git.monstr.eu Git - linux-2.6-microblaze.git/log

Merge tag 'kvmarm-fixes-6.18-1' of git://git./linux/kernel/git/kvmarm/kvmarm into HEAD

KVM/arm64 fixes for 6.18, take #1

Improvements and bug fixes:

- Fix the handling of ZCR_EL2 in NV VMs
  (20250926194108.84093-1-oliver.upton@linux.dev)

- Pick the correct translation regime when doing a PTW on
  the back of a SEA (20250926224246.731748-1-oliver.upton@linux.dev)

- Prevent userspace from injecting an event into a vcpu that isn't
  initialised yet (20250930085237.108326-1-oliver.upton@linux.dev)

- Move timer save/restore to the sysreg handling code, fixing EL2 timer
  access in the process (20250929160458.3351788-1-maz@kernel.org)

- Add FGT-based trapping of MDSCR_EL1 to reduce the overhead of debug
  (20250924235150.617451-1-oliver.upton@linux.dev)

- Fix trapping configuration when the host isn't GICv3
  (20251007160704.1673584-1-sascha.bischoff@arm.com)

- Improve the detection of HCR_EL2.E2H being RES1
  (20251009121239.29370-1-maz@kernel.org)

- Drop a spurious 'break' statement in the S1 PTW
  (20250930135621.162050-1-osama.abdelkader@gmail.com)

- Don't try to access SPE when owned by EL3
  (20251010174707.1684200-1-mukesh.ojha@oss.qualcomm.com)

Documentation updates:

- Document the failure modes of event injection
  (20250930233620.124607-1-oliver.upton@linux.dev)

- Document that a GICv3 guest can be created on a GICv5 host
  with FEAT_GCIE_LEGACY (20251007154848.1640444-1-sascha.bischoff@arm.com)

Selftest improvements:

- Add a selftest for the effective value of HCR_EL2.AMO
  (20250926224454.734066-1-oliver.upton@linux.dev)

- Address build warning in the timer selftest when building
  with clang (20250926155838.2612205-1-seanjc@google.com)

- Teach irq_fd selftests about non-x86 architectures
  (20250930193301.119859-1-oliver.upton@linux.dev)

- Add missing sysregs to the set_id_regs selftest
  (20251012154352.61133-1-zenghui.yu@linux.dev)

- Fix vcpu allocation in the vgic_lpi_stress selftest
  (20251008154520.54801-1-zenghui.yu@linux.dev)

- Correctly enable interrupts in the vgic_lpi_stress selftest
  (20251007195254.260539-1-oliver.upton@linux.dev)

arm64: Revamp HCR_EL2.E2H RES1 detection

We currently have two ways to identify CPUs that only implement FEAT_VHE
and not FEAT_E2H0:

- either they advertise it via ID_AA64MMFR4_EL1.E2H0,
- or the HCR_EL2.E2H bit is RAO/WI

However, there is a third category of "cpus" that fall between these
two cases: on CPUs that do not implement FEAT_FGT, it is IMPDEF whether
an access to ID_AA64MMFR4_EL1 can trap to EL2 when the register value
is zero.

A consequence of this is that on systems such as Neoverse V2, a NV
guest cannot reliably detect that it is in a VHE-only configuration
(E2H is writable, and ID_AA64MMFR0_EL1 is 0), despite the hypervisor's
best effort to repaint the id register.

Replace the RAO/WI test by a sequence that makes use of the VHE
register remnapping between EL1 and EL2 to detect this situation,
and work out whether we get the VHE behaviour even after having
set HCR_EL2.E2H to 0.

This solves the NV problem, and provides a more reliable acid test
for CPUs that do not completely follow the letter of the architecture
while providing a RES1 behaviour for HCR_EL2.E2H.

Suggested-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Tested-by: Jan Kotas <jank@cadence.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/15A85F2B-1A0C-4FA7-9FE4-EEC2203CC09E@global.cadence.com

KVM: arm64: nv: Use FGT write trap of MDSCR_EL1 when available

Marc reports that the performance of running an L3 guest has regressed
by 60% as a result of setting MDCR_EL2.TDA to hide bad architecture.
That's of course terrible for the single user of recursive NV ;-)

While there's nothing to be done on non-FGT systems, take advantage of
the precise write trap of MDSCR_EL1 and leave the rest of the debug
registers untrapped.

Reported-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Compute per-vCPU FGTs at vcpu_load()

To date KVM has used the fine-grained traps for the sake of UNDEF
enforcement (so-called FGUs), meaning the constituent parts could be
computed on a per-VM basis and folded into the effective value when
programmed.

Prepare for traps changing based on the vCPU context by computing the
whole mess of them at vcpu_load(). Aggressively inline all the helpers
to preserve the build-time checks that were there before.

Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: selftests: Fix misleading comment about virtual timer encoding

The userspace-visible encoding for CNTV_CVAL_EL0 and CNTVCNT_EL0
have been swapped for as long as usersapce has had access to the
registers. This is documented in arch/arm64/include/uapi/asm/kvm.h.

Despite that, the get_reg_list test has unhelpful comments indicating
the wrong register for the encoding.

Replace this with definitions exposed in the include file, and
a comment explaining again the brokenness.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: selftests: Add an E2H=0-specific configuration to get_reg_list

Add yet another configuration, this time dealing E2H=0.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: selftests: Make dependencies on VHE-specific registers explicit

The hyp virtual timer registers only exist when VHE is present,
Similarly, VNCR_EL2 only exists when NV2 is present.

Make these dependencies explicit.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Kill leftovers of ad-hoc timer userspace access

Now that the whole timer infrastructure is handled as system register
accesses, get rid of the now unused ad-hoc infrastructure.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Fix WFxT handling of nested virt

The spec for WFxT indicates that the parameter to the WFxT instruction
is relative to the reading of CNTVCT_EL0. This means that the implementation
needs to take the execution context into account, as CNTVOFF_EL2
does not always affect readings of CNTVCT_EL0 (such as when HCR_EL2.E2H
is 1 and that we're in host context).

This also rids us of the last instance of KVM_REG_ARM_TIMER_CNT
outside of the userspace interaction code.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Move CNT*CT_EL0 userspace accessors to generic infrastructure

Moving the counter registers is a bit more involved than for the control
and comparator (there is no shadow data for the counter), but still
pretty manageable.

Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Move CNT*_CVAL_EL0 userspace accessors to generic infrastructure

As for the control registers, move the comparator registers to
the common infrastructure.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Move CNT*_CTL_EL0 userspace accessors to generic infrastructure

Remove the handling of CNT*_CTL_EL0 from guest.c, and move it to
sys_regs.c, using a new TIMER_REG() definition to encapsulate it.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Add timer UAPI workaround to sysreg infrastructure

Amongst the numerous bugs that plague the KVM/arm64 UAPI, one of
the most annoying thing is that the userspace view of the virtual
timer has its CVAL and CNT encodings swapped.

In order to reduce the amount of code that has to know about this,
start by adding handling for this bug in the sys_reg code.

Nothing is making use of it yet, as the code responsible for userspace
interaction is catching the accesses early.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Make timer_set_offset() generally accessible

Move the timer_set_offset() helper to arm_arch_timer.h, so that it
is next to timer_get_offset(), and accessible by the rest of KVM.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Replace timer context vcpu pointer with timer_id

Having to follow a pointer to a vcpu is pretty dumb, when the timers
are are a fixed offset in the vcpu structure itself.

Trade the vcpu pointer for a timer_id, which can then be used to
compute the vcpu address as needed.

Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Introduce timer_context_to_vcpu() helper

We currently have a vcpu pointer nested into each timer context.

As we are about to remove this pointer, introduce a helper (aptly
named timer_context_to_vcpu()) that returns this pointer, at least
until we repaint the data structure.

Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Hide CNTHV_*_EL2 from userspace for nVHE guests

Although we correctly UNDEF any CNTHV_*_EL2 access from the guest
when E2H==0, we still expose these registers to userspace, which
is a bad idea.

Drop the ad-hoc UNDEF injection and switch to a .visibility()
callback which will also hide the register from userspace.

Fixes: 0e45981028550 ("KVM: arm64: timer: Don't adjust the EL2 virtual timer offset")
Signed-off-by: Marc Zyngier <maz@kernel.org>

Documentation: KVM: Update GICv3 docs for GICv5 hosts

GICv5 hosts optionally include FEAT_GCIE_LEGACY, which allows them to
execute GICv3-based VMs on GICv5 hardware. Update the GICv3
documentation to reflect this now that GICv3 guests are supports on
compatible GICv5 hosts.

Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: gic-v3: Only set ICH_HCR traps for v2-on-v3 or v3 guests

The ICH_HCR_EL2 traps are used when running on GICv3 hardware, or when
running a GICv3-based guest using FEAT_GCIE_LEGACY on GICv5
hardware. When running a GICv2 guest on GICv3 hardware the traps are
used to ensure that the guest never sees any part of GICv3 (only GICv2
is visible to the guest), and when running a GICv3 guest they are used
to trap in specific scenarios. They are not applicable for a
GICv2-native guest, and won't be applicable for a(n upcoming) GICv5
guest.

The traps themselves are configured in the vGIC CPU IF state, which is
stored as a union. Updating the wrong aperture of the union risks
corrupting state, and therefore needs to be avoided at all costs.

Bail early if we're not running a compatible guest (GICv2 on GICv3
hardware, GICv3 native, GICv3 on GICv5 hardware). Trap everything
unconditionally if we're running a GICv2 guest on GICv3
hardware. Otherwise, conditionally set up GICv3-native trapping.

Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: selftests: Actually enable IRQs in vgic_lpi_stress

vgic_lpi_stress rather hilariously leaves IRQs disabled for the duration
of the test. While the ITS translation of MSIs happens regardless of
this, for completeness the guest should actually handle the LPIs.

Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Reviewed-by: Zenghui Yu <zenghui.yu@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: selftests: Allocate vcpus with correct size

vcpus array contains pointers to struct kvm_vcpu {}. It is way overkill
to allocate the array with (nr_cpus * sizeof(struct kvm_vcpu)). Fix the
allocation by using the correct size.

Signed-off-by: Zenghui Yu <zenghui.yu@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Guard PMSCR_EL1 initialization with SPE presence check

Commit efad60e46057 ("KVM: arm64: Initialize PMSCR_EL1 when in VHE")
does not perform sufficient check before initializing PMSCR_EL1 to 0
when running in VHE mode. On some platforms, this causes the system to
hang during boot, as EL3 has not delegated access to the Profiling
Buffer to the Non-secure world, nor does it reinject an UNDEF on sysreg
trap.

To avoid this issue, restrict the PMSCR_EL1 initialization to CPUs that
support Statistical Profiling Extension (FEAT_SPE) and have the
Profiling Buffer accessible in Non-secure EL1. This is determined via a
new helper `cpu_has_spe()` which checks both PMSVer and PMBIDR_EL1.P.

This ensures the initialization only affects CPUs where SPE is
implemented and usable, preventing boot failures on platforms where SPE
is not properly configured.

