patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Christian Ehrhardt <christian.ehrhardt@canonical.com>
To: "Jiawei(Jonny) Wang" <jiaweiw@nvidia.com>
Cc: dpdk stable <stable@dpdk.org>, Ori Kam <orika@nvidia.com>
Subject: Re: [dpdk-stable] " app/testpmd: fix NVGRE encap configuration" incompatible with 19.11.x
Date: Tue, 8 Jun 2021 06:50:41 +0200	[thread overview]
Message-ID: <CAATJJ0KpHqfhHrcn6Dc_39s5X5GP6nOh_qt4nG0P6Rz+Zu64cw@mail.gmail.com> (raw)
In-Reply-To: <BN9PR12MB5034D31CA4805BA4147C29D1C63B9@BN9PR12MB5034.namprd12.prod.outlook.com>

On Fri, Jun 4, 2021 at 3:44 PM Jiawei(Jonny) Wang <jiaweiw@nvidia.com> wrote:
>
> Hello Christian,
>
> > -----Original Message-----
> > From: Christian Ehrhardt <christian.ehrhardt@canonical.com>
> > Sent: Friday, June 4, 2021 1:07 PM
> > To: Jiawei(Jonny) Wang <jiaweiw@nvidia.com>
> > Cc: dpdk stable <stable@dpdk.org>; Ori Kam <orika@nvidia.com>
> > Subject: Re: " app/testpmd: fix NVGRE encap configuration" incompatible
> > with 19.11.x
> >
> > On Thu, Jun 3, 2021 at 4:54 PM Jiawei(Jonny) Wang <jiaweiw@nvidia.com>
> > wrote:
> > >
> > > Hello Christian,
> > >
> > > I applied this patch on the 19.11.9-rc1, but not found this issue during
> > compilation.
> > >
> > > Could you send me your code branch then I can try on my side?
> >
> > Hi,
> > For me it fails when applied on top of the 19.11.x WIP branch.
> > You can use that from
> > https://github.com/cpaelzer/dpdk-stable-queue/commits/19.11
> >
> > But as I said it only fails in some releases/configurations.
> > That is why I have linked the failing error logs.
> > For example on my Ubuntu 20.04 it builds just fine.
> >
>
> Yes, I compared and found it's related with older gcc version.
> I had sent the code patch.

I've got the patch it applies and seems to work - thank you!
It isn't critical enough to respin the release candidates though.
If we hit a functional issue in testing that will force us to respin it will
be in 19.11.9-rc3, otherwise it will be the first patch that gets
queued for 19.11.10

