diff mbox series

[v4] ixgbe: let the xdpdrv work with more than 64 cpus

Message ID 20210826141623.8151-1-kerneljasonxing@gmail.com
State Superseded
Headers show
Series [v4] ixgbe: let the xdpdrv work with more than 64 cpus | expand

Commit Message

Jason Xing Aug. 26, 2021, 2:16 p.m. UTC
From: Jason Xing <xingwanli@kuaishou.com>

Originally, ixgbe driver doesn't allow the mounting of xdpdrv if the
server is equipped with more than 64 cpus online. So it turns out that
the loading of xdpdrv causes the "NOMEM" failure.

Actually, we can adjust the algorithm and then make it work through
mapping the current cpu to some xdp ring with the protect of @tx_lock.

v4:
- Update the wrong commit messages. (Jason)

v3:
- Change nr_cpu_ids to num_online_cpus() (Maciej)
- Rename MAX_XDP_QUEUES to IXGBE_MAX_XDP_QS (Maciej)
- Rename ixgbe_determine_xdp_cpu() to ixgbe_determine_xdp_q_idx() (Maciej)
- Wrap ixgbe_xdp_ring_update_tail() with lock into one function (Maciej)

v2:
- Adjust cpu id in ixgbe_xdp_xmit(). (Jesper)
- Add a fallback path. (Maciej)
- Adjust other parts related to xdp ring.

Fixes: 33fdc82f08 ("ixgbe: add support for XDP_TX action")
Co-developed-by: Shujin Li <lishujin@kuaishou.com>
Signed-off-by: Shujin Li <lishujin@kuaishou.com>
Signed-off-by: Jason Xing <xingwanli@kuaishou.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe.h           | 15 ++++-
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c       |  9 ++-
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c      | 64 ++++++++++++++++------
 .../net/ethernet/intel/ixgbe/ixgbe_txrx_common.h   |  1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c       |  9 +--
 5 files changed, 73 insertions(+), 25 deletions(-)

Comments

Jesse Brandeburg Aug. 26, 2021, 3:23 p.m. UTC | #1
On 8/26/2021 7:16 AM, kerneljasonxing@gmail.com wrote:
> From: Jason Xing <xingwanli@kuaishou.com>
> 
> Originally, ixgbe driver doesn't allow the mounting of xdpdrv if the
> server is equipped with more than 64 cpus online. So it turns out that
> the loading of xdpdrv causes the "NOMEM" failure.
> 
> Actually, we can adjust the algorithm and then make it work through
> mapping the current cpu to some xdp ring with the protect of @tx_lock.

Thank you very much for working on this!

you should put your sign off block here, and then end with a triple-dash
"---"

then have your vN: updates below that, so they will be dropped from
final git apply. It's ok to have more than one triple-dash.

> 
> v4:
> - Update the wrong commit messages. (Jason)
> 
> v3:
> - Change nr_cpu_ids to num_online_cpus() (Maciej)
> - Rename MAX_XDP_QUEUES to IXGBE_MAX_XDP_QS (Maciej)
> - Rename ixgbe_determine_xdp_cpu() to ixgbe_determine_xdp_q_idx() (Maciej)
> - Wrap ixgbe_xdp_ring_update_tail() with lock into one function (Maciej)
> 
> v2:
> - Adjust cpu id in ixgbe_xdp_xmit(). (Jesper)
> - Add a fallback path. (Maciej)
> - Adjust other parts related to xdp ring.
> 
> Fixes: 33fdc82f08 ("ixgbe: add support for XDP_TX action")
> Co-developed-by: Shujin Li <lishujin@kuaishou.com>
> Signed-off-by: Shujin Li <lishujin@kuaishou.com>
> Signed-off-by: Jason Xing <xingwanli@kuaishou.com>
> ---
>  drivers/net/ethernet/intel/ixgbe/ixgbe.h           | 15 ++++-
>  drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c       |  9 ++-
>  drivers/net/ethernet/intel/ixgbe/ixgbe_main.c      | 64 ++++++++++++++++------
>  .../net/ethernet/intel/ixgbe/ixgbe_txrx_common.h   |  1 +
>  drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c       |  9 +--
>  5 files changed, 73 insertions(+), 25 deletions(-)

...

