stmmac_reset() takes the stmmac_priv and an ioaddr. It has one call
site, which passes the priv pointer, and dereferences priv for the
ioaddr.
stmmac_reset() then checks whether priv is NULL. If it was, the caller
would have oopsed. Remove the checks for NULL, and move the dereference
for ioaddr into stmmac_reset().
* tag 'linux-can-next-for-6.19-20251112-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next:
can: bxcan: Fix a typo error for assign
dt-bindings: can: mcp251xfd: add gpio-controller property
can: mcp251xfd: add gpio functionality
can: mcp251xfd: only configure PIN1 when rx_int is set
can: mcp251xfd: add workaround for errata 5
can: mcp251xfd: utilize gather_write function for all non-CRC writes
can: mcp251xfd: move chip sleep mode into runtime pm
can: mcp251x: mcp251x_can_probe(): use dev_err_probe()
can: peak_usb: convert to use ndo_hwtstamp callbacks
can: peak_canfd: convert to use ndo_hwtstamp callbacks
can: convert generic HW timestamp ioctl to ndo_hwtstamp callbacks
====================
====================
net: stmmac: convert glue drivers to use stmmac_get_phy_intf_sel()
This series converts the remaining glue drivers that support
multi-interface to use stmmac_get_phy_intf_sel(). The reason these
drivers are not converted to the set_phy_intf_sel() method is
because it is unclear whether there are ordering dependencies that
would prevent it.
For example, reading the stm32mp2 documentation, it is required to
set the ETH1_SEL field while the dwmac core is in reset and before
clocks are enabled. This requirement can not be satsified at the
moment (but could with further changes.)
====================
net: stmmac: visconti: use stmmac_get_phy_intf_sel()
Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the
phy_intf_sel value, validate the result and use that to set the
control register to select the operating mode for the DWMAC core.
Note that this will allow GMII as well as MII as the phy_intf_sel
value is the same for both.
net: stmmac: visconti: use PHY_INTF_SEL_x to select PHY interface
Convert dwmac-visconti to use the PHY_INTF_SEL_x definitions. The
original definitions used constant 0, BIT(0) (==1) and BIT(2) (==4)
to define these, but the values of the bits corresponds with the
PHY_INTF_SEL_x values, so it is highly likely that these are not
individual bits, but the PHY_INTF_SEL_x bitfield.
Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the
phy_intf_sel value. As both configure functions would end up with the
same code, call this from stm32mp1_set_mode(), validate the result and
pass the resulting value into the stm32 configure function. Use this
value to set the operating mode for the DWMAC core.
Rather than defining separate constants for each, use the
PHY_INTF_SEL_x definitions in the switch()es configuring the
control register, and use one FIELD_PREP() to convert phy_intf_sel
to the register value.
net: stmmac: stm32: use PHY_INTF_SEL_x to select PHY interface
Convert dwmac-stm32 to use the PHY_INTF_SEL_x definitions.
For stm32mp1, the original definitions used constant 0 (GMII, 0 << 21),
BIT(21) (RGMII, 1 << 21) and BIT(23) (RMII, 4 << 21) to define these,
but from the values it can be clearly seen that these are the
PHY_INTF_SEL_x inputs to the dwmac.
For stm32mp2, the original definitions cover a bitfield 6:4 in the
SYSCFG Ethernet1 control register (according to documentation) and use
the PHY_INTF_SEL_x values.
Use the common dwmac definitions for the PHY interface selection field
by adding the bitfield mask, and using FIELD_PREP() for the bitfield
values.
net: stmmac: starfive: use stmmac_get_phy_intf_sel()
Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the
phy_intf_sel value, validate the result and use that to set the
control register to select the operating mode for the DWMAC core.
net: stmmac: mediatek: use stmmac_get_phy_intf_sel()
Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the
phy_intf_sel value, validate the result, and pass that into the
implementation specific ->dwmac_set_phy_interface() method. Use this
to configure the PHY interface selection field.
net: stmmac: loongson1: use stmmac_get_phy_intf_sel()
Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the
phy_intf_sel value, validate the result and use that to set the
control register to select the operating mode for the DWMAC core.
Note that this will allow GMII as well as MII as the phy_intf_sel
value is the same for both.
Use PHY_INTF_SEL_x definitions for phy_intf_sel bitfield.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
selftests: drv-net: Limit the max number of queues in procfs_downup_hammer
For NICs with a large (1024+) number of queues, this test can cause
excessive memory fragmentation. This results in OOM errors, and in the
worst case driver/kernel crashes. We don't need to test with the max number
of queues, just enough to create a high likelihood of races between
reconfiguration and stats getting read.
The mcp251xfd allows two pins to be configured as GPIOs. This series adds
support for this feature.
The GPIO functionality is controlled with the IOCON register which has an
erratum.
