* [PATCH net-next 0/2] Add TX stop/wake counters
@ 2024-05-09 16:32 Daniel Jurgens
0 siblings, 0 replies; 7+ messages in thread
From: Daniel Jurgens @ 2024-05-09 16:32 UTC (permalink / raw)
To: netdev
Cc: mst, jasowang, xuanzhuo, virtualization, davem, edumazet, kuba,
pabeni, jiri, Daniel Jurgens
Several drivers provide TX stop and wake counters via ethtool stats. Add
those to the netdev queue stats, and use them in virtio_net.
Daniel Jurgens (2):
netdev: Add queue stats for TX stop and wake
virtio_net: Add TX stopped and wake counters
Documentation/netlink/specs/netdev.yaml | 10 +++++++++
drivers/net/virtio_net.c | 28 +++++++++++++++++++++++--
include/net/netdev_queues.h | 3 +++
include/uapi/linux/netdev.h | 2 ++
net/core/netdev-genl.c | 4 +++-
tools/include/uapi/linux/netdev.h | 3 ++-
6 files changed, 46 insertions(+), 4 deletions(-)
--
2.44.0
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH net-next 0/2] Add TX stop/wake counters
@ 2024-05-10 20:19 Daniel Jurgens
2024-05-10 20:19 ` [PATCH net-next v2 1/2] netdev: Add queue stats for TX stop and wake Daniel Jurgens
` (2 more replies)
0 siblings, 3 replies; 7+ messages in thread
From: Daniel Jurgens @ 2024-05-10 20:19 UTC (permalink / raw)
To: netdev
Cc: mst, jasowang, xuanzhuo, virtualization, davem, edumazet, kuba,
pabeni, jiri, Daniel Jurgens
Several drivers provide TX stop and wake counters via ethtool stats. Add
those to the netdev queue stats, and use them in virtio_net.
v2:
- Fixed an accidental line deletion
- Enhanced documentation
Daniel Jurgens (2):
netdev: Add queue stats for TX stop and wake
virtio_net: Add TX stopped and wake counters
Documentation/netlink/specs/netdev.yaml | 14 +++++++++++++
drivers/net/virtio_net.c | 28 +++++++++++++++++++++++--
include/net/netdev_queues.h | 3 +++
include/uapi/linux/netdev.h | 2 ++
net/core/netdev-genl.c | 4 +++-
tools/include/uapi/linux/netdev.h | 2 ++
6 files changed, 50 insertions(+), 3 deletions(-)
--
2.45.0
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH net-next v2 1/2] netdev: Add queue stats for TX stop and wake
2024-05-10 20:19 [PATCH net-next 0/2] Add TX stop/wake counters Daniel Jurgens
@ 2024-05-10 20:19 ` Daniel Jurgens
2024-05-11 3:31 ` Jason Xing
2024-05-10 20:19 ` [PATCH net-next v2 2/2] virtio_net: Add TX stopped and wake counters Daniel Jurgens
2024-05-13 22:50 ` [PATCH net-next 0/2] Add TX stop/wake counters patchwork-bot+netdevbpf
2 siblings, 1 reply; 7+ messages in thread
From: Daniel Jurgens @ 2024-05-10 20:19 UTC (permalink / raw)
To: netdev
Cc: mst, jasowang, xuanzhuo, virtualization, davem, edumazet, kuba,
pabeni, jiri, Daniel Jurgens
TX queue stop and wake are counted by some drivers.
Support reporting these via netdev-genl queue stats.
Signed-off-by: Daniel Jurgens <danielj@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
---
Documentation/netlink/specs/netdev.yaml | 14 ++++++++++++++
include/net/netdev_queues.h | 3 +++
include/uapi/linux/netdev.h | 2 ++
net/core/netdev-genl.c | 4 +++-
tools/include/uapi/linux/netdev.h | 2 ++
5 files changed, 24 insertions(+), 1 deletion(-)
diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml
index 2be4b3714d17..11a32373365a 100644
--- a/Documentation/netlink/specs/netdev.yaml
+++ b/Documentation/netlink/specs/netdev.yaml
@@ -439,6 +439,20 @@ attribute-sets:
Number of the packets dropped by the device due to the transmit
packets bitrate exceeding the device rate limit.
type: uint
+ -
+ name: tx-stop
+ doc: |
+ Number of times driver paused accepting new tx packets
+ from the stack to this queue, because the queue was full.
