[dpdk-stable] [dpdk-dev] 19.11.9 patches review and test - V2

Christian Ehrhardt christian.ehrhardt at canonical.com
Mon Jun 14 07:20:06 CEST 2021


On Fri, Jun 11, 2021 at 10:55 AM Jiang, YuX <yux.jiang at intel.com> wrote:
>
> Hi Christian,
>
> Testing with dpdk v19.11.9-rc2 from Intel is almost finished, some build issues and telemetry issue are found.

Thank you for completing this on -rc2 and thereby giving us a preview
of the final results.

> # Basic Intel(R) NIC testing
> * PF(i40e, ixgbe): test scenarios including rte_flow/TSO/Jumboframe/checksum offload/Tunnel, etc. Listed but not all.
>         - Known issues as below:
>         1) https://bugs.dpdk.org/show_bug.cgi?id=731 – [dpdk-19.11.9] telemetry: python3 /root/dpdk/usertools/dpdk-telemetry-client.py prompt connection refused)
>                 Fixed commit  is af68c1d699be6c369e296b775bdbf13ae18b79cc eal: fix hang in control thread creation

This one was in the candidate list of patches for 19.11.x, but it
didn't apply cleanly.
I've asked for a backport in my mail on 20th May - in this thread [1]
But in the meantime that was all done and -rc2 that you have tested contains

34cc55cce6b1 ("eal: fix race in control thread creation")
af68c1d699be ("eal: fix hang in control thread creation")

In the form of these backports:
$ git tag --contains 555da24882e1ca836f9532304f5f58034e57ff5e
v19.11.9-rc2
$ git tag --contains 0b8f4f0462f2260fc618e297e91317a891c6d255
v19.11.9-rc2

So if af68c1d699be really was meant to fix this known issue then it
does not do so in 19.11.x.
@Luc - I'm setting you to CC in case you happen to know why.

>         2) https://bugs.dpdk.org/show_bug.cgi?id=687: unit_tests_power/power_cpufreq: unit test failed. This issue is found in 21.05 and dev has patches to fix it but not merged into main.

Ok, so hopefully this will eventually be fixed in 21.08 and then also
in the respective 19.11.x afterwards.

> * VF(i40e,ixgbe): test scenarios including vf-rte_flow/TSO/Jumboframe/checksum offload/Tunnel, Listed but not all.  - No new issues are found.
> * PF/VF(ice): test scenarios including switch features/Flow Director/Advanced RSS/ACL/DCF/Flexible Descriptor and so on, Listed but not all.
>         - One 4 known issues is found, but won't fix on LTS19.11. [dpdk-19.11.7] userspace_ethtool/retrieve_eeprom: Dumped eeprom not same as linux dumped. Fixed patches are based on 4 ice share code, not be contained in LTS19.11.
> * Build or compile:
>         * Build: cover the build test combination with different GCC/Clang and the popular OS revision such as Ubuntu20.04, CentOS8.3, SUSE15, Fedora34 and so on. Listed but not all.
>         - Two new issues are found
>                 1) https://bugs.dpdk.org/show_bug.cgi?id=732 [dpdk-19.11.9-rc2] DPDK build failed on Win10
>                 2) https://bugs.dpdk.org/show_bug.cgi?id=728 [dpdk-19.11.9-rc2]When compiling dpdk with make or meson under SUSE Linux Enterprise Server 15 SP2, the compilation fails

Known - currently we have to decide if we break SP3 (no change) or fix
SP3 but thereby break SP2 (that is what was in -rc2).
DIscussions and mail threads are ongoing, Ferruh and I tried to pull
in Suse people as well since it all started with a kernel change
backport.

>                 - Known issue is Fedora34 GCC11 and Clang12 build issue
>                 a.GCC11 issue is same as https://bugs.dpdk.org/show_bug.cgi?id=692 : Bug 692 - bnx2x build fail on Fedora 34 with gcc 11. This issue is found in 21.05 with Fedora34 GCC 11. Has patches to fix in 21.05 and merged into main.
>                         Applies below tree patches can build passed. Patches link:
>                                 [v3,1/4] net/bnx2x: fix build with gcc11 - Patchwork (dpdk.org)
>                                 [v3,2/4] net/bnx2x: fix build with gcc11 - Patchwork (dpdk.org)
>                                 test/table: fix build with GCC 11 - Patchwork (dpdk.org)
>                 b.Clang12 issue: app/test/dpdk-test.p/test_cmdline_num.c.o build failed on Fedora34 with Clang12.
>                         Fixed commits are 5ac070cfed test/cmdline: silence clang 12 warning &   414245bbc5 test/cmdline: fix inputs array(Refer to Kevin's reply)