Patch 1 from https://lore.kernel.org/linux-can/20240429-mcp251xfd-runtime_pm-v1-3-c26a93a66544@pengutronix.de/
Patch 2 refactor of no-crc functions to prepare workaround for non-crc writes
Patch 3 is the fix/workaround for the aforementioned erratum
Patch 4 only configure pin1 for rx-int
Patch 5 adds the gpio support
Patch 6 updates dt-binding
can: mcp251xfd: only configure PIN1 when rx_int is set
When rx_int is used th mcp251xfd_chip_rx_int_enable and
mcp251xfd_chip_rx_int_disable function configure both PIN0 and PIN1. To
prepare the support of the GPIOS only configure PIN1 with
regmap_update_bits.
This way PIN0 can be used as GPIO while PIN1 is used as rx_int
interrupt.
According to Errata DS80000789E 5 writing IOCON register using one SPI
write command clears LAT0/LAT1.
Errata Fix/Work Around suggests to write registers with single byte write
instructions. However, it seems that every write to the second byte
causes the overwrite of LAT0/LAT1.
Never write byte 2 of IOCON register to avoid clearing of LAT0/LAT1.
Jakub Kicinski [Wed, 12 Nov 2025 17:33:23 +0000 (09:33 -0800)]
Merge tag 'wireless-next-2025-11-12' of https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next
Johannes Berg says:
====================
More -next material, notably:
- split ieee80211.h file, it's way too big
- mac80211: initial chanctx work towards NAN
- mac80211: MU-MIMO sniffer improvements
- ath12k: statistics improvements
* tag 'wireless-next-2025-11-12' of https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next: (26 commits)
wifi: cw1200: Fix potential memory leak in cw1200_bh_rx_helper()
wifi: mac80211: make monitor link info check more specific
wifi: mac80211: track MU-MIMO configuration on disabled interfaces
wifi: cfg80211/mac80211: Add fallback mechanism for INDOOR_SP connection
wifi: cfg80211/mac80211: clean up duplicate ap_power handling
wifi: cfg80211: use a C99 initializer in wiphy_register
wifi: cfg80211: fix doc of struct key_params
wifi: mac80211: remove unnecessary vlan NULL check
wifi: mac80211: pass frame type to element parsing
wifi: mac80211: remove "disabling VHT" message
wifi: mac80211: add and use chanctx usage iteration
wifi: mac80211: simplify ieee80211_recalc_chanctx_min_def() API
wifi: mac80211: remove chanctx to link back-references
wifi: mac80211: make link iteration safe for 'break'
wifi: mac80211: fix EHT typo
wifi: cfg80211: fix EHT typo
wifi: ieee80211: split NAN definitions out
wifi: ieee80211: split P2P definitions out
wifi: ieee80211: split S1G definitions out
wifi: ieee80211: split EHT definitions out
...
====================
Improve the readibility of the code setting ndev->max_mtu. This depends
on the hardware specific maximum defined by the MAC core, and also a
platform provided maximum.
The code was originally checking that the platform specific maximum was
between ndev->min_mtu..MAC core maximum before reducing ndev->max_mtu,
otherwise if the platform specific maximum was less than ndev->min_mtu,
issuing a warning.
Re-order the code to handle the case where the platform specific max is
below ndev->min_mtu, which then means that the subsequent test is
simply reducing ndev->max_mtu.
Update the comment, and add a few blank lines to separate the blocks of
code.
Eric Dumazet [Tue, 11 Nov 2025 15:12:35 +0000 (15:12 +0000)]
net: clear skb->sk in skb_release_head_state()
skb_release_head_state() inlines skb_orphan().
We need to clear skb->sk otherwise we can freeze TCP flows
on a mostly idle host, because skb_fclone_busy() would
return true as long as the packet is not yet processed by
skb_defer_free_flush().
Fixes: 1fcf572211da ("net: allow skb_release_head_state() to be called multiple times") Fixes: e20dfbad8aab ("net: fix napi_consume_skb() with alien skbs") Signed-off-by: Eric Dumazet <edumazet@google.com> Tested-by: Aditya Garg <gargaditya@linux.microsoft.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com> Link: https://patch.msgid.link/20251111151235.1903659-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================
selftests/vsock: refactor and improve vmtest infrastructure
This patch series refactors the vsock selftest VM infrastructure to
improve test run times, improve logging, and prepare for future tests
which make heavy usage of these refactored functions and have new
requirements such as simultaneous QEMU processes.
====================
Bobby Eshleman [Sat, 8 Nov 2025 16:01:03 +0000 (08:01 -0800)]
selftests/vsock: disable shellcheck SC2317 and SC2119
Disable shellcheck rules SC2317 and SC2119. These rules are being
triggered due to false positives. For SC2317, many `return
"${KSFT_PASS}"` lines are reported as unreachable, even though they are
executed during normal runs. For SC2119, the fact that
log_guest/log_host accept either stdin or arguments triggers SC2119,
despite being valid.
