Saeed Mahameed says:
====================
mlx5-updates-2020-07-02
Rx and Tx devlink health reporters enhancements.
1) Code cleanup
2) devlink output format improvements
3) Print more useful info on devlink health diagnose output
4) TX timeout recovery, on a single SQ recover failure, stop the loop
and reset all rings (re-open netdev).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Upon a TX timeout handle, if the TX reporter was not able to recover
from the error, reopen the channels. If tried to reopen channels, do not
loop over TX queues for timeout.
With that, the reporters state and separation will better
expose the driver's state.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Add helper which retrieves the RQ WQE's head. Use this helper in RX
reporter diagnose callback.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Use txrx.h to contain helper function regarding TX/RX. In the coming
patches, I will add more RQ helpers.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Change the hierarchy of the RX reporter 'Common config' in the diagnose
output to match the 'Common config' of the TX reporter which reflects
that CQ is a helper to the traffic queues.
Before:
$ devlink health diagnose pci/0000:00:0b.0 reporter rx
Common config:
RQ:
type: 2 stride size: 2048 size: 8
CQ:
stride size: 64 size: 1024
RQs:
...
After:
$ devlink health diagnose pci/0000:00:0b.0 reporter rx
Common config:
RQ:
type: 2 stride size: 2048 size: 8
CQ:
stride size: 64 size: 1024
RQs:
...
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
When received a CQE error, the driver inspect the syndrome given by the
firmware. RQ recovery is initiated only as a result of a fatal syndrome;
syndrome which set the RQ into an error state. Hence no need to query
the RQ state at the beginning of the recovery process. Add additional
debug prints before recovering.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
During queue's recovery, driver waits for flush. The flush timeout is
set to 2 seconds. Add a define for this value for the benefit of RX and
TX reporters.
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Creation of devlink health reporters is not fatal for mlx5e instance load.
In case of error in reporter's creation, the return value is ignored.
Change all reporters creation functions to return void.
In addition, with this change, a failure in creating a reporter, will not
prevent the driver from trying to create the next reporter in the list.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Reviewed-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Edward Cree says:
====================
sfc: prerequisites for EF100 driver, part 3
Continuing on from [1] and [2], this series assembles the last pieces
of the common codebase that will be used by the forthcoming EF100
driver.
Patch #1 also adds a minor feature to EF10 (setting MTU on VFs) since
EF10 supports the same MCDI extension which that feature will use on
EF100.
Patches #5 & #7, while they should have no externally-visible effect
on driver functionality, change how that functionality is implemented
and how the driver represents TXQ configuration internally, so are
not mere cleanup/refactoring like most of these prerequisites have
(from the perspective of the existing sfc driver) been.
Changes in v2:
* Patch #1: use efx_mcdi_set_mtu() directly, instead of as a fallback,
in the mtu_only case (Jakub)
* Patch #3: fix symbol collision in non-modular builds by renaming
interrupt_mode to efx_interrupt_mode (kernel test robot)
* Patch #6: check for failure of netif_set_real_num_[tr]x_queues (Jakub)
* Patch #12: cleaner solution for ethtool drvinfo (Jakub, David)
[1]: https://lore.kernel.org/netdev/20200629.173812.1532344417590172093.davem@davemloft.net/T/
[2]: https://lore.kernel.org/netdev/20200630.130923.402514193016248355.davem@davemloft.net/T/
====================
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
We have yet another new scheme for NVRAM, and a corresponding new MCDI.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since ethtool_common.o will be built into both sfc and sfc_ef100 drivers,
it can't use KBUILD_MODNAME directly. Instead, make it reference a
string provided by the individual driver code.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Previously this was only happening in ef10-specific code.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
_down() merely removes all our filters and VLANs, it doesn't free
efx->filter_state itself.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since we only allocate VIs for the number of TXQs we actually need, we
cannot naively use "channel * TXQ_TYPES + txq" for the TXQ number, as
this has gaps (when efx->tx_queues_per_channel < EFX_TXQ_TYPES) and
thus overruns the driver's VI allocations, causing the firmware to
reject the MC_CMD_INIT_TXQ based on INSTANCE.
Thus, we distinguish INSTANCE (stored in tx_queue->queue) from LABEL
(tx_queue->label); the former is allocated starting from 0 in
efx_set_channels(), while the latter is simply the txq type (index in
channel->tx_queue array).
