[ Upstream commit e8797a058466b60fc5a3291b92430c93ba90eaff ]
Clear the interrupt credits before enabling the queue rather
than after to be sure that the enabled queue starts at 0 and
that we don't wipe away possible credits after enabling the
queue.
Fixes: 0f3154e6bc ("ionic: Add Tx and Rx handling")
Signed-off-by: Neel Patel <neel.patel@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 280c0f7cd0aa4d190619b18243110e052a90775c ]
A problem about ionic create debugfs failed is triggered with the
following log given:
[ 415.799514] debugfs: Directory 'ionic' with parent '/' already present!
The reason is that ionic_init_module() returns ionic_bus_register_driver()
directly without checking its return value, if ionic_bus_register_driver()
failed, it returns without destroy the newly created debugfs, resulting
the debugfs of ionic can never be created later.
ionic_init_module()
ionic_debugfs_create() # create debugfs directory
ionic_bus_register_driver()
pci_register_driver()
driver_register()
bus_add_driver()
priv = kzalloc(...) # OOM happened
# return without destroy debugfs directory
Fix by removing debugfs when ionic_bus_register_driver() returns error.
Fixes: fbfb803153 ("ionic: Add hardware init and device commands")
Signed-off-by: Yuan Can <yuancan@huawei.com>
Acked-by: Shannon Nelson <snelson@pensando.io>
Link: https://lore.kernel.org/r/20221113092929.19161-1-yuancan@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit aa1d7e1267c12e07d979aa34c613716a89029db2 ]
It's possible that the driver will dereference a qcq that doesn't exist
when calling ionic_reconfigure_queues(), which causes a page fault BUG.
If a reduction in the number of queues is followed by a different
reconfig such as changing the ring size, the driver can hit a NULL
pointer when trying to clean up non-existent queues.
Fix this by checking to make sure both the qcqs array and qcq entry
exists bofore trying to use and free the entry.
Fixes: 101b40a017 ("ionic: change queue count with no reset")
Signed-off-by: Brett Creeley <brett@pensando.io>
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Link: https://lore.kernel.org/r/20221017233123.15869-1-snelson@pensando.io
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 19058be7c48ceb3e60fa3948e24da1059bd68ee4 ]
Assign a random mac address to the VF interface station
address if it boots with a zero mac address in order to match
similar behavior seen in other VF drivers. Handle the errors
where the older firmware does not allow the VF to set its own
station address.
Newer firmware will allow the VF to set the station mac address
if it hasn't already been set administratively through the PF.
Setting it will also be allowed if the VF has trust.
Fixes: fbb39807e9 ("ionic: support sr-iov operations")
Signed-off-by: R Mohamed Shah <mohamed@pensando.io>
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0fc4dd452d6c14828eed6369155c75c0ac15bab3 ]
In looping on FW update tests we occasionally see the
FW_ACTIVATE_STATUS command fail while it is in its EAGAIN loop
waiting for the FW activate step to finsh inside the FW. The
firmware is complaining that the done bit is set when a new
dev_cmd is going to be processed.
Doing a clean on the cmd registers and doorbell before exiting
the wait-for-done and cleaning the done bit before the sleep
prevents this from occurring.
Fixes: fbfb803153 ("ionic: Add hardware init and device commands")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 9cb9dadb8f45c67e4310e002c2f221b70312b293 ]
There is a case found in heavy testing where a link flap happens just
before a firmware Recovery event and the driver gets stuck in the
BROKEN state. This comes from the driver getting interrupted by a FW
generation change when coming back up from the link flap, and the call
to ionic_start_queues() in ionic_link_status_check() fails. This can be
addressed by having the fw_up code clear the BROKEN bit if seen, rather
than waiting for a user to manually force the interface down and then
back up.
