Merge tag 'alpha-for-v7.1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/lindholm/alpha
Pull alpha updates from Magnus Lindholm:
"One fix to silence pgprot_modify() compiler warnings, and one patch
adding SECCOMP/SECCOMP_FILTER support together with the syscall and
ptrace fixes needed for it"
* tag 'alpha-for-v7.1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/lindholm/alpha:
alpha: Define pgprot_modify to silence tautological comparison warnings
alpha: add support for SECCOMP and SECCOMP_FILTER
Michael Margolin [Thu, 16 Apr 2026 20:14:08 +0000 (20:14 +0000)]
RDMA/core: Fix user CQ creation for drivers without create_cq
CQ creation is failing for drivers that only implement create_user_cq
(e.g. EFA), when buffer isn't provided by userspace. This because of a
leftover check that requires create_cq existence in such case.
Remove the create_cq existence check from the no-buffer path. The
buffer is optional and drivers that handle their own memory should work
through create_user_cq regardless.
Matt Roper [Thu, 16 Apr 2026 21:17:09 +0000 (14:17 -0700)]
drm/xe: Drop skip_mtcfg descriptor flag
The skip_mtcfg descriptor flag is unused and expected to remain that
way. Drop it.
Single-tile platforms are already identified by a zero/unset value for
max_remote_tiles and don't need/use this flag to avoid trying to read
out multi-tile configuration. PVC is currently the only multi-tile
platform, and PVC uses MTCFG so this flag is not set. The current
expectation is that if/when future multi-tile platforms show up, they
will also use the MTCFG register in the same manner as PVC, meaning that
they won't have any need to set 'skip_mtcfg' either.
Even if a future platform does change how multi-tile configuration gets
probed (e.g., using some different register), simply doing an early return
from xe_info_probe_tile_count() would probably not be the correct logic
to handle that anyway.
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull kvm updates from Paolo Bonzini:
"Arm:
- Add support for tracing in the standalone EL2 hypervisor code,
which should help both debugging and performance analysis. This
uses the new infrastructure for 'remote' trace buffers that can be
exposed by non-kernel entities such as firmware, and which came
through the tracing tree
- Add support for GICv5 Per Processor Interrupts (PPIs), as the
starting point for supporting the new GIC architecture in KVM
- Finally add support for pKVM protected guests, where pages are
unmapped from the host as they are faulted into the guest and can
be shared back from the guest using pKVM hypercalls. Protected
guests are created using a new machine type identifier. As the
elusive guestmem has not yet delivered on its promises, anonymous
memory is also supported
This is only a first step towards full isolation from the host; for
example, the CPU register state and DMA accesses are not yet
isolated. Because this does not really yet bring fully what it
promises, it is hidden behind CONFIG_ARM_PKVM_GUEST +
'kvm-arm.mode=protected', and also triggers TAINT_USER when a VM is
created. Caveat emptor
- Rework the dreaded user_mem_abort() function to make it more
maintainable, reducing the amount of state being exposed to the
various helpers and rendering a substantial amount of state
immutable
- Expand the Stage-2 page table dumper to support NV shadow page
tables on a per-VM basis
- Tidy up the pKVM PSCI proxy code to be slightly less hard to
follow
- Fix both SPE and TRBE in non-VHE configurations so that they do not
generate spurious, out of context table walks that ultimately lead
to very bad HW lockups
- A small set of patches fixing the Stage-2 MMU freeing in error
cases
- Tighten-up accepted SMC immediate value to be only #0 for host
SMCCC calls
- The usual cleanups and other selftest churn