To simplify things, rather than changing tx_queues_per_channel after
setting up TXQs, make Siena always probe its HIGHPRI queues at start
of day, rather than deferring it until tc mqprio enables them.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Siena needs four TX queues (csum * highpri), EF10 needs two (csum),
and EF100 only needs one (as checksumming is controlled entirely by
the transmit descriptor). Rather than having various bits of ad-hoc
code to decide which queues to set up etc., put the knowledge of how
many TXQs a channel has in one place.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of exposing this old module parameter on the new driver (thus
having to keep it forever after for compatibility), let's confine it
to the old one; if we find later that we need the feature, we ought
to support it properly, with ethtool set-channels.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
EF100 only supports MSI-X, so there's no need for the new driver to
expose this old module parameter.
Since it's now visible to the linker, we have to rename it internally
to efx_interrupt_mode to avoid symbol collisions in non-modular
builds.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All NICs supported by this driver are capable of MSI-X interrupts (only
Falcon A1 wasn't, and that's now hived off into its own driver), so no
need for a nic-type parameter. Besides, the code that checked it was
buggy anyway (the following assignment that checked min_interrupt_mode
overrode it).
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Unprivileged functions (such as VFs) may set their MTU by use of the
'control' field of MC_CMD_SET_MAC_EXT, as used in efx_mcdi_set_mtu().
If calling efx_ef10_mac_reconfigure() from efx_change_mtu(), and the
NIC supports the above (SET_MAC_ENHANCED capability), use it rather
than efx_mcdi_set_mac().
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The variable act is being initialized with a value that is
never read and it is being updated later with a new value. The
initialization is redundant and can be removed.
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Alex Elder says:
====================
net: ipa: simplify endpoint programming
Add tests to functions so they don't update undefined endpoint
registers, rather than requiring the caller to avoid calling them.
Move the call to a workaround function required when suspending
inside the function that puts an endpoint into suspend mode. This
requires moving a few functions (which are otherwise unchanged).
Then simplify ipa_endpoint_program() to call essentially all
endpoint register update functions unconditionally.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Have functions that write endpoint configuration registers return
immediately if they are not valid for the direction of transfer for
the endpoint. This allows most of the calls in ipa_endpoint_program()
to be made unconditionally. Reorder the register writes to match
the order of their definition (based on offset).
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
IPA version 4.0+ does not support endpoint suspend. Put a test at
the top of ipa_endpoint_program_suspend() that returns immediately
if suspend is not supported rather than making that check in the caller.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
IPA version 3.5.1 has a hardware quirk that requires special
handling if an RX endpoint is suspended while aggregation is active.
This handling is implemented by ipa_endpoint_suspend_aggr().
Have ipa_endpoint_program_suspend() be responsible for calling
ipa_endpoint_suspend_aggr() if suspend mode is being enabled on
an endpoint. If the endpoint does not support aggregation, or if
aggregation isn't active, this call will continue to have no effect.
Move the definition of ipa_endpoint_suspend_aggr() up in the file so
its definition precedes the new earlier reference to it. This
requires ipa_endpoint_aggr_active() and ipa_endpoint_force_close()
to be moved as well.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
IPA version 4.2 has a hardware quirk that affects endpoint delay
mode, so it isn't used there. Isolate the test that avoids using
delay mode for that version inside ipa_endpoint_program_delay(),
rather than making that check in the caller.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
In certain configurations without power management support, the
following warnings happen:
drivers/net/ethernet/mellanox/mlx4/main.c:4388:12:
warning: 'mlx4_resume' defined but not used [-Wunused-function]
4388 | static int mlx4_resume(struct device *dev_d)
| ^~~~~~~~~~~
drivers/net/ethernet/mellanox/mlx4/main.c:4373:12: warning:
'mlx4_suspend' defined but not used [-Wunused-function]
4373 | static int mlx4_suspend(struct device *dev_d)
| ^~~~~~~~~~~~
Mark these functions as __maybe_unused to make it clear to the
compiler that this is going to happen based on the configuration,
which is the standard for these types of functions.
Fixes: 0e3e206a3e ("mlx4: use generic power management")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In certain configurations without power management support, gcc report
the following warning:
drivers/net/ethernet/micrel/ksz884x.c:7182:12: warning:
'pcidev_suspend' defined but not used [-Wunused-function]
7182 | static int pcidev_suspend(struct device *dev_d)
| ^~~~~~~~~~~~~~
Mark pcidev_suspend() as __maybe_unused to make it clear.
Fixes: 64120615d1 ("ksz884x: use generic power management")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Claudiu Beznea says:
====================
net: macb: few code cleanups
Patches in this series cleanup a bit macb code.
Changes in v2:
- in patch 2/4 use hweight32() instead of hweight_long()
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove is_udp variable that is used in only one place and use
ip_hdr(skb)->protocol == IPPROTO_UDP check instead.
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Bit 0 of queue_mask is set at the beginning of
macb_probe_queues() function. Do not set it again after reading
DGFG6 but instead use "|=" operator.
