net: dsa: microchip: Don't embed struct phy_device to maintain the port state
The KSZ9477 maintains the SGMII port's state for speed, duplex and link
status to be able to fixup the accesses to its internal older version of
the Designware XPCS. However, it does so by embedding a full instance of
struct phy_device, only to use the 'speed', 'link' and 'duplex' fields.
This is also only used for the SGMII port, it's otherwise unused for all
other regular ports.
====================
net: phy: realtek: pair order and polarity
The RTL8224 PHY gives the manufacturer some flexbility with the pair
order and polarity to ease the wiring on the PCB. Then the correct pair
order and pair polarity must be provided to the PHY to function
properly. This series adds the support to configure the pair order and
the pair polarity to the Realtek PHY driver.
====================
Damien Dejean [Wed, 18 Mar 2026 21:55:01 +0000 (22:55 +0100)]
net: phy: realtek: add RTL8224 polarity support
The RTL8224 has a register to configure the polarity of every pair of
each port. It provides device designers more flexbility when wiring the
chip.
Unfortunately, the register is left in an unknown state after a reset.
Thus on devices where the bootloader don't initialize it, the driver has
to do it to detect and use a link.
The MDI polarity swap can be set in the device tree using the property
enet-phy-pair-polarity. The u32 value is a bitfield where bit[0..3]
control the polarity of pairs A..D.
Add the property enet-phy-pair-polarity to describe the polarity of the
PHY pairs. To ease PCB designs some manufacturers allow to wire the
pairs with a reverse polarity and provide a way to configure it.
The property 'enet-phy-pair-polarity' sets the polarity of each pair.
Bit 0 to 3 configure the polarity or pairs A to D, if set to 1 the
polarity is reversed for this pair.
Damien Dejean [Wed, 18 Mar 2026 21:54:59 +0000 (22:54 +0100)]
net: phy: realtek: add RTL8224 pair order support
The RTL8224 has a register to configure a pair swap (from ABCD order to
DCBA) providing PCB designers more flexbility when wiring the chip. The
swap parameter has to be set correctly for each of the 4 ports before
the chip can detect a link.
After a reset, this register is (unfortunately) left in a random state,
thus it has to be initialized. On most of the devices the bootloader
does it once for all and we can rely on the value set, on some other it
is not and the kernel has to do it.
The MDI pair swap can be set in the device tree using the property
enet-phy-pair-order. The property is set to 0 to keep the default order
(ABCD), or 1 to reverse the pairs (DCBA).
Add property enet-phy-pair-order to the device tree bindings to define
the pair order of the PHY. To simplify PCB design some manufacturers
allow to wire the pairs in a reverse order, and change the order in
software.
The property can be set to 0 to force the normal pair order (ABCD), or 1
to force the reverse pair order (DCBA).
One of the reasons is because bitset.h s lacking definitions for
nlattr, netlink_ext_ack, ETH_GSTRING_LEN, and types such as u32, bool,
etc.
Make bitset.h standalone by including <linux/ethtool.h> for
ETH_GSTRING_LEN, and <linux/netlink.h> for nlattr, netlink_ext_ack and
the rest.
While at it, take a pass on ethnl sources to re-order the local
includes :
- put them after the global includes
- add a newline between global and local includes
- alpha-sort the local includes
One notable exception is the cmis.h include, that needs definitions from
module_fw.h. Keep them in this order for now.
Justin Chen [Thu, 19 Mar 2026 23:48:13 +0000 (16:48 -0700)]
net: bcmasp: fix double disable of clk
Switch to devm_clk_get_optional() so we can manage the clock ourselves.
We dynamically control the clocks depending on the state of the interface
for power savings. The default state is clock disabled, so unbinding the
driver causes a double disable.
Fixes: 490cb412007d ("net: bcmasp: Add support for ASP2.0 Ethernet controller") Signed-off-by: Justin Chen <justin.chen@broadcom.com> Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com> Link: https://patch.msgid.link/20260319234813.1937315-3-justin.chen@broadcom.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Qi Tang [Wed, 18 Mar 2026 06:48:47 +0000 (14:48 +0800)]
net/smc: fix double-free of smc_spd_priv when tee() duplicates splice pipe buffer
smc_rx_splice() allocates one smc_spd_priv per pipe_buffer and stores
the pointer in pipe_buffer.private. The pipe_buf_operations for these
buffers used .get = generic_pipe_buf_get, which only increments the page
reference count when tee(2) duplicates a pipe buffer. The smc_spd_priv
pointer itself was not handled, so after tee() both the original and the
cloned pipe_buffer share the same smc_spd_priv *.