+ Note that if BQL is supported and enabled on the device
+ the networking stack will avoid queuing a lot of data at once.
+ type: uint
+ -
+ name: tx-wake
+ doc: |
+ Number of times driver re-started accepting send
+ requests to this queue from the stack.
+ type: uint
operations:
list:
diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h
index e7b84f018cee..a8a7e48dfa6c 100644
--- a/include/net/netdev_queues.h
+++ b/include/net/netdev_queues.h
@@ -41,6 +41,9 @@ struct netdev_queue_stats_tx {
u64 hw_gso_wire_bytes;
u64 hw_drop_ratelimits;
+
+ u64 stop;
+ u64 wake;
};
/**
diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h
index cf24f1d9adf8..a8188202413e 100644
--- a/include/uapi/linux/netdev.h
+++ b/include/uapi/linux/netdev.h
@@ -165,6 +165,8 @@ enum {
NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS,
NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES,
NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS,
+ NETDEV_A_QSTATS_TX_STOP,
+ NETDEV_A_QSTATS_TX_WAKE,
__NETDEV_A_QSTATS_MAX,
NETDEV_A_QSTATS_MAX = (__NETDEV_A_QSTATS_MAX - 1)
diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c
index 4b5054087309..1f6ae6379e0f 100644
--- a/net/core/netdev-genl.c
+++ b/net/core/netdev-genl.c
@@ -517,7 +517,9 @@ netdev_nl_stats_write_tx(struct sk_buff *rsp, struct netdev_queue_stats_tx *tx)
netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_BYTES, tx->hw_gso_bytes) ||
netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS, tx->hw_gso_wire_packets) ||
netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES, tx->hw_gso_wire_bytes) ||
- netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, tx->hw_drop_ratelimits))
+ netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, tx->hw_drop_ratelimits) ||
+ netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_STOP, tx->stop) ||
+ netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_WAKE, tx->wake))
return -EMSGSIZE;
return 0;
}
diff --git a/tools/include/uapi/linux/netdev.h b/tools/include/uapi/linux/netdev.h
index cf24f1d9adf8..a8188202413e 100644
--- a/tools/include/uapi/linux/netdev.h
+++ b/tools/include/uapi/linux/netdev.h
@@ -165,6 +165,8 @@ enum {
NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS,
NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES,
NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS,
+ NETDEV_A_QSTATS_TX_STOP,
+ NETDEV_A_QSTATS_TX_WAKE,
__NETDEV_A_QSTATS_MAX,
NETDEV_A_QSTATS_MAX = (__NETDEV_A_QSTATS_MAX - 1)
--
2.45.0
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH net-next v2 2/2] virtio_net: Add TX stopped and wake counters
2024-05-10 20:19 [PATCH net-next 0/2] Add TX stop/wake counters Daniel Jurgens
2024-05-10 20:19 ` [PATCH net-next v2 1/2] netdev: Add queue stats for TX stop and wake Daniel Jurgens
@ 2024-05-10 20:19 ` Daniel Jurgens
2024-05-11 3:38 ` Jason Xing
2024-05-13 22:50 ` [PATCH net-next 0/2] Add TX stop/wake counters patchwork-bot+netdevbpf
2 siblings, 1 reply; 7+ messages in thread
From: Daniel Jurgens @ 2024-05-10 20:19 UTC (permalink / raw)
To: netdev
Cc: mst, jasowang, xuanzhuo, virtualization, davem, edumazet, kuba,
pabeni, jiri, Daniel Jurgens
Add a tx queue stop and wake counters, they are useful for debugging.
$ ./tools/net/ynl/cli.py --spec netlink/specs/netdev.yaml \
--dump qstats-get --json '{"scope": "queue"}'
...
