Bobby Eshleman [Mon, 9 Jun 2025 16:39:24 +0000 (09:39 -0700)]
selftests/vsock: add initial vmtest.sh for vsock
This commit introduces a new vmtest.sh runner for vsock.
It uses virtme-ng/qemu to run tests in a VM. The tests validate G2H,
H2G, and loopback. The testing tools from tools/testing/vsock/ are
reused. Currently, only vsock_test is used.
VMCI and hyperv support is included in the config file to be built with
the -b option, though not used in the tests.
Only tested on x86.
To run:
$ make -C tools/testing/selftests TARGETS=vsock
$ tools/testing/selftests/vsock/vmtest.sh
or
$ make -C tools/testing/selftests TARGETS=vsock run_tests
Example runs (after make -C tools/testing/selftests TARGETS=vsock):
$ ./tools/testing/selftests/vsock/vmtest.sh
1..3
ok 0 vm_server_host_client
ok 1 vm_client_host_server
ok 2 vm_loopback
SUMMARY: PASS=3 SKIP=0 FAIL=0
Log: /tmp/vsock_vmtest_m7DI.log
$ mkdir -p ~/scratch
$ make -C tools/testing/selftests install TARGETS=vsock INSTALL_PATH=~/scratch
[... omitted ...]
$ cd ~/scratch
$ ./run_kselftest.sh
TAP version 13
1..1
# timeout set to 300
# selftests: vsock: vmtest.sh
# 1..3
# ok 0 vm_server_host_client
# ok 1 vm_client_host_server
# ok 2 vm_loopback
# SUMMARY: PASS=3 SKIP=0 FAIL=0
# Log: /tmp/vsock_vmtest_svEl.log
ok 1 selftests: vsock: vmtest.sh
Future work can include vsock_diag_test.
Because vsock requires a VM to test anything other than loopback, this
patch adds vmtest.sh as a kselftest itself. This is different than other
systems that have a "vmtest.sh", where it is used as a utility script to
spin up a VM to run the selftests as a guest (but isn't hooked into
kselftest).
net: stop napi kthreads when THREADED napi is disabled
Once the THREADED napi is disabled, the napi kthread should also be
stopped. Keeping the kthread intact after disabling THREADED napi makes
the PID of this kthread show up in the output of netlink 'napi-get' and
ps -ef output.
The is discussed in the patch below:
https://lore.kernel.org/all/20250502191548.559cc416@kernel.org
NAPI kthread should stop only if,
- There are no pending napi poll scheduled for this thread.
- There are no new napi poll scheduled for this thread while it has
stopped.
- The ____napi_schedule can correctly fallback to the softirq for napi
polling.
Since napi_schedule_prep provides mutual exclusion over STATE_SCHED bit,
it is safe to unset the STATE_THREADED when SCHED_THREADED is set or the
SCHED bit is not set. SCHED_THREADED being set means that SCHED is
already set and the kthread owns this napi.
To disable threaded napi, unset STATE_THREADED bit safely if
SCHED_THREADED is set or SCHED is unset. Once STATE_THREADED is unset
safely then wait for the kthread to unset the SCHED_THREADED bit so it
safe to stop the kthread.
Add a new test in nl_netdev to verify this behaviour.
Tested:
./tools/testing/selftests/net/nl_netdev.py
TAP version 13
1..6
ok 1 nl_netdev.empty_check
ok 2 nl_netdev.lo_check
ok 3 nl_netdev.page_pool_check
ok 4 nl_netdev.napi_list_check
ok 5 nl_netdev.dev_set_threaded
ok 6 nl_netdev.nsim_rxq_reset_down
# Totals: pass:6 fail:0 xfail:0 xpass:0 skip:0 error:0
Ran neper for 300 seconds and did enable/disable of thread napi in a
loop continuously.
Moon Yeounsu [Tue, 10 Jun 2025 00:01:30 +0000 (09:01 +0900)]
net: dlink: enable RMON MMIO access on supported devices
Enable memory-mapped I/O access to RMON statistics registers for devices
known to work correctly. Currently, only the D-Link DGE-550T (`0x4000`)
with PCI revision A3 (`0x0c`) is allowed.
To avoid issues on other hardware, a runtime check was added to restrict
MMIO usage. The `MEM_MAPPING` macro was removed in favor of runtime
detection.
To access RMON registers, the code `dw32(RmonStatMask, 0x0007ffff);`
must also be skipped, so this patch conditionally disables it as well.
The first 4 patches are by Vincent Mailhol and prepare the CAN netlink
interface for the introduction of CAN XL configuration.
Geert Uytterhoeven's patch updates the CAN networking documentation.
The last 2 patched are by Davide Caratti and introduce skb drop
reasons in the receive path of several CAN protocols.
* tag 'linux-can-next-for-6.17-20250610' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next:
can: add drop reasons in CAN protocols receive path
can: add drop reasons in the receive path of AF_CAN
documentation: networking: can: Document alloc_candev_mqs()
can: netlink: can_changelink(): rename tdc_mask into fd_tdc_flag_provided
can: bittiming: rename can_tdc_is_enabled() into can_fd_tdc_is_enabled()
can: bittiming: rename CAN_CTRLMODE_TDC_MASK into CAN_CTRLMODE_FD_TDC_MASK
can: netlink: replace tabulation by space in assignment
====================
octeontx2-pf: Avoid typecasts by simplifying otx2_atomic64_add macro
Just because otx2_atomic64_add is using u64 pointer as argument
all callers has to typecast __iomem void pointers which inturn
causing sparse warnings. Fix those by changing otx2_atomic64_add
argument to void pointer.
This patch removes unnecessary typecasts by marking the
mbox_regions array as __iomem since it is used to store
pointers to memory-mapped I/O (MMIO) regions. Also simplified
the call to readq() in PF driver by removing redundant type casts.
Add a new function, netif_subqueue_sent, which is a wrapper for
netdev_tx_sent_queue.
Drivers that use the subqueue variant macros, netif_subqueue_xxx,
identify queue by index and are not required to obtain
struct netdev_queue explicitly.
Such drivers still need to call netdev_tx_sent_queue which is a
counterpart of netif_subqueue_completed_wake. Allowing drivers to use a
subqueue variant for this purpose improves their code consistency by
always referring to queue by its index.
Jake moves from individual virtchnl RSS configuration values, for ice,
i40e, and iavf, to a common libie location and values.
Martyna and Dawid add counters for link_down_events to ice, i40e, and
ixgbe drivers. The counter increments only on actual physical link-down
events visible to the PHY. It does not increment when the user performs
a software-only interface down/up (e.g. ip link set dev down).
The counter does increment in cases where the interface is reinitialized
in a way that causes a real link drop - such as eg. when attaching
an XDP program, reconfiguring channels, or toggling certain priv-flags.
For ice:
Arkadiusz and Karol separate PTP and DPLL functionality to their
respective APIs.
Michal adds a separate handler for Flow Director command processing.
For iavf:
Ahmed converts driver to utilize core's IRQ affinity API.
For ixgbe:
Alok Tiwari fixes issues with some comments; typos, copy/paste errors,
etc.
* '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue:
ixgbe: Fix typos and clarify comments in X550 driver code
iavf: convert to NAPI IRQ affinity API
ice: add a separate Rx handler for flow director commands
ice: add ice driver PTP pin documentation
ice: change SMA pins to SDP in PTP API
ice: redesign dpll sma/u.fl pins control
ixgbe: add link_down_events statistic
i40e: add link_down_events statistic
ice: add link_down_events statistic
net: intel: move RSS packet classifier types to libie
net: intel: rename 'hena' to 'hashcfg' for clarity
====================
Jakub Kicinski [Mon, 9 Jun 2025 14:39:33 +0000 (07:39 -0700)]
uapi: in6: restore visibility of most IPv6 socket options
A decade ago commit 6d08acd2d32e ("in6: fix conflict with glibc")
hid the definitions of IPV6 options, because GCC was complaining
about duplicates. The commit did not list the warnings seen, but
trying to recreate them now I think they are (building iproute2):
In file included from ./include/uapi/rdma/rdma_user_cm.h:39,
from rdma.h:16,
from res.h:9,
from res-ctx.c:7:
../include/uapi/linux/in6.h:171:9: warning: ‘IPV6_ADD_MEMBERSHIP’ redefined
171 | #define IPV6_ADD_MEMBERSHIP 20
| ^~~~~~~~~~~~~~~~~~~
In file included from /usr/include/netinet/in.h:37,
from rdma.h:13:
/usr/include/bits/in.h:233:10: note: this is the location of the previous definition
233 | # define IPV6_ADD_MEMBERSHIP IPV6_JOIN_GROUP
| ^~~~~~~~~~~~~~~~~~~
../include/uapi/linux/in6.h:172:9: warning: ‘IPV6_DROP_MEMBERSHIP’ redefined
172 | #define IPV6_DROP_MEMBERSHIP 21
| ^~~~~~~~~~~~~~~~~~~~
/usr/include/bits/in.h:234:10: note: this is the location of the previous definition
234 | # define IPV6_DROP_MEMBERSHIP IPV6_LEAVE_GROUP
| ^~~~~~~~~~~~~~~~~~~~
Compilers don't complain about redefinition if the defines
are identical, but here we have the kernel using the literal
value, and glibc using an indirection (defining to a name
of another define, with the same numerical value).
