Difference between revisions of "VPP/AArch64"

Revision as of 05:00, 19 April 2018

Meeting Details

Weekly on Wednesdays, 6AM PT / 3PM CET / 7:30PM IST / 10PM CST. FD.io Zoom Meeting room

Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/5301185804 or https://zoom.us/my/fastdata

Or iPhone one-tap (US Toll):  +14086380968,,5301185804# or +16465588656,,5301185804#

Or Telephone:
    Dial: +1 408 638 0968 (US Toll) or +1 646 558 8656 (US Toll)
    +1 855 880 1246 (US Toll Free)
    +1 877 369 0926 (US Toll Free)
    Meeting ID: 530 118 5804
    International numbers available: https://zoom.us/zoomconference?m=ppBOQMQTVxGYmbxNsVemC6KNo8eX2ptF

IRC Channel

#fdio-arm on freenode.net

FD.io Lab

Build, unit test, packaging

The following is tracked manually until hardware is integrated into upstream FD.io CI

Status on commit: a38783e0d1ab1d4c661570a1ec90670a1fb0598d (Thu Feb 15 07:31:01 2018 +0000)

kubeproxy tests are broken on purpose: corresponding features are not fully implemented

Timing consideration on platform: Hierofalcon with Cortex-A57 & Fedora 26

• Might want to have a look at this patch which adds make config: https://gerrit.fd.io/r/#/c/9200/

CSIT

https://wiki.fd.io/view/CSIT/AArch64

AArch64 Tuning

Areas:

• Profiling analysis & optimization
• Runtime selection of code using existing methods in VPP

AArch64 Porting

Areas:

• Hardware topology variation, e.g. non PCIe NICs
• VPP-1215 - TC01 : Process untagged send tagged testcase failing due to same packet received as sent

All: JIRA issues with ARM64 label

Assigned and New:

Known Issues

Compilation may fail on systems with less than 1GB memory per core. One workaround is to search for -j in build-root/Makefile and multiply by 1 instead of 2.

GCC 5.3.x ICEs during FP register allocation. Please use GCC 5.4+.

Try disabling ASLR if experiencing random crashes: sysctl -w kernel.randomize_va_space=0

Recent Patches

Meeting Minutes

4/18/2018

• FD.io lab
  • Temporarily borrow 1x ThunderX to be used for ONAP demo at OpenStack Summit (end of May)? Yes.
  • OS exists on ThunderXs; Varun will keysign with EdW; need to resolve OS netdev connectivity over 10/40GbE
  • OS exists on TaiShan2280; no connectivity to the Internet
• VPP
  • RC2
    • 'make' passes, 'make test' fail, 'make test-all' ??? - MACCHIATObin (4GB DRAM)
    • 'make' passes, 'make test' pass, 'make test-all' fails - Centriq
    • 'make' passes, 'make test' pass, 'make test-all' fails - x86
  • Build
    • Testing Verify and Merge jobs for 18.04 master on arm64 today
    • Clang build fails on arm? 'CC=clang CXX=clang make'
• CSIT
  • Adarsh updated CSIT status in xls
  • CSIT-1023: decided to go with OpenSSL instead of ARMv8 crypto library, in DPDK, due to number of algorithms supported
    • e.g. AES-GCM not supported by ARMv8 crypto library
  • Nitin updated CSIT-990 (buildroot) with more information
• Action Items
  • Sirshak to add link to xls to wiki page
  • Brian to raise LF RT ticket about MACCHIATObins
  • Nitin to check 'make test' on MACCHIATObin (16GB DRAM)
  • Honnappa, Khem to check Clang build on arm64

4/11/2018

• Proposal to keep meeting at current time with additional overflow meeting at 8AM PST
• FD.io lab
  • MACCHIATObins just arrived at VEXXHOST
  • Nitin working on getting IPMI login credentials to provision OS on ThunderX
  • Need to connect Skylake TG machines to Arm machines
    • ETA: 1wk
  • Khem working with Aton (LF) to provision OS on TaiShan2280
    • ETA: 1wk, Ubuntu 17.10
• VPP
  • Brian to do more benchmarking on MACCHIATObin
  • Khem working on benchmarking clib_memcpy64_x4()
• CSIT
  • Lucian submitted patches for CSIT-1019, CSIT-1021
  • Lucian looking for contact for ARMv8 crypto driver in DPDK for CSIT-1023
    • See CSIT-1023 for details; looks like DPDK issue?
  • Nitin to add more details to CSIT-990
• Action Items
  • Sirshak to move JIRA tickets to xls
  • Lucian to work with Nitin/Jerin on CSIT-1023

