On Tue, Sep 2, 2014 at 9:05 PM, Jay Rolette <rolette at infiniteio.com> wrote:
> Couple of minor nits noted. The bigger feedback would be that if you could > keep the "whitespace / line-wrapping / formatting" only changes in separate > patches, these reviews would be easier to get through. > Hum, indeed, I should have done this. Will keep in mind. /* Free memory prior to re-allocation if needed */ >> if (dev->data->tx_queues[queue_idx] != NULL) { >> @@ -1271,7 +1271,7 @@ eth_igb_tx_queue_setup(struct rte_eth_dev *dev, >> igb_tx_queue_release(txq); >> return (-ENOMEM); >> } >> - PMD_INIT_LOG(DEBUG, "sw_ring=%p hw_ring=%p >> dma_addr=0x%"PRIx64"\n", >> + PMD_INIT_LOG(DEBUG, "sw_ring=%p hw_ring=%p dma_addr=0x%"PRIx64"", >> > The empty string after PRIx64 is extraneous > > >> txq->sw_ring, txq->tx_ring, txq->tx_ring_phys_addr); >> >> igb_reset_tx_queue(txq, dev); >> @@ -1409,7 +1409,7 @@ eth_igb_rx_queue_setup(struct rte_eth_dev *dev, >> igb_rx_queue_release(rxq); >> return (-ENOMEM); >> } >> - PMD_INIT_LOG(DEBUG, "sw_ring=%p hw_ring=%p >> dma_addr=0x%"PRIx64"\n", >> + PMD_INIT_LOG(DEBUG, "sw_ring=%p hw_ring=%p dma_addr=0x%"PRIx64"", >> > The empty string after PRIx64 is extraneous > > Ok, I will update. Thanks. -- David Marchand