When both pipes are subsequently released, smc_rx_pipe_buf_release() is
called twice against the same object:
KASAN reports a slab-use-after-free in smc_rx_pipe_buf_release(), which
then escalates to a NULL-pointer dereference and kernel panic via
smc_rx_update_consumer() when it chases the freed priv->smc pointer:
BUG: KASAN: slab-use-after-free in smc_rx_pipe_buf_release+0x78/0x2a0
Read of size 8 at addr ffff888004a45740 by task smc_splice_tee_/74
Call Trace:
<TASK>
dump_stack_lvl+0x53/0x70
print_report+0xce/0x650
kasan_report+0xc6/0x100
smc_rx_pipe_buf_release+0x78/0x2a0
free_pipe_info+0xd4/0x130
pipe_release+0x142/0x160
__fput+0x1c6/0x490
__x64_sys_close+0x4f/0x90
do_syscall_64+0xa6/0x1a0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
</TASK>
Beyond the memory-safety problem, duplicating an SMC splice buffer is
semantically questionable: smc_rx_update_cons() would advance the
consumer cursor twice for the same data, corrupting receive-window
accounting. A refcount on smc_spd_priv could fix the double-free, but
the cursor-accounting issue would still need to be addressed separately.
The .get callback is invoked by both tee(2) and splice_pipe_to_pipe()
for partial transfers; both will now return -EFAULT. Users who need
to duplicate SMC socket data must use a copy-based read path.
net: lan743x: fix SGMII detection on PCI1xxxx B0+ during warm reset
A warm reset on boards using an EEPROM-only strap configuration (where
no MAC address is set in the image) can cause the driver to incorrectly
revert to RGMII mode. This occurs because the ENET_CONFIG_LOAD_STARTED
bit may not persist or behave as expected.
Update pci11x1x_strap_get_status() to use revision-specific validation:
- For PCI11x1x A0: Continue using the legacy check (config load started
or reset protection) to validate the SGMII strap.
- For PCI11x1x B0 and later: Use the newly available
STRAP_READ_USE_SGMII_EN_ bit in the upper strap register to validate
the lower SGMII_EN bit.
This ensures the SGMII interface is correctly identified even after a
warm reboot.
validate_set() accepted OVS_KEY_ATTR_MPLS as variable-sized payload for
SET/SET_MASKED actions. In action handling, OVS expects fixed-size
MPLS key data (struct ovs_key_mpls).
Use the already normalized key_len (masked case included) and reject
non-matching MPLS action key sizes.
Reject invalid MPLS action payload lengths early.
Fixes: fbdcdd78da7c ("Change in Openvswitch to support MPLS label depth of 3 in ingress direction") Reported-by: Yifan Wu <yifanwucs@gmail.com> Reported-by: Juefei Pu <tomapufckgml@gmail.com> Tested-by: Ao Zhou <n05ec@lzu.edu.cn> Co-developed-by: Yuan Tan <tanyuan98@outlook.com> Signed-off-by: Yuan Tan <tanyuan98@outlook.com> Suggested-by: Xin Liu <bird@lzu.edu.cn> Signed-off-by: Yang Yang <n05ec@lzu.edu.cn> Reviewed-by: Ilya Maximets <i.maximets@ovn.org> Link: https://patch.msgid.link/20260319080228.3423307-1-n05ec@lzu.edu.cn Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Yang Yang [Thu, 19 Mar 2026 07:42:41 +0000 (07:42 +0000)]
openvswitch: defer tunnel netdev_put to RCU release
ovs_netdev_tunnel_destroy() may run after NETDEV_UNREGISTER already
detached the device. Dropping the netdev reference in destroy can race
with concurrent readers that still observe vport->dev.
Do not release vport->dev in ovs_netdev_tunnel_destroy(). Instead, let
vport_netdev_free() drop the reference from the RCU callback, matching
the non-tunnel destroy path and avoiding additional synchronization
under RTNL.
Fixes: a9020fde67a6 ("openvswitch: Move tunnel destroy function to oppenvswitch module.") Reported-by: Yifan Wu <yifanwucs@gmail.com> Reported-by: Juefei Pu <tomapufckgml@gmail.com> Tested-by: Ao Zhou <n05ec@lzu.edu.cn> Co-developed-by: Yuan Tan <tanyuan98@outlook.com> Signed-off-by: Yuan Tan <tanyuan98@outlook.com> Suggested-by: Xin Liu <bird@lzu.edu.cn> Signed-off-by: Yang Yang <n05ec@lzu.edu.cn> Reviewed-by: Ilya Maximets <i.maximets@ovn.org> Link: https://patch.msgid.link/20260319074241.3405262-1-n05ec@lzu.edu.cn Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================
selftests/vsock: support nested VM runner for vmtest.sh
This series fixes a few issues trying to launch vmtest.sh in a nested VM
environment and were discovered when trying to prepare the tests for
netdev CI/CD.
When taken together these patches make vmtest.sh work both on bare metal
and in nested VMs, regardless of the outer VM's user, coincidental path
overlaps, or filesystem settings.