> @@ -8539,21 +8539,32 @@ static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb,
>  int ixgbe_xmit_xdp_ring(struct ixgbe_adapter *adapter,
>  			struct xdp_frame *xdpf)
>  {
> -	struct ixgbe_ring *ring = adapter->xdp_ring[smp_processor_id()];
>  	struct ixgbe_tx_buffer *tx_buffer;
>  	union ixgbe_adv_tx_desc *tx_desc;
> +	struct ixgbe_ring *ring;
>  	u32 len, cmd_type;
>  	dma_addr_t dma;
> +	int index, ret;
>  	u16 i;
>  
>  	len = xdpf->len;
>  
> -	if (unlikely(!ixgbe_desc_unused(ring)))
> -		return IXGBE_XDP_CONSUMED;
> +	index = ixgbe_determine_xdp_q_idx(smp_processor_id());
> +	ring = adapter->xdp_ring[index];
> +
> +	if (static_branch_unlikely(&ixgbe_xdp_locking_key))
> +		spin_lock(&ring->tx_lock);
> +
> +	if (unlikely(!ixgbe_desc_unused(ring))) {
> +		ret = IXGBE_XDP_CONSUMED;
> +		goto out;
> +	}

This static key stuff is tricky code, but I guess if it works, then it's
better than nothing.

As Maciej also commented, I'd like to see some before/after numbers for
some of the xdp sample programs to make sure this doesn't cause a huge
regression on the xdp transmit path. A small regression would be ok,
since this *is* adding overhead.

Jesse
Jesse Brandeburg Aug. 26, 2021, 4:41 p.m. UTC | #2
On 8/26/2021 9:18 AM, Eric Dumazet wrote:

>> +static inline int ixgbe_determine_xdp_q_idx(int cpu)
>> +{
>> +	if (static_key_enabled(&ixgbe_xdp_locking_key))
>> +		return cpu % IXGBE_MAX_XDP_QS;
>> +	else
>> +		return cpu;
> 
> Even if num_online_cpus() is 8, the returned cpu here could be
> 
> 0, 32, 64, 96, 128, 161, 197, 224
> 
> Are we sure this will still be ok ?

I'm not sure about that one myself. Jason?

> 
>> +}
>> +
>>  static inline u8 ixgbe_max_rss_indices(struct ixgbe_adapter *adapter)
>>  {
>>  	switch (adapter->hw.mac.type) {
>> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
>> index 0218f6c..884bf99 100644
>> --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
>> +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
>> @@ -299,7 +299,10 @@ static void ixgbe_cache_ring_register(struct ixgbe_adapter *adapter)
>>  
>>  static int ixgbe_xdp_queues(struct ixgbe_adapter *adapter)
>>  {
>> -	return adapter->xdp_prog ? nr_cpu_ids : 0;
>> +	int queues;
>> +
>> +	queues = min_t(int, IXGBE_MAX_XDP_QS, num_online_cpus());
> 
> num_online_cpus() might change later...

I saw that too, but I wonder if it doesn't matter to the driver. If a
CPU goes offline or comes online after the driver loads, we will use
this logic to try to pick an available TX queue. But this is a
complicated thing that is easy to get wrong, is there a common example
of how to get it right?

A possible problem I guess is that if the "static_key_enabled" check
returned false in the past, we would need to update that if the number
of CPUs changes, do we need a notifier?

Also, now that I'm asking it, I dislike the global as it would apply to
all ixgbe ports and each PF would increment and decrement it
independently. Showing my ignorance here, but I haven't seen this
utility in the kernel before in detail. Not sure if this is "OK" from
multiple device (with the same driver / global namespace) perspective.
Eric Dumazet Aug. 26, 2021, 6:19 p.m. UTC | #3
On 8/26/21 10:03 AM, Jason Xing wrote:

> 
> Honestly, I'm a little confused right now. @nr_cpu_ids is the fixed
> number which means the total number of cpus the machine has.
> I think, using @nr_cpu_ids is safe one way or the other regardless of
> whether the cpu goes offline or not. What do you think?
> 

More exactly, nr_cpu_ids is the max number cpu id can reach,
even in presence of holes.

I think that most/many num_online_cpus() in drivers/net are simply broken
and should be replaced by nr_cpu_ids.

The assumptions of cpus being nicely numbered from [0 to X-1],
with X==num_online_cpus() is wrong.

Same remark for num_possible_cpus(), see commit
88d4f0db7fa8 ("perf: Fix alloc_callchain_buffers()") for reference.
Jason Xing Aug. 27, 2021, 12:25 a.m. UTC | #4
On Fri, Aug 27, 2021 at 2:19 AM Eric Dumazet <eric.dumazet@gmail.com> wrote:
>

>

>

> On 8/26/21 10:03 AM, Jason Xing wrote:

>

> >

> > Honestly, I'm a little confused right now. @nr_cpu_ids is the fixed

> > number which means the total number of cpus the machine has.

