Release Notes

Changes in CSIT-2202

  1. VPP PERFORMANCE TESTS

    • Regressions with DPDK drivers: Change from DPDK v21.08 to DPDK v21.11 introduced regression across all tests using dpdk drivers (with dpdk_plugin loaded). Compared to previous VPP release performance drop varies in the range of -15% to -6%, depending on test. It is related to updated MTU checks within DPDK code and associated VPP code changes. See VPP v2202 release notes and Known Issues.

    • Number of CSIT 9000B frame tests failing: tests with higher encapsulation overhead are failing due to exceeding default Ethernet Maximum Frame Size value that has been reduced by MTU related VPP code changes. See VPP v2202 release notes and Known Issues.

    • Intel Xeon Ice Lake: Performance test data for these platforms is now provided by testbeds newly installed in FD.io CSIT labs. For details about the physical setup see 2-Node Xeon Icelake (2n-icx) and 3-Node Xeon Icelake (3n-icx).

    • Reduction of tests: Removed certain test variations executed iteratively for the report (as well as in daily and weekly trending) due to physical testbeds overload.

    • RC2 coverage test data is used for 2n-icx test bed: There is only one 2n-icx test bed and the amount of tests is large (same as 2n-skx with 4 test beds), so we decided to use test data already available from RC2 testing.

  2. TEST FRAMEWORK

    • CSIT test environment version has been updated to ver. 9, see Environment Versioning.

    • CSIT PAPI support: Due to issues with PAPI performance, and deprecation of VAT, VPP CLI is used in CSIT for many VPP scale tests. See Known Issues.

    • General Code Housekeeping: Ongoing code optimizations and bug fixes.

  3. PRESENTATION AND ANALYTICS LAYER

    • Graphs improvements: Updated Packet Latency graphs, see Packet Latency.

Known Issues

New

#

JiraID

Issue Description

1

CSIT-1799

All NAT44-ED 16M scale tests fail while setting NAT44 address range.

2

CSIT-1800

All Geneve L3 mode scale tests (1024 tunnels) are failing.

3

CSIT-1801

9000B payload frames not forwarded over tunnels due to violating supported Max Frame Size (VxLAN, LISP, SRv6).

4

CSIT-1808

All tests with 9000B payload frames not forwarded over memif interfaces.

5

CSIT-1809

All tests with 9000B payload frames not forwarded over vhostuser interfaces.

6

CSIT-1802

AF-XDP - NDR tests failing from time to time.

7

CSIT-1803

3n-icx testbeds (Icelake): all IMIX aes128cbc-hmac512sha tests are failing due to excessive packet loss.

8

CSIT-1804

3n-tsh testbed (Taishan): NDR tests failing from time to time.

9

CSIT-1805

3n-skx testbeds (Skylake): all hoststack quic vppecho scale tests are failing.

10

CSIT-1810

DPDK - performance regression with DPDK driver when Max Frame Size is set to less than 2023.

Worse performance with DPDK driver when MTU is set to 2022 or less.

VPP-1876

11

CSIT-1811

All 9000B NAT44DET 64k 1m scale tests fail due to bps rate set to high on TRex.

12

CSIT-1812

All IMIX NAT44DET 4m 16m scale tests fail due to not creating required session count.

Fixed

Issues reported in previous releases which were fixed in this release:

#

JiraID

Issue Description

1

CSIT-1789

IPSEC SW async scheduler tests show bad behavior. VPP code is not behaving correctly when crypto workers are the bottleneck.

VPP-1998

Root Cause Analysis for Performance Changes

List of RCAs in CSIT-2202 for VPP performance changes:

#

JiraID

Issue Description

1

CSIT-1814

rls2202 regression: NAT44ED CPS larger scales.

2

CSIT-1815

rls2202 regression: l2bdbasemaclrn around 2021-11-11.

3

CSIT-1816

rls2202 regression: l2-input.

4

CSIT-1820

rls2202 regression: tx2 ip6base at dpdk upgrade.

5

CSIT-1821

rls2202 regression: tx2 ip6base before dpdk upgrade.

6

CSIT-1822

rls2202 regression: avf-ip4base around 2022-01-06.

7

CSIT-1823

rls2202 regression: vxlan-l2xcbase on tsh around 2022-01-07.

8

CSIT-1824

rls2202 regression: SRv6.