====================
Bobby Eshleman [Tue, 17 Mar 2026 22:09:36 +0000 (15:09 -0700)]
selftests/vsock: fix vsock_test path shadowing in nested VMs
The /root mount introduced for nested VM support shadows any host paths
under /root. This breaks systems where the outer VM runs as root and the
vsock_test binary path is something like:
Bobby Eshleman [Tue, 17 Mar 2026 22:09:35 +0000 (15:09 -0700)]
selftests/vsock: fix vmtest.sh for read-only nested VM runners
When running vmtest.sh inside a nested VM, there occurs a problem
with stacking two sets of virtiofs/overlay layers (the first set from
the outer VM and the second set from the inner VM). The virtme init
scripts (sshd, udhcpd, etc...) fail to execute basic programs (e.g.,
/bin/cat) and load library dependencies (e.g., libpam) due to ESTALE.
This only occurs when both layers (outer and inner) use virtiofs. Work
around this by using 9p in the inner VM via --force-9p.
Additionally, when the outer VM is read-only, the inner VM's attempt at
populating SSH keys to the root filesystem fails:
virtme-ng-init: mkdir: cannot create directory '/root/.cache': Read-only file system
Work around this by creating a temporary home directory with generated
SSH keys and passing it through to the guest as /root via --rwdir.
Disable strict host key checking in vm_ssh() since the VM will be seen
as a new host each run.
The --rw arg had to be removed to prevent a vng complaint about overlay
(in combination with the other parameters). The guest doesn't really
need write access anyway, so this was probably overly permissive to
begin with.
Breno Leitao [Wed, 18 Mar 2026 12:22:48 +0000 (05:22 -0700)]
bonding: remove bonding_priv.h
bonding_priv.h only defined DRV_NAME and DRV_DESCRIPTION, but caused
unnecessary recompilation: it included <generated/utsrelease.h> to
define bond_version, which is used solely in bond_procfs.c. With
CONFIG_LOCALVERSION_AUTO=y, utsrelease.h is regenerated on every git
commit, so any git operation triggered recompilation of bond_main.c
which also included bonding_priv.h.
Remove the header entirely, as suggested by Jakub, given the macros on
this file can be integrated into the C files directly.
Lorenzo Bianconi [Tue, 17 Mar 2026 16:40:47 +0000 (17:40 +0100)]
net: airoha: Reset PPE cpu port configuration in airoha_ppe_hw_init()
Before this patch, the PPE cpu port configuration used for a specific GDM
device was set just running ndo_init() callback during the device
initialization. The selected PPE cpu port configuration depends on the QDMA
block assigned to the GDM port. The QDMA block is selected according to
the GDM port LAN/WAN configuration as specified in the commit
'8737d7194d6d ("net: airoha: select QDMA block according LAN/WAN
configuration"). However, the user selected PPE cpu port configuration can
be different with respect to the one hardcoded in the NPU firmware binary.
The hardcoded NPU PPE cpu port configuration is loaded initializing the PPE
engine running the NPU ops ppe_init() callback in airoha_ppe_offload_setup
routine (this is executed at runtime by the netfilter flowtable
infrastructure during flow offloading).
Reset the PPE cpu port configuration in airoha_ppe_hw_init routine in
order to apply the user requested setup according to the device DTS.
Please note this patch is fixing an issue not visible to the user (so we
do not need to backport it) since airoha_eth driver currently supports just
the internal phy available via the MT7530 DSA switch and there are no WAN
interfaces officially supported since PCS/external phy is not merged
mainline yet (it will be posted with following patches).
Daniel Golle [Wed, 18 Mar 2026 03:07:52 +0000 (03:07 +0000)]
net: dsa: mxl862xx: don't read out-of-bounds
The write loop in mxl862xx_api_wrap() computes the word count as
(size + 1) / 2, rounding up for odd-sized structs.
On the last iteration of an odd-sized buffer it reads a full __le16
from data[i], accessing one byte past the end of the caller's struct.
KASAN catches this as a stack-out-of-bounds read during probe (e.g.
from mxl862xx_bridge_config_fwd() because of the odd length of
sizeof(struct mxl862xx_bridge_config) == 49).
The read-back loop already handles this case, it writes only a single
byte when (i * 2 + 1) == size. The write loop lacked the same guard.
In practice the over-read is harmless: the extra stack byte is sent to
the firmware which ignores trailing data beyond the command's declared
payload size.
Apply the same odd-size last-byte handling to the write path: when the
final word contains only one valid byte, send *(u8 *)&data[i] instead
of le16_to_cpu(data[i]). This is endian-safe because data is
__le16-encoded and the low byte is always at the lowest address
regardless of host byte order.
Kevin Hao [Wed, 18 Mar 2026 06:36:59 +0000 (14:36 +0800)]
net: macb: Protect access to net_device::ip_ptr with RCU lock
Access to net_device::ip_ptr and its associated members must be
protected by an RCU lock. Since we are modifying this piece of code,
let's also move it to execute only when WAKE_ARP is enabled.
To minimize the duration of the RCU lock, a local variable is used to
temporarily store the IP address. This change resolves the following
RCU check warning:
WARNING: suspicious RCU usage
7.0.0-rc3-next-20260310-yocto-standard+ #122 Not tainted
-----------------------------
drivers/net/ethernet/cadence/macb_main.c:5944 suspicious rcu_dereference_check() usage!