{'ifindex': 13,
'queue-id': 0,
'queue-type': 'tx',
'tx-bytes': 14756682850,
'tx-packets': 226465,
'tx-stop': 113208,
'tx-wake': 113208},
{'ifindex': 13,
'queue-id': 1,
'queue-type': 'tx',
'tx-bytes': 18167675008,
'tx-packets': 278660,
'tx-stop': 8632,
'tx-wake': 8632}]
Signed-off-by: Daniel Jurgens <danielj@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
---
drivers/net/virtio_net.c | 28 ++++++++++++++++++++++++++--
1 file changed, 26 insertions(+), 2 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 218a446c4c27..df6121c38a1b 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -95,6 +95,8 @@ struct virtnet_sq_stats {
u64_stats_t xdp_tx_drops;
u64_stats_t kicks;
u64_stats_t tx_timeouts;
+ u64_stats_t stop;
+ u64_stats_t wake;
};
struct virtnet_rq_stats {
@@ -145,6 +147,8 @@ static const struct virtnet_stat_desc virtnet_rq_stats_desc[] = {
static const struct virtnet_stat_desc virtnet_sq_stats_desc_qstat[] = {
VIRTNET_SQ_STAT_QSTAT("packets", packets),
VIRTNET_SQ_STAT_QSTAT("bytes", bytes),
+ VIRTNET_SQ_STAT_QSTAT("stop", stop),
+ VIRTNET_SQ_STAT_QSTAT("wake", wake),
};
static const struct virtnet_stat_desc virtnet_rq_stats_desc_qstat[] = {
@@ -1014,6 +1018,9 @@ static void check_sq_full_and_disable(struct virtnet_info *vi,
*/
if (sq->vq->num_free < 2+MAX_SKB_FRAGS) {
netif_stop_subqueue(dev, qnum);
+ u64_stats_update_begin(&sq->stats.syncp);
+ u64_stats_inc(&sq->stats.stop);
+ u64_stats_update_end(&sq->stats.syncp);
if (use_napi) {
if (unlikely(!virtqueue_enable_cb_delayed(sq->vq)))
virtqueue_napi_schedule(&sq->napi, sq->vq);
@@ -1022,6 +1029,9 @@ static void check_sq_full_and_disable(struct virtnet_info *vi,
free_old_xmit(sq, false);
if (sq->vq->num_free >= 2+MAX_SKB_FRAGS) {
netif_start_subqueue(dev, qnum);
+ u64_stats_update_begin(&sq->stats.syncp);
+ u64_stats_inc(&sq->stats.wake);
+ u64_stats_update_end(&sq->stats.syncp);
virtqueue_disable_cb(sq->vq);
}
}
@@ -2322,8 +2332,14 @@ static void virtnet_poll_cleantx(struct receive_queue *rq)
free_old_xmit(sq, true);
} while (unlikely(!virtqueue_enable_cb_delayed(sq->vq)));
- if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS)
+ if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS) {
+ if (netif_tx_queue_stopped(txq)) {
+ u64_stats_update_begin(&sq->stats.syncp);
+ u64_stats_inc(&sq->stats.wake);
+ u64_stats_update_end(&sq->stats.syncp);
+ }
netif_tx_wake_queue(txq);
+ }
__netif_tx_unlock(txq);
}
@@ -2473,8 +2489,14 @@ static int virtnet_poll_tx(struct napi_struct *napi, int budget)
virtqueue_disable_cb(sq->vq);
free_old_xmit(sq, true);
- if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS)
+ if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS) {
+ if (netif_tx_queue_stopped(txq)) {
+ u64_stats_update_begin(&sq->stats.syncp);
+ u64_stats_inc(&sq->stats.wake);
+ u64_stats_update_end(&sq->stats.syncp);
+ }
netif_tx_wake_queue(txq);
+ }
opaque = virtqueue_enable_cb_prepare(sq->vq);
@@ -4790,6 +4812,8 @@ static void virtnet_get_base_stats(struct net_device *dev,
tx->bytes = 0;
tx->packets = 0;
+ tx->stop = 0;
+ tx->wake = 0;
if (vi->device_stats_cap & VIRTIO_NET_STATS_TYPE_TX_BASIC) {
tx->hw_drops = 0;
--
2.45.0
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH net-next v2 1/2] netdev: Add queue stats for TX stop and wake
2024-05-10 20:19 ` [PATCH net-next v2 1/2] netdev: Add queue stats for TX stop and wake Daniel Jurgens
@ 2024-05-11 3:31 ` Jason Xing
0 siblings, 0 replies; 7+ messages in thread
From: Jason Xing @ 2024-05-11 3:31 UTC (permalink / raw)
To: Daniel Jurgens
Cc: netdev, mst, jasowang, xuanzhuo, virtualization, davem, edumazet,
kuba, pabeni, jiri
On Sat, May 11, 2024 at 4:20 AM Daniel Jurgens <danielj@nvidia.com> wrote:
>
> TX queue stop and wake are counted by some drivers.
> Support reporting these via netdev-genl queue stats.
>
> Signed-off-by: Daniel Jurgens <danielj@nvidia.com>
> Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Good to see this finally :)
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
> ---
> Documentation/netlink/specs/netdev.yaml | 14 ++++++++++++++
> include/net/netdev_queues.h | 3 +++
> include/uapi/linux/netdev.h | 2 ++
> net/core/netdev-genl.c | 4 +++-
> tools/include/uapi/linux/netdev.h | 2 ++
> 5 files changed, 24 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml
> index 2be4b3714d17..11a32373365a 100644
> --- a/Documentation/netlink/specs/netdev.yaml
> +++ b/Documentation/netlink/specs/netdev.yaml
> @@ -439,6 +439,20 @@ attribute-sets:
> Number of the packets dropped by the device due to the transmit
> packets bitrate exceeding the device rate limit.
> type: uint
> + -
> + name: tx-stop
> + doc: |
> + Number of times driver paused accepting new tx packets
> + from the stack to this queue, because the queue was full.
> + Note that if BQL is supported and enabled on the device
> + the networking stack will avoid queuing a lot of data at once.
> + type: uint
> + -
> + name: tx-wake
> + doc: |
> + Number of times driver re-started accepting send
> + requests to this queue from the stack.
> + type: uint
>
> operations:
> list:
> diff --git a/include/net/netdev_queues.h b/include/net/netdev_queues.h
> index e7b84f018cee..a8a7e48dfa6c 100644
> --- a/include/net/netdev_queues.h
> +++ b/include/net/netdev_queues.h
> @@ -41,6 +41,9 @@ struct netdev_queue_stats_tx {
> u64 hw_gso_wire_bytes;
>
> u64 hw_drop_ratelimits;
> +
> + u64 stop;
> + u64 wake;
> };
>
> /**
> diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h
> index cf24f1d9adf8..a8188202413e 100644
> --- a/include/uapi/linux/netdev.h
> +++ b/include/uapi/linux/netdev.h
> @@ -165,6 +165,8 @@ enum {
> NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS,
> NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES,
> NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS,
> + NETDEV_A_QSTATS_TX_STOP,
> + NETDEV_A_QSTATS_TX_WAKE,
>
> __NETDEV_A_QSTATS_MAX,
> NETDEV_A_QSTATS_MAX = (__NETDEV_A_QSTATS_MAX - 1)
> diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c
> index 4b5054087309..1f6ae6379e0f 100644
> --- a/net/core/netdev-genl.c
> +++ b/net/core/netdev-genl.c
> @@ -517,7 +517,9 @@ netdev_nl_stats_write_tx(struct sk_buff *rsp, struct netdev_queue_stats_tx *tx)
> netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_BYTES, tx->hw_gso_bytes) ||
> netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS, tx->hw_gso_wire_packets) ||
> netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES, tx->hw_gso_wire_bytes) ||
> - netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, tx->hw_drop_ratelimits))
> + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS, tx->hw_drop_ratelimits) ||
> + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_STOP, tx->stop) ||
> + netdev_stat_put(rsp, NETDEV_A_QSTATS_TX_WAKE, tx->wake))
> return -EMSGSIZE;
> return 0;
> }
> diff --git a/tools/include/uapi/linux/netdev.h b/tools/include/uapi/linux/netdev.h
> index cf24f1d9adf8..a8188202413e 100644
> --- a/tools/include/uapi/linux/netdev.h
