Commit graph

112,392 commits

Author SHA1 Message Date
Roi Dayan
fad5479069 net/mlx5e: Add tc action infrastructure
Add an infrastructure to help parsing tc actions in a generic way.

Supporting an action parser means implementing struct mlx5e_tc_act
for that action.

The infrastructure will give the possibility to be generic when parsing tc
actions, i.e. parse_tc_nic_actions() and parse_tc_fdb_actions().
To parse tc actions a user needs to allocate a parse_state instance
and pass it when iterating over the tc actions parsers.
If a parser doesn't exists then a user can treat it as unsupported.

To add an action parser a user needs to implement two callbacks.
The can_offload() callback to quickly check if an action can be offloaded.
The parse_action() callback to do actual parsing and prepare for offload.

Add implementation for drop, trap, mark and accept action parsers with this
commit to act as examples and implement usage of the new infrastructure for
those actions.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-12-14 21:29:41 -08:00
Kurt Kanzenbach
6cf01e4515 net: dsa: hellcreek: Add missing PTP via UDP rules
The switch supports PTP for UDP transport too. Therefore, add the missing static
FDB entries to ensure correct forwarding of these packets.

Fixes: ddd56dfe52 ("net: dsa: hellcreek: Add PTP clock support")
Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-14 18:46:29 -08:00
Kurt Kanzenbach
cad1798d2d net: dsa: hellcreek: Allow PTP P2P measurements on blocked ports
Allow PTP peer delay measurements on blocked ports by STP. In case of topology
changes the PTP stack can directly start with the correct delays.

Fixes: ddd56dfe52 ("net: dsa: hellcreek: Add PTP clock support")
Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-14 18:46:28 -08:00
Kurt Kanzenbach
b7ade35eb5 net: dsa: hellcreek: Add STP forwarding rule
Treat STP as management traffic. STP traffic is designated for the CPU port
only. In addition, STP traffic has to pass blocked ports.

Fixes: e4b27ebc78 ("net: dsa: Add DSA driver for Hirschmann Hellcreek switches")
Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-14 18:46:28 -08:00
Kurt Kanzenbach
4db4c3ea56 net: dsa: hellcreek: Fix insertion of static FDB entries
The insertion of static FDB entries ignores the pass_blocked bit. That bit is
evaluated with regards to STP. Add the missing functionality.

Fixes: e4b27ebc78 ("net: dsa: Add DSA driver for Hirschmann Hellcreek switches")
Signed-off-by: Kurt Kanzenbach <kurt@linutronix.de>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-12-14 18:46:28 -08:00
Kees Cook
c2ed5611af iw_cxgb4: Use memset_startat() for cpl_t5_pass_accept_rpl
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memset(), avoid intentionally writing across
neighboring fields.

Use memset_startat() so memset() doesn't get confused about writing beyond
the destination member that is intended to be the starting point of
zeroing through the end of the struct. Additionally, since everything
appears to perform a roundup (including allocation), just change the size
of the struct itself and add a build-time check to validate the expected
size.

Link: https://lore.kernel.org/r/20211213223331.135412-13-keescook@chromium.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-12-14 20:21:22 -04:00
Karol Kolacinski
37e738b6fd ice: Don't put stale timestamps in the skb
The driver has to check if it does not accidentally put the timestamp in
the SKB before previous timestamp gets overwritten.
Timestamp values in the PHY are read only and do not get cleared except
at hardware reset or when a new timestamp value is captured.
The cached_tstamp field is used to detect the case where a new timestamp
has not yet been captured, ensuring that we avoid sending stale
timestamp data to the stack.

Fixes: ea9b847cda ("ice: enable transmit timestamps for E810 devices")
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
2021-12-14 11:31:47 -08:00
Karol Kolacinski
0013881c11 ice: Use div64_u64 instead of div_u64 in adjfine
Change the division in ice_ptp_adjfine from div_u64 to div64_u64.
div_u64 is used when the divisor is 32 bit but in this case incval is
64 bit and it caused incorrect calculations and incval adjustments.