LoongArch:
- Use CSR_CRMD_PLV for kvm_arch_vcpu_in_kernel()
- Add DMSINTC irqchip in kernel support
RISC-V:
- Fix steal time shared memory alignment checks
- Fix vector context allocation leak
- Fix array out-of-bounds in pmu_ctr_read() and pmu_fw_ctr_read_hi()
- Fix double-free of sdata in kvm_pmu_clear_snapshot_area()
- Fix integer overflow in kvm_pmu_validate_counter_mask()
- Fix shift-out-of-bounds in make_xfence_request()
- Fix lost write protection on huge pages during dirty logging
- Split huge pages during fault handling for dirty logging
- Skip CSR restore if VCPU is reloaded on the same core
- Implement kvm_arch_has_default_irqchip() for KVM selftests
- Factored-out ISA checks into separate sources
- Added hideleg to struct kvm_vcpu_config
- Factored-out VCPU config into separate sources
- Support configuration of per-VM HGATP mode from KVM user space
s390:
- Support for ESA (31-bit) guests inside nested hypervisors
- Remove restriction on memslot alignment, which is not needed
anymore with the new gmap code
- Fix LPSW/E to update the bear (which of course is the breaking
event address register)
x86:
- Shut up various UBSAN warnings on reading module parameter before
they were initialized
- Don't zero-allocate page tables that are used for splitting
hugepages in the TDP MMU, as KVM is guaranteed to set all SPTEs in
the page table and thus write all bytes
- As an optimization, bail early when trying to unsync 4KiB mappings
if the target gfn can just be mapped with a 2MiB hugepage
x86 generic:
- Copy single-chunk MMIO write values into struct kvm_vcpu (more
precisely struct kvm_mmio_fragment) to fix use-after-free stack
bugs where KVM would dereference stack pointer after an exit to
userspace
- Clean up and comment the emulated MMIO code to try to make it
easier to maintain (not necessarily "easy", but "easier")
- Move VMXON+VMXOFF and EFER.SVME toggling out of KVM (not *all* of
VMX and SVM enabling) as it is needed for trusted I/O
- Advertise support for AVX512 Bit Matrix Multiply (BMM) instructions
- Immediately fail the build if a required #define is missing in one
of KVM's headers that is included multiple times
- Reject SET_GUEST_DEBUG with -EBUSY if there's an already injected
exception, mostly to prevent syzkaller from abusing the uAPI to
trigger WARNs, but also because it can help prevent userspace from
unintentionally crashing the VM
- Exempt SMM from CPUID faulting on Intel, as per the spec
- Misc hardening and cleanup changes
x86 (AMD):
- Fix and optimize IRQ window inhibit handling for AVIC; make it
per-vCPU so that KVM doesn't prematurely re-enable AVIC if multiple
vCPUs have to-be-injected IRQs
- Clean up and optimize the OSVW handling, avoiding a bug in which
KVM would overwrite state when enabling virtualization on multiple
CPUs in parallel. This should not be a problem because OSVW should
usually be the same for all CPUs
- Drop a WARN in KVM_MEMORY_ENCRYPT_REG_REGION where KVM complains
about a "too large" size based purely on user input
- Clean up and harden the pinning code for KVM_MEMORY_ENCRYPT_REG_REGION
- Disallow synchronizing a VMSA of an already-launched/encrypted
vCPU, as doing so for an SNP guest will crash the host due to an
RMP violation page fault
- Overhaul KVM's APIs for detecting SEV+ guests so that VM-scoped
queries are required to hold kvm->lock, and enforce it by lockdep.
Fix various bugs where sev_guest() was not ensured to be stable for
the whole duration of a function or ioctl
- Convert a pile of kvm->lock SEV code to guard()
- Play nicer with userspace that does not enable
KVM_CAP_EXCEPTION_PAYLOAD, for which KVM needs to set CR2 and DR6
as a response to ioctls such as KVM_GET_VCPU_EVENTS (even if the
payload would end up in EXITINFO2 rather than CR2, for example).
