From: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
To: "David S. Miller" <davem@davemloft.net>,
Jakub Kicinski <kuba@kernel.org>,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH net-next 5/7] net: korina: Use DMA API
Date: Tue, 13 Apr 2021 22:48:16 +0200 [thread overview]
Message-ID: <20210413204818.23350-6-tsbogend@alpha.franken.de> (raw)
In-Reply-To: <20210413204818.23350-1-tsbogend@alpha.franken.de>
Instead of messing with MIPS specific macros use DMA API for mapping
descriptors and skbs.
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
---
drivers/net/ethernet/korina.c | 143 ++++++++++++++++++++--------------
1 file changed, 85 insertions(+), 58 deletions(-)
diff --git a/drivers/net/ethernet/korina.c b/drivers/net/ethernet/korina.c
index 728c4baa3eee..0224cd179b03 100644
--- a/drivers/net/ethernet/korina.c
+++ b/drivers/net/ethernet/korina.c
@@ -109,6 +109,8 @@ struct korina_private {
struct dma_reg __iomem *tx_dma_regs;
struct dma_desc *td_ring; /* transmit descriptor ring */
struct dma_desc *rd_ring; /* receive descriptor ring */
+ dma_addr_t td_dma;
+ dma_addr_t rd_dma;
struct sk_buff *tx_skb[KORINA_NUM_TDS];
struct sk_buff *rx_skb[KORINA_NUM_RDS];
@@ -137,10 +139,21 @@ struct korina_private {
struct mii_if_info mii_if;
struct work_struct restart_task;
struct net_device *dev;
+ struct device *dmadev;
};
extern unsigned int idt_cpu_freq;
+static dma_addr_t korina_tx_dma(struct korina_private *lp, int idx)
+{
+ return lp->td_dma + (idx * sizeof(struct dma_desc));
+}
+
+static dma_addr_t korina_rx_dma(struct korina_private *lp, int idx)
+{
+ return lp->rd_dma + (idx * sizeof(struct dma_desc));
+}
+
static inline void korina_abort_dma(struct net_device *dev,
struct dma_reg *ch)
{
@@ -179,10 +192,12 @@ static int korina_send_packet(struct sk_buff *skb, struct net_device *dev)
u32 length;
u32 chain_prev, chain_next;
struct dma_desc *td;
+ int idx;
spin_lock_irqsave(&lp->lock, flags);
- td = &lp->td_ring[lp->tx_chain_tail];
+ idx = lp->tx_chain_tail;
+ td = &lp->td_ring[idx];
/* stop queue when full, drop pkts if queue already full */
if (lp->tx_count >= (KORINA_NUM_TDS - 2)) {
@@ -190,26 +205,23 @@ static int korina_send_packet(struct sk_buff *skb, struct net_device *dev)
if (lp->tx_count == (KORINA_NUM_TDS - 2))
netif_stop_queue(dev);
- else {
- dev->stats.tx_dropped++;
- dev_kfree_skb_any(skb);
- spin_unlock_irqrestore(&lp->lock, flags);
-
- return NETDEV_TX_OK;
- }
+ else
+ goto drop_packet;
}
lp->tx_count++;
- lp->tx_skb[lp->tx_chain_tail] = skb;
+ lp->tx_skb[idx] = skb;
length = skb->len;
- dma_cache_wback((u32)skb->data, skb->len);
/* Setup the transmit descriptor. */
- td->ca = CPHYSADDR(skb->data);
- chain_prev = (lp->tx_chain_tail - 1) & KORINA_TDS_MASK;
- chain_next = (lp->tx_chain_tail + 1) & KORINA_TDS_MASK;
+ td->ca = dma_map_single(lp->dmadev, skb->data, length, DMA_TO_DEVICE);
+ if (dma_mapping_error(lp->dmadev, td->ca))
+ goto drop_packet;
+
+ chain_prev = (idx - 1) & KORINA_TDS_MASK;
