diff options
Diffstat (limited to 'drivers/net/ethernet/freescale')
-rw-r--r-- | drivers/net/ethernet/freescale/fec_main.c | 41 | ||||
-rw-r--r-- | drivers/net/ethernet/freescale/fec_ptp.c | 6 | ||||
-rw-r--r-- | drivers/net/ethernet/freescale/gianfar.c | 502 | ||||
-rw-r--r-- | drivers/net/ethernet/freescale/gianfar.h | 77 | ||||
-rw-r--r-- | drivers/net/ethernet/freescale/gianfar_ethtool.c | 4 |
5 files changed, 363 insertions, 267 deletions
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 271bb58..787da8e 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -364,7 +364,7 @@ fec_enet_clear_csum(struct sk_buff *skb, struct net_device *ndev) return 0; } -static int +static struct bufdesc * fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq, struct sk_buff *skb, struct net_device *ndev) @@ -439,10 +439,7 @@ fec_enet_txq_submit_frag_skb(struct fec_enet_priv_tx_q *txq, bdp->cbd_sc = status; } - txq->cur_tx = bdp; - - return 0; - + return bdp; dma_mapping_error: bdp = txq->cur_tx; for (i = 0; i < frag; i++) { @@ -450,7 +447,7 @@ dma_mapping_error: dma_unmap_single(&fep->pdev->dev, bdp->cbd_bufaddr, bdp->cbd_datlen, DMA_TO_DEVICE); } - return NETDEV_TX_OK; + return ERR_PTR(-ENOMEM); } static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq, @@ -467,7 +464,6 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq, unsigned int estatus = 0; unsigned int index; int entries_free; - int ret; entries_free = fec_enet_get_free_txdesc_num(fep, txq); if (entries_free < MAX_SKB_FRAGS + 1) { @@ -485,6 +481,7 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq, /* Fill in a Tx ring entry */ bdp = txq->cur_tx; + last_bdp = bdp; status = bdp->cbd_sc; status &= ~BD_ENET_TX_STATS; @@ -513,9 +510,9 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq, } if (nr_frags) { - ret = fec_enet_txq_submit_frag_skb(txq, skb, ndev); - if (ret) - return ret; + last_bdp = fec_enet_txq_submit_frag_skb(txq, skb, ndev); + if (IS_ERR(last_bdp)) + return NETDEV_TX_OK; } else { status |= (BD_ENET_TX_INTR | BD_ENET_TX_LAST); if (fep->bufdesc_ex) { @@ -544,7 +541,6 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq, ebdp->cbd_esc = estatus; } - last_bdp = txq->cur_tx; index = fec_enet_get_bd_index(txq->tx_bd_base, last_bdp, fep); /* Save skb pointer */ txq->tx_skbuff[index] = skb; @@ -563,6 +559,10 @@ static int fec_enet_txq_submit_skb(struct fec_enet_priv_tx_q *txq, skb_tx_timestamp(skb); + /* Make sure the update to bdp and tx_skbuff are performed before + * cur_tx. + */ + wmb(); txq->cur_tx = bdp; /* Trigger transmission start */ @@ -1218,10 +1218,11 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id) /* get next bdp of dirty_tx */ bdp = fec_enet_get_nextdesc(bdp, fep, queue_id); - while (((status = bdp->cbd_sc) & BD_ENET_TX_READY) == 0) { - - /* current queue is empty */ - if (bdp == txq->cur_tx) + while (bdp != READ_ONCE(txq->cur_tx)) { + /* Order the load of cur_tx and cbd_sc */ + rmb(); + status = READ_ONCE(bdp->cbd_sc); + if (status & BD_ENET_TX_READY) break; index = fec_enet_get_bd_index(txq->tx_bd_base, bdp, fep); @@ -1275,6 +1276,10 @@ fec_enet_tx_queue(struct net_device *ndev, u16 queue_id) /* Free the sk buffer associated with this last transmit */ dev_kfree_skb_any(skb); + /* Make sure the update to bdp and tx_skbuff are performed + * before dirty_tx + */ + wmb(); txq->dirty_tx = bdp; /* Update pointer to next buffer descriptor to be transmitted */ @@ -1774,7 +1779,7 @@ static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) int ret = 0; ret = pm_runtime_get_sync(dev); - if (IS_ERR_VALUE(ret)) + if (ret < 0) return ret; fep->mii_timeout = 0; @@ -1813,7 +1818,7 @@ static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, int ret = 0; ret = pm_runtime_get_sync(dev); - if (IS_ERR_VALUE(ret)) + if (ret < 0) return ret; fep->mii_timeout = 0; @@ -2865,7 +2870,7 @@ fec_enet_open(struct net_device *ndev) int ret; ret = pm_runtime_get_sync(&fep->pdev->dev); - if (IS_ERR_VALUE(ret)) + if (ret < 0) return ret; pinctrl_pm_select_default_state(&fep->pdev->dev); diff --git a/drivers/net/ethernet/freescale/fec_ptp.c b/drivers/net/ethernet/freescale/fec_ptp.c