The EMAC driver had a custom IBM_EMAC_RX_SKB_HEADROOM
Kconfig option that reserved additional skb headroom for RX.
This patch removes the option and migrates the code
to use napi_alloc_skb() and netdev_alloc_skb_ip_align()
in its place.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current driver only enables Pdelay_Req and Pdelay_Resp when
HWTSTAMP_FILTER_PTP_V2_EVENT, HWTSTAMP_FILTER_PTP_V1_L4_EVENT or
HWTSTAMP_FILTER_PTP_V2_L4_EVENT is requested. This results in ptp sync on
slave mode to report 'received SYNC without timestamp' when using ptp4l.
Although the hardware can support Sync, Pdelay_Req and Pdelay_resp by
setting bit14 annd bits 17/16 to 01 this leaves Delay_Req timestamps out.
Fix this by enabling all event and general messages timestamps.
This includes SYNC, Follow_Up, Delay_Req, Delay_Resp, Pdelay_Req,
Pdelay_Resp and Pdelay_Resp_Follow_Up messages.
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Acked-by: Jose Abreu <joabreu@synopsys.com>
Tested-by: Alexandre TORGUE <alexandre.torgue@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Create a new FIB entry type for blackhole routes and set it in case the
type of the notified route is 'RTN_BLACKHOLE'.
Program such routes with a discard action and mark them as offloaded
since the device is dropping the packets instead of the kernel.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlxsw implements SWITCHDEV_ATTR_ID_PORT_PARENT_ID and we want to get rid
of switchdev_ops eventually, ease that migration by implementing a
ndo_get_port_parent_id() function which returns what
switchdev_port_attr_get() would do.
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
NFP only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it a
great candidate to be converted to use the ndo_get_port_parent_id() NDO
instead of implementing switchdev_port_attr_get().
Since NFP uses switchdev_port_same_parent_id() convert it to use
netdev_port_same_parent_id().
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ocelot only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID as a valid
switchdev attribute getter, convert it to use ndo_get_port_parent_id()
and get rid of the switchdev_ops::switchdev_port_attr_get altogether.
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlxsw implements SWITCHDEV_ATTR_ID_PORT_PARENT_ID and we want to get rid
of switchdev_ops eventually, ease that migration by implementing a
ndo_get_port_parent_id() function which returns what
switchdev_port_attr_get() would do.
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlx5e only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it a
great candidate to be converted to use the ndo_get_port_parent_id() NDO
instead of implementing switchdev_port_attr_get().
Since mlx5e makes use of switchdev_port_parent_id() convert it to use
netdev_port_same_parent_id().
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Liquidio only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it
a great candidate to be converted to use the ndo_get_port_parent_id()
NDO instead of implementing switchdev_port_attr_get().
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
BNXT only supports SWITCHDEV_ATTR_ID_PORT_PARENT_ID, which makes it a
great candidate to be converted to use the ndo_get_port_parent_id() NDO
instead of implementing switchdev_port_attr_get(). The conversion is
straight forward here since the PF and VF code use the same getter.
Since bnxt makes uses of switchdev_port_same_parent_id() convert it to
use netdev_port_same_parent_id().
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change t4fw_version.h to update latest firmware version
number to 1.22.9.0.
Signed-off-by: Vishal Kulkarni <vishal@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Read MAC address 32-bit at a time and manually extract the individual
bytes. This avoids pointer aliasing and gives the compiler a better
chance of optimizing the operation.
Suggested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the system was booted using a device tree and if the device tree
contains a MAC address, use it instead of reading one from the EEPROM.
This is useful in situations where the EEPROM isn't properly programmed
or where the firmware wants to override the existing MAC address.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Trace EMAD errors returned from HW.
Signed-off-by: Nir Dotan <nird@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add more buffers to the Rx buffer pool as soon as 7 of them
get consumed, instead of waiting for their number to drop
below a fixed threshold.
7 is the number of buffers that can be released in the pool
via a single DPIO command.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Starting with MC10.14.0, dpaa2_io_service_enqueue_fq() API is
functional. Since there are a number of cases where it offers
better performance compared to the currently used enqueue
function, switch to it for firmware versions that support it.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While in NAPI context, free skbs by calling napi_consume_skb()
instead of dev_kfree_skb(), to take advantage of the bulk freeing
mechanism.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of allocating page fragments via the network stack,
use the page allocator directly. For now, we consume one page
for each Rx buffer.
With the new memory model we are free to consider adding more
XDP support.
Performance decreases slightly in some IP forwarding cases.
No visible effect on termination traffic. The driver memory
footprint increases as a result of this change, but it is
still small enough to not really matter.
