> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-boun...@osuosl.org> On Behalf Of
> Mateusz Polchlopek
> Sent: Wednesday, April 3, 2024 1:11 PM
> To: intel-wired-...@lists.osuosl.org
> Cc: and...@lunn.ch; j...@resnulli.us; Wilczynski, Michal
> <michal.wilczyn...@intel.com>; Polchlopek, Mateusz
> <mateusz.polchlo...@intel.com>; net...@vger.kernel.org; Czapnik, Lukasz
> <lukasz.czap...@intel.com>; Raj, Victor <victor....@intel.com>; Nguyen,
> Anthony L <anthony.l.ngu...@intel.com>; ho...@kernel.org; Kitszel, Przemyslaw
> <przemyslaw.kits...@intel.com>; k...@kernel.org
> Subject: [Intel-wired-lan] [PATCH net-next v9 2/6] ice: Support 5 layer
> topology
>
> From: Raj Victor <victor....@intel.com>
>
> There is a performance issue when the number of VSIs are not multiple of 8.
> This is caused due to the max children limitation per node(8) in
> 9 layer topology. The BW credits are shared evenly among the children by
> default. Assume one node has 8 children and the other has 1.
> The parent of these nodes share the BW credit equally among them.
> Apparently this causes a problem for the first node which has 8 children.
> The 9th VM get more BW credits than the first 8 VMs.
>
> Example:
>
> 1) With 8 VM's:
> for x in 0 1 2 3 4 5 6 7;
> do taskset -c ${x} netperf -P0 -H 172.68.169.125 & sleep .1 ; done
>
> tx_queue_0_packets: 23283027
> tx_queue_1_packets: 23292289
> tx_queue_2_packets: 23276136
> tx_queue_3_packets: 23279828
> tx_queue_4_packets: 23279828
> tx_queue_5_packets: 23279333
> tx_queue_6_packets: 23277745
> tx_queue_7_packets: 23279950
> tx_queue_8_packets: 0
>
> 2) With 9 VM's:
> for x in 0 1 2 3 4 5 6 7 8;
> do taskset -c ${x} netperf -P0 -H 172.68.169.125 & sleep .1 ; done
>
> tx_queue_0_packets: 24163396
> tx_queue_1_packets: 24164623
> tx_queue_2_packets: 24163188
> tx_queue_3_packets: 24163701
> tx_queue_4_packets: 24163683
> tx_queue_5_packets: 24164668
> tx_queue_6_packets: 23327200
> tx_queue_7_packets: 24163853
> tx_queue_8_packets: 91101417
>
> So on average queue 8 statistics show that 3.7 times more packets were send
> there than to the other queues.
>
> The FW starting with version 3.20, has increased the max number of children
> per node by reducing the number of layers from 9 to 5. Reflect this on driver
> side.
>
> Signed-off-by: Raj Victor <victor....@intel.com>
> Co-developed-by: Michal Wilczynski <michal.wilczyn...@intel.com>
> Signed-off-by: Michal Wilczynski <michal.wilczyn...@intel.com>
> Co-developed-by: Mateusz Polchlopek <mateusz.polchlo...@intel.com>
> Signed-off-by: Mateusz Polchlopek <mateusz.polchlo...@intel.com>
> ---
> .../net/ethernet/intel/ice/ice_adminq_cmd.h | 23 ++
> drivers/net/ethernet/intel/ice/ice_common.c | 5 +
> drivers/net/ethernet/intel/ice/ice_ddp.c | 205 ++++++++++++++++++
> drivers/net/ethernet/intel/ice/ice_ddp.h | 2 +
> drivers/net/ethernet/intel/ice/ice_sched.h | 3 +
> drivers/net/ethernet/intel/ice/ice_type.h | 1 +
> 6 files changed, 239 insertions(+)
>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pu...@intel.com> (A
Contingent worker at Intel)