Fixes: efad60e46057 ("KVM: arm64: Initialize PMSCR_EL1 when in VHE")
Signed-off-by: Mukesh Ojha <mukesh.ojha@oss.qualcomm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: selftests: Sync ID_AA64PFR1, MPIDR, CLIDR in guest

We forgot to sync several registers (ID_AA64PFR1, MPIDR, CLIDR) in guest to
make sure that the guest had seen the written value.

Add them to the list.

Signed-off-by: Zenghui Yu <zenghui.yu@linux.dev>
Reviewed-By: Ben Horgan <ben.horgan@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Remove unreachable break after return

Remove an unnecessary 'break' statement that follows a 'return'
in arch/arm64/kvm/at.c. The break is unreachable.

Signed-off-by: Osama Abdelkader <osama.abdelkader@gmail.com>
Reviewed-by: Zenghui Yu <yuzenghui@huawei.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: selftests: Fix irqfd_test for non-x86 architectures

The KVM_IRQFD ioctl fails if no irqchip is present in-kernel, which
isn't too surprising as there's not much KVM can do for an IRQ if it
cannot resolve a destination.

As written the irqfd_test assumes that a 'default' VM created in
selftests has an in-kernel irqchip created implicitly. That may be the
case on x86 but it isn't necessarily true on other architectures.

Add an arch predicate indicating if 'default' VMs get an irqchip and
make the irqfd_test depend on it. Work around arm64 VGIC initialization
requirements by using vm_create_with_one_vcpu(), ignoring the created
vCPU as it isn't used for the test.

Reported-by: Sebastian Ott <sebott@redhat.com>
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Acked-by: Sean Christopherson <seanjc@google.com>
Fixes: 7e9b231c402a ("KVM: selftests: Add a KVM_IRQFD test to verify uniqueness requirements")
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Document vCPU event ioctls as requiring init'ed vCPU

KVM rejects calls to KVM_{GET,SET}_VCPU_EVENTS for an uninitialized vCPU
as of commit cc96679f3c03 ("KVM: arm64: Prevent access to vCPU events
before init"). Update the corresponding API documentation.

Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Prevent access to vCPU events before init

Another day, another syzkaller bug. KVM erroneously allows userspace to
pend vCPU events for a vCPU that hasn't been initialized yet, leading to
KVM interpreting a bunch of uninitialized garbage for routing /
injecting the exception.

In one case the injection code and the hyp disagree on whether the vCPU
has a 32bit EL1 and put the vCPU into an illegal mode for AArch64,
tripping the BUG() in exception_target_el() during the next injection:

  kernel BUG at arch/arm64/kvm/inject_fault.c:40!
  Internal error: Oops - BUG: 00000000f2000800 [#1]  SMP
  CPU: 3 UID: 0 PID: 318 Comm: repro Not tainted 6.17.0-rc4-00104-g10fd0285305d #6 PREEMPT
  Hardware name: linux,dummy-virt (DT)
  pstate: 21402009 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
  pc : exception_target_el+0x88/0x8c
  lr : pend_serror_exception+0x18/0x13c
  sp : ffff800082f03a10
  x29: ffff800082f03a10 x28: ffff0000cb132280 x27: 0000000000000000
  x26: 0000000000000000 x25: ffff0000c2a99c20 x24: 0000000000000000
  x23: 0000000000008000 x22: 0000000000000002 x21: 0000000000000004
  x20: 0000000000008000 x19: ffff0000c2a99c20 x18: 0000000000000000
  x17: 0000000000000000 x16: 0000000000000000 x15: 00000000200000c0
  x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000
  x11: 0000000000000000 x10: 0000000000000000 x9 : 0000000000000000
  x8 : ffff800082f03af8 x7 : 0000000000000000 x6 : 0000000000000000
  x5 : ffff800080f621f0 x4 : 0000000000000000 x3 : 0000000000000000
  x2 : 000000000040009b x1 : 0000000000000003 x0 : ffff0000c2a99c20
  Call trace:
   exception_target_el+0x88/0x8c (P)
   kvm_inject_serror_esr+0x40/0x3b4
   __kvm_arm_vcpu_set_events+0xf0/0x100
   kvm_arch_vcpu_ioctl+0x180/0x9d4
   kvm_vcpu_ioctl+0x60c/0x9f4
   __arm64_sys_ioctl+0xac/0x104
   invoke_syscall+0x48/0x110
   el0_svc_common.constprop.0+0x40/0xe0
   do_el0_svc+0x1c/0x28
   el0_svc+0x34/0xf0
   el0t_64_sync_handler+0xa0/0xe4
   el0t_64_sync+0x198/0x19c
  Code: f946bc01 b4fffe61 9101e020 17fffff2 (d4210000)

Reject the ioctls outright as no sane VMM would call these before
KVM_ARM_VCPU_INIT anyway. Even if it did the exception would've been
thrown away by the eventual reset of the vCPU's state.

Cc: stable@vger.kernel.org # 6.17
Fixes: b7b27facc7b5 ("arm/arm64: KVM: Add KVM_GET/SET_VCPU_EVENTS")
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: selftests: Track width of timer counter as "int", not "uint64_t"

Store the width of arm64's timer counter as an "int", not a "uint64_t".
ilog2() returns an "int", and more importantly using what is an "unsigned
long" under the hood makes clang unhappy due to a type mismatch when
clamping the width to a sane value.

  arm64/arch_timer_edge_cases.c:1032:10: error: comparison of distinct pointer types
     ('typeof (width) *' (aka 'unsigned long *') and 'typeof (56) *' (aka 'int *'))
     [-Werror,-Wcompare-distinct-pointer-types]
   1032 |         width = clamp(width, 56, 64);
        |                 ^~~~~~~~~~~~~~~~~~~~
  tools/include/linux/kernel.h:47:45: note: expanded from macro 'clamp'
     47 | #define clamp(val, lo, hi)      min((typeof(val))max(val, lo), hi)
        |                                                  ^~~~~~~~~~~~
  tools/include/linux/kernel.h:33:17: note: expanded from macro 'max'
     33 |         (void) (&_max1 == &_max2);              \
        |                 ~~~~~~ ^  ~~~~~~
  tools/include/linux/kernel.h:39:9: note: expanded from macro 'min'
     39 |         typeof(x) _min1 = (x);                  \
        |                ^

Fixes: fad4cf944839 ("KVM: arm64: selftests: Determine effective counter width in arch_timer_edge_cases")
Cc: Sebastian Ott <sebott@redhat.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: selftests: Test effective value of HCR_EL2.AMO

A defect against the architecture now allows an implementation to treat
AMO as 1 when HCR_EL2.{E2H, TGE} = {1, 0}. KVM now takes advantage of
this interpretation to address a quality of emulation issue w.r.t.
SError injection.

Add a corresponding test case and expect a pending SError to be taken.

Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: Use the in-context stage-1 in __kvm_find_s1_desc_level()

Running the external_aborts selftest at EL2 leads to an ugly splat due
to the stage-1 MMU being disabled for the walked context, owing to the
fact that __kvm_find_s1_desc_level() is hardcoded to the EL1&0 regime.

Select the appropriate translation regime for the stage-1 walk based on
the current vCPU context.

Fixes: b8e625167a32 ("KVM: arm64: Add S1 IPA to page table level walker")
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: nv: Don't advance PC when pending an SVE exception

Jan reports that running a nested guest on Neoverse-V2 leads to a WARN
in the host due to simultaneously pending an exception and PC increment
after an access to ZCR_EL2.

Returning true from a sysreg accessor is an indication that the sysreg
instruction has been retired. Of course this isn't the case when we've
pended a synchronous SVE exception for the guest. Fix the return value
and let the exception propagate to the guest as usual.

Reported-by: Jan Kotas <jank@cadence.com>
Closes: https://lore.kernel.org/kvmarm/865xd61tt5.wl-maz@kernel.org/
Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

KVM: arm64: nv: Don't treat ZCR_EL2 as a 'mapped' register

Unlike the other mapped EL2 sysregs ZCR_EL2 isn't guaranteed to be
resident when a vCPU is loaded as it actually follows the SVE
context. As such, the contents of ZCR_EL1 may belong to another guest if
the vCPU has been preempted before reaching sysreg emulation.

Unconditionally use the in-memory value of ZCR_EL2 and switch to the
memory-only accessors. The in-memory value is guaranteed to be valid as
fpsimd_lazy_switch_to_{guest,host}() will restore/save the register
appropriately.

Signed-off-by: Oliver Upton <oliver.upton@linux.dev>
Signed-off-by: Marc Zyngier <maz@kernel.org>

Linux 6.18-rc1

Merge tag 'i2c-for-6.18-rc1-hotfix' of git://git./linux/kernel/git/wsa/linux

Pull i2c fix from Wolfram Sang:
"One revert because of a regression in the I2C core which has sadly not
showed up during its time in -next"

* tag 'i2c-for-6.18-rc1-hotfix' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
Revert "i2c: boardinfo: Annotate code used in init phase only"

Merge tag 'irq_urgent_for_v6.18_rc1' of git://git./linux/kernel/git/tip/tip

Pull irq fixes from Borislav Petkov:

- Skip interrupt ID 0 in sifive-plic during suspend/resume because
   ID 0 is reserved and accessing reserved register space could result
   in undefined behavior

- Fix a function's retval check in aspeed-scu-ic

* tag 'irq_urgent_for_v6.18_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/sifive-plic: Avoid interrupt ID 0 handling during suspend/resume
  irqchip/aspeed-scu-ic: Fix an IS_ERR() vs NULL check

Merge tag 'trace-v6.18-3' of git://git./linux/kernel/git/trace/linux-trace

Pull tracing fixes from Steven Rostedt:
"The previous fix to trace_marker required updating trace_marker_raw as
  well. The difference between trace_marker_raw from trace_marker is
  that the raw version is for applications to write binary structures
  directly into the ring buffer instead of writing ASCII strings. This
  is for applications that will read the raw data from the ring buffer
  and get the data structures directly. It's a bit quicker than using
  the ASCII version.

  Unfortunately, it appears that our test suite has several tests that
  test writes to the trace_marker file, but lacks any tests to the
  trace_marker_raw file (this needs to be remedied). Two issues came
  about the update to the trace_marker_raw file that syzbot found:

   - Fix tracing_mark_raw_write() to use per CPU buffer

     The fix to use the per CPU buffer to copy from user space was
     needed for both the trace_maker and trace_maker_raw file.

     The fix for reading from user space into per CPU buffers properly
     fixed the trace_marker write function, but the trace_marker_raw
     file wasn't fixed properly. The user space data was correctly
     written into the per CPU buffer, but the code that wrote into the
     ring buffer still used the user space pointer and not the per CPU
     buffer that had the user space data already written.

   - Stop the fortify string warning from writing into trace_marker_raw

     After converting the copy_from_user_nofault() into a memcpy(),
     another issue appeared. As writes to the trace_marker_raw expects
     binary data, the first entry is a 4 byte identifier. The entry
     structure is defined as:

     struct {
    struct trace_entry ent;
    int id;
    char buf[];
     };

     The size of this structure is reserved on the ring buffer with:

       size = sizeof(*entry) + cnt;

     Then it is copied from the buffer into the ring buffer with:

       memcpy(&entry->id, buf, cnt);

     This use to be a copy_from_user_nofault(), but now converting it to
     a memcpy() triggers the fortify-string code, and causes a warning.

     The allocated space is actually more than what is copied, as the
     cnt used also includes the entry->id portion. Allocating
     sizeof(*entry) plus cnt is actually allocating 4 bytes more than
     what is needed.

     Change the size function to:

       size = struct_size(entry, buf, cnt - sizeof(entry->id));

     And update the memcpy() to unsafe_memcpy()"

* tag 'trace-v6.18-3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  tracing: Stop fortify-string from warning in tracing_mark_raw_write()
  tracing: Fix tracing_mark_raw_write() to use buf and not ubuf

Merge tag 'kbuild-fixes-6.18-1' of git://git./linux/kernel/git/kbuild/linux

Pull Kbuild fixes from Nathan Chancellor:

- Fix UAPI types check in headers_check.pl

- Only enable -Werror for hostprogs with CONFIG_WERROR / W=e

- Ignore fsync() error when output of gen_init_cpio is a pipe

- Several little build fixes for recent modules.builtin.modinfo series

* tag 'kbuild-fixes-6.18-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kbuild/linux:
  kbuild: Use '--strip-unneeded-symbol' for removing module device table symbols
  s390/vmlinux.lds.S: Move .vmlinux.info to end of allocatable sections
  kbuild: Add '.rel.*' strip pattern for vmlinux
  kbuild: Restore pattern to avoid stripping .rela.dyn from vmlinux
  gen_init_cpio: Ignore fsync() returning EINVAL on pipes
  scripts/Makefile.extrawarn: Respect CONFIG_WERROR / W=e for hostprogs
  kbuild: uapi: Strip comments before size type check

Revert "i2c: boardinfo: Annotate code used in init phase only"

This reverts commit 1a2b423be6a89dd07d5fc27ea042be68697a6a49 because we
got a regression report and need time to find out the details.

Reported-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Closes: https://lore.kernel.org/r/29ec0082-4dd4-4120-acd2-44b35b4b9487@oss.qualcomm.com
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

Merge tag 'rtc-6.18' of git://git./linux/kernel/git/abelloni/linux

Pull RTC updates from Alexandre Belloni:
"This cycle, we have a new RTC driver, for the SpacemiT P1. The optee
  driver gets alarm support. We also get a fix for a race condition that
  was fairly rare unless while stress testing the alarms.

  Subsystem:
   - Fix race when setting alarm
   - Ensure alarm irq is enabled when UIE is enabled
   - remove unneeded 'fast_io' parameter in regmap_config

  New driver:
   - SpacemiT P1 RTC

  Drivers:
   - efi: Remove wakeup functionality
   - optee: add alarms support
   - s3c: Drop support for S3C2410
   - zynqmp: Restore alarm functionality after kexec transition"

* tag 'rtc-6.18' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux: (29 commits)
  rtc: interface: Ensure alarm irq is enabled when UIE is enabled
  rtc: tps6586x: Fix initial enable_irq/disable_irq balance
  rtc: cpcap: Fix initial enable_irq/disable_irq balance
  rtc: isl12022: Fix initial enable_irq/disable_irq balance
  rtc: interface: Fix long-standing race when setting alarm
  rtc: pcf2127: fix watchdog interrupt mask on pcf2131
  rtc: zynqmp: Restore alarm functionality after kexec transition
  rtc: amlogic-a4: Optimize global variables
  rtc: sd2405al: Add I2C address.
  rtc: Kconfig: move symbols to proper section
  rtc: optee: make optee_rtc_pm_ops static
  rtc: optee: Fix error code in optee_rtc_read_alarm()
  rtc: optee: fix error code in probe()
  dt-bindings: rtc: Convert apm,xgene-rtc to DT schema
  rtc: spacemit: support the SpacemiT P1 RTC
  rtc: optee: add alarm related rtc ops to optee rtc driver
  rtc: optee: remove unnecessary memory operations
  rtc: optee: fix memory leak on driver removal
  rtc: x1205: Fix Xicor X1205 vendor prefix
  dt-bindings: rtc: Fix Xicor X1205 vendor prefix
  ...

Merge tag 'scsi-misc' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
"Fixes only in drivers (ufs, mvsas, qla2xxx, target) that came in just
  before or during the merge window.

  The most important one is the qla2xxx which reverts a conversion to
  fix flexible array member warnings, that went up in this merge window
  but which turned out on further testing to be causing data corruption"

* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: ufs: core: Include UTP error in INT_FATAL_ERRORS
  scsi: ufs: sysfs: Make HID attributes visible
  scsi: mvsas: Fix use-after-free bugs in mvs_work_queue
  scsi: ufs: core: Fix PM QoS mutex initialization
  scsi: ufs: core: Fix runtime suspend error deadlock
  Revert "scsi: qla2xxx: Fix memcpy() field-spanning write issue"
  scsi: target: target_core_configfs: Add length check to avoid buffer overflow

Merge tag 'x86_core_for_v6.18_rc1' of git://git./linux/kernel/git/tip/tip

Pull more x86 updates from Borislav Petkov:

- Remove a bunch of asm implementing condition flags testing in KVM's
   emulator in favor of int3_emulate_jcc() which is written in C

- Replace KVM fastops with C-based stubs which avoids problems with the
   fastop infra related to latter not adhering to the C ABI due to their
   special calling convention and, more importantly, bypassing compiler
   control-flow integrity checking because they're written in asm

- Remove wrongly used static branches and other ugliness accumulated
   over time in hyperv's hypercall implementation with a proper static
   function call to the correct hypervisor call variant

- Add some fixes and modifications to allow running FRED-enabled
   kernels in KVM even on non-FRED hardware

- Add kCFI improvements like validating indirect calls and prepare for
   enabling kCFI with GCC. Add cmdline params documentation and other
   code cleanups

- Use the single-byte 0xd6 insn as the official #UD single-byte
   undefined opcode instruction as agreed upon by both x86 vendors

- Other smaller cleanups and touchups all over the place

* tag 'x86_core_for_v6.18_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (24 commits)
  x86,retpoline: Optimize patch_retpoline()
  x86,ibt: Use UDB instead of 0xEA
  x86/cfi: Remove __noinitretpoline and __noretpoline
  x86/cfi: Add "debug" option to "cfi=" bootparam
  x86/cfi: Standardize on common "CFI:" prefix for CFI reports
  x86/cfi: Document the "cfi=" bootparam options
  x86/traps: Clarify KCFI instruction layout
  compiler_types.h: Move __nocfi out of compiler-specific header
  objtool: Validate kCFI calls
  x86/fred: KVM: VMX: Always use FRED for IRQs when CONFIG_X86_FRED=y
  x86/fred: Play nice with invoking asm_fred_entry_from_kvm() on non-FRED hardware
  x86/fred: Install system vector handlers even if FRED isn't fully enabled
  x86/hyperv: Use direct call to hypercall-page
  x86/hyperv: Clean up hv_do_hypercall()
  KVM: x86: Remove fastops
  KVM: x86: Convert em_salc() to C
  KVM: x86: Introduce EM_ASM_3WCL
  KVM: x86: Introduce EM_ASM_1SRC2
  KVM: x86: Introduce EM_ASM_2CL
  KVM: x86: Introduce EM_ASM_2W
  ...

Merge tag 'x86_cleanups_for_v6.18_rc1' of git://git./linux/kernel/git/tip/tip

Pull x86 cleanups from Borislav Petkov:

- Simplify inline asm flag output operands now that the minimum
   compiler version supports the =@ccCOND syntax

- Remove a bunch of AS_* Kconfig symbols which detect assembler support
   for various instruction mnemonics now that the minimum assembler
   version supports them all

- The usual cleanups all over the place

* tag 'x86_cleanups_for_v6.18_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/asm: Remove code depending on __GCC_ASM_FLAG_OUTPUTS__
  x86/sgx: Use ENCLS mnemonic in <kernel/cpu/sgx/encls.h>
  x86/mtrr: Remove license boilerplate text with bad FSF address
  x86/asm: Use RDPKRU and WRPKRU mnemonics in <asm/special_insns.h>
  x86/idle: Use MONITORX and MWAITX mnemonics in <asm/mwait.h>
  x86/entry/fred: Push __KERNEL_CS directly
  x86/kconfig: Remove CONFIG_AS_AVX512
  crypto: x86 - Remove CONFIG_AS_VPCLMULQDQ
  crypto: X86 - Remove CONFIG_AS_VAES
  crypto: x86 - Remove CONFIG_AS_GFNI
  x86/kconfig: Drop unused and needless config X86_64_SMP

Merge tag 'slab-for-6.18-rc1-hotfix' of git://git./linux/kernel/git/vbabka/slab

Pull slab fix from Vlastimil Babka:
"A NULL pointer deref hotfix"

* tag 'slab-for-6.18-rc1-hotfix' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab:
slab: fix barn NULL pointer dereference on memoryless nodes

Merge tag 'bpf-fixes' of git://git./linux/kernel/git/bpf/bpf

Pull bpf fixes from Alexei Starovoitov:

- Finish constification of 1st parameter of bpf_d_path() (Rong Tao)

- Harden userspace-supplied xdp_desc validation (Alexander Lobakin)

- Fix metadata_dst leak in __bpf_redirect_neigh_v{4,6}() (Daniel
   Borkmann)

- Fix undefined behavior in {get,put}_unaligned_be32() (Eric Biggers)

- Use correct context to unpin bpf hash map with special types (KaFai
   Wan)

* tag 'bpf-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
  selftests/bpf: Add test for unpinning htab with internal timer struct
  bpf: Avoid RCU context warning when unpinning htab with internal structs
  xsk: Harden userspace-supplied xdp_desc validation
  bpf: Fix metadata_dst leak __bpf_redirect_neigh_v{4,6}
  libbpf: Fix undefined behavior in {get,put}_unaligned_be32()
  bpf: Finish constification of 1st parameter of bpf_d_path()

Merge tag 'mm-nonmm-stable-2025-10-10-15-03' of git://git./linux/kernel/git/akpm/mm

Pull more updates from Andrew Morton:
"Just one series here - Mike Rappoport has taught KEXEC handover to
  preserve vmalloc allocations across handover"

* tag 'mm-nonmm-stable-2025-10-10-15-03' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  lib/test_kho: use kho_preserve_vmalloc instead of storing addresses in fdt
  kho: add support for preserving vmalloc allocations
  kho: replace kho_preserve_phys() with kho_preserve_pages()
  kho: check if kho is finalized in __kho_preserve_order()
  MAINTAINERS, .mailmap: update Umang's email address

Merge tag 'mm-hotfixes-stable-2025-10-10-15-00' of git://git./linux/kernel/git/akpm/mm

Pull misc fixes from Andrew Morton:
"7 hotfixes.  All 7 are cc:stable and all 7 are for MM.