Fixes: 06c16d89d2 ("ice: register 1588 PTP clock device object for E810 devices")
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
2021-12-14 11:29:34 -08:00
David Mosberger-Tang
09ed8bfc52 wilc1000: Rename workqueue from "WILC_wq" to "NETDEV-wq"
This follows normal Linux convention and is more useful since the new
name will make it apparent which network device the work-queue is for
(e.g., the name will be "wlan0-wq" for network device "wlan0").

hif_workqueue allocation has to move from
cfg80211.c:wilc_cfg80211_init() to netdev.c:wilc_netdev_ifc_init()
because the network device name is not known until after the netdev is
registered.  The move also makes sense because netdev.c is already
responsible for destroying the work queue when it is no longer needed.

Signed-off-by: David Mosberger-Tang <davidm@egauge.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209044411.3482259-5-davidm@egauge.net
2021-12-14 20:46:35 +02:00
David Mosberger-Tang
3cc23932ba wilc1000: Rename tx task from "K_TXQ_TASK" to NETDEV-tx
This follows normal Linux conventions and is also more useful because
the netdevice name is part of the task name (e.g., "wlan0-tx" for
network device "wlan0").

Signed-off-by: David Mosberger-Tang <davidm@egauge.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209044411.3482259-4-davidm@egauge.net
2021-12-14 20:46:34 +02:00
David Mosberger-Tang
30e08bc0a9 wilc1000: Rename irq handler from "WILC_IRQ" to netdev name
This change follows normal Linux convention and the new name is more
useful since it'll be clear which irq handler statistics correspond to
which net device.

Signed-off-by: David Mosberger-Tang <davidm@egauge.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209044411.3482259-3-davidm@egauge.net
2021-12-14 20:46:34 +02:00
David Mosberger-Tang
4347d34e6a wilc1000: Rename SPI driver from "WILC_SPI" to "wilc1000_spi"
The name "wilc1000_spi" follows normal Linux conventions and also is
analogous to the SDIO driver, which uses "wilc1000_sdio".

Signed-off-by: David Mosberger-Tang <davidm@egauge.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209044411.3482259-2-davidm@egauge.net
2021-12-14 20:46:33 +02:00
David Mosberger-Tang
73bbef64bc wilc1000: Fix spurious "FW not responding" error
When deinitializing the driver, one or more "FW not responding" error
appears on the console.  This appears to be due to wilc_wlan_stop()
disabling host/WILC1000 communication, but then right afterwards, it
tries to release the bus with chip-sleep enabled.  The problem is
enabling the chip-sleep cannot success once host/WILC1000
communication is disabled.  Fix by only releasing the bus.

Signed-off-by: David Mosberger-Tang <davidm@egauge.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211208062747.3405221-1-davidm@egauge.net
2021-12-14 20:45:27 +02:00
David Mosberger-Tang
dde02213fa wilc1000: Remove misleading USE_SPI_DMA macro
The USE_SPI_DMA macro name suggests that it could be set to 1 to
control whether or not SPI DMA should be used.  However, that's not
what it does.  If set to 1, it'll set the SPI messages'
"is_dma_mapped" flag to true, even though the tx/rx buffers aren't
actually DMA mapped by the driver.  In other words, setting this flag
to 1 will break the driver.

Best to clean up this confusion by removing the macro altogether.
There is no need to explicitly initialize "is_dma_mapped" because the
message is cleared to zero anyhow, so "is_dma_mapped" is set to false
by default.

Signed-off-by: David Mosberger-Tang <davidm@egauge.net>
Acked-by: Ajay Singh <ajay.kathat@microchip.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211207002453.3193737-1-davidm@egauge.net
2021-12-14 20:43:47 +02:00
David Mosberger-Tang
5ae660641d wilc1000: Fix missing newline in error message
Add missing newline in pr_err() message.

Signed-off-by: David Mosberger-Tang <davidm@egauge.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211206232709.3192856-3-davidm@egauge.net
2021-12-14 20:42:34 +02:00
David Mosberger-Tang
f92b9f9674 wilc1000: Fix copy-and-paste typo in wilc_set_mac_address
The messages appears to have been copied from wilc_get_mac_address and
says "get" when it should say "set".

Signed-off-by: David Mosberger-Tang <davidm@egauge.net>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211206232709.3192856-2-davidm@egauge.net
2021-12-14 20:42:33 +02:00
Ching-Te Ku
4c02043c5a rtw89: coex: Update COEX to 5.5.8
Update COEX version.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209083229.10815-8-pkshih@realtek.com
2021-12-14 20:36:02 +02:00
Ching-Te Ku
bd309c8b49 rtw89: coex: Cancel PS leaving while C2H comes
It's unnecessary to leave WL PS while C2H comes.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209083229.10815-7-pkshih@realtek.com
2021-12-14 20:36:01 +02:00
Ching-Te Ku
eb87d79911 rtw89: coex: Update BT counters while receiving report
Move _chk_btc_err and update_bt_cnt to _chk_btc_report(),
so we can update counter/info to COEX at a proper moment,
instead of relying on a user does cat the debug info periodically.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209083229.10815-6-pkshih@realtek.com
2021-12-14 20:36:01 +02:00
Ching-Te Ku
2200ff3f0d rtw89: coex: Define LPS state for BTC using
To distinguish three types of LPS state.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209083229.10815-5-pkshih@realtek.com
2021-12-14 20:36:01 +02:00
Ching-Te Ku
8c7e9ceb5b rtw89: coex: Add MAC API to get BT polluted counter
Add function to get and parse BT polluted counter.
When WLAN Tx was dropped by BT, the packet will be marked as BT polluted.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209083229.10815-4-pkshih@realtek.com
2021-12-14 20:36:00 +02:00
Ching-Te Ku
f8028a9a92 rtw89: coex: Not to send H2C when WL not ready and count H2C
Prevent to send H2C request to FW when BTC is not initialized or
WL is under power saving. Add counter to count the H2C success or fail.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209083229.10815-3-pkshih@realtek.com
2021-12-14 20:36:00 +02:00
Ching-Te Ku
b3131a41ac rtw89: coex: correct C2H header length
To resolve C2H handle length mismatch, or it will parse the c2h content
out of array.

Signed-off-by: Ching-Te Ku <ku920601@realtek.com>
Signed-off-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20211209083229.10815-2-pkshih@realtek.com
2021-12-14 20:35:59 +02:00
Tony Nguyen
f8a3bcceb4 ice: Remove unused ICE_FLOW_SEG_HDRS_L2_MASK
Remove the unused define ICE_FLOW_SEG_HDRS_L2_MASK.

Reported-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Paul Menzel <pmenzel@molgen.mpg.de>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
2021-12-14 10:19:14 -08:00
Dan Carpenter
e53a80835f ice: Remove unnecessary casts
The "bitmap" variable is already an unsigned long so there is no need
for this cast.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
2021-12-14 10:19:14 -08:00
Tony Nguyen
c14846914e ice: Propagate error codes
As all functions now return standard error codes, propagate the values
being returned instead of converting them to generic values.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
2021-12-14 10:19:14 -08:00
Tony Nguyen
2ccc1c1ccc ice: Remove excess error variables
ice_status previously had a variable to contain these values where other
error codes had a variable as well. With ice_status now being an int,
there is no need for two variables to hold error values. In cases where
this occurs, remove one of the excess variables and use a single one.
Some initialization of variables are no longer needed and have been
removed.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
2021-12-14 10:19:13 -08:00
Tony Nguyen
5518ac2a64 ice: Cleanup after ice_status removal
Clean up code after changing ice_status to int. Rearrange to fix reverse
Christmas tree and pull lines up where applicable.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
2021-12-14 10:19:13 -08:00
Tony Nguyen
d54699e27d ice: Remove enum ice_status
Replace uses of ice_status to, as equivalent as possible, error codes.
Remove enum ice_status and its helper conversion function as they are no
longer needed.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
2021-12-14 10:19:13 -08:00
Tony Nguyen
5e24d5984c ice: Use int for ice_status
To prepare for removal of ice_status, change the variables from
ice_status to int. This eases the transition when values are changed to
return standard int error codes over enum ice_status.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
2021-12-14 10:19:13 -08:00
Tony Nguyen
5f87ec4861 ice: Remove string printing for ice_status
Remove the ice_stat_str() function which prints the string
representation of the ice_status error code. With upcoming changes
moving away from ice_status, there will be no need for this function.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
2021-12-14 10:19:13 -08:00
Wojciech Drewek
247dd97d71 ice: Refactor status flow for DDP load
Before this change, final state of the DDP pkg load process was
dependent on many variables such as: ice_status, pkg version,
ice_aq_err. The last one had be stored in hw->pkg_dwnld_status.
It was impossible to conclude this state just from ice_status, that's
why logging process of DDP pkg load in the caller was a little bit
complicated.