> > I think, using @nr_cpu_ids is safe one way or the other regardless of

> > whether the cpu goes offline or not. What do you think?

> >

>

> More exactly, nr_cpu_ids is the max number cpu id can reach,

> even in presence of holes.

>

> I think that most/many num_online_cpus() in drivers/net are simply broken

> and should be replaced by nr_cpu_ids.

>


Thank you, Eric, really. I nearly made a terrible mistake.

> The assumptions of cpus being nicely numbered from [0 to X-1],

> with X==num_online_cpus() is wrong.

>

> Same remark for num_possible_cpus(), see commit

> 88d4f0db7fa8 ("perf: Fix alloc_callchain_buffers()") for reference.
diff mbox series

Patch

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index a604552..5f7f181 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -82,6 +82,8 @@ 
 #define IXGBE_2K_TOO_SMALL_WITH_PADDING \
 ((NET_SKB_PAD + IXGBE_RXBUFFER_1536) > SKB_WITH_OVERHEAD(IXGBE_RXBUFFER_2K))
 
+DECLARE_STATIC_KEY_FALSE(ixgbe_xdp_locking_key);
+
 static inline int ixgbe_compute_pad(int rx_buf_len)
 {
 	int page_size, pad_size;
@@ -351,6 +353,7 @@  struct ixgbe_ring {
 	};
 	u16 rx_offset;
 	struct xdp_rxq_info xdp_rxq;
+	spinlock_t tx_lock;	/* used in XDP mode */
 	struct xsk_buff_pool *xsk_pool;
 	u16 ring_idx;		/* {rx,tx,xdp}_ring back reference idx */
 	u16 rx_buf_len;
@@ -375,7 +378,7 @@  enum ixgbe_ring_f_enum {
 #define IXGBE_MAX_FCOE_INDICES		8
 #define MAX_RX_QUEUES			(IXGBE_MAX_FDIR_INDICES + 1)
 #define MAX_TX_QUEUES			(IXGBE_MAX_FDIR_INDICES + 1)
-#define MAX_XDP_QUEUES			(IXGBE_MAX_FDIR_INDICES + 1)
+#define IXGBE_MAX_XDP_QS		(IXGBE_MAX_FDIR_INDICES + 1)
 #define IXGBE_MAX_L2A_QUEUES		4
 #define IXGBE_BAD_L2A_QUEUE		3
 #define IXGBE_MAX_MACVLANS		63
@@ -629,7 +632,7 @@  struct ixgbe_adapter {
 
 	/* XDP */
 	int num_xdp_queues;
-	struct ixgbe_ring *xdp_ring[MAX_XDP_QUEUES];
+	struct ixgbe_ring *xdp_ring[IXGBE_MAX_XDP_QS];
 	unsigned long *af_xdp_zc_qps; /* tracks AF_XDP ZC enabled rings */
 
 	/* TX */
@@ -772,6 +775,14 @@  struct ixgbe_adapter {
 #endif /* CONFIG_IXGBE_IPSEC */
 };
 
+static inline int ixgbe_determine_xdp_q_idx(int cpu)
+{
+	if (static_key_enabled(&ixgbe_xdp_locking_key))
+		return cpu % IXGBE_MAX_XDP_QS;
+	else
+		return cpu;
+}
+
 static inline u8 ixgbe_max_rss_indices(struct ixgbe_adapter *adapter)
 {
 	switch (adapter->hw.mac.type) {
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
index 0218f6c..884bf99 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
@@ -299,7 +299,10 @@  static void ixgbe_cache_ring_register(struct ixgbe_adapter *adapter)
 
 static int ixgbe_xdp_queues(struct ixgbe_adapter *adapter)
 {
-	return adapter->xdp_prog ? nr_cpu_ids : 0;
+	int queues;
+
+	queues = min_t(int, IXGBE_MAX_XDP_QS, num_online_cpus());
+	return adapter->xdp_prog ? queues : 0;
 }
 
 #define IXGBE_RSS_64Q_MASK	0x3F
@@ -947,6 +950,7 @@  static int ixgbe_alloc_q_vector(struct ixgbe_adapter *adapter,
 		ring->count = adapter->tx_ring_count;
 		ring->queue_index = xdp_idx;
 		set_ring_xdp(ring);
+		spin_lock_init(&ring->tx_lock);
 