4/4/2018

• Propose to move the meeting +2 hours?
• RC1 cut today
• FD.io lab
  • Allocate 3 ThunderX for EdK to integrate into CI
    • JohnB from Cavium agreed to supply 3 more ThunderX for CSIT (will pre-install FW & OS)
  • Brian working on provisioning SSDs for MACCHIATObins
  • Khem can ping IPMI interfaces on TaiShan2280s; also needs an OS to be installed
• VPP
  • Discussed ONS slides
  • Khem has patch for clib_memcpy64_x4() and needs help benchmarking
• CSIT
  • Lucian found and created JIRA tickets for 3 issues while running CSIT
  • Nitin created JIRA ticket for buildroot issue
  • Khem seeing issues with VM
• Action Items
  • Nitin/Varun to help provision Ubuntu 16.04 and firmware update on ThunderX machines

3/28/2018

• Sachin Saxena from NXP joined the call, welcome
• FD.io lab
  • Khemendra is having issues with Rudy's emails. Hence, not been able to access Taishan servers
  • Nitin will try to access the servers this week
  • MACCHIATObin setup under progress
  • OD1000 is added to Jenkins slave. The build is failing currently. The build can be triggered manually.
• VPP
  • Discuss Single core, L3Fwd sample perf numbers and analysis next week
  • Sachin is working on compiling 18.01. Native compilation works fine, cross compilation is failing
  • Nitin still working on patch for cache line size
  • VPP-1126 is being used in DPDK input node. Khemendra will take a look at it this week.
  • VPP-1129 Brian/Sirshak will take a look. Looks like it can be closed.
  • VPP-1114 Patch under internal review
• CSIT
  • Khemendra having issues with interface bring up failing intermittently. Nitin suggested to add delay.
  • Nicolas/Lucian debugging TC-07
  • Khemendra having issues with TG VM crashing randomly with Ubuntu 16.04, QEMU 2.10. Solved by moving to Ubuntu 17.10, QEMU 2.10
  • Nitin using Ubuntu 16.04 with 4.13 kernel
• Action Items
  • Discuss Single core, L3Fwd sample perf numbers and analysis next week - Brian
  • VPP-1126 Take a look this week as it affects DPDK input node - Khemendra
  • Need more attention on solution for buildroot issue, need more information on failure CSIT-990 - Nitin
  • Create an excel sheet with the test case status - Nicolas/Lucian

3/21/2018

• Key signing party! Thank you Ed!
• FD.io lab
  • VEXXHOST currently working on getting another PDU because there are not enough power ports
  • Received SSDs for MACCHIATObins
• VPP
  • Discuss high level plan for VPP on Arm
  • Nitin still working on patch for cache line size
• CSIT
  • Need more attention on solution for buildroot issue CSIT-990
  • Nitin moving towards L2 & L3 perf test cases
  • VM crash due to buffer overflow when multiple VMs share NVRAM; resolved in Fedora27

3/14/2018

• Key signing party! Thank you Ed!
• FD.io lab
  • ToR switch issue resolved; confirm mgmt IP address assignment to racked Huawei/Cavium machines
  • Started provisioning MACCHIATObins; Andy ordered SSDs to go with them
• VPP
  • No updates
• CSIT
  • Adarsh started running CSIT on virtual topology; moved past a paramiko issue, seeing other test failures
  • Ongoing discussions on getting Adrian access to machines