Only set CR2 and DR6 when consumption of the payload is imminent,
but on the other hand force delivery of the payload in all paths
where userspace retrieves CR2 or DR6
- Use vcpu->arch.cr2 when updating vmcb12's CR2 on nested #VMEXIT
instead of vmcb02->save.cr2. The value is out of sync after a
save/restore or after a #PF is injected into L2
- Fix a class of nSVM bugs where some fields written by the CPU are
not synchronized from vmcb02 to cached vmcb12 after VMRUN, and so
are not up-to-date when saved by KVM_GET_NESTED_STATE
- Fix a class of bugs where the ordering between KVM_SET_NESTED_STATE
and KVM_SET_{S}REGS could cause vmcb02 to be incorrectly
initialized after save+restore
- Add a variety of missing nSVM consistency checks
- Fix several bugs where KVM failed to correctly update VMCB fields
on nested #VMEXIT
- Fix several bugs where KVM failed to correctly synthesize #UD or
#GP for SVM-related instructions
- Add support for save+restore of virtualized LBRs (on SVM)
- Refactor various helpers and macros to improve clarity and
(hopefully) make the code easier to maintain
- Aggressively sanitize fields when copying from vmcb12, to guard
against unintentionally allowing L1 to utilize yet-to-be-defined
features
- Fix several bugs where KVM botched rAX legality checks when
emulating SVM instructions. There are remaining issues in that KVM
doesn't handle size prefix overrides for 64-bit guests
- Fail emulation of VMRUN/VMLOAD/VMSAVE if mapping vmcb12 fails
instead of somewhat arbitrarily synthesizing #GP (i.e. don't double
down on AMD's architectural but sketchy behavior of generating #GP
for "unsupported" addresses)
- Cache all used vmcb12 fields to further harden against TOCTOU bugs
x86 (Intel):
- Drop obsolete branch hint prefixes from the VMX instruction macros
- Use ASM_INPUT_RM() in __vmcs_writel() to coerce clang into using a
register input when appropriate
- Code cleanups
guest_memfd:
- Don't mark guest_memfd folios as accessed, as guest_memfd doesn't
support reclaim, the memory is unevictable, and there is no storage
to write back to
LoongArch selftests:
- Add KVM PMU test cases
s390 selftests:
- Enable more memory selftests
x86 selftests:
- Add support for Hygon CPUs in KVM selftests
- Fix a bug in the MSR test where it would get false failures on
AMD/Hygon CPUs with exactly one of RDPID or RDTSCP
- Add an MADV_COLLAPSE testcase for guest_memfd as a regression test
for a bug where the kernel would attempt to collapse guest_memfd
folios against KVM's will"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (373 commits)
KVM: x86: use inlines instead of macros for is_sev_*guest
x86/virt: Treat SVM as unsupported when running as an SEV+ guest
KVM: SEV: Goto an existing error label if charging misc_cg for an ASID fails
KVM: SVM: Move lock-protected allocation of SEV ASID into a separate helper
KVM: SEV: use mutex guard in snp_handle_guest_req()
KVM: SEV: use mutex guard in sev_mem_enc_unregister_region()
KVM: SEV: use mutex guard in sev_mem_enc_ioctl()
KVM: SEV: use mutex guard in snp_launch_update()
KVM: SEV: Assert that kvm->lock is held when querying SEV+ support
KVM: SEV: Document that checking for SEV+ guests when reclaiming memory is "safe"
KVM: SEV: Hide "struct kvm_sev_info" behind CONFIG_KVM_AMD_SEV=y
KVM: SEV: WARN on unhandled VM type when initializing VM
KVM: LoongArch: selftests: Add PMU overflow interrupt test
KVM: LoongArch: selftests: Add basic PMU event counting test
KVM: LoongArch: selftests: Add cpucfg read/write helpers
LoongArch: KVM: Add DMSINTC inject msi to vCPU
LoongArch: KVM: Add DMSINTC device support
LoongArch: KVM: Make vcpu_is_preempted() as a macro rather than function
LoongArch: KVM: Move host CSR_GSTAT save and restore in context switch
LoongArch: KVM: Move host CSR_EENTRY save and restore in context switch
...
drm/panel: visionox-rm69299: Make use of prepare_prev_first
The DSI link must be powered up to let panel driver to talk to the panel
during prepare() callback execution. Set the prepare_prev_first flag to
guarantee this.
