From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Thu, 9 Feb 2023 07:48:05 -0500 From: "Michael S. Tsirkin" Subject: Re: [virtio-comment] [PATCH v3] virtio-net: Improve introductory description Message-ID: <20230209074712-mutt-send-email-mst@kernel.org> References: <20230209014108.209950-1-parav@nvidia.com> MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=us-ascii Content-Disposition: inline To: Parav Pandit Cc: David Edmondson , "virtio-dev@lists.oasis-open.org" , "cohuck@redhat.com" , Shahaf Shuler , "virtio-comment@lists.oasis-open.org" List-ID: On Thu, Feb 09, 2023 at 12:45:09PM +0000, Parav Pandit wrote: > > > From: David Edmondson > > Sent: Thursday, February 9, 2023 7:03 AM > > > > [..] > > > +to the Ethernet network. The device has transmit and receive queues. > > > +The driver posts empty buffers in the receive virtqueue. > > > +The device receives the incoming packets from the link; the device > > > +places these incoming packets in the receive virtqueue buffers. > > > +The driver enqueues outgoing packets to the transmit virtqueue. The > > > device > > > > We "post" receive buffers, yet "enqueue" transmit buffers. These actions seem > > the same. > > > > Maybe "submits" in both cases? > > > At virtqueue level it is same. > In context of virtio net device, transmit operation is queued that awaits processing (to send) by device. > While receive side, its unsolicited operation done by driver that doesn't await processing by device. > Whenever/if in future when a packet arrives, the posted buffer is consumed. > > So enqueue and post were capturing this contextual detail. > But maybe I am trying to be too precise here but above wording were more natural to me at net level. :) > Let me know.. Just "add" - this is what happens at vq level: Driver makes requests available to device by adding an available buffer to the queue, i.e., adding a buffer -- MST