 		/* assign ring to adapter */
 		WRITE_ONCE(adapter->xdp_ring[xdp_idx], ring);
@@ -1032,6 +1036,9 @@  static void ixgbe_free_q_vector(struct ixgbe_adapter *adapter, int v_idx)
 	adapter->q_vector[v_idx] = NULL;
 	__netif_napi_del(&q_vector->napi);
 
+	if (static_key_enabled(&ixgbe_xdp_locking_key))
+		static_branch_dec(&ixgbe_xdp_locking_key);
+
 	/*
 	 * after a call to __netif_napi_del() napi may still be used and
 	 * ixgbe_get_stats64() might access the rings on this vector,
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 14aea40..a878f40 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -165,6 +165,9 @@  static int ixgbe_notify_dca(struct notifier_block *, unsigned long event,
 MODULE_DESCRIPTION("Intel(R) 10 Gigabit PCI Express Network Driver");
 MODULE_LICENSE("GPL v2");
 
+DEFINE_STATIC_KEY_FALSE(ixgbe_xdp_locking_key);
+EXPORT_SYMBOL(ixgbe_xdp_locking_key);
+
 static struct workqueue_struct *ixgbe_wq;
 
 static bool ixgbe_check_cfg_remove(struct ixgbe_hw *hw, struct pci_dev *pdev);
@@ -2422,13 +2425,10 @@  static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 		xdp_do_flush_map();
 
 	if (xdp_xmit & IXGBE_XDP_TX) {
-		struct ixgbe_ring *ring = adapter->xdp_ring[smp_processor_id()];
+		int index = ixgbe_determine_xdp_q_idx(smp_processor_id());
+		struct ixgbe_ring *ring = adapter->xdp_ring[index];
 
-		/* Force memory writes to complete before letting h/w
-		 * know there are new descriptors to fetch.
-		 */
-		wmb();
-		writel(ring->next_to_use, ring->tail);
+		ixgbe_xdp_ring_update_tail_locked(ring);
 	}
 
 	u64_stats_update_begin(&rx_ring->syncp);
@@ -6320,7 +6320,7 @@  static int ixgbe_sw_init(struct ixgbe_adapter *adapter,
 	if (ixgbe_init_rss_key(adapter))
 		return -ENOMEM;
 
-	adapter->af_xdp_zc_qps = bitmap_zalloc(MAX_XDP_QUEUES, GFP_KERNEL);
+	adapter->af_xdp_zc_qps = bitmap_zalloc(IXGBE_MAX_XDP_QS, GFP_KERNEL);
 	if (!adapter->af_xdp_zc_qps)
 		return -ENOMEM;
 
@@ -8539,21 +8539,32 @@  static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb,
 int ixgbe_xmit_xdp_ring(struct ixgbe_adapter *adapter,
 			struct xdp_frame *xdpf)
 {
-	struct ixgbe_ring *ring = adapter->xdp_ring[smp_processor_id()];
 	struct ixgbe_tx_buffer *tx_buffer;
 	union ixgbe_adv_tx_desc *tx_desc;
+	struct ixgbe_ring *ring;
 	u32 len, cmd_type;
 	dma_addr_t dma;
+	int index, ret;
 	u16 i;
 
 	len = xdpf->len;
 
-	if (unlikely(!ixgbe_desc_unused(ring)))
-		return IXGBE_XDP_CONSUMED;
+	index = ixgbe_determine_xdp_q_idx(smp_processor_id());
+	ring = adapter->xdp_ring[index];
+
+	if (static_branch_unlikely(&ixgbe_xdp_locking_key))
+		spin_lock(&ring->tx_lock);
+
+	if (unlikely(!ixgbe_desc_unused(ring))) {
+		ret = IXGBE_XDP_CONSUMED;
+		goto out;
+	}
 
 	dma = dma_map_single(ring->dev, xdpf->data, len, DMA_TO_DEVICE);
-	if (dma_mapping_error(ring->dev, dma))
-		return IXGBE_XDP_CONSUMED;
+	if (dma_mapping_error(ring->dev, dma)) {
+		ret = IXGBE_XDP_CONSUMED;
+		goto out;
+	}
 
 	/* record the location of the first descriptor for this packet */
 	tx_buffer = &ring->tx_buffer_info[ring->next_to_use];
@@ -8590,7 +8601,11 @@  int ixgbe_xmit_xdp_ring(struct ixgbe_adapter *adapter,
 	tx_buffer->next_to_watch = tx_desc;
 	ring->next_to_use = i;
 