Fixes: 9e8eaf8427 ("ionic: stop watchdog when in broken state")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 2624d95972dbebe5f226361bfc51a83bdb68c93b ]
Widen the coverage of the queue_lock to be sure the lif init
and lif deinit actions are protected. This addresses a hang
seen when a Tx Timeout action was attempted at the same time
as a FW Reset was started.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit a86e86db5e6d72c82724a63ca1c5293409a21518 ]
The prototype of input features of ionic_set_nic_features() is
netdev_features_t, but the vlan_flags is using the private
definition of ionic drivers. It should use the variable
ctx.cmd.lif_setattr.features, rather than features to check
the vlan flags. So fixes it.
Fixes: beead698b1 ("ionic: Add the basic NDO callbacks for netdev support")
Signed-off-by: Jian Shen <shenjian15@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Acked-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b8fd0271dad00b953caaabe73474788d3d19e252 ]
It's possible the FW is already shutting down while the driver is being
removed and/or when the driver is going through reset. This can cause
unexpected/unnecessary errors to be printed:
eth0: DEV_CMD IONIC_CMD_PORT_RESET (12) error, IONIC_RC_ERROR (29) failed
eth1: DEV_CMD IONIC_CMD_RESET (3) error, IONIC_RC_ERROR (29) failed
Fix this by checking the FW status register before issuing the reset
commands.
Also, since err may not be assigned in ionic_port_reset(), assign it a
default value of 0, and remove an unnecessary log message.
Fixes: fbfb803153 ("ionic: Add hardware init and device commands")
Signed-off-by: Brett Creeley <brett@pensando.io>
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 9ad2939a1525962a79a2fd974ec7e3a71455b964 ]
The watchdog expects the lif to fully exist when it goes off,
so lets not start the watchdog until all is ready in case there
is some quirky time dialation that makes probe take multiple
seconds.
Fixes: 089406bc5a ("ionic: add a watchdog timer to monitor heartbeat")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit bc0bf9de6f48268f4ee59e57fb42ac751be3ecda ]
Sparse seems to have gotten a little more picky lately and
we need to revisit this bit of code to make sparse happy.
warning: incorrect type in initializer (different address spaces)
expected union ionic_dev_cmd_regs *regs
got union ionic_dev_cmd_regs [noderef] __iomem *dev_cmd_regs
warning: incorrect type in argument 2 (different address spaces)
expected void [noderef] __iomem *
got unsigned int *
warning: incorrect type in argument 1 (different address spaces)
expected void volatile [noderef] __iomem *
got union ionic_dev_cmd *
Fixes: d701ec326a ("ionic: clean up sparse complaints")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Bridging, and possibly other upper stack gizmos, adds the
lower device's netdev->dev_addr to its own uc list, and
then requests it be deleted when the upper bridge device is
removed. This delete request also happens with the bridging
vlan_filtering is enabled and then disabled.
Bonding has a similar behavior with the uc list, but since it
also uses set_mac to manage netdev->dev_addr, it doesn't have
the same the failure case.
Because we store our netdev->dev_addr in our uc list, we need
to ignore the delete request from dev_uc_sync so as to not
lose the address and all hope of communicating. Note that
ndo_set_mac_address is expressly changing netdev->dev_addr,
so no limitation is set there.
Fixes: 2a654540be ("ionic: Add Rx filter and rx_mode ndo support")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the setting of the filter-sync-needed bit to the error
case in the filter add routine to be sure we're checking the
live filter status rather than a copy of the pre-sync status.
Fixes: 969f843946 ("ionic: sync the filters in the work task")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't print stats for which we haven't reserved space as it can
cause nasty memory bashing and related bad behaviors.
