On Sat, Nov 11, 2017 at 06:45:04PM +0900, David Miller wrote: > From: Simon Guinot <simon.guinot@xxxxxxxxxxxx> > Date: Wed, 8 Nov 2017 17:58:35 +0100 > > > @@ -2413,8 +2416,7 @@ static int mvneta_tx(struct sk_buff *skb, struct net_device *dev) > > if (txq->count >= txq->tx_stop_threshold) > > netif_tx_stop_queue(nq); > > > > - if (!skb->xmit_more || netif_xmit_stopped(nq) || > > - txq->pending + frags > MVNETA_TXQ_DEC_SENT_MASK) > > + if (!skb->xmit_more || netif_xmit_stopped(nq)) > > mvneta_txq_pend_desc_add(pp, txq, frags); > > else > > txq->pending += frags; > > As David Laight said, you should not allow unlimited amounts of > ->xmit_more frames to be queued without a TX doorbell update. > > Therefore, please keep some kind of limit here otherwise latency > will spike in some circumstances. Hi David, IIUC the driver stops the queue if a threshold of 316 Tx descriptors is reached (default and worst value). Is that a "good enough" kind of limit ? Or do you want me to keep the condition above ? Simon
Attachment:
signature.asc
Description: PGP signature