Kevin Hao [Wed, 18 Mar 2026 06:36:58 +0000 (14:36 +0800)]
net: macb: Move devm_{free,request}_irq() out of spin lock area
The devm_free_irq() and devm_request_irq() functions should not be
executed in an atomic context.
During device suspend, all userspace processes and most kernel threads
are frozen. Additionally, we flush all tx/rx status, disable all macb
interrupts, and halt rx operations. Therefore, it is safe to split the
region protected by bp->lock into two independent sections, allowing
devm_free_irq() and devm_request_irq() to run in a non-atomic context.
This modification resolves the following lockdep warning:
BUG: sleeping function called from invalid context at kernel/locking/mutex.c:591
in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 501, name: rtcwake
preempt_count: 1, expected: 0
RCU nest depth: 1, expected: 0
7 locks held by rtcwake/501:
#0: ffff0008038c3408 (sb_writers#5){.+.+}-{0:0}, at: vfs_write+0xf8/0x368
#1: ffff0008049a5e88 (&of->mutex#2){+.+.}-{4:4}, at: kernfs_fop_write_iter+0xbc/0x1c8
#2: ffff00080098d588 (kn->active#70){.+.+}-{0:0}, at: kernfs_fop_write_iter+0xcc/0x1c8
#3: ffff800081c84888 (system_transition_mutex){+.+.}-{4:4}, at: pm_suspend+0x1ec/0x290
#4: ffff0008009ba0f8 (&dev->mutex){....}-{4:4}, at: device_suspend+0x118/0x4f0
#5: ffff800081d00458 (rcu_read_lock){....}-{1:3}, at: rcu_lock_acquire+0x4/0x48
#6: ffff0008031fb9e0 (&bp->lock){-.-.}-{3:3}, at: macb_suspend+0x144/0x558
irq event stamp: 8682
hardirqs last enabled at (8681): [<ffff8000813c7d7c>] _raw_spin_unlock_irqrestore+0x44/0x88
hardirqs last disabled at (8682): [<ffff8000813c7b58>] _raw_spin_lock_irqsave+0x38/0x98
softirqs last enabled at (7322): [<ffff8000800f1b4c>] handle_softirqs+0x52c/0x588
softirqs last disabled at (7317): [<ffff800080010310>] __do_softirq+0x20/0x2c
CPU: 1 UID: 0 PID: 501 Comm: rtcwake Not tainted 7.0.0-rc3-next-20260310-yocto-standard+ #125 PREEMPT
Hardware name: ZynqMP ZCU102 Rev1.1 (DT)
Call trace:
show_stack+0x24/0x38 (C)
__dump_stack+0x28/0x38
dump_stack_lvl+0x64/0x88
dump_stack+0x18/0x24
__might_resched+0x200/0x218
__might_sleep+0x38/0x98
__mutex_lock_common+0x7c/0x1378
mutex_lock_nested+0x38/0x50
free_irq+0x68/0x2b0
devm_irq_release+0x24/0x38
devres_release+0x40/0x80
devm_free_irq+0x48/0x88
macb_suspend+0x298/0x558
device_suspend+0x218/0x4f0
dpm_suspend+0x244/0x3a0
dpm_suspend_start+0x50/0x78
suspend_devices_and_enter+0xec/0x560
pm_suspend+0x194/0x290
state_store+0x110/0x158
kobj_attr_store+0x1c/0x30
sysfs_kf_write+0xa8/0xd0
kernfs_fop_write_iter+0x11c/0x1c8
vfs_write+0x248/0x368
ksys_write+0x7c/0xf8
__arm64_sys_write+0x28/0x40
invoke_syscall+0x4c/0xe8
el0_svc_common+0x98/0xf0
do_el0_svc+0x28/0x40
el0_svc+0x54/0x1e0
el0t_64_sync_handler+0x84/0x130
el0t_64_sync+0x198/0x1a0
Fixes: 558e35ccfe95 ("net: macb: WoL support for GEM type of Ethernet controller") Cc: stable@vger.kernel.org Reviewed-by: Théo Lebrun <theo.lebrun@bootlin.com> Signed-off-by: Kevin Hao <haokexin@gmail.com> Link: https://patch.msgid.link/20260318-macb-irq-v2-1-f1179768ab24@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Linus Torvalds [Sat, 21 Mar 2026 01:21:27 +0000 (18:21 -0700)]
Merge tag 'drm-fixes-2026-03-21' of https://gitlab.freedesktop.org/drm/kernel
Pull drm fixes from Dave Airlie:
"Regular weekly pull request, from sunny San Diego. Usual suspects in
xe/i915/amdgpu with small fixes all over, then some minor fixes across
a few other drivers. It's probably a bit on the heavy side, but most
of the fix seem well contained,
core:
- drm_dev_unplug UAF fix
pagemap:
- lock handling fix
xe:
- A number of teardown fixes
- Skip over non-leaf PTE for PRL generation
- Fix an uninitialized variable
- Fix a missing runtime PM reference
i915/display:
- Fix #15771: Screen corruption and stuttering on P14s w/ 3K display
- Fix for PSR entry setup frames count on rejected commit
- Fix OOPS if firmware is not loaded and suspend is attempted
- Fix unlikely NULL deref due to DC6 on probe
* tag 'drm-fixes-2026-03-21' of https://gitlab.freedesktop.org/drm/kernel: (40 commits)
drm/xe: Fix missing runtime PM reference in ccs_mode_store
drm/xe: Open-code GGTT MMIO access protection
drm/xe/lrc: Fix uninitialized new_ts when capturing context timestamp
drm/xe/oa: Allow reading after disabling OA stream
drm/xe: Skip over non leaf pte for PRL generation
