#fdio-vpp: fdio-vpp
Meeting started by DaveBarach at 16:04:17 UTC
(full logs).
Meeting summary
- 
  - https://docs.releng.linuxfoundation.org/en/latest/meetbot.html
    (vrpolak,
    16:04:34)
 
 
- CSIT (Maciek) (DaveBarach, 16:04:47)
  - CSIT-2009 report published on 15-Oct, see
    announce email from me (mackonstan,
    16:05:10)
 
  - All but AMD and nat44ed test data are
    there (mackonstan,
    16:06:25)
 
  - https://docs.fd.io/csit/rls2009/report/
    (mackonstan,
    16:06:34)
 
  - Ongoing work on: nat44ed cps tunings, gso
    tests, geneve tests (mackonstan,
    16:07:06)
 
  - CSIT-2009 RCA update by Vratko, no big enough
    regressions (< -5%), spent time on some smaller regressions
    (l2patch, see ticket) (mackonstan,
    16:08:02)
 
  - Trending update by Vratko, one regression
    that's not real (introduced by nat44ed tests), side-effect of not
    trusting TRex when it comes to test time measurement ( <1%
    change) (mackonstan,
    16:08:55)
 
  - Vratko, some progressions noticed but not
    investigated (mackonstan,
    16:09:35)
 
 
- Host Stack (Florin) (DaveBarach, 16:10:52)
  - various cleanups, to avoid bit-rot (DaveBarach,
    16:11:16)
 
  - to improve compile time, add {tcp,udp}_packet.h
    vs. protocol implementation headers (DaveBarach,
    16:11:57)
 
  - per-worker cut-through sessions. Gives much
    improved cut-through session setup performance (DaveBarach,
    16:12:45)
 
  - host stack no longer supports legacy binary API
    (shared memory) connections. Security measure. (DaveBarach,
    16:13:28)
 
  - Question: why does dpdk map 128gb of
    memory? (DaveBarach,
    16:14:04)
 
  - Damjan talked to Anatoly (dpdk guy) yesterday
    about that (DaveBarach,
    16:14:23)
 
  - dpdk maps a huge chunk, then finds an
    appropriate boundary (DaveBarach,
    16:14:57)
 
  - host stack can trip over the dpdk etc.
    (DaveBarach,
    16:16:04)
 
  - vcl improved handling of out-of-order events.
    Small cleanups / fixes in addition. (DaveBarach,
    16:16:55)
 
 
- Docs (Ray) (DaveBarach, 16:17:03)
  - Ray points to Ole... ;-)... (DaveBarach,
    16:17:36)
 
 
- Release Management (Andrew, DaveW) (DaveBarach, 16:18:34)
  - 21.01 roughly 1 month to [sic] API freeze
    pre-throttle pull cool-down (DaveBarach,
    16:19:31)
 
  - Actual release in January, most of a month
    between RC1 and RC2 (DaveBarach,
    16:19:55)
 
  - 20.09 one outstanding question: maintenance
    release 20.09.1 (DaveBarach,
    16:20:54)
 
  - Any requests for the dot release?
    Timing? (DaveBarach,
    16:21:12)
 
  - DaveW suggests that waiting until February
    would be quite a while since the formal release (DaveBarach,
    16:22:08)
 
  - 19.08 Job removal in January? (DaveBarach,
    16:22:45)
 
  - Not planning to do another 19.08 point release.
    Revisit in December. (DaveBarach,
    16:23:10)
 
  - Cadence after 21.01: September is tricky due to
    summer vacation. (DaveBarach,
    16:24:36)
 
  - Possibly 21.06 and 21.10. (DaveBarach,
    16:25:17)
 
  - 22.01, etc. (DaveBarach,
    16:25:41)
 
 
- Infra Status (Dave W) (DaveBarach, 16:26:01)
  - 4 [sic] emergency maintenance ops in the last
    month (DaveBarach,
    16:26:17)
 
  - All of the LF instances have increased in
    capacity (DaveBarach,
    16:26:40)
 
  - Some positive results from the exercise
    (DaveBarach,
    16:27:01)
 
  - Finally root-caused the multiple cherry-pick
    issues. (DaveBarach,
    16:27:23)
 
  - Need to reduce the Nomad cluster cleanup
    interval. (DaveBarach,
    16:27:48)
 
  - Docker image creation scripts merged into
    ci-mgmt (DaveBarach,
    16:28:18)
 
  - Centos-7 job will go away in the next release;
    removing from master builds later today (DaveBarach,
    16:30:26)
 
  - Debian 9 is old. Would like to switch to Debian
    10 in the 21.0 time frame (DaveBarach,
    16:31:04)
 
  - DaveB reports > 20 static analysis
    complaints (DaveBarach,
    16:32:47)
 
  - Ray reports Intel internal discussion about
    moving away from Centos-7 (DaveBarach,
    16:33:39)
 
  - Centos-8 patch validation online (DaveBarach,
    16:33:59)
 
 
- Kubecon (Jerome) (DaveBarach, 16:38:04)
  - Presenting VPP / Calico integration at Kubecon
    North America (DaveBarach,
    16:38:28)
 
  - joint preso between Aloys and the Calico
    guys (DaveBarach,
    16:38:42)
 
  - three-way comparison between Calico/VPP,
    Calico/Linux and Calico/eBPF (DaveBarach,
    16:39:13)
 
  - expect recording to be available (DaveBarach,
    16:39:27)
 
  - $50 to join live (DaveBarach,
    16:40:12)
 
 
Meeting ended at 16:43:42 UTC
(full logs).
Action items
  - (none)
 
People present (lines said)
  - DaveBarach (44)
 
  - mackonstan (7)
 
  - collab-meetbot (3)
 
  - vrpolak (1)
 
Generated by MeetBot 0.1.4.