Le lun. 30 oct. 2017 à 22:23, Bill Fischofer <bill.fischo...@linaro.org> a écrit :
> On Mon, Oct 30, 2017 at 2:54 PM, Honnappa Nagarahalli < > honnappa.nagaraha...@linaro.org> wrote: > > > Bill mentioned that the packets are flying through net-mdev, good news :) > > > > I mentioned that Ilias and Mykyta have gotten packet I/O through net_mdev. > I'll let them characterize whether it's "flying". We're still in early days > with this work. :) > > > > > > I see clear distinction in what needs to be done. > > > > 1) DDF - scanning/enumerating the devices, matching the drivers for > > the devices etc > > > > We need to decide how much, if any, of this work is actually needed as it > is duplicative of what Linux already does. > > [FF] it is not about replicating this is about leveraging. And for the Xth time, Linux does NOT enumerate vdev (vhostuser) Qemu does. We can also think of a pktio string : "DPDK:0" which is port 0 of DPDK. Other ways are possible but I hope this illustrate the idea of enumeration. > > > 2) Driver/packet I/O - Device initialization/configuration and actual > > packet send/recv > > > > This is the main thing we need. > > > > > > As the discussion on DDF continues, I think we can work on 2). We > > could enable/disable the packet I/O ops during compilation based on > > the NIC in the platform. For ex: if the platform has x710 card, we can > > enable the respective packet I/O during compilation. This portion of > > the code will remain more or less intact irrespective of the outcome > > of 1). > > > > The enable/disable of the packet I/O during compilation can be removed > > once we have the automation from 1) > > > > Once a device is named (something the application does when it calls > odp_pktio_open()), it either needs to be connected via net_mdev or else > have ownership transferred to it. In either case this is an OS > responsibility. The only question is whether we use existing system calls > or if new ones are needed to do this. > > [FF] not when we deal with full VFIO. OS responsability is not on device but on device configuration space and IO. > > > > > Please let me know what you think. > > > > Thank you, > > Honnappa > > > -- [image: Linaro] <http://www.linaro.org/> François-Frédéric Ozog | *Director Linaro Networking Group* T: +33.67221.6485 <javascript:void(0);> francois.o...@linaro.org | Skype: ffozog