Problem is, the commit in question hid all the IPV6 socket
options, and glibc has a pretty sparse list. For instance
it lacks Flow Label related options. Willem called this out
in commit 3fb321fde22d ("selftests/net: ipv6 flowlabel"):
/* uapi/glibc weirdness may leave this undefined */
#ifndef IPV6_FLOWINFO
#define IPV6_FLOWINFO 11
#endif
More interestingly some applications (socat) use
a #ifdef IPV6_FLOWINFO to gate compilation of thier
rudimentary flow label support. (For added confusion
socat misspells it as IPV4_FLOWINFO in some places.)
Hide only the two defines we know glibc has a problem
with. If we discover more warnings we can hide more
but we should avoid covering the entire block of
defines for "IPV6 socket options".
The counters are exposed using ethtool_ops->get_ethtool_stats and
ethtool_ops->get_strings. This feature/counters are not available
to all versions of hardware.
====================
netconsole: Optimize console registration and improve testing
During performance analysis of console subsystem latency, I discovered that
netconsole registers console handlers even when no active targets exist.
These orphaned console handlers are invoked on every printk() call, get
the lock, iterate through empty target lists, and consume CPU cycles
without performing any useful work.
This patch series addresses the inefficiency by:
1. Implementing dynamic console registration/unregistration based on target
availability, ensuring console handlers are only active when needed
2. Adding automatic cleanup of unused console registrations when targets
are disabled or removed
3. Extending the selftest suite to cover non-extended console format,
which was previously untested
The optimization reduces printk() overhead by eliminating unnecessary
function calls and list traversals when netconsole targets are not
configured, improving overall system performance during heavy logging
scenarios.
Breno Leitao [Mon, 9 Jun 2025 09:46:29 +0000 (02:46 -0700)]
selftests: netconsole: Add support for basic netconsole target format
Extend the netconsole selftest to validate both basic and extended
target formats. The basic format is a simpler variant that doesn't
support userdata or release functionality.
The test now validates that netconsole works correctly in both
configurations, improving test coverage for different netconsole
deployment scenarios.
Breno Leitao [Mon, 9 Jun 2025 09:46:28 +0000 (02:46 -0700)]
selftests: netconsole: Do not exit from inside the validation function
Remove the exit call from validate_result() function and move the
test exit logic to the main script. This allows the function to
be reused in scenarios where the test needs to continue execution
after validation, rather than terminating immediately.
The validate_result() function should focus on validation logic
only, while the calling script maintains control over program
flow and exit conditions. This change improves code modularity
and prepares for potential future enhancements where multiple
validations might be needed in a single test run.
Breno Leitao [Mon, 9 Jun 2025 09:46:27 +0000 (02:46 -0700)]
netconsole: Add automatic console unregistration on target removal
Add unregister_netcons_consoles() function to automatically unregister
console handlers when no targets of the corresponding type remain active.
The function iterates through the target list to determine which console
types (basic vs extended) are still needed, and unregisters any console
handlers that are no longer required. This prevents having registered
console handlers without corresponding active targets.
The function is called when a target is disabled and moved to the cleanup
list, ensuring proper cleanup of unused console registrations.
Breno Leitao [Mon, 9 Jun 2025 09:46:26 +0000 (02:46 -0700)]
netconsole: Only register console drivers when targets are configured
The netconsole driver currently registers the basic console driver
unconditionally during initialization, even when only extended targets
are configured. This results in unnecessary console registration and
performance overhead, as the write_msg() callback is invoked for every
log message only to return early when no matching targets are found.
Optimize the driver by conditionally registering console drivers based
on the actual target configuration. The basic console driver is now
registered only when non-extended targets exist, same as the extended
console. The implementation also handles dynamic target creation through
the configfs interface.
This change eliminates unnecessary console driver registrations,
redundant write_msg() callbacks for unused console types, and associated
lock contention and target list iterations. The optimization is
particularly beneficial for systems using only the most common extended
console type.
Merge patch series "can: add drop reasons in the receive path"
Davide Caratti <dcaratti@redhat.com> says:
drop reasons have been recently introduced to improve debuggability of
networking stack. This series introduces drop reasons in the RX path
of the CAN protocol stack.
Davide Caratti [Wed, 4 Jun 2025 16:06:05 +0000 (18:06 +0200)]
can: add drop reasons in CAN protocols receive path
sock_queue_rcv_skb() can fail because of lack of memory resources: use
drop reasons and pass the receiving socket to the tracepoint, so that
it's possible to better locate/debug such events.
Tested with:
| # modprobe vcan echo=1
| # ip link add name vcan2 type vcan
| # ip link set dev vcan2 up
| # ./netlayer/tst-proc 1 &
| # bg
| # while true ; do perf record -e skb:kfree_skb -aR -- \
| > ./raw/tst-raw-sendto vcan2 ; perf script ; done
| [...]
| tst-raw-sendto 10942 [000] 506428.431856: skb:kfree_skb: skbaddr=0xffff97cec38b4200 rx_sk=0xffff97cf0f75a800 protocol=12 location=raw_rcv+0x20e reason: SOCKET_RCVBUF
Davide Caratti [Wed, 4 Jun 2025 16:06:04 +0000 (18:06 +0200)]
can: add drop reasons in the receive path of AF_CAN
Besides the existing pr_warn_once(), use skb drop reasons in case AF_CAN
layer drops non-conformant CAN{,FD,XL} frames, or conformant frames
received by "wrong" devices, so that it's possible to debug (and count)
such events using existing tracepoints:
Since the introduction of alloc_candev_mqs() and friends, there is no
longer a need to allocate a generic network device and perform explicit
CAN-specific setup. Remove the code showing this setup, and document
alloc_candev_mqs() instead.
Merge patch series "can: netlink: preparation before introduction of CAN XL"
Vincent Mailhol <mailhol.vincent@wanadoo.fr> says:
An RFC was sent last weekend to kick-off the discussion of the
introduction of CAN XL [1]. While the series received some positive
feedback, it is far from completion. Some work is still needed to:
- adjust the nesting of the IFLA_CAN_XL_DATA_BITTIMING_CONST in the
netlink interface
- add the CAN XL PWM configuration
and this TODO list may grow if more feedback is received.
Regardless of this, the RFC started with a tree wide refactor followed
by a set of trivial patches to do some clean-up and some renaming in
preparation of the introduction of CAN XL.
This series just contains those preparation patch which were cherry
picked from the RFC and rebased on of top of linux-can-next/main:
- the first patch is purely cosmetic and fixes a trivial tabulation
mistake.
- the last three patches do some renaming: both the CAN FD and the
CAN XL have databittiming parameters. In order not to get confused
once CAN XL will be introduced, many symbols are modified to
explicitly add CAN FD in their names.
The goal is to have those merged first to remove some overhead from
the netlink CAN XL main series before tacking care of the other
comments.
Vincent Mailhol [Tue, 12 Nov 2024 16:50:20 +0000 (01:50 +0900)]
can: netlink: can_changelink(): rename tdc_mask into fd_tdc_flag_provided
The only purpose of the tdc_mask variable is to check whether or not
any tdc flags (CAN_CTRLMODE_TDC_{AUTO,MANUAL}) were provided. At this
point, the actual value of the flags do no matter any more because
these can be deduced from some other information.
Rename the tdc_mask variable into fd_tdc_flag_provided to make this
more explicit. Note that the fd_ prefix is added in preparation of the
introduction of CAN XL.
Vincent Mailhol [Tue, 12 Nov 2024 16:50:19 +0000 (01:50 +0900)]
can: bittiming: rename can_tdc_is_enabled() into can_fd_tdc_is_enabled()
With the introduction of CAN XL, a new can_xl_tdc_is_enabled() helper
function will be introduced later on. Rename can_tdc_is_enabled() into
can_fd_tdc_is_enabled() to make it more explicit that this helper is
meant for CAN FD.
Vincent Mailhol [Tue, 12 Nov 2024 16:50:18 +0000 (01:50 +0900)]
can: bittiming: rename CAN_CTRLMODE_TDC_MASK into CAN_CTRLMODE_FD_TDC_MASK
With the introduction of CAN XL, a new CAN_CTRLMODE_XL_TDC_MASK will
be introduced later on. Because CAN_CTRLMODE_TDC_MASK is not part of
the uapi, rename it to CAN_CTRLMODE_FD_TDC_MASK to make it more
explicit that this mask is meant for CAN FD.
Vincent Mailhol [Tue, 12 Nov 2024 16:50:17 +0000 (01:50 +0900)]
can: netlink: replace tabulation by space in assignment
commit cfd98c838cbe ("can: netlink: move '=' operators back to
previous line (checkpatch fix)") inadvertently introduced a tabulation
between the IFLA_CAN_DATA_BITTIMING_CONST array index and the equal
sign.