Another side effect is that now Rx buffer alignment requirements
are naturally satisfied without any additional actions needed.
Remove alignment related code, except in the buffer layout
information conveyed to MC, as hardware still needs to know the
alignment value we guarantee.
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The qede driver supports for ethtool_rx_flow_spec and flower, both
codebases look very similar.
This patch uses the ethtool_rx_flow_rule() infrastructure to remove the
duplicated ethtool_rx_flow_spec parser and consolidate ACL offload
support around the flow_rule infrastructure.
Furthermore, more code can be consolidated by merging
qede_add_cls_rule() and qede_add_tc_flower_fltr(), these two functions
also look very similar.
This driver currently provides simple ACL support, such as 5-tuple
matching, drop policy and queue to CPU.
Drivers that support more features can benefit from this infrastructure
to save even more redundant codebase.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is a preparation patch to reuse the existing TC flower codebase
from ethtool_rx_flow_spec.
This patch is merely moving the core ethtool_rx_flow_spec parser after
tc flower offload driver code so we can skip a few forward function
declarations in the follow up patch.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch updates drivers to use the new flow action infrastructure.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch provides the flow_stats structure that acts as container for
tc_cls_flower_offload, then we can use to restore the statistics on the
existing TC actions. Hence, tcf_exts_stats_update() is not used from
drivers anymore.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds pedit_headers_action structure to store the result of
parsing tc pedit actions. Then, it calls alloc_tc_pedit_action() to
populate the mlx5e hardware intermediate representation once all actions
have been parsed.
This patch comes in preparation for the new flow_action infrastructure,
where each packet mangling comes in an separated action, ie. not packed
as in tc pedit.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch wraps the dissector key and mask - that flower uses to
represent the matching side - around the flow_match structure.
To avoid a follow up patch that would edit the same LoCs in the drivers,
this patch also wraps this new flow match structure around the flow rule
object. This new structure will also contain the flow actions in follow
up patches.
This introduces two new interfaces:
bool flow_rule_match_key(rule, dissector_id)
that returns true if a given matching key is set on, and:
flow_rule_match_XYZ(rule, &match);
To fetch the matching side XYZ into the match container structure, to
retrieve the key and the mask with one single call.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds basic ethtool support to the device to allow
for configuration.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
With recent changes, need to bump the driver version to reflect the
changes.
Signed-off-by: Todd Fujinaka <todd.fujinaka@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Remove the redundant 'igc_get_phy_id_base' method and use
the 'igc_get_phy_id' method directly instead.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Remove the redundant 'igc_read_mac_addr_base' method and use
the 'igc_read_mac_addr' method directly instead.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
I'm seeing series of e1000e resets (sometimes endless) at system boot
if something generates tx traffic at this time. In my case this is
netconsole who sends message "e1000e 0000:02:00.0: Some CPU C-states
have been disabled in order to enable jumbo frames" from e1000e itself.
As result e1000_watchdog_task sees used tx buffer while carrier is off
and start this reset cycle again.
[ 17.794359] e1000e: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[ 17.794714] IPv6: ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
[ 22.936455] e1000e 0000:02:00.0 eth1: changing MTU from 1500 to 9000
[ 23.033336] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 26.102364] e1000e: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[ 27.174495] 8021q: 802.1Q VLAN Support v1.8
[ 27.174513] 8021q: adding VLAN 0 to HW filter on device eth1
[ 30.671724] cgroup: cgroup: disabling cgroup2 socket matching due to net_prio or net_cls activation
[ 30.898564] netpoll: netconsole: local port 6666
[ 30.898566] netpoll: netconsole: local IPv6 address 2a02:6b8:0:80b:beae:c5ff:fe28:23f8
[ 30.898567] netpoll: netconsole: interface 'eth1'
[ 30.898568] netpoll: netconsole: remote port 6666
[ 30.898568] netpoll: netconsole: remote IPv6 address 2a02:6b8:b000:605c:e61d:2dff:fe03:3790
[ 30.898569] netpoll: netconsole: remote ethernet address b0:a8:6e:f4:ff:c0
[ 30.917747] console [netcon0] enabled
[ 30.917749] netconsole: network logging started
[ 31.453353] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 34.185730] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 34.321840] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 34.465822] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 34.597423] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 34.745417] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 34.877356] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 35.005441] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 35.157376] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 35.289362] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 35.417441] e1000e 0000:02:00.0: Some CPU C-states have been disabled in order to enable jumbo frames
[ 37.790342] e1000e: eth1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
This patch flushes tx buffers only once when carrier is off
rather than at each watchdog iteration.