Add vsock_loopback module loading to the loopback test so that vmtest.sh
can be used for kernels built with loopback as a module.
This is not technically a fix as kselftest expects loopback to be
built-in already (defined in selftests/vsock/config). This is useful
only for using vmtest.sh outside of kselftest.
Bobby Eshleman [Sat, 8 Nov 2025 16:01:00 +0000 (08:01 -0800)]
selftests/vsock: add BUILD=0 definition
Add the definition for BUILD and initialize it to zero. This avoids
'bash -u vmtest.sh` from throwing 'unbound variable' when BUILD is not
set to 1 and is later checked for its value.
Bobby Eshleman [Sat, 8 Nov 2025 16:00:58 +0000 (08:00 -0800)]
selftests/vsock: add check_result() for pass/fail counting
Add check_result() function to reuse logic for incrementing the
pass/fail counters. This function will get used by different callers as
we add different types of tests in future patches (namely, namespace and
non-namespace tests will be called at different places, and re-use this
function).
Bobby Eshleman [Sat, 8 Nov 2025 16:00:57 +0000 (08:00 -0800)]
selftests/vsock: speed up tests by reducing the QEMU pidfile timeout
Reduce the time waiting for the QEMU pidfile from three minutes to five
seconds. The three minute time window was chosen to make sure QEMU had
enough time to fully boot up. This, however, is an unreasonably long
delay for QEMU to write the pidfile, which happens earlier when the QEMU
process starts (not after VM boot). The three minute delay becomes
noticeably wasteful in future tests that expect QEMU to fail and wait a
full three minutes for a pidfile that will never exist.
Bobby Eshleman [Sat, 8 Nov 2025 16:00:56 +0000 (08:00 -0800)]
selftests/vsock: do not unconditionally die if qemu fails
If QEMU fails to boot, then set the returncode (via timeout) instead of
unconditionally dying. This is in preparation for tests that expect QEMU
to fail to boot. In that case, we just want to know if the boot failed
or not so we can test the pass/fail criteria, and continue executing the
next test.
Bobby Eshleman [Sat, 8 Nov 2025 16:00:55 +0000 (08:00 -0800)]
selftests/vsock: avoid multi-VM pidfile collisions with QEMU
Change QEMU to use generated pidfile names instead of just a single
globally-defined pidfile. This allows multiple QEMU instances to
co-exist with different pidfiles. This is required for future tests that
use multiple VMs to check for CID collissions.
Additionally, this also places the burden of killing the QEMU process
and cleaning up the pidfile on the caller of vm_start(). To help with
this, a function terminate_pidfiles() is introduced that callers use to
perform the cleanup. The terminate_pidfiles() function supports multiple
pidfile removals because future patches will need to process two
pidfiles at a time.
Change QEMU_OPTS to be initialized inside the vm_start(). This allows
the generated pidfile to be passed to the string assignment, and
prepares for future vm-specific options as well (e.g., cid).
Bobby Eshleman [Sat, 8 Nov 2025 16:00:54 +0000 (08:00 -0800)]
selftests/vsock: reuse logic for vsock_test through wrapper functions
Add wrapper functions vm_vsock_test() and host_vsock_test() to invoke
the vsock_test binary. This encapsulates several items of repeat logic,
such as waiting for the server to reach listening state and
enabling/disabling the bash option pipefail to avoid pipe-style logging
from hiding failures.
Bobby Eshleman [Sat, 8 Nov 2025 16:00:53 +0000 (08:00 -0800)]
selftests/vsock: make wait_for_listener() work even if pipefail is on
Rewrite wait_for_listener()'s pattern matching to avoid tripping the
if-condition when pipefail is on.
awk doesn't gracefully handle SIGPIPE with a non-zero exit code, so grep
exiting upon finding a match causes false-positives when the pipefail
option is used (grep exits, SIGPIPE emits, and awk complains with a
non-zero exit code). Instead, move all of the pattern matching into awk
so that SIGPIPE cannot happen and the correct exit code is returned.
Bobby Eshleman [Sat, 8 Nov 2025 16:00:52 +0000 (08:00 -0800)]
selftests/vsock: improve logging in vmtest.sh
Improve usability of logging functions. Remove the test name prefix from
logging functions so that logging calls can be made deeper into the call
stack without passing down the test name or setting some global. Teach
log function to accept a LOG_PREFIX variable to avoid unnecessary
argument shifting.
Remove log_setup() and instead use log_host(). The host/guest prefixes
are useful to show whether a failure happened on the guest or host side,
but "setup" doesn't really give additional useful information. Since all
log_setup() calls happen on the host, lets just use log_host() instead.
can: mcp251xfd: utilize gather_write function for all non-CRC writes
This is a preparation patch to add errata workaround for non crc writes.
Currently for non-crc writes to the chip can go through the
.gather_write, .write or the reg_update_bits callback.