> +++ b/tools/include/uapi/linux/netdev.h
> @@ -165,6 +165,8 @@ enum {
> NETDEV_A_QSTATS_TX_HW_GSO_WIRE_PACKETS,
> NETDEV_A_QSTATS_TX_HW_GSO_WIRE_BYTES,
> NETDEV_A_QSTATS_TX_HW_DROP_RATELIMITS,
> + NETDEV_A_QSTATS_TX_STOP,
> + NETDEV_A_QSTATS_TX_WAKE,
>
> __NETDEV_A_QSTATS_MAX,
> NETDEV_A_QSTATS_MAX = (__NETDEV_A_QSTATS_MAX - 1)
> --
> 2.45.0
>
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH net-next v2 2/2] virtio_net: Add TX stopped and wake counters
2024-05-10 20:19 ` [PATCH net-next v2 2/2] virtio_net: Add TX stopped and wake counters Daniel Jurgens
@ 2024-05-11 3:38 ` Jason Xing
0 siblings, 0 replies; 7+ messages in thread
From: Jason Xing @ 2024-05-11 3:38 UTC (permalink / raw)
To: Daniel Jurgens
Cc: netdev, mst, jasowang, xuanzhuo, virtualization, davem, edumazet,
kuba, pabeni, jiri
On Sat, May 11, 2024 at 4:20 AM Daniel Jurgens <danielj@nvidia.com> wrote:
>
> Add a tx queue stop and wake counters, they are useful for debugging.
>
> $ ./tools/net/ynl/cli.py --spec netlink/specs/netdev.yaml \
> --dump qstats-get --json '{"scope": "queue"}'
> ...
> {'ifindex': 13,
> 'queue-id': 0,
> 'queue-type': 'tx',
> 'tx-bytes': 14756682850,
> 'tx-packets': 226465,
> 'tx-stop': 113208,
> 'tx-wake': 113208},
> {'ifindex': 13,
> 'queue-id': 1,
> 'queue-type': 'tx',
> 'tx-bytes': 18167675008,
> 'tx-packets': 278660,
> 'tx-stop': 8632,
> 'tx-wake': 8632}]
>
> Signed-off-by: Daniel Jurgens <danielj@nvidia.com>
> Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Thanks!
> ---
> drivers/net/virtio_net.c | 28 ++++++++++++++++++++++++++--
> 1 file changed, 26 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 218a446c4c27..df6121c38a1b 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -95,6 +95,8 @@ struct virtnet_sq_stats {
> u64_stats_t xdp_tx_drops;
> u64_stats_t kicks;
> u64_stats_t tx_timeouts;
> + u64_stats_t stop;
> + u64_stats_t wake;
> };
>
> struct virtnet_rq_stats {
> @@ -145,6 +147,8 @@ static const struct virtnet_stat_desc virtnet_rq_stats_desc[] = {
> static const struct virtnet_stat_desc virtnet_sq_stats_desc_qstat[] = {
> VIRTNET_SQ_STAT_QSTAT("packets", packets),
> VIRTNET_SQ_STAT_QSTAT("bytes", bytes),
> + VIRTNET_SQ_STAT_QSTAT("stop", stop),
> + VIRTNET_SQ_STAT_QSTAT("wake", wake),
> };
>
> static const struct virtnet_stat_desc virtnet_rq_stats_desc_qstat[] = {
> @@ -1014,6 +1018,9 @@ static void check_sq_full_and_disable(struct virtnet_info *vi,
> */
> if (sq->vq->num_free < 2+MAX_SKB_FRAGS) {
> netif_stop_subqueue(dev, qnum);
> + u64_stats_update_begin(&sq->stats.syncp);
> + u64_stats_inc(&sq->stats.stop);
> + u64_stats_update_end(&sq->stats.syncp);
> if (use_napi) {
> if (unlikely(!virtqueue_enable_cb_delayed(sq->vq)))
> virtqueue_napi_schedule(&sq->napi, sq->vq);
> @@ -1022,6 +1029,9 @@ static void check_sq_full_and_disable(struct virtnet_info *vi,
> free_old_xmit(sq, false);
> if (sq->vq->num_free >= 2+MAX_SKB_FRAGS) {
> netif_start_subqueue(dev, qnum);
> + u64_stats_update_begin(&sq->stats.syncp);
> + u64_stats_inc(&sq->stats.wake);
> + u64_stats_update_end(&sq->stats.syncp);
> virtqueue_disable_cb(sq->vq);
> }
> }
> @@ -2322,8 +2332,14 @@ static void virtnet_poll_cleantx(struct receive_queue *rq)