Fixes: 9e15123eca79 ("drm/msm/dsi: Stop unconditionally powering up DSI hosts at modeset") Signed-off-by: Guido Günther <agx@sigxcpu.org> Signed-off-by: David Heidelberg <david@ixit.cz> Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org> Reviewed-by: Douglas Anderson <dianders@chromium.org> Signed-off-by: Douglas Anderson <dianders@chromium.org> Link: https://patch.msgid.link/20260417-axolotl-display-v2-1-8ce5341e46c2@ixit.cz
Guangshuo Li [Wed, 15 Apr 2026 17:05:15 +0000 (01:05 +0800)]
parisc: led: fix reference leak on failed device registration
When platform_device_register() fails in startup_leds(), the embedded
struct device in platform_leds has already been initialized by
device_initialize(), but the failure path only reports the error and
does not drop the device reference for the current platform device:
module.lds.S: Fix modules on 32-bit parisc architecture
On the 32-bit parisc architecture, we always used the
-ffunction-sections compiler option to tell the compiler to put the
functions into seperate text sections. This is necessary, otherwise
"big" kernel modules like ext4 or ipv6 fail to load because some
branches won't be able to reach their stubs.
Commit 1ba9f8979426 ("vmlinux.lds: Unify TEXT_MAIN, DATA_MAIN, and related
macros") broke this for parisc because all text sections will get
unconditionally merged now.
Introduce the ARCH_WANTS_MODULES_TEXT_SECTIONS config option which
avoids the text section merge for modules, and fix this issue by
enabling this option by default for 32-bit parisc.
Fixes: 1ba9f8979426 ("vmlinux.lds: Unify TEXT_MAIN, DATA_MAIN, and related macros") Cc: Josh Poimboeuf <jpoimboe@kernel.org> Cc: stable@vger.kernel.org # v6.19+ Suggested-by: Sami Tolvanen <samitolvanen@google.com> Reviewed-by: Petr Pavlu <petr.pavlu@suse.com> Signed-off-by: Helge Deller <deller@gmx.de>
parisc: Fix signal code to depend on CONFIG_COMPAT instead of CONFIG_64BIT
The signal handler code used CONFIG_64BIT to decide if compat handling
code should be compiled in. Fix it to use CONFIG_COMPAT instead.
This allows to disable CONFIG_COMPAT even when running a 64-bit kernel.
Zen1's hardware divider can leave, under certain circumstances, partial
results from previous operations. Those results can be leaked by
another, attacker thread.
Remove member no longer used by the scheduler core.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Lucas Stach <l.stach@pengutronix.de> Cc: Russell King <linux+etnaviv@armlinux.org.uk> Cc: Christian Gmeiner <christian.gmeiner@gmail.com> Cc: etnaviv@lists.freedesktop.org Reviewed-by: Christian Gmeiner <cgmeiner@igalia.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-20-tvrtko.ursulin@igalia.com
Remove member no longer used by the scheduler core.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: amd-gfx@lists.freedesktop.org Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-19-tvrtko.ursulin@igalia.com
drm/sched: Embed run queue singleton into the scheduler
Now that the run queue to scheduler relationship is always 1:1 we can
embed it (the run queue) directly in the scheduler struct and save on
some allocation error handling code and such.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-15-tvrtko.ursulin@igalia.com
drm/sched: Remove FIFO and RR and simplify to a single run queue
Since the new FAIR policy is in general better than FIFO and almost as
good as round-robin in interactive use cases, plus the latter has not been
the default policy in a long time, we can afford to remove both and leave
just FAIR.
By doing so we can simplify the scheduler code by making the scheduler to
run queue relationship always 1:1 and remove some code.
Also, now that the FIFO policy is gone the tree of entities is not a FIFO
tree any more so rename it to just the tree.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-14-tvrtko.ursulin@igalia.com
FAIR policy works better than FIFO for all known use cases and either
matches or gets close to RR. Lets make it a default to improve the user
experience especially with interactive workloads competing with heavy
clients.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-13-tvrtko.ursulin@igalia.com
GPUs do not always implement preemption and DRM scheduler definitely
does not support it at the front end scheduling level. This means
execution quanta can be quite long and is controlled by userspace,
consequence of which is picking the "wrong" entity to run can have a
larger negative effect than it would have with a virtual runtime based CPU
scheduler.
Another important consideration is that rendering clients often have
shallow submission queues, meaning they will be entering and exiting the
scheduler's runnable queue often.
Relevant scenario here is what happens when an entity re-joins the
runnable queue with other entities already present. One cornerstone of the
virtual runtime algorithm is to let it re-join at the head and rely on the
virtual runtime accounting and timeslicing to sort it out.
However, as explained above, this may not work perfectly in the GPU world.
