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-13745 Lustre <p><tt>copy_file_range(2)</tt> reads from a Lustre file hang.</p> <p>With no <tt>.copy_file_range</tt> VFS API implemented it calls <tt>do&#95;splice&#95;direct()&#45;&gt;splice&#95;direct&#95;to&#95;actor()&#45;&gt;do&#95;splice&#95;to()&#45;&gt;ll&#95;file&#95;splice&#95;read()</tt>.</p> <p>While the call chain of <tt>ll&#95;file&#95;splice&#95;read()&#45;&gt;ll&#95;file&#95;io&#95;generic()&#45;&gt;generic&#95;file&#95;splice&#95;read()&#45;&gt;ll&#95;file&#95;read&#95;iter()&#45;&gt;ll&#95;file&#95;io&#95;generic()</tt>.</p> <p>And that would try to get LDLM lock twice in <tt>ll&#95;file&#95;io&#95;generic()</tt>, so that hang ensued. </p> LU-13745

tasks hang with copy_file_range: ll_file_splice_read()

Bug Major Open Unresolved Zhenyu Xu Zhenyu Xu always_except Thu, 2 Jul 2020 14:23:58 +0000 Wed, 23 Dec 2020 23:58:25 +0000 0 11 <p>Bobi Jam (bobijam@hotmail.com) uploaded a new patch: <a href="https://review.whamcloud.com/39246" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/39246</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> llite: avoid lock ldlm twice in splice read<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 4f31a54bad899b69d775f971d572b6536b6da0b0</p> <p>James Simmons (jsimmons@infradead.org) uploaded a new patch: <a href="https://review.whamcloud.com/39272" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/39272</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> llite: switch generic_file_splice_read() to use of -&gt;read_iter()<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 5a957f5de38543418c5af9b0435299aace81062c</p> <p>Wang Shilong (wshilong@whamcloud.com) uploaded a new patch: <a href="https://review.whamcloud.com/39695" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/39695</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> test: add splice test for lustre<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: f997041face6a8ce5c60af1e37da3a5368194083</p> <p>James Simmons (jsimmons@infradead.org) uploaded a new patch: <a href="https://review.whamcloud.com/39910" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/39910</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> test: add PCC test for splice handling<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 81abcfe3241090a61af2dcaee3bc1dd9b1023d4b</p> <p>Oleg Drokin (green@whamcloud.com) merged in patch <a href="https://review.whamcloud.com/39272/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/39272/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> llite: switch generic_file_splice_read() to use of -&gt;read_iter()<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 1635dc9de0bc1d6701ca5f4bc0d342fca416f89a</p> <p>New test need to land still</p> <p>Oleg Drokin (green@whamcloud.com) merged in patch <a href="https://review.whamcloud.com/39695/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/39695/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> test: add splice test for lustre<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 7ad8bf544f4132c7a291c2e8a2035afcb98f94c6</p> <p>The landing of patch <a href="https://review.whamcloud.com/39695" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/39695</a> "<tt><a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> test: add splice test for lustre</tt>" has caused 100% sanity.sh test_426 failure on aarch64.</p> <p>Andreas Dilger (adilger@whamcloud.com) uploaded a new patch: <a href="https://review.whamcloud.com/40326" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/40326</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> tests: skip sanity test_426 for 4.18+<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 37563cfa713b2bfe8a71dc1b23fea66f6329f4c5</p> <p><a href="https://jira.whamcloud.com/secure/ViewProfile.jspa?name=adilger" class="user-hover" rel="adilger">adilger</a> That means the newly added test is helpful<img class="emoticon" src="https://jira.whamcloud.com/images/icons/emoticons/smile.png" height="16" width="16" align="absmiddle" alt="" border="0"/> and something we need fix still</p> <p><a href="https://jira.whamcloud.com/secure/ViewProfile.jspa?name=adilger" class="user-hover" rel="adilger">adilger</a> this may be of interest, but I think my patch <a href="https://review.whamcloud.com/35856" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/35856</a> has also failed 100% of "test review-ldiskfs-ubuntu on CentOS 7.8/x86_64, Ubuntu 18.04/x86_64" stage I have attempted, for the same sanity/test_426 crash on Client side that you have provided the significant stack in <a href="https://jira.whamcloud.com/browse/LU-14045" class="external-link" rel="nofollow">LU-14045</a>. So it should be more a Kernel v4.x related issue than an arch related one.</p> <p>Oleg Drokin (green@whamcloud.com) merged in patch <a href="https://review.whamcloud.com/40326/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/40326/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> tests: skip sanity test_426 for 4.18+<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 010425898fa4b2abc6325a8073e20cb994ce7947</p> <p>This is also failing on Ubuntu 18.04 which is using 4.15.0-72-generic. See <a href="https://testing.whamcloud.com/test_sets/bdfb8c4b-a6a2-493a-ab57-6a9923f96e7c" class="external-link" target="_blank" rel="nofollow noopener">https://testing.whamcloud.com/test_sets/bdfb8c4b-a6a2-493a-ab57-6a9923f96e7c</a>.</p> <p><a href="https://jira.whamcloud.com/secure/ViewProfile.jspa?name=adilger" class="user-hover" rel="adilger">adilger</a>, after I have rebased my patch on top of change #40326, it appears that same crash during sanity/test_426 still occurs with Ubuntu Client because the used Kernel version is 4.15 so sub-test is not skipped !!</p> <p>John L. Hammond (jhammond@whamcloud.com) uploaded a new patch: <a href="https://review.whamcloud.com/40366" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/40366</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> tests: skip sanity test_426 for 4.15+<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: a8a6a55a1e4e11ee3b61aa7e230d752b5c1a476a</p> <p>Oleg Drokin (green@whamcloud.com) merged in patch <a href="https://review.whamcloud.com/40366/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/40366/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> tests: skip sanity test_426 for 4.15+<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: f8a8d3f83db67be9dcc724ff49757cce81b13a5e</p> <p>Yingjin Qian (qian@ddn.com) uploaded a new patch: <a href="https://review.whamcloud.com/40396" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/40396</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> pcc: fall back normal splice read for detached file<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 7b45c8d9c77a9c45862bd61ea05b8c46117cffa4</p> <p>Oleg Drokin (green@whamcloud.com) merged in patch <a href="https://review.whamcloud.com/40396/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/40396/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13745" title="tasks hang with copy_file_range: ll_file_splice_read()" class="issue-link" data-issue-key="LU-13745">LU-13745</a> pcc: fall back normal splice read for detached file<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: cca45ad8aeaa8e124e9e48361bf7cff89a035f82</p> <p>The fix itself has landed for 2.14. All that remains tracked by this ticket is a test. Are there still plans to land that test imminently or can we either abandon that changeset /move it to a new JIRA?</p> Related LU-12425 LU-14060 LU-14045 Development Rank 1|i014br: Rank (Obsolete) 9223372036854775807 Severity