index f457a23..1543cf0 100644 --- a/drivers/net/ethernet/freescale/fec_ptp.c +++ b/drivers/net/ethernet/freescale/fec_ptp.c @@ -506,12 +506,6 @@ int fec_ptp_set(struct net_device *ndev, struct ifreq *ifr) break; default: - /* - * register RXMTRL must be set in order to do V1 packets, - * therefore it is not possible to time stamp both V1 Sync and - * Delay_Req messages and hardware does not support - * timestamping all packets => return error - */ fep->hwts_rx_en = 1; config.rx_filter = HWTSTAMP_FILTER_ALL; break; diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c index 10b3bbbb..4b69d061 100644 --- a/drivers/net/ethernet/freescale/gianfar.c +++ b/drivers/net/ethernet/freescale/gianfar.c @@ -109,15 +109,15 @@ #define TX_TIMEOUT (1*HZ) -const char gfar_driver_version[] = "1.3"; +const char gfar_driver_version[] = "2.0"; static int gfar_enet_open(struct net_device *dev); static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev); static void gfar_reset_task(struct work_struct *work); static void gfar_timeout(struct net_device *dev); static int gfar_close(struct net_device *dev); -static struct sk_buff *gfar_new_skb(struct net_device *dev, - dma_addr_t *bufaddr); +static void gfar_alloc_rx_buffs(struct gfar_priv_rx_q *rx_queue, + int alloc_cnt); static int gfar_set_mac_address(struct net_device *dev); static int gfar_change_mtu(struct net_device *dev, int new_mtu); static irqreturn_t gfar_error(int irq, void *dev_id); @@ -141,8 +141,7 @@ static void gfar_netpoll(struct net_device *dev); #endif int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit); static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue); -static void gfar_process_frame(struct net_device *dev, struct sk_buff *skb, - int amount_pull, struct napi_struct *napi); +static void gfar_process_frame(struct net_device *ndev, struct sk_buff *skb); static void gfar_halt_nodisable(struct gfar_private *priv); static void gfar_clear_exact_match(struct net_device *dev); static void gfar_set_mac_for_addr(struct net_device *dev, int num, @@ -169,17 +168,15 @@ static void gfar_init_rxbdp(struct gfar_priv_rx_q *rx_queue, struct rxbd8 *bdp, bdp->lstatus = cpu_to_be32(lstatus); } -static int gfar_init_bds(struct net_device *ndev) +static void gfar_init_bds(struct net_device *ndev) { struct gfar_private *priv = netdev_priv(ndev); struct gfar __iomem *regs = priv->gfargrp[0].regs; struct gfar_priv_tx_q *tx_queue = NULL; struct gfar_priv_rx_q *rx_queue = NULL; struct txbd8 *txbdp; - struct rxbd8 *rxbdp; u32 __iomem *rfbptr; int i, j; - dma_addr_t bufaddr; for (i = 0; i < priv->num_tx_queues; i++) { tx_queue = priv->tx_queue[i]; @@ -207,40 +204,26 @@ static int gfar_init_bds(struct net_device *ndev) rfbptr = ®s->rfbptr0; for (i = 0; i < priv->num_rx_queues; i++) { rx_queue = priv->rx_queue[i]; - rx_queue->cur_rx = rx_queue->rx_bd_base; - rx_queue->skb_currx = 0; - rxbdp = rx_queue->rx_bd_base; - for (j = 0; j < rx_queue->rx_ring_size; j++) { - struct sk_buff *skb = rx_queue->rx_skbuff[j]; + rx_queue->next_to_clean = 0; + rx_queue->next_to_use = 0; + rx_queue->next_to_alloc = 0; - if (skb) { - bufaddr = be32_to_cpu(rxbdp->bufPtr); - } else { - skb = gfar_new_skb(ndev, &bufaddr); - if (!skb) { - netdev_err(ndev, "Can't allocate RX buffers\n"); - return -ENOMEM; - } - rx_queue->rx_skbuff[j] = skb; - } - - gfar_init_rxbdp(rx_queue, rxbdp, bufaddr); - rxbdp++; - } + /* make sure next_to_clean != next_to_use after this + * by leaving at least 1 unused descriptor + */ + gfar_alloc_rx_buffs(rx_queue, gfar_rxbd_unused(rx_queue)); rx_queue->rfbptr = rfbptr; rfbptr += 2; } - - return 0; } static int gfar_alloc_skb_resources(struct net_device *ndev) { void *vaddr; dma_addr_t addr; - int i, j, k; + int i, j; struct gfar_private *priv = netdev_priv(ndev); struct device *dev = priv->dev; struct gfar_priv_tx_q *tx_queue = NULL; @@ -279,7 +262,8 @@ static int gfar_alloc_skb_resources(struct net_device *ndev) rx_queue = priv->rx_queue[i]; rx_queue->rx_bd_base = vaddr; rx_queue->rx_bd_dma_base = addr; - rx_queue->dev = ndev; + rx_queue->ndev = ndev; + rx_queue->dev = dev; addr += sizeof(struct rxbd8) * rx_queue->rx_ring_size; vaddr += sizeof(struct rxbd8) * rx_queue->rx_ring_size; } @@ -294,25 +278,20 @@ static int gfar_alloc_skb_resources(struct net_device *ndev) if (!tx_queue->tx_skbuff) goto cleanup; - for (k = 0; k < tx_queue->tx_ring_size; k++) - tx_queue->tx_skbuff[k] = NULL; + for (j = 0; j < tx_queue->tx_ring_size; j++) + tx_queue->tx_skbuff[j] = NULL; } for (i = 0; i < priv->num_rx_queues; i++) { rx_queue = priv->rx_queue[i]; - rx_queue->rx_skbuff = - kmalloc_array(rx_queue->rx_ring_size, - sizeof(*rx_queue->rx_skbuff), - GFP_KERNEL); - if (!rx_queue->rx_skbuff) + rx_queue->rx_buff = kcalloc(rx_queue->rx_ring_size, + sizeof(*rx_queue->rx_buff), + GFP_KERNEL); + if (!rx_queue->rx_buff) goto cleanup; - - for (j = 0; j < rx_queue->rx_ring_size; j++) - rx_queue->rx_skbuff[j] = NULL; } - if (gfar_init_bds(ndev)) - goto cleanup; + gfar_init_bds(ndev); return 0; @@ -354,10 +333,8 @@ static void gfar_init_rqprm(struct gfar_private *priv) } } -static void gfar_rx_buff_size_config(struct gfar_private *priv) +static void gfar_rx_offload_en(struct gfar_private *priv) { - int frame_size = priv->ndev->mtu + ETH_HLEN + ETH_FCS_LEN; - /* set this when rx hw offload (TOE) functions are being used */ priv->uses_rxfcb = 0; @@ -366,16 +343,6 @@ static void gfar_rx_buff_size_config(struct gfar_private *priv) if (priv->hwts_rx_en) priv->uses_rxfcb = 1; - - if (priv->uses_rxfcb) - frame_size += GMAC_FCB_LEN; - - frame_size += priv->padding; - - frame_size = (frame_size & ~(INCREMENTAL_BUFFER_SIZE - 1)) + - INCREMENTAL_BUFFER_SIZE; - - priv->rx_buffer_size = frame_size; } static void gfar_mac_rx_config(struct gfar_private *priv) @@ -593,9 +560,8 @@ static int gfar_alloc_rx_queues(struct gfar_private *priv) if (!priv->rx_queue[i]) return -ENOMEM; - priv->rx_queue[i]->rx_skbuff = NULL; priv->rx_queue[i]->qindex = i; - priv->rx_queue[i]->dev = priv->ndev; + priv->rx_queue[i]->ndev = priv->ndev; } return 0; } @@ -1187,12 +1153,11 @@ void gfar_mac_reset(struct gfar_private *priv) udelay(3); - /* Compute rx_buff_size based on config flags */ - gfar_rx_buff_size_config(priv); + gfar_rx_offload_en(priv); /* Initialize the max receive frame/buffer lengths */ - gfar_write(®s->maxfrm, priv->rx_buffer_size); - gfar_write(®s->mrblr, priv->rx_buffer_size); + gfar_write(®s->maxfrm, GFAR_JUMBO_FRAME_SIZE); + gfar_write(®s->mrblr, GFAR_RXB_SIZE); /* Initialize the Minimum Frame Length Register */ gfar_write(®s->minflr, MINFLR_INIT_SETTINGS); @@ -1200,12 +1165,11 @@ void gfar_mac_reset(struct gfar_private *priv) /* Initialize MACCFG2. */ tempval = MACCFG2_INIT_SETTINGS; - /* If the mtu is larger than the max size for standard - * ethernet frames (ie, a jumbo frame), then set maccfg2 - * to allow huge frames, and to check the length + /* eTSEC74 erratum: Rx frames of length MAXFRM or MAXFRM-1 + * are marked as truncated. Avoid this by MACCFG2[Huge Frame]=1, + * and by checking RxBD[LG] and discarding larger than MAXFRM. */ - if (priv->rx_buffer_size > DEFAULT_RX_BUFFER_SIZE || - gfar_has_errata(priv, GFAR_ERRATA_74)) + if (gfar_has_errata(priv, GFAR_ERRATA_74)) tempval |= MACCFG2_HUGEFRAME | MACCFG2_LENGTHCHECK; gfar_write(®s->maccfg2, tempval); @@ -1415,8 +1379,6 @@ static int gfar_probe(struct platform_device *ofdev) priv->device_flags & FSL_GIANFAR_DEV_HAS_TIMER) dev->needed_headroom = GMAC_FCB_LEN; - priv->rx_buffer_size = DEFAULT_RX_BUFFER_SIZE; - /* Initializing some of the rx/tx queue level parameters */ for (i = 0; i < priv->num_tx_queues; i++) { priv->tx_queue[i]->tx_ring_size = DEFAULT_TX_RING_SIZE; @@ -1599,10 +1561,7 @@ static int gfar_restore(struct device *dev) return 0; } - if (gfar_init_bds(ndev)) { - free_skb_resources(priv); - return -ENOMEM; - } + gfar_init_bds(ndev); gfar_mac_reset(priv); @@ -1893,26 +1852,32 @@ static void free_skb_tx_queue(struct gfar_priv_tx_q *tx_queue) static void free_skb_rx_queue(struct gfar_priv_rx_q *rx_queue) { - struct rxbd8 *rxbdp; - struct gfar_private *priv = netdev_priv(rx_queue->dev); int i; - rxbdp = rx_queue->rx_bd_base; + struct rxbd8 *rxbdp = rx_queue->rx_bd_base; + + if (rx_queue->skb) + dev_kfree_skb(rx_queue->skb); for (i = 0; i < rx_queue->rx_ring_size; i++) { - if (rx_queue->rx_skbuff[i]) { - dma_unmap_single(priv->dev, be32_to_cpu(rxbdp->bufPtr), - priv->rx_buffer_size, - DMA_FROM_DEVICE); - dev_kfree_skb_any(rx_queue->rx_skbuff[i]); - rx_queue->rx_skbuff[i] = NULL; - } + struct gfar_rx_buff *rxb = &rx_queue->rx_buff[i]; + rxbdp->lstatus = 0; rxbdp->bufPtr = 0; rxbdp++; + + if (!rxb->page) + continue; + + dma_unmap_single(rx_queue->dev, rxb->dma, + PAGE_SIZE, DMA_FROM_DEVICE); + __free_page(rxb->page); + + rxb->page = NULL; } - kfree(rx_queue->rx_skbuff); - rx_queue->rx_skbuff = NULL; + + kfree(rx_queue->rx_buff); + rx_queue->rx_buff = NULL; } /* If there are any tx skbs or rx skbs still around, free them. @@ -1937,7 +1902,7 @@ static void free_skb_resources(struct gfar_private *priv) for (i = 0; i < priv->num_rx_queues; i++) { rx_queue = priv->rx_queue[i]; - if (rx_queue->rx_skbuff) + if (rx_queue->rx_buff) free_skb_rx_queue(rx_queue); } @@ -2500,7 +2465,7 @@ static int gfar_change_mtu(struct net_device *dev, int new_mtu) struct gfar_private *priv = netdev_priv(dev); int frame_size = new_mtu + ETH_HLEN; - if ((frame_size < 64) || (frame_size > JUMBO_FRAME_SIZE)) { + if ((frame_size < 64) || (frame_size > GFAR_JUMBO_FRAME_SIZE)) { netif_err(priv, drv, dev, "Invalid MTU setting\n"); return -EINVAL; } @@ -2554,15 +2519,6 @@ static void gfar_timeout(struct net_device *dev) schedule_work(&priv->reset_task); } -static void gfar_align_skb(struct sk_buff *skb) -{ - /* We need the data buffer to be aligned properly. We will reserve - * as many bytes as needed to align the data properly - */ - skb_reserve(skb, RXBUF_ALIGNMENT - - (((unsigned long) skb->data) & (RXBUF_ALIGNMENT - 1))); -} - /* Interrupt Handler for Transmit complete */ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) { @@ -2620,7 +2576,8 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) { struct skb_shared_hwtstamps shhwtstamps; - u64 *ns = (u64*) (((u32)skb->data + 0x10) & ~0x7); + u64 *ns = (u64 *)(((uintptr_t)skb->data + 0x10) & + ~0x7UL); memset(&shhwtstamps, 0, sizeof(shhwtstamps)); shhwtstamps.hwtstamp = ns_to_ktime(*ns); @@ -2669,49 +2626,85 @@ static void gfar_clean_tx_ring(struct gfar_priv_tx_q *tx_queue) netdev_tx_completed_queue(txq, howmany, bytes_sent); } -static struct sk_buff *gfar_alloc_skb(struct net_device *dev) +static bool gfar_new_page(struct gfar_priv_rx_q *rxq, struct gfar_rx_buff *rxb) { - struct gfar_private *priv = netdev_priv(dev); - struct sk_buff *skb; + struct page *page; + dma_addr_t addr; - skb = netdev_alloc_skb(dev, priv->rx_buffer_size + RXBUF_ALIGNMENT); - if (!skb) - return NULL; + page = dev_alloc_page(); + if (unlikely(!page)) + return false; - gfar_align_skb(skb); + addr = dma_map_page(rxq->dev, page, 0, PAGE_SIZE, DMA_FROM_DEVICE); + if (unlikely(dma_mapping_error(rxq->dev, addr))) { + __free_page(page); - return skb; + return false; + } + + rxb->dma = addr; + rxb->page = page; + rxb->page_offset = 0; + + return true; } -static struct sk_buff *gfar_new_skb(struct net_device *dev, dma_addr_t *bufaddr) +static void gfar_rx_alloc_err(struct gfar_priv_rx_q *rx_queue) { - struct gfar_private *priv = netdev_priv(dev); - struct sk_buff *skb; - dma_addr_t addr; + struct gfar_private *priv = netdev_priv(rx_queue->ndev); + struct gfar_extra_stats *estats = &priv->extra_stats; - skb = gfar_alloc_skb(dev); - if (!skb) - return NULL; + netdev_err(rx_queue->ndev, "Can't alloc RX buffers\n"); + atomic64_inc(&estats->rx_alloc_err); +} - addr = dma_map_single(priv->dev, skb->data, - priv->rx_buffer_size, DMA_FROM_DEVICE); - if (unlikely(dma_mapping_error(priv->dev, addr))) { - dev_kfree_skb_any(skb); - return NULL; +static void gfar_alloc_rx_buffs(struct gfar_priv_rx_q *rx_queue, + int alloc_cnt) +{ + struct rxbd8 *bdp; + struct gfar_rx_buff *rxb; + int i; + + i = rx_queue->next_to_use; + bdp = &rx_queue->rx_bd_base[i]; + rxb = &rx_queue->rx_buff[i]; + + while (alloc_cnt--) { + /* try reuse page */ + if (unlikely(!rxb->page)) { + if (unlikely(!gfar_new_page(rx_queue, rxb))) { + gfar_rx_alloc_err(rx_queue); + break; + } + } + + /* Setup