On Mon, Jun 17, 2024 at 5:18 PM Jiri Pirko [off-list ref] wrote:

Mon, Jun 17, 2024 at 04:34:26AM CEST, jasowang@redhat.com wrote:

quoted

On Thu, Jun 13, 2024 at 1:09 AM Jiri Pirko [off-list ref] wrote:

quoted

From: Jiri Pirko <redacted>

Add support for Byte Queue Limits (BQL).

Tested on qemu emulated virtio_net device with 1, 2 and 4 queues.
Tested with fq_codel and pfifo_fast. Super netperf with 50 threads is
running in background. Netperf TCP_RR results:

NOBQL FQC 1q:  159.56  159.33  158.50  154.31    agv: 157.925
NOBQL FQC 2q:  184.64  184.96  174.73  174.15    agv: 179.62
NOBQL FQC 4q:  994.46  441.96  416.50  499.56    agv: 588.12
NOBQL PFF 1q:  148.68  148.92  145.95  149.48    agv: 148.2575
NOBQL PFF 2q:  171.86  171.20  170.42  169.42    agv: 170.725
NOBQL PFF 4q: 1505.23 1137.23 2488.70 3507.99    agv: 2159.7875
  BQL FQC 1q: 1332.80 1297.97 1351.41 1147.57    agv: 1282.4375
  BQL FQC 2q:  768.30  817.72  864.43  974.40    agv: 856.2125
  BQL FQC 4q:  945.66  942.68  878.51  822.82    agv: 897.4175
  BQL PFF 1q:  149.69  151.49  149.40  147.47    agv: 149.5125
  BQL PFF 2q: 2059.32  798.74 1844.12  381.80    agv: 1270.995
  BQL PFF 4q: 1871.98 4420.02 4916.59 13268.16   agv: 6119.1875

Signed-off-by: Jiri Pirko <redacted>
---
v1->v2:
- moved netdev_tx_completed_queue() call into __free_old_xmit(),
  propagate use_napi flag to __free_old_xmit() and only call
  netdev_tx_completed_queue() in case it is true
- added forgotten call to netdev_tx_reset_queue()
- fixed stats for xdp packets
- fixed bql accounting when __free_old_xmit() is called from xdp path
- handle the !use_napi case in start_xmit() kick section
---
 drivers/net/virtio_net.c | 50 +++++++++++++++++++++++++---------------
 1 file changed, 32 insertions(+), 18 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 61a57d134544..5863c663ccab 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c

@@ -84,7 +84,9 @@ struct virtnet_stat_desc {

 struct virtnet_sq_free_stats {
        u64 packets;
+       u64 xdp_packets;
        u64 bytes;
+       u64 xdp_bytes;
 };

 struct virtnet_sq_stats {

@@ -506,29 +508,33 @@ static struct xdp_frame *ptr_to_xdp(void *ptr)
        return (struct xdp_frame *)((unsigned long)ptr & ~VIRTIO_XDP_FLAG);
 }

-static void __free_old_xmit(struct send_queue *sq, bool in_napi,
+static void __free_old_xmit(struct send_queue *sq, struct netdev_queue *txq,
+                           bool in_napi, bool use_napi,
                            struct virtnet_sq_free_stats *stats)
 {
        unsigned int len;
        void *ptr;

        while ((ptr = virtqueue_get_buf(sq->vq, &len)) != NULL) {
-               ++stats->packets;
-
                if (!is_xdp_frame(ptr)) {
                        struct sk_buff *skb = ptr;

                        pr_debug("Sent skb %p\n", skb);

+                       stats->packets++;
                        stats->bytes += skb->len;
                        napi_consume_skb(skb, in_napi);
                } else {
                        struct xdp_frame *frame = ptr_to_xdp(ptr);

-                       stats->bytes += xdp_get_frame_len(frame);
+                       stats->xdp_packets++;
+                       stats->xdp_bytes += xdp_get_frame_len(frame);
                        xdp_return_frame(frame);
                }
        }
+       if (use_napi)
+               netdev_tx_completed_queue(txq, stats->packets, stats->bytes);
+
 }

I wonder if this works correctly, for example NAPI could be enabled
after queued but before sent. So __netdev_tx_sent_queue() is not
called before.

How is that possible? Napi weight can't change when link is up. Or am I
missing something?

Something like this:

1) packet were queued
2) if down
3) enable NAPI
4) if up
5) packet were sent

?

Thanks

quoted

Thanks