diff options
author | Achiad Shochat | 2015-07-23 22:35:59 +0200 |
---|---|---|
committer | David S. Miller | 2015-07-27 09:29:17 +0200 |
commit | 88a85f99e51fb2373259ab83c8bb130a9bbf3804 (patch) | |
tree | f27980b2cdd3c9601a5953c7d268eabc18452a2e /drivers/net/ethernet/mellanox/mlx5/core/en_main.c | |
parent | net/mlx5e: Support TX packet copy into WQE (diff) | |
download | kernel-qcow2-linux-88a85f99e51fb2373259ab83c8bb130a9bbf3804.tar.gz kernel-qcow2-linux-88a85f99e51fb2373259ab83c8bb130a9bbf3804.tar.xz kernel-qcow2-linux-88a85f99e51fb2373259ab83c8bb130a9bbf3804.zip |
net/mlx5e: TX latency optimization to save DMA reads
A regular TX WQE execution involves two or more DMA reads -
one to fetch the WQE, and another one per WQE gather entry.
These DMA reads obviously increase the TX latency.
There are two mlx5 mechanisms to bypass these DMA reads:
1) Inline WQE
2) Blue Flame (BF)
An inline WQE contains a whole packet, thus saves the DMA read/s
of the regular WQE gather entry/s. Inline WQE support was already
added in the previous commit.
A BF WQE is written directly to the device I/O mapped memory, thus
enables saving the DMA read that fetches the WQE.
The BF WQE I/O write must be in cache line granularity, thus uses
the CPU write combining mechanism.
A BF WQE I/O write acts also as a TX doorbell for notifying the
device of new TX WQEs.
A BF WQE is written to the same I/O mapped address as the regular TX
doorbell, thus this address is being mapped twice - once by ioremap()
and once by io_mapping_map_wc().
While both mechanisms reduce the TX latency, they both consume more CPU
cycles than a regular WQE:
- A BF WQE must still be written to host memory, in addition to being
written directly to the device I/O mapped memory.
- An inline WQE involves copying the SKB data into it.
To handle this tradeoff, we introduce here a heuristic algorithm that
strives to avoid using these two mechanisms in case the TX queue is
being back-pressured by the device, and limit their usage rate otherwise.
An inline WQE will always be "Blue Flamed" (written directly to the
device I/O mapped memory) while a BF WQE may not be inlined (may contain
gather entries).
Preliminary testing using netperf UDP_RR shows that the latency goes down
from 17.5us to 16.9us, while the message rate (tested with pktgen) stays
the same.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlx5/core/en_main.c')
-rw-r--r-- | drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 12 |
1 files changed, 7 insertions, 5 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index c55fad431cbf..4a87e9dcf52c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -514,6 +514,7 @@ static int mlx5e_create_sq(struct mlx5e_channel *c, sq->wq.db = &sq->wq.db[MLX5_SND_DBR]; sq->uar_map = sq->uar.map; + sq->uar_bf_map = sq->uar.bf_map; sq->bf_buf_size = (1 << MLX5_CAP_GEN(mdev, log_bf_reg_size)) / 2; sq->max_inline = param->max_inline; @@ -524,11 +525,12 @@ static int mlx5e_create_sq(struct mlx5e_channel *c, txq_ix = c->ix + tc * priv->params.num_channels; sq->txq = netdev_get_tx_queue(priv->netdev, txq_ix); - sq->pdev = c->pdev; - sq->mkey_be = c->mkey_be; - sq->channel = c; - sq->tc = tc; - sq->edge = (sq->wq.sz_m1 + 1) - MLX5_SEND_WQE_MAX_WQEBBS; + sq->pdev = c->pdev; + sq->mkey_be = c->mkey_be; + sq->channel = c; + sq->tc = tc; + sq->edge = (sq->wq.sz_m1 + 1) - MLX5_SEND_WQE_MAX_WQEBBS; + sq->bf_budget = MLX5E_SQ_BF_BUDGET; priv->txq_to_sq_map[txq_ix] = sq; return 0; |