3/7/2018

• FD.io lab
  • Trishan (LF) to help follow up on progress in FD.io lab
• VPP
  • More discussion on patch for cache line size; use MIDR register exported by proc fs
  • Decision has been made to use wrappers for atomics
  • Damjan reworked PCI handling code and added native driver for Intel AVF (XL710 i.e. Fortville)
    • Measuring 132 clocks per packet on Skylake (ip4 routing) with VLIB_FRAME_SIZE 256 (default); +1Mpps over DPDK avf/i40e PMD
  • Damjan reworked memcpy() in MEMIF; achieve 2x25GbE line rate with these changes
  • Sirshak working on getting VPP running on Qualcomm Centriq with Mellanox NIC
    • Seeing issues with external DPDK; static works but not shared; is VPP build system missing -libverbs -lmlx5 in LDFLAGS?
    • Nitin noticed DPDK 17.11 Mellanox PMD does not compile
    • Mellanox recently submitted a patch to VPP to support dynamic loading of Mellanox libraries
• CSIT
  • Adrian does not have machines to work with in Bucharest; machine in Paris that Gabriel was using no longer available
    • AndyW to help resolve
  • Adarsh moved past VM issues; able to launch VPP in VM with virtio interface; starting to run CSIT scripts

2/28/2018

• FD.io lab
  • Ed Kern to try containerized CI on one OD1000 in parallel with Vanessa
  • Received MACCHIATObins in Austin
• VPP
  • Adarsh trying to run VPP in VM but getting PCI mapping issue; trying to connect to Linux bridge on host
  • Patches for build breakage were committed; arm64 build stable now
  • Brian able to reproduce low PPS numbers seen on MACCHIATObin
• CSIT
  • Adarsh can reproduce a crash in qemu 2.10 Ubuntu 16.04; going to try Ubuntu 17.10
  • Need to partition func test cases across people

2/21/2018

• FD.io lab
• CSIT
  • Gabriel updated CSIT/AArch64 wiki with PASS/FAIL/OTHER list
    • OTHER - failure due to expect-like parsing of output(?)
    • FAIL - ssh timeout during PCIe rescan(?)
  • Moved past first UEFI crash; still seeing crashing on startup (Gabriel)
    • Setup new Ubuntu environment
    • Continue debugging UEFI issue on Fedora with JeremyL
  • Ubuntu is used pretty much everywhere except for additional CentOS CSIT perf
  • Nitin working on upstreaming changes to CSIT
  • Adarsh working on getting VM interfaces working
• VPP
  • More discussion on how to handle cache line size
  • Sync'd on patches for build breakage

2/14/2018

• FD.io lab
  • Working on getting access to LF lab in order to setup OD1000 environment
  • Check with tykeal & zxiiro on trust policy for getting others access (Brian)
  • VEXXHOST
    • Mohammed says they do not have extra rack shelf - we need to send one for 3x MACCHIATObin
    • LF RT tickets: #52434 (ThunderX), #52435 (TaiShan2280), #52436 (MACCHIATObin)
• VPP
  • Build, unit test, deb/rpm
    • 64B/128B cache line size - working on passing this configuration to rest of build system i.e. DPDK (Nitin)
    • RPi3 32-bit
      • Some parts of patch are 32-bit related, some RPi3 related
      • If there is justification, look into maintaining a 32-bit build on ARM
  • Porting & Tuning
    • If patches need to be tested on multiple Arm chips, please use DO_NOT_MERGE and Code Review -2
    • Two NEON related patches merged, working in progress on others, Nitin testing CLASSIFY_USE_SSE
• CSIT
  • Please open JIRA ticket with details on VM crashing on startup. DONE: CSIT-922
  • Khem working on running VPP func tests on internal setup

2/7/2018

• LF lab
  • OD1000 - last machine was racked; Vanessa needs credentials
  • Taishan2280 - machines arrived at Vexxhost; confirm with Rudy/Mohammed
  • ThunderX - machines arrived at Vexxhost; send board details to Mohammed
  • MACCHIATObin - boards arrived in Arm SJC waiting for enclosures (Andy)
• Build, unit test, packaging
  • 64B/128B cache line size - working on it (Nitin)
  • Interest in ILP32 from Cavium; customer coming from MIPS32
    • BKK16-305B ILP32 Performance on AArch64
• VPP
  • NEON usage in vhost - sent first patch for review (Nitin)
    • Need to verify how it performs on other Arm-based machines (Brian)
    • VPP maintainers prefer to use SIMD wrappers, but it might not always be possible
      • Cavium/Arm had to rewrite algorithm for AArch64 instead of use SIMD wrappers in DPDK
  • CLIB_HAVE_VEC128 - working on it (Gabriel)
  • Discussed compiler builtins for atomics in VPP call; need to spin another patch with wrappers based on architecture (Kevin)
  • Seeing prefetch hostspots on TX2+MlnxCX4en (similar to Armada8040) (Nitin)
• CSIT
  • libvirt crashing on VM startup (Hierofalcon) (Gabriel)
    • Need someone who can reproduce this issue (Arm TBD)
  • Huawei also seeing VM issues (Khem)
  • buildroot doesn't work on Arm (Nitin)
    • Root issue: no support in GRUB for AArch64 in buildroot (?)
      • Need someone who can reproduce this issue (Arm TBD)
    • Peter Mikus replied to Nitin on csit-dev mail list
    • Using a temporary workaround: use a different VM image (Ubuntu Cloud) instead of one produced by buildroot
      • Working on patching DPDK in VM image (Ubuntu Cloud) just like done in buildroot
• Misc
  • OpenFlow (Nitin, Damjan)
    • Is there an OpenFlow agent for VPP, and can VPP implement OpenFlow rules/tables?
    • VPP is not flow-based like OVS is; they are different
    • Can ODL/Honeycomb be used?

