On 25/02/2026 18:49, Long Li wrote:
MANA hardware requires at least one doorbell ring every 8 wraparounds
of the CQ. The driver rings the doorbell as a form of flow control to
inform hardware that CQEs have been consumed.
The NAPI poll functions mana_poll_tx_cq() and mana_poll_rx_cq() can
poll up to CQE_POLLING_BUFFER (512) completions per call. If the CQ
has fewer than 512 entries, a single poll call can process more than
4 wraparounds without ringing the doorbell. The doorbell threshold
check also uses ">" instead of ">=", delaying the ring by one extra
CQE beyond 4 wraparounds. Combined, these issues can cause the driver
to exceed the 8-wraparound hardware limit, leading to missed
completions and stalled queues.
Fix this by capping the number of CQEs polled per call to 4 wraparounds
of the CQ in both TX and RX paths. Also change the doorbell threshold
from ">" to ">=" so the doorbell is rung as soon as 4 wraparounds are
reached.
Cc: [email protected]
Fixes: 58a63729c957 ("net: mana: Fix doorbell out of order violation and avoid
unnecessary doorbell rings")
Signed-off-by: Long Li <[email protected]>
---
drivers/net/ethernet/microsoft/mana/mana_en.c | 23 +++++++++++++++----
1 file changed, 18 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c
b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 9919183ad39e..fe667e0d930d 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -1770,8 +1770,14 @@ static void mana_poll_tx_cq(struct mana_cq *cq)
ndev = txq->ndev;
apc = netdev_priv(ndev);
+ /* Limit CQEs polled to 4 wraparounds of the CQ to ensure the
+ * doorbell can be rung in time for the hardware's requirement
+ * of at least one doorbell ring every 8 wraparounds.
+ */
comp_read = mana_gd_poll_cq(cq->gdma_cq, completions,
- CQE_POLLING_BUFFER);
+ min_t(u32, (cq->gdma_cq->queue_size /
no need for min_t, simple min() can be used, queue_size is already u32
+ COMP_ENTRY_SIZE) * 4,
+ CQE_POLLING_BUFFER));
if (comp_read < 1)
return;
@@ -2156,7 +2162,14 @@ static void mana_poll_rx_cq(struct mana_cq *cq)
struct mana_rxq *rxq = cq->rxq;
int comp_read, i;
- comp_read = mana_gd_poll_cq(cq->gdma_cq, comp, CQE_POLLING_BUFFER);
+ /* Limit CQEs polled to 4 wraparounds of the CQ to ensure the
+ * doorbell can be rung in time for the hardware's requirement
+ * of at least one doorbell ring every 8 wraparounds.
+ */
+ comp_read = mana_gd_poll_cq(cq->gdma_cq, comp,
+ min_t(u32, (cq->gdma_cq->queue_size /
same here
+ COMP_ENTRY_SIZE) * 4,
+ CQE_POLLING_BUFFER));
WARN_ON_ONCE(comp_read > CQE_POLLING_BUFFER);
rxq->xdp_flush = false;
