The 'MODULE_FIBER' value replaced several other FIBER values
in newer management firmware images, so existing code would
fail to properly reflect its mode.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Driver has 2 sets of entries for handling ramrod configurations
toward firmware - a regular pre-allocated set of entires and a
possible 'unlimited' list of additional pending entries.
In most scenarios the 'unlimited' list would not be used, but
when it does the handling of the ramrod completion doesn't
properly handle the release of the entry.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several user APIs can cause driver to perform an inner-reload.
Currently, doing this would cause the HW/FW statistics of the
adapter to reset, which isn't the expected behavior [statistics
should only reset on explicit unloads].
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Internal loopback in driver is based on two things - first
is the willingness of transmitter to use it [in case of VFs,
this can be forced based on VEPA/VEB] and secondly on another
vport classification configuration which should match the
packet's destination.
Current code allows non-linux VFs to configure a 'promisc'
mode on Tx, meaning all traffic sent by VF would be loopbacked
internally by firmware; This isn't considered a valid mode and
as such should be prevented by PF.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When no vlan filter is configured, firmware has a configurable
default on whether to pass only untagged packets or all packets
regardless of their tagging. Driver currently doesn't set this
field in the necessary ramrod, causing the default to always be
'receive all'.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The conversion to the 64-bit time based ptp methods left two instances
of 'struct timespec' in place. This is harmless because 64-bit
architectures define timespec64 as timespec, and this driver is
not used on 32-bit machines.
However, using 'struct timespec64' directly is obviously the right
thing to do, and will help us remove 'struct timespec' in the future.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: b9acf24f77 ("ptp: tilegx: convert to the 64 bit get/set time methods.")
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stop the SW TX counter from counting the TX header bytes
since they are not being sent out.
Fixes: e577516b9d ("mlxsw: Fix use-after-free bug in mlxsw_sx_port_xmit")
Signed-off-by: Nogah Frankel <nogahf@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stop the SW TX counter from counting the TX header bytes
since they are not being sent out.
Fixes: 56ade8fe3f ("mlxsw: spectrum: Add initial support for Spectrum ASIC")
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Nogah Frankel <nogahf@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We were using an incorrect define to get the irq vector number.
NFP_NET_CFG_LSC is a control BAR offset, LSC interrupt vector
index is called NFP_NET_IRQ_LSC_IDX. For machines with less
than 30 CPUs this meant that we were disabling/enabling IRQ 0.
For bigger hosts we were just playing with the 31st RX/TX
interrupt.
Fixes: 0ba40af963 ("nfp: move link state interrupt request/free calls")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The mlx4e driver does not support more than one port for VXLAN offload. As
such expecting the hardware to offload other ports is invalid since it
appears the parsing logic is used to perform Tx checksum and segmentation
offloads. Use the vxlan_port number to determine in which cases we can
apply the offload and in which cases we can not.
Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When building with -Wextra, we get a harmless warning from the
EFX_EXTRACT_OWORD32 macro:
ethernet/sfc/farch.c: In function 'efx_farch_test_registers':
ethernet/sfc/farch.c:119:30: error: comparison of unsigned expression < 0 is always false [-Werror=type-limits]
ethernet/sfc/farch.c:124:144: error: comparison of unsigned expression < 0 is always false [-Werror=type-limits]
ethernet/sfc/farch.c:124:392: error: comparison of unsigned expression < 0 is always false [-Werror=type-limits]
ethernet/sfc/farch.c:124:731: error: comparison of unsigned expression < 0 is always false [-Werror=type-limits]
The macro and the caller are both correct, but we can avoid the
warning by changing the index variable to a signed type.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Bert Kenward <bkenward@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Maciej Żenczykowski reported lockdep warning a spinlock
was not registered before being held in mlx4_cmd_wake_completions()
cmd.context_lock initialization is not at the right place.
1) mlx4_cmd_use_events() can be called multiple times.
Calling spin_lock_init() on a live spinlock can lead
to hangs.
2) mlx4_cmd_wake_completions() can be called while lock
has not been initialized.
Lockdep complains, and current logic is not race prone.
