[dpdk-dev] [PATCH] net/mlx5: fix packet inline on Tx queue wraparound

[dpdk-dev] [PATCH] net/mlx5: fix packet inline on Tx queue wraparound

[Shahaf Shuler]

Inlining a packet to WQE that cross the WQ wraparound, i.e. the WQE
starts on the end of the ring and ends on the beginning, is not
supported and blocked by the data path logic.

However, in case of TSO, an extra inline header is required before
inlining. This inline header is not taken into account when checking if
there is enough room left for the required inline size.
On some corner cases were
(ring_tailroom - inline header) < inline size < ring_tailroom ,
this can lead to WQE being written outsize of the ring buffer.

Fixing it by always assuming the worse case that inline of packet will
require the inline header.

Fixes: 3f13f8c23a7c ("net/mlx5: support hardware TSO")
Cc: stable@dpdk.org

Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
---
 drivers/net/mlx5/mlx5_rxtx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
index baa4079c14..38ce0e29a2 100644
--- a/drivers/net/mlx5/mlx5_rxtx.c
+++ b/drivers/net/mlx5/mlx5_rxtx.c
@@ -693,7 +693,8 @@ mlx5_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
 						   RTE_CACHE_LINE_SIZE);
 			copy_b = (addr_end > addr) ?
 				 RTE_MIN((addr_end - addr), length) : 0;
-			if (copy_b && ((end - (uintptr_t)raw) > copy_b)) {
+			if (copy_b && ((end - (uintptr_t)raw) >
+				       (copy_b + sizeof(inl)))) {
 				/*
 				 * One Dseg remains in the current WQE.  To
 				 * keep the computation positive, it is
-- 
2.12.0

Re: [dpdk-dev] [PATCH] net/mlx5: fix packet inline on Tx queue wraparound

[Yongseok Koh]

> On Mar 10, 2019, at 12:14 AM, Shahaf Shuler <shahafs@mellanox.com> wrote:
> 
> Inlining a packet to WQE that cross the WQ wraparound, i.e. the WQE
> starts on the end of the ring and ends on the beginning, is not
> supported and blocked by the data path logic.
> 
> However, in case of TSO, an extra inline header is required before
> inlining. This inline header is not taken into account when checking if
> there is enough room left for the required inline size.
> On some corner cases were
> (ring_tailroom - inline header) < inline size < ring_tailroom ,
> this can lead to WQE being written outsize of the ring buffer.
> 
> Fixing it by always assuming the worse case that inline of packet will
> require the inline header.
> 
> Fixes: 3f13f8c23a7c ("net/mlx5: support hardware TSO")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
> ---

Acked-by: Yongseok Koh <yskoh@mellanox.com>
 
Thanks

> drivers/net/mlx5/mlx5_rxtx.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
> index baa4079c14..38ce0e29a2 100644
> --- a/drivers/net/mlx5/mlx5_rxtx.c
> +++ b/drivers/net/mlx5/mlx5_rxtx.c
> @@ -693,7 +693,8 @@ mlx5_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
> 						   RTE_CACHE_LINE_SIZE);
> 			copy_b = (addr_end > addr) ?
> 				 RTE_MIN((addr_end - addr), length) : 0;
> -			if (copy_b && ((end - (uintptr_t)raw) > copy_b)) {
> +			if (copy_b && ((end - (uintptr_t)raw) >
> +				       (copy_b + sizeof(inl)))) {
> 				/*
> 				 * One Dseg remains in the current WQE.  To
> 				 * keep the computation positive, it is
> -- 
> 2.12.0
> 

Re: [dpdk-dev] [PATCH] net/mlx5: fix packet inline on Tx queue wraparound

[Shahaf Shuler]

Monday, March 11, 2019 9:58 PM, Yongseok Koh:
> Subject: Re: [PATCH] net/mlx5: fix packet inline on Tx queue wraparound
> 
> 
> > On Mar 10, 2019, at 12:14 AM, Shahaf Shuler <shahafs@mellanox.com>
> wrote:
> >
> > Inlining a packet to WQE that cross the WQ wraparound, i.e. the WQE
> > starts on the end of the ring and ends on the beginning, is not
> > supported and blocked by the data path logic.
> >
> > However, in case of TSO, an extra inline header is required before
> > inlining. This inline header is not taken into account when checking
> > if there is enough room left for the required inline size.
> > On some corner cases were
> > (ring_tailroom - inline header) < inline size < ring_tailroom , this
> > can lead to WQE being written outsize of the ring buffer.
> >
> > Fixing it by always assuming the worse case that inline of packet will
> > require the inline header.
> >
> > Fixes: 3f13f8c23a7c ("net/mlx5: support hardware TSO")
> > Cc: stable@dpdk.org
> >
> > Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
> > ---
> 
> Acked-by: Yongseok Koh <yskoh@mellanox.com>

Applied to next-net-mlx, thanks. 

Re: [dpdk-dev] [PATCH] net/mlx5: fix packet inline on Tx queue wraparound

[Shahaf Shuler]

Monday, March 11, 2019 9:58 PM, Yongseok Koh:
> Subject: Re: [PATCH] net/mlx5: fix packet inline on Tx queue wraparound
> 
> 
> > On Mar 10, 2019, at 12:14 AM, Shahaf Shuler <shahafs@mellanox.com>
> wrote:
> >
> > Inlining a packet to WQE that cross the WQ wraparound, i.e. the WQE
> > starts on the end of the ring and ends on the beginning, is not
> > supported and blocked by the data path logic.
> >
> > However, in case of TSO, an extra inline header is required before
> > inlining. This inline header is not taken into account when checking
> > if there is enough room left for the required inline size.
> > On some corner cases were
> > (ring_tailroom - inline header) < inline size < ring_tailroom , this
> > can lead to WQE being written outsize of the ring buffer.
> >
> > Fixing it by always assuming the worse case that inline of packet will
> > require the inline header.
> >
> > Fixes: 3f13f8c23a7c ("net/mlx5: support hardware TSO")
> > Cc: stable@dpdk.org
> >
> > Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
> > ---
> 
> Acked-by: Yongseok Koh <yskoh@mellanox.com>

Applied to next-net-mlx, thanks.