  All singletons, please see the changelogs for details"

* tag 'mm-hotfixes-stable-2025-10-10-15-00' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  mm: hugetlb: avoid soft lockup when mprotect to large memory area
  fsnotify: pass correct offset to fsnotify_mmap_perm()
  mm/ksm: fix flag-dropping behavior in ksm_madvise
  mm/damon/vaddr: do not repeat pte_offset_map_lock() until success
  mm/rmap: fix soft-dirty and uffd-wp bit loss when remapping zero-filled mTHP subpage to shared zeropage
  mm/thp: fix MTE tag mismatch when replacing zero-filled subpages
  memcg: skip cgroup_file_notify if spinning is not allowed

tracing: Stop fortify-string from warning in tracing_mark_raw_write()

The way tracing_mark_raw_write() records its data is that it has the
following structure:

  struct {
struct trace_entry;
int id;
char buf[];
  };

But memcpy(&entry->id, buf, size) triggers the following warning when the
size is greater than the id:

------------[ cut here ]------------
memcpy: detected field-spanning write (size 6) of single field "&entry->id" at kernel/trace/trace.c:7458 (size 4)
WARNING: CPU: 7 PID: 995 at kernel/trace/trace.c:7458 write_raw_marker_to_buffer.isra.0+0x1f9/0x2e0
Modules linked in:
CPU: 7 UID: 0 PID: 995 Comm: bash Not tainted 6.17.0-test-00007-g60b82183e78a-dirty #211 PREEMPT(voluntary)
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.17.0-debian-1.17.0-1 04/01/2014
RIP: 0010:write_raw_marker_to_buffer.isra.0+0x1f9/0x2e0
Code: 04 00 75 a7 b9 04 00 00 00 48 89 de 48 89 04 24 48 c7 c2 e0 b1 d1 b2 48 c7 c7 40 b2 d1 b2 c6 05 2d 88 6a 04 01 e8 f7 e8 bd ff <0f> 0b 48 8b 04 24 e9 76 ff ff ff 49 8d 7c 24 04 49 8d 5c 24 08 48
RSP: 0018:ffff888104c3fc78 EFLAGS: 00010292
RAX: 0000000000000000 RBX: 0000000000000006 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 1ffffffff6b363b4 RDI: 0000000000000001
RBP: ffff888100058a00 R08: ffffffffb041d459 R09: ffffed1020987f40
R10: 0000000000000007 R11: 0000000000000001 R12: ffff888100bb9010
R13: 0000000000000000 R14: 00000000000003e3 R15: ffff888134800000
FS:  00007fa61d286740(0000) GS:ffff888286cad000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000560d28d509f1 CR3: 00000001047a4006 CR4: 0000000000172ef0
Call Trace:
  <TASK>
  tracing_mark_raw_write+0x1fe/0x290
  ? __pfx_tracing_mark_raw_write+0x10/0x10
  ? security_file_permission+0x50/0xf0
  ? rw_verify_area+0x6f/0x4b0
  vfs_write+0x1d8/0xdd0
  ? __pfx_vfs_write+0x10/0x10
  ? __pfx_css_rstat_updated+0x10/0x10
  ? count_memcg_events+0xd9/0x410
  ? fdget_pos+0x53/0x5e0
  ksys_write+0x182/0x200
  ? __pfx_ksys_write+0x10/0x10
  ? do_user_addr_fault+0x4af/0xa30
  do_syscall_64+0x63/0x350
  entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x7fa61d318687
Code: 48 89 fa 4c 89 df e8 58 b3 00 00 8b 93 08 03 00 00 59 5e 48 83 f8 fc 74 1a 5b c3 0f 1f 84 00 00 00 00 00 48 8b 44 24 10 0f 05 <5b> c3 0f 1f 80 00 00 00 00 83 e2 39 83 fa 08 75 de e8 23 ff ff ff
RSP: 002b:00007ffd87fe0120 EFLAGS: 00000202 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 00007fa61d286740 RCX: 00007fa61d318687
RDX: 0000000000000006 RSI: 0000560d28d509f0 RDI: 0000000000000001
RBP: 0000560d28d509f0 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000000006
R13: 00007fa61d4715c0 R14: 00007fa61d46ee80 R15: 0000000000000000
  </TASK>
---[ end trace 0000000000000000 ]---

This is because fortify string sees that the size of entry->id is only 4
bytes, but it is writing more than that. But this is OK as the
dynamic_array is allocated to handle that copy.

The size allocated on the ring buffer was actually a bit too big:

  size = sizeof(*entry) + cnt;

But cnt includes the 'id' and the buffer data, so adding cnt to the size
of *entry actually allocates too much on the ring buffer.

Change the allocation to:

  size = struct_size(entry, buf, cnt - sizeof(entry->id));

and the memcpy() to unsafe_memcpy() with an added justification.

Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: https://lore.kernel.org/20251011112032.77be18e4@gandalf.local.home
Fixes: 64cf7d058a00 ("tracing: Have trace_marker use per-cpu data to read user space")
Reported-by: syzbot+9a2ede1643175f350105@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/all/68e973f5.050a0220.1186a4.0010.GAE@google.com/
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>

slab: fix barn NULL pointer dereference on memoryless nodes

Phil reported a boot failure once sheaves become used in commits
59faa4da7cd4 ("maple_tree: use percpu sheaves for maple_node_cache") and
3accabda4da1 ("mm, vma: use percpu sheaves for vm_area_struct cache"):

BUG: kernel NULL pointer dereference, address: 0000000000000040
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
PGD 0 P4D 0
Oops: Oops: 0000 [#1] SMP NOPTI
CPU: 21 UID: 0 PID: 818 Comm: kworker/u398:0 Not tainted 6.17.0-rc3.slab+ #5 PREEMPT(voluntary)
Hardware name: Dell Inc. PowerEdge R7425/02MJ3T, BIOS 1.26.0 07/30/2025
RIP: 0010:__pcs_replace_empty_main+0x44/0x1d0
Code: ec 08 48 8b 46 10 48 8b 76 08 48 85 c0 74 0b 8b 48 18 85 c9 0f 85 e5 00 00 00 65 48 63 05 e4 ee 50 02 49 8b 84 c6 e0 00 00 00 <4c> 8b 68 40 4c 89 ef e8 b0 81 ff ff 48 89 c5 48 85 c0 74 1d 48 89
RSP: 0018:ffffd2d10950bdb0 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff8a775dab74b0 RCX: 00000000ffffffff
RDX: 0000000000000cc0 RSI: ffff8a6800804000 RDI: ffff8a680004e300
RBP: ffffd2d10950be40 R08: 0000000000000060 R09: ffffffffb9367388
R10: 00000000000149e8 R11: ffff8a6f87a38000 R12: 0000000000000cc0
R13: 0000000000000cc0 R14: ffff8a680004e300 R15: 00000000000000c0
FS:  0000000000000000(0000) GS:ffff8a77a3541000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000040 CR3: 0000000e1aa24000 CR4: 00000000003506f0
Call Trace:
  <TASK>
  ? srso_return_thunk+0x5/0x5f
  ? vm_area_alloc+0x1e/0x60
  kmem_cache_alloc_noprof+0x4ec/0x5b0
  vm_area_alloc+0x1e/0x60
  create_init_stack_vma+0x26/0x210
  alloc_bprm+0x139/0x200
  kernel_execve+0x4a/0x140
  call_usermodehelper_exec_async+0xd0/0x190
  ? __pfx_call_usermodehelper_exec_async+0x10/0x10
  ret_from_fork+0xf0/0x110
  ? __pfx_call_usermodehelper_exec_async+0x10/0x10
  ret_from_fork_asm+0x1a/0x30
  </TASK>
Modules linked in:
CR2: 0000000000000040
---[ end trace 0000000000000000 ]---
RIP: 0010:__pcs_replace_empty_main+0x44/0x1d0
Code: ec 08 48 8b 46 10 48 8b 76 08 48 85 c0 74 0b 8b 48 18 85 c9 0f 85 e5 00 00 00 65 48 63 05 e4 ee 50 02 49 8b 84 c6 e0 00 00 00 <4c> 8b 68 40 4c 89 ef e8 b0 81 ff ff 48 89 c5 48 85 c0 74 1d 48 89
RSP: 0018:ffffd2d10950bdb0 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff8a775dab74b0 RCX: 00000000ffffffff
RDX: 0000000000000cc0 RSI: ffff8a6800804000 RDI: ffff8a680004e300
RBP: ffffd2d10950be40 R08: 0000000000000060 R09: ffffffffb9367388
R10: 00000000000149e8 R11: ffff8a6f87a38000 R12: 0000000000000cc0
R13: 0000000000000cc0 R14: ffff8a680004e300 R15: 00000000000000c0
FS:  0000000000000000(0000) GS:ffff8a77a3541000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000040 CR3: 0000000e1aa24000 CR4: 00000000003506f0
Kernel panic - not syncing: Fatal exception
Kernel Offset: 0x36a00000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
---[ end Kernel panic - not syncing: Fatal exception ]---

And noted "this is an AMD EPYC 7401 with 8 NUMA nodes configured such
that memory is only on 2 of them."

# numactl --hardware
available: 8 nodes (0-7)
node 0 cpus: 0 8 16 24 32 40 48 56 64 72 80 88
node 0 size: 0 MB
node 0 free: 0 MB
node 1 cpus: 2 10 18 26 34 42 50 58 66 74 82 90
node 1 size: 31584 MB
node 1 free: 30397 MB
node 2 cpus: 4 12 20 28 36 44 52 60 68 76 84 92
node 2 size: 0 MB
node 2 free: 0 MB
node 3 cpus: 6 14 22 30 38 46 54 62 70 78 86 94
node 3 size: 0 MB
node 3 free: 0 MB
node 4 cpus: 1 9 17 25 33 41 49 57 65 73 81 89
node 4 size: 0 MB
node 4 free: 0 MB
node 5 cpus: 3 11 19 27 35 43 51 59 67 75 83 91
node 5 size: 32214 MB
node 5 free: 31625 MB
node 6 cpus: 5 13 21 29 37 45 53 61 69 77 85 93
node 6 size: 0 MB
node 6 free: 0 MB
node 7 cpus: 7 15 23 31 39 47 55 63 71 79 87 95
node 7 size: 0 MB
node 7 free: 0 MB

Linus decoded the stacktrace to get_barn() and get_node() and determined
that kmem_cache->node[numa_mem_id()] is NULL.

The problem is due to a wrong assumption that memoryless nodes only
exist on systems with CONFIG_HAVE_MEMORYLESS_NODES, where numa_mem_id()
points to the nearest node that has memory. SLUB has been allocating its
kmem_cache_node structures only on nodes with memory and so it does with
struct node_barn.