To allow the addition of the errata fix at a single location use
mcp251xfd_regmap_nocrc_gather_write for all non-CRC write instructions,
similar to the crc regmap.
can: mcp251xfd: move chip sleep mode into runtime pm
This is a preparation patch to add GPIO support.
Up to now, the Vdd regulator and the clocks have been managed by
Runtime-PM (on systems without CONFIG_PM these remain permanently
switched on).
During the mcp251xfd_open() callback the mcp251xfd is powered,
soft-reset and configured. In mcp251xfd_stop() the chip is shut down
again. To support the on-chip GPIOs, the chip must be supplied with
power while GPIOs are being requested, even if the networking
interface is down.
To support this, move the functions mcp251xfd_chip_softreset() and
mcp251xfd_chip_clock_init() from mcp251xfd_chip_start() to
mcp251xfd_runtime_resume(). Instead of setting the controller to sleep
mode in mcp251xfd_chip_stop(), bring it into configuration mode. This
way it doesn't take part in bus activity and doesn't enter sleep mode.
====================
net: stmmac: convert meson8b to use stmmac_get_phy_intf_sel()
This series splits out meson8b from the previous 16 patch series
as that now has r-b tags.
This series converts meson8b to use stmmac_get_phy_intf_sel(). This
driver is not converted to the set_phy_intf_sel() method as it is
unclear whether there are ordering dependencies that would prevent
it. I would appreciate the driver author looking in to whether this
conversion is possible.
====================
net: stmmac: meson8b: use stmmac_get_phy_intf_sel()
Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the
phy_intf_sel value, validate the result and use that to set the
control register to select the operating mode for the DWMAC core.
net: netcp: ethss: Fix type of first parameter in hwtstamp stubs
When building without CONFIG_TI_CPTS, there are a series of errors from
-Wincompatible-pointer-types:
drivers/net/ethernet/ti/netcp_ethss.c:3831:27: error: initialization of 'int (*)(void *, struct kernel_hwtstamp_config *)' from incompatible pointer type 'int (*)(struct gbe_intf *, struct kernel_hwtstamp_config *)' [-Wincompatible-pointer-types]
3831 | .hwtstamp_get = gbe_hwtstamp_get,
| ^~~~~~~~~~~~~~~~
drivers/net/ethernet/ti/netcp_ethss.c:3831:27: note: (near initialization for 'gbe_module.hwtstamp_get')
drivers/net/ethernet/ti/netcp_ethss.c:2758:19: note: 'gbe_hwtstamp_get' declared here
2758 | static inline int gbe_hwtstamp_get(struct gbe_intf *gbe_intf,
| ^~~~~~~~~~~~~~~~
drivers/net/ethernet/ti/netcp_ethss.c:3832:27: error: initialization of 'int (*)(void *, struct kernel_hwtstamp_config *, struct netlink_ext_ack *)' from incompatible pointer type 'int (*)(struct gbe_intf *, struct kernel_hwtstamp_config *, struct netlink_ext_ack *)' [-Wincompatible-pointer-types]
3832 | .hwtstamp_set = gbe_hwtstamp_set,
| ^~~~~~~~~~~~~~~~
drivers/net/ethernet/ti/netcp_ethss.c:3832:27: note: (near initialization for 'gbe_module.hwtstamp_set')
drivers/net/ethernet/ti/netcp_ethss.c:2764:19: note: 'gbe_hwtstamp_set' declared here
2764 | static inline int gbe_hwtstamp_set(struct gbe_intf *gbe_intf,
| ^~~~~~~~~~~~~~~~
In a recent conversion to ndo_hwtstamp, the type of the first parameter
was updated for the CONFIG_TI_CPTS=y implementations of
gbe_hwtstamp_get() and gbe_hwtstamp_set() but not the CONFIG_TI_CPTS=n
ones.
Update the type of the first parameter in the CONFIG_TI_CPTS=n stubs to
resolve the errors.
Before having traffic flow through an eswitch, a user may want to have the
ability to block traffic towards the FDB until FDB is fully programmed and the
user is ready to send traffic to it. For example: when two eswitches are present
for vports in a multi-PF setup, one eswitch may take over the traffic from the
other when the user chooses. Before this take over, a user may want to first
program the inactive eswitch and then once ready redirect traffic to this new
eswitch.
This series introduces a user-configurable mode for an eswitch that allows
dynamically switching between active and inactive modes. When inactive, traffic
does not flow through the eswitch. While inactive, steering pipeline
configuration can be done (e.g. adding TC rules, discovering representors,
enabling the desired SDN modes such as bridge/OVS/DPDK/etc). Once configuration
is completed, a user can set the eswitch mode to active and have traffic flow
through. This allows admins to upgrade forwarding pipeline rules with very
minimal downtime and packet drops.
A user can start the eswitch in switchdev or switchdev_inactive mode.
Active: Traffic is enabled on this eswitch FDB.