1/31/2018

• LF lab
  • OD1000 - 1 replacement being installed this week
  • Huawei & Cavium boards should arrive at colo this week; confirm with Rudy
• Build, unit test, packaging
  • Kubeproxy/NAT failures
    • Not arch related
    • Part of extended unit tests, so does not block CI
  • `make test` passes on D03 & D05 (Ubuntu)
• MACCHIATObin
  • Seeing hotspots in VPP graph nodes
    • L3 forwarding - ip4 rewrite node
    • L2 cross-connect
    • Try reducing quad loop to a dual loop
    • dpdk-input node highly opt for x86 (could contribute to low perf) but hotspots still in rte_mbuf_t conversion(?)
  • Some examples of runtime code selection based on uarch exist in the codebase
• CSIT
  • Adrian Oanca join from Enea
  • Gabriel seeing VM crashing during boot; related to # interfaces assigned (6)
  • Nitin ran into issue with buildroot on arm64; see thread on csit-dev

1/24/2018

• VPP
  • DPDK issue with non-pci network cards
  • build & test status updated
  • VPP-1127 (VEC_128 enable) under discussion. Should we enable this by default ?
  • add Nitin to review Neon commits
  • VPP-1114 currently internal review
  • VPP-1064 under rework after review by Damjan
• CSIT
  • first 3-nodes functional tests status list
  • TODO Gabriel: share CSIT VM setup env
  • nested VM: build-root package support for ARM. Create Jira ticket for Brian.

1/17/2018

• Tina to send calendar invite for meeting
• FD.io lab
  • Cavium shipping
• VPP
  • Kubeproxy tests failing
  • Khem trying to find out the PCIe address for a given netdev interface
• CSIT
  • Gabriel setting up 3 node topo with VMs
  • Gabriel working on PASS/FAIL status
• CSIT 17.10 report

1/10/2018

• Meeting moved 2 hours earlier - 6AM PT / 3PM CET / 7:30PM IST / 10PM CST
• FD.io lab
  • Cavium ThunderX shipping soon
• VPP
  • Kumar to look at VPP-1126
  • Gabriel proposed https://gerrit.fd.io/r/#/c/10049/ as follow-up to Damjan's patch
• CSIT
  • Gabriel's patch for aarch64 support in CSIT merged
  • VirtualBox not supported on Arm / Vagrant unknown
    • This is OK for upstream since automation expects VMs to already exist
• Performance
  • Need plan for 1T; use TaiShans that were sent to lab
• AIs
  • Brian: Follow up with Vanessa and EdW regarding 'resource issue'
  • Gabriel: Update CSIT wiki page; which tests are passing/failing?
  • Brian: Check with Vanessa how to split machines between CI jobs and CSIT jobs

1/3/2018

• FD.io lab
  • One OD1000 sent for RMA
  • Huawei PO sent out
  • Cavium PO sent out (?)
• VPP
  • Gabriel working on patch for "show cpu" to display MIDR as human readable
  • Nitin sent preliminary patch for vhost-user NEON impl
    • Seeing perf differences on different cores; tradeoff is single-threaded perf vs. NEON
  • Kumar built and unit test successfully on D03
  • Nitin to resume patch for supporting different cache line sizes for the same arch
• CSIT
  • Gabriel cleaned up WIP patch; ready for review
  • Kumar starting CSIT func tests with Ubuntu VMs
    • Scripts for running on dedicated hardware need to be modified, e.g. PCIe resources
  • Kumar to send doc on testing
• Performance
  • Kumar to start thread on performance testing
• AIs
  • Brian: Check with Tina on shipping and open LF RT ticket once they have arrived
  • Brian: Need a way to choose either SW or NEON impl based on chip
  • Gabriel: Create list of broken CSIT tests for 2-node topology