Alok Tiwari [Thu, 22 May 2025 07:47:26 +0000 (00:47 -0700)]
ixgbe: Fix typos and clarify comments in X550 driver code
Corrected spelling errors such as "simular" -> "similar",
"excepted" -> "accepted", and "Determime" -> "Determine".
Fixed including incorrect word usage ("to MAC" -> "two MAC")
and improved awkward phrasing.
Aligned function header descriptions with their actual functionality
(e.g., "Writes a value" -> "Reads a value").
Corrected typo in error code from -ENIVAL to -EINVAL.
Improved overall clarity and consistency in comment across various
functions.
These changes improve maintainability and readability of the code
without affecting functionality.
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Ahmed Zaki [Fri, 16 May 2025 22:19:09 +0000 (16:19 -0600)]
iavf: convert to NAPI IRQ affinity API
Commit bd7c00605ee0 ("net: move aRFS rmap management and CPU affinity
to core") allows the drivers to delegate the IRQ affinity to the NAPI
instance. However, the driver needs to use a persistent NAPI config
and explicitly set/unset the NAPI<->IRQ association.
Convert to the new IRQ affinity API.
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Signed-off-by: Ahmed Zaki <ahmed.zaki@intel.com> Tested-by: Rafal Romanowski <rafal.romanowski@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Michal Kubiak [Wed, 14 May 2025 12:37:24 +0000 (14:37 +0200)]
ice: add a separate Rx handler for flow director commands
The "ice" driver implementation uses the control VSI to handle
the flow director configuration for PFs and VFs.
Unfortunately, although a separate VSI type was created to handle flow
director queues, the Rx queue handler was shared between the flow
director and a standard NAPI Rx handler.
Such a design approach was not very flexible. First, it mixed hotpath
and slowpath code, blocking their further optimization. It also created
a huge overkill for the flow director command processing, which is
descriptor-based only, so there is no need to allocate Rx data buffers.
For the above reasons, implement a separate Rx handler for the control
VSI. Also, remove from the NAPI handler the code dedicated to
configuring the flow director rules on VFs.
Do not allocate Rx data buffers to the flow director queues because
their processing is descriptor-based only.
Finally, allow Rx data queues to be allocated only for VSIs that have
netdev assigned to them.
This handler splitting approach is the first step in converting the
driver to use the Page Pool (which can only be used for data queues).
Test hints:
1. Create a VF for any PF managed by the ice driver.
2. In a loop, add and delete flow director rules for the VF, e.g.:
for i in {1..128}; do
q=$(( i % 16 ))
ethtool -N ens802f0v0 flow-type tcp4 dst-port "$i" action "$q"
done
for i in {0..127}; do
ethtool -N ens802f0v0 delete "$i"
done
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Suggested-by: Michal Swiatkowski <michal.swiatkowski@intel.com> Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Michal Kubiak <michal.kubiak@intel.com> Tested-by: Rafal Romanowski <rafal.romanowski@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Karol Kolacinski [Tue, 22 Apr 2025 16:01:49 +0000 (18:01 +0200)]
ice: add ice driver PTP pin documentation
Add a description of PTP pins support by the adapters to ice driver
documentation.
Reviewed-by: Milena Olech <milena.olech@intel.com> Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com> Signed-off-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Karol Kolacinski [Tue, 22 Apr 2025 16:01:48 +0000 (18:01 +0200)]
ice: change SMA pins to SDP in PTP API
This change aligns E810 PTP pin control to all other products.
Currently, SMA/U.FL port expanders are controlled together with SDP pins
connected to 1588 clock. To align this, separate this control by
exposing only SDP20..23 pins in PTP API on adapters with DPLL.
Clear error for all E810 on absent NVM pin section or other errors to
allow proper initialization on SMA E810 with NVM section.
Use ARRAY_SIZE for pin array instead of internal definition.
Reviewed-by: Milena Olech <milena.olech@intel.com> Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com> Signed-off-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
DPLL-enabled E810 NIC driver provides user with list of input and output
pins. Hardware internal design impacts user control over SMA and U.FL
pins. Currently end-user view on those dpll pins doesn't provide any layer
of abstraction. On the hardware level SMA and U.FL pins are tied together
due to existence of direction control logic for each pair:
- SMA1 (bi-directional) and U.FL1 (only output)
- SMA2 (bi-directional) and U.FL2 (only input)
The user activity on each pin of the pair may impact the state of the
other.
Previously all the pins were provided to the user as is, without the
control over SMA pins direction.
Introduce a software controlled layer of abstraction over external board
pins, instead of providing the user with access to raw pins connected to
the dpll:
- new software controlled SMA and U.FL pins,
- callback operations directing user requests to corresponding hardware
pins according to the runtime configuration,
- ability to control SMA pins direction.
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Signed-off-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Introduce a link_down_events counter to the ixgbe driver, incremented
each time the link transitions from up to down.
This counter can help diagnose issues related to link stability,
such as port flapping or unexpected link drops.
The value is exposed via ethtool's get_link_ext_stats() interface.
Reviewed-by: Kory Maincent <kory.maincent@bootlin.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Signed-off-by: Martyna Szapar-Mudlaw <martyna.szapar-mudlaw@linux.intel.com> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Dawid Osuchowski [Wed, 21 May 2025 14:23:32 +0000 (16:23 +0200)]
i40e: add link_down_events statistic
Introduce a link_down_events counter to the i40e driver, incremented
each time the link transitions from up to down.
This counter can help diagnose issues related to link stability,
such as port flapping or unexpected link drops.
The value is exposed via ethtool's get_link_ext_stats() interface.
Co-developed-by: Martyna Szapar-Mudlaw <martyna.szapar-mudlaw@linux.intel.com> Signed-off-by: Martyna Szapar-Mudlaw <martyna.szapar-mudlaw@linux.intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Signed-off-by: Dawid Osuchowski <dawid.osuchowski@linux.intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Introduce a link_down_events counter to the ice driver, incremented
each time the link transitions from up to down.
This counter can help diagnose issues related to link stability,
such as port flapping or unexpected link drops.
The value is exposed via ethtool's get_link_ext_stats() interface.
Reviewed-by: Kory Maincent <kory.maincent@bootlin.com> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Martyna Szapar-Mudlaw <martyna.szapar-mudlaw@linux.intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Jacob Keller [Mon, 5 May 2025 20:14:23 +0000 (13:14 -0700)]
net: intel: move RSS packet classifier types to libie
The Intel i40e, iavf, and ice drivers all include a definition of the
packet classifier filter types used to program RSS hash enable bits. For
i40e, these bits are used for both the PF and VF to configure the PFQF_HENA
and VFQF_HENA registers.
For ice and iAVF, these bits are used to communicate the desired hash
enable filter over virtchnl via its struct virtchnl_rss_hashena. The
virtchnl.h header makes no mention of where the bit definitions reside.
Maintaining a separate copy of these bits across three drivers is
cumbersome. Move the definition to libie as a new pctype.h header file.
Each driver can include this, and drop its own definition.
The ice implementation also defined a ICE_AVF_FLOW_FIELD_INVALID, intending
to use this to indicate when there were no hash enable bits set. This is
confusing, since the enumeration is using bit positions. A value of 0
*should* indicate the first bit. Instead, rewrite the code that uses
ICE_AVF_FLOW_FIELD_INVALID to just check if the avf_hash is zero. From
context this should be clear that we're checking if none of the bits are
set.
The values are kept as bit positions instead of encoding the BIT_ULL
directly into their value. While most users will simply use BIT_ULL
immediately, i40e uses the macros both with BIT_ULL and test_bit/set_bit
calls.
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Rafal Romanowski <rafal.romanowski@intel.com> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Jacob Keller [Mon, 5 May 2025 20:14:22 +0000 (13:14 -0700)]
net: intel: rename 'hena' to 'hashcfg' for clarity
i40e, ice, and iAVF all use 'hena' as a shorthand for the "hash enable"
configuration. This comes originally from the X710 datasheet 'xxQF_HENA'
registers. In the context of the registers the meaning is fairly clear.
However, on its own, hena is a weird name that can be more difficult to
understand. This is especially true in ice. The E810 hardware doesn't even
have registers with HENA in the name.
Replace the shorthand 'hena' with 'hashcfg'. This makes it clear the
variables deal with the Hash configuration, not just a single boolean
on/off for all hashing.
Do not update the register names. These come directly from the datasheet
for X710 and X722, and it is more important that the names can be searched.
