X-Git-Url: http://pilppa.org/gitweb/?a=blobdiff_plain;f=drivers%2Fnet%2Fpasemi_mac.c;h=ef1ebb4b8af27af194bb1ba06d3f9e037fcae81f;hb=7ddeae2c6ceed7f786344731dda27d4277957780;hp=b2861e0df86c3631b81446c873c4853dacae5ebb;hpb=18eec695427ce1258fb5dad0ac180fa4d6f64af7;p=linux-2.6-omap-h63xx.git diff --git a/drivers/net/pasemi_mac.c b/drivers/net/pasemi_mac.c index b2861e0df86..ef1ebb4b8af 100644 --- a/drivers/net/pasemi_mac.c +++ b/drivers/net/pasemi_mac.c @@ -37,21 +37,25 @@ #include "pasemi_mac.h" +/* We have our own align, since ppc64 in general has it at 0 because + * of design flaws in some of the server bridge chips. However, for + * PWRficient doing the unaligned copies is more expensive than doing + * unaligned DMA, so make sure the data is aligned instead. + */ +#define LOCAL_SKB_ALIGN 2 /* TODO list * - * - Get rid of pci_{read,write}_config(), map registers with ioremap - * for performance - * - PHY support * - Multicast support * - Large MTU support - * - Other performance improvements + * - SW LRO + * - Multiqueue RX/TX */ /* Must be a power of two */ -#define RX_RING_SIZE 512 -#define TX_RING_SIZE 512 +#define RX_RING_SIZE 4096 +#define TX_RING_SIZE 4096 #define DEFAULT_MSG_ENABLE \ (NETIF_MSG_DRV | \ @@ -63,10 +67,10 @@ NETIF_MSG_RX_ERR | \ NETIF_MSG_TX_ERR) -#define TX_DESC(mac, num) ((mac)->tx->desc[(num) & (TX_RING_SIZE-1)]) -#define TX_DESC_INFO(mac, num) ((mac)->tx->desc_info[(num) & (TX_RING_SIZE-1)]) -#define RX_DESC(mac, num) ((mac)->rx->desc[(num) & (RX_RING_SIZE-1)]) -#define RX_DESC_INFO(mac, num) ((mac)->rx->desc_info[(num) & (RX_RING_SIZE-1)]) +#define TX_RING(mac, num) ((mac)->tx->ring[(num) & (TX_RING_SIZE-1)]) +#define TX_RING_INFO(mac, num) ((mac)->tx->ring_info[(num) & (TX_RING_SIZE-1)]) +#define RX_RING(mac, num) ((mac)->rx->ring[(num) & (RX_RING_SIZE-1)]) +#define RX_RING_INFO(mac, num) ((mac)->rx->ring_info[(num) & (RX_RING_SIZE-1)]) #define RX_BUFF(mac, num) ((mac)->rx->buffers[(num) & (RX_RING_SIZE-1)]) #define RING_USED(ring) (((ring)->next_to_fill - (ring)->next_to_clean) \ @@ -160,6 +164,30 @@ static int pasemi_get_mac_addr(struct pasemi_mac *mac) return 0; } +static int pasemi_mac_unmap_tx_skb(struct pasemi_mac *mac, + struct sk_buff *skb, + dma_addr_t *dmas) +{ + int f; + int nfrags = skb_shinfo(skb)->nr_frags; + + pci_unmap_single(mac->dma_pdev, dmas[0], skb_headlen(skb), + PCI_DMA_TODEVICE); + + for (f = 0; f < nfrags; f++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[f]; + + pci_unmap_page(mac->dma_pdev, dmas[f+1], frag->size, + PCI_DMA_TODEVICE); + } + dev_kfree_skb_irq(skb); + + /* Freed descriptor slot + main SKB ptr + nfrags additional ptrs, + * aligned up to a power of 2 + */ + return (nfrags + 3) & ~1; +} + static int pasemi_mac_setup_rx_resources(struct