drm/xe/guc: Ensure CT state transitions via STOP before DISABLED
drm/xe: Trigger queue cleanup if not in wedged mode 2
drm/xe: Forcefully tear down exec queues in GuC submit fini
drm/xe: Always kill exec queues in xe_guc_submit_pause_abort
drm/xe/guc: Fail immediately on GuC load error
drm/i915/gt: Check set_default_submission() before deferencing
drm/radeon: apply state adjust rules to some additional HAINAN vairants
drm/amdgpu: apply state adjust rules to some additional HAINAN vairants
drm/amdgpu: rework how we handle TLB fences
drm/bridge: dw-hdmi-qp: fix multi-channel audio output
drm: Fix use-after-free on framebuffers and property blobs when calling drm_dev_unplug
drm/amdgpu: Fix ISP segfault issue in kernel v7.0
drm/amdgpu/gmc9.0: add bounds checking for cid
drm/amdgpu/mmhub4.2.0: add bounds checking for cid
drm/amdgpu/mmhub4.1.0: add bounds checking for cid
...
ARM: dts: broadcom: bcm2835-rpi: Move non simple-bus nodes to root level
The 'gpu' and 'firmware' nodes are not MMIO devices, so they should not be
under a 'simple-bus'. Additionally, the "raspberrypi,bcm2835-power" node
is part of the firmware, so move it under the 'rpi-firmware' node.
The valid range for the pulses-per-revolution devicetree property is
1..4. The current code checks for a range of 1..5. Fix it.
Declare the variable used to retrieve pulses per revolution from
devicetree as u32 (unsigned) to match the of_property_read_u32() API.
The current code uses a postfix decrement when writing the pulses per
resolution into the chip. This has no effect since the value is evaluated
before it is decremented. Fix it by decrementing before evaluating the
value.
Fixes: 7506ebcd662b ("hwmon: (max6639) : Configure based on DT property") Cc: Naresh Solanki <naresh.solanki@9elements.com> Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Biju Das [Wed, 4 Mar 2026 07:17:28 +0000 (07:17 +0000)]
dt-bindings: i2c: renesas,riic: Document the R9A08G046 support
Document the Renesas RZ/G3L (R9A08G046) RIIC IP. This is compatible with
the version available on Renesas RZ/V2H (R9A09G057).
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com> Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Johan Hovold [Fri, 6 Mar 2026 09:41:14 +0000 (10:41 +0100)]
i2c: cp2615: rename disconnect callback
Rename the driver disconnect function so that it reflects the callback
name for consistency with the rest of the kernel (e.g. makes it easier
to grep for).
Signed-off-by: Johan Hovold <johan@kernel.org> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Arnd Bergmann [Fri, 6 Mar 2026 16:33:09 +0000 (17:33 +0100)]
check-uapi: use dummy libc includes
Based on Thomas Weißschuh's series to kernel headers to no longer require
an installed libc when build testing the uapi headers, the same can now
be done for the scripts/check-uapi.sh script.
The only required change here is to add the usr/dummy-include include
path.
Arnd Bergmann [Fri, 6 Mar 2026 16:33:08 +0000 (17:33 +0100)]
check-uapi: honor ${CROSS_COMPILE} setting
When ${CROSS_COMPILE} is set, but ${CC} is not set, the logic in
check-uapi.sh is different from the top-level Makefile, which defaults
to using the cross gcc. This leads to using the native gcc instead of the
cross version, resulting in unexpected false-positive and false-negative
output.
Use the same logic here that we use in Kbuild for consistency.
Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Thomas Weißschuh <linux@weissschuh.net> Acked-by: Nathan Chancellor <nathan@kernel.org> Link: https://patch.msgid.link/20260306163309.2015837-3-arnd@kernel.org Signed-off-by: Nicolas Schier <nsc@kernel.org>
Arnd Bergmann [Fri, 6 Mar 2026 16:33:07 +0000 (17:33 +0100)]
check-uapi: link into shared objects
While testing ABI changes across all architectures, I found that abidiff
sometimes produces nonsensical output. Further debugging identified
missing or broken libelf support for architecture specific relocations
in ET_REL binaries as the source of the problem[1].
Change the script to no longer produce a relocatable object file but
instead create a shared library for each header. This makes abidiff
work for all of the architectures in upstream linux kernels.
Yi Lai [Fri, 20 Mar 2026 01:09:29 +0000 (09:09 +0800)]
vfio: selftests: Support DMR and GNR-D DSA devices
Currently, the VFIO DSA driver test only supports the SPR DSA device ID.