Suggested-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Rafal Romanowski <rafal.romanowski@intel.com> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
- fix net_devmem_bind_dmabuf() stub when DEVMEM not compiled
- eth: airoha: fixes for config / accel in bridge mode
Previous releases - regressions:
- Bluetooth: hci_qca: move the SoC type check to the right place, fix
GPIO integration
- prevent a NULL deref in rtnl_create_link() after locking changes
- fix udp gso skb_segment after pull from frag_list
- hv_netvsc: fix potential deadlock in netvsc_vf_setxdp()
Previous releases - always broken:
- netfilter:
- nf_nat: also check reverse tuple to obtain clashing entry
- nf_set_pipapo_avx2: fix initial map fill (zeroing)
- fix the helper for incremental update of packet checksums after
modifying the IP address, used by ILA and BPF
- eth:
- stmmac: prevent div by 0 when clock rate is misconfigured
- ice: fix Tx scheduler handling of XDP and changing queue count
- eth: fix support for the RGMII interface when delays configured"
* tag 'net-6.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (76 commits)
calipso: unlock rcu before returning -EAFNOSUPPORT
seg6: Fix validation of nexthop addresses
net: prevent a NULL deref in rtnl_create_link()
net: annotate data-races around cleanup_net_task
selftests: drv-net: tso: make bkg() wait for socat to quit
selftests: drv-net: tso: fix the GRE device name
selftests: drv-net: add configs for the TSO test
wireguard: device: enable threaded NAPI
netlink: specs: rt-link: decode ip6gre
netlink: specs: rt-link: add missing byte-order properties
net: wwan: mhi_wwan_mbim: use correct mux_id for multiplexing
wifi: cfg80211/mac80211: correctly parse S1G beacon optional elements
net: dsa: b53: do not touch DLL_IQQD on bcm53115
net: dsa: b53: allow RGMII for bcm63xx RGMII ports
net: dsa: b53: do not configure bcm63xx's IMP port interface
net: dsa: b53: do not enable RGMII delay on bcm63xx
net: dsa: b53: do not enable EEE on bcm63xx
net: ti: icssg-prueth: Fix swapped TX stats for MII interfaces.
selftests: netfilter: nft_nat.sh: add test for reverse clash with nat
netfilter: nf_nat: also check reverse tuple to obtain clashing entry
...
Linus Torvalds [Thu, 5 Jun 2025 18:45:33 +0000 (11:45 -0700)]
Merge tag 'uml-for-linux-6.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/uml/linux
Pull UML updates from Johannes Berg:
"The only really new thing is the long-standing seccomp work
(originally from 2021!). Wven if it still isn't enabled by default due
to security concerns it can still be used e.g. for tests.
- remove obsolete network transports
- remove PCI IO port support
- start adding seccomp-based process handling instead of ptrace"
* tag 'uml-for-linux-6.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/uml/linux: (29 commits)
um: remove "extern" from implementation of sigchld_handler
um: fix unused variable warning
um: fix SECCOMP 32bit xstate register restore
um: pass FD for memory operations when needed
um: Add SECCOMP support detection and initialization
um: Implement kernel side of SECCOMP based process handling
um: Track userspace children dying in SECCOMP mode
um: Add helper functions to get/set state for SECCOMP
um: Add stub side of SECCOMP/futex based process handling
um: Move faultinfo extraction into userspace routine
um: vector: Use mac_pton() for MAC address parsing
um: vector: Clean up and modernize log messages
um: chan_kern: use raw spinlock for irqs_to_free_lock
MAINTAINERS: remove obsolete file entry in TUN/TAP DRIVER
um: Fix tgkill compile error on old host OSes
um: stop using PCI port I/O
um: Remove legacy network transport infrastructure
um: vector: Eliminate the dependency on uml_net
um: Remove obsolete legacy network transports
um/asm: Replace "REP; NOP" with PAUSE mnemonic
...
Linus Torvalds [Thu, 5 Jun 2025 18:39:17 +0000 (11:39 -0700)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"We've got a couple of build fixes when using LLD, a missing TLB
invalidation and a workaround for broken firmware on SoCs with CPUs
that implement MPAM:
- Disable problematic linker assertions for broken versions of LLD
- Work around sporadic link failure with LLD and various randconfig
builds
- Fix missing invalidation in the TLB batching code when reclaim
races with mprotect() and friends
- Add a command-line override for MPAM to allow booting on systems
with broken firmware"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: Add override for MPAM
arm64/mm: Close theoretical race where stale TLB entry remains valid
arm64: Work around convergence issue with LLD linker
arm64: Disable LLD linker ASSERT()s for the time being
Linus Torvalds [Thu, 5 Jun 2025 18:33:09 +0000 (11:33 -0700)]
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rmk/linux
Pull ARM fixes from Russell King:
- Fix arch_memremap_can_ram_remap() which incorrectly passed a PFN to
memblock_is_map_memory rather than the actual address.
- Disallow kernel mode NEON when IRQs are disabled
Explanation:
"To avoid having to preserve/restore kernel mode NEON state when
such a softirq is taken softirqs are now disabled when using the
NEON from task context."
should explain that it's nested kernel mode.
In other words, softirqs from user mode are fine, because the context
will be preserved. softirqs from kernel mode may be from a context
that has already saved the user NEON state, and thus we would need to
preserve the NEON state for the parent kernel mode context, and this
we don't allow.
The problem occurs when the kernel context disables hard IRQs, and
then uses NEON. When it's finished, and restores the userspace NEON
state, we call local_bh_enable() with hard IRQs disabled, which
causes a warning.
This commit addresses that by disallowing the use of NEON with hard
IRQs disabled.
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rmk/linux:
ARM: 9446/1: Disallow kernel mode NEON when IRQs are disabled
ARM: 9447/1: arm/memremap: fix arch_memremap_can_ram_remap()
Linus Torvalds [Thu, 5 Jun 2025 15:54:47 +0000 (08:54 -0700)]
Merge tag 'rtc-6.16' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux
Pull RTC updates from Alexandre Belloni:
"There are two new drivers this cycle. There is also support for a
negative offset for RTCs that have been shipped with a date set using
an epoch that is before 1970. This unfortunately happens with some
products that ship with a vendor kernel and an out of tree driver.
Core:
- support negative offsets for RTCs that have shipped with an epoch
earlier than 1970
New drivers:
- NXP S32G2/S32G3
- Sophgo CV1800
Drivers:
- loongson: fix missing alarm notifications for ACPI
- m41t80: kickstart ocillator upon failure
- mt6359: mt6357 support
- pcf8563: fix wrong alarm register
- sh: cleanups"
* tag 'rtc-6.16' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux: (39 commits)
rtc: mt6359: Add mt6357 support
rtc: test: Test date conversion for dates starting in 1900
rtc: test: Also test time and wday outcome of rtc_time64_to_tm()
rtc: test: Emit the seconds-since-1970 value instead of days-since-1970
rtc: Fix offset calculation for .start_secs < 0
rtc: Make rtc_time64_to_tm() support dates before 1970
rtc: pcf8563: fix wrong alarm register
rtc: rzn1: support input frequencies other than 32768Hz
rtc: rzn1: Disable controller before initialization
dt-bindings: rtc: rzn1: add optional second clock
rtc: m41t80: reduce verbosity
rtc: m41t80: kickstart ocillator upon failure
rtc: s32g: add NXP S32G2/S32G3 SoC support
dt-bindings: rtc: add schema for NXP S32G2/S32G3 SoCs
dt-bindings: at91rm9260-rtt: add microchip,sama7d65-rtt
dt-bindings: rtc: at91rm9200: add microchip,sama7d65-rtc
rtc: loongson: Add missing alarm notifications for ACPI RTC events
rtc: sophgo: add rtc support for Sophgo CV1800 SoC
rtc: stm32: drop unused module alias
rtc: s3c: drop unused module alias
...
Linus Torvalds [Thu, 5 Jun 2025 15:49:30 +0000 (08:49 -0700)]
Merge tag 'dmaengine-6.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine
Pull dmaengine updates from Vinod Koul:
"A fairly small update for the dmaengine subsystem. This has a new ARM
dmaengine driver and couple of new device support and few driver
changes:
New support:
- Renesas RZ/V2H(P) dma support for r9a09g057
- Arm DMA-350 driver
- Tegra Tegra264 ADMA support
Linus Torvalds [Thu, 5 Jun 2025 15:20:21 +0000 (08:20 -0700)]
Merge tag 'phy-for-6.16' of git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy
Pull phy updates from Vinod Koul:
"As usual featuring couple of new driver and bunch of new device
support and some driver changes to Freescale, rockchip driver along
with couple of yaml binding conversions.
New Support:
- Qualcomm IPQ5424 qusb2 support, IPQ5018 uniphy-pcie driver
- Rockchip usb2 support for RK3562, RK3036 usb2 phy support
- Samsung exynos2200 eusb2 phy support and driver refactoring for
this support, exynos7870 USBDRD support
- Mediatek MT7988 xs-phy support
- Broadcom BCM74110 usb phy support
- Renesas RZ/V2H(P) usb2 phy support
Updates:
- Freescale phy rate claculation updates, i.MX95 tuning support
- Better error handling for amlogic pcie phy
- Rockchip color depth configuration and management support
- Yaml binding conversion for RK3399 Type-C and PCIe Phy"
* tag 'phy-for-6.16' of git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy: (77 commits)
phy: tegra: p2u: Broaden architecture dependency
phy: rockchip: inno-usb2: Add usb2 phy support for rk3562
dt-bindings: phy: rockchip,inno-usb2phy: add rk3562
phy: rockchip: inno-usb2: add phy definition for rk3036
dt-bindings: phy: rockchip,inno-usb2phy: add rk3036 compatible
phy: freescale: fsl-samsung-hdmi: Improve LUT search for best clock
phy: freescale: fsl-samsung-hdmi: Refactor finding PHY settings
phy: freescale: fsl-samsung-hdmi: Rename phy_clk_round_rate
phy: renesas: phy-rcar-gen3-usb2: Add USB2.0 PHY support for RZ/V2H(P)
phy: renesas: phy-rcar-gen3-usb2: Sort compatible entries by SoC part number
dt-bindings: phy: renesas,usb2-phy: Document RZ/V2H(P) SoC
dt-bindings: phy: renesas,usb2-phy: Add clock constraint for RZ/G2L family
phy: exynos5-usbdrd: support Exynos USBDRD 3.2 4nm controller
phy: phy-snps-eusb2: add support for exynos2200
phy: phy-snps-eusb2: refactor reference clock init
phy: phy-snps-eusb2: make reset control optional
phy: phy-snps-eusb2: make repeater optional
phy: phy-snps-eusb2: split phy init code
phy: phy-snps-eusb2: refactor constructs names
phy: move phy-qcom-snps-eusb2 out of its vendor sub-directory
...
