On 2021/1/18 下午8:03, Xuan Zhuo wrote:
On Mon, 18 Jan 2021 17:10:24 +0800, Jason Wang wrote:
On 2021/1/16 上午10:59, Xuan Zhuo wrote:
virtnet_xsk_run will be called in the tx interrupt handling function
virtnet_poll_tx.
The sending process gets desc from the xsk tx queue, and assembles it to
s
On Mon, Jan 18, 2021 at 05:10:24PM +0800, Jason Wang wrote:
>
> On 2021/1/16 上午10:59, Xuan Zhuo wrote:
> > virtnet_xsk_run will be called in the tx interrupt handling function
> > virtnet_poll_tx.
> >
> > The sending process gets desc from the xsk tx queue, and assembles it to
> > send the data.
On 2021/1/16 上午10:59, Xuan Zhuo wrote:
virtnet_xsk_run will be called in the tx interrupt handling function
virtnet_poll_tx.
The sending process gets desc from the xsk tx queue, and assembles it to
send the data.
Compared with other drivers, a special place is that the page of the
data in xsk
On Sat, 16 Jan 2021 10:59:26 +0800 Xuan Zhuo wrote:
> + idx = sq->xsk.hdr_con % sq->xsk.hdr_n;
The arguments here are 64 bit, this code will not build on 32 bit
machines:
ERROR: modpost: "__umoddi3" [drivers/net/virtio_net.ko] undefined!
There's also a sparse warning in this patch:
drivers/
virtnet_xsk_run will be called in the tx interrupt handling function
virtnet_poll_tx.
The sending process gets desc from the xsk tx queue, and assembles it to
send the data.
Compared with other drivers, a special place is that the page of the
data in xsk is used here instead of the dma address. B