Port-arm archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: High system times on Scaleway Cloud server
Jason Thorpe wrote:
> I would start by using dtrace to get some profiling information of where the time is being spent while on-CPU and generate a flame graph for it.
I'm not sure that dtrace is available but tprof works!Â
# tprof monitor -e CPU_CYCLES:k -o tprof.out sleep 30
# tprof analyze tprof.out
File: tprof.out
Number of samples: 74
percentage nsamples pid lwp cpu k address symbol
------------ -------- ------ ---- ---- - ---------------- ------
2.702703% 2 0 4 0 1 ffffffc0000040a4 splx+0xbc
2.702703% 2 0 92 0 1 ffffffc000457908 vfs_vnode_iterator_next1+0x78
1.351351% 1 27653 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 27141 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 25997 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 25057 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 24441 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 23893 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 23337 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 22489 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 18477 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 17802 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 16665 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 16638 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 16429 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 16342 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 16173 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 15047 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 6328 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 5980 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 5054 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 4933 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 3791 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 2512 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 2506 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 1340 1 0 1 ffffffc0000040a4 splx+0xbc
1.351351% 1 628 1 0 1 ffffffc0000040a4 splx+0xbc
...
--
Alex
Home |
Main Index |
Thread Index |
Old Index