diff options
author | Achiad Shochat <achiad@mellanox.com> | 2015-07-23 23:35:59 +0300 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2015-07-27 00:29:17 -0700 |
commit | 88a85f99e51fb2373259ab83c8bb130a9bbf3804 (patch) | |
tree | f27980b2cdd3c9601a5953c7d268eabc18452a2e /drivers/net/ethernet/mellanox/mlx5/core/uar.c | |
parent | 58d522912ac7d25b63f468fa4a4e8bb059c5144e (diff) | |
download | linux-88a85f99e51fb2373259ab83c8bb130a9bbf3804.tar.bz2 |
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/uar.c')
-rw-r--r-- | drivers/net/ethernet/mellanox/mlx5/core/uar.c | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/uar.c b/drivers/net/ethernet/mellanox/mlx5/core/uar.c index 9ef85873ceea..eb05c845ece9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/uar.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/uar.c @@ -32,6 +32,7 @@ #include <linux/kernel.h> #include <linux/module.h> +#include <linux/io-mapping.h> #include <linux/mlx5/driver.h> #include <linux/mlx5/cmd.h> #include "mlx5_core.h" @@ -246,6 +247,10 @@ int mlx5_alloc_map_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar) goto err_free_uar; } + if (mdev->priv.bf_mapping) + uar->bf_map = io_mapping_map_wc(mdev->priv.bf_mapping, + uar->index << PAGE_SHIFT); + return 0; err_free_uar: @@ -257,6 +262,7 @@ EXPORT_SYMBOL(mlx5_alloc_map_uar); void mlx5_unmap_free_uar(struct mlx5_core_dev *mdev, struct mlx5_uar *uar) { + io_mapping_unmap(uar->bf_map); iounmap(uar->map); mlx5_cmd_free_uar(mdev, uar->index); } |