For illustration, I am putting up a snippet of TOP output during one of my sessions --
top - 18:54:23 up 1 day, 23:33, 8 users, load average: 1.23, 1.08, 1.02
Tasks: 295 total, 3 running, 289 sleeping, 0 stopped, 3 zombie
Cpu(s): 28.4%us, 1.9%sy, 0.0%ni, 67.8%id, 1.8%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 7941828k total, 7741560k used, 200268k free, 90992k buffers
Swap: 8388604k total, 186344k used, 8202260k free, 1479116k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
61 root 39 19 0 0 0 S 0.0 0.0 0:07.02 khugepaged
60 root 25 5 0 0 0 S 0.0 0.0 0:00.00 ksmd
4358 rtkit 21 1 164m 976 940 S 0.0 0.0 0:02.09 rtkit-daemon
1 root 20 0 19352 1260 1032 S 0.0 0.0 0:05.40 init
2 root 20 0 0 0 0 S 0.0 0.0 0:00.40 kthreadd
4 root 20 0 0 0 0 S 0.0 0.0 0:04.24 ksoftirqd/0
19 root 20 0 0 0 0 S 0.0 0.0 0:37.61 events/0
23 root 20 0 0 0 0 S 0.0 0.0 0:00.00 cgroup
I wanted to figure out the difference between the values under PR and NI columns, and I found the answer in the man page of C library function getpriority. The following snippet from the man page is the most relevant and explains the difference very well --
The actual priority range varies between kernel versions. Linux before 1.3.36 had -infinity..15. Since kernel 1.3.43 Linux has the range -20..19. Within the kernel, nice values are actually represented using the corresponding range 40..1 (since negative numbers are error codes) and these are the values employed by the setpriority() and getpriority() system calls. The glibc wrapper functions for these system calls handle the translations between the user-land and kernel representations of the nice value according to the formula unice = 20 - knice.
top - 18:54:23 up 1 day, 23:33, 8 users, load average: 1.23, 1.08, 1.02
Tasks: 295 total, 3 running, 289 sleeping, 0 stopped, 3 zombie
Cpu(s): 28.4%us, 1.9%sy, 0.0%ni, 67.8%id, 1.8%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 7941828k total, 7741560k used, 200268k free, 90992k buffers
Swap: 8388604k total, 186344k used, 8202260k free, 1479116k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
61 root 39 19 0 0 0 S 0.0 0.0 0:07.02 khugepaged
60 root 25 5 0 0 0 S 0.0 0.0 0:00.00 ksmd
4358 rtkit 21 1 164m 976 940 S 0.0 0.0 0:02.09 rtkit-daemon
1 root 20 0 19352 1260 1032 S 0.0 0.0 0:05.40 init
2 root 20 0 0 0 0 S 0.0 0.0 0:00.40 kthreadd
4 root 20 0 0 0 0 S 0.0 0.0 0:04.24 ksoftirqd/0
19 root 20 0 0 0 0 S 0.0 0.0 0:37.61 events/0
23 root 20 0 0 0 0 S 0.0 0.0 0:00.00 cgroup
I wanted to figure out the difference between the values under PR and NI columns, and I found the answer in the man page of C library function getpriority. The following snippet from the man page is the most relevant and explains the difference very well --
The actual priority range varies between kernel versions. Linux before 1.3.36 had -infinity..15. Since kernel 1.3.43 Linux has the range -20..19. Within the kernel, nice values are actually represented using the corresponding range 40..1 (since negative numbers are error codes) and these are the values employed by the setpriority() and getpriority() system calls. The glibc wrapper functions for these system calls handle the translations between the user-land and kernel representations of the nice value according to the formula unice = 20 - knice.