Entity could always get to overtake the existing entities, or not,
depending on the submission order and rbtree equal key insertion
behaviour.
Allow interactive jobs to overtake entities already queued up for the
limited case when interactive entity is re-joining the queue after
being idle.
This gives more opportunity for the compositors to have their rendering
executed before the GPU hogs even if they have been configured with the
same scheduling priority.
To classify a client as interactive we look at its average job duration
versus the average for the whole scheduler. We can track this easily by
plugging into the existing job runtime tracking and applying the
exponential moving average window on the past submissions. Then, all other
things being equal, we let the more interactive jobs go first.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-12-tvrtko.ursulin@igalia.com
The FAIR scheduling policy is built upon the same concepts as the well
known CFS CPU scheduler - entity run queue is sorted by the virtual GPU
time consumed by entities in a way that the entity with least vruntime
runs first.
It is able to avoid total priority starvation, which is one of the
problems with FIFO, and it also does not need for per priority run queues.
As it scales the actual GPU runtime by an exponential factor as the
priority decreases, the virtual runtime for low priority entities grows
faster than for normal priority, pushing them further down the runqueue
order for the same real GPU time spent.
Apart from this fundamental fairness, fair policy is especially strong in
oversubscription workloads where it is able to give more GPU time to short
and bursty workloads when they are running in parallel with GPU heavy
clients submitting deep job queues.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-11-tvrtko.ursulin@igalia.com
There is no need to keep entities with no jobs in the tree so lets remove
it once the last job is consumed. This keeps the tree smaller which is
nicer and more efficient as entities are removed and re-added on every
popped job.
Apart from that, the upcoming fair scheduling algorithm will rely on the
tree only containing runnable entities.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-10-tvrtko.ursulin@igalia.com
To implement fair scheduling we need a view into the GPU time consumed by
entities. Problem we have is that jobs and entities objects have decoupled
lifetimes, where at the point we have a view into accurate GPU time, we
cannot link back to the entity any longer.
Solve this by adding a light weight entity stats object which is reference
counted by both entity and the job and hence can safely be used from
either side.
With that, the only other thing we need is to add a helper for adding the
job's GPU time into the respective entity stats object, and call it once
the accurate GPU time has been calculated.
The most convenient place to do that is the free job worker for several
reasons. Doing the accounting from the job completion callback would mean
a few locks would need to become irq safe and we would also need to worry
about out of order completions (via dma_fence_is_signaled calls which we
cannot control). In-order completions are critical for GPU time accuracy
which is currently adjusted per fence in the free worker and requires
looking at the next job in the scheduler pending list. We would also need
to add a new lock to protect the scheduler average stats update.
In contrast to those complications, having the accounting done from the
free worker is serialized by definition and all the above complications
are avoided. Downside is there is potential for a time lag between job
completions and GPU time being accounted against the entity. Since that is
partly alleviated by batch processing the completed job queue, and the
scheduling algorithm does not attempt to be completely fair, which would
even be rather impossible to achieve in the GPU world with the current
DRM scheduler design and hardware with no or poor preemption support,
this downside is not considered critical. Plus, in practice the scheduler
is also affected by worker scheduling delays from other angles too. Not
least being able to promptly feed the GPU with new work.
We therefore choose the simple option and can later consider improving
upon it if the need arises.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-9-tvrtko.ursulin@igalia.com
To implement fair scheduling we will need as accurate as possible view
into per entity GPU time utilisation. Because sched fence execution time
are only adjusted for accuracy in the free worker we need to process
completed jobs as soon as possible so the metric is most up to date when
view from the submission side of things.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-8-tvrtko.ursulin@igalia.com
Round-robin being the non-default policy and unclear how much it is used,
we can notice that it can be implemented using the FIFO data structures if
we only invent a fake submit timestamp which is monotonically increasing
inside drm_sched_rq instances.
So instead of remembering which was the last entity the scheduler worker
picked we can simply bump the picked one to the bottom of the tree, which
ensures round-robin behaviour between all active queued jobs.
If the picked job was the last from a given entity, we remember the
assigned fake timestamp and use it to re-insert the job once it re-joins
the queue. This ensures the job neither overtakes all already queued jobs,
neither it goes last. Instead it keeps the position after the currently
queued jobs and before the ones which haven't yet been queued at the point
the entity left the queue.
