From: Sven Van Asbroeck <[email protected]>
Tree: git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git # e4b62cf7559f
v1 -> v2:
- Andrew Lunn:
+ always keep to Reverse Christmas Tree.
+ "changing the cache operations to operate on the received length" should
go in its own, separate patch, so it can be easily backed out if
"interesting things" should happen with it.
- Bryan Whitehead:
+ multi-buffer patch concept "looks good".
As a result, I will squash the intermediate "dma buffer only" patch which
demonstrated the speed boost using an inflexible solution
(w/o multi-buffers).
+ Rename lan743x_rx_process_buffer() to lan743x_rx_process_packet()
+ Remove unused RX_PROCESS_RESULT_PACKET_DROPPED
+ Rename RX_PROCESS_RESULT_BUFFER_RECEIVED to
RX_PROCESS_RESULT_PACKET_RECEIVED
+ Fold "unmap from dma" into lan743x_rx_init_ring_element() to prevent
use-after-dma-unmap issue
+ ensure that skb allocation issues do not result in the driver sending
incomplete packets to the OS. E.g. a three-buffer packet, with the
middle buffer missing
- Willem De Bruyn: skb_hwtstamps(skb) always returns a non-null value, if the
skb parameter points to a valid skb.
Summary of my tests below.
Suggestions for better tests are very welcome.
Tests with debug logging enabled (add #define DEBUG).
1. Limit rx buffer size to 500, so mtu (1500) takes 3 buffers.
Ping to chip, verify correct packet size is sent to OS.
Ping large packets to chip (ping -s 1400), verify correct
packet size is sent to OS.
Ping using packets around the buffer size, verify number of
buffers is changing, verify correct packet size is sent
to OS:
$ ping -s 472
$ ping -s 473
$ ping -s 992
$ ping -s 993
Verify that each packet is followed by extension processing.
2. Limit rx buffer size to 500, so mtu (1500) takes 3 buffers.
Run iperf3 -s on chip, verify that packets come in 3 buffers
at a time.
Verify that packet size is equal to mtu.
Verify that each packet is followed by extension processing.
3. Set mtu to 2000 on chip and host.
Limit rx buffer size to 500, so mtu (2000) takes 4 buffers.
Run iperf3 -s on chip, verify that packets come in 4 buffers
at a time.
Verify that packet size is equal to mtu.
Verify that each packet is followed by extension processing.
Tests with debug logging DISabled (remove #define DEBUG).
4. Limit rx buffer size to 500, so mtu (1500) takes 3 buffers.
Run iperf3 -s on chip, note sustained rx speed.
Set mtu to 2000, so mtu takes 4 buffers.
Run iperf3 -s on chip, note sustained rx speed.
Verify no packets are dropped in both cases.
Verify speeds are roughly comparable.
Tests with DEBUG_KMEMLEAK on:
$ mount -t debugfs nodev /sys/kernel/debug/
$ echo scan > /sys/kernel/debug/kmemleak
5. Limit rx buffer size to 500, so mtu (1500) takes 3 buffers.
Run the following tests concurrently for at least one hour:
- iperf3 -s on chip
- ping -> chip
Monitor reported memory leaks.
6. Set mtu to 2000.
Limit rx buffer size to 500, so mtu (2000) takes 4 buffers.
Run the following tests concurrently for at least one hour:
- iperf3 -s on chip
- ping -> chip
Monitor reported memory leaks.
7. Simulate low-memory in lan743x_rx_allocate_skb(): fail once every
100 allocations.
Repeat (5) and (6).
Monitor reported memory leaks.
8. Simulate low-memory in lan743x_rx_allocate_skb(): fail 10
allocations in a row in every 100.
Repeat (5) and (6).
Monitor reported memory leaks.
9. Simulate low-memory in lan743x_rx_trim_skb(): fail 1 allocation
in every 100.
Repeat (5) and (6).
Monitor reported memory leaks.
Tests with debug logging enabled (add #define DEBUG).
10. Set the chip mtu to 1500, generate lots of network traffic.
Stop all network traffic.
Set the chip and remote mtus to 8000.
Ping remote -> chip: $ ping <chip ip> -s 7000
Verify that the first few received packets are multi-buffer.
Verify no pings are dropped.
Tests with DEBUG_KMEMLEAK on:
$ mount -t debugfs nodev /sys/kernel/debug/
$ echo scan > /sys/kernel/debug/kmemleak
11. Start with chip mtu at 1500, host mtu at 8000.
Run concurrently:
- iperf3 -s on chip
- ping -> chip
Cycle the chip mtu between 1500 and 8000 every 10 seconds.
Scan kmemleak periodically to watch for memory leaks.
Verify that the mtu changeover happens smoothly, i.e.
the iperf3 test does not report periods where speed
drops and recovers suddenly.
Note: iperf3 occasionally reports dropped packets on
changeover. This behaviour also occurs on the original
driver, it's not a regression. Possibly related to the
chip's mac rx being disabled when the mtu is changed.
