I have 2 Raspberry Pi2s running FreeBSD 11.2-RELEASE, one is using the latest image (https://download.freebsd.org/ftp/releases/ISO-IMAGES/11.2/FreeBSD-11.2-RELEASE-arm-armv6-RPI2.img.xz), the other I have built from source (r338421). Both seemed to function perfectly with no stability issues but I noticed some odd load values. Despite being 98%-100% idle, the systems show load values in the region of but never less than 1.00. After some brief investigation only CPU 0,1 and 3 are being utilised despite all 4 cores being discovered on boot. CPU 2 remains 100% idle at all times on both machines. root@pi2:~ # sysctl hw.ncpu hw.ncpu: 4 Running top -P at idle shows :- last pid: 9083; load averages: 1.09, 1.09, 1.03 up 3+01:04:56 06:25:53 21 processes: 1 running, 20 sleeping CPU 0: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU 1: 5.3% user, 0.0% nice, 0.0% system, 0.0% interrupt, 94.7% idle CPU 2: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU 3: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle Mem: 4464K Active, 31M Inact, 72M Wired, 29M Buf, 688M Free Swap: Running top -P with make -j8 buildworld :- last pid: 10475; load averages: 8.62, 3.53, 2.06 up 3+01:16:06 14:29:56 67 processes: 12 running, 55 sleeping CPU 0: 73.7% user, 0.0% nice, 12.5% system, 13.7% interrupt, 0.0% idle CPU 1: 89.0% user, 0.0% nice, 11.0% system, 0.0% interrupt, 0.0% idle CPU 2: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU 3: 82.0% user, 0.0% nice, 6.7% system, 11.4% interrupt, 0.0% idle Mem: 113M Active, 92M Inact, 93M Wired, 38M Buf, 498M Free Swap:
(In reply to Steve Clarke from comment #0) I do not not normally use RPI2's any more but I will note that historically I've seen this "only uses 3 of 4 cpus" multiple times in the past, going back well before 11.2 . Mostly this note is just letting you know that it is not only you, at least from a history going back before 11.2 . I never managed to isolate anything about controlling when it happened or just what was happening internally when it did happen.
^Triage: I'm sorry that this PR did not get addressed in a timely fashion. By now, the version that it was created against is long out of support. Please re-open if it is still a problem on a supported version.