<div dir="ltr"><div class="gmail_default" style="font-size:small">We can look into that, but that now will need to be tested to work across all the different OS distros in the containers. </div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">For now, we can install the update on the ubuntu 18.04 worker that is running the production and remake the reference cache.</div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">Cheers,</div><div class="gmail_default" style="font-size:small">Lincoln</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, Apr 16, 2021 at 8:58 AM Thomas Monjalon <<a href="mailto:thomas@monjalon.net">thomas@monjalon.net</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">16/04/2021 14:43, Lincoln Lavoie:<br> > All of the UNH ABI testing is moving info containers, so it can be run on<br> > top of each OS, alongside the other compile and unit testing. This is<br> > actually ready now, but hasn't been pushed live this week, because of the<br> > backlog in the system because of the DTS failure. The additional compile<br> > jobs are already online now, it's just ABI that hasn't been pushed live.<br> > <br> > This means the current ABI (what is reporting right now) is running on<br> > 18.04 for x86 and 20.04 for aarch64. The aarch64 one will continue<br> > forward, because we're not going to moving to emulated environments for<br> > testing on that architecture.<br> > <br> > This has two implications, first, the scripts for running ABI (and the<br> > other tests) become part of the container definition, and at the last<br> > meeting we talked about moving those definitions into the dpdk-ci repo,<br> > which I think makes sense. Second, there isn't an operating system to<br> > "maintain" since it's what's inside the container images, which are<br> > periodically rebuilt, but pretty much treated as ephemeral. Assuming the<br> > container bases / distros have the updated libabigail version packaged with<br> > them.<br> <br> No, the version packaged in the OS is not recent enough.<br> Please check what is done in Travis and GitHub CI<br> in the shell function install_libabigail():<br> <a href="https://git.dpdk.org/dpdk/tree/.ci/linux-build.sh#n22" rel="noreferrer" target="_blank">https://git.dpdk.org/dpdk/tree/.ci/linux-build.sh#n22</a><br> <br> <br> > On Fri, Apr 16, 2021 at 8:32 AM Aaron Conole <<a href="mailto:aconole@redhat.com" target="_blank">aconole@redhat.com</a>> wrote:<br> > > David Marchand <<a href="mailto:david.marchand@redhat.com" target="_blank">david.marchand@redhat.com</a>> writes:<br> > ><br> > > > On Tue, Apr 13, 2021 at 5:15 AM Xueming Li <<a href="mailto:xuemingl@nvidia.com" target="_blank">xuemingl@nvidia.com</a>> wrote:<br> > > >> diff --git a/lib/librte_eal/include/rte_devargs.h<br> > > b/lib/librte_eal/include/rte_devargs.h<br> > > >> index 296f19324f..134b44a887 100644<br> > > >> --- a/lib/librte_eal/include/rte_devargs.h<br> > > >> +++ b/lib/librte_eal/include/rte_devargs.h<br> > > >> @@ -60,16 +60,16 @@ struct rte_devargs {<br> > > >> /** Name of the device. */<br> > > >> char name[RTE_DEV_NAME_MAX_LEN];<br> > > >> RTE_STD_C11<br> > > >> - union {<br> > > >> - /** Arguments string as given by user or "" for no argument. */<br> > > >> - char *args;<br> > > >> + union { /**< driver-related part of device string. */<br> > > >> + const char *args; /**< legacy name. */<br> > > >> const char *drv_str;<br> > > >> };<br> > > >> struct rte_bus *bus; /**< bus handle. */<br> > > >> struct rte_class *cls; /**< class handle. */<br> > > >> const char *bus_str; /**< bus-related part of device string. */<br> > > >> const char *cls_str; /**< class-related part of device string.<br> > > */<br> > > >> - const char *data; /**< Device string storage. */<br> > > >> + char *data;<br> > > >> + /**< Raw string including bus, class and driver arguments. */<br> > > >> };<br> > > >><br> > > >> /**<br> > > ><br> > > > - Flagging this patch for info and its impact on UNH jobs.<br> > > ><br> > > > This change is fine, but older libabigail versions could not deal with<br> > > > such changes (anonymous union, changes of const fields).<br> > > > This results in an ABI check failure in the UNH x86 job on Ubuntu<br> > > > 18.04 (and for some people not using recent libabigail).<br> > > > I can see the ARM job passes fine, so I suppose it is using a more<br> > > > recent libabigail (running Ubuntu 20.04 maybe?).<br> > > ><br> > > > We either need to disable this x86 job or update its libabigail<br> > > > package (maybe aligning with what we have for public CI which is<br> > > > libabigail 1.8 manually compiled).<br> > > ><br> > > ><br> > > > - For the longer term, what do you think of using/extending the .ci/<br> > > > scripts for use by UNH jobs?<br> > ><br> > > I think it would be great if we had some of the scripts shared as a<br> > > common resource. That would also help us to look at fixes / changes<br> > > when needed.<br> > ><br> > ><br> > <br> > <br> <br> <br> <br> <br> <br> </blockquote></div><br clear="all"><div><br></div>-- <br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div><b>Lincoln Lavoie</b><br></div><div>Principal Engineer, Broadband Technologies</div><div>21 Madbury Rd., Ste. 100, Durham, NH 03824</div><div><a href="mailto:lylavoie@iol.unh.edu" target="_blank">lylavoie@iol.unh.edu</a></div><div><a href="https://www.iol.unh.edu" target="_blank">https://www.iol.unh.edu</a></div><div>+1-603-674-2755 (m)</div><div><a href="https://www.iol.unh.edu" target="_blank"><img src="https://docs.google.com/uc?export=download&id=1j_iI6anwrnbQWNpTyuvukMLSNJJ8_8QU&revid=0B_0ujwABDnFZTmJiR3EzK0d1VjFKTjQvMENBWVM0QnA4ajhjPQ" width="200" height="43"></a><br></div></div></div></div></div></div></div></div></div></div></div>