Thanks for confirming, the coming -rc3 (still a while to give everone
a chance for more backports) will have those fixed.
If anyone needs it asap, it already is in the WIP branch on
https://github.com/cpaelzer/dpdk-stable-queue/commits/19.11

> * Intel NIC single core/NIC performance: test scenarios including PF/VF single core performance test(AVX2+AVX512) test and so on. Listed but not all.   - All passed. No big data drop.
>
> # Basic cryptodev and virtio testing
> * Virtio: both function and performance test are covered. Such as PVP/Virtio_loopback/virtio-user loopback/virtio-net VM2VM perf testing, etc.. Listed but not all.
>         - One known issues about The UDP fragmentation offload feature of Virtio-net device can’t be turned on in the VM, kernel issue, bugzilla has been submited: https://bugzilla.kernel.org/show_bug.cgi?id=207075, not fixed yet.
> * Cryptodev:
>         - Function test: test scenarios including Cryptodev API testing/CompressDev ISA-L/QAT/ZLIB PMD Testing/FIPS, etc. Listed but not all. - All passed.
>         - Performance test: test scenarios including Thoughput Performance /Cryptodev Latency, etc. Listed but not all.         - No big data drop.
>
> Best regards,
> Yu Jiang
>
> >-----Original Message-----
> >From: dev <dev-bounces at dpdk.org> On Behalf Of Christian Ehrhardt
> >Sent: Thursday, June 10, 2021 9:57 PM
> >To: Kevin Traynor <ktraynor at redhat.com>
> >Cc: Ali Alnubani <alialnu at nvidia.com>; dpdk stable <stable at dpdk.org>; dev
> ><dev at dpdk.org>; Abhishek Marathe <Abhishek.Marathe at microsoft.com>;
> >Akhil Goyal <akhil.goyal at nxp.com>; Walker, Benjamin
> ><benjamin.walker at intel.com>; David Christensen <drc at linux.vnet.ibm.com>;
> >Govindharajan, Hariprasad <hariprasad.govindharajan at intel.com>; Hemant
> >Agrawal <hemant.agrawal at nxp.com>; Stokes, Ian <ian.stokes at intel.com>;
> >Jerin Jacob <jerinj at marvell.com>; Mcnamara, John
> ><john.mcnamara at intel.com>; Ju-Hyoung Lee <juhlee at microsoft.com>; Luca
> >Boccassi <bluca at debian.org>; Pei Zhang <pezhang at redhat.com>; Yu, PingX
> ><pingx.yu at intel.com>; Xu, Qian Q <qian.q.xu at intel.com>; Raslan Darawsheh
> ><rasland at nvidia.com>; NBU-Contact-Thomas Monjalon
> ><thomas at monjalon.net>; Peng, Yuan <yuan.peng at intel.com>; Chen,
> >Zhaoyan <zhaoyan.chen at intel.com>
> >Subject: Re: [dpdk-dev] 19.11.9 patches review and test - V2
> >
> >On Thu, Jun 10, 2021 at 1:03 PM Kevin Traynor <ktraynor at redhat.com> wrote:
> >>
> >> On 10/06/2021 11:29, Kevin Traynor wrote:
> >> > On 10/06/2021 11:05, Christian Ehrhardt wrote:
> >> >> On Thu, Jun 10, 2021 at 11:05 AM Ali Alnubani <alialnu at nvidia.com>
> >wrote:
> >> >>>
> >> >>> Hi Christian,
> >> >>>
> >> >>>> -----Original Message-----
> >> >>>> From: Christian Ehrhardt <christian.ehrhardt at canonical.com>
> >> >>>> Sent: Thursday, June 10, 2021 11:49 AM
> >> >>>> To: dpdk stable <stable at dpdk.org>
> >> >>>> Cc: dev <dev at dpdk.org>; Abhishek Marathe
> >> >>>> <Abhishek.Marathe at microsoft.com>; Akhil Goyal
> >> >>>> <akhil.goyal at nxp.com>; Ali Alnubani <alialnu at nvidia.com>; Walker,
> >> >>>> Benjamin <benjamin.walker at intel.com>; David Christensen
> >> >>>> <drc at linux.vnet.ibm.com>; Govindharajan, Hariprasad
> >> >>>> <hariprasad.govindharajan at intel.com>; Hemant Agrawal
> >> >>>> <hemant.agrawal at nxp.com>; Ian Stokes <ian.stokes at intel.com>;
> >> >>>> Jerin Jacob <jerinj at marvell.com>; John McNamara
> >> >>>> <john.mcnamara at intel.com>; Ju-Hyoung Lee
> ><juhlee at microsoft.com>;
> >> >>>> Kevin Traynor <ktraynor at redhat.com>; Luca Boccassi
> >> >>>> <bluca at debian.org>; Pei Zhang <pezhang at redhat.com>; Yu, PingX
> >> >>>> <pingx.yu at intel.com>; Xu, Qian Q <qian.q.xu at intel.com>; Raslan
> >> >>>> Darawsheh <rasland at nvidia.com>; NBU- Contact-Thomas Monjalon