With this patch new status enum is introduced - ice_ddp_state.
It covers all the possible final states of the loading process.
What's tricky for ice_ddp_state is that not only
ICE_DDP_PKG_SUCCESS(=0) means that load was successful. Actually
three states mean that:
 - ICE_DDP_PKG_SUCCESS
 - ICE_DDP_PKG_SAME_VERSION_ALREADY_LOADED
 - ICE_DDP_PKG_COMPATIBLE_ALREADY_LOADED
ice_is_init_pkg_successful can tell that information.

One ddp_state should not be used outside of ice_init_pkg which is
ICE_DDP_PKG_ALREADY_LOADED. It is more generic, it is used in
ice_dwnld_cfg_bufs to see if pkg is already loaded. At this point
we can't use one of the specific one (SAME_VERSION, COMPATIBLE,
NOT_SUPPORTED) because we don't have information on the package
currently loaded in HW (we are before calling ice_get_pkg_info).

We can get rid of hw->pkg_dwnld_status because we are immediately
mapping aq errors to ice_ddp_state in ice_dwnld_cfg_bufs.

Other errors like ICE_ERR_NO_MEMORY, ICE_ERR_PARAM are mapped the
generic ICE_DDP_PKG_ERR.

Suggested-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Wojciech Drewek <wojciech.drewek@intel.com>
Tested-by: Tony Brelinski <tony.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
2021-12-14 10:19:13 -08:00
Brett Creeley
fabf480bf9 ice: Refactor promiscuous functions
Some of the promiscuous mode functions take a boolean to indicate
set/clear, which affects readability. Refactor and provide an
interface for the promiscuous mode code with explicit set and clear
promiscuous mode operations.

Signed-off-by: Brett Creeley <brett.creeley@intel.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
2021-12-14 10:18:52 -08:00
Jeff Guo
60f44fe4cd ice: refactor PTYPE validating
Since the capability of a PTYPE within a specific package could be
negotiated by checking the HW bit map, it means that there's no need
to maintain a different PTYPE list for each type of the package when
parsing PTYPE. So refactor the PTYPE validating mechanism.

Signed-off-by: Jeff Guo <jia.guo@intel.com>
Tested-by: Tony Brelinski <tony.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
2021-12-14 08:06:47 -08:00
Haiyue Wang
8818b95409 ice: Add package PTYPE enable information
Scan the 'Marker Ptype TCAM' section to retrieve the Rx parser PTYPE
enable information from the current package.

Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Tested-by: Tony Brelinski <tony.brelinski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
2021-12-14 08:06:47 -08:00
Seevalamuthu Mariappan
beefee7133 ath11k: Change qcn9074 fw to operate in mode-2
In mode-2 QCN9074 firmware uses 15MB of host memory and firmware
request 1MB size segements in QMI, whereas in mode-0 firmware
uses 45MB of host memory and each segment is of 2MB size.
In mode-2 firmware operates with reduced number of vdevs and peers.

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.4.0.1-01838-QCAHKSWPL_SILICONZ-1

Signed-off-by: Seevalamuthu Mariappan <seevalam@codeaurora.org>
Signed-off-by: Anilkumar Kolli <akolli@codeaurora.org>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20211209104351.9811-3-kvalo@kernel.org
2021-12-14 17:32:34 +02:00
Kalle Valo
523aafd0f0 ath11k: add ab to TARGET_NUM_VDEVS & co
The next patch changes TARGET_NUM_VDEVS to be dynamic and need access to ab.
Add ab separately to keep the next patch simple.

Compile tested only.

Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20211209104351.9811-2-kvalo@kernel.org
2021-12-14 17:32:33 +02:00
Zhou Qingyang
eccd251363 ath11k: Fix a NULL pointer dereference in ath11k_mac_op_hw_scan()
In ath11k_mac_op_hw_scan(), the return value of kzalloc() is directly
used in memcpy(), which may lead to a NULL pointer dereference on
failure of kzalloc().

