On Fri, Mar 4, 2016 at 4:37 PM, Doug Ledford <dledford@xxxxxxxxxx> wrote: > On 03/01/2016 02:46 PM, Eli Cohen wrote: >> On Tue, Mar 01, 2016 at 12:31:53PM -0700, Jason Gunthorpe wrote: >>> >>> I mean opensm supports the GUID Alias scheme for virtualization, this >>> new virtualization scheme is not compatible, and we shouldn't have the >>> kernel drop support for existing working SMs, by, eg, replacing the >>> mlx4 guid alias scheme with this new scheme. >>> >>> I'm guessing a user controlled switch is going to be necessary here to >>> pick GUID alias or port port virtualization. >> The alias GUID mechanism remains and can be used with mlx4 devices. >> With this scheme the admin configures the port and node GUIDs using >> iprout2 which ends up in the hardware driver configuring the deivce. A >> virtualization aware SM can read this configuration through MADs. > If the alias GUID mechanism is to be retained, then we need another NDO > entry point for setting P_Keys on alias GUID VFs. If we are going to > switch over to using iproute2, then the solution needs to be complete. Doug, as you commented on the net/core patch, iproute2 et al can't be really used today to build mlx4 based SRIOV cloud systems. This series doesn't introduce regressions for the way mlx4 setups are made. Or. -- 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