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Remove the 'igc_get_link_up_info_base method' from igc_base.c file.
Use the 'igc_get_speed_and_duplex_copper' method directly and reduce
the code redundancy.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Remove unused igc_adv_data_desc definition from igc_base.h file.
Descriptors definition will be added per demand.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
The change is based on the issue found by Kangjie Lu <kjlu@umn.edu> where
we not checking the return value of a register read/write which could result
in a NULL pointer dereference if the read/write fails.
Since we are only trying to disable the far-end loopback, if the read
and write of register fails, we do not want to bail out of the function.
We just want to log that it failed to disable and continue on.
CC: Sasha Neftin <sasha.neftin@intel.com>
CC: Kangjie Lu <kjlu@umn.edu>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
The function comment for fm10k_iov_msg_msix_pf has an extra space in
a sentence, which is unnecessary.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
ixgbe_reset_hw_82599() resets the value of hw->mac.num_rar_entries to
pre-defined value of 128. Let's get rid of that hardcoded literal, and use
IXGBE_82599_RAR_ENTRIES instead, the same way the normal initialization
path does.
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Remove redundant igc_check_for_link_base code and replace it with
an igc_check_for_copper_link method.
Fix duplication of IGC_ADVTXD_PAYLEN_SHIFT mask declaration.
Remove obsolete IGC_SCVPC register definition.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Address community comment.
Remove the unreachable code leads to the static checker warning.
PHY functionality will be added later per demand.
Reported by Dan Carpenter.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
e1000e sets different WoL settings in system suspend callback and
runtime suspend callback.
The suspend direct complete optimization leaves e1000e in runtime
suspended state with wrong WoL setting during system suspend.
To fix this, we need to disable suspend direct complete optimization to
let e1000e always use suspend callback to set correct WoL during system
suspend.
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Sven Auhagen reports that if he changes a SFP+ module for a SFP module
on the Macchiatobin Single Shot, the link does not come back up. For
Sven, it is as easy as:
- Insert a SFP+ module connected, and use ping6 to verify link is up.
- Remove SFP+ module
- Insert SFP 1000base-X module use ping6 to verify link is up: Link
up event did not trigger and the link is down
but that doesn't show the problem for me. Locally, this has been
reproduced by:
- Boot with no modules.
- Insert SFP+ module, confirm link is up.
- Replace module with 25000base-X module. Confirm link is up.
- Set remote end down, link is reported as dropped at both ends.
- Set remote end up, link is reported up at remote end, but not local
end due to lack of link interrupt.
Fix this by setting up both GMAC and XLG interrupts for port 0, but
only unmasking the appropriate interrupt according to the current mode
set in the mac_config() method. However, only do the mask/unmask
dance when we are really changing the link mode to avoid missing any
link interrupts.
Tested-by: Sven Auhagen <sven.auhagen@voleatech.de>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the phy_interface_mode_is_8023z() helper for detecting interface
modes that use 802.3z serial encoding. This is equivalent to testing
for both 1000base-X and 2500base-X.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for fixed-link configurations to nixge driver.
Signed-off-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make MDIO child optional and only instantiate the
MDIO bus if the child is actually present.
There are currently no (in-tree) users of this
binding; all (out-of-tree) users use overlays that
get shipped together with the FPGA images that contain
the IP.
This will significantly increase maintainabilty
of future revisions of this IP.
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
All users of the fixed_phy_add() pass -1 as GPIO number
to the fixed phy driver, and all users of fixed_phy_register()
pass -1 as GPIO number as well, except for the device
tree MDIO bus.
Any new users should create a proper device and pass the
GPIO as a descriptor associated with the device so delete
the GPIO argument from the calls and drop the code looking
requesting a GPIO in fixed_phy_add().
In fixed phy_register(), investigate the "fixed-link"
node and pick the GPIO descriptor from "link-gpios" if
this property exists. Move the corresponding code out
of of_mdio.c as the fixed phy code anyways requires
OF to be in use.
Tested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the tab before '}' and keep the code style consistent.
