[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <1374769283-18580-1-git-send-email-amirv@mellanox.com>
Date: Thu, 25 Jul 2013 19:21:23 +0300
From: Amir Vadai <amirv@...lanox.com>
To: "David S. Miller" <davem@...emloft.net>
Cc: netdev@...r.kernel.org, Amir Vadai <amirv@...lanox.com>,
Eugenia Emantayev <eugenia@...lanox.com>
Subject: [PATCH net-next] net/mlx4_en: Fix BlueFlame race
From: Eugenia Emantayev <eugenia@...lanox.com>
Fix a race between BlueFlame flow and stamping in post send flow.
Example:
SW: Build WQE 0 on the TX buffer, except the ownership bit
SW: Set ownership for WQE 0 on the TX buffer
SW: Ring doorbell for WQE 0
SW: Build WQE 1 on the TX buffer, except the ownership bit
SW: Set ownership for WQE 1 on the TX buffer
HW: Read WQE 0 and then WQE 1, before doorbell was rung/BF was done for WQE 1
HW: Produce CQEs for WQE 0 and WQE 1
SW: Process the CQEs, and stamp WQE 0 and WQE 1 accordingly (on the TX buffer)
SW: Copy WQE 1 from the TX buffer to the BF register - ALREADY STAMPED!
HW: CQE error with index 0xFFFF - the BF WQE's control segment is STAMPED,
so the BF index is 0xFFFF. Error: Invalid Opcode.
As a result QP enters the error state and no traffic can be sent.
Solution:
When stamping - do not stamp last completed wqe.
Signed-off-by: Eugenia Emantayev <eugenia@...lanox.com>
Signed-off-by: Amir Vadai <amirv@...lanox.com>
---
drivers/net/ethernet/mellanox/mlx4/en_tx.c | 61 ++++++++++++++++++++----------
1 file changed, 42 insertions(+), 19 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 7c49238..6dcca98 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -191,6 +191,39 @@ void mlx4_en_deactivate_tx_ring(struct mlx4_en_priv *priv,
MLX4_QP_STATE_RST, NULL, 0, 0, &ring->qp);
}
+static void mlx4_en_stamp_wqe(struct mlx4_en_priv *priv,
+ struct mlx4_en_tx_ring *ring, int index,
+ u8 owner)
+{
+ __be32 stamp = cpu_to_be32(STAMP_VAL | (!!owner << STAMP_SHIFT));
+ struct mlx4_en_tx_desc *tx_desc = ring->buf + index * TXBB_SIZE;
+ struct mlx4_en_tx_info *tx_info = &ring->tx_info[index];
+ void *end = ring->buf + ring->buf_size;
+ __be32 *ptr = (__be32 *)tx_desc;
+ int i;
+
+ /* Optimize the common case when there are no wraparounds */
+ if (likely((void *)tx_desc + tx_info->nr_txbb * TXBB_SIZE <= end)) {
+ /* Stamp the freed descriptor */
+ for (i = 0; i < tx_info->nr_txbb * TXBB_SIZE;
+ i += STAMP_STRIDE) {
+ *ptr = stamp;
+ ptr += STAMP_DWORDS;
+ }
+ } else {
+ /* Stamp the freed descriptor */
+ for (i = 0; i < tx_info->nr_txbb * TXBB_SIZE;
+ i += STAMP_STRIDE) {
+ *ptr = stamp;
+ ptr += STAMP_DWORDS;
+ if ((void *)ptr >= end) {
+ ptr = ring->buf;
+ stamp ^= cpu_to_be32(0x80000000);
+ }
+ }
+ }
+}
+
static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
struct mlx4_en_tx_ring *ring,
@@ -205,8 +238,6 @@ static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
void *end = ring->buf + ring->buf_size;
int frags = skb_shinfo(skb)->nr_frags;
int i;
- __be32 *ptr = (__be32 *)tx_desc;
- __be32 stamp = cpu_to_be32(STAMP_VAL | (!!owner << STAMP_SHIFT));
struct skb_shared_hwtstamps hwts;
if (timestamp) {
@@ -232,12 +263,6 @@ static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
skb_frag_size(frag), PCI_DMA_TODEVICE);
}
}
- /* Stamp the freed descriptor */
- for (i = 0; i < tx_info->nr_txbb * TXBB_SIZE; i += STAMP_STRIDE) {
- *ptr = stamp;
- ptr += STAMP_DWORDS;
- }
-
} else {
if (!tx_info->inl) {
if ((void *) data >= end) {
@@ -263,16 +288,6 @@ static u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv,
++data;
}
}
- /* Stamp the freed descriptor */
- for (i = 0; i < tx_info->nr_txbb * TXBB_SIZE; i += STAMP_STRIDE) {
- *ptr = stamp;
- ptr += STAMP_DWORDS;
- if ((void *) ptr >= end) {
- ptr = ring->buf;
- stamp ^= cpu_to_be32(0x80000000);
- }
- }
-
}
dev_kfree_skb_any(skb);
return tx_info->nr_txbb;
@@ -318,8 +333,9 @@ static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq)
struct mlx4_en_tx_ring *ring = &priv->tx_ring[cq->ring];
struct mlx4_cqe *cqe;
u16 index;
- u16 new_index, ring_index;
+ u16 new_index, ring_index, stamp_index;
u32 txbbs_skipped = 0;
+ u32 txbbs_stamp = 0;
u32 cons_index = mcq->cons_index;
int size = cq->size;
u32 size_mask = ring->size_mask;
@@ -335,6 +351,7 @@ static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq)
index = cons_index & size_mask;
cqe = &buf[(index << factor) + factor];
ring_index = ring->cons & size_mask;
+ stamp_index = ring_index;
/* Process all completed CQEs */
while (XNOR(cqe->owner_sr_opcode & MLX4_CQE_OWNER_MASK,
@@ -359,6 +376,12 @@ static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq)
priv, ring, ring_index,
!!((ring->cons + txbbs_skipped) &
ring->size), timestamp);
+
+ mlx4_en_stamp_wqe(priv, ring, stamp_index,
+ !!((ring->cons + txbbs_stamp) &
+ ring->size));
+ stamp_index = ring_index;
+ txbbs_stamp = txbbs_skipped;
packets++;
bytes += ring->tx_info[ring_index].nr_bytes;
} while (ring_index != new_index);
--
1.8.3.251.g1462b67
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists