On 2014/08/22 16:17, Prashant Sreedharan wrote:
> Benjamin, thanks for the patch. Broadcom QA will be testing the changes.
> Couple of comments below.
> > segs = skb_gso_segment(skb, tp->dev->features &
> > ~(NETIF_F_TSO | NETIF_F_TSO6));
> > - if (IS_ERR(segs) || !segs)
> > + if (IS_ERR_OR_NULL(segs))
> > goto tg3_tso_bug_end;
> >
> > do {
> > + unsigned int desc_cnt = skb_shinfo(segs)->nr_frags + 1;
> > +
> > nskb = segs;
> > segs = segs->next;
> > nskb->next = NULL;
> > - tg3_start_xmit(nskb, tp->dev);
> > +
> > + if (tg3_tx_avail(tnapi) <= segs_remaining - 1 + desc_cnt &&
> > + skb_linearize(nskb)) {
> > + nskb->next = segs;
> > + segs = nskb;
> > + do {
> > + nskb = segs->next;
> > +
> > + dev_kfree_skb_any(segs);
> > + segs = nskb;
> > + } while (segs);
>
> If skb_linearize() fails need to increment the tp->tx_dropped count
Sorry for the delay, while testing this error path I noticed a potential
problem. There should be an additional check here to stop the queue with
the default threshold. Otherwise, the netdev_err message at the start of
__tg3_start_xmit() could be triggered when the next frame is
transmitted. That is because the previous calls to __tg3_start_xmit() in
tg3_tso_bug() may have been using a stop_thresh=segs_remaining that is <
MAX_SKB_FRAGS + 1.
>
> > + goto tg3_tso_bug_end;
> > + }
> > + segs_remaining--;
> > + if (segs_remaining)
> > + __tg3_start_xmit(nskb, tp->dev, segs_remaining);
>
> To clarify passing segs_remaining will make sure the queue is never
> stopped correct ?
It makes sure the queue is not stopped before we are finished submitting
all gso segments.
This is what's alluded to in this part of the commit message:
This puts us in the exceptional situation that a single skb that
triggers tg3_tso_bug() may require the entire tx ring. [...]
Likewise, usually the tx queue is stopped as soon as an skb with
max frags may overrun it. Since the skbs submitted from
tg3_tso_bug() use a controlled number of descriptors, the tx
queue stop threshold may be lowered.
>
> > + else
> > + tg3_start_xmit(nskb, tp->dev);
> > } while (segs);
> >
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
