Messages in this thread Patch in this message | | | From | Vladimir Oltean <> | Subject | [PATCH v3 net-next 05/13] net/sched: mqprio: add an extack message to mqprio_parse_opt() | Date | Mon, 20 Feb 2023 14:23:35 +0200 |
| |
Ferenc reports that a combination of poor iproute2 defaults and obscure cases where the kernel returns -EINVAL make it difficult to understand what is wrong with this command:
$ ip link add veth0 numtxqueues 8 numrxqueues 8 type veth peer name veth1 $ tc qdisc add dev veth0 root mqprio num_tc 8 map 0 1 2 3 4 5 6 7 \ queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 RTNETLINK answers: Invalid argument
Hopefully with this patch, the cause is clearer:
Error: Device does not support hardware offload.
The kernel was (and still is) rejecting this because iproute2 defaults to "hw 1" if this command line option is not specified.
Link: https://lore.kernel.org/netdev/ede5e9a2f27bf83bfb86d3e8c4ca7b34093b99e2.camel@inf.elte.hu/ Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Ferenc Fejes <fejes@inf.elte.hu> --- v2->v3: change link from patchwork to lore v1->v2: slightly reword last paragraph of commit message
net/sched/sch_mqprio.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index 5a9261c38b95..9ee5a9a9b9e9 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -133,8 +133,11 @@ static int mqprio_parse_opt(struct net_device *dev, struct tc_mqprio_qopt *qopt, /* If ndo_setup_tc is not present then hardware doesn't support offload * and we should return an error. */ - if (qopt->hw && !dev->netdev_ops->ndo_setup_tc) + if (qopt->hw && !dev->netdev_ops->ndo_setup_tc) { + NL_SET_ERR_MSG(extack, + "Device does not support hardware offload"); return -EINVAL; + } return 0; } -- 2.34.1
| |