Inactive: Traffic is ignored/dropped on this eswitch FDB.
An example use case:
$ devlink dev eswitch set pci/0000:08:00.1 mode switchdev_inactive
Setup FDB pipeline and netdev representors
...
Once ready to start receiving traffic
$ devlink dev eswitch set pci/0000:08:00.1 mode switchdev
Saeed Mahameed [Sat, 8 Nov 2025 07:04:03 +0000 (23:04 -0800)]
net/mlx5: MPFS, add support for dynamic enable/disable
MPFS (Multi PF Switch) is enabled by default in Multi-Host environments,
the driver keeps a list of desired unicast mac addresses of all vports
(vfs/Sfs) and applied to HW via L2_table FW command.
Add API to dynamically apply the list of MACs to HW when needed for next
patches, to utilize this new API in devlink eswitch active/in-active uAPI.
Adds DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE attribute to UAPI and
documentation.
Before having traffic flow through an eswitch, a user may want to have the
ability to block traffic towards the FDB until FDB is fully programmed and
the user is ready to send traffic to it. For example: when two eswitches
are present for vports in a multi-PF setup, one eswitch may take over the
traffic from the other when the user chooses.
Before this take over, a user may want to first program the inactive
eswitch and then once ready redirect traffic to this new eswitch.
switchdev modes transition semantics:
legacy->switchdev_inactive: Create switchdev mode normally, traffic not
allowed to flow yet.
switchdev_inactive->switchdev: Enable traffic to flow.
switchdev->switchdev_inactive: Block traffic on the FDB, FDB and
representros state and content is preserved.
When eswitch is configured to this mode, traffic is ignored/dropped on
this eswitch FDB, while current configuration is kept, e.g FDB rules and
netdev representros are kept available, FDB programming is allowed.
Example:
# start inactive switchdev
devlink dev eswitch set pci/0000:08:00.1 mode switchdev_inactive
# setup TC rules, representors etc ..
# activate
devlink dev eswitch set pci/0000:08:00.1 mode switchdev
====================
tools: ynl: turn the page-pool sample into a real tool
The page-pool YNL sample is quite useful. It's helps calculate
recycling rate and memory consumption. Since we still haven't
figured out a way to integrate with iproute2 (not for the lack
of thinking how to solve it) - create a ynltool command in ynl.
Add page-pool and qstats support.
Most commands can use the Python YNL CLI directly but low level
stats often need aggregation or some math on top to be useful.
Specifically in this patch set:
- page pool stats are aggregated and recycling rate computed
- per-queue stats are used to compute traffic balance across queues
Jakub Kicinski [Fri, 7 Nov 2025 16:22:24 +0000 (08:22 -0800)]
tools: ynltool: create skeleton for the C command
Based on past discussions it seems like integration of YNL into
iproute2 is unlikely. YNL itself is not great as a C library,
since it has no backward compat (we routinely change types).
Most of the operations can be performed with the generic Python
CLI directly. There is, however, a handful of operations where
summarization of kernel output is very useful (mostly related
to stats: page-pool, qstat).
Create a command (inspired by bpftool, I think it stood the test
of time reasonably well) to be able to plug the subcommands into.
Benjamin Berg [Mon, 10 Nov 2025 12:20:20 +0000 (14:20 +0200)]
wifi: mac80211: make monitor link info check more specific
Verify that only one of the permitted change flags is set when changing
the link of a monitor interface. Before the WARN_ON_ONCE would accept
anything if mu_mimo_owner was set.
Also, split out the mu_mimo_owner flag and enable it for all interface
types. The option is set during association when VHT is available and it
is not expected that any configuration of the MU groups is done without
it being set.
Benjamin Berg [Mon, 10 Nov 2025 12:18:20 +0000 (14:18 +0200)]
wifi: mac80211: track MU-MIMO configuration on disabled interfaces
For monitoring, userspace will try to configure the VIF sdata, while the
driver may see the monitor_sdata that is created when only monitor
interfaces are up. This causes the odd situation that it may not be
possible to store the MU-MIMO configuration on monitor_sdata.
Fix this by storing that information on the VIF sdata and updating the
monitor_sdata when available and the interface is up. Also, adjust the
code that adds monitor_sdata so that it will configure MU-MIMO based on
the newly added interface or one of the existing ones.
This should give a mostly consistent behaviour when configuring MU-MIMO
on sniffer interfaces. Should the user configure MU-MIMO on multiple
sniffer interfaces, then mac80211 will simply select one of the
configurations. This behaviour should be good enough and avoids breaking
user expectations in the common scenarios.
wifi: cfg80211/mac80211: Add fallback mechanism for INDOOR_SP connection
Implement fallback to LPI mode when SP mode is not permitted
by regulatory constraints for INDOOR_SP connections.
