diff options
author | 2020-09-17 13:13:38 -0700 | |
---|---|---|
committer | 2021-02-22 11:19:15 -0800 | |
commit | a6aa7c8f998f4afddd73410aa043dad38162ce9e (patch) | |
tree | 056ea991eb27ee8c6cf829d5b39536c2763961f0 /tools/perf/scripts/python/export-to-postgresql.py | |
parent | ice: Set trusted VF as default VSI when setting allmulti on (diff) | |
download | wireguard-linux-a6aa7c8f998f4afddd73410aa043dad38162ce9e.tar.xz wireguard-linux-a6aa7c8f998f4afddd73410aa043dad38162ce9e.zip |
ice: Account for port VLAN in VF max packet size calculation
Currently if an AVF driver doesn't account for the possibility of a
port VLAN when determining its max packet size then packets at MTU will
be dropped. It is not the VF driver's responsibility to account for a
port VLAN so fix this. To fix this, do the following:
1. Add a function that determines the max packet size a VF is allowed by
using the port's max packet size and whether the VF is in a port
VLAN. If a port VLAN is configured then a VF's max packet size will
always be the port's max packet size minus VLAN_HLEN. Otherwise it
will be the port's max packet size.
2. Use this function to verify the max packet size from the VF.
3. If there is a port VLAN configured then add 4 bytes (VLAN_HLEN) to
the VF's max packet size configuration.
Also, the VIRTCHNL_OP_GET_VF_RESOURCES message provides the capability
to communicate a VF's max packet size. Use the new function for this
purpose.
Fixes: 1071a8358a28 ("ice: Implement virtchnl commands for AVF support")
Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Tested-by: Tony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Diffstat (limited to 'tools/perf/scripts/python/export-to-postgresql.py')
0 files changed, 0 insertions, 0 deletions