+ chain_next = (idx + 1) & KORINA_TDS_MASK;
if (readl(&(lp->tx_dma_regs->dmandptr)) == 0) {
if (lp->tx_chain_status == desc_empty) {
@@ -219,8 +231,8 @@ static int korina_send_packet(struct sk_buff *skb, struct net_device *dev)
/* Move tail */
lp->tx_chain_tail = chain_next;
/* Write to NDPTR */
- writel(CPHYSADDR(&lp->td_ring[lp->tx_chain_head]),
- &lp->tx_dma_regs->dmandptr);
+ writel(korina_tx_dma(lp, lp->tx_chain_head),
+ &lp->tx_dma_regs->dmandptr);
/* Move head to tail */
lp->tx_chain_head = lp->tx_chain_tail;
} else {
@@ -231,12 +243,12 @@ static int korina_send_packet(struct sk_buff *skb, struct net_device *dev)
lp->td_ring[chain_prev].control &=
~DMA_DESC_COF;
/* Link to prev */
- lp->td_ring[chain_prev].link = CPHYSADDR(td);
+ lp->td_ring[chain_prev].link = korina_tx_dma(lp, idx);
/* Move tail */
lp->tx_chain_tail = chain_next;
/* Write to NDPTR */
- writel(CPHYSADDR(&lp->td_ring[lp->tx_chain_head]),
- &(lp->tx_dma_regs->dmandptr));
+ writel(korina_tx_dma(lp, lp->tx_chain_head),
+ &lp->tx_dma_regs->dmandptr);
/* Move head to tail */
lp->tx_chain_head = lp->tx_chain_tail;
lp->tx_chain_status = desc_empty;
@@ -255,7 +267,7 @@ static int korina_send_packet(struct sk_buff *skb, struct net_device *dev)
DMA_DESC_COF | DMA_DESC_IOF;
lp->td_ring[chain_prev].control &=
~DMA_DESC_COF;
- lp->td_ring[chain_prev].link = CPHYSADDR(td);
+ lp->td_ring[chain_prev].link = korina_tx_dma(lp, idx);
lp->tx_chain_tail = chain_next;
}
}
@@ -263,6 +275,13 @@ static int korina_send_packet(struct sk_buff *skb, struct net_device *dev)
netif_trans_update(dev);
spin_unlock_irqrestore(&lp->lock, flags);
+ return NETDEV_TX_OK;
+
+drop_packet:
+ dev->stats.tx_dropped++;
+ dev_kfree_skb_any(skb);
+ spin_unlock_irqrestore(&lp->lock, flags);
+
return NETDEV_TX_OK;
}
@@ -345,8 +364,8 @@ static int korina_rx(struct net_device *dev, int limit)
struct korina_private *lp = netdev_priv(dev);
struct dma_desc *rd = &lp->rd_ring[lp->rx_next_done];
struct sk_buff *skb, *skb_new;
- u8 *pkt_buf;
u32 devcs, pkt_len, dmas;
+ dma_addr_t ca;
int count;
for (count = 0; count < limit; count++) {
@@ -358,6 +377,8 @@ static int korina_rx(struct net_device *dev, int limit)
if ((KORINA_RBSIZE - (u32)DMA_COUNT(rd->control)) == 0)
break;
+ ca = rd->ca;
+
/* check that this is a whole packet
* WARNING: DMA_FD bit incorrectly set
* in Rc32434 (errata ref #077) */
@@ -382,20 +403,21 @@ static int korina_rx(struct net_device *dev, int limit)
goto next;
}
- pkt_len = RCVPKT_LENGTH(devcs);
-
- /* must be the (first and) last
- * descriptor then */
- pkt_buf = (u8 *)lp->rx_skb[lp->rx_next_done]->data;
-
- /* invalidate the cache */
- dma_cache_inv((unsigned long)pkt_buf, pkt_len - 4);
-
/* Malloc up new buffer. */
skb_new = netdev_alloc_skb_ip_align(dev, KORINA_RBSIZE);
-
if (!skb_new)
break;
+
+ ca = dma_map_single(lp->dmadev, skb_new->data, KORINA_RBSIZE,
+ DMA_FROM_DEVICE);
+ if (dma_mapping_error(lp->dmadev, ca)) {
+ dev_kfree_skb_any(skb_new);
+ break;
+ }
+
+ pkt_len = RCVPKT_LENGTH(devcs);
