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-4266 Lustre <p>Build lustre using OFED 3.5-1 fail</p> <p>++ echo 3.5-1<br/> ++ sed <del>re 's/</del>(<span class="error">&#91;0-9&#93;</span><b>-<span class="error">&#91;rR&#93;</span><span class="error">&#91;cC&#93;</span><span class="error">&#91;0-9&#93;</span></b>)$//'<br/> + ofed_version=3.5-1<br/> ++ ls '/mnt/build/build/b25/BUILD/RPMS/<b>/compat-rdma-devel-3.5-1-2.6.32_358.18.1.el6.x86_64.</b>.rpm'<br/> ls: cannot access /mnt/build/build/b25/BUILD/RPMS/<b>/compat-rdma-devel-3.5-1-2.6.32_358.18.1.el6.x86_64.</b>.rpm: No such file or directory<br/> + local rpm=<br/> + rpm2cpio<br/> + cpio -id</p> LU-4266

fix lbuild script to work with OFED 3.5-x

Bug Minor Resolved Fixed Minh Diep Minh Diep Mon, 18 Nov 2013 16:31:30 +0000 Thu, 22 May 2014 18:35:22 +0000 Mon, 16 Dec 2013 16:48:14 +0000 Lustre 2.6.0 Lustre 2.5.2 0 10 <p><a href="http://review.whamcloud.com/#/c/8319/" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/#/c/8319/</a></p> <p>but build failed with the following unrelated to OFED. It's interesting that with in-kernel ofed, it passed.</p> <p>In file included from include/trace/ftrace.h:440,<br/> from include/trace/define_trace.h:73,<br/> from /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h:904,<br/> from /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/super.c:56:<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h: In function 'ftrace_profile_enable_ldiskfs_free_inode':<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h:18: error: implicit declaration of function 'register_trace_ldiskfs_free_inode'<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h: In function 'ftrace_profile_disable_ldiskfs_free_inode':<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h:18: error: implicit declaration of function 'unregister_trace_ldiskfs_free_inode'<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h: In function 'ftrace_profile_enable_ldiskfs_request_inode':<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h:47: error: implicit declaration of function 'register_trace_ldiskfs_request_inode'<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h: In function 'ftrace_profile_disable_ldiskfs_request_inode':<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h:47: error: implicit declaration of function 'unregister_trace_ldiskfs_request_inode'<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h: In function 'ftrace_profile_enable_ldiskfs_allocate_inode':<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h:69: error: implicit declaration of function 'register_trace_ldiskfs_allocate_inode'<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h: In function 'ftrace_profile_disable_ldiskfs_allocate_inode':<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h:69: error: implicit declaration of function 'unregister_trace_ldiskfs_allocate_inode'<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h: In function 'ftrace_profile_enable_ldiskfs_write_begin':<br/> /mnt/build/build/master-ofed/BUILD/BUILD/lustre-2.5.51/ldiskfs/trace/events/ldiskfs.h:121: error: implicit declaration of function 'register_trace_ldiskfs_write_begin'<br/> ....</p> <p>There was a mention about this failure in <a href="https://jira.whamcloud.com/browse/LU-3462" title="Eliminate ldiskfs recursive &amp; independent rpm packaging" class="issue-link" data-issue-key="LU-3462"><del>LU-3462</del></a> but no further follow up</p> <p>The patch <a href="http://review.whamcloud.com/8404" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/8404</a> should fix this issue.</p> <p>the above patch did not solve the issue. there seems to be something else. investigating</p> <p>I fixed it with fixing OFED codes. This causes tracepoint evenets are backported in RHEL6's kernel below.</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>TRACE_EVENT and DEFINE_EVENT are defined in ofa_kernel/default/include/linux/tracepoint.h #if (LINUX_VERSION_CODE &lt;= KERNEL_VERSION(2,6,32)) /* * Disable all tracing for older kernels * &lt; 2.6.27 had no tracing * 2.6.27 had broken tracing * 2.6.28-2.6.32 didn't have anything like DECLARE_EVENT_CLASS * and faking it would be extremely difficult */backported #if (LINUX_VERSION_CODE &gt;= KERNEL_VERSION(2,6,28)) /* * For 2.6.28+ include the original tracepoint.h but override * the defines new code uses to disable tracing completely. */ #include_next &lt;linux/tracepoint.h&gt; #endif #undef TRACE_EVENT #define TRACE_EVENT(name, proto, ...) \ static inline void trace_ ## name(proto) {} #undef DECLARE_EVENT_CLASS #define DECLARE_EVENT_CLASS(...) #undef DEFINE_EVENT #define DEFINE_EVENT(evt_class, name, proto, ...) \ static inline void trace_ ## name(proto) {} </pre> </div></div> <p>We might need more proper fixes, but as a quick workaround is adding "!defined(CONFIG_COMPAT_RHEL_6_4)" to skip trace events for patched lustre kernel below. </p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>--- ofa_kernel/default/include/linux/tracepoint.h.orig 2013-11-05 03:11:23.000000000 -0800 +++ ofa_kernel/default/include/linux/tracepoint.h 2013-11-05 