> >> >>>> <thomas at monjalon.net>; Peng, Yuan <yuan.peng at intel.com>; Chen,
> >> >>>> Zhaoyan <zhaoyan.chen at intel.com>
> >> >>>> Subject: Re: [dpdk-dev] 19.11.9 patches review and test - V2
> >> >>>>
> >> >>>> On Fri, Jun 4, 2021 at 7:52 AM Christian Ehrhardt
> >> >>>> <christian.ehrhardt at canonical.com> wrote:
> >> >>>>>
> >> >>>>> Hi all,
> >> >>>>>
> >> >>>>> Here is version 2 of the list of patches targeted for stable release
> >19.11.9.
> >> >>>>> Thanks to plenty of helpful developers we've collected a few
> >> >>>>> more backports by now and sorted out a few rare compile time
> >> >>>>> issues that were
> >> >>>> found with -rc1.
> >> >>>>>
> >> >>>>> The planned date for the final release of 19.11.9 is now 18th of June.
> >> >>>>>
> >> >>>>> Please help with testing and validation of your use cases and
> >> >>>>> report any issues/results with reply-all to this mail. For the
> >> >>>>> final release the fixes and reported validations will be added to the
> >release notes.
> >> >>>>>
> >> >>>>> A release candidate tarball can be found at:
> >> >>>>>
> >> >>>>>     https://dpdk.org/browse/dpdk-stable/tag/?id=v19.11.9-rc2
> >> >>>>
> >> >>>> Unfortunately there were further build issues with new suse kernels.
> >> >>>> The fix for some of them, now breaks others :-/ The final fix
> >> >>>> didn't finalize yet, but as a TL;DR it means there will be a -rc3 down the
> >road.
> >> >>>>
> >> >>>> I beg your pardon for the extra work.
> >> >>>> If you are testing non-suse and are already deep into the tests
> >> >>>> you likely can continue that.
> >> >>>> If you didn't start your tests yet then I'd recommend holding
> >> >>>> back until -rc3 exists.
> >> >>>>
> >> >>>
> >> >>> I see some build failures with gcc 11. Do we expect this release to
> >support it?
> >> >>
> >> >> I have added all that were targeted at stable at dpdk, but I remember
> >> >> that Kevin mentioned a few missing in 20.11.x due to the
> >> >> submissions missing to go to stable at dpdk.
> >> >> The same could be the case here.
> >> >> @Kevin could you have a look at 19.11 and if there are some missing
> >> >> point me to the commits to pick up?
> >> >>
> >> >> @Ali - if you have a particular bug signature that might help to
> >> >> identify the right patch.
> >> >>
> >> >
> >> > They will all note 'GCC' (or gcc) in the commit msg. Most will have
> >> > an example of the 'warning' they resolved too. There was also
> >> > bugzillas opened for them all i think, which note the fixing commit.
> >> > So if you find any issues when building, you could cross-reference
> >> > between warning in commit/warning in Bz and fixing commit. Of course
> >> > there may be some new issues pop up in older code that was later
> >changed/replaced in main.
> >> >
> >> > There was also 2 commits needed for clang 12 support (my bad for
> >> > missing tags on these).
> >>
> >> > 5ac070cfed test/cmdline: silence clang 12 warning
> >> ^ just sent backport of this with upstream commit and correct tag
> >> http://inbox.dpdk.org/stable/20210610110028.230803-1-ktraynor@redhat.c
> >> om/T/#u
> >
> >I've already applied it a few hours ago, but the content is the same
> >
> >> > 414245bbc5 test/cmdline: fix inputs array
> >> ^ this one was already merged on 19.11 and 20.11 branches
> >
> >Yes, as are all the other GCC 11 fixes we discussed.
> >
> >Thank you Kevin!
> >
> >> >
> >> >>> I also see some doxygen errors with version 1.9.1 in Fedora Rawhide.
> >Will open a Bugzilla ticket.
> >> >>
> >> >> Rawhide builds of 19.11.x were broken for a while now, but yeah
> >> >> filing the bug surely is the right path.
> >> >>
> >> >>> Thanks,
> >> >>> Ali
> >> >>
> >> >>
> >> >>
> >> >
> >>
> >
> >
> >--
> >Christian Ehrhardt
> >Staff Engineer, Ubuntu Server
> >Canonical Ltd



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd


More information about the stable mailing list