Hi Jason, On Tue, 14 Jan 2025 16:48:28 -0400 Jason Gunthorpe <jgg@xxxxxxxxxx> wrote: > > I think we need to retain the ib_get_curr_port_state() call: Why? It is no longer used to determine the attr->phys-state value and then attr->state is set again just below. And ib_get_curr_port_state(ndev) dose not appear to have side effects. > --- drivers/infiniband/sw/siw/siw_verbs.c 2025-01-14 16:37:02.023738738 -0400 > +++ /home/jgg/oss/testing-k.o/drivers/infiniband/sw/siw/siw_verbs.c 2025-01-14 16:42:50.797896811 -0400 > @@ -189,6 +189,7 @@ > attr->max_msg_sz = -1; > attr->max_mtu = ib_mtu_int_to_enum(ndev->max_mtu); > attr->active_mtu = ib_mtu_int_to_enum(READ_ONCE(ndev->mtu)); > + attr->state = ib_get_curr_port_state(ndev); > attr->phys_state = (netif_running(ndev) && netif_carrier_ok(ndev)) ? > IB_PORT_PHYS_STATE_LINK_UP : IB_PORT_PHYS_STATE_DISABLED; > attr->state = attr->phys_state == IB_PORT_PHYS_STATE_LINK_UP ? -- Cheers, Stephen Rothwell
Attachment:
pgp4DdD7Pipx7.pgp
Description: OpenPGP digital signature
