<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;"><br class=""><blockquote type="cite" class="">On Jun 20, 2018, at 7:22 PM, Lonnie Abelbeck <<a href="mailto:lists@lonnie.abelbeck.com" class="">lists@lonnie.abelbeck.com</a>> wrote:<br class=""><br class=""><br class=""><blockquote type="cite" class="">On Jun 20, 2018, at 6:47 PM, Jason A. Donenfeld <<a href="mailto:Jason@zx2c4.com" class="">Jason@zx2c4.com</a>> wrote:<br class=""><br class="">Hey Lonnie,<br class=""><br class="">Thanks for helping to debug this.<br class=""><br class="">On Thu, Jun 21, 2018 at 12:37 AM Lonnie Abelbeck<br class=""><<a href="mailto:lists@lonnie.abelbeck.com" class="">lists@lonnie.abelbeck.com</a>> wrote:<br class=""><blockquote type="cite" class="">Hunk #1 only does the trick, though performance is ever so slightly slower than before overall.<br class=""></blockquote><br class="">It's good to hear that hunks #2 and #3 don't have much an effect,<br class="">though it does still seem to have _some_ effect.<br class=""><br class="">Looks like hunk 1 is rather worrisome though. Can you try out<br class="">https://א.cc/eaxxpxbB and let me know if it has any effect?<br class=""></blockquote><br class="">That patch, as is, is very bad<br class="">--<br class="">[SUM] 0.00-30.00 sec 1.26 GBytes 360 Mbits/sec 98 sender<br class="">[SUM] 0.00-30.03 sec 1.25 GBytes 358 Mbits/sec receiver<br class=""><br class="">I then edited the patch to add back in local_bh_disable() / local_bh_enable(), much better<br class="">--<br class="">[SUM] 0.00-30.00 sec 2.62 GBytes 751 Mbits/sec 1389 sender<br class="">[SUM] 0.00-30.00 sec 2.61 GBytes 748 Mbits/sec receiver<br class=""><br class="">essentially back to 0.0.20180531 performance, hunk #1 from previous patch and hunk #1 from the latest patch.</blockquote><div class=""><br class=""></div>Hey Jason,<div class=""><br class=""></div><div class="">I'm not sure if this helps, but the 0.0.20180620 performance loss is most noticeable on the box performing "iperf3 -s".</div><div class=""><br class=""></div><div class="">Also, here are a couple .png CPU utilizations of the "iperf3 -s" box via htop during the iperf3 test ...<br class=""><div class=""><br class=""></div><div class="">Test: 0.0.20180620</div><div class=""><br class=""></div><div class=""><img apple-inline="yes" id="7DF220A8-E066-4677-A820-65F8696CC6F3" height="127" width="782" apple-width="yes" apple-height="yes" src="cid:350BC1F9-29A2-49BA-8339-18173E46FDE7@priv.abelbeck.com" class=""></div><div class=""><br class=""></div><div class="">Test: 0.0.20180620 + hunk #1 from previous patch and hunk #1 from the latest patch</div><div class=""><br class=""></div><div class=""><img apple-inline="yes" id="67E4EBBF-302C-4CA7-BBD5-C40D1140F760" height="129" width="780" apple-width="yes" apple-height="yes" src="cid:3CF68584-069A-4728-9F19-86658095735E@priv.abelbeck.com" class=""></div><div class=""><br class=""></div><div class="">Lonnie</div><div class=""><br class=""><br class=""></div></div></body></html>