03:13:08.000000000 -0800 @@ -3,7 +3,7 @@ #include &lt;linux/version.h&gt; -#if (LINUX_VERSION_CODE &lt;= KERNEL_VERSION(2,6,32)) +#if (LINUX_VERSION_CODE &lt;= KERNEL_VERSION(2,6,32) &amp;&amp; !defined(CONFIG_COMPAT_RHEL_6_4)) /* * Disable all tracing for older kernels * &lt; 2.6.27 had no tracing @@ -35,6 +35,6 @@ static inline void trace_ ## name(proto) /* since 2.6.33, tracing hasn't changed, so just include the kernel's file */ #include_next &lt;linux/tracepoint.h&gt; -#endif /* (LINUX_VERSION_CODE &lt;= KERNEL_VERSION(2,6,32)) */ +#endif /* (LINUX_VERSION_CODE &lt;= KERNEL_VERSION(2,6,32) &amp;&amp; !defined(CONFIG_COMPAT_RHEL_6_4)) */ #endif /* _COMPAT_LINUX_TRACEPOINT_H */ </pre> </div></div> <p>I can confirm with this patch I was able to compile OFED-3.5-1 and OFED-3.5-2-rc2. But we have an issue with specifying our patch for OFED in "contrib/patches/ofed". The script "01-remove-mlx4-erroneous-modprobe-config-<a href="file:rhel6.ed" class="external-link" target="_blank" rel="nofollow noopener">file:rhel6.ed</a>" is not applied to OFED spec and thanks this other patched will not be applied also.</p> <p>that's differnt problem. please use OFED-3.5.2-daily build. previous version doesn't work on RHEL6.4</p> <p>Ihara, could you explain why OFED-3.5-1 doesn't work on RHEL6.4?</p> <p>I have a patch here that patch ofed and built OFED-3.5-1 successfully.</p> <p><a href="http://review.whamcloud.com/#/c/8319/" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/#/c/8319/</a></p> <p>Ihara, Dmitry,</p> <p>please review the patch. I will do another around of clean up the comments and remove the hack to build OFED 3.5 (since our lab doesn't do that at the moment)</p> <p>Actually I think we used wrong way to adopt OFED for our build. Why we use OFED headers for fdiskfs build? Probably we should use OFED headers only for sources that required them. Otherwise such backport issues will always affect us. Could we change build scripts to avoid using OFED headers for all sources?</p> <p>Include any OFED backport headers in all compile commands was introduced in commit 70eca9ed3d7408bebeda59ad65ea4fc1ca9bf57b with following comment:</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java">+# it's ugly to be doing anything with OFED outside of the lnet module, but +# <span class="code-keyword">this</span> has to be done here so that the backports path is set before all of +# the LN_PROG_LINUX checks are done +LB_CONFIG_OFED_BACKPORTS </pre> </div></div> <p>Does anybody knows why this done? Why we need backport headers in all compile commands?</p> <p>Just removed OFED backport headers for all components except lnet and move all OFED related autoconf testing into lnet/autoconf in patch <a href="http://review.whamcloud.com/8451" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/8451</a></p> <p>Hello Dmitry,<br/> I tested your patch <a href="http://review.whamcloud.com/#/c/8451/1" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/#/c/8451/1</a> to build OFED-3.5-2-rc2 ( for another issue <a href="https://jira.whamcloud.com/browse/LU-3389" title="Lustre b2_1 build failed on RHEL6.4 with OFA IB stack" class="issue-link" data-issue-key="LU-3389"><del>LU-3389</del></a> reported by Wally ). The build completed successfully. I did remove the hard coded value of OFED_VERSION.<br/> Let me know if I should be updating the logs for your reference.</p> <p>Thanks</p> <p>Thanks for testing. But it looks we have an issue with OFED-1.5.3.1:</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java">/mnt/build/build/lu4266/BUILD/BUILD/lustre-2.5.52/lnet/klnds/o2iblnd/o2iblnd.c: In function <span class="code-quote">'kiblnd_dev_need_failover'</span>: /mnt/build/build/lu4266/BUILD/BUILD/lustre-2.5.52/lnet/klnds/o2iblnd/o2iblnd.c:2565: error: too few arguments to function <span class="code-quote">'rdma_create_id'</span> /mnt/build/build/lu4266/BUILD/BUILD/lustre-2.5.52/lnet/klnds/o2iblnd/o2iblnd.c: In function <span class="code-quote">'kiblnd_dev_failover'</span>: /mnt/build/build/lu4266/BUILD/BUILD/lustre-2.5.52/lnet/klnds/o2iblnd/o2iblnd.c:2639: error: too few arguments to function <span class="code-quote">'rdma_create_id'</span> make[7]: *** [/mnt/build/build/lu4266/BUILD/BUILD/lustre-2.5.52/lnet/klnds/o2iblnd/o2iblnd.o] Error 1 make[6]: *** [/mnt/build/build/lu4266/BUILD/BUILD/lustre-2.5.52/lnet/klnds/o2iblnd] Error 2 make[5]: *** [/mnt/build/build/lu4266/BUILD/BUILD/lustre-2.5.52/lnet/klnds] Error 2 make[4]: *** [/mnt/build/build/lu4266/BUILD/BUILD/lustre-2.5.52/lnet] Error 2 make[4]: *** Waiting <span class="code-keyword">for</span> unfinished jobs.... </pre> </div></div> <p>I need to fix this.</p> <p>The Patch Set 2 works for all OFED versions.</p> <p>Same here. Your patch address my issues as well. I have these same issues on the 2.5 branch as well so this will need to be backported.</p> <p>Yes, let's update the fix version to 2.5.1</p> <p>Patch for b2_5 is at <a href="http://review.whamcloud.com/#/c/9109" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/#/c/9109</a></p> Blocker Development Rank 1|hzw9lz: Rank (Obsolete) 11719 Severity