-	return IXGBE_XDP_TX;
+	ret = IXGBE_XDP_TX;
+out:
+	if (static_branch_unlikely(&ixgbe_xdp_locking_key))
+		spin_unlock(&ring->tx_lock);
+	return ret;
 }
 
 netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb,
@@ -10130,8 +10145,13 @@  static int ixgbe_xdp_setup(struct net_device *dev, struct bpf_prog *prog)
 			return -EINVAL;
 	}
 
-	if (nr_cpu_ids > MAX_XDP_QUEUES)
+	/* if the number of cpus is much larger than the maximum of queues,
+	 * we should stop it and then return with NOMEM like before.
+	 */
+	if (num_online_cpus() > IXGBE_MAX_XDP_QS * 2)
 		return -ENOMEM;
+	else if (num_online_cpus() > IXGBE_MAX_XDP_QS)
+		static_branch_inc(&ixgbe_xdp_locking_key);
 
 	old_prog = xchg(&adapter->xdp_prog, prog);
 	need_reset = (!!prog != !!old_prog);
@@ -10195,12 +10215,22 @@  void ixgbe_xdp_ring_update_tail(struct ixgbe_ring *ring)
 	writel(ring->next_to_use, ring->tail);
 }
 
+void ixgbe_xdp_ring_update_tail_locked(struct ixgbe_ring *ring)
+{
+	if (static_branch_unlikely(&ixgbe_xdp_locking_key))
+		spin_lock(&ring->tx_lock);
+	ixgbe_xdp_ring_update_tail(ring);
+	if (static_branch_unlikely(&ixgbe_xdp_locking_key))
+		spin_unlock(&ring->tx_lock);
+}
+
 static int ixgbe_xdp_xmit(struct net_device *dev, int n,
 			  struct xdp_frame **frames, u32 flags)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(dev);
 	struct ixgbe_ring *ring;
 	int nxmit = 0;
+	int index;
 	int i;
 
 	if (unlikely(test_bit(__IXGBE_DOWN, &adapter->state)))
@@ -10209,10 +10239,12 @@  static int ixgbe_xdp_xmit(struct net_device *dev, int n,
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
 		return -EINVAL;
 
+	index = ixgbe_determine_xdp_q_idx(smp_processor_id());
+
 	/* During program transitions its possible adapter->xdp_prog is assigned
 	 * but ring has not been configured yet. In this case simply abort xmit.
 	 */
-	ring = adapter->xdp_prog ? adapter->xdp_ring[smp_processor_id()] : NULL;
+	ring = adapter->xdp_prog ? adapter->xdp_ring[index] : NULL;
 	if (unlikely(!ring))
 		return -ENXIO;
 
@@ -10230,7 +10262,7 @@  static int ixgbe_xdp_xmit(struct net_device *dev, int n,
 	}
 
 	if (unlikely(flags & XDP_XMIT_FLUSH))
-		ixgbe_xdp_ring_update_tail(ring);
+		ixgbe_xdp_ring_update_tail_locked(ring);
 
 	return nxmit;
 }
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h
index 2aeec78..f6426d9 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h
@@ -23,6 +23,7 @@  void ixgbe_process_skb_fields(struct ixgbe_ring *rx_ring,
 void ixgbe_rx_skb(struct ixgbe_q_vector *q_vector,
 		  struct sk_buff *skb);
 void ixgbe_xdp_ring_update_tail(struct ixgbe_ring *ring);
+void ixgbe_xdp_ring_update_tail_locked(struct ixgbe_ring *ring);
 void ixgbe_irq_rearm_queues(struct ixgbe_adapter *adapter, u64 qmask);
 
 void ixgbe_txrx_ring_disable(struct ixgbe_adapter *adapter, int ring);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
index b1d22e4..82d00e4 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
@@ -334,13 +334,10 @@  int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector,
 		xdp_do_flush_map();
 
 	if (xdp_xmit & IXGBE_XDP_TX) {
-		struct ixgbe_ring *ring = adapter->xdp_ring[smp_processor_id()];
+		int index = ixgbe_determine_xdp_q_idx(smp_processor_id());
+		struct ixgbe_ring *ring = adapter->xdp_ring[index];
 
-		/* Force memory writes to complete before letting h/w
-		 * know there are new descriptors to fetch.
-		 */
-		wmb();
-		writel(ring->next_to_use, ring->tail);
+		ixgbe_xdp_ring_update_tail_locked(ring);
 	}
 
 	u64_stats_update_begin(&rx_ring->syncp);