Linus Torvalds [Thu, 5 Jun 2025 15:07:24 +0000 (08:07 -0700)]
Merge tag 'soundwire-6.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire
Pull soundwire updates from Vinod Koul:
"A couple of small core changes and an Intel driver change:
- sdw_assign_device_num() logic simplification, using internal slave
id for irqs and optimizing computing of port params in specific
stream states
- Intel driver updates for ACE3+ microphone privacy status reporting
and enabling the status in HDA Intel driver"
* tag 'soundwire-6.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire:
soundwire: only compute port params in specific stream states
ASoC: SOF: Intel: hda: Set the mic_privacy flag for soundwire with ACE3+
soundwire: intel: Add awareness of ACE3+ microphone privacy
soundwire: bus: Add internal slave ID and use for IRQs
soundwire: bus: Simplify sdw_assign_device_num()
Ido Schimmel [Wed, 4 Jun 2025 11:32:52 +0000 (14:32 +0300)]
seg6: Fix validation of nexthop addresses
The kernel currently validates that the length of the provided nexthop
address does not exceed the specified length. This can lead to the
kernel reading uninitialized memory if user space provided a shorter
length than the specified one.
Fix by validating that the provided length exactly matches the specified
one.
Fixes: d1df6fd8a1d2 ("ipv6: sr: define core operations for seg6local lightweight tunnel") Reviewed-by: Petr Machata <petrm@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: David Ahern <dsahern@kernel.org> Link: https://patch.msgid.link/20250604113252.371528-1-idosch@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Eric Dumazet [Wed, 4 Jun 2025 10:58:15 +0000 (10:58 +0000)]
net: prevent a NULL deref in rtnl_create_link()
At the time rtnl_create_link() is running, dev->netdev_ops is NULL,
we must not use netdev_lock_ops() or risk a NULL deref if
CONFIG_NET_SHAPER is defined.
Jakub Kicinski [Wed, 4 Jun 2025 01:20:55 +0000 (18:20 -0700)]
selftests: drv-net: tso: make bkg() wait for socat to quit
Commit 846742f7e32f ("selftests: drv-net: add a warning for
bkg + shell + terminate") added a warning for bkg() used
with terminate=True. The tso test was missed as we didn't
have it running anywhere in NIPA. Add exit_wait=True, to avoid:
# Warning: combining shell and terminate is risky!
# SIGTERM may not reach the child on zsh/ksh!
Jakub Kicinski [Wed, 4 Jun 2025 01:20:31 +0000 (18:20 -0700)]
selftests: drv-net: tso: fix the GRE device name
The device type for IPv4 GRE is "gre" not "ipgre",
unlike for IPv6 which uses "ip6gre".
Not sure how I missed this when writing the test, perhaps
because all HW I have access to is on an IPv6-only network.
Fixes: 0d0f4174f6c8 ("selftests: drv-net: add a simple TSO test") Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Willem de Bruijn <willemb@google.com> Link: https://patch.msgid.link/20250604012031.891242-1-kuba@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub Kicinski [Wed, 4 Jun 2025 00:16:52 +0000 (17:16 -0700)]
selftests: drv-net: add configs for the TSO test
Add missing config options for the tso.py test, specifically
to make sure the kernel is built with vxlan and gre tunnels.
I noticed this while adding a TSO-capable device QEMU to the CI.
Previously we only run virtio tests and it doesn't report LSO
stats on the QEMU we have.
Jakub Kicinski [Thu, 5 Jun 2025 14:59:31 +0000 (07:59 -0700)]
Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:
====================
iavf: get rid of the crit lock
Przemek Kitszel says:
Fix some deadlocks in iavf, and make it less error prone for the future.
Patch 1 is simple and independent from the rest.
Patches 2, 3, 4 are strictly a refactor, but it enables the last patch
to be much smaller.
(Technically Jake given his RB tags not knowing I will send it to -net).
Patch 5 just adds annotations, this also helps prove last patch to be correct.
Patch 6 removes the crit lock, with its unusual try_lock()s.
I have more refactoring for scheduling done for -next, to be sent soon.
There is a simple test:
add VF; decrease number of queueus; remove VF
that was way too hard to pass without this series :)
* '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
iavf: get rid of the crit lock
iavf: sprinkle netdev_assert_locked() annotations
iavf: extract iavf_watchdog_step() out of iavf_watchdog_task()
iavf: simplify watchdog_task in terms of adminq task scheduling
iavf: centralize watchdog requeueing itself
iavf: iavf_suspend(): take RTNL before netdev_lock()
====================
Mirco Barone [Thu, 5 Jun 2025 12:06:16 +0000 (14:06 +0200)]
wireguard: device: enable threaded NAPI
Enable threaded NAPI by default for WireGuard devices in response to low
performance behavior that we observed when multiple tunnels (and thus
multiple wg devices) are deployed on a single host. This affects any
kind of multi-tunnel deployment, regardless of whether the tunnels share
the same endpoints or not (i.e., a VPN concentrator type of gateway
would also be affected).
The problem is caused by the fact that, in case of a traffic surge that
involves multiple tunnels at the same time, the polling of the NAPI
instance of all these wg devices tends to converge onto the same core,
causing underutilization of the CPU and bottlenecking performance.
This happens because NAPI polling is hosted by default in softirq
context, but the WireGuard driver only raises this softirq after the rx
peer queue has been drained, which doesn't happen during high traffic.
In this case, the softirq already active on a core is reused instead of
raising a new one.
As a result, once two or more tunnel softirqs have been scheduled on
the same core, they remain pinned there until the surge ends.
In our experiments, this almost always leads to all tunnel NAPIs being
handled on a single core shortly after a surge begins, limiting
scalability to less than 3× the performance of a single tunnel, despite
plenty of unused CPU cores being available.
The proposed mitigation is to enable threaded NAPI for all WireGuard
devices. This moves the NAPI polling context to a dedicated per-device
kernel thread, allowing the scheduler to balance the load across all
available cores.
On our 32-core gateways, enabling threaded NAPI yields a ~4× performance
improvement with 16 tunnels, increasing throughput from ~13 Gbps to
~48 Gbps. Meanwhile, CPU usage on the receiver (which is the bottleneck)
jumps from 20% to 100%.
We have found no performance regressions in any scenario we tested.
Single-tunnel throughput remains unchanged.
Paolo Abeni [Thu, 5 Jun 2025 11:37:02 +0000 (13:37 +0200)]
Merge tag 'nf-25-06-05' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following patchset contains Netfilter fixes for net:
1) Zero out the remainder in nft_pipapo AVX2 implementation, otherwise
next lookup could bogusly report a mismatch. This is followed by two
patches to update nft_pipapo selftests to cover for the previous bug.
From Florian Westphal.
2) Check for reverse tuple too in case of esoteric NAT collisions for
UDP traffic and extend selftest coverage. Also from Florian.
netfilter pull request 25-06-05
* tag 'nf-25-06-05' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
selftests: netfilter: nft_nat.sh: add test for reverse clash with nat
netfilter: nf_nat: also check reverse tuple to obtain clashing entry
selftests: netfilter: nft_concat_range.sh: add datapath check for map fill bug
selftests: netfilter: nft_concat_range.sh: prefer per element counters for testing
netfilter: nf_set_pipapo_avx2: fix initial map fill
====================
Adding GRE tunnels to the .config for driver tests caused
some unhappiness in YNL, as it can't decode all the link
attrs on the system. Add ip6gre support to fix the tests.
This is similar to commit 6ffdbb93a59c ("netlink: specs:
rt_link: decode ip6tnl, vti and vti6 link attrs").
====================
Jakub Kicinski [Tue, 3 Jun 2025 13:53:57 +0000 (06:53 -0700)]
netlink: specs: rt-link: decode ip6gre
Driver tests now require GRE tunnels, while we don't configure
them with YNL, YNL will complain when it sees link types it
doesn't recognize. Teach it decoding ip6gre tunnels. The attrs
are largely the same as IPv4 GRE.
Correct the type of encap-limit, but note that this attr is
only used in ip6gre, so the mistake didn't matter until now.
