.
Wolf, in your cudaminer fork for BBR, I see quite huge amount of context switches per second,
what cause high cpu utilisation:
Celeron + 3x 750Ti:
root@kopiemtu:~# vmstat -n 1
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
5 0 0 2269904 64944 1106396 0 0 0 4 11 7 45 20 35 0 0
3 0 0 2269896 64944 1106396 0 0 0 26 590 257053 57 43 0 0 0
3 0 0 2269896 64944 1106396 0 0 0 0 594 256664 57 43 0 0 0
3 0 0 2269864 64944 1106396 0 0 0 0 588 257120 57 43 0 0 0
root@kopiemtu:~# uptime
21:41:52 up 22 days, 7:03, 2 users, load average: 3,13, 3,09, 3,06
and:
Xeon + 5x 750Ti
root@kopiemtu:~# vmstat -n 1
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
6 0 0 2148372 59916 589464 0 0 0 1 9 5 24 15 61 0 0
5 0 0 2148364 59916 589464 0 0 0 0 1177 367276 61 39 0 0 0
5 0 0 2148364 59916 589464 0 0 0 0 1158 366147 61 39 0 0 0
5 1 0 2148364 59916 589464 0 0 0 23 1171 365873 61 39 0 0 0
root@kopiemtu:~# uptime
21:41:35 up 13 days, 4:21, 2 users, load average: 5,17, 5,16, 5,14
what is the reason of it? maybe there is a way to avoid it?