Advantage is that we can consolidate to a single code path and remove a
bunch of code. Downside is round-robin mode now needs to lock on the job
pop path but that should not have a measurable performance impact.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-7-tvrtko.ursulin@igalia.com
Every 100ms for the duration of the test it logs how many jobs each
client had completed, prefixed by minimum, average and maximum numbers.
When finished overall average delta between max and min is output as a
rough indicator to scheduling fairness.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-6-tvrtko.ursulin@igalia.com
To make evaluating different scheduling policies easier (no need for
external benchmarks) and perfectly repeatable, lets add some synthetic
workloads built upon mock scheduler unit test infrastructure.
Focus is on two parallel clients (two threads) submitting different job
patterns and logging their progress and some overall metrics. This is
repeated for both scheduler credit limit 1 and 2.
There we have two clients represented in the two respective columns, with
their progress logged roughly every 100 milliseconds. The metrics are:
- pct - Percentage progress of the job submit part
- cps - Cycles per second
- qd - Queue depth - number of submitted unfinished jobs
The cycles per second metric is inherent to the fact that workload
patterns are a data driven cycling sequence of:
- Submit 1..N jobs
- Wait for Nth job to finish (optional)
- Sleep (optional)
- Repeat from start
In this particular example we have a normal priority and a low priority
client both spamming the scheduler with 8ms jobs with no sync and no
sleeping. Hence they build very deep queues and we can see how the low
priority client is completely starved until the normal finishes.
Note that the PCT and CPS metrics are irrelevant for "unsync" clients
since they manage to complete all of their cycles instantaneously.
In this case client one is submitting 3x 2.5ms jobs, waiting for the 3rd
and then sleeping for 2.5ms (in effect causing 75% GPU load, minus the
overheads). Second client is submitting 1ms jobs, waiting for each to
finish and sleeping for 9ms (effective 10% GPU load). Here we can see
the PCT and CPS reflecting real progress.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-5-tvrtko.ursulin@igalia.com
drm/sched: Move run queue related code into a separate file
Lets move all the code dealing with struct drm_sched_rq into a separate
compilation unit. Advantage being sched_main.c is left with a clearer set
of responsibilities.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Reviewed-by: Matthew Brost <matthew.brost@intel.com> # v1 Acked-by: Danilo Krummrich <dakr@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-4-tvrtko.ursulin@igalia.com
drm/sched: Disallow initializing entities with no schedulers
Since we have removed the case where amdgpu was initializing entitites
with either no schedulers on the list, or with a single NULL scheduler,
and there appears no other drivers which rely on this, we can simplify the
scheduler by explicitly rejecting that early.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Cc: Christian König <christian.koenig@amd.com> Cc: Danilo Krummrich <dakr@kernel.org> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Philipp Stanner <phasta@kernel.org> Reviewed-by: Christian König <christian.koenig@amd.com> Acked-by: Philipp Stanner <phasta@kernel.org> Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Philipp Stanner <phasta@kernel.org> Link: https://patch.msgid.link/20260417103744.76020-2-tvrtko.ursulin@igalia.com
Junrui Luo [Wed, 15 Apr 2026 09:26:55 +0000 (17:26 +0800)]
KVM: s390: pci: fix GAIT table indexing due to double-scaling pointer arithmetic
kvm_s390_pci_aif_enable(), kvm_s390_pci_aif_disable(), and
aen_host_forward() index the GAIT by manually multiplying the index
with sizeof(struct zpci_gaite).
Since aift->gait is already a struct zpci_gaite pointer, this
double-scales the offset, accessing element aisb*16 instead of aisb.
This causes out-of-bounds accesses when aisb >= 32 (with
ZPCI_NR_DEVICES=512)
Fix by removing the erroneous sizeof multiplication.
