Date: Fri, 20 Aug 2004 23:37:08 -0400 From: "Alexandre \"Sunny\" Kovalenko" <Alex.Kovalenko@verizon.net> To: Toxa <postfix@senmail.ru> Cc: freebsd-mobile@freebsd.org Subject: Re: Enhanced SpeedStep driver available Message-ID: <1093059427.701.7.camel@RabbitsDen> In-Reply-To: <20040818182931.GA1926@laptoxa.toxa.lan> References: <6.1.0.6.1.20040816074348.03f99338@popserver.sfu.ca> <4122A34F.2020607@nikiforov.ru> <20040818082141.5d2fcee7@duality.bytephobia.de> <20040818182931.GA1926@laptoxa.toxa.lan>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 2004-08-18 at 14:29, Toxa wrote: > On Wed, Aug 18, 2004 at 08:21:41AM +0200, Patrick Hurrelmann wrote: > > > Pentium 4 M is the mobile version of a Pentium 4 and capable of speed step. > > Pentium M is a complete new processor and not a mobile version of any other. Indeed it is based on the Pentium 3 layout (but only based ;) ). > > > Centrino is no processor at all. Centrino is the name for a hardware bundle: > > - Intel Pentium M (no Pentium 4!) > > - Intel Chipset (with or without onboard graphic) > > - Intel Wireless-LAN adapter > > > If one of this partsis missing on a system it must not be called Centrino (that's Intel politics). > > Thanks for clearing it out for us all, but the fact is that Pentium > 4 M (not Pentium M from Centrino "toolkit") is capable of SpeedStep > (Not Enhanced SpeedStep, right?). But freebsd lacks of speedstep support > (AFAIK, hw.acpi.cpu_throttle_state plays only with processor C-states, > which are not SpeedStep), so I'm womdering if est.ko would do > something with Pentium 4 M processors. Ahem... some time last fall I was using CPU throttling on my ThinkPad 560Z (Pentium II 266MHz). Worked nicely in 8 steps (12.5% increments). I don't know if it was Pentium II specific, but it definitely worked out of the box on then -CURRENT. CPU states were (and still are) separate from that. I have since switched to AMD laptop and have to use separate module (powernow_k7), but I am pretty sure that at least some CPU throttling was there and was working. Controls, AFAICR, were somewhere in hw.acpi.cpu space. Sorry for vague recollections, but maybe they will elicit response from somebody, who has better grip on the subject. --- Alexandre "Sunny" Kovalenko.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1093059427.701.7.camel>