Limit fallback mechanism to client mode.
wifi: cfg80211: use a C99 initializer in wiphy_register
struct regulatory request was not fully initialized. While this is not
really a big deal because nl80211_send_reg_change_event doesn't look at
the other fields, it still makes sense to zero all the other fields as
Coverity suggests.
Jason Xing [Fri, 31 Oct 2025 10:33:28 +0000 (18:33 +0800)]
xsk: add indirect call for xsk_destruct_skb
Since Eric proposed an idea about adding indirect call wrappers for
UDP and managed to see a huge improvement[1], the same situation can
also be applied in xsk scenario.
This patch adds an indirect call for xsk and helps current copy mode
improve the performance by around 1% stably which was observed with
IXGBE at 10Gb/sec loaded. If the throughput grows, the positive effect
will be magnified. I applied this patch on top of batch xmit series[2],
and was able to see <5% improvement from our internal application
which is a little bit unstable though.
Use INDIRECT wrappers to keep xsk_destruct_skb static as it used to
be when the mitigation config is off.
Be aware of the freeing path that can be very hot since the frequency
can reach around 2,000,000 times per second with the xdpsock test.
Suggested-by: Alexander Lobakin <aleksander.lobakin@intel.com> Signed-off-by: Jason Xing <kernelxing@tencent.com> Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com> Link: https://patch.msgid.link/20251031103328.95468-1-kerneljasonxing@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
isdn: kcapi: add WQ_PERCPU to alloc_workqueue users
Currently if a user enqueues a work item using schedule_delayed_work() the
used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use
WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to
schedule_work() that is using system_wq and queue_work(), that makes use
again of WORK_CPU_UNBOUND.
This lack of consistency cannot be addressed without refactoring the API.
alloc_workqueue() treats all queues as per-CPU by default, while unbound
workqueues must opt-in via WQ_UNBOUND.
This default is suboptimal: most workloads benefit from unbound queues,
allowing the scheduler to place worker threads where they’re needed and
reducing noise when CPUs are isolated.
This continues the effort to refactor workqueue APIs, which began with
the introduction of new workqueues and a new alloc_workqueue flag in:
commit 128ea9f6ccfb ("workqueue: Add system_percpu_wq and system_dfl_wq")
commit 930c2ea566af ("workqueue: Add new WQ_PERCPU flag")
This change adds a new WQ_PERCPU flag to explicitly request
alloc_workqueue() to be per-cpu when WQ_UNBOUND has not been specified.
With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND),
any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND
must now use WQ_PERCPU.
Once migration is complete, WQ_UNBOUND can be removed and unbound will
become the implicit default.
This patch series improves the management of the RX buffer length for
the DQO queue format in the gve driver. The goal is to make RX buffer
length config more explicit, easy to change, and performant by default.
We accomplish that in four patches:
1. Currently, the buffer length is implicitly coupled with the header
split setting, which is an unintuitive and restrictive design. The
first patch decouples the RX buffer length from the header split
configuration.
2. The second patch is a preparatory step for third. It converts the XDP
config verification method to use extack for better error reporting.
3. The third patch exposes the `rx_buf_len` parameter to userspace via
ethtool, allowing user to directly view or modify the RX buffer length
if supported by the device.
4. The final patch improves the out-of-the-box RX single stream throughput
by >10% by changing the driver's default behavior to select the
maximum supported RX buffer length advertised by the device during
initialization.
====================
Ankit Garg [Thu, 6 Nov 2025 19:27:46 +0000 (11:27 -0800)]
gve: Default to max_rx_buffer_size for DQO if device supported
Change the driver's default behavior to prefer the largest available RX
buffer length supported by the device for DQO format, rather than always
using the hardcoded 2K default.
Previously, the driver would initialize with
`GVE_DEFAULT_RX_BUFFER_SIZE` (2K), even if the device advertised support
for a larger length (e.g., 4K).
Performance observations:
- With LRO disabled, we observed >10% improvement in RX single stream
throughput when MTU >=2048.
- With LRO enabled, we observed >10% improvement in RX single stream
throughput when MTU >=1460.
- No regressions were observed.
Signed-off-by: Ankit Garg <nktgrg@google.com> Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com> Reviewed-by: Jordan Rhee <jordanrhee@google.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Joshua Washington <joshwash@google.com> Link: https://patch.msgid.link/20251106192746.243525-5-joshwash@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Ankit Garg [Thu, 6 Nov 2025 19:27:45 +0000 (11:27 -0800)]
gve: Allow ethtool to configure rx_buf_len
Add support for getting and setting the RX buffer length via the
ethtool ring parameters (`ethtool -g`/`-G`). The driver restricts the
allowed buffer length to 2048 (SZ_2K) by default and allows 4096 (SZ_4K)
based on device options.
As XDP is only supported when the `rx_buf_len` is 2048, the driver now
enforces this in two places:
1. In `gve_xdp_set`, rejecting XDP programs if the current buffer
length is not 2048.
