[dpdk-dev] [PATCH] net/mlx5: fix mbufs replenishment check for zipped CQEs

Alexander Kozyrev akozyrev at nvidia.com
Wed Aug 4 08:23:16 CEST 2021


A core dump is being generated with the following call stack:
0 _mm256_storeu_si256 (__A=..., __P=0x80)
1 rte_mov32 (src=0x2299c9140 "", dst=0x80)
2 rte_memcpy_aligned (n=60, src=0x2299c9140, dst=0x80)
3 rte_memcpy (n=60, src=0x2299c9140, dst=0x80)
4 mprq_buf_to_pkt (strd_cnt=1, strd_idx=0, buf=0x2299c8a00, len=60,
pkt=0x18345f0c0, rxq=0x18345ef40)
5 rxq_copy_mprq_mbuf_v (rxq=0x18345ef40, pkts=0x7f76e0ff6d18, pkts_n=5)
6 rxq_burst_mprq_v (rxq=0x18345ef40, pkts=0x7f76e0ff6d18, pkts_n=46,
err=0x7f76e0ff6a28, no_cq=0x7f76e0ff6a27)
7 mlx5_rx_burst_mprq_vec (dpdk_rxq=0x18345ef40, pkts=0x7f76e0ff6a88,
pkts_n=128)
8 rte_eth_rx_burst (nb_pkts=128, rx_pkts=0x7f76e0ff6a88,
queue_id=<optimized out>, port_id=<optimized out>)

This crash is caused by an attempt to copy previously uncompressed CQEs
into non-allocated mbufs. There is a check to make sure we only use
allocated mbufs in the rxq_burst_mprq_v() function, but it is done only
before the main processing loop. Leftovers of compressed CQEs session are
handled before that loop and may lead to the mbufs overflow as seen.

Move the check for replenished mbufs up to protect uncompressed CQEs
session leftovers from accessing non-allocated mbufs after the
mlx5_rx_mprq_replenish_bulk_mbuf() function is invoked.

Bugzilla ID: 746
Fixes: 0f20acbf5e ("net/mlx5: implement vectorized MPRQ burst")
Cc: stable at dpdk.org

Signed-off-by: Alexander Kozyrev <akozyrev at nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo at nvidia.com>
---
 drivers/net/mlx5/mlx5_rxtx_vec.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_rxtx_vec.c b/drivers/net/mlx5/mlx5_rxtx_vec.c
index e1b6d5422a..ecd273e00a 100644
--- a/drivers/net/mlx5/mlx5_rxtx_vec.c
+++ b/drivers/net/mlx5/mlx5_rxtx_vec.c
@@ -448,6 +448,8 @@ rxq_burst_mprq_v(struct mlx5_rxq_data *rxq, struct rte_mbuf **pkts,
 	rte_prefetch0(cq + 3);
 	pkts_n = RTE_MIN(pkts_n, MLX5_VPMD_RX_MAX_BURST);
 	mlx5_rx_mprq_replenish_bulk_mbuf(rxq);
+	/* Not to move past the allocated mbufs. */
+	pkts_n = RTE_MIN(pkts_n, rxq->elts_ci - rxq->rq_pi);
 	/* See if there're unreturned mbufs from compressed CQE. */
 	rcvd_pkt = rxq->decompressed;
 	if (rcvd_pkt > 0) {
@@ -463,8 +465,6 @@ rxq_burst_mprq_v(struct mlx5_rxq_data *rxq, struct rte_mbuf **pkts,
 	/* Not to cross queue end. */
 	pkts_n = RTE_MIN(pkts_n, elts_n - elts_idx);
 	pkts_n = RTE_MIN(pkts_n, q_n - cq_idx);
-	/* Not to move past the allocated mbufs. */
-	pkts_n = RTE_MIN(pkts_n, rxq->elts_ci - rxq->rq_pi);
 	if (!pkts_n) {
 		*no_cq = !cp_pkt;
 		return cp_pkt;
-- 
2.18.2



More information about the dev mailing list