https://bugzilla.kernel.org/show_bug.cgi?id=217965 --- Comment #62 from Matthew Stapleton (matthew4196@xxxxxxxxx) --- With top at 1 second intervals , with thunderbird source extract, I can see the flush process with around 70% to 100% for 2 1 second intervals during the tar extract. I don't this this much cpu usage was present with CR_BEST_AVAIL_LEN commented out. Hopefully the perf probes will show where the real usage is. I still haven't tried out the perf probes, but I'm still planning on doing that hopefully sometime in the next few days. -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.
