https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-15718 Lustre <p>High cpu usage seen while running a lnet_selftest with 3k-4k message size.<br/> based on perf top it cause a spin lock contention in some area - 60%-70% cpu time spent for spin lock.<br/> Looking a lnet selftest code founds an global spin lock protected whole work - new match bits allocation, stats, etc..<br/> let's replace this lock is with atomic counters. this is reduce a lock contention by two.<br/> second half separated by lnet resource and </p> LU-15718

lnet_selftest performance issues.

Bug Critical Resolved Fixed Alexey Lyashkov Alexey Lyashkov Tue, 5 Apr 2022 15:00:40 +0000 Mon, 6 Jun 2022 13:26:34 +0000 Mon, 6 Jun 2022 13:26:34 +0000 Lustre 2.16.0 0 3 <p>Thanks for the report shadow, are you working on a patch for this now?</p> <p>Colinm yes - I working on patch now. I should be ready son.</p> <p>"Oleg Drokin &lt;green@whamcloud.com&gt;" merged in patch <a href="https://review.whamcloud.com/47002/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/47002/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-15718" title="lnet_selftest performance issues." class="issue-link" data-issue-key="LU-15718"><del>LU-15718</del></a> lnet: improve lnet_selftest speed<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: dd5aa640781d8c6dcbbba46135e8b28b532a3840</p> <p>Landed for 2.16</p> Related Development Rank 1|i02mi7: Rank (Obsolete) 9223372036854775807 Severity