Fix this bug by adding a check of arg.extraie.ptr.

This bug was found by a static analyzer. The analysis employs
differential checking to identify inconsistent security operations
(e.g., checks or kfrees) between two code paths and confirms that the
inconsistent operations are not recovered in the current function or
the callers, so they constitute bugs.

Note that, as a bug found by static analysis, it can be a false
positive or hard to trigger. Multiple researchers have cross-reviewed
the bug.

Builds with CONFIG_ATH11K=m show no new warnings, and our static
analyzer no longer warns about this code.

Fixes: d5c65159f2 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Zhou Qingyang <zhou1615@umn.edu>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20211202155348.71315-1-zhou1615@umn.edu
2021-12-14 17:30:26 +02:00
Dany Madden
fe4c82a7e0 ibmvnic: remove unused defines
IBMVNIC_STATS_TIMEOUT and IBMVNIC_INIT_FAILED are not used in the driver.
Remove them.

Suggested-by: Sukadev Bhattiprolu <sukadev@linux.ibm.com>
Signed-off-by: Dany Madden <drt@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:58:02 +00:00
Dany Madden
b6ee566cf3 ibmvnic: Update driver return codes
Update return codes to be more informative.

Signed-off-by: Jacob Root <otis@otisroot.com>
Signed-off-by: Dany Madden <drt@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:58:02 +00:00
Danielle Ratson
b442f2ea84 mlxsw: spectrum_router: Consolidate MAC profiles when possible
Currently, when setting a router interface (RIF) MAC address while the
MAC profile is not shared with other RIFs, the profile is edited so that
the new MAC address is assigned to it.

This does not take into account a situation in which the new MAC address
already matches an existing MAC profile. In that situation, two MAC
profiles will be occupied even though they hold MAC addresses from the
same profile.

In order to prevent that, add a check to ensure that editing a MAC
profile takes place only when the new MAC address does not match an
existing profile.

Fixes: 605d25cd78 ("mlxsw: spectrum_router: Add RIF MAC profiles support")
Reported-by: Maksym Yaremchuk <maksymy@nvidia.com>
Tested-by: Maksym Yaremchuk <maksymy@nvidia.com>
Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:56:10 +00:00
Vladimir Oltean
c8a2a011cd net: dsa: sja1105: fix broken connection with the sja1110 tagger
The driver was incorrectly converted assuming that "sja1105" is the only
tagger supported by this driver. This results in SJA1110 switches
failing to probe:

sja1105 spi1.0: Unable to connect to tag protocol "sja1110": -EPROTONOSUPPORT
sja1105: probe of spi1.2 failed with error -93

Add DSA_TAG_PROTO_SJA1110 to the list of supported taggers by the
sja1105 driver. The sja1105_tagger_data structure format is common for
the two tagging protocols.

Fixes: c79e84866d ("net: dsa: tag_sja1105: convert to tagger-owned data")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:45:16 +00:00
David S. Miller
a41c4d96ae Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:

====================
Intel Wired LAN Driver Updates 2021-12-13

This series contains updates to iavf driver only.

Dan Carpenter fixes some missing mutex unlocking.

Stefan Assmann restores stopping watchdog from overriding to reset state.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:41:05 +00:00
Guillaume Nault
b4bffa4cea bareudp: Add extack support to bareudp_configure()
Add missing extacks for common configuration errors.

Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:39:17 +00:00
Ong Boon Leong
aeb7c75cb7 net: stmmac: fix tc flower deletion for VLAN priority Rx steering
To replicate the issue:-

1) Add 1 flower filter for VLAN Priority based frame steering:-
$ IFDEVNAME=eth0
$ tc qdisc add dev $IFDEVNAME ingress
$ tc qdisc add dev $IFDEVNAME root mqprio num_tc 8 \
   map 0 1 2 3 4 5 6 7 0 0 0 0 0 0 0 0 \
   queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 hw 0
$ tc filter add dev $IFDEVNAME parent ffff: protocol 802.1Q \
   flower vlan_prio 0 hw_tc 0

