Date: Thu, 29 Apr 2021 20:33:03 +0200 From: Jesper Schmitz Mouridsen <jsm@FreeBSD.org> To: freebsd-hackers@freebsd.org Subject: Re: hwpstate_intel cpufreq and est. Message-ID: <f6fe1344-a7b2-2dc8-f8ec-2a55f3b5fd24@FreeBSD.org> In-Reply-To: <226bfdd6-2d24-2dc7-79e3-7dc33a56f2a2@FreeBSD.org> References: <226bfdd6-2d24-2dc7-79e3-7dc33a56f2a2@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 29.04.2021 20.09, Jesper Schmitz Mouridsen wrote: > Hi list. > > I am working on sysutils/libsysstat [1], I have maintained it over the > years. > > I even have a little testprogram [2] > > Recently I got a new CPU which uses freq_driver: hwpstate_intel. > > It seems to me from values from dev.cpu.X.freq that this drivers > supports different freq levels per core? > > And that the BUG section in man cpufreq stating that > > "When multiple CPUs offer frequency control, they cannot be set to > different levels and must all offer the same frequency settings." > > Does not apply to hwpstate_intel? Is this correct. My current code > relies on > > this same freq on all cores bug. > > [1] https://www.freshports.org/sysutils/libsysstat > > [2] https://people.freebsd.org/~jsm/libsysstat-test.tar.gz > > Thanks > > Jesper Schmtiz Mouridsen. > I think I found the answer in https://reviews.freebsd.org/D30004: Users running newer Intel(R) CPUs may find some differences in dynamic frequency control when upgrading to FreeBSD 13.0. A new driver for the Intel(R) Speed Shift(TM) feature set, available on certain SKUs, exposes the ability for the hardware to dynamically vary the core frequencies, including on a per core basis. > _______________________________________________ > freebsd-hackers@freebsd.org mailing list > https://lists.freebsd.org/mailman/listinfo/freebsd-hackers > To unsubscribe, send any mail to > "freebsd-hackers-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?f6fe1344-a7b2-2dc8-f8ec-2a55f3b5fd24>