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The SH7763 SoC manual describes the Ether MAC's RX checksum offload
the same way as it's implemented in the EtherAVB MACs...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The SH7734 SoC manual describes the Ether MAC's RX checksum offload
the same way as it's implemented in the EtherAVB MACs...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The R-Car V3H (R8A77980) SoC manual describes the Ether MAC's RX checksum
offload the same way as it's implemented in the EtherAVB MAC...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The R-Mobile A1 (R8A7740) SoC manual describes the Ether MAC's RX checksum
offload the same way as it's implemented in the EtherAVB MAC...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: David S. Miller <davem@davemloft.net>
The RZ/A1H (R7S721000) SoC manual describes the Ether MAC's RX checksum
offload the same way as it's implemented in the EtherAVB MACs...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for the RX checksum offload. This is enabled by default and
may be disabled and re-enabled using 'ethtool':
# ethtool -K eth0 rx off
# ethtool -K eth0 rx on
Some Ether MACs provide a simple checksumming scheme which appears to be
completely compatible with CHECKSUM_COMPLETE: sum of all packet data after
the L2 header is appended to packet data; this may be trivially read by
the driver and used to update the skb accordingly. The same checksumming
scheme is implemented in the EtherAVB MACs and now supported by the 'ravb'
driver.
In terms of performance, throughput is close to gigabit line rate with the
RX checksum offload both enabled and disabled. The 'perf' output, however,
appears to indicate that significantly less time is spent in do_csum() --
this is as expected.
Test results with RX checksum offload enabled:
~/netperf-2.2pl4# perf record -a ./netperf -t TCP_MAERTS -H 192.168.2.4
TCP MAERTS TEST to 192.168.2.4
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
131072 16384 16384 10.01 933.93
[ perf record: Woken up 8 times to write data ]
[ perf record: Captured and wrote 1.955 MB perf.data (41940 samples) ]
~/netperf-2.2pl4# perf report
Samples: 41K of event 'cycles:ppp', Event count (approx.): 9915302763
Overhead Command Shared Object Symbol
9.44% netperf [kernel.kallsyms] [k] __arch_copy_to_user
7.75% swapper [kernel.kallsyms] [k] _raw_spin_unlock_irq
6.31% swapper [kernel.kallsyms] [k] default_idle_call
5.89% swapper [kernel.kallsyms] [k] arch_cpu_idle
4.37% swapper [kernel.kallsyms] [k] tick_nohz_idle_exit
4.02% netperf [kernel.kallsyms] [k] _raw_spin_unlock_irq
2.52% netperf [kernel.kallsyms] [k] preempt_count_sub
1.81% netperf [kernel.kallsyms] [k] tcp_recvmsg
1.80% netperf [kernel.kallsyms] [k] _raw_spin_unlock_irqres
1.78% netperf [kernel.kallsyms] [k] preempt_count_add
1.36% netperf [kernel.kallsyms] [k] __tcp_transmit_skb
1.20% netperf [kernel.kallsyms] [k] __local_bh_enable_ip
1.10% netperf [kernel.kallsyms] [k] sh_eth_start_xmit
Test results with RX checksum offload disabled:
~/netperf-2.2pl4# perf record -a ./netperf -t TCP_MAERTS -H 192.168.2.4
TCP MAERTS TEST to 192.168.2.4
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
131072 16384 16384 10.01 932.04
[ perf record: Woken up 14 times to write data ]
[ perf record: Captured and wrote 3.642 MB perf.data (78817 samples) ]
~/netperf-2.2pl4# perf report
Samples: 78K of event 'cycles:ppp', Event count (approx.): 18091442796
Overhead Command Shared Object Symbol
7.00% swapper [kernel.kallsyms] [k] do_csum
3.94% swapper [kernel.kallsyms] [k] sh_eth_poll
3.83% ksoftirqd/0 [kernel.kallsyms] [k] do_csum
3.23% swapper [kernel.kallsyms] [k] _raw_spin_unlock_irq
2.87% netperf [kernel.kallsyms] [k] __arch_copy_to_user
2.86% swapper [kernel.kallsyms] [k] arch_cpu_idle
2.13% swapper [kernel.kallsyms] [k] default_idle_call
2.12% ksoftirqd/0 [kernel.kallsyms] [k] sh_eth_poll
2.02% swapper [kernel.kallsyms] [k] _raw_spin_unlock_irqrestore
1.84% swapper [kernel.kallsyms] [k] __softirqentry_text_start
1.64% swapper [kernel.kallsyms] [k] tick_nohz_idle_exit
1.53% netperf [kernel.kallsyms] [k] _raw_spin_unlock_irq
1.32% netperf [kernel.kallsyms] [k] preempt_count_sub
1.27% swapper [kernel.kallsyms] [k] __pi___inval_dcache_area
1.22% swapper [kernel.kallsyms] [k] check_preemption_disabled
1.01% ksoftirqd/0 [kernel.kallsyms] [k] _raw_spin_unlock_irqrestore
The above results collected on the R-Car V3H Starter Kit board.
Based on the commit 4d86d38186 ("ravb: RX checksum offload")...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
