During Rx queue error recovery, the vectorized path in
mlx5_rx_err_handle() reallocates an mbuf for every queue element. When
rte_mbuf_raw_alloc() fails (for example, the mempool is exhausted), the
rollback loop frees the mbufs allocated so far, but masks the element
ring index with "& elts_n" instead of "& (elts_n - 1)".
elts_n is a power-of-two element count, so "x & elts_n" isolates a
single bit and can only evaluate to 0 or elts_n, regardless of the loop
counter. The rollback therefore never frees the mbufs just allocated in
this pass (they are leaked); instead it repeatedly frees elts[0], a live
mbuf still posted to the NIC (use-after-free / double free), and
elts[elts_n], the fake_mbuf padding entry used by the vector datapath.
Mask with the existing e_mask (elts_n - 1), as already done in the
matching forward allocation loop just above.
Fixes: 0f20acbf5eda ("net/mlx5: implement vectorized MPRQ burst")
Cc: [email protected]
Signed-off-by: Borys Tsyrulnikov <[email protected]>
---
.mailmap | 1 +
drivers/net/mlx5/mlx5_rx.c | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/.mailmap b/.mailmap
index 4001e5fb0e..0b09243c45 100644
--- a/.mailmap
+++ b/.mailmap
@@ -222,6 +222,7 @@ Boleslav Stankevich <[email protected]>
Boon Ang <[email protected]> <[email protected]>
Boris Ouretskey <[email protected]>
Boris Pismenny <[email protected]>
+Borys Tsyrulnikov <[email protected]>
Brad Larson <[email protected]>
Brandon Lo <[email protected]>
Brendan Ryan <[email protected]>
diff --git a/drivers/net/mlx5/mlx5_rx.c b/drivers/net/mlx5/mlx5_rx.c
index ce50087b70..c0ad8d6701 100644
--- a/drivers/net/mlx5/mlx5_rx.c
+++ b/drivers/net/mlx5/mlx5_rx.c
@@ -662,7 +662,7 @@ mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec,
if (!*elt) {
for (i--; i >= 0; --i) {
elt_idx = (elts_ci +
- i) & elts_n;
+ i) & e_mask;
elt = &(*rxq->elts)
[elt_idx];
rte_pktmbuf_free_seg
--
2.34.1