For kmem_cache_node, get_partial_node() checks if get_node() result is
not NULL, which I assumed was for protection from a bogus node id passed
to kmalloc_node() but apparently it's also for systems where
numa_mem_id() (used when no specific node is given) might return a
memoryless node.

Fix the sheaves code the same way by checking the result of get_node()
and bailing out if it's NULL. Note that cpus on such memoryless nodes
will have degraded sheaves performance, which can be improved later,
preferably by making numa_mem_id() work properly on such systems.

Fixes: 2d517aa09bbc ("slab: add opt-in caching layer of percpu sheaves")
Reported-and-tested-by: Phil Auld <pauld@redhat.com>
Closes: https://lore.kernel.org/all/20251010151116.GA436967@pauld.westford.csb/
Analyzed-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/all/CAHk-%3Dwg1xK%2BBr%3DFJ5QipVhzCvq7uQVPt5Prze6HDhQQ%3DQD_BcQ@mail.gmail.com/
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>

tracing: Fix tracing_mark_raw_write() to use buf and not ubuf

The fix to use a per CPU buffer to read user space tested only the writes
to trace_marker. But it appears that the selftests are missing tests to
the trace_maker_raw file. The trace_maker_raw file is used by applications
that writes data structures and not strings into the file, and the tools
read the raw ring buffer to process the structures it writes.

The fix that reads the per CPU buffers passes the new per CPU buffer to
the trace_marker file writes, but the update to the trace_marker_raw write
read the data from user space into the per CPU buffer, but then still used
then passed the user space address to the function that records the data.

Pass in the per CPU buffer and not the user space address.

TODO: Add a test to better test trace_marker_raw.

Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: https://lore.kernel.org/20251011035243.386098147@kernel.org
Fixes: 64cf7d058a00 ("tracing: Have trace_marker use per-cpu data to read user space")
Reported-by: syzbot+9a2ede1643175f350105@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/all/68e973f5.050a0220.1186a4.0010.GAE@google.com/
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>

kbuild: Use '--strip-unneeded-symbol' for removing module device table symbols

After commit 5ab23c7923a1 ("modpost: Create modalias for builtin
modules"), relocatable RISC-V kernels with CONFIG_KASAN=y start failing
when attempting to strip the module device table symbols:

  riscv64-linux-objcopy: not stripping symbol `__mod_device_table__kmod_irq_starfive_jh8100_intc__of__starfive_intc_irqchip_match_table' because it is named in a relocation
  make[4]: *** [scripts/Makefile.vmlinux:97: vmlinux] Error 1

The relocation appears to come from .LASANLOC5 in .data.rel.local:

  $ llvm-objdump --disassemble-symbols=.LASANLOC5 --disassemble-all -r drivers/irqchip/irq-starfive-jh8100-intc.o

  drivers/irqchip/irq-starfive-jh8100-intc.o:   file format elf64-littleriscv

  Disassembly of section .data.rel.local:

  0000000000000180 <.LASANLOC5>:
  ...
       1d0: 0000          unimp
                  00000000000001d0:  R_RISCV_64   __mod_device_table__kmod_irq_starfive_jh8100_intc__of__starfive_intc_irqchip_match_table
  ...

This section appears to come from GCC for including additional
information about global variables that may be protected by KASAN.

There appears to be no way to opt out of the generation of these symbols
through either a flag or attribute. Attempting to remove '.LASANLOC*'
with '--strip-symbol' results in the same error as above because these
symbols may refer to (thus have relocation between) each other.

Avoid this build breakage by switching to '--strip-unneeded-symbol' for
removing __mod_device_table__ symbols, as it will only remove the symbol
when there is no relocation pointing to it. While this may result in a
little more bloat in the symbol table in certain configurations, it is
not as bad as outright build failures.

Fixes: 5ab23c7923a1 ("modpost: Create modalias for builtin modules")
Reported-by: Charles Mirabile <cmirabil@redhat.com>
Closes: https://lore.kernel.org/20251007011637.2512413-1-cmirabil@redhat.com/
Suggested-by: Alexey Gladkov <legion@kernel.org>
Tested-by: Nicolas Schier <nsc@kernel.org>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>

Merge tag 'for-6.18/hpfs-changes' of git://git./linux/kernel/git/device-mapper/linux-dm

Pull hpfs updates from Mikulas Patocka:

- Avoid -Wflex-array-member-not-at-end warnings

- Replace simple_strtoul with kstrtoint

- Fix error code for new_inode() failure

* tag 'for-6.18/hpfs-changes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  fs/hpfs: Fix error code for new_inode() failure in mkdir/create/mknod/symlink
  hpfs: Replace simple_strtoul with kstrtoint in hpfs_parse_param
  fs: hpfs: Avoid multiple -Wflex-array-member-not-at-end warnings

Merge tag 'drm-next-2025-10-11-1' of https://gitlab.freedesktop.org/drm/kernel

Pull more drm fixes from Dave Airlie:
"Just the follow up fixes for rc1 from the next branch, amdgpu and xe
  mostly with a single v3d fix in there.

  amdgpu:
   - DC DCE6 fixes
   - GPU reset fixes
   - Secure diplay messaging cleanup
   - MES fix
   - GPUVM locking fixes
   - PMFW messaging cleanup
   - PCI US/DS switch handling fix
   - VCN queue reset fix
   - DC FPU handling fix
   - DCN 3.5 fix
   - DC mirroring fix

  amdkfd:
   - Fix kfd process ref leak
   - mmap write lock handling fix
   - Fix comments in IOCTL

  xe:
   - Fix build with clang 16
   - Fix handling of invalid configfs syntax usage and spell out the
     expected syntax in the documentation
   - Do not try late bind firmware when running as VF since it shouldn't
     handle firmware loading
   - Fix idle assertion for local BOs
   - Fix uninitialized variable for late binding
   - Do not require perfmon_capable to expose free memory at page
     granularity. Handle it like other drm drivers do
   - Fix lock handling on suspend error path
   - Fix I2C controller resume after S3

  v3d:
   - fix fence locking"

* tag 'drm-next-2025-10-11-1' of https://gitlab.freedesktop.org/drm/kernel: (34 commits)
  drm/amd/display: Incorrect Mirror Cositing
  drm/amd/display: Enable Dynamic DTBCLK Switch
  drm/amdgpu: Report individual reset error
  drm/amdgpu: partially revert "revert to old status lock handling v3"
  drm/amd/display: Fix unsafe uses of kernel mode FPU
  drm/amd/pm: Disable VCN queue reset on SMU v13.0.6 due to regression
  drm/amdgpu: Fix general protection fault in amdgpu_vm_bo_reset_state_machine
  drm/amdgpu: Check swus/ds for switch state save
  drm/amdkfd: Fix two comments in kfd_ioctl.h
  drm/amd/pm: Avoid interface mismatch messaging
  drm/amdgpu: Merge amdgpu_vm_set_pasid into amdgpu_vm_init
  drm/amd/amdgpu: Fix the mes version that support inv_tlbs
  drm/amd: Check whether secure display TA loaded successfully
  drm/amdkfd: Fix mmap write lock not release
  drm/amdkfd: Fix kfd process ref leaking when userptr unmapping
  drm/amdgpu: Fix for GPU reset being blocked by KIQ I/O.
  drm/amd/display: Disable scaling on DCE6 for now
  drm/amd/display: Properly disable scaling on DCE6
  drm/amd/display: Properly clear SCL_*_FILTER_CONTROL on DCE6
  drm/amd/display: Add missing DCE6 SCL_HORZ_FILTER_INIT* SRIs
  ...

Merge tag 'drm-fixes-2025-10-11' of https://gitlab.freedesktop.org/drm/kernel

Pull drm fixes from Dave Airlie:
"Some fixes leftover from our fixes branch, just nouveau and vmwgfx:

  nouveau:
   - Return errno code from TTM move helper

  vmwgfx:
   - Fix null-ptr access in cursor code
   - Fix UAF in validation
   - Use correct iterator in validation"

* tag 'drm-fixes-2025-10-11' of https://gitlab.freedesktop.org/drm/kernel:
  drm/nouveau: fix bad ret code in nouveau_bo_move_prep
  drm/vmwgfx: Fix copy-paste typo in validation
  drm/vmwgfx: Fix Use-after-free in validation
  drm/vmwgfx: Fix a null-ptr access in the cursor snooper

Merge tag 'drm-misc-fixes-2025-10-09' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes

Short summary of fixes pull:

nouveau:
- Return errno code from TTM move helper

vmwgfx:
- Fix null-ptr access in cursor code
- Fix UAF in validation
- Use correct iterator in validation

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20251009120004.GA17570@linux.fritz.box

Merge tag 'devicetree-fixes-for-6.18-1' of git://git./linux/kernel/git/robh/linux

Pull devicetree fixes from Rob Herring:

- Allow child nodes on renesas-bsc bus binding

- Drop node name pattern on allwinner,sun50i-a64-de2 bus binding

- Switch DT patchwork to kernel.org from ozlabs.org

- Fix some typos in docs and bindings

- Fix reference count in PCI node unittest

* tag 'devicetree-fixes-for-6.18-1' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
  dt-bindings: bus: renesas-bsc: allow additional properties
  dt-bindings: bus: allwinner,sun50i-a64-de2: don't check node names
  MAINTAINERS: Move DT patchwork to kernel.org
  of: unittest: Fix device reference count leak in of_unittest_pci_node_verify
  of: doc: Fix typo in doc comments.
  dt-bindings: mmc: Correct typo "upto" to "up to"

dt-bindings: bus: renesas-bsc: allow additional properties

Allow additional properties to enable devices attached to the bus.
Fixes warnings like these:

arch/arm/boot/dts/renesas/sh73a0-kzm9g.dtb: bus@fec10000 (renesas,bsc-sh73a0): Unevaluated properties are not allowed ('ethernet@10000000' was unexpected)
arch/arm/boot/dts/renesas/r8a73a4-ape6evm.dtb: bus@fec10000 (renesas,bsc-r8a73a4): Unevaluated properties are not allowed ('ethernet@8000000', 'flash@0' were unexpected)

Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>

dt-bindings: bus: allwinner,sun50i-a64-de2: don't check node names

Node names are already and properly checked by the core schema. No need
to do it again.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
[robh: Also drop [A-F] in unit address]
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>

Merge tag 'ceph-for-6.18-rc1' of https://github.com/ceph/ceph-client

Pull ceph updates from Ilya Dryomov:

- some messenger improvements (Eric and Max)

- address an issue (also affected userspace) of incorrect permissions
   being granted to users who have access to multiple different CephFS
   instances within the same cluster (Kotresh)

- a bunch of assorted CephFS fixes (Slava)

