So I'm emerging things with -j8 1 [|||||||||||||||||||||||||||||||||||||||||||||||||||||||||100.0%] 5 [ 0.0%] 2 [ 0.0%] 6 [ 0.0%] 3 [ 0.0%] 7 [ 0.0%] 4 [ 0.0%] 8 [ 0.0%] Mem[||||||||||||| 3282/32064MB] Tasks: 89, 86 thr; 9 running Swp[ 0/0MB] Load average: 8.16 8.24 8.08 Uptime: 03:45:2 htop shows loadavg 8, but only one core utilized, I've waited some time to be sure that loadavg won't fall back, but I would say that it's reasonable to assume that emerging webkit-gtk happens with 8 jobs top - 13:13:06 up 3:45, 8 users, load average: 8.20, 8.25, 8.09 Tasks: 216 total, 10 running, 206 sleeping, 0 stopped, 0 zombie %Cpu0 : 95.7 us, 4.0 sy, 0.0 ni, 0.0 id, 0.0 wa, 0.3 hi, 0.0 si, 0.0 st %Cpu1 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu2 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu3 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu4 : 0.0 us, 0.0 sy, 0.0 ni, 99.7 id, 0.0 wa, 0.3 hi, 0.0 si, 0.0 st %Cpu5 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu6 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu7 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st KiB Mem : 32833772 total, 26752872 free, 3112392 used, 2968508 buff/cache KiB Swap: 0 total, 0 free, 0 used. 29444172 avail Mem also top, whichever is easier to read ;) Reproducible: Always
Probably also of interest actual processes: PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 5480 portage 20 0 566348 529692 17600 R 12.6 1.6 0:04.11 cc1plus 5566 portage 20 0 252852 211396 14168 R 12.6 0.6 0:01.06 cc1plus 5409 portage 20 0 776824 743420 14384 R 12.3 2.3 0:05.98 cc1plus 5529 portage 20 0 479280 444876 14348 R 12.3 1.4 0:02.74 cc1plus 5549 portage 20 0 374664 336524 14128 R 12.3 1.0 0:01.70 cc1plus 5583 portage 20 0 176828 141052 14176 R 12.3 0.4 0:00.63 cc1plus 5586 portage 20 0 165856 130544 14164 R 12.3 0.4 0:00.57 cc1plus 5589 portage 20 0 141248 99680 14188 R 12.3 0.3 0:00.48 cc1plus
there seems to be problems with this kernel. try 4.2.4 which i added to the tree, upstream will want to see the latest tested, and that is the latest. in the mean time i'll let upstream know about this kernel
Created attachment 415448 [details] kernel .config file same things also happens on 4.2.4 top - 14:30:39 up 10 min, 7 users, load average: 8.09, 4.09, 1.74 Tasks: 242 total, 12 running, 230 sleeping, 0 stopped, 0 zombie %Cpu0 : 91.7 us, 8.3 sy, 0.0 ni, 0.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu1 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu2 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu3 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu4 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu5 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu6 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu7 : 0.0 us, 0.0 sy, 0.0 ni,100.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st KiB Mem : 32833772 total, 31056744 free, 881872 used, 895156 buff/cache KiB Swap: 0 total, 0 free, 0 used. 31749732 avail Mem PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 7499 portage 20 0 68700 42252 17120 R 7.3 0.1 0:00.22 cc1plus 7505 portage 20 0 63808 38988 17160 R 7.0 0.1 0:00.21 cc1plus 7507 portage 20 0 70064 41224 14468 R 7.0 0.1 0:00.21 cc1plus 7512 portage 20 0 61384 34264 15992 R 4.0 0.1 0:00.12 cc1plus 7515 portage 20 0 58696 29520 13464 R 3.0 0.1 0:00.09 cc1plus 7519 portage 20 0 56552 28208 13472 R 2.7 0.1 0:00.08 cc1plus 7522 portage 20 0 33972 18496 3688 R 2.7 0.1 0:00.08 ld 1546 root 20 0 446552 79316 60692 S 0.3 0.2 0:03.54 X 1962 amade 20 0 116016 11200 8856 S 0.3 0.0 0:00.09 irssi 2059 amade 20 0 1064240 286972 97444 S 0.3 0.9 0:23.76 firefox 29314 root 20 0 319228 227184 8576 S 0.3 0.7 0:22.50 emerge 30432 portage 20 0 13552 4932 2268 S 0.3 0.0 0:00.06 make 30740 root 20 0 36600 3868 3092 R 0.3 0.0 0:00.39 top attaching .config in case it's some kind of misconfiguration on my side
Same on 4.2.5 series, retested on vanilla where I also reproduced the problem, seems like upstream have changed something with NO_HZ_FULL, changing to NO_HZ_IDLE fixes things (as I predicted in previous comment misconfiguration ;) http://thread.gmane.org/gmane.linux.kernel/2052957 seems like this is expeted behaviour, so closing as invalid.