mwolff accepted this revision. mwolff added a comment.
OK, cool! That clearly shows that this patch _is_ valuable: Before we have ~6% CPU cycle cost, now it's down to 1.5% (inclusively). This is a significant reduction, so I'm all for it. But note once again the stark difference in these numbers vs. what valgrind reports. Please, try to use `perf record --call-graph dwarf` + hotspot flamegraphs more in favor of callgrind/kcachegrind. Note that you can even use hotspot's record page for some more features, most notably the off-CPU profiling. This can give you more food for potential optimizations (finding sleep time, I/O, lock contention, ...). Happy profiling! REPOSITORY R241 KIO REVISION DETAIL https://phabricator.kde.org/D12511 To: jtamate, #kate, cullmann, mwolff Cc: mwolff, cullmann, #frameworks, michaelh, kevinapavew, ngraham, bruns, demsking, sars, dhaumann