* tag 'ceph-for-6.18-rc1' of https://github.com/ceph/ceph-client:
  ceph: add bug tracking system info to MAINTAINERS
  ceph: fix multifs mds auth caps issue
  ceph: cleanup in ceph_alloc_readdir_reply_buffer()
  ceph: fix potential NULL dereference issue in ceph_fill_trace()
  libceph: add empty check to ceph_con_get_out_msg()
  libceph: pass the message pointer instead of loading con->out_msg
  libceph: make ceph_con_get_out_msg() return the message pointer
  ceph: fix potential race condition on operations with CEPH_I_ODIRECT flag
  ceph: refactor wake_up_bit() pattern of calling
  ceph: fix potential race condition in ceph_ioctl_lazyio()
  ceph: fix overflowed constant issue in ceph_do_objects_copy()
  ceph: fix wrong sizeof argument issue in register_session()
  ceph: add checking of wait_for_completion_killable() return value
  ceph: make ceph_start_io_*() killable
  libceph: Use HMAC-SHA256 library instead of crypto_shash

Merge tag 'v6.18-rc-part2-smb-client-fixes' of git://git.samba.org/sfrench/cifs-2.6

Pull more smb client updates from Steve French:

- fix i_size in fallocate

- two truncate fixes

- utime fix

- minor cleanups

- SMB1 fixes

- improve error check in read

- improve perf of copy file_range (copy_chunk)

* tag 'v6.18-rc-part2-smb-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: update internal version number
  cifs: Add comments for DeletePending assignments in open functions
  cifs: Add fallback code path for cifs_mkdir_setinfo()
  cifs: Allow fallback code in smb_set_file_info() also for directories
  cifs: Query EA $LXMOD in cifs_query_path_info() for WSL reparse points
  smb: client: remove cfids_invalidation_worker
  smb: client: remove redudant assignment in cifs_strict_fsync()
  smb: client: fix race with fallocate(2) and AIO+DIO
  smb: client: fix missing timestamp updates after utime(2)
  smb: client: fix missing timestamp updates after ftruncate(2)
  smb: client: fix missing timestamp updates with O_TRUNC
  cifs: Fix copy_to_iter return value check
  smb: client: batch SRV_COPYCHUNK entries to cut round trips
  smb: client: Omit an if branch in smb2_find_smb_tcon()
  smb: client: Return directly after a failed genlmsg_new() in cifs_swn_send_register_message()
  smb: client: Use common code in cifs_do_create()
  smb: client: Improve unlocking of a mutex in cifs_get_swn_reg()
  smb: client: Return a status code only as a constant in cifs_spnego_key_instantiate()
  smb: client: Use common code in cifs_lookup()
  smb: client: Reduce the scopes for a few variables in two functions

Merge tag 'xtensa-20251010' of https://github.com/jcmvbkbc/linux-xtensa

Pull Xtensa updates from Max Filippov:

- minor cleanups

* tag 'xtensa-20251010' of https://github.com/jcmvbkbc/linux-xtensa:
xtensa: use HZ_PER_MHZ in platform_calibrate_ccount
xtensa: simdisk: add input size check in proc_write_simdisk

Merge tag 'block-6.18-20251009' of git://git./linux/kernel/git/axboe/linux

Pull block fixes from Jens Axboe:

- Don't include __GFP_NOWARN for loop worker allocation, as it already
   uses GFP_NOWAIT which has __GFP_NOWARN set already

- Small series cleaning up the recent bio_iov_iter_get_pages() changes

- loop fix for leaking the backing reference file, if validation fails

- Update of a comment pertaining to disk/partition stat locking

* tag 'block-6.18-20251009' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
  loop: remove redundant __GFP_NOWARN flag
  block: move bio_iov_iter_get_bdev_pages to block/fops.c
  iomap: open code bio_iov_iter_get_bdev_pages
  block: rename bio_iov_iter_get_pages_aligned to bio_iov_iter_get_pages
  block: remove bio_iov_iter_get_pages
  block: Update a comment of disk statistics
  loop: fix backing file reference leak on validation error

Merge tag 'io_uring-6.18-20251009' of git://git./linux/kernel/git/axboe/linux

Pull io_uring fixes from Jens Axboe:

- Fixup indentation in the UAPI header

- Two fixes for zcrx. One fixes receiving too much in some cases, and
   the other deals with not correctly incrementing the source in the
   fallback copy loop

- Fix for a race in the IORING_OP_WAITID command, where there was a
   small window where the request would be left on the wait_queue_head
   list even though it was being canceled/completed

- Update liburing git URL in the kernel tree

* tag 'io_uring-6.18-20251009' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
  io_uring/zcrx: increment fallback loop src offset
  io_uring/zcrx: fix overshooting recv limit
  io_uring: use tab indentation for IORING_SEND_VECTORIZED comment
  io_uring/waitid: always prune wait queue entry in io_waitid_wait()
  io_uring: update liburing git URL

Merge patch series "kbuild: Fixes for fallout from recent modules.builtin.modinfo series"

This is a series to address some problems that were exposed by the
recent modules.builtin.modinfo series that landed in commit c7d3dd9163e6
("Merge patch series "Add generated modalias to
modules.builtin.modinfo"").

The third patch is not directly related to the aforementioned series, as
the warning it fixes happens prior to the series but commit 8d18ef04f940
("s390: vmlinux.lds.S: Reorder sections") from the series creates
conflicts in this area, so I included it here.

Link: https://patch.msgid.link/20251008-kbuild-fix-modinfo-regressions-v1-0-9fc776c5887c@kernel.org
Signed-off-by: Nathan Chancellor <nathan@kernel.org>

s390/vmlinux.lds.S: Move .vmlinux.info to end of allocatable sections

When building s390 defconfig with binutils older than 2.32, there are
several warnings during the final linking stage:

  s390-linux-ld: .tmp_vmlinux1: warning: allocated section `.got.plt' not in segment
  s390-linux-ld: .tmp_vmlinux2: warning: allocated section `.got.plt' not in segment
  s390-linux-ld: vmlinux.unstripped: warning: allocated section `.got.plt' not in segment
  s390-linux-objcopy: vmlinux: warning: allocated section `.got.plt' not in segment
  s390-linux-objcopy: st7afZyb: warning: allocated section `.got.plt' not in segment

binutils commit afca762f598 ("S/390: Improve partial relro support for
64 bit") [1] in 2.32 changed where .got.plt is emitted, avoiding the
warning.

The :NONE in the .vmlinux.info output section description changes the
segment for subsequent allocated sections. Move .vmlinux.info right
above the discards section to place all other sections in the previously
defined segment, .data.

Fixes: 30226853d6ec ("s390: vmlinux.lds.S: explicitly handle '.got' and '.plt' sections")
Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=afca762f598d453c563f244cd3777715b1a0cb72
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Acked-by: Alexey Gladkov <legion@kernel.org>
Acked-by: Nicolas Schier <nsc@kernel.org>
Link: https://patch.msgid.link/20251008-kbuild-fix-modinfo-regressions-v1-3-9fc776c5887c@kernel.org
Signed-off-by: Nathan Chancellor <nathan@kernel.org>

kbuild: Add '.rel.*' strip pattern for vmlinux

Prior to binutils commit c12d9fa2afe ("Support objcopy
--remove-section=.relaFOO") [1] in 2.32, stripping relocation sections
required the trailing period (i.e., '.rel.*') to work properly.

After commit 3e86e4d74c04 ("kbuild: keep .modinfo section in
vmlinux.unstripped"), there is an error with binutils 2.31.1 or earlier
because these sections are not properly removed:

s390-linux-objcopy: st6tO8Ev: symbol `.modinfo' required but not present
s390-linux-objcopy:st6tO8Ev: no symbols

Add the old pattern to resolve this issue (along with a comment to allow
cleaning this when binutils 2.32 or newer is the minimum supported
version). While the aforementioned kbuild change exposes this, the
pattern was originally changed by commit 71d815bf5dfd ("kbuild: Strip
runtime const RELA sections correctly"), where it would still be
incorrect with binutils older than 2.32.

Fixes: 71d815bf5dfd ("kbuild: Strip runtime const RELA sections correctly")
Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=c12d9fa2afe7abcbe407a00e15719e1a1350c2a7
Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
Closes: https://lore.kernel.org/CA+G9fYvVktRhFtZXdNgVOL8j+ArsJDpvMLgCitaQvQmCx=hwOQ@mail.gmail.com/
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Acked-by: Alexey Gladkov <legion@kernel.org>
Acked-by: Nicolas Schier <nsc@kernel.org>
Link: https://patch.msgid.link/20251008-kbuild-fix-modinfo-regressions-v1-2-9fc776c5887c@kernel.org
Signed-off-by: Nathan Chancellor <nathan@kernel.org>

kbuild: Restore pattern to avoid stripping .rela.dyn from vmlinux

Commit 0ce5139fd96e ("kbuild: always create intermediate
vmlinux.unstripped") removed the pattern to avoid stripping .rela.dyn
sections added by commit e9d86b8e17e7 ("scripts: Do not strip .rela.dyn
section"). Restore it so that .rela.dyn sections remain in the final
vmlinux.

Fixes: 0ce5139fd96e ("kbuild: always create intermediate vmlinux.unstripped")
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Acked-by: Alexey Gladkov <legion@kernel.org>
Acked-by: Nicolas Schier <nsc@kernel.org>
Link: https://patch.msgid.link/20251008-kbuild-fix-modinfo-regressions-v1-1-9fc776c5887c@kernel.org
Signed-off-by: Nathan Chancellor <nathan@kernel.org>

Merge branch 'bpf-avoid-rcu-context-warning-when-unpinning-htab-with-internal-structs'

KaFai Wan says:

====================
bpf: Avoid RCU context warning when unpinning htab with internal structs

This small patchset is about avoid RCU context warning when unpinning
htab with internal structs (timer, workqueue, or task_work).

v3:
  - fix nit (Yonghong Song)
  - add Acked-by: Yonghong Song <yonghong.song@linux.dev>

v2:
  - rename bpf_free_inode() to bpf_destroy_inode() (Andrii)
https://lore.kernel.org/all/20251007012235.755853-1-kafai.wan@linux.dev/

v1:
https://lore.kernel.org/all/20251003084528.502518-1-kafai.wan@linux.dev/
---
====================

Link: https://patch.msgid.link/20251008102628.808045-1-kafai.wan@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

selftests/bpf: Add test for unpinning htab with internal timer struct

Add test to verify that unpinning hash tables containing internal timer
structures does not trigger context warnings.

Each subtest (timer_prealloc and timer_no_prealloc) can trigger the
context warning when unpinning, but the warning cannot be triggered
twice within a short time interval (a HZ), which is expected behavior.

Signed-off-by: KaFai Wan <kafai.wan@linux.dev>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20251008102628.808045-3-kafai.wan@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

bpf: Avoid RCU context warning when unpinning htab with internal structs

When unpinning a BPF hash table (htab or htab_lru) that contains internal
structures (timer, workqueue, or task_work) in its values, a BUG warning
is triggered:
BUG: sleeping function called from invalid context at kernel/bpf/hashtab.c:244
in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 14, name: ksoftirqd/0
...

The issue arises from the interaction between BPF object unpinning and
RCU callback mechanisms:
1. BPF object unpinning uses ->free_inode() which schedules cleanup via
   call_rcu(), deferring the actual freeing to an RCU callback that
   executes within the RCU_SOFTIRQ context.
2. During cleanup of hash tables containing internal structures,
   htab_map_free_internal_structs() is invoked, which includes
   cond_resched() or cond_resched_rcu() calls to yield the CPU during
   potentially long operations.

