d/numam-dpdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

net/mlx4: remove scatter mode compilation option

Browse Source 
This option both sets the maximum number of segments for Rx/Tx packets and
whether scattered mode is supported at all. This commit removes the latter
as well as configuration file exposure since the most appropriate value
should be decided at run-time.

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
This commit is contained in:
Adrien Mazarguil 2017-09-01 10:06:23 +02:00 committed by Ferruh Yigit
parent 31a76ab0df
commit 367b31cd5b
5 changed files with 0 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
config/common_base
View File

@ -215,7 +215,6 @@ CONFIG_RTE_LIBRTE_FM10K_INC_VECTOR=y
CONFIG_RTE_LIBRTE_MLX4_PMD=n
CONFIG_RTE_LIBRTE_MLX4_DEBUG=n
CONFIG_RTE_LIBRTE_MLX4_DEBUG_BROKEN_VERBS=n
CONFIG_RTE_LIBRTE_MLX4_SGE_WR_N=4
CONFIG_RTE_LIBRTE_MLX4_MAX_INLINE=0
CONFIG_RTE_LIBRTE_MLX4_TX_MP_CACHE=8

7
doc/guides/nics/mlx4.rst
View File

@ -124,13 +124,6 @@ These options can be modified in the ``.config`` file.
to abort with harmless debugging messages as a workaround.
Relevant only when CONFIG_RTE_LIBRTE_MLX4_DEBUG is enabled.
- ``CONFIG_RTE_LIBRTE_MLX4_SGE_WR_N`` (default **4**)
Number of scatter/gather elements (SGEs) per work request (WR). Lowering
this number improves performance but also limits the ability to receive
scattered packets (packets that do not fit a single mbuf). The default
value is a safe tradeoff.
- ``CONFIG_RTE_LIBRTE_MLX4_MAX_INLINE`` (default **0**)
Amount of data to be inlined during TX operations. Improves latency but

4
drivers/net/mlx4/Makefile
View File

@ -68,10 +68,6 @@ else
CFLAGS += -DNDEBUG -UPEDANTIC
endif
ifdef CONFIG_RTE_LIBRTE_MLX4_SGE_WR_N
CFLAGS += -DMLX4_PMD_SGE_WR_N=$(CONFIG_RTE_LIBRTE_MLX4_SGE_WR_N)
endif
ifdef CONFIG_RTE_LIBRTE_MLX4_MAX_INLINE
CFLAGS += -DMLX4_PMD_MAX_INLINE=$(CONFIG_RTE_LIBRTE_MLX4_MAX_INLINE)
endif

10
drivers/net/mlx4/mlx4.c
View File

@ -1176,8 +1176,6 @@ txq_mp2mr_iter(struct rte_mempool *mp, void *arg)
txq_mp2mr(txq, mp);
}
#if MLX4_PMD_SGE_WR_N > 1
/**
* Copy scattered mbuf contents to a single linear buffer.
*
@ -1324,8 +1322,6 @@ stop:
};
}
#endif /* MLX4_PMD_SGE_WR_N > 1 */
/**
* DPDK callback for TX.
*
@ -1451,7 +1447,6 @@ mlx4_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
goto stop;
sent_size += length;
} else {
#if MLX4_PMD_SGE_WR_N > 1
struct ibv_sge sges[MLX4_PMD_SGE_WR_N];
struct tx_burst_sg_ret ret;
@ -1469,11 +1464,6 @@ mlx4_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
if (unlikely(err))
goto stop;
sent_size += ret.length;
#else /* MLX4_PMD_SGE_WR_N > 1 */
DEBUG("%p: TX scattered buffers support not"
" compiled in", (void *)txq);
goto stop;
#endif /* MLX4_PMD_SGE_WR_N > 1 */
}
elts_head = elts_head_next;
/* Increment sent bytes counter. */

2
drivers/net/mlx4/mlx4.h
View File

@ -81,9 +81,7 @@
#define MLX4_PMD_TX_PER_COMP_REQ 64
/* Maximum number of Scatter/Gather Elements per Work Request. */
#ifndef MLX4_PMD_SGE_WR_N
#define MLX4_PMD_SGE_WR_N 4
#endif
/* Maximum size for inline data. */
#ifndef MLX4_PMD_MAX_INLINE