It seems better to move the initialization earlier in
mlx4_load_one()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Maciej Żenczykowski <maze@google.com>
Cc: Eugenia Emantayev <eugenia@mellanox.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 26c5f03 uses a new skb allocator to avoid the RFD overflow
issue.
But from debugging without datasheet, we found the error always
happen when the DMA RX address is set to 0x....fc0, which is very
likely to be a HW/silicon problem.
So one idea is instead of adding a new allocator, why not just
hitting the right target by avaiding the error-prone DMA address?
This patch will actually
* Remove the commit 26c5f03
* Apply rx skb with 64 bytes longer space, and if the allocated skb
has a 0x...fc0 address, it will use skb_resever(skb, 64) to
advance the address, so that the RX overflow can be avoided.
In theory this method should also apply to atl1c driver, which
I can't find anyone who can help to test on real devices.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=70761
Signed-off-by: Feng Tang <feng.tang@intel.com>
Suggested-by: Eric Dumazet <edumazet@google.com>
Tested-by: Ole Lukoie <olelukoie@mail.ru>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no reason to destroy channels that are destroyed while
cpdma_ctlr destroy. In this case no need to remember how much
channels where created and destroy them by one, as cpdma_ctlr
destroys all of them.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 7f854420fb
("phy: Add API for {un}registering an mdio device to a bus.")
broke PHY detection on this driver with a copy-paste bug:
The code is looking 32 times for a PHY at address 0.
Fixes ethernet on AMD DB1100/DB1500/DB1550 boards which have
their (autodetected) PHYs at address 31.
Cc: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Manuel Lauss <manuel.lauss@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 7f854420fb
("phy: Add API for {un}registering an mdio device to a bus.")
broke PHY detection on this driver with a copy-paste bug:
The code is looking 32 times for a PHY at address 0.
Fixes ethernet on AMD DB1100/DB1500/DB1550 boards which have
their (autodetected) PHYs at address 31.
Cc: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Manuel Lauss <manuel.lauss@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The code checks if the queue should be stopped because we are below the
threshold of free descriptors only to check if it should be started again.
If we do end up in a state where we are at the threshold limit, it makes
more sense to just stop the queue and wait for the next IRQ to trigger the
TX housekeeping again. There is no rush in enqueuing the next packet, it
needs to wait for all the others in the queue to be dispatched first
anyway.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current code unconditionally wakes up the queue at the end of each
tx_poll action. Change the code to only wake up the queues if any of
them have actually been stopped before.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The TX ring setup has an off by one error causing it to not utilise all
descriptors. This has the side effect that we need to reset the next
pointer at runtime to make it work. Fix the off by one and remove the
code fixing the ring at runtime.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
During stress testing, after reducing the threshold value, we have seen
TX timeouts that were caused by the watchdog_timeo value being too low.
Increase the value to 5 * HZ which is a value commonly used by many other
drivers.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The logic to calculate the threshold value for stopping the TX queue is
bad. Currently it will always use 1/2 of the rings size, which is way too
much. Set the threshold to MAX_SKB_FRAGS. This makes sure that the queue
is stopped when there is not enough room to accept an additional segment.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current code only disables those IRQs that we will later use. To
ensure that we have a predefined state, we really want to disable all IRQs.
Change the code to disable all IRQs to achieve this.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The QDMA engine can fail to update the register pointing to the next TX
descriptor if this bit does not get set in the QDMA configuration register.
Not setting this bit can result in invalid values inside the TX rings
registers which will causes TX stalls.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are two places inside mtk_poll_rx where rx_dropped is not being
incremented properly. Fix this by adding the missing code to increment
the counter.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The lookup of the tx_buffer in the error path inside mtk_tx_map() uses the
wrong descriptor pointer. This looks like a copy & paste error. Change the
code to use the correct pointer.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Scratch memory gets allocated in mtk_init_fq_dma() but the corresponding
code to free it is missing inside mtk_dma_free() causing a memory leak.
With this patch applied, we can run ifconfig up/down several thousand
times without any problems.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The code fails to check if the scratch memory was properly allocated. Add
this check and return with an error if the allocation failed.