the new RxBD */ + gfar_init_rxbdp(rx_queue, bdp, + rxb->dma + rxb->page_offset + RXBUF_ALIGNMENT); + + /* Update to the next pointer */ + bdp++; + rxb++; + + if (unlikely(++i == rx_queue->rx_ring_size)) { + i = 0; + bdp = rx_queue->rx_bd_base; + rxb = rx_queue->rx_buff; + } } - *bufaddr = addr; - return skb; + rx_queue->next_to_use = i; + rx_queue->next_to_alloc = i; } -static inline void count_errors(unsigned short status, struct net_device *dev) +static void count_errors(u32 lstatus, struct net_device *ndev) { - struct gfar_private *priv = netdev_priv(dev); - struct net_device_stats *stats = &dev->stats; + struct gfar_private *priv = netdev_priv(ndev); + struct net_device_stats *stats = &ndev->stats; struct gfar_extra_stats *estats = &priv->extra_stats; /* If the packet was truncated, none of the other errors matter */ - if (status & RXBD_TRUNCATED) { + if (lstatus & BD_LFLAG(RXBD_TRUNCATED)) { stats->rx_length_errors++; atomic64_inc(&estats->rx_trunc); @@ -2719,25 +2712,25 @@ static inline void count_errors(unsigned short status, struct net_device *dev) return; } /* Count the errors, if there were any */ - if (status & (RXBD_LARGE | RXBD_SHORT)) { + if (lstatus & BD_LFLAG(RXBD_LARGE | RXBD_SHORT)) { stats->rx_length_errors++; - if (status & RXBD_LARGE) + if (lstatus & BD_LFLAG(RXBD_LARGE)) atomic64_inc(&estats->rx_large); else atomic64_inc(&estats->rx_short); } - if (status & RXBD_NONOCTET) { + if (lstatus & BD_LFLAG(RXBD_NONOCTET)) { stats->rx_frame_errors++; atomic64_inc(&estats->rx_nonoctet); } - if (status & RXBD_CRCERR) { + if (lstatus & BD_LFLAG(RXBD_CRCERR)) { atomic64_inc(&estats->rx_crcerr); stats->rx_crc_errors++; } - if (status & RXBD_OVERRUN) { + if (lstatus & BD_LFLAG(RXBD_OVERRUN)) { atomic64_inc(&estats->rx_overrun); - stats->rx_crc_errors++; + stats->rx_over_errors++; } } @@ -2788,6 +2781,93 @@ static irqreturn_t gfar_transmit(int irq, void *grp_id) return IRQ_HANDLED; } +static bool gfar_add_rx_frag(struct gfar_rx_buff *rxb, u32 lstatus, + struct sk_buff *skb, bool first) +{ + unsigned int size = lstatus & BD_LENGTH_MASK; + struct page *page = rxb->page; + + /* Remove the FCS from the packet length */ + if (likely(lstatus & BD_LFLAG(RXBD_LAST))) + size -= ETH_FCS_LEN; + + if (likely(first)) + skb_put(skb, size); + else + skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page, + rxb->page_offset + RXBUF_ALIGNMENT, + size, GFAR_RXB_TRUESIZE); + + /* try reuse page */ + if (unlikely(page_count(page) != 1)) + return false; + + /* change offset to the other half */ + rxb->page_offset ^= GFAR_RXB_TRUESIZE; + + atomic_inc(&page->_count); + + return true; +} + +static void gfar_reuse_rx_page(struct gfar_priv_rx_q *rxq, + struct gfar_rx_buff *old_rxb) +{ + struct gfar_rx_buff *new_rxb; + u16 nta = rxq->next_to_alloc; + + new_rxb = &rxq->rx_buff[nta]; + + /* find next buf that can reuse a page */ + nta++; + rxq->next_to_alloc = (nta < rxq->rx_ring_size) ? nta : 0; + + /* copy page reference */ + *new_rxb = *old_rxb; + + /* sync for use by the device */ + dma_sync_single_range_for_device(rxq->dev, old_rxb->dma, + old_rxb->page_offset, + GFAR_RXB_TRUESIZE, DMA_FROM_DEVICE); +} + +static struct sk_buff *gfar_get_next_rxbuff(struct gfar_priv_rx_q *rx_queue, + u32 lstatus, struct sk_buff *skb) +{ + struct gfar_rx_buff *rxb = &rx_queue->rx_buff[rx_queue->next_to_clean]; + struct page *page = rxb->page; + bool first = false; + + if (likely(!skb)) { + void *buff_addr = page_address(page) + rxb->page_offset; + + skb = build_skb(buff_addr, GFAR_SKBFRAG_SIZE); + if (unlikely(!skb)) { + gfar_rx_alloc_err(rx_queue); + return NULL; + } + skb_reserve(skb, RXBUF_ALIGNMENT); + first = true; + } + + dma_sync_single_range_for_cpu(rx_queue->dev, rxb->dma, rxb->page_offset, + GFAR_RXB_TRUESIZE, DMA_FROM_DEVICE); + + if (gfar_add_rx_frag(rxb, lstatus, skb, first)) { + /* reuse the free half of the page */ + gfar_reuse_rx_page(rx_queue, rxb); + } else { + /* page cannot be reused, unmap it */ + dma_unmap_page(rx_queue->dev, rxb->dma, + PAGE_SIZE, DMA_FROM_DEVICE); + } + + /* clear rxb content */ + rxb->page = NULL; + + return skb; +} + static inline void gfar_rx_checksum(struct sk_buff *skb, struct rxfcb *fcb) { /* If valid headers were found, and valid sums @@ -2802,10 +2882,9 @@ static inline void gfar_rx_checksum(struct sk_buff *skb, struct rxfcb *fcb) } /* gfar_process_frame() -- handle one incoming packet if skb isn't NULL. */ -static void gfar_process_frame(struct net_device *dev, struct sk_buff *skb, - int amount_pull, struct napi_struct *napi) +static void gfar_process_frame(struct net_device *ndev, struct sk_buff *skb) { - struct gfar_private *priv = netdev_priv(dev); + struct gfar_private *priv = netdev_priv(ndev); struct rxfcb *fcb = NULL; /* fcb is at the beginning if exists */ @@ -2814,10 +2893,8 @@ static void gfar_process_frame(struct net_device *dev, struct sk_buff *skb, /* Remove the FCB from the skb * Remove the padded bytes, if there are any */ - if (amount_pull) { - skb_record_rx_queue(skb, fcb->rq); - skb_pull(skb, amount_pull); - } + if (priv->uses_rxfcb) + skb_pull(skb, GMAC_FCB_LEN); /* Get receive timestamp from the skb */ if (priv->hwts_rx_en) { @@ -2831,24 +2908,20 @@ static void gfar_process_frame(struct net_device *dev, struct sk_buff *skb, if (priv->padding) skb_pull(skb, priv->padding); - if (dev->features & NETIF_F_RXCSUM) + if (ndev->features & NETIF_F_RXCSUM) gfar_rx_checksum(skb, fcb); /* Tell the skb what kind of packet this is */ - skb->protocol = eth_type_trans(skb, dev); + skb->protocol = eth_type_trans(skb, ndev); /* There's need to check for NETIF_F_HW_VLAN_CTAG_RX here. * Even if vlan rx accel is disabled, on some chips * RXFCB_VLN is pseudo randomly set. */ - if (dev->features & NETIF_F_HW_VLAN_CTAG_RX && + if (ndev->features & NETIF_F_HW_VLAN_CTAG_RX && be16_to_cpu(fcb->flags) & RXFCB_VLN) __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), be16_to_cpu(fcb->vlctl)); - - /* Send the packet up the stack */ - napi_gro_receive(napi, skb); - } /* gfar_clean_rx_ring() -- Processes each frame in the rx ring @@ -2857,91 +2930,89 @@ static void gfar_process_frame(struct net_device *dev, struct sk_buff *skb, */ int gfar_clean_rx_ring(struct gfar_priv_rx_q *rx_queue, int rx_work_limit) { - struct net_device *dev = rx_queue->dev; - struct rxbd8 *bdp, *base; - struct sk_buff *skb; - int pkt_len; - int amount_pull; - int howmany = 0; - struct gfar_private *priv = netdev_priv(dev); + struct net_device *ndev = rx_queue->ndev; + struct gfar_private *priv = netdev_priv(ndev); + struct rxbd8 *bdp; + int i, howmany = 0; + struct sk_buff *skb = rx_queue->skb; + int cleaned_cnt = gfar_rxbd_unused(rx_queue); + unsigned int total_bytes = 0, total_pkts = 0; /* Get the first full descriptor */ - bdp = rx_queue->cur_rx; - base = rx_queue->rx_bd_base; + i = rx_queue->next_to_clean; - amount_pull = priv->uses_rxfcb ? GMAC_FCB_LEN : 0; + while (rx_work_limit--) { + u32 lstatus; + + if (cleaned_cnt >= GFAR_RX_BUFF_ALLOC) { + gfar_alloc_rx_buffs(rx_queue, cleaned_cnt); + cleaned_cnt = 0; + } - while (!(be16_to_cpu(bdp->status) & RXBD_EMPTY) && rx_work_limit--) { - struct sk_buff *newskb; - dma_addr_t bufaddr; + bdp = &rx_queue->rx_bd_base[i]; + lstatus = be32_to_cpu(bdp->lstatus); + if (lstatus & BD_LFLAG(RXBD_EMPTY)) + break; + /* order rx buffer descriptor reads */ rmb(); - /* Add another skb for the future */ - newskb = gfar_new_skb(dev, &bufaddr); + /* fetch next to clean buffer from the ring */ + skb = gfar_get_next_rxbuff(rx_queue, lstatus, skb); + if (unlikely(!skb)) + break; - skb = rx_queue->rx_skbuff[rx_queue->skb_currx]; + cleaned_cnt++; + howmany++; - dma_unmap_single(priv->dev, be32_to_cpu(bdp->bufPtr), - priv->rx_buffer_size, DMA_FROM_DEVICE); - - if (unlikely(!(be16_to_cpu(bdp->status) & RXBD_ERR) && - be16_to_cpu(bdp->length) > priv->rx_buffer_size)) - bdp->status = cpu_to_be16(RXBD_LARGE); - - /* We drop the frame if we failed to allocate a new buffer */ - if (unlikely(!newskb || - !