net/mlx5: fix legacy non-inline multi-packet performance

Message ID 1574857992-28499-1-git-send-email-viacheslavo@mellanox.com (mailing list archive)
State Accepted, archived
Delegated to: Raslan Darawsheh
Headers
Series net/mlx5: fix legacy non-inline multi-packet performance |

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/iol-compilation fail Compilie Testing issues
ci/iol-intel-Performance success Performance Testing PASS
ci/iol-mellanox-Performance fail Performance Testing issues

Commit Message

Slava Ovsiienko Nov. 27, 2019, 12:33 p.m. UTC
  The legacy multi-packet write is the feature allowing to
put multiple packets into one transmitting descriptor,
this feature is supported by only NIC ConnectX-4LX.
The number of packets should be limited to provide
optimal size descriptor and better performance.

Fixes: 18a1c20044c0 ("net/mlx5: implement Tx burst template")

Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
---
 drivers/net/mlx5/mlx5_prm.h  | 2 ++
 drivers/net/mlx5/mlx5_rxtx.c | 4 +++-
 2 files changed, 5 insertions(+), 1 deletion(-)
  

Comments

Matan Azrad Nov. 27, 2019, 1:53 p.m. UTC | #1
From: Viacheslav Ovsiienko 
> The legacy multi-packet write is the feature allowing to put multiple packets
> into one transmitting descriptor, this feature is supported by only NIC
> ConnectX-4LX.
> The number of packets should be limited to provide optimal size descriptor
> and better performance.
> 
> Fixes: 18a1c20044c0 ("net/mlx5: implement Tx burst template")
> 
> Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Matan Azrad <matan@mellanox.com>
  
Thomas Monjalon Nov. 27, 2019, 3:26 p.m. UTC | #2
27/11/2019 14:53, Matan Azrad:
> From: Viacheslav Ovsiienko 
> > The legacy multi-packet write is the feature allowing to put multiple packets
> > into one transmitting descriptor, this feature is supported by only NIC
> > ConnectX-4LX.
> > The number of packets should be limited to provide optimal size descriptor
> > and better performance.
> > 
> > Fixes: 18a1c20044c0 ("net/mlx5: implement Tx burst template")
> > 
> > Signed-off-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
> Acked-by: Matan Azrad <matan@mellanox.com>

Applied, thanks
  

Patch

diff --git a/drivers/net/mlx5/mlx5_prm.h b/drivers/net/mlx5/mlx5_prm.h
index 651006b..2bac6a7 100644
--- a/drivers/net/mlx5/mlx5_prm.h
+++ b/drivers/net/mlx5/mlx5_prm.h
@@ -99,6 +99,8 @@ 
  * in tx burst routine at the moment of freeing multiple mbufs.
  */
 #define MLX5_EMPW_MAX_PACKETS MLX5_TX_COMP_THRESH
+#define MLX5_MPW_MAX_PACKETS 6
+
 /*
  * Default packet length threshold to be inlined with
  * ordinary SEND. Inlining saves the MR key search
diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
index 73c16de..25d915c 100644
--- a/drivers/net/mlx5/mlx5_rxtx.c
+++ b/drivers/net/mlx5/mlx5_rxtx.c
@@ -3888,7 +3888,9 @@  enum mlx5_txcmp_code {
 
 next_empw:
 		assert(NB_SEGS(loc->mbuf) == 1);
-		part = RTE_MIN(pkts_n, MLX5_EMPW_MAX_PACKETS);
+		part = RTE_MIN(pkts_n, MLX5_TXOFF_CONFIG(MPW) ?
+				       MLX5_MPW_MAX_PACKETS :
+				       MLX5_EMPW_MAX_PACKETS);
 		if (unlikely(loc->elts_free < part)) {
 			/* We have no enough elts to save all mbufs. */
 			if (unlikely(loc->elts_free < MLX5_EMPW_MIN_PACKETS))