Fixes: 0d0f4174f6c8 ("selftests: drv-net: add a simple TSO test") Signed-off-by: Jakub Kicinski <kuba@kernel.org> Reviewed-by: Donald Hunter <donald.hunter@gmail.com> Link: https://patch.msgid.link/20250603135357.502626-3-kuba@kernel.org Signed-off-by: Paolo Abeni <pabeni@redhat.com>
A number of fields in the ip tunnels are lacking the big-endian
designation. I suspect this is not intentional, as decoding
the ports with the right endian seems objectively beneficial.
Fixes: 6ffdbb93a59c ("netlink: specs: rt_link: decode ip6tnl, vti and vti6 link attrs") Fixes: 077b6022d24b ("doc/netlink/specs: Add sub-message type to rt_link family") Signed-off-by: Jakub Kicinski <kuba@kernel.org> Reviewed-by: Donald Hunter <donald.hunter@gmail.com> Link: https://patch.msgid.link/20250603135357.502626-2-kuba@kernel.org Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Paolo Abeni [Thu, 5 Jun 2025 10:37:10 +0000 (12:37 +0200)]
Merge tag 'ovpn-net-20250603' of https://github.com/OpenVPN/ovpn-net-next
Antonio Quartulli says:
====================
In this batch you can find the following bug fixes:
Patch 1: when releasing a UDP socket we were wrongly invoking
setup_udp_tunnel_sock() with an empty config. This was not
properly shutting down the UDP encap state.
With this patch we simply undo what was done during setup.
Patch 2: ovpn was holding a reference to a 'struct socket'
without increasing its reference counter. This was intended
and worked as expected until we hit a race condition where
user space tries to close the socket while kernel space is
also releasing it. In this case the (struct socket *)->sk
member would disappear under our feet leading to a null-ptr-deref.
This patch fixes this issue by having struct ovpn_socket hold
a reference directly to the sk member while also increasing
its reference counter.
Patch 3: in case of errors along the TCP RX path (softirq)
we want to immediately delete the peer, but this operation may
sleep. With this patch we move the peer deletion to a scheduled
worker.
Patch 4 and 5 are instead fixing minor issues in the ovpn
kselftests.
* tag 'ovpn-net-20250603' of https://github.com/OpenVPN/ovpn-net-next:
selftest/net/ovpn: fix missing file
selftest/net/ovpn: fix TCP socket creation
ovpn: avoid sleep in atomic context in TCP RX error path
ovpn: ensure sk is still valid during cleanup
ovpn: properly deconfigure UDP-tunnel
====================
Daniele Palmas [Tue, 3 Jun 2025 09:12:04 +0000 (11:12 +0200)]
net: wwan: mhi_wwan_mbim: use correct mux_id for multiplexing
Recent Qualcomm chipsets like SDX72/75 require MBIM sessionId mapping
to muxId in the range (0x70-0x8F) for the PCIe tethered use.
This has been partially addressed by the referenced commit, mapping
the default data call to muxId = 112, but the multiplexed data calls
scenario was not properly considered, mapping sessionId = 1 to muxId
1, while it should have been 113.
Fix this by moving the session_id assignment logic to mhi_mbim_newlink,
in order to map sessionId = n to muxId = n + WDS_BIND_MUX_DATA_PORT_MUX_ID.
Fixes: 65bc58c3dcad ("net: wwan: mhi: make default data link id configurable") Signed-off-by: Daniele Palmas <dnlplm@gmail.com> Reviewed-by: Loic Poulain <loic.poulain@oss.qualcomm.com> Link: https://patch.msgid.link/20250603091204.2802840-1-dnlplm@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Lachlan Hodges [Tue, 3 Jun 2025 05:35:38 +0000 (15:35 +1000)]
wifi: cfg80211/mac80211: correctly parse S1G beacon optional elements
S1G beacons are not traditional beacons but a type of extension frame.
Extension frames contain the frame control and duration fields, followed
by zero or more optional fields before the frame body. These optional
fields are distinct from the variable length elements.
The presence of optional fields is indicated in the frame control field.
To correctly locate the elements offset, the frame control must be parsed
to identify which optional fields are present. Currently, mac80211 parses
S1G beacons based on fixed assumptions about the frame layout, without
inspecting the frame control field. This can result in incorrect offsets
to the "variable" portion of the frame.
Properly parse S1G beacon frames by using the field lengths defined in
IEEE 802.11-2024, section 9.3.4.3, ensuring that the elements offset is
calculated accurately.
Fixes: 9eaffe5078ca ("cfg80211: convert S1G beacon to scan results") Fixes: cd418ba63f0c ("mac80211: convert S1G beacon to scan results") Signed-off-by: Lachlan Hodges <lachlan.hodges@morsemicro.com> Link: https://patch.msgid.link/20250603053538.468562-1-lachlan.hodges@morsemicro.com Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Benjamin Berg [Thu, 5 Jun 2025 05:03:24 +0000 (07:03 +0200)]
um: fix unused variable warning
The code was updated to access the PID of the userspace stub process in
a different way, making the local cpu variable obsolete. Remove it.
Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202506050008.AwXLNxQX-lkp@intel.com/ Fixes: 406d17c6c370 ("um: Implement kernel side of SECCOMP based process handling") Signed-off-by: Benjamin Berg <benjamin.berg@intel.com> Link: https://patch.msgid.link/20250605050325.1077208-1-benjamin@sipsolutions.net Signed-off-by: Johannes Berg <johannes.berg@intel.com>
RGMII ports on BCM63xx were not really working, especially with PHYs
that support EEE and are capable of configuring their own RGMII delays.
So let's make them work, and fix additional minor rgmii related issues
found while working on it.
With a BCM96328BU-P300:
Before:
[ 3.580000] b53-switch 10700000.switch GbE3 (uninitialized): validation of rgmii with support 0000000,00000000,00000000,000062ff and advertisement 0000000,00000000,00000000,000062ff failed: -EINVAL
[ 3.600000] b53-switch 10700000.switch GbE3 (uninitialized): failed to connect to PHY: -EINVAL
[ 3.610000] b53-switch 10700000.switch GbE3 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 4
[ 3.620000] b53-switch 10700000.switch GbE1 (uninitialized): validation of rgmii with support 0000000,00000000,00000000,000062ff and advertisement 0000000,00000000,00000000,000062ff failed: -EINVAL
[ 3.640000] b53-switch 10700000.switch GbE1 (uninitialized): failed to connect to PHY: -EINVAL
[ 3.650000] b53-switch 10700000.switch GbE1 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 5
[ 3.660000] b53-switch 10700000.switch GbE4 (uninitialized): validation of rgmii with support 0000000,00000000,00000000,000062ff and advertisement 0000000,00000000,00000000,000062ff failed: -EINVAL
[ 3.680000] b53-switch 10700000.switch GbE4 (uninitialized): failed to connect to PHY: -EINVAL
[ 3.690000] b53-switch 10700000.switch GbE4 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 6
[ 3.700000] b53-switch 10700000.switch GbE5 (uninitialized): validation of rgmii with support 0000000,00000000,00000000,000062ff and advertisement 0000000,00000000,00000000,000062ff failed: -EINVAL
[ 3.720000] b53-switch 10700000.switch GbE5 (uninitialized): failed to connect to PHY: -EINVAL
[ 3.730000] b53-switch 10700000.switch GbE5 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 7
Jonas Gorski [Mon, 2 Jun 2025 19:39:53 +0000 (21:39 +0200)]
net: dsa: b53: do not touch DLL_IQQD on bcm53115
According to OpenMDK, bit 2 of the RGMII register has a different
meaning for BCM53115 [1]:
"DLL_IQQD 1: In the IDDQ mode, power is down0: Normal function
mode"
Configuring RGMII delay works without setting this bit, so let's keep it
at the default. For other chips, we always set it, so not clearing it
is not an issue.
One would assume BCM53118 works the same, but OpenMDK is not quite sure
what this bit actually means [2]:
"BYPASS_IMP_2NS_DEL #1: In the IDDQ mode, power is down#0: Normal
function mode1: Bypass dll65_2ns_del IP0: Use
dll65_2ns_del IP"
Jonas Gorski [Mon, 2 Jun 2025 19:39:52 +0000 (21:39 +0200)]
net: dsa: b53: allow RGMII for bcm63xx RGMII ports
Add RGMII to supported interfaces for BCM63xx RGMII ports so they can be
actually used in RGMII mode.
Without this, phylink will fail to configure them:
[ 3.580000] b53-switch 10700000.switch GbE3 (uninitialized): validation of rgmii with support 0000000,00000000,00000000,000062ff and advertisement 0000000,00000000,00000000,000062ff failed: -EINVAL
[ 3.600000] b53-switch 10700000.switch GbE3 (uninitialized): failed to connect to PHY: -EINVAL
[ 3.610000] b53-switch 10700000.switch GbE3 (uninitialized): error -22 setting up PHY for tree 0, switch 0, port 4
Fixes: ce3bf94871f7 ("net: dsa: b53: add support for BCM63xx RGMIIs") Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com> Link: https://patch.msgid.link/20250602193953.1010487-5-jonas.gorski@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Jonas Gorski [Mon, 2 Jun 2025 19:39:49 +0000 (21:39 +0200)]
net: dsa: b53: do not enable EEE on bcm63xx
BCM63xx internal switches do not support EEE, but provide multiple RGMII
ports where external PHYs may be connected. If one of these PHYs are EEE
capable, we may try to enable EEE for the MACs, which then hangs the
system on access of the (non-existent) EEE registers.