Attempting to run the test on newer platforms like DMR or GNR-D results
in a "No driver found" error, causing the test to be skipped.
Refactor dsa_probe() to use a switch statement for checking device IDs.
This improves maintainability and makes it easier to add new device IDs
in the future.
Add the following DSA device IDs to the supported list:
PCI_DEVICE_ID_INTEL_DSA_DMR (0x1212)
PCI_DEVICE_ID_INTEL_DSA_GNRD (0x11fb)
* tag 'execve-v7.0-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
fs/tests: exec: Remove bad test vector
binfmt_elf_fdpic: fix AUXV size calculation for ELF_HWCAP3 and ELF_HWCAP4
Alberto Garcia [Mon, 9 Mar 2026 17:39:42 +0000 (18:39 +0100)]
PM: hibernate: return -ENODATA if the snapshot image is not loaded
snapshot_image_loaded() is used in both the in-kernel and the
userspace restore path to ensure that the snapshot image has been
completely loaded. However the latter path returns -EPERM in such
situations, which is meant for cases where the operation is neither
write-only nor ready.
This patch updates the check so the returned error code is -ENODATA in
both cases.
init/Kconfig: Require a release version of clang-22 for CC_HAS_COUNTED_BY_PTR
Commit 150a04d817d8 ("compiler_types.h: Attributes: Add __counted_by_ptr
macro") used Clang 22.0.0 as a minimum supported version for
__counted_by_ptr, which made sense while 22.0.0 was the version of
LLVM's main branch to allow developers to easily test and develop uses
of __counted_by_ptr in their code. However, __counted_by_ptr requires a
change [1] merged towards the end of the 22 development cycle to avoid
errors when applied to void pointers.
In file included from fs/xfs/xfs_attr_inactive.c:18:
fs/xfs/libxfs/xfs_attr.h:59:2: error: 'counted_by' cannot be applied to a pointer with pointee of unknown size because 'void' is an incomplete type
59 | void *buffer __counted_by_ptr(bufsize);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This is disruptive for deployed prerelease clang-22 builds (such as
Android LLVM) or when bisecting between llvmorg-21-init and the fix.
Require a released version of clang-22 (i.e., 21.1.0 or newer) to
enabled __counted_by_ptr to ensure all fixes needed for proper support
are present.
drm/xe: Fix format specifier for printing pointer differences
GCC and clang warn (or error with CONFIG_WERROR=y / W=e) several times
when targeting 32-bit platforms along the lines of
drivers/gpu/drm/xe/xe_lrc.c: In function 'dump_mi_command':
drivers/gpu/drm/xe/xe_lrc.c:1921:40: error: format '%lx' expects argument of type 'long unsigned int', but argument 3 has type 'int' [-Werror=format=]
1921 | drm_printf(p, "LRC[%#5lx] = [%#010x] MI_NOOP (%d dwords)\n",
| ~~~~^
| |
| long unsigned int
| %#5x
1922 | dw - num_noop - start, inst_header, num_noop);
| ~~~~~~~~~~~~~~~~~~~~~
| |
| int
drivers/gpu/drm/xe/xe_lrc.c:1922:7: error: format specifies type 'unsigned long' but the argument has type '__ptrdiff_t' (aka 'int') [-Werror,-Wformat]
1921 | drm_printf(p, "LRC[%#5lx] = [%#010x] MI_NOOP (%d dwords)\n",
| ~~~~~
| %#5tx
1922 | dw - num_noop - start, inst_header, num_noop);
| ^~~~~~~~~~~~~~~~~~~~~
Use the '%tx' specifier for printing pointer differences, which clears
up the warnings for 32-bit platforms while introducing no regressions
for 64-bit platforms.
Remove any GFP_KERNEL arguments found in the new kmalloc_obj-family
helpers. This captures the script used in commit 189f164e573e ("Convert
remaining multi-line kmalloc_obj/flex GFP_KERNEL uses").
Linus Torvalds [Fri, 20 Mar 2026 16:58:56 +0000 (09:58 -0700)]
Merge tag 'io_uring-7.0-20260320' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux
Pull io_uring fixes from Jens Axboe:
- A bit of a work-around for AF_UNIX recv multishot, as the in-kernel
implementation doesn't properly signal EOF. We'll likely rework this
one going forward, but the fix is sufficient for now
- Two fixes for incrementally consumed buffers, for non-pollable files
and for 0 byte reads
* tag 'io_uring-7.0-20260320' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
io_uring/kbuf: propagate BUF_MORE through early buffer commit path
io_uring/kbuf: fix missing BUF_MORE for incremental buffers at EOF
io_uring/poll: fix multishot recv missing EOF on wakeup race
Thomas Weißschuh [Tue, 17 Mar 2026 08:40:36 +0000 (09:40 +0100)]
tools/nolibc: MIPS: fix clobbers of 'lo' and 'hi' registers on different ISAs
All MIPS ISAs before r6 use the 'lo' and 'hi' special registers.