To: Bryan Whitehead <[email protected]>
To: [email protected]
To: "David S. Miller" <[email protected]>
To: Jakub Kicinski <[email protected]>
Cc: Andrew Lunn <[email protected]>
Cc: Alexey Denisov <[email protected]>
Cc: Sergej Bauer <[email protected]>
Cc: Tim Harvey <[email protected]>
Cc: Anders Rønningen <[email protected]>
Cc: Hillf Danton <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Willem de Bruijn <[email protected]>
Cc: [email protected]
Cc: [email protected]
Sven Van Asbroeck (5):
lan743x: boost performance on cpu archs w/o dma cache snooping
lan743x: sync only the received area of an rx ring buffer
TEST ONLY: lan743x: limit rx ring buffer size to 500 bytes
TEST ONLY: lan743x: skb_alloc failure test
TEST ONLY: lan743x: skb_trim failure test
drivers/net/ethernet/microchip/lan743x_main.c | 350 +++++++++---------
drivers/net/ethernet/microchip/lan743x_main.h | 5 +-
2 files changed, 172 insertions(+), 183 deletions(-)
--
2.17.1
From: Sven Van Asbroeck <[email protected]>
Simulate low-memory in lan743x_rx_allocate_skb(): fail 10
allocations in a row in every 100.
Signed-off-by: Sven Van Asbroeck <[email protected]>
---
To: Bryan Whitehead <[email protected]>
To: [email protected]
To: "David S. Miller" <[email protected]>
To: Jakub Kicinski <[email protected]>
Cc: Andrew Lunn <[email protected]>
Cc: Alexey Denisov <[email protected]>
Cc: Sergej Bauer <[email protected]>
Cc: Tim Harvey <[email protected]>
Cc: Anders Rønningen <[email protected]>
Cc: Hillf Danton <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Willem de Bruijn <[email protected]>
Cc: [email protected]
Cc: [email protected]
drivers/net/ethernet/microchip/lan743x_main.c | 21 +++++++++++++++----
1 file changed, 17 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c
index 90d49231494d..0094ecac5741 100644
--- a/drivers/net/ethernet/microchip/lan743x_main.c
+++ b/drivers/net/ethernet/microchip/lan743x_main.c
@@ -1963,7 +1963,20 @@ static void lan743x_rx_update_tail(struct lan743x_rx *rx, int index)
index);
}
-static int lan743x_rx_init_ring_element(struct lan743x_rx *rx, int index)
+static struct sk_buff *
+lan743x_alloc_skb(struct net_device *netdev, int length, bool can_fail)
+{
+ static int rx_alloc;
+ int counter = rx_alloc++ % 100;
+
+ if (can_fail && counter >= 20 && counter < 30)
+ return NULL;
+
+ return __netdev_alloc_skb(netdev, length, GFP_ATOMIC | GFP_DMA);
+}
+
+static int
+lan743x_rx_init_ring_element(struct lan743x_rx *rx, int index, bool can_fail)
{
struct net_device *netdev = rx->adapter->netdev;
struct device *dev = &rx->adapter->pdev->dev;
@@ -1977,7 +1990,7 @@ static int lan743x_rx_init_ring_element(struct lan743x_rx *rx, int index)
descriptor = &rx->ring_cpu_ptr[index];
buffer_info = &rx->buffer_info[index];
- skb = __netdev_alloc_skb(netdev, buffer_length, GFP_ATOMIC | GFP_DMA);
+ skb = lan743x_alloc_skb(netdev, buffer_length, can_fail);
if (!skb)
return -ENOMEM;
dma_ptr = dma_map_single(dev, skb->data, buffer_length, DMA_FROM_DEVICE);
@@ -2137,7 +2150,7 @@ static int lan743x_rx_process_buffer(struct lan743x_rx *rx)
/* save existing skb, allocate new skb and map to dma */
skb = buffer_info->skb;
- if (lan743x_rx_init_ring_element(rx, rx->last_head)) {
+ if (lan743x_rx_init_ring_element(rx, rx->last_head, true)) {
/* failed to allocate next skb.
* Memory is very low.
* Drop this packet and reuse buffer.