> > > Code base:
> > > commit 15d8f64d18a891a70d664d18eb918eee0705993e
> > > Author: Christian Ehrhardt <christian.ehrhardt@canonical.com>
> > > Date:   Wed Jun 2 16:33:26 2021 +0200
> > >
> > >     version: 19.11.9-rc1
> > >
> > >     Signed-off-by: Christian Ehrhardt
> > > <christian.ehrhardt@canonical.com>
> > >
> > >
> > > #make O=x86_64-default-linux-gcc T=x86_64-native-linux-gcc config #sed
> > > -ri 's,(RTE_MACHINE=).*,\1default,' x86_64-default-linux-gcc/.config
> > > #sed -ri 's,(RTE_APP_TEST=).*,\1n,' x86_64-default-linux-gcc/.config
> > > #sed -ri 's,(RTE_BUILD_SHARED_LIB=).*,\1y,'
> > > x86_64-default-linux-gcc/.config #sed -ri 's,(RTE_NEXT_ABI=).*,\1n,'
> > > x86_64-default-linux-gcc/.config #sed -ri 's,(LIBRTE_VHOST=).*,\1y,'
> > > x86_64-default-linux-gcc/.config #sed -ri 's,(LIBRTE_VHOST=).*,\1y,'
> > > x86_64-default-linux-gcc/.config #sed -ri 's,(RTE_KNI_KMOD=).*,\1y,'
> > > x86_64-default-linux-gcc/.config #sed -ri
> > > 's,(RTE_EAL_IGB_UIO=).*,\1y,' x86_64-default-linux-gcc/.config # #make
> > > O=x86_64-default-linux-gcc -j3 EXTRA_CFLAGS=-Werror
> >
> > Yeah as I said it seems to only affect some releases/configs - I hope the logs
> > help to find what is different between good/bad cases.
> >
> Yes, it's helpful, I had updated the code to make sure it works with older gcc version.
> Thanks.
>
> > > ......
> > >
> > >   INSTALL-APP dpdk-test-crypto-perf
> > >   INSTALL-MAP dpdk-test-crypto-perf.map
> > >   CC test_perf_queue.o
> > >   CC test_perf_atq.o
> > >   CC test_pipeline_common.o
> > >   CC test_pipeline_queue.o
> > >   CC test_pipeline_atq.o
> > >   LD dpdk-test-eventdev
> > >   INSTALL-APP dpdk-test-eventdev
> > >   INSTALL-MAP dpdk-test-eventdev.map
> > > Build complete [x86_64-default-linuxapp-gcc]
> > >
> > > Thanks.
> > > B.R.
> > >
> > > Jonny
> > >
> > >
> > > > -----Original Message-----
> > > > From: Christian Ehrhardt <christian.ehrhardt@canonical.com>
> > > > Sent: Wednesday, June 2, 2021 9:25 PM
> > > > To: Jiawei(Jonny) Wang <jiaweiw@nvidia.com>
> > > > Cc: dpdk stable <stable@dpdk.org>; Ori Kam <orika@nvidia.com>
> > > > Subject: " app/testpmd: fix NVGRE encap configuration" incompatible
> > > > with 19.11.x
> > > >
> > > > Hi,
> > > > this change was targeted for the stable releases
> > > >
> > > > commit 9b0da816bdec98c0d6c5d17ccdd337bdbafc3f75
> > > > Author: Jiawei Wang <jiaweiw@nvidia.com>
> > > > Date:   Tue Mar 16 06:18:27 2021 +0200
> > > >
> > > >     app/testpmd: fix NVGRE encap configuration
> > > >
> > > > But when applied it causes:
> > > >
> > > > [  263s] /home/abuild/rpmbuild/BUILD/dpdk-
> > > > 1622611368.078834f40/app/test-pmd/cmdline_flow.c:
> > > > In function 'parse_vc_action_nvgre_encap':
> > > > [  263s] /home/abuild/rpmbuild/BUILD/dpdk-
> > > > 1622611368.078834f40/app/test-pmd/cmdline_flow.c:4254:3:
> > > > error: missing initializer for field 'protocol' of 'struct
> > > > rte_flow_item_nvgre' [- Werror=missing-field-initializers]
> > > > [  263s]    .item_nvgre.protocol = RTE_BE16(RTE_ETHER_TYPE_TEB),
> > > > [  263s]    ^
> > > > [  263s] In file included from
> > > > /home/abuild/rpmbuild/BUILD/dpdk-1622611368.078834f40/x86_64-
> > > > default-linux-gcc/include/rte_eth_ctrl.h:11:0,
> > > > [  263s]                  from
> > > > /home/abuild/rpmbuild/BUILD/dpdk-1622611368.078834f40/x86_64-
> > > > default-linux-gcc/include/rte_ethdev.h:978,
> > > > [  263s]                  from
> > > > /home/abuild/rpmbuild/BUILD/dpdk-1622611368.078834f40/app/test-
> > > > pmd/cmdline_flow.c:18:
> > > > [  263s] /home/abuild/rpmbuild/BUILD/dpdk-
> > > > 1622611368.078834f40/x86_64-default-linux-
> > gcc/include/rte_flow.h:917:13:
> > > > note: 'protocol' declared here
> > > > [  263s]   rte_be16_t protocol; /**< Protocol type (0x6558). */
> > > > [  263s]              ^
> > > >
> > > > This is failing not everywhere but in:
> > > > - SLE_12_SP4/x86_64
> > > > - RHEL_7/x86_64
> > > > - openSUSE_Leap_42.3/x86_64
> > > >
> > > > The type and the init code looks the same here and in the recent main
> > branch.
> > > > Neither do I see recent changes that would cause this.
> > > > And the init just looks right to me, so I'm sure I'm overlooking something.
> > > >
> > > > So I wondered (if you consider this patch important for 19.11.x) if
> > > > you could have a look and consider backporting it so that it works
> > > > well (or as an alternative let me know that it won't be needed for
> > > > 19.11.x)
> > > >
> > > > References/Logs:
> > > > 19.11.8 working: https://paste.ubuntu.com/p/vvm4qCcmJT/
> > > > 19.11.9 fail RH7: https://paste.ubuntu.com/p/2wph5nfbCg/
> > > > 19.11.9 fail SLE12SP4: https://paste.ubuntu.com/p/nGHhkPJbR3/
> > > > 19.11.9 fail openSuse42.3: https://paste.ubuntu.com/p/43mMx3sY2n/
> > > > 19.11.9 witzh the patch dropped works:
> > > > https://paste.ubuntu.com/p/JGZ44BhkC6/
> > > >
> > > >
> > > > --
> > > > Christian Ehrhardt
> > > > Staff Engineer, Ubuntu Server
> > > > Canonical Ltd
> >
> >
> >
> > --
> > Christian Ehrhardt
> > Staff Engineer, Ubuntu Server
> > Canonical Ltd



--
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

  reply	other threads:[~2021-06-08  4:51 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-02 13:25 Christian Ehrhardt
2021-06-03 14:54 ` Jiawei(Jonny) Wang
2021-06-04  5:07   ` Christian Ehrhardt
2021-06-04 13:44     ` Jiawei(Jonny) Wang
2021-06-08  4:50       ` Christian Ehrhardt [this message]
2021-06-08  5:59         ` Jiawei(Jonny) Wang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAATJJ0KpHqfhHrcn6Dc_39s5X5GP6nOh_qt4nG0P6Rz+Zu64cw@mail.gmail.com \
    --to=christian.ehrhardt@canonical.com \
    --cc=jiaweiw@nvidia.com \
    --cc=orika@nvidia.com \
    --cc=stable@dpdk.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).