2. In `gve_set_rx_buf_len_config`, rejecting buffer length changes if XDP
is loaded and the new length is not 2048.
Signed-off-by: Ankit Garg <nktgrg@google.com> Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com> Reviewed-by: Jordan Rhee <jordanrhee@google.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Joshua Washington <joshwash@google.com> Link: https://patch.msgid.link/20251106192746.243525-4-joshwash@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Ankit Garg [Thu, 6 Nov 2025 19:27:43 +0000 (11:27 -0800)]
gve: Decouple header split from RX buffer length
Previously, enabling header split via `gve_set_hsplit_config` also
implicitly changed the RX buffer length to 4K (if supported by the
device). This coupled two settings that should be orthogonal; this patch
removes that side effect.
After this change, `gve_set_hsplit_config` only toggles the header
split configuration. The RX buffer length is no longer affected and
must be configured independently.
Signed-off-by: Ankit Garg <nktgrg@google.com> Reviewed-by: Harshitha Ramamurthy <hramamurthy@google.com> Reviewed-by: Jordan Rhee <jordanrhee@google.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Joshua Washington <joshwash@google.com> Link: https://patch.msgid.link/20251106192746.243525-2-joshwash@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net: stmmac: ingenic: pass ingenic_mac struct rather than plat_dat
It no longer makes sense to pass a pointer to struct
plat_stmmacenet_data when calling the set_mode() methods to only use it
to get a pointer to the ingenic_mac structure that we already had in
the caller. Simplify this by passing the struct ingenic_mac pointer.
As per the previous commit, we have validated that the phy_intf_sel
value is one that is permissible for this SoC, so there is no need to
handle invalid PHY interface modes. We can also apply the other
configuration based upon the phy_intf_sel value rather than the
PHY interface mode.
x1000, x1600 and x1830 only accept RMII mode. PHY_INTF_SEL_RMII is only
selected with PHY_INTERFACE_MODE_RMII, and PHY_INTF_SEL_RMII has been
validated by the SoC's .valid_phy_intf_sel bitmask. Thus, checking the
interface mode in these functions becomes unnecessary. Remove these.
jz4775 is similar, except for a greater set of PHY_INTF_SEL_x valies.
Also remove the switch statement here.
net: stmmac: ingenic: move "MAC PHY control register" debug
Move the printing of the MAC PHY control register interface mode
setting into ingenic_set_phy_intf_sel(), and use phy_modes() to
print the string rather than using the enum name.
net: stmmac: ingenic: use stmmac_get_phy_intf_sel()
Use stmmac_get_phy_intf_sel() to decode the PHY interface mode to the
phy_intf_sel value, validate the result against the SoC specific
supported phy_intf_sel values, and pass into the SoC specific
set_mode() methods, replacing the local phy_intf_sel variable. This
provides the value for the MACPHYC_PHY_INFT_MASK field.
Use the PHY_INTF_SEL_x values directly in each of the mac_set_mode
methods rather than the driver private MACPHYC_PHY_INFT_x definitions.
Remove the MACPHYC_PHY_INFT_x definitions.
Such large transmit queues can be problematic, especially for cellular
modems. For example, with a typical celluar link speed of 10 Mbit/s, a
fully occupied USB3 transmit queue results in:
454.80 KB / (10 Mbit/s / 8 bit/byte) = 363.84 ms
of additional latency.
This patch adds support for Byte Queue Limits (BQL) [1] to dynamically
manage the transmit queue size and reduce latency without sacrificing
throughput.
Testing was performed on various devices using the usbnet driver for
packet transmission:
No performance degradation was observed for iperf3 TCP or UDP traffic,
while latency for a prioritized ping application was significantly
reduced. For example, using the USB3 to 2.5 GbE adapter, which was fully
utilized by iperf3 UDP traffic, the prioritized ping was improved from
1.6 ms to 0.6 ms. With the same setup but with a 100 Mbit/s Ethernet
connection, the prioritized ping was improved from 35 ms to 5 ms.
====================
net: dsa: b53: add support for BCM5389/97/98 and BCM63XX ARL formats
Currently b53 assumes that all switches apart from BCM5325/5365 use the
same ARL formats, but there are actually multiple formats in use.
Older switches use a format apparently introduced with BCM5387/BCM5389,
while newer chips use a format apparently introduced with BCM5395.
Note that these numbers are not linear, BCM5397/BCM5398 use the older
format.
In addition to that the switches integrated into BCM63XX SoCs use their
own format. While accessing these normal read/write ARL entries are the
same format as BCM5389 one, the search format is different.
So in order to support all these different format, split all code
accessing these entries into chip-family specific functions, and collect
them in appropriate arl ops structs to keep the code cleaner.
Sent as net-next since the ARL accesses have never worked before, and
the extensive refactoring might be too much to warrant a fix.
