On Thu, Jun 27, 2019 at 02:23:05PM +0200, Jesper Dangaard Brouer wrote:
> On Tue, 25 Jun 2019 18:06:19 +0300
> Ilias Apalodimas <ilias.apalodi...@linaro.org> wrote:
>
> > @@ -609,6 +639,9 @@ static bool netsec_clean_tx_dring(struct netsec_priv
> > *priv)
> > int tail = dring->tail;
> > int cnt = 0;
> >
> > + if (dring->is_xdp)
> > + spin_lock(&dring->lock);
> > +
> > pkts = 0;
> > bytes = 0;
> > entry = dring->vaddr + DESC_SZ * tail;
> > @@ -622,16 +655,24 @@ static bool netsec_clean_tx_dring(struct netsec_priv
> > *priv)
> > eop = (entry->attr >> NETSEC_TX_LAST) & 1;
> > dma_rmb();
> >
> > - dma_unmap_single(priv->dev, desc->dma_addr, desc->len,
> > - DMA_TO_DEVICE);
> > - if (eop) {
> > - pkts++;
> > + if (!eop)
> > + goto next;
> > +
> > + if (desc->buf_type == TYPE_NETSEC_SKB) {
> > + dma_unmap_single(priv->dev, desc->dma_addr, desc->len,
> > + DMA_TO_DEVICE);
>
> I don't think this is correct. If I read the code correctly, you will
> miss the DMA unmap for !eop packets.
>
You are reading it correct, thanks for catching this.
I'll fix it on the proper patch
Thanks
/Ilias