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Horatiu Vultur says:
====================
bridge: mrp: Add support for getting the status
This patch series extends the MRP netlink interface to allow the userspace
daemon to get the status of the MRP instances in the kernel.
v3:
- remove misleading comment
- fix to use correctly the RCU
v2:
- fix sparse warnings
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch extends the function br_fill_ifinfo to return also the MRP
status for each instance on a bridge. It also adds a new filter
RTEXT_FILTER_MRP to return the MRP status only when this is set, not to
interfer with the vlans. The MRP status is return only on the bridge
interfaces.
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Acked-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add MRP attribute IFLA_BRIDGE_MRP_INFO to allow the userspace to get the
current state of the MRP instances. This is a nested attribute that
contains other attributes like, ring id, index of primary and secondary
port, priority, ring state, ring role.
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Acked-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal says:
====================
mptcp: add receive buffer auto-tuning
First patch extends the test script to allow for reproducible results.
Second patch adds receive auto-tuning. Its based on what TCP is doing,
only difference is that we use the largest RTT of any of the subflows
and that we will update all subflows with the new value.
Else, we get spurious packet drops because the mptcp work queue might
not be able to move packets from subflow socket to master socket
fast enough. Without the adjustment, TCP may drop the packets because
the subflow socket is over its rcvbuffer limit.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
When mptcp is used, userspace doesn't read from the tcp (subflow)
socket but from the parent (mptcp) socket receive queue.
skbs are moved from the subflow socket to the mptcp rx queue either from
'data_ready' callback (if mptcp socket can be locked), a work queue, or
the socket receive function.
This means tcp_rcv_space_adjust() is never called and thus no receive
buffer size auto-tuning is done.
An earlier (not merged) patch added tcp_rcv_space_adjust() calls to the
function that moves skbs from subflow to mptcp socket.
While this enabled autotuning, it also meant tuning was done even if
userspace was reading the mptcp socket very slowly.
This adds mptcp_rcv_space_adjust() and calls it after userspace has
read data from the mptcp socket rx queue.
Its very similar to tcp_rcv_space_adjust, with two differences:
1. The rtt estimate is the largest one observed on a subflow
2. The rcvbuf size and window clamp of all subflows is adjusted
to the mptcp-level rcvbuf.
Otherwise, we get spurious drops at tcp (subflow) socket level if
the skbs are not moved to the mptcp socket fast enough.
Before:
time mptcp_connect.sh -t -f $((4*1024*1024)) -d 300 -l 0.01% -r 0 -e "" -m mmap
[..]
ns4 MPTCP -> ns3 (10.0.3.2:10108 ) MPTCP (duration 40823ms) [ OK ]
ns4 MPTCP -> ns3 (10.0.3.2:10109 ) TCP (duration 23119ms) [ OK ]
ns4 TCP -> ns3 (10.0.3.2:10110 ) MPTCP (duration 5421ms) [ OK ]
ns4 MPTCP -> ns3 (dead:beef:3::2:10111) MPTCP (duration 41446ms) [ OK ]
ns4 MPTCP -> ns3 (dead:beef:3::2:10112) TCP (duration 23427ms) [ OK ]
ns4 TCP -> ns3 (dead:beef:3::2:10113) MPTCP (duration 5426ms) [ OK ]
Time: 1396 seconds
After:
ns4 MPTCP -> ns3 (10.0.3.2:10108 ) MPTCP (duration 5417ms) [ OK ]
ns4 MPTCP -> ns3 (10.0.3.2:10109 ) TCP (duration 5427ms) [ OK ]
ns4 TCP -> ns3 (10.0.3.2:10110 ) MPTCP (duration 5422ms) [ OK ]
ns4 MPTCP -> ns3 (dead:beef:3::2:10111) MPTCP (duration 5415ms) [ OK ]
ns4 MPTCP -> ns3 (dead:beef:3::2:10112) TCP (duration 5422ms) [ OK ]
ns4 TCP -> ns3 (dead:beef:3::2:10113) MPTCP (duration 5423ms) [ OK ]
Time: 296 seconds
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The script generates two random files that are then sent via tcp and
mptcp connections.
In order to compare throughput over consecutive runs add an option
to provide the file size on the command line: "-f 128000".
Also add an option, -t, to enable tcp tests. This is useful to
compare throughput of mptcp connections and tcp connections.
Example: run tests with a 4mb file size, 300ms delay 0.01% loss,
default gso/tso/gro settings and with large write/blocking io:
mptcp_connect.sh -t -f $((4 * 1024 * 1024)) -d 300 -l 0.01% -r 0 -e "" -m mmap
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
