My new 800MHz titanium powerbook is consistently reported by /proc/cpuinfo at 667MHz. Is this true or a reporting error?
Apple's hardware docs seem to indicate that the 800MHz model can switch on the fly to 667MHz as a power-saving feature. Poking around the kernel sources, it appears that the report in cpuinfo comes from Open Firmware, so I looked through the Open Firmware device tree. In the CPU node, the properties that seem to be relevant are min_clock_frequency (667MHz), max_clock_frequency (800MHz), clock_frequency (667MHz), and force_slower_speed (1). Presumably it is the setting of clock_frequency that is being reported in /proc/cpuinfo (I can't think of an easy objective way to verify what the actual clock speed is). Booting into Open Firmware and checking these properties indicates that Linux is not messing with them - clock-frequency is set to 667 before the kernel is loaded. When running OS X (10.1.5), the Apple utilities show the CPU at 800MHz, and checking the OF device tree at this point shows clock-frequency to be up to 800MHz. How did they do it? A quick grep through the Darwin sources and I couldn't find any reference to max-clock-frequency, so I can't see where or how it's being set. Again, maybe the chip really is running at 800MHz under Linux, in which case /proc/cpuinfo (and OF) are simply lying to me, which isn't really that big of a deal. But if it is really running at 667MHz, it would be great to figure out a way to step the clock up. Any ideas? -- Robert Coie <[EMAIL PROTECTED]> Implementor, Apropos Ltd. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