> free_old_xmit(sq, true);
> } while (unlikely(!virtqueue_enable_cb_delayed(sq->vq)));
>
> - if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS)
> + if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS) {
> + if (netif_tx_queue_stopped(txq)) {
> + u64_stats_update_begin(&sq->stats.syncp);
> + u64_stats_inc(&sq->stats.wake);
> + u64_stats_update_end(&sq->stats.syncp);
> + }
> netif_tx_wake_queue(txq);
> + }
>
> __netif_tx_unlock(txq);
> }
> @@ -2473,8 +2489,14 @@ static int virtnet_poll_tx(struct napi_struct *napi, int budget)
> virtqueue_disable_cb(sq->vq);
> free_old_xmit(sq, true);
>
> - if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS)
> + if (sq->vq->num_free >= 2 + MAX_SKB_FRAGS) {
> + if (netif_tx_queue_stopped(txq)) {
> + u64_stats_update_begin(&sq->stats.syncp);
> + u64_stats_inc(&sq->stats.wake);
> + u64_stats_update_end(&sq->stats.syncp);
> + }
> netif_tx_wake_queue(txq);
> + }
>
> opaque = virtqueue_enable_cb_prepare(sq->vq);
>
> @@ -4790,6 +4812,8 @@ static void virtnet_get_base_stats(struct net_device *dev,
>
> tx->bytes = 0;
> tx->packets = 0;
> + tx->stop = 0;
> + tx->wake = 0;
>
> if (vi->device_stats_cap & VIRTIO_NET_STATS_TYPE_TX_BASIC) {
> tx->hw_drops = 0;
> --
> 2.45.0
>
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH net-next 0/2] Add TX stop/wake counters
2024-05-10 20:19 [PATCH net-next 0/2] Add TX stop/wake counters Daniel Jurgens
2024-05-10 20:19 ` [PATCH net-next v2 1/2] netdev: Add queue stats for TX stop and wake Daniel Jurgens
2024-05-10 20:19 ` [PATCH net-next v2 2/2] virtio_net: Add TX stopped and wake counters Daniel Jurgens
@ 2024-05-13 22:50 ` patchwork-bot+netdevbpf
2 siblings, 0 replies; 7+ messages in thread
From: patchwork-bot+netdevbpf @ 2024-05-13 22:50 UTC (permalink / raw)
To: Dan Jurgens
Cc: netdev, mst, jasowang, xuanzhuo, virtualization, davem, edumazet,
kuba, pabeni, jiri
Hello:
This series was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:
On Fri, 10 May 2024 23:19:25 +0300 you wrote:
> Several drivers provide TX stop and wake counters via ethtool stats. Add
> those to the netdev queue stats, and use them in virtio_net.
>
> v2:
> - Fixed an accidental line deletion
> - Enhanced documentation
>
> [...]
Here is the summary with links:
- [net-next,v2,1/2] netdev: Add queue stats for TX stop and wake
https://git.kernel.org/netdev/net-next/c/b56035101e1c
- [net-next,v2,2/2] virtio_net: Add TX stopped and wake counters
https://git.kernel.org/netdev/net-next/c/c39add9b2423
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2024-05-13 22:50 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-05-10 20:19 [PATCH net-next 0/2] Add TX stop/wake counters Daniel Jurgens
2024-05-10 20:19 ` [PATCH net-next v2 1/2] netdev: Add queue stats for TX stop and wake Daniel Jurgens
2024-05-11 3:31 ` Jason Xing
2024-05-10 20:19 ` [PATCH net-next v2 2/2] virtio_net: Add TX stopped and wake counters Daniel Jurgens
2024-05-11 3:38 ` Jason Xing
2024-05-13 22:50 ` [PATCH net-next 0/2] Add TX stop/wake counters patchwork-bot+netdevbpf
-- strict thread matches above, loose matches on Subject: below --
2024-05-09 16:32 Daniel Jurgens
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).