-----Original Message-----
From: Lance Richardson <redacted>
Sent: Tuesday, January 26, 2021 11:45
To: Yigit, Ferruh <redacted>
Cc: Lu, Wenzhuo <redacted>; Li, Xiaoyun <redacted>;
Iremonger, Bernard [off-list ref]; Yang, SteveX
[off-list ref]; dev@dpdk.org; stable@dpdk.org;
oulijun@huawei.com; wisamm@mellanox.com; lihuisong@huawei.com
Subject: Re: [PATCH v5] app/testpmd: fix setting maximum packet length

On Mon, Jan 25, 2021 at 7:44 PM Ferruh Yigit [off-list ref] wrote:

quoted

quoted

quoted

+       if (rx_offloads != port->dev_conf.rxmode.offloads) {
+               uint16_t qid;
+
+               port->dev_conf.rxmode.offloads = rx_offloads;
+
+               /* Apply JUMBO_FRAME offload configuration to Rx queue(s) */
+               for (qid = 0; qid < port->dev_info.nb_rx_queues; qid++) {
+                       if (on)
+                               port->rx_conf[qid].offloads |=

DEV_RX_OFFLOAD_JUMBO_FRAME;

quoted

quoted

quoted

+                       else
+                               port->rx_conf[qid].offloads &=

~DEV_RX_OFFLOAD_JUMBO_FRAME;

quoted

quoted

quoted

+               }

Is it correct to set per-queue offloads that aren't advertised by the PMD
as supported in rx_queue_offload_capa?

'port->rx_conf[]' is testpmd struct, and 'port->dev_conf.rxmode.offloads'

values

quoted

are reflected to 'port->rx_conf[].offloads' for all queues.

We should set the offload in 'port->rx_conf[].offloads' if it is set in
'port->dev_conf.rxmode.offloads'.

If a port has capability for 'JUMBO_FRAME', 'port->rx_conf[].offloads' can

have

quoted

it. And the port level capability is already checked above.

I'm still not 100% clear about the per-queue offload question.

With this patch, and jumbo max packet size configured (on the command
line in this case), I see:

testpmd> show port 0 rx_offload configuration
Rx Offloading Configuration of port 0 :
  Port : JUMBO_FRAME
  Queue[ 0] : JUMBO_FRAME

testpmd> show port 0 rx_offload capabilities
Rx Offloading Capabilities of port 0 :
  Per Queue :
  Per Port  : VLAN_STRIP IPV4_CKSUM UDP_CKSUM TCP_CKSUM TCP_LRO
OUTER_IPV4_CKSUM VLAN_FILTER VLAN_EXTEND JUMBO_FRAME SCATTER
TIMESTAMP
KEEP_CRC OUTER_UDP_CKSUM RSS_HASH

Yet if I configure a jumbo MTU starting with standard max packet size,
jumbo is only enabled at the port level:
testpmd> port config mtu 0 9000
testpmd> port start all

testpmd> show port 0 rx_offload configuration
Rx Offloading Configuration of port 0 :
  Port : JUMBO_FRAME
  Queue[ 0] :

It still seems odd for a per-queue offload to be enabled on a PMD that
doesn't support per-queue receive offloads.

In struct rte_eth_dev_info, rx_offload_capa means All RX offload capabilities including all per-queue ones.
And rx_queue_offload_capa means Device per-queue RX offload capabilities.

The meaning of rx_queue_offload_capa is a bit of confusing between here and driver.
I think here rx_queue_offload_capa means whether a queue supports offloads.

But some drivers like i40e don't use rx_queue_offload_capa, set this as 0 and only use global rx_offload_capa.
I guess it's because the driver doesn't want to support different offloads settings for different queues?
Then rx_queue_offload_capa means differently.

Actually for i40e, it can support different offloads for different queues since there is 'offloads' in struct i40e_rx_queue.
I40e can just set rx_queue_offload_capa as the value for rx_offload_capa.
But maybe some drivers really don't want this? Not sure on this.