Dne 23.11.2016 v 02:08 tang.junhui@xxxxxxxxxx napsal(a):
So, now we have at least 4 ways to improve mutliapth efficiency:
1) Filtering uevents;
2) Merger uevents;
3) Using separate locks for mpvec and pathvec;
4) Get rid of the gazillion waiter threads.
This is exciting, but how do we achieve this blueprint?
Can we set up some working groups and develop it in parallel
to implement each improvement since most of them are independent?
I'm missing here the analysis of impact on performance.
Do we have 'perf' traces and other reports where is the CPU wasted ?
While surely solving all issues make sense - there are some tasks which have
major impact on speed - while others user will most not notice at all.
Regards
Zdenek
--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel
