From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id 92DC7A04BC for ; Thu, 8 Oct 2020 16:29:52 +0200 (CEST) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 8083E1BB5C; Thu, 8 Oct 2020 16:29:51 +0200 (CEST) Received: from mail-ed1-f46.google.com (mail-ed1-f46.google.com [209.85.208.46]) by dpdk.org (Postfix) with ESMTP id 437FF2BE1 for ; Thu, 8 Oct 2020 16:29:50 +0200 (CEST) Received: by mail-ed1-f46.google.com with SMTP id p13so6054022edi.7 for ; Thu, 08 Oct 2020 07:29:50 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=iol.unh.edu; s=unh-iol; h=mime-version:from:date:message-id:subject:to :content-transfer-encoding; bh=Q9QzyqMwXNw0VMVbiWF91ZSLNLJYK1Up+AAtDgyyJTI=; b=BRSxwOKkF55Gh2SrYsAsHL63XK2aD0+03z6BpiZOhtMRSPLJr43e4GMF+NL6daFYRQ 7GHsE7OEl8C4uTQd3YSoPbnwN9i1nEwnnlUQ4BlBICsec/aiKOnCLtjQEs9Zm9Gd31dL MESo0XfmPrn2Zm0XCaf51CmjoL80RR8RXPxBs= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:from:date:message-id:subject:to :content-transfer-encoding; bh=Q9QzyqMwXNw0VMVbiWF91ZSLNLJYK1Up+AAtDgyyJTI=; b=OshqRTYSkBdqaORhcHSIIccw/d+oubCT4PK1p/HLZOCmHRpuNvtOHiB+Xa1BKKQdBQ JH/Bhpyw52fCOj5cr/fTcOwPvjRRd664cpKMxrM8s6jJSa0Ci+RW8A8a7nD0eBdX4hNO PfjMWkyQH0cyB7xEh6t2TtK//VWOaOQlIQ35JNVxVP8qAskshYqaJVJEWnTNC3Zn0NvI Rd3CInQIo74Cp2DmDjypqfDdH/6FRjX/m7sbU8W6a2RFUo43XyWRMipK1tkkZO2N+dkP 9m4Ya6YFSwsua/26cvES2CTtmhAyoLnXeSe0oul6unsuz1752O9lt5o3P1DQ9FIIjJrX VzEA== X-Gm-Message-State: AOAM532lUz3NGBeGXRx9jKiMndeyUKnqL2ipjqWIg/I4ln+dpE9VmW7s DjpdrcjN6zF7ImFl0QgFV/DL7naXx8zQ1R6FYGW6CUpCG3ZsLQ== X-Google-Smtp-Source: ABdhPJzJk7IbtrXWIKt8EmoVgxQlR7L4bO/LQoOZu+x/J0OC9/OXGKudmg5jzAi67TWSPsftDrojETxKrhAp3N1+csM= X-Received: by 2002:aa7:c2c4:: with SMTP id m4mr9661411edp.172.1602167388266; Thu, 08 Oct 2020 07:29:48 -0700 (PDT) MIME-Version: 1.0 From: Lincoln Lavoie Date: Thu, 8 Oct 2020 10:28:37 -0400 Message-ID: To: ci@dpdk.org Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Subject: [dpdk-ci] October 8 Community CI Minutes X-BeenThere: ci@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK CI discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: ci-bounces@dpdk.org Sender: "ci" October 8, 2020 ################################################### Attendees 1. Lincoln Lavoie 2. Trishan de Lanerolle 3. Honnappa Nagarahalli 4. Thomas Monjalon 5. Ali Alnubani 6. Brandon Lo 7. Juraj Linkes 8. Kananyev 9. Tomasz Zawadzki ################################################### Agenda 1. CI Status 2. Test Development 3. Any other business ################################################### Minutes =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D CI Status -------------------------------- General * BUG-210 (Patches with dependency fails to apply): Syntax is now official in the documentation, should be possible to update the script to look for / parse this information. * UNH-IOL will look into implementing this, either in git-pw or as a wrapper to that. * BUG-511 (Add check if performance tests are needed): This can also be implemented to avoid running unnecessary testing. * UNH-IOL will look into implementing this. -------------------------------- UNH-IOL Community Lab * Fedora Rawhide compile / unit testing has been added. Failures being flagged as warnings right now, should disappear once all branches catch up to main. * Previous fail results have been changed accordingly. * We are still planning to look into adding GCC 11 compile to this testing as well. * Need to hold running this regularly if it generates failures, and file bugzilla tickets to fix those issues. * Adding Fedora Rawhide has also prompted a look into how the container images are maintained, to ensure they are kept current (i.e. how often is a new base image pulled for =E2=80=9CDocker Hub=E2=80=9D). * Issues caused on the bare-metal systems by change of repos for kernel modules, working on correcting this today. * Recommended solutions by DPDK/DTS team? * DTS will still rmmod and insmod the nonexistent kernel module file. * SPDK compile / unit testing should be back into production today, fixed after change to the meson build process. * Glitch in OpenSUSE testing fixed, caused by issue with container creation. Fixed on Tuesday. * Arm hardware is being installed and will be added to the CI pipelines as soon as possible, including 25Gbe to 100Gbe NICs from Broadcom, Intel, and Mellanox. * Functionality testing * Slowed down by the kernel module issue listed above. * Will continue to expand the coverage to include the DTS. -------------------------------- Intel Lab * Didn=E2=80=99t attend the call (holiday for the team). -------------------------------- Travis CI & OBS * Update from Aaron (via email): OBS still being worked on. * Need to check if the system is also checking the kernel modules repo as well as the mainline repo. * Patch to enable running Travis CI on AWS Graviton2 has not been merged, Honnappa wanted to check on the status of that patch. Looks like it was merged this week. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Test Development -------------------------------- Completed Features 1. MTU Update 2. Basic stats 3. Speed Capabilities 4. Checksum offload checks 5. Multicast Mac Filtering 6. RSS Key Update 7. L3 and L4 Checksum Offload 8. Multicast mac address filter 9. Linux Kernel Module VFIO Testing 10. Linux Kernel Module UIO Testing 11. FW Version 12. Rx Interrupt 13. Module EEPROM Dump -------------------------------- Work in Progress 1. rte_flow 1. Testing framework is completed, and is being tested on some bare-metal systems in UNH-IOL. 2. We have seen some systems that will =E2=80=9Cvalidate=E2=80=9D a rule= , but fail when =E2=80=9Ccreate=E2=80=9D is called. This seems like it should be a va= lid failure. We=E2=80=99re looking into this today and will reach out to some = of the developers / hardware folks. 1. Sounds like this could actually be bugs, if so, the team will create bugzilla tickets. 3. Question asked to the DTS list about runtime code generation, which is used to automatically generate the pattern match test cases. This allows one test case per pattern (i.e. match IPv4 source address), which in turn allows each consumption of results for what passed / failed within the DTS framework. But, this would be a newer approach within DTS. 4. There are some gaps in testpmd for what patterns can be created. The UNH-IOL team will create a bug for this, as well as documenting in the DTS RST files. 5. Still aiming to submit the DTS patch by the end of this week (October= 9). 6. Test_flow_perf - new DPDK sample app to test the performance and driver efficiency. Could be interesting to add to DTS, as a future patch / target. Create a feature bug ticket for this time. 2. Speed Capabilities 1. Pending updates to testpmd (BUG-496), just keeping here to track status, etc. 3. Inner RSS 1. Currently stalled by BUG-530. There doesn=E2=80=99t appear to be a wa= y to access information about this through testpmd, the BUG-530 proposes the feature to testpmd. The above will represent the conclusion of the contracted development work for the expansion of DTS coverage. -------------------------------- Future Work * DTS Feedback - Some initial (rough draft) user and developer feedback has been captured here: https://docs.google.com/document/d/1c5S0_mZzFvzZfYkqyORLT2-qNvUb-fBdjA6DGus= y4yM/edit * Tech board is looking into DTS as a priority for 2021, and this feedback is extremely valuable to that process. Please contribute! =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Any other business * https://bugs.dpdk.org/buglist.cgi?bug_status=3DUNCONFIRMED&bug_status=3DC= ONFIRMED&bug_status=3DIN_PROGRESS&columnlist=3Dproduct%2Ccomponent%2Cpriori= ty%2Cbug_status%2Cassigned_to%2Cshort_desc%2Cchangeddate&component=3Djob%20= scripts&component=3DUNH%20infra&component=3DIntel%20Lab&component=3DTravis%= 20CI&list_id=3D2663&order=3Dpriority%2Cchangeddate%20DESC&product=3Dlab&que= ry_format=3Dadvanced&resolution=3D--- * Look into running performance unit testing on bare metal hardware in UNH-IOL Community Lab. Honnappa will create a bugzilla ticket for this item. * Next Meeting: October 22, 2020 --=20 Lincoln Lavoie Senior Engineer, Broadband Technologies 21 Madbury Rd., Ste. 100, Durham, NH 03824 lylavoie@iol.unh.edu https://www.iol.unh.edu +1-603-674-2755 (m)