+ dma_unmap_single(lp->dmadev, rd->ca, pkt_len, DMA_FROM_DEVICE);
+
/* Do not count the CRC */
skb_put(skb, pkt_len - 4);
skb->protocol = eth_type_trans(skb, dev);
@@ -415,10 +437,7 @@ static int korina_rx(struct net_device *dev, int limit)
rd->devcs = 0;
/* Restore descriptor's curr_addr */
- if (skb_new)
- rd->ca = CPHYSADDR(skb_new->data);
- else
- rd->ca = CPHYSADDR(skb->data);
+ rd->ca = ca;
rd->control = DMA_COUNT(KORINA_RBSIZE) |
DMA_DESC_COD | DMA_DESC_IOD;
@@ -439,9 +458,8 @@ static int korina_rx(struct net_device *dev, int limit)
lp->dma_halt_cnt++;
rd->devcs = 0;
- skb = lp->rx_skb[lp->rx_next_done];
- rd->ca = CPHYSADDR(skb->data);
- writel(CPHYSADDR(rd), &lp->rx_dma_regs->dmandptr);
+ writel(korina_rx_dma(lp, rd - lp->rd_ring),
+ &lp->rx_dma_regs->dmandptr);
}
return count;
@@ -564,6 +582,9 @@ static void korina_tx(struct net_device *dev)
/* We must always free the original skb */
if (lp->tx_skb[lp->tx_next_done]) {
+ dma_unmap_single(lp->dmadev, td->ca,
+ lp->tx_skb[lp->tx_next_done]->len,
+ DMA_TO_DEVICE);
dev_kfree_skb_any(lp->tx_skb[lp->tx_next_done]);
lp->tx_skb[lp->tx_next_done] = NULL;
}
@@ -610,8 +631,8 @@ korina_tx_dma_interrupt(int irq, void *dev_id)
if (lp->tx_chain_status == desc_filled &&
(readl(&(lp->tx_dma_regs->dmandptr)) == 0)) {
- writel(CPHYSADDR(&lp->td_ring[lp->tx_chain_head]),
- &(lp->tx_dma_regs->dmandptr));
+ writel(korina_tx_dma(lp, lp->tx_chain_head),
+ &lp->tx_dma_regs->dmandptr);
lp->tx_chain_status = desc_empty;
lp->tx_chain_head = lp->tx_chain_tail;
netif_trans_update(dev);
@@ -753,13 +774,17 @@ static int korina_alloc_ring(struct net_device *dev)
lp->rd_ring[i].control = DMA_DESC_IOD |
DMA_COUNT(KORINA_RBSIZE);
lp->rd_ring[i].devcs = 0;
- lp->rd_ring[i].ca = CPHYSADDR(skb->data);
- lp->rd_ring[i].link = CPHYSADDR(&lp->rd_ring[i+1]);
+ lp->rd_ring[i].ca = dma_map_single(lp->dmadev, skb->data,
+ KORINA_RBSIZE,
+ DMA_FROM_DEVICE);
+ if (dma_mapping_error(lp->dmadev, lp->rd_ring[i].ca))
+ return -ENOMEM;
+ lp->rd_ring[i].link = korina_rx_dma(lp, i + 1);
}
/* loop back receive descriptors, so the last
* descriptor points to the first one */
- lp->rd_ring[i - 1].link = CPHYSADDR(&lp->rd_ring[0]);
+ lp->rd_ring[i - 1].link = lp->rd_dma;
lp->rd_ring[i - 1].control |= DMA_DESC_COD;
lp->rx_next_done = 0;
@@ -777,16 +802,22 @@ static void korina_free_ring(struct net_device *dev)
for (i = 0; i < KORINA_NUM_RDS; i++) {
lp->rd_ring[i].control = 0;
- if (lp->rx_skb[i])
+ if (lp->rx_skb[i]) {
+ dma_unmap_single(lp->dmadev, lp->rd_ring[i].ca,
+ KORINA_RBSIZE, DMA_FROM_DEVICE);
dev_kfree_skb_any(lp->rx_skb[i]);
- lp->rx_skb[i] = NULL;
+ lp->rx_skb[i] = NULL;
+ }
}
for (i = 0; i < KORINA_NUM_TDS; i++) {
lp->td_ring[i].control = 0;
- if (lp->tx_skb[i])
+ if (lp->tx_skb[i]) {
+ dma_unmap_single(lp->dmadev, lp->td_ring[i].ca,
+ lp->tx_skb[i]->len, DMA_TO_DEVICE);
dev_kfree_skb_any(lp->tx_skb[i]);
- lp->tx_skb[i] = NULL;
+ lp->tx_skb[i] = NULL;
+ }
}
}
@@ -819,7 +850,7 @@ static int korina_init(struct net_device *dev)