However, cond_resched() or cond_resched_rcu() cannot be safely called from
atomic RCU softirq context, leading to the BUG warning when attempting
to reschedule.

Fix this by changing from ->free_inode() to ->destroy_inode() and rename
bpf_free_inode() to bpf_destroy_inode() for BPF objects (prog, map, link).
This allows direct inode freeing without RCU callback scheduling,
avoiding the invalid context warning.

Reported-by: Le Chen <tom2cat@sjtu.edu.cn>
Closes: https://lore.kernel.org/all/1444123482.1827743.1750996347470.JavaMail.zimbra@sjtu.edu.cn/
Fixes: 68134668c17f ("bpf: Add map side support for bpf timers.")
Suggested-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: KaFai Wan <kafai.wan@linux.dev>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20251008102628.808045-2-kafai.wan@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

xsk: Harden userspace-supplied xdp_desc validation

Turned out certain clearly invalid values passed in xdp_desc from
userspace can pass xp_{,un}aligned_validate_desc() and then lead
to UBs or just invalid frames to be queued for xmit.

desc->len close to ``U32_MAX`` with a non-zero pool->tx_metadata_len
can cause positive integer overflow and wraparound, the same way low
enough desc->addr with a non-zero pool->tx_metadata_len can cause
negative integer overflow. Both scenarios can then pass the
validation successfully.
This doesn't happen with valid XSk applications, but can be used
to perform attacks.

Always promote desc->len to ``u64`` first to exclude positive
overflows of it. Use explicit check_{add,sub}_overflow() when
validating desc->addr (which is ``u64`` already).

bloat-o-meter reports a little growth of the code size:

add/remove: 0/0 grow/shrink: 2/1 up/down: 60/-16 (44)
Function                                     old     new   delta
xskq_cons_peek_desc                          299     330     +31
xsk_tx_peek_release_desc_batch               973    1002     +29
xsk_generic_xmit                            3148    3132     -16

but hopefully this doesn't hurt the performance much.

Fixes: 341ac980eab9 ("xsk: Support tx_metadata_len")
Cc: stable@vger.kernel.org # 6.8+
Signed-off-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20251008165659.4141318-1-aleksander.lobakin@intel.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Merge tag 'parisc-for-6.18-rc1' of git://git./linux/kernel/git/deller/parisc-linux

Pull parisc updates from Helge Deller:
"Minor enhancements and fixes, specifically:

   - report emulation and alignment faults via perf

   - add initial kernel-side support for perf_events

   - small initialization fixes in the parisc firmware layer

   - adjust TC* constants and avoid referencing termio structs to avoid
     userspace build errors"

* tag 'parisc-for-6.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Fix iodc and device path return values on old machines
  parisc: Firmware: Fix returned path for PDC_MODULE_FIND on older machines
  parisc: Add initial kernel-side perf_event support
  parisc: Report software alignment faults via perf
  parisc: Report emulation faults via perf
  parisc: don't reference obsolete termio struct for TC* constants
  parisc: Remove spurious if statement from raw_copy_from_user()

Merge tag 'sound-fix-6.18-rc1' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
"A few more small fixes for 6.18-rc1.

  Most of changes are about ASoC Intel and SOF drivers, while a few
  other device-specific fixes are found for HD-audio, USB-audio, ASoC
  RT722VB and Meson"

* tag 'sound-fix-6.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ASoC: rt722: add settings for rt722VB
  ASoC: meson: aiu-encoder-i2s: fix bit clock polarity
  ALSA: usb: fpc: replace kmalloc_array followed by copy_from_user with memdup_array_user
  ALSA: hda/tas2781: Enable init_profile_id for device initialization
  ALSA: emu10k1: Fix typo in docs
  ALSA: hda/realtek: Add quirk for ASUS ROG Zephyrus Duo
  ASoC: SOF: Intel: Read the LLP via the associated Link DMA channel
  ASoC: SOF: ipc4-pcm: do not report invalid delay values
  ASoC: SOF: sof-audio: add dev_dbg_ratelimited wrapper
  ASoC: SOF: Intel: hda-pcm: Place the constraint on period time instead of buffer time
  ASoC: SOF: ipc4-topology: Account for different ChainDMA host buffer size
  ASoC: SOF: ipc4-topology: Correct the minimum host DMA buffer size
  ASoC: SOF: ipc4-pcm: fix start offset calculation for chain DMA
  ASoC: SOF: ipc4-pcm: fix delay calculation when DSP resamples
  ASoC: SOF: ipc3-topology: Fix multi-core and static pipelines tear down
  ALSA: hda/hdmi: Add pin fix for HP ProDesk model

Merge tag 'fbdev-for-6.18-rc1' of git://git./linux/kernel/git/deller/linux-fbdev

Pull fbdev updates from Helge Deller:
"Beside the usual bunch of smaller bug fixes, the majority of changes
  were by Zsolt Kajtar to improve the s3fb driver.

  Bug fixes:
   - Bounds checking to fix vmalloc-out-of-bounds (Albin Babu Varghese)
   - Fix logic error in "offb" name match (Finn Thain)
   - simplefb: Fix use after free in (Janne Grunau)
   - s3fb: Various fixes and powersave improvements (Zsolt Kajtar)

  Enhancements & code cleanups:
   - Various fixes in the documentation (Bagas Sanjaya)
   - Use string choices helpers (Chelsy Ratnawat)
   - xenfb: Use vmalloc_array to simplify code (Qianfeng Rong)
   - mb862xxfb: use signed type for error codes (Qianfeng Rong)
   - Make drivers depend on LCD_CLASS_DEVICE (Thomas Zimmermann)
   - radeonfb: Remove stale product link in Kconfig (Sukrut Heroorkar)"

* tag 'fbdev-for-6.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/linux-fbdev:
  fbdev: Fix logic error in "offb" name match
  fbdev: Add bounds checking in bit_putcs to fix vmalloc-out-of-bounds
  fbdev: Make drivers depend on LCD_CLASS_DEVICE
  fbdev: radeonfb: Remove stale product link in Kconfig
  Documentation: fb: Retitle driver docs
  Documentation: fb: ep93xx: Demote section headings
  Documentation: fb: Split toctree
  fbdev: simplefb: Fix use after free in simplefb_detach_genpds()
  fbdev: s3fb: Revert mclk stop in suspend
  fbdev: mb862xxfb: Use int type to store negative error codes
  fbdev: Use string choices helpers
  fbdev: core: Fix ubsan warning in pixel_to_pat
  fbdev: s3fb: Implement 1 and 2 BPP modes, improve 4 BPP
  fbdev: s3fb: Implement powersave for S3 FB
  fbdev: xenfb: Use vmalloc_array to simplify code

Merge tag 'gpio-fixes-for-v6.18-rc1' of git://git./linux/kernel/git/brgl/linux

Pull gpio fixes from Bartosz Golaszewski:

- add a missing ACPI ID for MTL-CVF devices in gpio-usbio

- mark the gpio-wcd934x controller as "sleeping" as it uses a mutex for
   locking internally

* tag 'gpio-fixes-for-v6.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
  gpio: wcd934x: mark the GPIO controller as sleeping
  gpio: usbio: Add ACPI device-id for MTL-CVF devices

Merge tag 'ntb-6.18' of https://github.com/jonmason/ntb

Pull NTB updates from Jon Mason:

- Add support for Renesas R-Car and allow arbitrary BAR mapping in EPF

- Update ntb_hw_amd to support the latest generation secondary topology
   and add a new maintainer

- Fix a bug by adding a mutex to ensure `link_event_callback` executes
   sequentially

* tag 'ntb-6.18' of https://github.com/jonmason/ntb:
  NTB: epf: Add Renesas rcar support
  NTB: epf: Allow arbitrary BAR mapping
  ntb: Add mutex to make link_event_callback executed linearly.
  MAINTAINERS: Update for the NTB AMD driver maintainer
  ntb_hw_amd: Update amd_ntb_get_link_status to support latest generation secondary topology

Merge tag 'i2c-for-6.18-rc1-part2' of git://git./linux/kernel/git/wsa/linux

Pull more i2c updates from Wolfram Sang:

- Second part of rtl9300 updates since dependencies are in now:
    - general cleanups
    - implement block read/write support
    - add RTL9310 support

- DT schema conversion of hix5hd2 binding

- namespace cleanup for i2c-algo-pca

- minor simplification for mt65xx

* tag 'i2c-for-6.18-rc1-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  dt-bindings: i2c: hisilicon,hix5hd2: convert to DT schema
  i2c: mt65xx: convert set_speed function to void
  i2c: rename wait_for_completion callback to wait_for_completion_cb
  i2c: rtl9300: add support for RTL9310 I2C controller
  dt-bindings: i2c: realtek,rtl9301-i2c: extend for RTL9310 support
  i2c: rtl9300: use scoped guard instead of explicit lock/unlock
  i2c: rtl9300: separate xfer configuration and execution
  i2c: rtl9300: do not set read mode on every transfer
  i2c: rtl9300: move setting SCL frequency to config_io
  i2c: rtl9300: rename internal sda_pin to sda_num
  dt-bindings: i2c: realtek,rtl9301-i2c: fix wording and typos
  i2c: rtl9300: use regmap fields and API for registers
  i2c: rtl9300: Implement I2C block read and write

cifs: update internal version number

to 2.57

Signed-off-by: Steve French <stfrench@microsoft.com>

Merge tag 'v6.18-p3' of git://git./linux/kernel/git/herbert/crypto-2.6

Pull crypto fixes from Herbert Xu:

- Fix bug in crypto_skcipher that breaks the new ti driver

- Check for invalid assoclen in essiv

* tag 'v6.18-p3' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: essiv - Check ssize for decryption and in-place encryption
crypto: skcipher - Fix reqsize handling

Merge tag 'tpmdd-next-v6.18-2' of git://git./linux/kernel/git/jarkko/linux-tpmdd

Pull tpm updates from Jarkko Sakkinen:

- Disable TCG_TPM2_HMAC from defconfig

   It causes performance issues, and breaks some atypical
   configurations.

- simplify code using the new crypto library

- misc fixes and cleanups

* tag 'tpmdd-next-v6.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd:
  tpm: Prevent local DOS via tpm/tpm0/ppi/*operations
  tpm: use a map for tpm2_calc_ordinal_duration()
  tpm_tis: Fix incorrect arguments in tpm_tis_probe_irq_single
  tpm: Use HMAC-SHA256 library instead of open-coded HMAC
  tpm: Compare HMAC values in constant time
  tpm: Disable TPM2_TCG_HMAC by default

MAINTAINERS: Move DT patchwork to kernel.org

The ozlabs.org PW instance is slow due to being geographically far away
from any of the maintainers and seems to have gotten slower as of late
(AI scrapers perhaps). The kernel.org PW also has some additional
features (i.e. pwbot) we want to use.

