Skip to content

L
linux
Commit 9061cb02 authored by David S. Miller's avatar David S. Miller
Browse files
Options

Merge branch 'gianfar_rx_sg' 

Claudiu Manoil says:

====================
gianfar: Add Rx S/G

This patch-set introduces scatter/gather support
on the Rx side, addressing Rx path performance
issues in the driver.
Thanks.

As an example, two boards connected back-to-back
were used to measure the throughput, running the
same kernel 4.1, before and after applying these
patches.
The netperf UDP_STREAM results below show that the
bottleneck lies on the Rx side BEFORE applying the
patches, and that the Rx throughput is even lower
with a larger MTU.  AFTER applying the patches the
Rx bottleneck is gone (Rx throughput matches the
Tx one) and the RX throughput is not influenced by
MTU size any longer (as expected).

BEFORE:

1) MTU 1500 (default)

root@p1010rdb-pb:~# netperf -l 150 -cC -H 192.85.1.1 -p 12867 -t UDP_STREAM -- -m 512
MIGRATED UDP STREAM TEST from 0.0.0.0 () port 0 AF_INET to 192.85.1.1 () port 0 AF_INET
Socket  Message  Elapsed      Messages                   CPU      Service
Size    Size     Time         Okay Errors   Throughput   Util     Demand
bytes   bytes    secs            #      #   10^6bits/sec % SS     us/KB

163840     512   150.00    20119124      0      549.4     100.00   14.911
163840           150.00    14057349             383.9     100.00   14.911

root@p1010rdb-pb:~# netperf -l 150 -cC -H 192.85.1.1 -p 12867 -t UDP_STREAM -- -m 64
MIGRATED UDP STREAM TEST from 0.0.0.0 () port 0 AF_INET to 192.85.1.1 () port 0 AF_INET
Socket  Message  Elapsed      Messages                   CPU      Service
Size    Size     Time         Okay Errors   Throughput   Util     Demand
bytes   bytes    secs            #      #   10^6bits/sec % SS     us/KB

163840      64   150.00    23654013      0       80.7     100.00   101.463
163840           150.00    15875288              54.2     100.00   101.463

2) MTU 8000

root@p1010rdb-pb:~# netperf -l 150 -cC -H 192.85.1.1 -p 12867 -t UDP_STREAM -- -m 512
MIGRATED UDP STREAM TEST from 0.0.0.0 () port 0 AF_INET to 192.85.1.1 () port 0 AF_INET
Socket  Message  Elapsed      Messages                   CPU      Service
Size    Size     Time         Okay Errors   Throughput   Util     Demand
bytes   bytes    secs            #      #   10^6bits/sec % SS     us/KB

163840     512   150.00    20067232      0      548.0     100.00   14.950
163840           150.00    6113498             166.9     99.95    14.942

root@p1010rdb-pb:~# netperf -l 150 -cC -H 192.85.1.1 -p 12867 -t UDP_STREAM -- -m 64
MIGRATED UDP STREAM TEST from 0.0.0.0 () port 0 AF_INET to 192.85.1.1 () port 0 AF_INET
Socket  Message  Elapsed      Messages                   CPU      Service
Size    Size     Time         Okay Errors   Throughput   Util     Demand
bytes   bytes    secs            #      #   10^6bits/sec % SS     us/KB

163840      64   150.00    23621279      0       80.6     100.00   101.604
163840           150.00    5868602              20.0     99.96    101.563

AFTER:
(both MTU 1500 and MTU 8000)

root@p1010rdb-pb:~# netperf -l 150 -cC -H 192.85.1.1 -p 12867 -t UDP_STREAM -- -m 512
MIGRATED UDP STREAM TEST from 0.0.0.0 () port 0 AF_INET to 192.85.1.1 () port 0 AF_INET
Socket  Message  Elapsed      Messages                   CPU      Service
Size    Size     Time         Okay Errors   Throughput   Util     Demand
bytes   bytes    secs            #      #   10^6bits/sec % SS     us/KB

163840     512   150.00    19914969      0      543.8     100.00   15.064
163840           150.00    19914969             543.8     99.35    14.966

root@p1010rdb-pb:~# netperf -l 150 -cC -H 192.85.1.1 -p 12867 -t UDP_STREAM -- -m 64
MIGRATED UDP STREAM TEST from 0.0.0.0 () port 0 AF_INET to 192.85.1.1 () port 0 AF_INET
Socket  Message  Elapsed      Messages                   CPU      Service
Size    Size     Time         Okay Errors   Throughput   Util     Demand
bytes   bytes    secs            #      #   10^6bits/sec % SS     us/KB

