperature. The forum > describes the function of the drivers as followed: > > They correspond direct with the datasheet of the w83627thf [...] There is this new idea in lm-sensors to try to make these things hardware-independant (although at the state it is right now, it just broke some drivers, and took away functionality). I have a working setup that is reasonably hardware independant for smart fan control, for the LM85 and friends. Since there is NO proper documentation for this "device independant" interface, and any queries about it to this list are met with silence, I suppose it is up to us who need it to fix that. Are you willing to work with me so that we can draft such an interface (based on what already is in lm-sensors)? Your patch would change a bit, as it is 100% different from what is already in use in lm-sensors, though. Also, from what I can see, your chip can do a *great* deal many things, we would probably have to automate a few of them (they wouldn't fit an useable generic interface), and if you need to expose the underlying controls anyway, we'd need a hardware_specific namespace to do so. What do you think? -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh _______________________________________________ lm-sensors mailing list lm-sensors at lm-sensors.org http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