@@ -2342,7 +2355,7 @@ static int lan743x_rx_ring_init(struct lan743x_rx *rx)
rx->last_head = 0;
for (index = 0; index < rx->ring_size; index++) {
- ret = lan743x_rx_init_ring_element(rx, index);
+ ret = lan743x_rx_init_ring_element(rx, index, false);
if (ret)
goto cleanup;
}
--
2.17.1
From: Sven Van Asbroeck <[email protected]>
Signed-off-by: Sven Van Asbroeck <[email protected]>
---
To: Bryan Whitehead <[email protected]>
To: [email protected]
To: "David S. Miller" <[email protected]>
To: Jakub Kicinski <[email protected]>
Cc: Andrew Lunn <[email protected]>
Cc: Alexey Denisov <[email protected]>
Cc: Sergej Bauer <[email protected]>
Cc: Tim Harvey <[email protected]>
Cc: Anders Rønningen <[email protected]>
Cc: Hillf Danton <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Willem de Bruijn <[email protected]>
Cc: [email protected]
Cc: [email protected]
drivers/net/ethernet/microchip/lan743x_main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c
index 36cc67c72851..90d49231494d 100644
--- a/drivers/net/ethernet/microchip/lan743x_main.c
+++ b/drivers/net/ethernet/microchip/lan743x_main.c
@@ -1973,7 +1973,7 @@ static int lan743x_rx_init_ring_element(struct lan743x_rx *rx, int index)
struct sk_buff *skb;
dma_addr_t dma_ptr;
- buffer_length = netdev->mtu + ETH_HLEN + 4 + RX_HEAD_PADDING;
+ buffer_length = 500 + ETH_HLEN + 4 + RX_HEAD_PADDING;
descriptor = &rx->ring_cpu_ptr[index];
buffer_info = &rx->buffer_info[index];
--
2.17.1
From: Sven Van Asbroeck <[email protected]>
Simulate low-memory in lan743x_rx_trim_skb(): fail one allocation
in every 100.
Signed-off-by: Sven Van Asbroeck <[email protected]>
---
To: Bryan Whitehead <[email protected]>
To: [email protected]
To: "David S. Miller" <[email protected]>
To: Jakub Kicinski <[email protected]>
Cc: Andrew Lunn <[email protected]>
Cc: Alexey Denisov <[email protected]>
Cc: Sergej Bauer <[email protected]>
Cc: Tim Harvey <[email protected]>
Cc: Anders Rønningen <[email protected]>
Cc: Hillf Danton <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Willem de Bruijn <[email protected]>
Cc: [email protected]
Cc: [email protected]
drivers/net/ethernet/microchip/lan743x_main.c | 28 ++++++++-----------
1 file changed, 11 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c
index 0094ecac5741..53c2b93b82b4 100644
--- a/drivers/net/ethernet/microchip/lan743x_main.c
+++ b/drivers/net/ethernet/microchip/lan743x_main.c
@@ -1963,20 +1963,7 @@ static void lan743x_rx_update_tail(struct lan743x_rx *rx, int index)
index);
}
-static struct sk_buff *
-lan743x_alloc_skb(struct net_device *netdev, int length, bool can_fail)
-{
- static int rx_alloc;
- int counter = rx_alloc++ % 100;
-
- if (can_fail && counter >= 20 && counter < 30)
- return NULL;
-
- return __netdev_alloc_skb(netdev, length, GFP_ATOMIC | GFP_DMA);
-}
-
-static int
-lan743x_rx_init_ring_element(struct lan743x_rx *rx, int index, bool can_fail)
+static int lan743x_rx_init_ring_element(struct lan743x_rx *rx, int index)
{
struct net_device *netdev = rx->adapter->netdev;
struct device *dev = &rx->adapter->pdev->dev;
@@ -1990,7 +1977,7 @@ lan743x_rx_init_ring_element(struct lan743x_rx *rx, int index, bool can_fail)
descriptor = &rx->ring_cpu_ptr[index];
buffer_info = &rx->buffer_info[index];
- skb = lan743x_alloc_skb(netdev, buffer_length, can_fail);
+ skb = __netdev_alloc_skb(netdev, buffer_length, GFP_ATOMIC | GFP_DMA);
if (!skb)
return -ENOMEM;
dma_ptr = dma_map_single(dev, skb->data, buffer_length, DMA_FROM_DEVICE);
@@ -2075,6 +2062,13 @@ static void lan743x_rx_release_ring_element(struct lan743x_rx *rx, int index)
static struct sk_buff *
lan743x_rx_trim_skb(struct sk_buff *skb, int frame_length)
{
+ static int trim_cnt;
+
+ if ((trim_cnt++ % 100) == 77) {
+ dev_kfree_skb_irq(skb);
+ return NULL;
+ }
+
if (skb_linearize(skb)) {
dev_kfree_skb_irq(skb);
return NULL;
@@ -2150,7 +2144,7 @@ static int lan743x_rx_process_buffer(struct lan743x_rx *rx)
/* save existing skb, allocate new skb and map to dma */
skb = buffer_info->skb;
- if (lan743x_rx_init_ring_element(rx, rx->last_head, true)) {
+ if (lan743x_rx_init_ring_element(rx, rx->last_head)) {
/* failed to allocate next skb.
* Memory is very low.
* Drop this packet and reuse buffer.
@@ -2355,7 +2349,7 @@ static int lan743x_rx_ring_init(struct lan743x_rx *rx)
rx->last_head = 0;
for (index = 0; index < rx->ring_size; index++) {
- ret = lan743x_rx_init_ring_element(rx, index, false);
+ ret = lan743x_rx_init_ring_element(rx, index);
if (ret)
goto cleanup;
}
--
2.17.1