Date: Fri, 19 Jun 2020 17:22:14 +0200 From: Michael Tuexen <tuexen@freebsd.org> To: Justin Hibbits <chmeeedalf@gmail.com> Cc: Dennis Clarke <dclarke@blastwave.org>, freebsd-ppc <freebsd-ppc@freebsd.org> Subject: Re: recent FreeBSD 13.0-CURRENT has fans screaming again on PowerMac quad Message-ID: <FFD9F04A-DE16-4EEF-B2F2-FD95EAAF4C07@freebsd.org> In-Reply-To: <20200619093447.469d669c@titan.knownspace> References: <250e4677-c442-3b41-5796-6b79b07ac12c@blastwave.org> <1692AB48-2438-4B4B-82BB-F8BAEF04B459@freebsd.org> <20200619093447.469d669c@titan.knownspace>
next in thread | previous in thread | raw e-mail | index | archive | help
> On 19. Jun 2020, at 16:34, Justin Hibbits <chmeeedalf@gmail.com> = wrote: >=20 > On Fri, 19 Jun 2020 15:55:30 +0200 > Michael Tuexen <tuexen@freebsd.org> wrote: >=20 >>> On 19. Jun 2020, at 14:22, Dennis Clarke via freebsd-ppc >>> <freebsd-ppc@freebsd.org> wrote: >>>=20 >>>=20 >>> Dear ppc64 big endian types: >>>=20 >>> Sort of the same problem we had before. Except now the screaming >>> fans seem to mysteriously die down for about thirty seconds before >>> they start up again and run full blast for five minutes. It makes >>> the machine a nasty blower fan thing to put down the hallway and >>> run a 30 meter CAT6 ethernet to it just to deal with it at all. >>>=20 >>> Anyone else seeing this ? It is really quite horrible to listen to. >>>=20 >>> At the moment there is nothing running at all with the exception of >>> a svnlite checkout of http://svn.freebsd.org/base/head and the >>> machine is screaming as if its life may end at any moment. >>> Thankfully I have spare parts for when these fans blow up : >>>=20 >>>=20 >>> dclarke@enceladus:~ $ uptime >>> 12:21PM up 33 mins, 2 users, load averages: 0.97, 0.84, 0.48 >>>=20 >>> dclarke@enceladus:~ $ uname -apKU >>> FreeBSD enceladus 13.0-CURRENT FreeBSD 13.0-CURRENT #0 r362037: Thu >>> Jun 11 06:09:20 UTC 2020 >>> = root@releng1.nyi.freebsd.org:/usr/obj/usr/src/powerpc.powerpc64/sys/GENERI= C64 >>> powerpc powerpc64 1300097 1300097 >>>=20 >>> dclarke@enceladus:~ $ cat /etc/rc.conf >>> clear_tmp_enable=3D"YES" >>> syslogd_flags=3D"-ss" >>> hostname=3D"enceladus" >>> ifconfig_bge0=3D"inet 172.16.35.8 netmask 255.255.255.192" >>> defaultrouter=3D"172.16.35.1" >>> sshd_enable=3D"YES" >>> ntpd_enable=3D"YES" >>> # Set dumpdev to "AUTO" to enable crash dumps, "NO" to disable >>> dumpdev=3D"AUTO" >>> powerd_enable=3D"YES" >>> dclarke@enceladus:~ $ >>> dclarke@enceladus:~ $ >>>=20 >>> top -CSP shows : >>>=20 >>> last pid: 1113; load averages: 1.00, 0.70, 0.35 >>> up 0+00:29:42 12:17:57 55 processes: 3 >>> running, 51 sleeping, 1 waiting CPU 0: 8.6% user, 0.0% nice, >>> 44.7% system, 0.3% interrupt, 46.3% idle CPU 1: 0.3% user, 0.0% >>> nice, 2.8% system, 0.6% interrupt, 96.4% idle CPU 2: 0.0% user, >>> 0.0% nice, 0.0% system, 0.0% interrupt, 0.0% idle CPU 3: 0.0% >>> user, 0.0% nice, 0.0% system, 0.0% interrupt, 0.0% idle Mem: >>> 28M Active, 696M Inact, 626M Wired, 344M Buf, 6406M Free Swap: >>> 3615M Total, 3615M Free >>>=20 >>> PID USERNAME THR PRI NICE SIZE RES STATE C TIME >>> CPU COMMAND 11 root 4 155 ki31 0B 192K RUN 0 >>> 112:42 307.44% idle 1104 root 1 103 0 32M 19M >>> CPU3 3 4:57 87.60% svnlite 12 root 22 -52 - >>> 0B 1056K WAIT 1 0:12 2.65% intr 7 root 2 -16 - >>> 0B 96K - 0 0:04 1.17% cam 19 root 6 20 >>> - 0B 288K psleep 2 0:02 0.50% bufdaemon 1050 dclarke >>> 1 20 0 23M 5252K select 0 0:01 0.23% sshd 1113 >>> dclarke 1 20 0 16M 4408K CPU0 0 0:00 0.01% >>> top 931 root 1 20 0 23M 4676K select 2 0:00 >>> 0.01% ntpd 1109 dclarke 1 20 0 23M 5260K select 2 >>> 0:00 0.00% sshd 15 root 15 -68 - 0B 720K - >>> 0 0:00 0.00% usb 971 root 1 20 0 19M 5696K >>> select 3 0:00 0.00% sendmail 761 root 1 20 0 >>> 14M 2384K select 2 0:00 0.00% syslogd 0 root 21 -16 >>> - 0B 1008K swapin 3 0:00 0.00% kernel 1062 root >>> 1 20 0 15M 4944K pause 2 0:00 0.00% csh 1107 root >>> 1 20 0 23M 11M select 1 0:00 0.00% sshd >>> 1048 root 1 39 0 23M 11M select 2 0:00 >>> 0.00% sshd 940 root 1 20 0 14M 1260K select 3 >>> 0:00 0.00% powerd 1057 dclarke 1 20 0 15M 3848K >>> wait 1 0:00 0.00% su 1058 dante 1 20 0 15M >>> 3988K wait 3 0:00 0.00% sh 1051 dclarke 1 20 0 >>> 15M 3988K wait 3 0:00 0.00% sh 1 root 1 20 0 >>> 13M 1228K wait 3 0:00 0.00% init 1061 dante 1 >>> 20 0 15M 3848K wait 1 0:00 0.00% su 13 root >>> 3 -8 - 0B 144K - 3 0:00 0.00% geom 978 root >>> 1 20 0 14M 2400K nanslp 3 0:00 0.00% cron 1110 >>> dclarke 1 20 0 15M 4028K wait 1 0:00 0.00% sh >>> 23 root 1 -16 - 0B 48K pmac_t 3 0:00 0.00% >>> pmac_thermal 1037 root 1 52 0 14M 2808K ttyin 2 >>> 0:00 0.00% getty 1035 root 1 52 0 14M 2808K >>> ttyin 0 0:00 0.00% getty 1033 root 1 52 0 >>> 14M 2808K ttyin 1 0:00 0.00% getty 1032 root 1 52 >>> 0 14M 2808K ttyin 3 0:00 0.00% getty 968 root >>> 1 20 0 22M 4108K select 2 0:00 0.00% sshd 1036 root >>> 1 52 0 14M 2808K ttyin 3 0:00 0.00% getty >>>=20 >>> There must be a trivial fix to all this as we had it all working >>> pretty darn well some few months ago. =20 >> Disable SMP? >>=20 >> Best regards >> Michael >>>=20 >>> --=20 >>> Dennis Clarke >>> RISC-V/SPARC/PPC/ARM/CISC >>> UNIX and Linux spoken >>> GreyBeard and suspenders optional >>> _______________________________________________ >>> freebsd-ppc@freebsd.org mailing list >>> https://lists.freebsd.org/mailman/listinfo/freebsd-ppc >>> To unsubscribe, send any mail to >>> "freebsd-ppc-unsubscribe@freebsd.org" =20 >>=20 >> _______________________________________________ >> freebsd-ppc@freebsd.org mailing list >> https://lists.freebsd.org/mailman/listinfo/freebsd-ppc >> To unsubscribe, send any mail to = "freebsd-ppc-unsubscribe@freebsd.org" >=20 > If it's SMP timebase sync related, the problem actually looks > relatively easy to solve, at least for G4 and PowerMac11,2 G5's > (PowerMac7,3 looks trickier), from examining the Linux source and the > OFW device tree. The PowerMac11,2 (which the Quad is), and the G4s, > use a GPIO to disable and enable the timebase, so a very similar thing > can be done with the powermac platform driver as is done with the > mpc85xx platform driver, which effectively is: >=20 > All CPUs rendezvous in the smp_timebase_sync() function > BSP disables timebase, sets the timebase sync to non-zero > APs check for non-zero timebase, set their timebases to it > All CPUs rendezvous at the end > BSP re-enables timebase, and unleashes CPUs >=20 > A couple hours of work for someone with the hardware, I'd say. I can > probably provide a patch to be tested by someone in the next few days, > if someone else doesn't get around to it first. I would be happy to test. Need to check what kind of PowerMac I have in = my lab... Best regards Michael >=20 > - Justin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?FFD9F04A-DE16-4EEF-B2F2-FD95EAAF4C07>