====================
Jonas Gorski [Fri, 7 Nov 2025 08:07:49 +0000 (09:07 +0100)]
net: dsa: b53: add support for bcm63xx ARL entry format
The ARL registers of BCM63XX embedded switches are somewhat unique. The
normal ARL table access registers have the same format as BCM5389, but
the ARL search registers differ:
* SRCH_CTL is at the same offset of BCM5389, but 16 bits wide. It does
not have more fields, just needs to be accessed by a 16 bit read.
* SRCH_RSLT_MACVID and SRCH_RSLT are aligned to 32 bit, and have shifted
offsets.
* SRCH_RSLT has a different format than the normal ARL data entry
register.
* There is only one set of ENTRY_N registers, implying a 1 bin layout.
So add appropriate ops for bcm63xx and let it use it.
Jonas Gorski [Fri, 7 Nov 2025 08:07:48 +0000 (09:07 +0100)]
net: dsa: b53: add support for 5389/5397/5398 ARL entry format
BCM5389, BCM5397 and BCM5398 use a different ARL entry format with just
a 16 bit fwdentry register, as well as different search control and data
offsets.
So add appropriate ops for them and switch those chips to use them.
Jonas Gorski [Fri, 7 Nov 2025 08:07:47 +0000 (09:07 +0100)]
net: dsa: b53: move ARL entry functions into ops struct
Now that the differences in ARL entry formats are neatly contained into
functions per chip family, wrap them into an ops struct and add wrapper
functions to access them.
Jonas Gorski [Fri, 7 Nov 2025 08:07:43 +0000 (09:07 +0100)]
net: dsa: b53: move reading ARL entries into their own function
Instead of duplicating the whole code iterating over all bins for
BCM5325, factor out reading and parsing the entry into its own
functions, and name it the modern one after the first chip with that ARL
format, (BCM53)95.
We've added 19 non-merge commits during the last 3 day(s) which contain
a total of 22 files changed, 1345 insertions(+), 197 deletions(-).
The main changes are:
1) Preserve skb metadata after a TC BPF program has changed the skb,
from Jakub Sitnicki.
This allows a TC program at the end of a TC filter chain to still see
the skb metadata, even if another TC program at the front of the chain
has changed the skb using BPF helpers.
2) Initial af_smc bpf_struct_ops support to control the smc specific
syn/synack options, from D. Wythe.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next:
bpf/selftests: Add selftest for bpf_smc_hs_ctrl
net/smc: bpf: Introduce generic hook for handshake flow
bpf: Export necessary symbols for modules with struct_ops
selftests/bpf: Cover skb metadata access after bpf_skb_change_proto
selftests/bpf: Cover skb metadata access after change_head/tail helper
selftests/bpf: Cover skb metadata access after bpf_skb_adjust_room
selftests/bpf: Cover skb metadata access after vlan push/pop helper
selftests/bpf: Expect unclone to preserve skb metadata
selftests/bpf: Dump skb metadata on verification failure
selftests/bpf: Verify skb metadata in BPF instead of userspace
bpf: Make bpf_skb_change_head helper metadata-safe
bpf: Make bpf_skb_change_proto helper metadata-safe
bpf: Make bpf_skb_adjust_room metadata-safe
bpf: Make bpf_skb_vlan_push helper metadata-safe
bpf: Make bpf_skb_vlan_pop helper metadata-safe
vlan: Make vlan_remove_tag return nothing
bpf: Unclone skb head on bpf_dynptr_write to skb metadata
net: Preserve metadata on pskb_expand_head
net: Helper to move packet data and metadata after skb_push/pull
====================
net: ravb: Correct bad check of timestamp control flags
When converting the Renesas network drivers to use flags from enum
hwtstamp_rx_filters to control when to timestamp packages instead of a
driver specific schema with bit-wise flags an error was made.
The bit-wise driver specific flags correct logic to set get_ts was:
This change restores the converted flag check to the correct logic of
the bit-wise driver specific flags.
Reported-by: Simon Horman <horms@kernel.org> Closes: https://lore.kernel.org/linux-renesas-soc/aQ4xSv9629XF-Bt3@horms.kernel.org/ Fixes: 16e2e6cf75e6 ("net: ravb: Use common defines for time stamping control") Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Link: https://patch.msgid.link/20251107200100.3637869-1-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Zhongqiu Han [Fri, 7 Nov 2025 07:45:33 +0000 (15:45 +0800)]
ptp: ocp: Document sysfs output format for backward compatibility
Add a comment to ptp_ocp_tty_show() explaining that the sysfs output
intentionally does not include a trailing newline. This is required for
backward compatibility with existing userspace software that reads the
sysfs attribute and uses the value directly as a device path.
A previous attempt to add a newline to align with common kernel
conventions broke userspace applications that were opening device paths
like "/dev/ttyS4\n" instead of "/dev/ttyS4", resulting in ENOENT errors.
This comment prevents future attempts to "fix" this behavior, which would
break existing userspace applications.
projects / thirdparty / linux.git / log
