[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20220817193017.44063-4-gerhard@engleder-embedded.com>
Date: Wed, 17 Aug 2022 21:30:15 +0200
From: Gerhard Engleder <gerhard@...leder-embedded.com>
To: davem@...emloft.net, kuba@...nel.org
Cc: netdev@...r.kernel.org,
Gerhard Engleder <gerhard@...leder-embedded.com>
Subject: [PATCH net-next 3/5] tsnep: Improve TX length handling
TX length can by calculated more efficient during map and unmap of
fragments. Another reason is that, by moving TX statistic counting to
tsnep_tx_poll() it can be used there for XDP too.
Signed-off-by: Gerhard Engleder <gerhard@...leder-embedded.com>
---
drivers/net/ethernet/engleder/tsnep_main.c | 32 ++++++++++++++--------
1 file changed, 21 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index 7fe6a897ce00..9f8ca6d9a010 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -262,14 +262,14 @@ static int tsnep_tx_ring_init(struct tsnep_tx *tx)
return retval;
}
-static void tsnep_tx_activate(struct tsnep_tx *tx, int index, bool last)
+static void tsnep_tx_activate(struct tsnep_tx *tx, int index, int length,
+ bool last)
{
struct tsnep_tx_entry *entry = &tx->entry[index];
entry->properties = 0;
if (entry->skb) {
- entry->properties =
- skb_pagelen(entry->skb) & TSNEP_DESC_LENGTH_MASK;
+ entry->properties = length & TSNEP_DESC_LENGTH_MASK;
entry->properties |= TSNEP_DESC_INTERRUPT_FLAG;
if (skb_shinfo(entry->skb)->tx_flags & SKBTX_IN_PROGRESS)
entry->properties |= TSNEP_DESC_EXTENDED_WRITEBACK_FLAG;
@@ -334,6 +334,7 @@ static int tsnep_tx_map(struct sk_buff *skb, struct tsnep_tx *tx, int count)
struct tsnep_tx_entry *entry;
unsigned int len;
dma_addr_t dma;
+ int map_len = 0;
int i;
for (i = 0; i < count; i++) {
@@ -356,15 +357,18 @@ static int tsnep_tx_map(struct sk_buff *skb, struct tsnep_tx *tx, int count)
dma_unmap_addr_set(entry, dma, dma);
entry->desc->tx = __cpu_to_le64(dma);
+
+ map_len += len;
}
- return 0;
+ return map_len;
}
-static void tsnep_tx_unmap(struct tsnep_tx *tx, int index, int count)
+static int tsnep_tx_unmap(struct tsnep_tx *tx, int index, int count)
{
struct device *dmadev = tx->adapter->dmadev;
struct tsnep_tx_entry *entry;
+ int map_len = 0;
int i;
for (i = 0; i < count; i++) {
@@ -381,9 +385,12 @@ static void tsnep_tx_unmap(struct tsnep_tx *tx, int index, int count)
dma_unmap_addr(entry, dma),
dma_unmap_len(entry, len),
DMA_TO_DEVICE);
+ map_len += entry->len;
entry->len = 0;
}
}
+
+ return map_len;
}
static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb,
@@ -392,6 +399,7 @@ static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb,
unsigned long flags;
int count = 1;
struct tsnep_tx_entry *entry;
+ int length;
int i;
int retval;
@@ -415,7 +423,7 @@ static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb,
entry->skb = skb;
retval = tsnep_tx_map(skb, tx, count);
- if (retval != 0) {
+ if (retval < 0) {
tsnep_tx_unmap(tx, tx->write, count);
dev_kfree_skb_any(entry->skb);
entry->skb = NULL;
@@ -428,12 +436,13 @@ static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb,
return NETDEV_TX_OK;
}
+ length = retval;
if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)
skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
for (i = 0; i < count; i++)
- tsnep_tx_activate(tx, (tx->write + i) % TSNEP_RING_SIZE,
+ tsnep_tx_activate(tx, (tx->write + i) % TSNEP_RING_SIZE, length,
i == (count - 1));
tx->write = (tx->write + count) % TSNEP_RING_SIZE;
@@ -449,9 +458,6 @@ static netdev_tx_t tsnep_xmit_frame_ring(struct sk_buff *skb,
netif_stop_queue(tx->adapter->netdev);
}
- tx->packets++;
- tx->bytes += skb_pagelen(entry->skb) + ETH_FCS_LEN;
-
spin_unlock_irqrestore(&tx->lock, flags);
return NETDEV_TX_OK;
@@ -463,6 +469,7 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
int budget = 128;
struct tsnep_tx_entry *entry;
int count;
+ int length;
spin_lock_irqsave(&tx->lock, flags);
@@ -485,7 +492,7 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
if (skb_shinfo(entry->skb)->nr_frags > 0)
count += skb_shinfo(entry->skb)->nr_frags;
- tsnep_tx_unmap(tx, tx->read, count);
+ length = tsnep_tx_unmap(tx, tx->read, count);
if ((skb_shinfo(entry->skb)->tx_flags & SKBTX_IN_PROGRESS) &&
(__le32_to_cpu(entry->desc_wb->properties) &
@@ -512,6 +519,9 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
tx->read = (tx->read + count) % TSNEP_RING_SIZE;
+ tx->packets++;
+ tx->bytes += length + ETH_FCS_LEN;
+
budget--;
} while (likely(budget));
--
2.30.2
Powered by blists - more mailing lists