net_device *dev) { struct pasemi_mac_rxring *ring; @@ -174,22 +202,21 @@ static int pasemi_mac_setup_rx_resources(struct net_device *dev) spin_lock_init(&ring->lock); ring->size = RX_RING_SIZE; - ring->desc_info = kzalloc(sizeof(struct pasemi_mac_buffer) * + ring->ring_info = kzalloc(sizeof(struct pasemi_mac_buffer) * RX_RING_SIZE, GFP_KERNEL); - if (!ring->desc_info) - goto out_desc_info; + if (!ring->ring_info) + goto out_ring_info; /* Allocate descriptors */ - ring->desc = dma_alloc_coherent(&mac->dma_pdev->dev, - RX_RING_SIZE * - sizeof(struct pas_dma_xct_descr), + ring->ring = dma_alloc_coherent(&mac->dma_pdev->dev, + RX_RING_SIZE * sizeof(u64), &ring->dma, GFP_KERNEL); - if (!ring->desc) - goto out_desc; + if (!ring->ring) + goto out_ring_desc; - memset(ring->desc, 0, RX_RING_SIZE * sizeof(struct pas_dma_xct_descr)); + memset(ring->ring, 0, RX_RING_SIZE * sizeof(u64)); ring->buffers = dma_alloc_coherent(&mac->dma_pdev->dev, RX_RING_SIZE * sizeof(u64), @@ -203,7 +230,7 @@ static int pasemi_mac_setup_rx_resources(struct net_device *dev) write_dma_reg(mac, PAS_DMA_RXCHAN_BASEU(chan_id), PAS_DMA_RXCHAN_BASEU_BRBH(ring->dma >> 32) | - PAS_DMA_RXCHAN_BASEU_SIZ(RX_RING_SIZE >> 2)); + PAS_DMA_RXCHAN_BASEU_SIZ(RX_RING_SIZE >> 3)); write_dma_reg(mac, PAS_DMA_RXCHAN_CFG(chan_id), PAS_DMA_RXCHAN_CFG_HBU(2)); @@ -216,7 +243,9 @@ static int pasemi_mac_setup_rx_resources(struct net_device *dev) PAS_DMA_RXINT_BASEU_SIZ(RX_RING_SIZE >> 3)); write_dma_reg(mac, PAS_DMA_RXINT_CFG(mac->dma_if), - PAS_DMA_RXINT_CFG_DHL(2)); + PAS_DMA_RXINT_CFG_DHL(3) | + PAS_DMA_RXINT_CFG_L2 | + PAS_DMA_RXINT_CFG_LW); ring->next_to_fill = 0; ring->next_to_clean = 0; @@ -229,11 +258,11 @@ static int pasemi_mac_setup_rx_resources(struct net_device *dev) out_buffers: dma_free_coherent(&mac->dma_pdev->dev, - RX_RING_SIZE * sizeof(struct pas_dma_xct_descr), - mac->rx->desc, mac->rx->dma); -out_desc: - kfree(ring->desc_info); -out_desc_info: + RX_RING_SIZE * sizeof(u64), + mac->rx->ring, mac->rx->dma); +out_ring_desc: + kfree(ring->ring_info); +out_ring_info: kfree(ring); out_ring: return -ENOMEM; @@ -254,25 +283,24 @@ static int pasemi_mac_setup_tx_resources(struct net_device *dev) spin_lock_init(&ring->lock); ring->size = TX_RING_SIZE; - ring->desc_info = kzalloc(sizeof(struct pasemi_mac_buffer) * + ring->ring_info = kzalloc(sizeof(struct pasemi_mac_buffer) * TX_RING_SIZE, GFP_KERNEL); - if (!ring->desc_info) - goto out_desc_info; + if (!ring->ring_info) + goto out_ring_info; /* Allocate descriptors */ - ring->desc = dma_alloc_coherent(&mac->dma_pdev->dev, - TX_RING_SIZE * - sizeof(struct pas_dma_xct_descr), + ring->ring = dma_alloc_coherent(&mac->dma_pdev->dev, + TX_RING_SIZE * sizeof(u64), &ring->dma, GFP_KERNEL); - if (!ring->desc) - goto