(be16_to_cpu(bdp->status) & RXBD_LAST) || - be16_to_cpu(bdp->status) & RXBD_ERR)) { - count_errors(be16_to_cpu(bdp->status), dev); - - if (unlikely(!newskb)) { - newskb = skb; - bufaddr = be32_to_cpu(bdp->bufPtr); - } else if (skb) - dev_kfree_skb(skb); - } else { - /* Increment the number of packets */ - rx_queue->stats.rx_packets++; - howmany++; - - if (likely(skb)) { - pkt_len = be16_to_cpu(bdp->length) - - ETH_FCS_LEN; - /* Remove the FCS from the packet length */ - skb_put(skb, pkt_len); - rx_queue->stats.rx_bytes += pkt_len; - skb_record_rx_queue(skb, rx_queue->qindex); - gfar_process_frame(dev, skb, amount_pull, - &rx_queue->grp->napi_rx); + if (unlikely(++i == rx_queue->rx_ring_size)) + i = 0; - } else { - netif_warn(priv, rx_err, dev, "Missing skb!\n"); - rx_queue->stats.rx_dropped++; - atomic64_inc(&priv->extra_stats.rx_skbmissing); - } + rx_queue->next_to_clean = i; + + /* fetch next buffer if not the last in frame */ + if (!(lstatus & BD_LFLAG(RXBD_LAST))) + continue; + + if (unlikely(lstatus & BD_LFLAG(RXBD_ERR))) { + count_errors(lstatus, ndev); + /* discard faulty buffer */ + dev_kfree_skb(skb); + skb = NULL; + rx_queue->stats.rx_dropped++; + continue; } - rx_queue->rx_skbuff[rx_queue->skb_currx] = newskb; + /* Increment the number of packets */ + total_pkts++; + total_bytes += skb->len; - /* Setup the new bdp */ - gfar_init_rxbdp(rx_queue, bdp, bufaddr); + skb_record_rx_queue(skb, rx_queue->qindex); - /* Update Last Free RxBD pointer for LFC */ - if (unlikely(rx_queue->rfbptr && priv->tx_actual_en)) - gfar_write(rx_queue->rfbptr, (u32)bdp); + gfar_process_frame(ndev, skb); - /* Update to the next pointer */ - bdp = next_bd(bdp, base, rx_queue->rx_ring_size); + /* Send the packet up the stack */ + napi_gro_receive(&rx_queue->grp->napi_rx, skb); - /* update to point at the next skb */ - rx_queue->skb_currx = (rx_queue->skb_currx + 1) & - RX_RING_MOD_MASK(rx_queue->rx_ring_size); + skb = NULL; } - /* Update the current rxbd pointer to be the next one */ - rx_queue->cur_rx = bdp; + /* Store incomplete frames for completion */ + rx_queue->skb = skb; + + rx_queue->stats.rx_packets += total_pkts; + rx_queue->stats.rx_bytes += total_bytes; + + if (cleaned_cnt) + gfar_alloc_rx_buffs(rx_queue, cleaned_cnt); + + /* Update Last Free RxBD pointer for LFC */ + if (unlikely(priv->tx_actual_en)) { + u32 bdp_dma = gfar_rxbd_dma_lastfree(rx_queue); + + gfar_write(rx_queue->rfbptr, bdp_dma); + } return howmany; } @@ -3459,7 +3530,6 @@ static noinline void gfar_update_link_state(struct gfar_private *priv) struct phy_device *phydev = priv->phydev; struct gfar_priv_rx_q *rx_queue = NULL; int i; - struct rxbd8 *bdp; if (unlikely(test_bit(GFAR_RESETTING, &priv->state))) return; @@ -3516,15 +3586,11 @@ static noinline void gfar_update_link_state(struct gfar_private *priv) /* Turn last free buffer recording on */ if ((tempval1 & MACCFG1_TX_FLOW) && !tx_flow_oldval) { for (i = 0; i < priv->num_rx_queues; i++) { + u32 bdp_dma; + rx_queue = priv->rx_queue[i]; - bdp = rx_queue->cur_rx; - /* skip to previous bd */ - bdp = skip_bd(bdp, rx_queue->rx_ring_size - 1, - rx_queue->rx_bd_base, - rx_queue->rx_ring_size); - - if (rx_queue->rfbptr) - gfar_write(rx_queue->rfbptr, (u32)bdp); + bdp_dma = gfar_rxbd_dma_lastfree(rx_queue); + gfar_write(rx_queue->rfbptr, bdp_dma); } priv->tx_actual_en = 1; diff --git a/drivers/net/ethernet/freescale/gianfar.h b/drivers/net/ethernet/freescale/gianfar.h index 5545e41..8c19948 100644 --- a/drivers/net/ethernet/freescale/gianfar.h +++ b/drivers/net/ethernet/freescale/gianfar.h @@ -71,11 +71,6 @@ struct ethtool_rx_list { /* Number of bytes to align the rx bufs to */ #define RXBUF_ALIGNMENT 64 -/* The number of bytes which composes a unit for the purpose of - * allocating data buffers. ie-for any given MTU, the data buffer - * will be the next highest multiple of 512 bytes. */ -#define INCREMENTAL_BUFFER_SIZE 512 - #define PHY_INIT_TIMEOUT 100000 #define DRV_NAME "gfar-enet" @@ -92,6 +87,8 @@ extern const char gfar_driver_version[]; #define DEFAULT_TX_RING_SIZE 256 #define DEFAULT_RX_RING_SIZE 256 +#define GFAR_RX_BUFF_ALLOC 16 + #define GFAR_RX_MAX_RING_SIZE 256 #define GFAR_TX_MAX_RING_SIZE 256 @@ -103,11 +100,14 @@ extern const char gfar_driver_version[]; #define DEFAULT_RX_LFC_THR 16 #define DEFAULT_LFC_PTVVAL 4 -#define