15:09:08 <DaveBarach> #startmeeting fdio-vpp
15:09:08 <collab-meetbot> Meeting started Tue Oct 13 15:09:08 2020 UTC and is due to finish in 60 minutes.  The chair is DaveBarach. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:09:08 <collab-meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:09:08 <collab-meetbot> The meeting name has been set to 'fdio_vpp'
15:10:50 <DaveBarach> #topic CSIT (Maciek reporting)
15:11:23 <mackonstan> #info New tests:
15:11:56 <DaveBarach> #info still onboarding new tests, including nat44 tests
15:12:03 <mackonstan> #info Lots of nat44: i) nat44det throughput (stateless, bi-dir), ii) nat44ed throughput (stateless, uni-dir), iii) nat44ed CPS and PPS throughput (stateful, bi-dir).
15:12:11 <mackonstan> #info All nat44 tests at varying scale from 64k to 16mil UDP (stateless, stateful) and TCP (stateful) sessions.
15:12:35 <mackonstan> #info TRex upgraded to v2.82, calibration work for STL and ASTF.
15:13:15 <mackonstan> #info AMD 2n-zen2 EPYC 7532 testbed onboarded, wip dry run tests, CSIT-2009 tests to start shortly.
15:13:31 <mackonstan> #info CSIT-2009 bulk of VPP v20.09 tests done, initial report with results to be published tomorrow, wip draft report at the link below
15:13:38 <mackonstan> #info https://docs.fd.io/csit/master/report/
15:13:48 <mackonstan> #link https://docs.fd.io/csit/master/report/
15:14:02 <mackonstan> #info Report to be updated next week and week after with more runs/tests incl. AMD and nat44ed stateful.
15:15:31 <DaveBarach> #info Maciek to communicate location of AMD epyc tuning notes
15:15:41 <DaveBarach> #topic Host Stack (Florin)
15:15:56 <DaveBarach> #info session layer added pure socket-based APIs
15:16:09 <DaveBarach> #info will deprecate binary API attachment interface
15:16:26 <DaveBarach> #info cleaner / more secure way of integrating apps, particularly in containers
15:16:56 <DaveBarach> #info TCP small timer wheel refactor, to support 100us timer-wheel resolution
15:17:35 <DaveBarach> #info to support more advanced RTT-level timing
15:18:14 <DaveBarach> #info Envoy integration upstream refactoring, to facilitate host-stack / Envoy comms
15:18:42 <DaveBarach> #info overview paper will be presented at EnvoyCon in two days
15:19:23 <DaveBarach> #info florin to share recording, slides etc.
15:20:13 <DaveBarach> #topic Documentation
15:20:29 <DaveBarach> #info need to find an owner: John DeNisco has retired
15:20:40 <DaveBarach> #topic Release Management (Andrew / Dave W)
15:21:18 <DaveBarach> #info 20.09 update: increased "calm-down" interval definitely helped
15:21:48 <DaveBarach> #info Proposed plan for 21.01 dates on wiki
15:22:38 <DaveBarach> #info F0 on 12/09, formal release on 1/27
15:24:51 <DaveBarach> #info no comments on the posted schedule, so we will go with them...
15:26:26 <DaveBarach> #info 19.08 deprecation discussion, should turn off the jobs around the first of the year
15:26:55 <DaveBarach> #info Shoot for about one quarter's worth of LTS release overlap
15:27:12 <DaveBarach> #topic Infra (Dave W)
15:27:24 <DaveBarach> #info a couple of additional Jenkins outages
15:27:42 <DaveBarach> #info pushing the LF to root-cause the outages
15:28:19 <florinc> #info link to EnvoyCon2020 presentation https://wiki.fd.io/images/0/08/Using_vpp_as_envoys_network_stack.pdf
15:28:24 <DaveBarach> #info Might want to stop tracking "rechecks"
15:31:06 <DaveBarach> #info adding verify jobs for Debian 9, Debian 10
15:31:44 <DaveBarach> #info Nomad infra was manually managed, transitioned to management by Ansible
15:32:06 <DaveBarach> #info jobs running on Ubuntu 18.04
15:32:36 <DaveBarach> #info generation of Docker containers automated
15:40:31 <DaveBarach> #info keep Centos 7 jobs for LTS release(s) only, deprecate for master
15:48:39 <DaveBarach> #topic Static Analysis
15:48:47 <DaveBarach> #info 17 outstanding, please clean up
15:49:09 <DaveBarach> #topic Main heap / bihash-related changes
15:49:24 <DaveBarach> #info bihash using main heap, allows folks to use hugepages
15:49:59 <DaveBarach> #info main heap usage increased, but private mmap areas gone, and gone from core dumps
15:50:17 <DaveBarach> #info ip4 mtrie, same story
15:50:50 <DaveBarach> #info startup config: can specify main heap size, main heap page size
15:51:56 <DaveBarach> #info hugepage backing store / lack of TLB miss walks significantly improves forwarding perf w/ big tables
15:52:23 <DaveBarach> 1G main heap size still a reasonable default
15:53:43 <DaveBarach> #topic long compilation issues
15:54:31 <DaveBarach> #info lots of inlining. Most of the time for a good reason, but we need to avoid making the compiler go crazy
16:00:11 <mackonstan> #topic AMD EPYC 7532 zen2 server settings in FD.io CSIT labs
16:00:11 <mackonstan> #info amd 2n-zn2 BIOS settings
16:00:11 <mackonstan> #link https://gerrit.fd.io/r/gitweb?p=csit.git;f=docs/lab/testbeds_sm_zn2_hw_bios_cfg.md;hb=refs/changes/33/29033/2#l483
16:00:11 <mackonstan> #info zen2 servers firmware inventory
16:00:11 <mackonstan> #link https://gerrit.fd.io/r/gitweb?p=csit.git;f=docs/lab/testbeds_sm_zn2_hw_bios_cfg.md;hb=refs/changes/33/29033/2#l627
16:00:11 <mackonstan> #info amd zen2 GRUB cmd line:
16:00:11 <mackonstan> #info GRUB_CMDLINE_LINUX="isolcpus=0-15,16-31,32-47,48-63 rcu_nocbs=0-15,16-31,32-47,48-63 transparent_hugepages=never iommu=pt amd_iommu=on nohz=off numa_balancing=disable nmi_watchdog=0 audit=0 nosoftlockup processor.max_cstate=0"
16:00:44 <DaveBarach> #endmeeting