These are clobbered by system calls and need to be marked as such to
avoid miscompilations. Currently nolibc ties the clobbers to the ABI.
But this is wrong and leads to ISA<->ABI combinations which are not
handled correctly, leading to compiler errors or miscompilations.
David Laight [Sun, 8 Mar 2026 11:37:42 +0000 (11:37 +0000)]
selftests/nolibc: Use printf variable field widths and precisions
Now that printf supports '*' for field widths and precisions
then can be used to simplify the test output.
- aligning the "[OK]" strings.
- reporting the expected sprintf() output when there is a mismatch.
David Laight [Sun, 8 Mar 2026 11:37:41 +0000 (11:37 +0000)]
tools/nolibc/printf: Add support for octal output
Octal output isn't often used, but adding it costs very little.
Supporting "%#o" is mildly annoying, it has to add a leading '0' if
there isn't one present. In simple cases this is the same as adding a sign
of '0' - but that adds an extra '0' in a few places.
So you need 3 tests, %o, # and no leading '0' (which can only be checked
after the zero pad for precision).
If all the test are deferred until after zero padding then too many values
are 'live' across the call to _nolibc_u64toa_base() and get spilled to stack.
Hence the check that ignores the 'sign' if it is the same as the first
character of the output string.
David Laight [Sun, 8 Mar 2026 11:37:38 +0000 (11:37 +0000)]
tools/nolibc/printf: Special case 0 and add support for %#x
The output for %#x is almost the same as that for %p, both output in
hexadecimal with a leading "0x".
However for zero %#x should just output "0" (the same as decimal and ocal).
For %p match glibc and output "(nil)" rather than "0x0" or "0".
Add tests for "%#x", "% d", "%+d" and passing NULL to "%p".
David Laight [Sun, 8 Mar 2026 11:37:36 +0000 (11:37 +0000)]
tools/nolibc/printf: Prepend sign to converted number
Instead of appending the converted number to the sign, convert first
and then prepend the sign (or "0x").
Use the length returned by u64toh_r() instead of calling strlen().
Needed so that zero padding can be inserted between the sign and digits
in an upcoming patch.
David Laight [Sun, 8 Mar 2026 11:37:34 +0000 (11:37 +0000)]
tools/nolibc/printf: Add support for length modifiers tzqL and formats iX
Length modifiers t (ptrdiff_t) and z (size_t) are aliases for l (long),
q and L are 64bit the same as j (intmax).
Format i is an alias for d and X similar to x but upper case.
Supporting them is mostly just adding the relevant bit to the bit
pattern used for matching characters.
Although %X is detected the output will be lower case.
Change/add tests to use conversions i and X, and length modifiers L and ll.
Use the correct minimum value for "%Li".
Linus Torvalds [Fri, 20 Mar 2026 16:54:40 +0000 (09:54 -0700)]
Merge tag 'spi-fix-v7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi
Pull spi fixes from Mark Brown:
"There's a couple of core fixes here from Johan, fixing a race
condition and an error handling path, plus a bunch of driver specific
fixups.
The Qualcomm issues could be nasty if you ran into them, especially
the DMA ordering one"
* tag 'spi-fix-v7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: geni-qcom: Check DMA interrupts early in ISR
spi: fix statistics allocation
spi: fix use-after-free on controller registration failure
spi: geni-qcom: Fix CPHA and CPOL mode change detection
spi: axiado: Fix double-free in ax_spi_probe()
spi: amlogic-spisg: Fix memory leak in aml_spisg_probe()
spi: amlogic: spifc-a4: Remove redundant clock cleanup
Linus Torvalds [Fri, 20 Mar 2026 16:52:45 +0000 (09:52 -0700)]
Merge tag 'regulator-fix-v7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator
Pull regulator fix from Mark Brown:
"Just one fix here from Hugo Villeneuve, the documentation for some of
the regulator DT properties had been cut'n'pasted so that if anyone
actually read it they'd be informed that those properties had
completely incorrect meanings"
* tag 'regulator-fix-v7.0-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
regulator: dt-bindings: fix typos in regulator-uv-* descriptions
Linus Torvalds [Fri, 20 Mar 2026 16:46:15 +0000 (09:46 -0700)]
Merge tag 'pmdomain-v7.0-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm
Pull pmdomain fixes from Ulf Hansson:
- bcm: increase ASB control timeout for bcm2835
- mediatek: fix power domain count
* tag 'pmdomain-v7.0-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm:
pmdomain: bcm: bcm2835-power: Increase ASB control timeout
pmdomain: mediatek: Fix power domain count
David Laight [Sun, 8 Mar 2026 11:37:32 +0000 (11:37 +0000)]
tools/nolibc/printf: Use goto and reduce indentation
Upcoming changes will need to use goto to jump to the code that
outputs characters.
Use 'goto do_output' to output a known number of characters.
Use 'goto do_strlen_output' to output a '\0' terminated string.
Removes a level of indentation from the format processing code.
The change is best reviewed using 'git diff -b' after applying it.
