Date: Sat, 17 Jan 2015 11:45:18 -0800 From: Mark Millard <markmi@dsl-only.net> To: Adrian Chadd <adrian@freebsd.org>, Justin Hibbits <chmeeedalf@gmail.com> Cc: FreeBSD PowerPC ML <freebsd-ppc@freebsd.org> Subject: Re: PowerMac G5 quad-core, CPU A1 DIODE TEMP: 90.8 C (for example): How to handle? [more data] Message-ID: <42358897-0AC2-4B35-BE01-1D4EB2CC2F47@dsl-only.net> In-Reply-To: <CAJ-Vmo=0-tzB0U2evzU9XNE6oQU-7_WnqmE%2B9pHWnvRTpkh-Tg@mail.gmail.com> References: <42CF1E40-5BD5-4B00-86E9-C62AEB9B8B93@dsl-only.net> <15A6D627-9DC7-48AF-B133-94980AFCE46A@dsl-only.net> <20150115231129.1b28c8d0@zhabar.attlocal.net> <0631235D-A505-4C37-87D7-6F46A14552AB@dsl-only.net> <20150116233145.6708cc6f@zhabar.attlocal.net> <31331F84-63CC-48B7-81B5-E70A22E88CB7@dsl-only.net> <604BAA0A-FD15-4310-88B2-DFEE9988D1EB@dsl-only.net> <20150117080916.3e321a4f@zhabar.attlocal.net> <CAJ-Vmo=0-tzB0U2evzU9XNE6oQU-7_WnqmE%2B9pHWnvRTpkh-Tg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
While I mentioned forced idle time as a technical possibility I'm not so = sure that FreeBSD would want to automatically drop performance in order = to keep a machine running. In my case such would not be enough for me to = decide to continue to use the problematical quad-core G5. (I doubt that = it would be a minor amount of idle time that was required to get =E2=80=9C= make -j 8 buildworld buildkernel" or other such to work reliably on this = G5.) I will try something to put the problematical G5 under load under Mac OS = X 10.5, say 4+ copies of the HINT benchmark running concurrently if I = still have that around. But I'm not aware of a pre-existing way to see = the fans speeds, pump speeds, and other such in that context. I may only = learn if it automatically shuts down or not. For FreeBSD I was fairly = sure I'd be able to readily find such extra information (and I did). I'd say that what I reported for "01:45:51 to 02:13:50" was near = full-throttle over that time and it started at 70.3C. Looks like Justin = got it programmed with the properties he wanted. =3D=3D=3D Mark Millard markmi@dsl-only.net On 2015-Jan-17, at 08:35 AM, Adrian Chadd <adrian@freebsd.org> wrote: On 17 January 2015 at 08:09, Justin Hibbits <chmeeedalf@gmail.com> = wrote: > The new algorithm in this patch is supposed to put the fans at full > throttle around the midpoint, so around 70C they should be at full > blast. >=20 > Have you tried the same machine under heavy load with OS X or Linux? > We can keep adjusting the algorithm, as long as your machine is known > to be good enough under one of those OSs. If they also overheat, > there's nothing we can do, since OS X at least should run fine with > heavy load, as long as the hardware can handle it. Can you force introduce halt cycles to compensate for the increasing rise in temperature? -adrian > - Justin >=20 > On Sat, 17 Jan 2015 03:38:33 -0800 > Mark Millard <markmi@dsl-only.net> wrote: >=20 >> I've now tested the new patch and strongly expect that the new code >> is much better but it is not sufficient to keep the temperature down >> on the problematical quad-core G5 that I've been testing via >> buildworld. It did last over 30 minutes and there were no 0 rpm >> figures sampled for this new code. >>=20 >> Note: it turns out the two sysctl dumps of samples (dev.smu and >> dev.smusat) are done (as a pair) more like once every 3 to 4 seconds, >> not once a second. I include timestamps this time. >>=20 >> The new code ended up keeping the intake and exhaust fans near the >> 3200 maxrpm for a long time and the pump near the maxrpm 3600 too. >> But that was not enough for this G5 to keep its cool. It would >> apparently take giving the CPUs/cores some periodic idle time to cool >> down in order to keep this G5 cool enough. (Really just cpu a1 >> needing such.) >>=20 >> Over the interval below (01:45:51 to 02:13:50) no .rpm figure was >> sampled as < 3100 and no pump.rpm was sampled as < 3400. >>=20 >> Sat Jan 17 01:45:51 PST 2015 >> dev.smu.0.fans.cpu_a_intake.rpm: 3131 >> dev.smu.0.fans.cpu_b_intake.rpm: 3131 >> dev.smu.0.fans.cpu_a_exhaust.rpm: 3127 >> dev.smu.0.fans.cpu_b_exhaust.rpm: 3143 >> dev.smu.0.fans.cpu_a_pump.rpm: 3549 >> dev.smusat.0.cpu_a0_diode_temp: 59.4C >> dev.smusat.0.cpu_a1_diode_temp: 70.3C >> dev.smusat.1.cpu_b0_diode_temp: 59.1C >> dev.smusat.1.cpu_b1_diode_temp: 64.8C >> Sat Jan 17 01:45:54 PST 2015 >> ... Over the below range pump.rpm is sampled as < 3500 only 6 times >> (each > 3400)... Sat Jan 17 02:01:46 PST 2015 >> dev.smu.0.fans.cpu_a_intake.rpm: 3197 >> dev.smu.0.fans.cpu_b_intake.rpm: 3202 >> dev.smu.0.fans.cpu_a_exhaust.rpm: 3200 >> dev.smu.0.fans.cpu_b_exhaust.rpm: 3198 >> dev.smu.0.fans.cpu_a_pump.rpm: 3636 >> dev.smusat.0.cpu_a0_diode_temp: 67.6C >> dev.smusat.0.cpu_a1_diode_temp: 81.6C >> dev.smusat.1.cpu_b0_diode_temp: 61.0C >> dev.smusat.1.cpu_b1_diode_temp: 65.6C >> Sat Jan 17 02:01:50 PST 2015 >> ... The last lines that made it into the file before automatic >> shutdown were ... Sat Jan 17 02:13:50 PST 2015 >> dev.smu.0.fans.cpu_a_intake.rpm: 3200 >> dev.smu.0.fans.cpu_b_intake.rpm: 3199 >> dev.smu.0.fans.cpu_a_exhaust.rpm: 3199 >> dev.smu.0.fans.cpu_b_exhaust.rpm: 3197 >> dev.smu.0.fans.cpu_a_pump.rpm: 3614 >> dev.smusat.0.cpu_a0_diode_temp: 73.0C >> dev.smusat.0.cpu_a1_diode_temp: 90.9C >> dev.smusat.1.cpu_b0_diode_temp: 62.6C >> dev.smusat.1.cpu_b1_diode_temp: 67.1C >> Sat Jan 17 02:13:54 PST 2015 >>=20 >> So for the last 12 minutes or so the cpu intakes, exhausts, and pump >> were going basically full blast. It was not enough for my context but >> likely would help some other contexts.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?42358897-0AC2-4B35-BE01-1D4EB2CC2F47>