DEFAULT_RX_BUFFER_SIZE 1536 +#define GFAR_RXB_SIZE 1536 +#define GFAR_SKBFRAG_SIZE (RXBUF_ALIGNMENT + GFAR_RXB_SIZE \ + + SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) +#define GFAR_RXB_TRUESIZE 2048 + #define TX_RING_MOD_MASK(size) (size-1) #define RX_RING_MOD_MASK(size) (size-1) -#define JUMBO_BUFFER_SIZE 9728 -#define JUMBO_FRAME_SIZE 9600 +#define GFAR_JUMBO_FRAME_SIZE 9600 #define DEFAULT_FIFO_TX_THR 0x100 #define DEFAULT_FIFO_TX_STARVE 0x40 @@ -640,6 +640,7 @@ struct rmon_mib }; struct gfar_extra_stats { + atomic64_t rx_alloc_err; atomic64_t rx_large; atomic64_t rx_short; atomic64_t rx_nonoctet; @@ -651,7 +652,6 @@ struct gfar_extra_stats { atomic64_t eberr; atomic64_t tx_babt; atomic64_t tx_underrun; - atomic64_t rx_skbmissing; atomic64_t tx_timeout; }; @@ -1012,34 +1012,42 @@ struct rx_q_stats { unsigned long rx_dropped; }; +struct gfar_rx_buff { + dma_addr_t dma; + struct page *page; + unsigned int page_offset; +}; + /** * struct gfar_priv_rx_q - per rx queue structure - * @rx_skbuff: skb pointers - * @skb_currx: currently use skb pointer + * @rx_buff: Array of buffer info metadata structs * @rx_bd_base: First rx buffer descriptor - * @cur_rx: Next free rx ring entry + * @next_to_use: index of the next buffer to be alloc'd + * @next_to_clean: index of the next buffer to be cleaned * @qindex: index of this queue - * @dev: back pointer to the dev structure + * @ndev: back pointer to net_device * @rx_ring_size: Rx ring size * @rxcoalescing: enable/disable rx-coalescing * @rxic: receive interrupt coalescing vlaue */ struct gfar_priv_rx_q { - struct sk_buff **rx_skbuff __aligned(SMP_CACHE_BYTES); - dma_addr_t rx_bd_dma_base; + struct gfar_rx_buff *rx_buff __aligned(SMP_CACHE_BYTES); struct rxbd8 *rx_bd_base; - struct rxbd8 *cur_rx; - struct net_device *dev; - struct gfar_priv_grp *grp; + struct net_device *ndev; + struct device *dev; + u16 rx_ring_size; + u16 qindex; + struct gfar_priv_grp *grp; + u16 next_to_clean; + u16 next_to_use; + u16 next_to_alloc; + struct sk_buff *skb; struct rx_q_stats stats; - u16 skb_currx; - u16 qindex; - unsigned int rx_ring_size; - /* RX Coalescing values */ + u32 __iomem *rfbptr; unsigned char rxcoalescing; unsigned long rxic; - u32 __iomem *rfbptr; + dma_addr_t rx_bd_dma_base; }; enum gfar_irqinfo_id { @@ -1109,7 +1117,6 @@ struct gfar_private { struct device *dev; struct net_device *ndev; enum gfar_errata errata; - unsigned int rx_buffer_size; u16 uses_rxfcb; u16 padding; @@ -1292,6 +1299,28 @@ static inline void gfar_clear_txbd_status(struct txbd8 *bdp) bdp->lstatus = cpu_to_be32(lstatus); } +static inline int gfar_rxbd_unused(struct gfar_priv_rx_q *rxq) +{ + if (rxq->next_to_clean > rxq->next_to_use) + return rxq->next_to_clean - rxq->next_to_use - 1; + + return rxq->rx_ring_size + rxq->next_to_clean - rxq->next_to_use - 1; +} + +static inline u32 gfar_rxbd_dma_lastfree(struct gfar_priv_rx_q *rxq) +{ + struct rxbd8 *bdp; + u32 bdp_dma; + int i; + + i = rxq->next_to_use ? rxq->next_to_use - 1 : rxq->rx_ring_size - 1; + bdp = &rxq->rx_bd_base[i]; + bdp_dma = lower_32_bits(rxq->rx_bd_dma_base); + bdp_dma += (uintptr_t)bdp - (uintptr_t)rxq->rx_bd_base; + + return bdp_dma; +} + irqreturn_t gfar_receive(int irq, void *dev_id); int startup_gfar(struct net_device *dev); void stop_gfar(struct net_device *dev); diff --git a/drivers/net/ethernet/freescale/gianfar_ethtool.c b/drivers/net/ethernet/freescale/gianfar_ethtool.c index 5b90fcf..6bdc891 100644 --- a/drivers/net/ethernet/freescale/gianfar_ethtool.c +++ b/drivers/net/ethernet/freescale/gianfar_ethtool.c @@ -61,6 +61,8 @@ static void gfar_gdrvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo); static const char stat_gstrings[][ETH_GSTRING_LEN] = { + /* extra stats */ + "rx-allocation-errors", "rx-large-frame-errors", "rx-short-frame-errors", "rx-non-octet-errors", @@ -72,8 +74,8 @@ static const char stat_gstrings[][ETH_GSTRING_LEN] = { "ethernet-bus-error", "tx-babbling-errors", "tx-underrun-errors", - "rx-skb-missing-errors", "tx-timeout-errors", + /* rmon stats */ "tx-rx-64-frames", "tx-rx-65-127-frames", "tx-rx-128-255-frames", |