Fixes: aa620993b1 ("ionic: pull per-q stats work out of queue loops")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Deadlock seen in an instance where the hwstamp configuration
is changed while the driver is running:
[ 3988.736671] schedule_preempt_disabled+0xe/0x10
[ 3988.736676] __mutex_lock.isra.5+0x276/0x4e0
[ 3988.736683] __mutex_lock_slowpath+0x13/0x20
[ 3988.736687] ? __mutex_lock_slowpath+0x13/0x20
[ 3988.736692] mutex_lock+0x2f/0x40
[ 3988.736711] ionic_stop_queues_reconfig+0x16/0x40 [ionic]
[ 3988.736726] ionic_reconfigure_queues+0x43e/0xc90 [ionic]
[ 3988.736738] ionic_lif_config_hwstamp_rxq_all+0x85/0x90 [ionic]
[ 3988.736751] ionic_lif_hwstamp_set_ts_config+0x29c/0x360 [ionic]
[ 3988.736763] ionic_lif_hwstamp_set+0x76/0xf0 [ionic]
[ 3988.736776] ionic_eth_ioctl+0x33/0x40 [ionic]
[ 3988.736781] dev_ifsioc+0x12c/0x420
[ 3988.736785] dev_ioctl+0x316/0x720
This can be demonstrated with "ptp4l -m -i <intf>"
To fix this, we pull the use of the queue_lock further up above the
callers of ionic_reconfigure_queues() and ionic_stop_queues_reconfig().
Fixes: 7ee99fc5ed ("ionic: pull hwstamp queue_lock up a level")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
The queues can be freed in ionic_close(). They need to be recreated
after ionic_open(). It doesn't need to replay the whole config. It
only needs to create the timestamping queues again.
Signed-off-by: Allen Hubbe <allenbh@pensando.io>
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the hwstamp configuration use of queue_lock up
a level to simplify use and error handling.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the queue configuration lock to ionic_open() and
ionic_stop() so that they don't collide with other in parallel
queue configuration actions such as MTU changes as can be
demonstrated with a tight loop of ifup/change-mtu/ifdown.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make sure the ctx struct has the new mac address before
any save operations happen.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since the heartbeat check will already have complained about
the firmware status, don't bother complaining about the
DEVCMD failing. We'll keep the print message but demote it
to a debug messages so that we normally no longer see it.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
In some cases of fw_down it was called because there was a
fw_generation change, and the firmware is already back up.
In order to keep the down time to a minimum, don't wait for
the next watchdog polling cycle, fire another watchdog off
as soon as we can - an out-of-cycle check won't hurt, and
may well speed up the recovery.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make sure we go into PROMISC mode when we have too many
filters by specifically counting the filters that successfully
get saved to the firmware.
The device advertises max_ucast_filters and max_mcast_filters,
but really only has max_ucast_filters slots available for
uc and mc filters combined.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
The filter counting in ionic_lif_addr() really isn't useful,
and potentially misleading, especially when we're checking in
ionic_lif_rx_mode() to see if we need to go into PROMISC mode.
We can safely refactor this and remove a calling layer.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to separate the atomic needs of __dev_uc_sync()
and __dev_mc_sync() from the safe rx_mode handling, we need
to have the ndo handler manipulate the driver's filter list,
and later have the driver sync the filters to the firmware,
outside of the atomic context.
Here we put __dev_mc_sync() and __dev_uc_sync() back into the
ndo callback to give them their netif_addr_lock context and
have them update the driver's filter list, flagging changes
that should be made to the device filter list. Later, in the
rx_mode handler, we read those hints and sync up the device's
list as needed.
It is possible for multiple add/delete requests to come from
the stack before the rx_mode task processes the list, but the
handling of the sync status flag should keep everything sorted
correctly. For example, if a delete of an existing filter is
followed by another add before the rx_mode task is run, as can
happen when going in and out of a bond, the add will cancel
the delete and no actual changes will be sent to the device.
We also add a check in the watchdog to see if there are any
stray unsync'd filters, possibly left over from a filter
overflow and waiting to get sync'd after some other filter
gets removed to make room.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since only two functions call through ionic_set_rx_mode(), one
that can sleep and one that can't, we can split the function
and put the bits of code into the callers. This removes an
unnecessary calling layer.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
With the move of mac filter handling to outside of the
ndo_rx_mode context using the IONIC_DW_TYPE_RX_MODE,
we no longer are using IONIC_DW_TYPE_RX_ADDR_ADD and
IONIC_DW_TYPE_RX_ADDR_DEL and they can be removed.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to support more coalesce parameters through netlink,
add two new parameter kernel_coal and extack for .set_coalesce
and .get_coalesce, then some extra info can return to user with
the netlink API.