out_desc; + if (!ring->ring) + goto out_ring_desc; - memset(ring->desc, 0, TX_RING_SIZE * sizeof(struct pas_dma_xct_descr)); + memset(ring->ring, 0, TX_RING_SIZE * sizeof(u64)); write_dma_reg(mac, PAS_DMA_TXCHAN_BASEL(chan_id), PAS_DMA_TXCHAN_BASEL_BRBL(ring->dma)); val = PAS_DMA_TXCHAN_BASEU_BRBH(ring->dma >> 32); - val |= PAS_DMA_TXCHAN_BASEU_SIZ(TX_RING_SIZE >> 2); + val |= PAS_DMA_TXCHAN_BASEU_SIZ(TX_RING_SIZE >> 3); write_dma_reg(mac, PAS_DMA_TXCHAN_BASEU(chan_id), val); @@ -291,9 +319,9 @@ static int pasemi_mac_setup_tx_resources(struct net_device *dev) return 0; -out_desc: - kfree(ring->desc_info); -out_desc_info: +out_ring_desc: + kfree(ring->ring_info); +out_ring_info: kfree(ring); out_ring: return -ENOMEM; @@ -302,33 +330,37 @@ out_ring: static void pasemi_mac_free_tx_resources(struct net_device *dev) { struct pasemi_mac *mac = netdev_priv(dev); - unsigned int i; + unsigned int i, j; struct pasemi_mac_buffer *info; - struct pas_dma_xct_descr *dp; - - for (i = 0; i < TX_RING_SIZE; i++) { - info = &TX_DESC_INFO(mac, i); - dp = &TX_DESC(mac, i); - if (info->dma) { - if (info->skb) { - pci_unmap_single(mac->dma_pdev, - info->dma, - info->skb->len, - PCI_DMA_TODEVICE); - dev_kfree_skb_any(info->skb); - } - info->dma = 0; - info->skb = NULL; - dp->mactx = 0; - dp->ptr = 0; - } + dma_addr_t dmas[MAX_SKB_FRAGS+1]; + int freed; + int start, limit; + + start = mac->tx->next_to_clean; + limit = mac->tx->next_to_fill; + + /* Compensate for when fill has wrapped and clean has not */ + if (start > limit) + limit += TX_RING_SIZE; + + for (i = start; i < limit; i += freed) { + info = &TX_RING_INFO(mac, i+1); + if (info->dma && info->skb) { + for (j = 0; j <= skb_shinfo(info->skb)->nr_frags; j++) + dmas[j] = TX_RING_INFO(mac, i+1+j).dma; + freed = pasemi_mac_unmap_tx_skb(mac, info->skb, dmas); + } else + freed = 2; } + for (i = 0; i < TX_RING_SIZE; i++) + TX_RING(mac, i) = 0; + dma_free_coherent(&mac->dma_pdev->dev, - TX_RING_SIZE * sizeof(struct pas_dma_xct_descr), - mac->tx->desc, mac->tx->dma); + TX_RING_SIZE * sizeof(u64), + mac->tx->ring, mac->tx->dma); - kfree(mac->tx->desc_info); + kfree(mac->tx->ring_info); kfree(mac->tx); mac->tx = NULL; } @@ -338,34 +370,31 @@ static void pasemi_mac_free_rx_resources(struct net_device *dev) struct pasemi_mac *mac = netdev_priv(dev); unsigned int i; struct pasemi_mac_buffer *info; - struct pas_dma_xct_descr *dp; for (i = 0; i < RX_RING_SIZE; i++) { - info = &RX_DESC_INFO(mac, i); - dp = &RX_DESC(mac, i); - if (info->skb) { - if (info->dma) { - pci_unmap_single(mac->dma_pdev, - info->dma, - info->skb->len, - PCI_DMA_FROMDEVICE); - dev_kfree_skb_any(info->skb); - } - info->dma = 0; - info->skb = NULL; - dp->macrx = 0; - dp->ptr = 0; + info = &RX_RING_INFO(mac, i); + if (info->skb && info->dma) { + pci_unmap_single(mac->dma_pdev, + info->dma, + info->skb->len, + PCI_DMA_FROMDEVICE); + dev_kfree_skb_any(info->skb); } + info->dma = 0; + info->skb = NULL; } + for (i = 0; i < RX_RING_SIZE; i++) + RX_RING(mac, i) = 0; + dma_free_coherent(&mac->dma_pdev->dev, - RX_RING_SIZE * sizeof(struct pas_dma_xct_descr), - mac->rx->desc, mac->rx->dma); + RX_RING_SIZE * sizeof(u64), + mac->rx->ring, mac->rx->dma); dma_free_coherent(&mac->dma_pdev->dev, RX_RING_SIZE * sizeof(u64), mac->rx->buffers, mac->rx->buf_dma); - kfree(mac->rx->desc_info); + kfree(mac->rx->ring_info); kfree(mac->rx); mac->rx = NULL; } @@ -373,30 +402,35 @@ static void pasemi_mac_free_rx_resources(struct net_device *dev) static void pasemi_mac_replenish_rx_ring(struct net_device *dev, int limit) { struct pasemi_mac *mac = netdev_priv(dev); - unsigned int i; int start = mac->rx->next_to_fill; - int count; + unsigned int fill, count; if (limit <= 0) return; - i = start; + fill = start; for (count = 0; count < limit; count++) { - struct pasemi_mac_buffer *info = &RX_DESC_INFO(mac, i); - u64 *buff = &RX_BUFF(mac, i); + struct pasemi_mac_buffer *info = &RX_RING_INFO(mac, fill); + u64 *buff = &RX_BUFF(mac, fill); struct sk_buff *skb; dma_addr_t dma; + /* Entry in use? */ + WARN_ON(*buff); + /* skb might still be in there for recycle on short receives */ if (info->skb) skb = info->skb; - else + else { skb = dev_alloc_skb(BUF_SIZE); + skb_reserve(skb, LOCAL_SKB_ALIGN); + } if (unlikely(!skb)) break; - dma = pci_map_single(mac->dma_pdev, skb->data, BUF_SIZE, + dma = pci_map_single(mac->dma_pdev, skb->data, + BUF_SIZE - LOCAL_SKB_ALIGN, PCI_DMA_FROMDEVICE); if (unlikely(dma_mapping_error(dma))) { @@ -407,7 +441,7 @@ static void pasemi_mac_replenish_rx_ring(struct net_device *dev, int limit) info->skb = skb; info->dma = dma; *buff = XCT_RXB_LEN(BUF_SIZE) | XCT_RXB_ADDR(dma); - i++; + fill++; } wmb(); @@ -481,7 +515,6 @@ static int pasemi_mac_clean_rx(struct pasemi_mac *mac, int limit) { unsigned int n; int count; - struct pas_dma_xct_descr *dp; struct pasemi_mac_buffer *info; struct sk_buff *skb; unsigned int i, len; @@ -493,12 +526,7 @@ static int pasemi_mac_clean_rx(struct pasemi_mac *mac, int limit) n = mac->rx->next_to_clean; for (count = limit; count; count--) { - - rmb(); - - dp = &RX_DESC(mac, n); - prefetchw(dp); - macrx = dp->macrx; + macrx = RX_RING(mac, n); if ((macrx & XCT_MACRX_E) || (*mac->rx_status & PAS_STATUS_ERROR)) @@ -516,28 +544,29 @@ static int pasemi_mac_clean_rx(struct pasemi_mac *mac, int limit) * interface ring. */ - dma = (dp->ptr & XCT_PTR_ADDR_M); - for (i = n; i < (n + RX_RING_SIZE); i++) { - info = &RX_DESC_INFO(mac, i); + dma = (RX_RING(mac, n+1) & XCT_PTR_ADDR_M); + for (i = mac->rx->next_to_fill; + i < (mac->rx->next_to_fill + RX_RING_SIZE); + i++) { + info = &RX_RING_INFO(mac, i); if (info->dma == dma) break; } - prefetchw(info); skb = info->skb; - prefetchw(skb); - info->dma = 0; - pci_unmap_single(mac->dma_pdev, dma, skb->len, - PCI_DMA_FROMDEVICE); + prefetch(skb); + prefetch(&skb->data_len); len = (macrx & XCT_MACRX_LLEN_M) >> XCT_MACRX_LLEN_S; if (len < 256) { - struct sk_buff *new_skb = - netdev_alloc_skb(mac->netdev, len + NET_IP_ALIGN); + struct sk_buff *new_skb; + + new_skb = netdev_alloc_skb(mac->netdev, + len + LOCAL_SKB_ALIGN); if (new_skb) { - skb_reserve(new_skb, NET_IP_ALIGN); + skb_reserve(new_skb, LOCAL_SKB_ALIGN); memcpy(new_skb->data, skb->data, len); /* save the skb in buffer_info as good */ skb = new_skb; @@ -546,6 +575,10 @@ static int pasemi_mac_clean_rx(struct pasemi_mac *mac, int limit) } else info->skb = NULL; + pci_unmap_single(mac->dma_pdev, dma, len, PCI_DMA_FROMDEVICE); + + info->dma = 0; + skb_put(skb, len); if (likely((macrx & XCT_MACRX_HTY_M) == XCT_MACRX_HTY_IPV4_OK)) { @@ -561,13 +594,23 @@ static int pasemi_mac_clean_rx(struct pasemi_mac *mac, int limit) skb->protocol = eth_type_trans(skb, mac->netdev); netif_receive_skb(skb); - dp->ptr = 0; - dp->macrx = 0; + RX_RING(mac, n) = 0; + RX_RING(mac, n+1) = 0; - n++; + /* Need to zero it out since hardware doesn't, since the + * replenish loop uses it to tell when it's done. + */ + RX_BUFF(mac, i) = 0; + + n += 2; } - mac->rx->next_to_clean += limit - count; + if (n > RX_RING_SIZE) { + /* Errata 5971 workaround: L2 target of headers */ + write_iob_reg(mac, PAS_IOB_COM_PKTHDRCNT, 0); + n &= (RX_RING_SIZE-1); + } + mac->rx->next_to_clean = n; pasemi_mac_replenish_rx_ring(mac->netdev, limit-count); spin_unlock(&mac->rx->lock); @@ -575,62 +618,77 @@ static int pasemi_mac_clean_rx(struct pasemi_mac *mac, int limit) return count; } +/* Can't make this too large or we blow the kernel stack limits */ +#define TX_CLEAN_BATCHSIZE (128/MAX_SKB_FRAGS) + static int pasemi_mac_clean_tx(struct pasemi_mac *mac) { - int i; - struct pasemi_mac_buffer *info; - struct pas_dma_xct_descr *dp; - unsigned int start, count, limit; + int i, j; + unsigned int start, descr_count, buf_count, batch_limit; + unsigned int ring_limit; unsigned int total_count; unsigned long flags; - struct sk_buff *skbs[32]; - dma_addr_t dmas[32]; + struct sk_buff *skbs[TX_CLEAN_BATCHSIZE]; + dma_addr_t dmas[TX_CLEAN_BATCHSIZE][MAX_SKB_FRAGS+1]; total_count = 0; + batch_limit = TX_CLEAN_BATCHSIZE; restart: spin_lock_irqsave(&mac->tx->lock, flags); start = mac->tx->next_to_clean; - limit = min(mac->tx->next_to_fill, start+32); + ring_limit = mac->tx->next_to_fill; + + /* Compensate for when fill has wrapped but clean has not */ + if (start > ring_limit) + ring_limit += TX_RING_SIZE; - count = 0; + buf_count = 0; + descr_count = 0; - for (i = start; i < limit; i++) { - dp = &TX_DESC(mac, i); + for (i = start; + descr_count < batch_limit && i < ring_limit; + i += buf_count) { + u64 mactx = TX_RING(mac, i); + struct sk_buff *skb; - if ((dp->mactx & XCT_MACTX_E) || + if ((mactx & XCT_MACTX_E) || (*mac->tx_status & PAS_STATUS_ERROR)) - pasemi_mac_tx_error(mac, dp->mactx); + pasemi_mac_tx_error(mac, mactx); - if (unlikely(dp->mactx & XCT_MACTX_O)) + if (unlikely(mactx & XCT_MACTX_O)) /* Not yet transmitted */ break; - info = &TX_DESC_INFO(mac, i); - skbs[count] = info->skb; - dmas[count] = info->dma; + skb = TX_RING_INFO(mac, i+1).skb; + skbs[descr_count] = skb; - info->skb = NULL; - info->dma = 0; - dp->mactx = 0; - dp->ptr = 0; + buf_count = 2 + skb_shinfo(skb)->nr_frags; + for (j = 0; j <= skb_shinfo(skb)->nr_frags; j++) + dmas[descr_count][j] = TX_RING_INFO(mac, i+1+j).dma; + + TX_RING(mac, i) = 0; + TX_RING(mac, i+1) = 0; - count++; + /* Since we always fill with an even number of entries, make + * sure we skip any unused one at the end as well. + */ + if (buf_count & 1) + buf_count++; + descr_count++; } - mac->tx->next_to_clean += count; + mac->tx->next_to_clean = i & (TX_RING_SIZE-1); + spin_unlock_irqrestore(&mac->tx->lock, flags); netif_wake_queue(mac->netdev); - for (i = 0; i < count; i++) { - pci_unmap_single(mac->dma_pdev, dmas[i], - skbs[i]->len, PCI_DMA_TODEVICE); - dev_kfree_skb_irq(skbs[i]); - } + for (i = 0; i < descr_count; i++) + pasemi_mac_unmap_tx_skb(mac, skbs[i], dmas[i]); - total_count += count; + total_count += descr_count; /* If the batch was full, try to clean more */ - if (count == 32) + if (descr_count == batch_limit) goto restart; return total_count; @@ -1001,11 +1059,11 @@ static int pasemi_mac_start_tx(struct sk_buff *skb, struct net_device *dev) { struct pasemi_mac *mac = netdev_priv(dev); struct pasemi_mac_txring *txring; - struct pasemi_mac_buffer *info; - struct pas_dma_xct_descr *dp; - u64 dflags, mactx, ptr; - dma_addr_t map; + u64 dflags, mactx; + dma_addr_t map[MAX_SKB_FRAGS+1]; + unsigned int map_size[MAX_SKB_FRAGS+1]; unsigned long flags; + int i, nfrags; dflags = XCT_MACTX_O | XCT_MACTX_ST | XCT_MACTX_SS | XCT_MACTX_CRC_PAD; @@ -1026,54 +1084,78 @@ static int pasemi_mac_start_tx(struct sk_buff *skb, struct net_device *dev) } } - map = pci_map_single(mac->dma_pdev, skb->data, skb->len, PCI_DMA_TODEVICE); + nfrags = skb_shinfo(skb)->nr_frags; + + map[0] = pci_map_single(mac->dma_pdev, skb->data, skb_headlen(skb), + PCI_DMA_TODEVICE); + map_size[0] = skb_headlen(skb); + if (dma_mapping_error(map[0])) + goto out_err_nolock; - if (dma_mapping_error(map)) - return NETDEV_TX_BUSY; + for (i = 0; i < nfrags; i++) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; + + map[i+1] = pci_map_page(mac->dma_pdev, frag->page, + frag->page_offset, frag->size, + PCI_DMA_TODEVICE); + map_size[i+1] = frag->size; + if (dma_mapping_error(map[i+1])) { + nfrags = i; + goto out_err_nolock; + } + } mactx = dflags | XCT_MACTX_LLEN(skb->len); - ptr = XCT_PTR_LEN(skb->len) | XCT_PTR_ADDR(map); txring = mac->tx; spin_lock_irqsave(&txring->lock, flags); - if (RING_AVAIL(txring) <= 1) { - spin_unlock_irqrestore(&txring->lock, flags); - pasemi_mac_clean_tx(mac); - pasemi_mac_restart_tx_intr(mac); - spin_lock_irqsave(&txring->lock, flags); - - if (RING_AVAIL(txring) <= 1) { - /* Still no room -- stop the queue and wait for tx - * intr when there's room. - */ - netif_stop_queue(dev); - goto out_err; - } + /* Avoid stepping on the same cache line that the DMA controller + * is currently about to send, so leave at least 8 words available. + * Total free space needed is mactx + fragments + 8 + */ + if (RING_AVAIL(txring) < nfrags + 10) { + /* no room -- stop the queue and wait for tx intr */ + netif_stop_queue(dev); + goto out_err; } - dp = &TX_DESC(mac, txring->next_to_fill); - info = &TX_DESC_INFO(mac, txring->next_to_fill); + TX_RING(mac, txring->next_to_fill) = mactx; + txring->next_to_fill++; + TX_RING_INFO(mac, txring->next_to_fill).skb = skb; + for (i = 0; i <= nfrags; i++) { + TX_RING(mac, txring->next_to_fill+i) = + XCT_PTR_LEN(map_size[i]) | XCT_PTR_ADDR(map[i]); + TX_RING_INFO(mac, txring->next_to_fill+i).dma = map[i]; + } - dp->mactx = mactx; - dp->ptr = ptr; - info->dma = map; - info->skb = skb; + /* We have to add an even number of 8-byte entries to the ring + * even if the last one is unused. That means always an odd number + * of pointers + one mactx descriptor. + */ + if (nfrags & 1) + nfrags++; + + txring->next_to_fill = (txring->next_to_fill + nfrags + 1) & + (TX_RING_SIZE-1); - txring->next_to_fill++; dev->stats.tx_packets++; dev->stats.tx_bytes += skb->len; spin_unlock_irqrestore(&txring->lock, flags); - write_dma_reg(mac, PAS_DMA_TXCHAN_INCR(mac->dma_txch), 1); + write_dma_reg(mac, PAS_DMA_TXCHAN_INCR(mac->dma_txch), (nfrags+2) >> 1); return NETDEV_TX_OK; out_err: spin_unlock_irqrestore(&txring->lock, flags); - pci_unmap_single(mac->dma_pdev, map, skb->len, PCI_DMA_TODEVICE); +out_err_nolock: + while (nfrags--) + pci_unmap_single(mac->dma_pdev, map[nfrags], map_size[nfrags], + PCI_DMA_TODEVICE); + return NETDEV_TX_BUSY; } @@ -1209,7 +1291,7 @@ pasemi_mac_probe(struct pci_dev *pdev, const struct pci_device_id *ent) netif_napi_add(dev, &mac->napi, pasemi_mac_poll, 64); - dev->features = NETIF_F_HW_CSUM | NETIF_F_LLTX; + dev->features = NETIF_F_HW_CSUM | NETIF_F_LLTX | NETIF_F_SG; /* These should come out of the device tree eventually */ mac->dma_txch = index;