writel(0, &lp->rx_dma_regs->dmas);
/* Start Rx DMA */
writel(0, &lp->rx_dma_regs->dmandptr);
- writel(CPHYSADDR(&lp->rd_ring[0]), &lp->rx_dma_regs->dmadptr);
+ writel(korina_rx_dma(lp, 0), &lp->rx_dma_regs->dmadptr);
writel(readl(&lp->tx_dma_regs->dmasm) &
~(DMA_STAT_FINI | DMA_STAT_ERR),
@@ -1054,21 +1085,21 @@ static int korina_probe(struct platform_device *pdev)
}
lp->tx_dma_regs = p;
- lp->td_ring = kmalloc(TD_RING_SIZE + RD_RING_SIZE, GFP_KERNEL);
+ lp->td_ring = dmam_alloc_coherent(&pdev->dev, TD_RING_SIZE,
+ &lp->td_dma, GFP_KERNEL);
if (!lp->td_ring)
return -ENOMEM;
- dma_cache_inv((unsigned long)(lp->td_ring),
- TD_RING_SIZE + RD_RING_SIZE);
-
- /* now convert TD_RING pointer to KSEG1 */
- lp->td_ring = (struct dma_desc *)KSEG1ADDR(lp->td_ring);
- lp->rd_ring = &lp->td_ring[KORINA_NUM_TDS];
+ lp->rd_ring = dmam_alloc_coherent(&pdev->dev, RD_RING_SIZE,
+ &lp->rd_dma, GFP_KERNEL);
+ if (!lp->rd_ring)
+ return -ENOMEM;
spin_lock_init(&lp->lock);
/* just use the rx dma irq */
dev->irq = lp->rx_irq;
lp->dev = dev;
+ lp->dmadev = &pdev->dev;
dev->netdev_ops = &korina_netdev_ops;
dev->ethtool_ops = &netdev_ethtool_ops;
@@ -1086,7 +1117,6 @@ static int korina_probe(struct platform_device *pdev)
if (rc < 0) {
printk(KERN_ERR DRV_NAME
": cannot register net device: %d\n", rc);
- kfree((struct dma_desc *)KSEG0ADDR(lp->td_ring));
return rc;
}
timer_setup(&lp->media_check_timer, korina_poll_media, 0);
@@ -1101,9 +1131,6 @@ static int korina_probe(struct platform_device *pdev)
static int korina_remove(struct platform_device *pdev)
{
struct korina_device *bif = platform_get_drvdata(pdev);
- struct korina_private *lp = netdev_priv(bif->dev);
-
- kfree((struct dma_desc *)KSEG0ADDR(lp->td_ring));
unregister_netdev(bif->dev);
--
2.29.2
next prev parent reply other threads:[~2021-04-13 20:48 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-13 20:48 [PATCH net-next 0/7] net: Korina improvements Thomas Bogendoerfer
2021-04-13 20:48 ` [PATCH net-next 1/7] net: korina: Fix MDIO functions Thomas Bogendoerfer
2021-04-14 19:36 ` Andrew Lunn
2021-04-14 20:36 ` Thomas Bogendoerfer
2021-04-13 20:48 ` [PATCH net-next 2/7] net: korina: Use devres functions Thomas Bogendoerfer
2021-04-14 19:38 ` Andrew Lunn
2021-04-13 20:48 ` [PATCH net-next 3/7] net: korina: Remove not needed cache flushes Thomas Bogendoerfer
2021-04-13 20:48 ` [PATCH net-next 4/7] net: korina: Remove nested helpers Thomas Bogendoerfer
2021-04-13 20:48 ` Thomas Bogendoerfer [this message]
2021-04-13 20:48 ` [PATCH net-next 6/7] net: korina: Only pass mac address via platform data Thomas Bogendoerfer
2021-04-13 20:48 ` [PATCH net-next 7/7] net: korina: Make driver COMPILE_TESTable Thomas Bogendoerfer
2021-04-14 2:06 ` kernel test robot
2021-04-14 2:37 ` kernel test robot
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210413204818.23350-6-tsbogend@alpha.franken.de \
--to=tsbogend@alpha.franken.de \
--cc=davem@davemloft.net \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).