Messages in this thread | | | Date | Sat, 21 Feb 2015 18:46:25 +0100 | From | Florian Westphal <> | Subject | Re: 1e918876 breaks r8169 (linux-3.18+) |
| |
Eric Dumazet <eric.dumazet@gmail.com> wrote: > Hold on. > > I believe there is one race in the way you access skb->xmit_more _after_ > > txd->opts1 = cpu_to_le32(status); > > After this point, TX might have completed and TX completion already have > freed skb
Hmm, I _thought_ HW would not start xmit of this descriptor/skb until after
RTL_W8(TxPoll, NPQ);
call a bit later... your patch surely looks safer though, thanks for looking into this.
> Could Tomas try following fix ?
In addition -- Tomas, on your affected board -- did you try turning off gso/tso?
Holger Hoffstätte mentioned on lkml that his affected r8169 nic is stable with xmit_more+bql after disabling gso/tso on the nic
(ethtool -k $dev to display settings, "-K $dev tso off gso off" to disable offloads).
| |