DT core patches also go into PW, so add the PW link for it.

Signed-off-by: Rob Herring (Arm) <robh@kernel.org>

gpio: wcd934x: mark the GPIO controller as sleeping

The slimbus regmap passed to the GPIO driver down from MFD does not use
fast_io. This means a mutex is used for locking and thus this GPIO chip
must not be used in atomic context. Change the can_sleep switch in
struct gpio_chip to true.

Fixes: 59c324683400 ("gpio: wcd934x: Add support to wcd934x gpio controller")
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

tpm: Prevent local DOS via tpm/tpm0/ppi/*operations

Reads on tpm/tpm0/ppi/*operations can become very long on
misconfigured systems. Reading the TPM is a blocking operation,
thus a user could effectively trigger a DOS.

Resolve this by caching the results and avoiding the blocking
operations after the first read.

[ jarkko: fixed atomic sleep:
sed -i 's/spin_/mutex_/g' drivers/char/tpm/tpm_ppi.c
sed -i 's/DEFINE_SPINLOCK/DEFINE_MUTEX/g' drivers/char/tpm/tpm_ppi.c ]

Signed-off-by: Denis Aleksandrov <daleksan@redhat.com>
Reported-by: Jan Stancek <jstancek@redhat.com>
Closes: https://lore.kernel.org/linux-integrity/20250915210829.6661-1-daleksan@redhat.com/T/#u
Suggested-by: Jarkko Sakkinen <jarkko@kernel.org>
Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>

tpm: use a map for tpm2_calc_ordinal_duration()

The current shenanigans for duration calculation introduce too much
complexity for a trivial problem, and further the code is hard to patch and
maintain.

Address these issues with a flat look-up table, which is easy to understand
and patch. If leaf driver specific patching is required in future, it is
easy enough to make a copy of this table during driver initialization and
add the chip parameter back.

'chip->duration' is retained for TPM 1.x.

As the first entry for this new behavior address TCG spec update mentioned
in this issue:

https://github.com/raspberrypi/linux/issues/7054

Therefore, for TPM_SelfTest the duration is set to 3000 ms.

This does not categorize a as bug, given that this is introduced to the
spec after the feature was originally made.

Reviewed-by: Serge Hallyn <serge@hallyn.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>

tpm_tis: Fix incorrect arguments in tpm_tis_probe_irq_single

The tpm_tis_write8() call specifies arguments in wrong order. Should be
(data, addr, value) not (data, value, addr). The initial correct order
was changed during the major refactoring when the code was split.

Fixes: 41a5e1cf1fe1 ("tpm/tpm_tis: Split tpm_tis driver into a core and TCG TIS compliant phy")
Signed-off-by: Gunnar Kudrjavets <gunnarku@amazon.com>
Reviewed-by: Justinien Bouron <jbouron@amazon.com>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>

tpm: Use HMAC-SHA256 library instead of open-coded HMAC

Now that there are easy-to-use HMAC-SHA256 library functions, use these
in tpm2-sessions.c instead of open-coding the HMAC algorithm.

Note that the new implementation correctly handles keys longer than 64
bytes (SHA256_BLOCK_SIZE), whereas the old implementation handled such
keys incorrectly. But it doesn't appear that such keys were being used.

Signed-off-by: Eric Biggers <ebiggers@kernel.org>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>

tpm: Compare HMAC values in constant time

In tpm_buf_check_hmac_response(), compare the HMAC values in constant
time using crypto_memneq() instead of in variable time using memcmp().

This is worthwhile to follow best practices and to be consistent with
MAC comparisons elsewhere in the kernel. However, in this driver the
side channel seems to have been benign: the HMAC input data is
guaranteed to always be unique, which makes the usual MAC forgery via
timing side channel not possible. Specifically, the HMAC input data in
tpm_buf_check_hmac_response() includes the "our_nonce" field, which was
generated by the kernel earlier, remains under the control of the
kernel, and is unique for each call to tpm_buf_check_hmac_response().

Signed-off-by: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>

tpm: Disable TPM2_TCG_HMAC by default

After reading all the feedback, right now disabling the TPM2_TCG_HMAC
is the right call.

Other views discussed:

A. Having a kernel command-line parameter or refining the feature
   otherwise. This goes to the area of improvements.  E.g., one
   example is my own idea where the null key specific code would be
   replaced with a persistent handle parameter (which can be
   *unambigously* defined as part of attestation process when
   done correctly).

B. Removing the code. I don't buy this because that is same as saying
   that HMAC encryption cannot work at all (if really nitpicking) in
   any form. Also I disagree on the view that the feature could not
   be refined to something more reasoable.

Also, both A and B are worst options in terms of backporting.

Thuss, this is the best possible choice.

Cc: stable@vger.kernel.or # v6.10+
Fixes: d2add27cf2b8 ("tpm: Add NULL primary creation")
Suggested-by: Chris Fenner <cfenn@google.com>
Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>

cifs: Add comments for DeletePending assignments in open functions

On more places is set DeletePending member to 0. Add comments why is 0 the
correct value. Paths in DELETE_PENDING state cannot be opened by new calls.
So if the newly issued open for that path succeed then it means that the
path cannot be in DELETE_PENDING state.

Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>

cifs: Add fallback code path for cifs_mkdir_setinfo()

Use SMBSetInformation() as a fallback function (when CIFSSMBSetPathInfo()
fails) which can set attribudes on the directory, including changing
read-only attribute.

Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>

cifs: Allow fallback code in smb_set_file_info() also for directories

On NT systems, it is possible to do SMB open call also for directories.
Open argument CREATE_NOT_DIR disallows opening directories. So in fallback
code path in smb_set_file_info() remove CREATE_NOT_DIR restriction to allow
it also for directories.

Similar fallback is implemented also in CIFSSMBSetPathInfoFB() function and
this function already allows to call operation for directories.

Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>

cifs: Query EA $LXMOD in cifs_query_path_info() for WSL reparse points

EA $LXMOD is required for WSL non-symlink reparse points.

Fixes: ef86ab131d91 ("cifs: Fix querying of WSL CHR and BLK reparse points over SMB1")
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>

fbdev: Fix logic error in "offb" name match

A regression was reported to me recently whereby /dev/fb0 had disappeared
from a PowerBook G3 Series "Wallstreet". The problem shows up when the
"video=ofonly" parameter is passed to the kernel, which is what the
bootloader does when "no video driver" is selected. The cause of the
problem is the "offb" string comparison, which got mangled when it got
refactored. Fix it.

Cc: stable@vger.kernel.org
Fixes: 93604a5ade3a ("fbdev: Handle video= parameter in video/cmdline.c")
Reported-and-tested-by: Stan Johnson <userm57@yahoo.com>
Signed-off-by: Finn Thain <fthain@linux-m68k.org>
Signed-off-by: Helge Deller <deller@gmx.de>

parisc: Fix iodc and device path return values on old machines

Older machines may not fully initialize the return values when asking for IODC
and device path data when building the inventory. Work around possible
firmware leaks by proper initialization of the variables.

Signed-off-by: Helge Deller <deller@gmx.de>

parisc: Firmware: Fix returned path for PDC_MODULE_FIND on older machines

Older machines (like my 715/64) don't correctly initialize the
device path when returning from the PDC_MODULE_FIND firmware call.
Work around that shortcoming by initializing the path with the
known values.

Signed-off-by: Helge Deller <deller@gmx.de>

rtc: interface: Ensure alarm irq is enabled when UIE is enabled

When setting a normal alarm, user-space is responsible for using
RTC_AIE_ON/RTC_AIE_OFF to control if alarm irq should be enabled.

But when RTC_UIE_ON is used, interrupts must be enabled so that the
requested irq events are generated.
When RTC_UIE_OFF is used, alarm irq is disabled if there are no other
alarms queued, so this commit brings symmetry to that.

Signed-off-by: Esben Haabendal <esben@geanix.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20250516-rtc-uie-irq-fixes-v2-5-3de8e530a39e@geanix.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

rtc: tps6586x: Fix initial enable_irq/disable_irq balance

Interrupts are automatically enabled when requested, so we need to
initialize irq_en accordingly to avoid causing an unbalanced enable
warning.

Signed-off-by: Esben Haabendal <esben@geanix.com>
Link: https://lore.kernel.org/r/20250516-rtc-uie-irq-fixes-v2-4-3de8e530a39e@geanix.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

rtc: cpcap: Fix initial enable_irq/disable_irq balance

Interrupts are automatically enabled when requested, so we need to
initialize alarm_enabled accordingly to avoid causing an unbalanced enable
warning.

Signed-off-by: Esben Haabendal <esben@geanix.com>
Link: https://lore.kernel.org/r/20250516-rtc-uie-irq-fixes-v2-3-3de8e530a39e@geanix.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

rtc: isl12022: Fix initial enable_irq/disable_irq balance

Interrupts are automatically enabled when requested, so we need to
initialize irq_enabled accordingly to avoid causing an unbalanced enable
warning.

Fixes: c62d658e5253 ("rtc: isl12022: Add alarm support")
Signed-off-by: Esben Haabendal <esben@geanix.com>
Link: https://lore.kernel.org/r/20250516-rtc-uie-irq-fixes-v2-2-3de8e530a39e@geanix.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

rtc: interface: Fix long-standing race when setting alarm

As described in the old comment dating back to
commit 6610e0893b8b ("RTC: Rework RTC code to use timerqueue for events")
from 2010, we have been living with a race window when setting alarm
with an expiry in the near future (i.e. next second).
With 1 second resolution, it can happen that the second ticks after the
check for the timer having expired, but before the alarm is actually set.
When this happen, no alarm IRQ is generated, at least not with some RTC
chips (isl12022 is an example of this).

With UIE RTC timer being implemented on top of alarm irq, being re-armed
every second, UIE will occasionally fail to work, as an alarm irq lost
due to this race will stop the re-arming loop.

For now, I have limited the additional expiry check to only be done for
alarms set to next seconds. I expect it should be good enough, although I
don't know if we can now for sure that systems with loads could end up
causing the same problems for alarms set 2 seconds or even longer in the
future.

I haven't been able to reproduce the problem with this check in place.

Cc: stable@vger.kernel.org
Signed-off-by: Esben Haabendal <esben@geanix.com>
Link: https://lore.kernel.org/r/20250516-rtc-uie-irq-fixes-v2-1-3de8e530a39e@geanix.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Merge tag 'amd-drm-next-6.18-2025-10-09' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.18-2025-10-09:

amdgpu:
- DC DCE6 fixes
- GPU reset fixes
- Secure diplay messaging cleanup
- MES fix
- GPUVM locking fixes
- PMFW messaging cleanup
- PCI US/DS switch handling fix
- VCN queue reset fix
- DC FPU handling fix
- DCN 3.5 fix
- DC mirroring fix

amdkfd:
- Fix kfd process ref leak
- mmap write lock handling fix
- Fix comments in IOCTL

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20251009162915.981503-1-alexander.deucher@amd.com