12/20/2017

No meeting next week - Dec 27

• FD.io lab
  • OD1000s - build only
    • 1 of 3 needs to be RMAd
    • Can these be up in time to show 'make test' passes on ARM for 18.01 release report?
  • TaiShan
    • PO in progress
  • ThunderX - build only
    • PO went out
• VPP
  • Patches / JIRAs
    • Patch for extended test failure, but still more (new) extended test failures - Gabriel
    • Nitin to post vhost-user.c changes for NEON
      • Nitin will finish Gabriel's original NEON patch to add CLIB_HAVE_VEC_128
  • Can we share code on Github e.g. NEON perf tests?
• CSIT
  • Leading question: How many CSIT test cases are passing/failing?
  • Environment issues preventing running through all CSIT test cases; Gabriel needs dedicated machines or more RAM
  • Cavium & Huawei will join Gabriel in CSIT replication on ARM hardware next week
    • Cavium previously ran vhost test cases manually, now moving to CSIT

12/13/2017

• VPP
  • Quick overview of work items
  • Waiting to hear back from LF about OD1000 connectivity
    • Changes needed to ci-mgmt
• CSIT
  • Starting to reproduce CSIT on x86 and ARM (with Gabriel's WIP patch)
    • Some issues with environment variables (perf tests on 2-node)
  • Need Nexus to support aarch64 packages
    • Need a contact for Nexus
• Share known issues on wiki!
• Request CSIT 'deep dive'

12/06/2017

• Can we access the OD1000 in csit lab ?
  • currently mainly working with VMs
• added dedicated wiki page for CSIT : https://wiki.fd.io/view/CSIT/AArch64
• WIP : https://gerrit.fd.io/r/#/c/9474/

11/29/2017

• VPP
  • vhost-user.c - SSE4.2 only. Implement range search using NEON. (nitin)
  • OD1000 status ?
    • build only
    • can we access them ?
    • what wan we do to help in general ?
  • x86 intrinsic review
  • build VPP on ARM VM on x86
• CSIT
  • what platforms wil lbe made available

11/22/2017

• VPP CI
  • 3 ThunderX for Chrismas
• CSIT
  • func on VM vs perfs on HW
  • func on x86 VMs OK with 2 nodes
  • DPDK integration WIP : https://gerrit.fd.io/r/#/c/9474/
  • issues
    • how to access the lab ?
• Next steps
  • VPP
  • CSIT
    • structure work & send email (Gabriel)
    • is xxhash vs crc32 finished ? (Gabriel)
    • ask Maciek & setup a presentation meeting with someone from CSIT (Tina)
    • find a time to reschedule this meeting before the CSIT weekly call (Brian)

11/15/2017

• VPP upstream status
  • build && build-release OK
  • "make test" && "make test-debug" OK
  • packaging:
    • Ubuntu 16.04 OK
    • Ubuntu 17.10 ? (TBC)
    • fedora-26 OK
• vpp continuous test
  • all task required for jenkin's "verify" job are ready
  • TODO: request gerrit hook to Dave Barachs / vpp-dev (NB & GG)
  • set up ci in fdio lab
• CSIT
  • setting up env
  • ThunderX platforms should arrive this week
  • csit work sharing

11/8/2017

• Unit tests
  • Tests pass except for random initialization failures
  • Need to hear back from upstream about Extended unit tests
• Should we run plugins such as NSH SFC?
• Hardware to lab
  • Huawei h/w stalled
  • 3x ThunderX shipping to FD.io lab
• CSIT replication
  • Cavium replicating on ThunderX2; getting started
• Let's track our work in Jira; Brian to migrate tasks to Jira

10/25/2017

• Gabriel working on vpp init failure in linux_pci_init()
• Kumar to check with GeorgeZ on Huawei boards shipped to CSIT; need to verify tests also on this environment (package versions from distro)
• Brian to check whether anything else needs to be done besides 'make test' for upstream enablement