Signed-off-by: Yufeng Mo <moyufeng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
All kernel devlink implementations call to devlink_alloc() during
initialization routine for specific device which is used later as
a parent device for devlink_register().
Such late device assignment causes to the situation which requires us to
call to device_register() before setting other parameters, but that call
opens devlink to the world and makes accessible for the netlink users.
Any attempt to move devlink_register() to be the last call generates the
following error due to access to the devlink->dev pointer.
[ 8.758862] devlink_nl_param_fill+0x2e8/0xe50
[ 8.760305] devlink_param_notify+0x6d/0x180
[ 8.760435] __devlink_params_register+0x2f1/0x670
[ 8.760558] devlink_params_register+0x1e/0x20
The simple change of API to set devlink device in the devlink_alloc()
instead of devlink_register() fixes all this above and ensures that
prior to call to devlink_register() everything already set.
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Prefix the log output with the function string as in other
debug messages.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
If there's only one queue, there is no need to enable
the rxhashing.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are a few things that we can't safely do when the fw is
resetting, as the driver may be in the middle of rebuilding
queue structures.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
We don't use these fields, so remove them from
the definition.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the new VF to our internal count before we start configuring it.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Initialize err to 0 instead of ENOMEM, and specifically set
err to ENOMEM in the devm_kcalloc() failure cases.
Also, add an error message to the end of reconfig.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Print the version of the DSC firmware seen when we do a fresh
ident check. Because the FW can be updated by the external
orchestration system, this helps us track that FW has been
updated on the DSC.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
The top 4 bits of the fw_status in dev_info_regs is reserved
for the status generation. This generation number is an
arbitrary value defined when firmware starts up. If the FW
is killed/crashed/stopped and then restarted, it will create
a different generation number. With this mechanism, the host
driver can detect that the FW has crashed and restarted, and
the driver can then take steps to re-initialize its connection.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
When running in a small kdump kernel, we can play nice and
minimize our resource use to help make sure that kdump is
successful in its mission.
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Be sure to count the csum_none cases when csum offload is
enabled.
Fixes: 0f3154e6bc ("ionic: Add Tx and Rx handling")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to count the correct Tx and/or Rx packets for dynamic
interrupt moderation, depending on which we're processing on
the queue interrupt.
Fixes: 04a834592b ("ionic: dynamic interrupt moderation")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the interrupt coalesce value update out of the napi
thread and into the dim_work thread and set it only when it
has actually changed.
Fixes: 04a834592b ("ionic: dynamic interrupt moderation")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
If PTP configuration is attempted on ports that don't support
it, such as VF ports, the driver will return an error status
-95, or EOPNOSUPP and print an error message
enp98s0: hwstamp set failed: -95
Because some daemons can retry every few seconds, this can end
up filling the dmesg log and pushing out other more useful
messages.
We can catch this issue earlier in our handling and return
the error without a log message.
Fixes: 829600ce5e ("ionic: add ts_config replay")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the bulk of the code from ionic_set_rx_mode(), which
can be called from atomic context, into ionic_lif_rx_mode()
which is a safe context.
A call from the stack will get pushed off into a work thread,
but it is also possible to simultaneously have a call driven
by a queue reconfig request from an ethtool command or fw
recovery event. We add a mutex around the rx_mode work to be
sure they don't collide.
Fixes: 81dbc24147 ("ionic: change set_rx_mode from_ndo to can_sleep")
Signed-off-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
The failure to register devlink will leave the system with dangled
devlink resource, which is not cleaned if devlink_port_register() fails.
In order to remove access to ".registered" field of struct devlink_port,
require both devlink_register and devlink_port_register to success and
check it through device pointer.
Fixes: fbfb803153 ("ionic: Add hardware init and device commands")
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Acked-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
The driver core will call to .remove callback only if .probe succeeded
and it will ensure that driver data has pointer to struct ionic.
There is no need to check it again.
Fixes: fbfb803153 ("ionic: Add hardware init and device commands")
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Acked-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: David S. Miller <davem@davemloft.net>