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
The dwmac4_set_umac_addr() takes a struct mac_device_info as
the first parameter, but is being passed a ioaddr instead from
dwmac4_set_filter(). Fix the warning/bug by changing the first
parameter.
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c:159:46: warning: incorrect type in argument 1 (different address spaces)
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c:159:46: expected struct mac_device_info *hw
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c:159:46: got void [noderef] <asn:2>*ioaddr
Note, only compile tested this as do not have any
hardware with it in.
Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Acked-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Blue flame is a latency enhancement feature that allows the driver to
write the packet data directly to the NIC's registers thus making the
read of the packet data from host memory redundant.
We maintain a quota for the blue flame which is reloaded whenever we
identify that the hardware is processing send requests and processes
them fast enough so by the time we post the next send request it was
able to process all the pending ones. This indicates that the hardware
is capable of processing more blue flame requests efficiently. The blue
flame quota is decremented whenever we send using blue flame.
The current code erroneously clears the budget if we did not use blue
flame for the current post send operation and we fix it here.
Fixes: 88a85f99e5 ('net/mlx5e: TX latency optimization to save DMA reads')
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current implementation copies the flow of ndo_stop instead of
calling it explicitly, Fixed it.
Fixes: 5fc7197d3a ("net/mlx5: Add pci shutdown callback")
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Set the mc_promisc flag also in the case of adding new mc address to
existing allmulti vport.
Fixes: a35f71f27a ('net/mlx5: E-Switch, Implement promiscuous rx modes vf request handling')
Signed-off-by: Mohamad Haj Yahia <mohamad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In RoCE, the RDMA-CM needs the node guid to establish connection
between nodes.
Today, the node guid exposed to mlx5 Ethernet VFs is zero, therefore
RDMA-CM on the VF is broken.
Whenever the administrator sets a MAC for a VF, derive the node guid
from it and set it as well in the following way:
MAC: e4:1d:2d:b3:f4:01 -> node_guid: e4:1d:2d:ff:fe:b3:f4:01
Fixes: 77256579c6 ('net/mlx5: E-Switch, Introduce Vport...')
Signed-off-by: Noa Osherovich <noaos@mellanox.com>
Signed-off-by: Majd Dibbiny <majd@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reorder vport enable flow to mark the vport as enabled before calling
the vport change handler which was modified to handle the case for
when vport is not enabled.
This fixes the case for when the PF netdev is open before sriov is
enabled, once sriov is enabled at esw_enable_vport,
esw_vport_change_handle_locked didn't read the PF context since it
thought the PF vport was not enabled.
When we enable the vport, arming for events is not required anymore,
since it's done on the vport change handle
Fixes: 586cfa7f1d ('net/mlx5: E-Switch, Use vport event handler for vport cleanup')
Signed-off-by: Mohamad Haj Yahia <mohamad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The mlx5 flow-steering API (mlx5_create_flow_table/group/rule) never
returns null pointer on error. Even if it was doing that, checking
for IS_ERR_OR_NULL(p) and then returning PTR_ERR(p) would have cause
bugs, since PTR_ERR(NULL) --> success, crash.
To make things more robust and protect against related future bugs,
convert all IS_ERR_OR_NULL checks on returned values to IS_ERR.
Fixes: 5742df0f7d ('net/mlx5: E-Switch, Introduce VST vport ingress/egress ACLs')
Fixes: 86d722ad2c ('net/mlx5: Use flow steering infrastructure for mlx5_en')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reported-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We must use kvfree() for something that could have been allocated with vzalloc(),
do that.
Fixes: 5742df0f7d ('net/mlx5: E-Switch, Introduce VST vport ingress/egress ACLs')
Fixes: 86d722ad2c ('net/mlx5: Use flow steering infrastructure for mlx5_en')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reported-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add missing capabilities check for E-Switch FDB and ACLs flow
tables before creating their namespace in flow steering.
Fixes: efdc810ba3 ('net/mlx5: Flow steering, Add vport ACL support')
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Flow steering infrastructure is currently used only on link layer
ethernet, therefore the driver should initialize the flow steering
when the device link layer is ethernet.
