Date: Sun, 25 Aug 2024 00:12:13 +0100 From: Bob Bishop <rb@gid.co.uk> To: Poul-Henning Kamp <phk@phk.freebsd.dk> Cc: "current@freebsd.org" <current@FreeBSD.org> Subject: Re: powerd forgets top CPU frequency ? Message-ID: <AEEBC43B-71F5-4974-87CB-3B05F096B07A@gid.co.uk> In-Reply-To: <202408241706.47OH6LfG023651@critter.freebsd.dk> References: <202408241706.47OH6LfG023651@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi, > On 24 Aug 2024, at 18:06, Poul-Henning Kamp <phk@phk.freebsd.dk> = wrote: >=20 > On T14s thinkpad running: >=20 > FreeBSD critter.freebsd.dk 15.0-CURRENT FreeBSD 15.0-CURRENT #7 = main-n271455-a1740cb93639-dirty: Thu Aug 1 09:45:40 UTC 2024 = root@critter.freebsd.dk:/critter/obj/critter/FreeBSD/src/amd64.amd64/sys/G= ENERIC-NODEBUG amd64 >=20 > When I run two 100% CPU loaded processes, the machine gradually gets = slower and slower, until eventually it limps along at 500-ish MHz rates. >=20 > Killing powerd and starting it again seems to bring the machine back = to full speed (for some time). >=20 > Looking at the powerd -v output, my gut feeling is that powerd = sometimes looses track of what the (current) top frequency is and thus = gradually drags the machine slower or slower... Possibly buggy ACPI on the Thinkpad not reporting any frequency higher = than it=E2=80=99s actually running at? > I have had a faint feeling for some time that this machine runs = significantly faster after a reboot, but I have no observations to = support that. >=20 > --=20 > Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 > phk@FreeBSD.ORG | TCP/IP since RFC 956 > FreeBSD committer | BSD since 4.3-tahoe > Never attribute to malice what can adequately be explained by = incompetence. -- Bob Bishop rb@gid.co.uk
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?AEEBC43B-71F5-4974-87CB-3B05F096B07A>