Fix this by checking if the switch actually supports EEE before
attempting to configure it.
Fixes: 22256b0afb12 ("net: dsa: b53: Move EEE functions to b53") Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com> Tested-by: Álvaro Fernández Rojas <noltari@gmail.com> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com> Link: https://patch.msgid.link/20250602193953.1010487-2-jonas.gorski@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Meghana Malladi [Tue, 3 Jun 2025 05:29:04 +0000 (10:59 +0530)]
net: ti: icssg-prueth: Fix swapped TX stats for MII interfaces.
In MII mode, Tx lines are swapped for port0 and port1, which means
Tx port0 receives data from PRU1 and the Tx port1 receives data from
PRU0. This is an expected hardware behavior and reading the Tx stats
needs to be handled accordingly in the driver. Update the driver to
read Tx stats from the PRU1 for port0 and PRU0 for port1.
Florian Westphal [Fri, 30 May 2025 10:34:03 +0000 (12:34 +0200)]
selftests: netfilter: nft_nat.sh: add test for reverse clash with nat
This will fail without the previous bug fix because we erronously
believe that the clashing entry went way.
However, the clash exists in the opposite direction due to an
existing nat mapping:
PASS: IP statless for ns2-LgTIuS
ERROR: failed to test udp ns1-x4iyOW to ns2-LgTIuS with dnat rule step 2, result: ""
This is partially adapted from test instructions from the below
ubuntu tracker.
Florian Westphal [Fri, 30 May 2025 10:34:02 +0000 (12:34 +0200)]
netfilter: nf_nat: also check reverse tuple to obtain clashing entry
The logic added in the blamed commit was supposed to only omit nat source
port allocation if neither the existing nor the new entry are subject to
NAT.
However, its not enough to lookup the conntrack based on the proposed
tuple, we must also check the reverse direction.
Otherwise there are esoteric cases where the collision is in the reverse
direction because that colliding connection has a port rewrite, but the
new entry doesn't. In this case, we only check the new entry and then
erronously conclude that no clash exists anymore.
The existing (udp) tuple is:
a:p -> b:P, with nat translation to s:P, i.e. pure daddr rewrite,
reverse tuple in conntrack table is s:P -> a:p.
When another UDP packet is sent directly to s, i.e. a:p->s:P, this is
correctly detected as a colliding entry: tuple is taken by existing reply
tuple in reverse direction.
But the colliding conntrack is only searched for with unreversed
direction, and we can't find such entry matching a:p->s:P.
The incorrect conclusion is that the clashing entry has timed out and
that no port address translation is required.
Such conntrack will then be discarded at nf_confirm time because the
proposed reverse direction clashes with an existing mapping in the
conntrack table.
Search for the reverse tuple too, this will then check the NAT bits of
the colliding entry and triggers port reallocation.
Followp patch extends nft_nat.sh selftest to cover this scenario.
The IPS_SEQ_ADJUST change is also a bug fix:
Instead of checking for SEQ_ADJ this tested for SEEN_REPLY and ASSURED
by accident -- _BIT is only for use with the test_bit() API.
This bug has little consequence in practice, because the sequence number
adjustments are only useful for TCP which doesn't support clash resolution.
The existing test case (conntrack_reverse_clash.sh) exercise a race
condition path (parallel conntrack creation on different CPUs), so
the colliding entries have neither SEEN_REPLY nor ASSURED set.
Thanks to Yafang Shao and Shaun Brady for an initial investigation
of this bug.
Fixes: d8f84a9bc7c4 ("netfilter: nf_nat: don't try nat source port reallocation for reverse dir clash") Closes: https://bugzilla.netfilter.org/show_bug.cgi?id=1795 Reported-by: Yafang Shao <laoar.shao@gmail.com> Reported-by: Shaun Brady <brady.1345@gmail.com> Signed-off-by: Florian Westphal <fw@strlen.de> Tested-by: Yafang Shao <laoar.shao@gmail.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Fri, 23 May 2025 12:20:46 +0000 (14:20 +0200)]
selftests: netfilter: nft_concat_range.sh: add datapath check for map fill bug
commit 0935ee6032df ("selftests: netfilter: add test case for recent mismatch bug")
added a regression check for incorrect initial fill of the result map
that was fixed with 791a615b7ad2 ("netfilter: nf_set_pipapo: fix initial map fill").
The test used 'nft get element', i.e., control plane checks for
match/nomatch results.
The control plane however doesn't use avx2 version, so we need to
send+match packets.
As the additional packet match/nomatch is slow, don't do this for
every element added/removed: add and use maybe_send_(no)match
helpers and use them.
Florian Westphal [Fri, 23 May 2025 12:20:45 +0000 (14:20 +0200)]
selftests: netfilter: nft_concat_range.sh: prefer per element counters for testing
The selftest uses following rule:
... @test counter name "test"
Then sends a packet, then checks if the named counter did increment or
not.
This is fine for the 'no-match' test case: If anything matches the
counter increments and the test fails as expected.
But for the 'should match' test cases this isn't optimal.
Consider buggy matching, where the packet matches entry x, but it
should have matched entry y.
In that case the test would erronously pass.
Rework the selftest to use per-element counters to avoid this.
After sending packet that should have matched entry x, query the
relevant element via 'nft reset element' and check that its counter
had incremented.
The 'nomatch' case isn't altered, no entry should match so the named
counter must be 0, changing it to the per-element counter would then
pass if another entry matches.
The downside of this change is a slight increase in test run-time by
a few seconds.
Linus Torvalds [Thu, 5 Jun 2025 04:18:37 +0000 (21:18 -0700)]
Merge tag 'rust-6.16' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux
Pull Rust updates from Miguel Ojeda:
"Toolchain and infrastructure:
- KUnit '#[test]'s:
- Support KUnit-mapped 'assert!' macros.
The support that landed last cycle was very basic, and the
'assert!' macros panicked since they were the standard library
ones. Now, they are mapped to the KUnit ones in a similar way to
how is done for doctests, reusing the infrastructure there.
# my_first_test: ASSERTION FAILED at rust/kernel/lib.rs:251
Expected 42 == 43 to be true, but is false
# my_first_test.speed: normal
not ok 1 my_first_test
- Support tests with checked 'Result' return types.
The return value of test functions that return a 'Result' will
be checked, thus one can now easily catch errors when e.g. using
the '?' operator in tests.
With this, a failing test like:
#[test]
fn my_test() -> Result {
f()?;
Ok(())
}
will report:
# my_test: ASSERTION FAILED at rust/kernel/lib.rs:321
Expected is_test_result_ok(my_test()) to be true, but is false
# my_test.speed: normal
not ok 1 my_test
- Add 'kunit_tests' to the prelude.
- Clarify the remaining language unstable features in use.
- Compile 'core' with edition 2024 for Rust >= 1.87.
- Workaround 'bindgen' issue with forward references to 'enum' types.
- objtool: relax slice condition to cover more 'noreturn' functions.
- Use absolute paths in macros referencing 'core' and 'kernel'
crates.
- Skip '-mno-fdpic' flag for bindgen in GCC 32-bit arm builds.
- Clean some 'doc_markdown' lint hits -- we may enable it later on.
'kernel' crate:
- 'alloc' module:
- 'Box': support for type coercion, e.g. 'Box<T>' to 'Box<dyn U>'
if 'T' implements 'U'.
- 'Vec': implement new methods (prerequisites for nova-core and
binder): 'truncate', 'resize', 'clear', 'pop',
'push_within_capacity' (with new error type 'PushError'),
'drain_all', 'retain', 'remove' (with new error type
'RemoveError'), insert_within_capacity' (with new error type
'InsertError').
In addition, simplify 'push' using 'spare_capacity_mut', split
'set_len' into 'inc_len' and 'dec_len', add type invariant 'len
<= capacity' and simplify 'truncate' using 'dec_len'.
- 'time' module:
- Morph the Rust hrtimer subsystem into the Rust timekeeping
subsystem, covering delay, sleep, timekeeping, timers. This new
subsystem has all the relevant timekeeping C maintainers listed
in the entry.
- Replace 'Ktime' with 'Delta' and 'Instant' types to represent a
duration of time and a point in time.
- Temporarily add 'Ktime' to 'hrtimer' module to allow 'hrtimer'
to delay converting to 'Instant' and 'Delta'.
- 'xarray' module:
- Add a Rust abstraction for the 'xarray' data structure. This
abstraction allows Rust code to leverage the 'xarray' to store
types that implement 'ForeignOwnable'. This support is a
dependency for memory backing feature of the Rust null block
driver, which is waiting to be merged.
- Set up an entry in 'MAINTAINERS' for the XArray Rust support.
Patches will go to the new Rust XArray tree and then via the
Rust subsystem tree for now.
- Allow 'ForeignOwnable' to carry information about the pointed-to
type. This helps asserting alignment requirements for the
pointer passed to the foreign language.