In addition, add missing capability check before initializing the
namespace of NIC RX flow tables.
Fixes: 2530236303 ('net/mlx5_core: Flow steering tree initialization')
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When we destroy the last flow table we need to update
the root_ft to NULL.
It fixes an issue for when the last flow table is destroyed
and recreated again, root_ft pointer will not be updated,
as a result traffic will be dropped.
Fixes: 2cc43b494a ('net/mlx5_core: Managing root flow table')
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Mask the reserved bits when reading the number of newly
created XRCD.
Fixes: e126ba97db ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Majd Dibbiny <majd@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When rtnl_fill_ifinfo() is called for a certain netdevice it queries its
various parameters such as switch id and physical port name. The
function might get called in an atomic context, which means the
underlying driver must not sleep during the query operation.
Don't query the device and sleep during ndo_get_phys_port_name(), but
instead store the needed parameters in port creation time.
Fixes: 2bf9a58675 ("mlxsw: spectrum: Add support for physical port names")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a port is created following a split / unsplit we need to map it to
the correct module and lane, enable it and then continue to initialize
its various parameters such as MTU and VLAN filters.
Under certain conditions, such as trying to split ports at the bottom
row of the front panel by four, we get firmware errors.
After evaluating this with the firmware team it was decided to alter the
split / unsplit flow, so that first all the affected ports are mapped,
then enabled and finally each is initialized separately.
Fix the split / unsplit flow by first mapping and enabling all the
affected ports. Newer firmware versions will support both flows.
Fixes: 18f1e70c41 ("mlxsw: spectrum: Introduce port splitting")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
7000-series SFC NICs connected with an SFP+ module currently fail to
report any supported link speeds.
Reported-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
Reviewed-by: Jarod Wilson <jarod@redhat.com>
Tested-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
trivial fix to spelling mistakes and add missing newline in pr_err
messages
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Acked-by: Fugang Duan <fugang.duan@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since both CTAG and STAG rx acceleration must be enabled together, we
only need to check one feature flag (NETIF_F_HW_VLAN_CTAG_RX) before
calling __vlan_hwaccel_put_tag().
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The hardware can only be set to strip or not strip both the VLAN CTAG and
STAG. It cannot strip one and not strip the other. Add logic to
bnxt_fix_features() to toggle both feature flags when the user is toggling
one of them.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Set the is_push flag in the software BD before the tx data is pushed to
the chip. It is possible to get the tx interrupt as soon as the tx data
is pushed. The tx handler will not handle the event properly if the
is_push flag is not set and it will crash.
Signed-off-by: Michael Chan <michael.chan@broadocm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since implementing VLAN filtering in commit 05cc5a39dd
("bnx2x: add vlan filtering offload") bnx2x refuses to add a VLAN while
the interface is down:
# ip link add link enp3s0f0 enp3s0f0_10 type vlan id 10
RTNETLINK answers: Bad address
and in dmesg (with bnx2x.debug=0x20):
bnx2x: [bnx2x_vlan_rx_add_vid:12941(enp3s0f0)]Ignoring VLAN
configuration the interface is down
Other drivers have no problem with this.
Fix this peculiar behavior in the following way:
- Accept requests to add/kill VID regardless of the device state.
Maintain the requested list of VIDs in the bp->vlan_reg list.
- If the device is up, try to configure the VID list into the hardware.
If we run out of VLAN credits or encounter a failure configuring an
entry, fall back to accepting all VLANs.
If we successfully configure all entries from the list, turn the
fallback off.
- Use the same code for reconfiguring VLANs during NIC load.
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Acked-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If an external Gigabit PHY is connected to either of the MACs we need to
be able to tell the PHY to use a delay. Not doing so will result in heavy
packet loss and/or data corruption when using PHYs such as the IC+ IP1001.
We tell the PHY which MII delay mode to use via the devictree.
The ethernet driver needs to be adapted to handle all 3 rgmii-*id modes
in the same way as normal rgmii when setting up the MAC.
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
