Age | Commit message (Collapse) | Author |
|
On xmit path we use skb_frag_dma_map() which is using dma_map_page(),
while upon completion we dma-unmap the skb fragments using
dma_unmap_single() rather than dma_unmap_page().
To fix this, we now save the dma map type on xmit path and use this
info to call the right dma unmap method upon TX completion.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Consider MLX5E_MAX_NUM_CHANNELS @ethtool set/get_channels
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
To avoid multiply/division operations on the data path,
we hold a {channel, tc}==>txq mapping table.
We held this mapping table inside the channel object that is
being destroyed upon some configuration operations (e.g MTU change).
So in case ndo_select_queue occurs during such a configuration operation,
it may access a NULL channel pointer, resulting in kernel panic.
To fix this issue we moved the {channel, tc}==>txq mapping table
outside the channel object so that it will be available also
during such configuration operations.
Signed-off-by: Rana Shahout <ranas@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Only for packets with first ethertype set to IPv4/6 for now.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
- Change the maximum LRO session size from 16KB to 64KB
- Reduce the LRO session timeout from 512us to 32us in
order to reduce the TCP latency of non-LRO'ed flows.
- Fix skb_shinfo(skb)->gso_size and set skb_shinfo(skb)->gso_type.
- Fix a bug accessing un-initialized mdev pointer.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
We un-intentionally limited the minimum rings size too much.
TX minimum ring size reduced from 128 to 64.
RX minimum ring size reduced from 128 to 2.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
- get_rxfh_key_size
- get_rxfh_indir_size
- get/set_rxfh indirection table and RSS Toeplitz hash key
- get_rxnfc
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The indirection table size was defined by a variable that
was actually assigned a constant value.
Since we do not have any forseen intension to make it configurable
we simply made it a constant.
We also limit the number of channels such that the RSS indirection
table could always populate all RX rings.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
No need to generate a unique key per TIR.
Generating a single key per netdev and copying it to all
its TIRs.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Added physical port counters in the following standard formats to
ethtool statistics:
- IEEE 802.3
- RFC2863
- RFC2819
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Now that TIRs, TISs and flow tables are kept alive while the netdev is
stopped (after executing ndo_stop()) we can do the following
improvements:
- Obsolete the active_vlans SW shadow.
- Do not delete/add flow table rules upon ndo_stop/open.
In addition to simplifying the flow, this change also fastens
the ndo_open/close operations.
- Obsolete synchronization of threads accessing the flow tables
with the netdev stop/open threads.
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>
|
|
Rename some functions that used to be invoked upon ndo_open/stop and
are now invoked upon create/destroy_netdev() in order to better hint
their place in the flow.
Change some functions location in the file so that functions involved
in ndo_open/stop flow will not be interleaved with other functions.
This is a cosmetic change, no logical change here.
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>
|
|
RX traffic routed to this RQ will be silently dropped, at the NIC HW
level.
This is in preparation for netdev "light-weight" open/stop flow
change described in previous commit.
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>
|
|
Generally an RX packet flows through the following objects:
Flow table --> TIR --> RQT --> RQ
Where:
- TIR stands for "Transport Interface Receive", defining the RSS and
LRO paramaters.
- RQT stands for "RQ Table", implementing the RSS indirection table.
- RQ stands for "Receive Queue"
For flows that do not need LRO, nor RSS, the driver made a shortcut to
the above RX flow by pointing to the RQ directly from the TIR, yielding
this flow:
Flow table --> TIR --> RQ
In this commit we remove this shortcut by "inserting" a single-RQ RQT
between the TIR and the RQ, i.e RX packets will reach the same RQ but
will go through an RQT of size 1, pointing to just a single RQ.
This way the RX traffic re-direction to/from the "Drop RQ" will be more
uniform (AKA "one flow"), as it will involve only RQTs re-direction and
no TIRs re-direction.
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>
|
|
It was used to update netdev priv parameters that require stopping
and re-opening the device in a generic way - it got the new
parameters and did: ndo_stop(), copy new parameters into current
parameters, ndo_open().
We chose to remove it for two reasons:
1) It requires additional instance of struct mlx5e_params on the
stack and looking forward we expect this struct to grow.
2) Sometimes we want to do additional operations (besides
just updating the priv parameters) while the netdev is stopped.
For example, updating netdev->mtu @mlx5e_change_mtu() should
be done while the netdev is stopped (done in this commit).
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>
|
|
This field already exists under the mlx5e_params struct
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>
|
|
In addition to the source/destination IP which are already hashed.
Only for unicast traffic for now.
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>
|
|
No logical change in this commit.
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>
|
|
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>
|
|
AKA inline WQE.
A TX latency optimization to save data gather DMA reads.
Controlled by ETHTOOL_TX_COPYBREAK.
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>
|
|
The ConnectX-4 HW implements inverted XOR8.
To make it act as XOR we re-order the HW RSS indirection table.
Set XOR to be the default RSS hash function and add ethtool API to
control it.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Use container_of() instead.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
To save per-packet calculations, we use the following static mappings:
1) priv {channel, tc} to netdev txq (used @mlx5e_selec_queue())
2) netdev txq to priv sq (used @mlx5e_xmit())
Thanks to these static mappings, no more need for a separate implementation
of ndo_start_xmit when multiple TCs are configured.
We believe the performance improvement of such separation would be negligible, if any.
The previous way of dynamically calculating the above mappings required
allocating more TX queues than actually used (@alloc_etherdev_mqs()),
which is now no longer needed.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Allocate and use transport domain by the Ethernet driver code.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
When NETIF_F_SG is set, each send WQE may have a different size since
each skb can have different number of fragments as of LSO header etc.
This implies that a given WQE may wrap around the send queue, i.e begin
at its end and continue at its start. While it is legal by the device spec,
we preferred a solution that avoids it - when building of current WQE is
done, if the next WQE may wrap around the send queue, fill the send queue
with NOPs WQEs till its end, so that the next WQE will begin at send queue
start.
NOP WQE for itself cannot wrap around the send queue since it is of
minimal size - 64 bytes, and all send WQEs are a multiple of that size.
Signed-off-by: Achiad Shochat <achiad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Previously we configured HW MTU to be netdev->mtu, actually we
need to configure netdev->mtu + (ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN).
Also, query MTU can not fail, hence make the relevant helper a
void functionm, add mlx5e_set_dev_port_mtu, helper function to
handle MTU setting.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Move the vport header file to be under include/linux/mlx5, such that
the mlx5 IB can use it as well.
Also add nic_ prefix to the vport NIC commands to differeniate between
HCA vport commands and NIC vport commands.
Signed-off-by: Majd Dibbiny <majd@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This is the Ethernet part of the driver for the Mellanox ConnectX(R)-4
Single/Dual-Port Adapter supporting 100Gb/s with VPI. The driver
extends the existing mlx5 driver with Ethernet functionality.
This patch contains the driver entry points but does not include
transmit and receive (see the previous patch in the series) routines.
It also adds the option MLX5_CORE_EN to Kconfig to enable/disable the
Ethernet functionality. Currently, Kconfig is programmed to make
Ethernet and Infiniband functionality mutally exclusive.
Also changed MLX5_INFINIBAND to be depandant on MLX5_CORE instead of
selecting it, since MLX5_CORE could be selected without MLX5_INFINIBAND
being selected.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|