David Laight [Sun, 8 Mar 2026 11:37:31 +0000 (11:37 +0000)]
tools/nolibc/printf: Simplify __nolibc_printf()
Move the check for the length modifiers into the format processing
between the field width and conversion specifier.
This lets the loop be simplified and a 'fast scan' for a format start
used.
If an error is detected (eg an invalid conversion specifier) then
copy the invalid format to the output buffer.
Reduces code size by about 10% on x86-64.
Some versions of gcc bloat this version by generating a jump table.
All goes away in the later patches.
David Laight [Sun, 8 Mar 2026 11:37:29 +0000 (11:37 +0000)]
tools/nolibc: Rename the 'errnum' parameter to strerror()
Change the parameter variable name from 'errno' to 'errnum'.
Matches any documentation and avoids any issues that might happen
if errno is actually a #define (which is not uncommon).
David Laight [Sun, 8 Mar 2026 11:37:28 +0000 (11:37 +0000)]
tools/nolibc: Implement strerror() in terms of strerror_r()
strerror() can be the only part of a program that has a .data section.
This requires 4k in the program file.
Add a simple implementation of strerror_r() and use that in strerror()
so that the "errno=" string is copied at run-time.
Use __builtin_memcpy() because that optimises away the input string
and just writes the required constants to the target buffer.
Code size change largely depends on whether the inlining decision for
strerror() changes.
Change the tests to use the normal EXPECT_VFPRINTF() when testing %m.
Skip the tests when !is_nolibc.
David Laight [Mon, 2 Mar 2026 10:17:59 +0000 (10:17 +0000)]
selftests/nolibc: Let EXPECT_VFPRINTF() tests be skipped
Tests that check explicit nolibc behavior (eg "%m") or test places
where the nolibc behaviour deviates from the libc need skipping
when compiled to use the host libc.
David Laight [Mon, 2 Mar 2026 10:17:57 +0000 (10:17 +0000)]
selftests/nolibc: Use length of 'expected' string to check snprintf() output
Instead of requiring the test cases specifying both the length and
expected output, take the length from the expected output.
Tests that expect the output be truncated are changed to specify
the un-truncated output.
Change the strncmp() to a memcmp() with an extra check that the
output is actually terminated.
Append a '+' to the printed output (after the final ") when the output
is truncated.
David Laight [Mon, 2 Mar 2026 10:17:55 +0000 (10:17 +0000)]
selftests/nolibc: Return correct value when printf test fails
Correctly return 1 (the number of errors) when strcmp()
fails rather than the return value from strncmp() which is the
signed difference between the mismatching characters.
David Laight [Mon, 23 Feb 2026 10:17:20 +0000 (10:17 +0000)]
tools/nolibc: Optimise and common up the number to ascii functions
Implement u[64]to[ah]_r() using a common function that uses multiply
by reciprocal to generate the least significant digit first and then
reverses the string.
On 32bit this is five multiplies (with 64bit product) for each output
digit. I think the old utoa_r() always did 36 multiplies and a lot
of subtracts - so this is likely faster even for 32bit values.
Definitely better for 64bit values (especially small ones).
Clearly shifts are faster for base 16, but reversing the output buffer
makes a big difference.
Sharing the code reduces the footprint (unless gcc decides to constant
fold the functions).
Definitely helps vfprintf() where the constants get loaded and a single
call is done.
Also makes it cheap to add octal support to vfprintf for completeness.
David Laight [Mon, 23 Feb 2026 10:17:21 +0000 (10:17 +0000)]
selftests/nolibc: Fix build with host headers and libc
Many systems don't have strlcpy() or strlcat() and readdir_r() is
deprecated. This makes the tests fail to build with the host headers.
Disable the 'directories' test and define strlcpy(), strlcat() and
readdir_r() using #defines so that the code compiles.
Thomas Weißschuh [Wed, 18 Mar 2026 17:00:33 +0000 (18:00 +0100)]
selftests/nolibc: fix test_file_stream() on musl libc
fwrite() modifying errno is non-standard.
Only validate this behavior on those libc implementations which
implement it.
Fixes: a5f00be9b3b0 ("tools/nolibc: Add a simple test for writing to a FILE and reading it back") Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Linus Torvalds [Fri, 20 Mar 2026 16:38:12 +0000 (09:38 -0700)]
Merge tag 'ata-7.0-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux
Pull ata fixes from Niklas Cassel:
- ADATA SU680 SSDs are causing command timeouts when LPM is enabled.
Enable the ATA_QUIRK_NOLPM quirk to prevent LPM from being enabled
on these devices (Damien)
- When receiving a REPORT SUPPORTED OPERATION CODES command with an
invalid REPORTING OPTIONS format, sense data should have the field
pointer set to byte 2 (the location of the REPORTING OPTIONS field)
instead of incorrectly pointing to byte 1 (Damien)
* tag 'ata-7.0-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux:
ata: libata-scsi: report correct sense field pointer in ata_scsiop_maint_in()
ata: libata-core: disable LPM on ADATA SU680 SSD
projects / thirdparty / kernel / stable.git / log
