On Fri, Dec 29, 2017 at 02:52:18PM -0600, Daniel Jurgens wrote: > On 12/29/2017 12:03 PM, Jason Gunthorpe wrote: > > On Wed, Dec 27, 2017 at 05:20:26PM -0600, Daniel Jurgens wrote: > >> On 12/27/2017 3:22 PM, Or Gerlitz wrote: > >>> On Wed, Dec 27, 2017 at 5:22 PM, Daniel Jurgens <danielj@xxxxxxxxxxxx> wrote: > >>> > >>>> If the PCI device were a slave there would be no IB device for it in the VM. > >>> This creates a regression for users that update their kernel and now > >>> their VMs stop working > >> The feature is not enabled by default. > > You never described how it is enabled in the commit messages either.. > > It's enabled vi FW ini file/PSID or mlxconfig. Do we ever describe that? If you add a new feature the commit message should briefly discussed how it is used, yes. Just saying 'Dual port mode is enabled by configuring the mlx5 firmware' is probably sufficient in this case. Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