- 'container_of!': retain pointer mut-ness and add a compile-time
check of the type of the first parameter ('$field_ptr').
- Support optional message in 'static_assert!'.
- Add C FFI types (e.g. 'c_int') to the prelude.
- 'str' module: simplify KUnit tests 'format!' macro, convert
'rusttest' tests into KUnit, take advantage of the '-> Result'
support in KUnit '#[test]'s.
- 'list' module: add examples for 'List', fix path of
'assert_pinned!' (so far unused macro rule).
- 'workqueue' module: remove 'HasWork::OFFSET'.
- 'page' module: add 'inline' attribute.
'macros' crate:
- 'module' macro: place 'cleanup_module()' in '.exit.text' section.
'pin-init' crate:
- Add 'Wrapper<T>' trait for creating pin-initializers for wrapper
structs with a structurally pinned value such as 'UnsafeCell<T>' or
'MaybeUninit<T>'.
- Add 'MaybeZeroable' derive macro to try to derive 'Zeroable', but
not error if not all fields implement it. This is needed to derive
'Zeroable' for all bindgen-generated structs.
- Add 'unsafe fn cast_[pin_]init()' functions to unsafely change the
initialized type of an initializer. These are utilized by the
'Wrapper<T>' implementations.
- Add support for visibility in 'Zeroable' derive macro.
- Add support for 'union's in 'Zeroable' derive macro.
- Upstream dev news: streamline CI, fix some bugs. Add new workflows
to check if the user-space version and the one in the kernel tree
have diverged. Use the issues tab [1] to track them, which should
help folks report and diagnose issues w.r.t. 'pin-init' better.
Linus Torvalds [Thu, 5 Jun 2025 02:23:37 +0000 (19:23 -0700)]
Merge tag '6.16-rc-ksmbd-server-fixes' of git://git.samba.org/ksmbd
Pull smb server updates from Steve French:
"Four smb3 server fixes:
- Fix for special character handling when mounting with "posix"
- Fix for mounts from Mac for fs that don't provide unique inode
numbers
- Two cleanup patches (e.g. for crypto calls)"
* tag '6.16-rc-ksmbd-server-fixes' of git://git.samba.org/ksmbd:
ksmbd: allow a filename to contain special characters on SMB3.1.1 posix extension
ksmbd: provide zero as a unique ID to the Mac client
ksmbd: remove unnecessary softdep on crc32
ksmbd: use SHA-256 library API instead of crypto_shash API
Linus Torvalds [Thu, 5 Jun 2025 02:14:24 +0000 (19:14 -0700)]
Merge tag 'bcachefs-2025-06-04' of git://evilpiepirate.org/bcachefs
Pull more bcachefs updates from Kent Overstreet:
"More bcachefs updates:
- More stack usage improvements (~600 bytes)
- Define CLASS()es for some commonly used types, and convert most
rcu_read_lock() uses to the new lock guards
- New introspection:
- Superblock error counters are now available in sysfs:
previously, they were only visible with 'show-super', which
doesn't provide a live view
- New tracepoint, error_throw(), which is called any time we
return an error and start to unwind
- Repair
- check_fix_ptrs() can now repair btree node roots
- We can now repair when we've somehow ended up with the journal
using a superblock bucket
- Revert some leftovers from the aborted directory i_size feature,
and add repair code: some userspace programs (e.g. sshfs) were
getting confused
It seems in 6.15 there's a bug where i_nlink on the vfs inode has been
getting incorrectly set to 0, with some unfortunate results;
list_journal analysis showed bch2_inode_rm() being called (by
bch2_evict_inode()) when it clearly should not have been.
- bch2_inode_rm() now runs "should we be deleting this inode?" checks
that were previously only run when deleting unlinked inodes in
recovery
- check_subvol() was treating a dangling subvol (pointing to a
missing root inode) like a dangling dirent, and deleting it. This
was the really unfortunate one: check_subvol() will now recreate
the root inode if necessary
This took longer to debug than it should have, and we lost several
filesystems unnecessarily, because users have been ignoring the
release notes and blindly running 'fsck -y'. Debugging required
reconstructing what happened through analyzing the journal, when
ideally someone would have noticed 'hey, fsck is asking me if I want
to repair this: it usually doesn't, maybe I should run this in dry run
mode and check what's going on?'
As a reminder, fsck errors are being marked as autofix once we've
verified, in real world usage, that they're working correctly; blindly
running 'fsck -y' on an experimental filesystem is playing with fire
Up to this incident we've had an excellent track record of not losing
data, so let's try to learn from this one
This is a community effort, I wouldn't be able to get this done
without the help of all the people QAing and providing excellent bug
reports and feedback based on real world usage. But please don't
ignore advice and expect me to pick up the pieces
If an error isn't marked as autofix, and it /is/ happening in the
wild, that's also something I need to know about so we can check it
out and add it to the autofix list if repair looks good. I haven't
been getting those reports, and I should be; since we don't have any
sort of telemetry yet I am absolutely dependent on user reports
Now I'll be spending the weekend working on new repair code to see if
I can get a filesystem back for a user who didn't have backups"
* tag 'bcachefs-2025-06-04' of git://evilpiepirate.org/bcachefs: (69 commits)
bcachefs: add cond_resched() to handle_overwrites()
bcachefs: Make journal read log message a bit quieter
bcachefs: Fix subvol to missing root repair
bcachefs: Run may_delete_deleted_inode() checks in bch2_inode_rm()
bcachefs: delete dead code from may_delete_deleted_inode()
bcachefs: Add flags to subvolume_to_text()
bcachefs: Fix oops in btree_node_seq_matches()
bcachefs: Fix dirent_casefold_mismatch repair
bcachefs: Fix bch2_fsck_rename_dirent() for casefold
bcachefs: Redo bch2_dirent_init_name()
bcachefs: Fix -Wc23-extensions in bch2_check_dirents()
bcachefs: Run check_dirents second time if required
bcachefs: Run snapshot deletion out of system_long_wq
bcachefs: Make check_key_has_snapshot safer
bcachefs: BCH_RECOVERY_PASS_NO_RATELIMIT
bcachefs: bch2_require_recovery_pass()
bcachefs: bch_err_throw()
bcachefs: Repair code for directory i_size
bcachefs: Kill un-reverted directory i_size code
bcachefs: Delete redundant fsck_err()
...
Kent Overstreet [Tue, 3 Jun 2025 13:31:58 +0000 (09:31 -0400)]
bcachefs: Make journal read log message a bit quieter
Users seem to be assuming that the 'dropped unflushed entries' message
at the end of journal read indicates some sort of problem, when it does
not - we expect there to be entries in the journal that weren't
commited, it's purely informational so that we can correlate journal
sequence numbers elsewhere when debugging.
Shorten the log message a bit to hopefully make this clearer.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Kent Overstreet [Mon, 2 Jun 2025 23:48:27 +0000 (19:48 -0400)]
bcachefs: Fix subvol to missing root repair
We had a bug where the root inode of a subvolume was erronously deleted:
bch2_evict_inode() called bch2_inode_rm(), meaning the VFS inode's
i_nlink was somehow set to 0 when it shouldn't have - the inode in the
btree indicated it clearly was not unlinked.
This has been addressed with additional safety checks in
bch2_inode_rm() - pulling in the safety checks we already were doing
when deleting unlinked inodes in recovery - but the really disastrous
bug was in check_subvols(), which on finding a dangling subvol (subvol
with a missing root inode) would delete the subvolume.
I assume this bug dates from early check_directory_structure() code,
which originally handled subvolumes and normal paths - the idea being
that still live contents of the subvolume would get reattached
somewhere.
But that's incorrect, and disastrously so; deleting a subvolume triggers
deleting the snapshot ID it points to, deleting the entire contents.
The correct way to repair is to recreate the root inode if it's missing;
then any contents will get reattached under that subvolume's lost+found.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Kent Overstreet [Mon, 2 Jun 2025 13:26:20 +0000 (09:26 -0400)]
bcachefs: Fix oops in btree_node_seq_matches()
btree_update_nodes_written() needs to wait on in-flight writes to old
nodes before marking them as freed. But it has no reason to pin those
old nodes in memory, so some trickyness ensues.
The update we're completing deleted references to those nodes from the
btree, so we know if they've been evicted they can't be pulled back in.
We just have to check if the nodes we have pointers to are still those
old nodes, and haven't been reused.
To do that we check the node's "sequence number" (actually a random 64
bit cookie), but that lives in the node's data buffer. 'struct btree'
can't be freed until filesystem shutdown (as they're quite small), but
the data buffers can be freed or swapped around.
Commit 1f88c3567495, which was fixing a kmsan warning, assumed that we
could safely do this locklessly with just a READ_ONCE() - if we've got a
non-null ptr it would be safe to read from.
But that's not true if the data buffer is a vmalloc allocation, so we
need to restore the locking that commit deleted (or alternatively RCU
free those data buffers, but there's no other reason for that).
Fixes: 1f88c3567495 ("bcachefs: Fix a KMSAN splat in btree_update_nodes_written()") Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
projects / thirdparty / linux.git / log