2) Get the 'pref' id
$ tc filter show dev $IFDEVNAME ingress

3) Delete a specific tc flower record (say pref 49151)
$ tc filter del dev $IFDEVNAME parent ffff: pref 49151

From dmesg, we will observe kernel NULL pointer ooops

[  197.170464] BUG: kernel NULL pointer dereference, address: 0000000000000000
[  197.171367] #PF: supervisor read access in kernel mode
[  197.171367] #PF: error_code(0x0000) - not-present page
[  197.171367] PGD 0 P4D 0
[  197.171367] Oops: 0000 [#1] PREEMPT SMP NOPTI

<snip>

[  197.171367] RIP: 0010:tc_setup_cls+0x20b/0x4a0 [stmmac]

<snip>

[  197.171367] Call Trace:
[  197.171367]  <TASK>
[  197.171367]  ? __stmmac_disable_all_queues+0xa8/0xe0 [stmmac]
[  197.171367]  stmmac_setup_tc_block_cb+0x70/0x110 [stmmac]
[  197.171367]  tc_setup_cb_destroy+0xb3/0x180
[  197.171367]  fl_hw_destroy_filter+0x94/0xc0 [cls_flower]

The above issue is due to previous incorrect implementation of
tc_del_vlan_flow(), shown below, that uses flow_cls_offload_flow_rule()
to get struct flow_rule *rule which is no longer valid for tc filter
delete operation.

  struct flow_rule *rule = flow_cls_offload_flow_rule(cls);
  struct flow_dissector *dissector = rule->match.dissector;

So, to ensure tc_del_vlan_flow() deletes the right VLAN cls record for
earlier configured RX queue (configured by hw_tc) in tc_add_vlan_flow(),
this patch introduces stmmac_rfs_entry as driver-side flow_cls_offload
record for 'RX frame steering' tc flower, currently used for VLAN
priority. The implementation has taken consideration for future extension
to include other type RX frame steering such as EtherType based.

v2:
 - Clean up overly extensive backtrace and rewrite git message to better
   explain the kernel NULL pointer issue.

Fixes: 0e039f5cf8 ("net: stmmac: add RX frame steering based on VLAN priority in tc flower")
Tested-by: Kurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:32:19 +00:00
Thomas Gleixner
09eb3ad55f Merge branch 'irq/urgent' into irq/msi
to pick up the PCI/MSI-x fixes.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2021-12-14 13:30:34 +01:00
Hangbin Liu
085d610008 Bonding: force user to add HWTSTAMP_FLAG_BONDED_PHC_INDEX when get/set HWTSTAMP
When there is a failover, the PHC index of bond active interface will be
changed. This may break the user space program if the author didn't aware.

By setting this flag, the user should aware that the PHC index get/set
by syscall is not stable. And the user space is able to deal with it.
Without this flag, the kernel will reject the request forwarding to
bonding.

Reported-by: Jakub Kicinski <kuba@kernel.org>
Fixes: 94dd016ae5 ("bond: pass get_ts_info and SIOC[SG]HWTSTAMP ioctl to active device")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:28:24 +00:00
Hangbin Liu
9c9211a3fc net_tstamp: add new flag HWTSTAMP_FLAG_BONDED_PHC_INDEX
Since commit 94dd016ae5 ("bond: pass get_ts_info and SIOC[SG]HWTSTAMP
ioctl to active device") the user could get bond active interface's
PHC index directly. But when there is a failover, the bond active
interface will change, thus the PHC index is also changed. This may
break the user's program if they did not update the PHC timely.

This patch adds a new hwtstamp_config flag HWTSTAMP_FLAG_BONDED_PHC_INDEX.
When the user wants to get the bond active interface's PHC, they need to
add this flag and be aware the PHC index may be changed.

With the new flag. All flag checks in current drivers are removed. Only
the checking in net_hwtstamp_validate() is kept.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2021-12-14 12:28:24 +00:00
Maor Gottlieb
c7d5fa105b net/mlx5: Create more priorities for FDB bypass namespace
Create 16 flow steering priorities for FDB bypass users.

Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-12-13 16:03:00 -08:00
Maor Gottlieb
4588fed7be net/mlx5: Refactor mlx5_get_flow_namespace
Have all the namespace type check in the same switch case.

Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-12-13 16:02:59 -08:00