163840      64   150.00    23433989      0       80.0     100.00   102.416
163840           150.00    23433989              80.0     99.62    102.023
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parents aad0d51e 75354148
Show whitespace changes
Inline Side-by-side
Showing with 330 additions and 241 deletions
drivers/net/ethernet/freescale/gianfar.c
View file @ 9061cb02
......@@ -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 = &regs->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),
rx_queue->rx_buff = kcalloc(rx_queue->rx_ring_size,
sizeof(*rx_queue->rx_buff),
GFP_KERNEL);
if (!rx_queue->rx_skbuff)
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)
......@@ -609,9 +576,8 @@ 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;
}
......@@ -1203,12 +1169,11 @@ 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(&regs->maxfrm, priv->rx_buffer_size);
gfar_write(&regs->mrblr, priv->rx_buffer_size);
gfar_write(&regs->maxfrm, GFAR_JUMBO_FRAME_SIZE);
gfar_write(&regs->mrblr, GFAR_RXB_SIZE);
/* Initialize the Minimum Frame Length Register */
gfar_write(&regs->minflr, MINFLR_INIT_SETTINGS);
......@@ -1216,12 +1181,11 @@ 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(&regs->maccfg2, tempval);
......@@ -1432,8 +1396,6 @@ 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;
......@@ -1639,10 +1601,7 @@ 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);
......@@ -1933,26 +1892,32 @@ 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.
......@@ -1977,7 +1942,7 @@ 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);
}
......@@ -2535,7 +2500,7 @@ 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;
}
......@@ -2589,15 +2554,6 @@ 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)
{
......@@ -2704,49 +2660,85 @@ 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;
}
}
*bufaddr = addr;
return skb;
/* 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;
}
}
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);
......@@ -2754,25 +2746,25 @@ 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++;
}
}
......@@ -2823,6 +2815,93 @@ 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
......@@ -2837,10 +2916,9 @@ 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 */
......@@ -2849,10 +2927,8 @@ 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) {
......@@ -2866,24 +2942,20 @@ 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
......@@ -2892,92 +2964,89 @@ 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)
if (unlikely(++i == rx_queue->rx_ring_size))
i = 0;
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);
} else {
skb = NULL;
rx_queue->stats.rx_dropped++;
continue;
}
/* Increment the number of packets */
rx_queue->stats.rx_packets++;
howmany++;
total_pkts++;
total_bytes += skb->len;
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);
} else {
netif_warn(priv, rx_err, dev, "Missing skb!\n");
rx_queue->stats.rx_dropped++;
atomic64_inc(&priv->extra_stats.rx_skbmissing);
}
gfar_process_frame(ndev, skb);
/* Send the packet up the stack */
napi_gro_receive(&rx_queue->grp->napi_rx, skb);
skb = NULL;
}
rx_queue->rx_skbuff[rx_queue->skb_currx] = newskb;
/* Store incomplete frames for completion */
rx_queue->skb = skb;
/* Setup the new bdp */
gfar_init_rxbdp(rx_queue, bdp, bufaddr);
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(rx_queue->rfbptr && priv->tx_actual_en))
if (unlikely(priv->tx_actual_en)) {
bdp = gfar_rxbd_lastfree(rx_queue);
gfar_write(rx_queue->rfbptr, (u32)bdp);
/* Update to the next pointer */
bdp = next_bd(bdp, base, rx_queue->rx_ring_size);
/* 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);
}
/* Update the current rxbd pointer to be the next one */
rx_queue->cur_rx = bdp;
return howmany;
}
......@@ -3552,13 +3621,7 @@ static noinline void gfar_update_link_state(struct gfar_private *priv)
if ((tempval1 & MACCFG1_TX_FLOW) && !tx_flow_oldval) {
for (i = 0; i < priv->num_rx_queues; i++) {
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)
bdp = gfar_rxbd_lastfree(rx_queue);
gfar_write(rx_queue->rfbptr, (u32)bdp);
}
......
drivers/net/ethernet/freescale/gianfar.h
View file @ 9061cb02
......@@ -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 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;
......@@ -1295,6 +1302,23 @@ 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 struct rxbd8 *gfar_rxbd_lastfree(struct gfar_priv_rx_q *rxq)
{
int i;
i = rxq->next_to_use ? rxq->next_to_use - 1 : rxq->rx_ring_size - 1;
return &rxq->rx_bd_base[i];
}
irqreturn_t gfar_receive(int irq, void *dev_id);
int startup_gfar(struct net_device *dev);
void stop_gfar(struct net_device *dev);
......
drivers/net/ethernet/freescale/gianfar_ethtool.c
View file @ 9061cb02
......@@ -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",
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7