Fixes: 3c5a1b6f0a18 ("KVM: s390: pci: provide routines for enabling/disabling interrupt forwarding") Fixes: 73f91b004321 ("KVM: s390: pci: enable host forwarding of Adapter Event Notifications") Reported-by: Yuhao Jiang <danisjiang@gmail.com> Cc: stable@vger.kernel.org Signed-off-by: Junrui Luo <moonafterrain@outlook.com> Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com> Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com> Tested-by: Matthew Rosato <mjrosato@linux.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@linux.ibm.com>
regmap: ram: fix memory leaks in __regmap_init_ram() on error
Two allocations in __regmap_init_ram() are not cleaned up on failure.
If the kzalloc_objs() for data->written fails, data->read is returned
with no way for the caller to free it.
If __regmap_init() fails, neither data->read nor data->written is freed
because its error paths do not call bus->free_context() (which is
regmap_ram_free_context() here). Only regmap_exit() does, and that is
never reached on an init failure.
Free the allocated arrays before returning any error.
parisc: Drop ip_fast_csum() inline assembly implementation
The assembly code of ip_fast_csum() triggers unaligned access warnings
if the IP header isn't correctly aligned:
Kernel: unaligned access to 0x173d22e76 in inet_gro_receive+0xbc/0x2e8 (iir 0x0e8810b6)
Kernel: unaligned access to 0x173d22e7e in inet_gro_receive+0xc4/0x2e8 (iir 0x0e88109a)
Kernel: unaligned access to 0x173d22e82 in inet_gro_receive+0xc8/0x2e8 (iir 0x0e90109d)
Kernel: unaligned access to 0x173d22e7a in inet_gro_receive+0xd0/0x2e8 (iir 0x0e9810b8)
Kernel: unaligned access to 0x173d22e86 in inet_gro_receive+0xdc/0x2e8 (iir 0x0e8810b8)
We have the option to a) ignore the warnings, b) work around it by
adding more code to check for alignment, or c) to switch to the generic
implementation and rely on the compiler to optimize the code.
Let's go with c), because a) isn't nice, and b) would effectively lead
to an implementation which is basically equal to c).
Kexin Sun [Sat, 21 Mar 2026 10:58:31 +0000 (18:58 +0800)]
parisc: update outdated comments for renamed ccio_alloc_consistent()
The function ccio_alloc_consistent() was renamed to ccio_alloc() by commit 79387179e2e4 ("parisc: convert to dma_map_ops"). Update the three stale
references in ccio-dma.c.
Also replace the obsolete PCI_DMA_TODEVICE constant name with DMA_TO_DEVICE in
a nearby comment to match the code.
Marco Elver [Thu, 16 Apr 2026 13:25:07 +0000 (15:25 +0200)]
slub: fix data loss and overflow in krealloc()
Commit 2cd8231796b5 ("mm/slub: allow to set node and align in
k[v]realloc") introduced the ability to force a reallocation if the
original object does not satisfy new alignment or NUMA node, even when
the object is being shrunk.
This introduced two bugs in the reallocation fallback path:
1. Data loss during NUMA migration: The jump to 'alloc_new' happens
before 'ks' and 'orig_size' are initialized. As a result, the
memcpy() in the 'alloc_new' block would copy 0 bytes into the new
allocation.
2. Buffer overflow during shrinking: When shrinking an object while
forcing a new alignment, 'new_size' is smaller than the old size.
However, the memcpy() used the old size ('orig_size ?: ks'), leading
to an out-of-bounds write.
The same overflow bug exists in the kvrealloc() fallback path, where the
old bucket size ksize(p) is copied into the new buffer without being
bounded by the new size.
A simple reproducer:
// e.g. add to lkdtm as KREALLOC_SHRINK_OVERFLOW
while (1) {
void *p = kmalloc(128, GFP_KERNEL);
p = krealloc_node_align(p, 64, 256, GFP_KERNEL, NUMA_NO_NODE);
kfree(p);
}
demonstrates the issue:
==================================================================
BUG: KFENCE: out-of-bounds write in memcpy_orig+0x68/0x130
Out-of-bounds write at 0xffff8883ad757038 (120B right of kfence-#47):
memcpy_orig+0x68/0x130
krealloc_node_align_noprof+0x1c8/0x340
lkdtm_KREALLOC_SHRINK_OVERFLOW+0x8c/0xc0 [lkdtm]
lkdtm_do_action+0x3a/0x60 [lkdtm]
...
allocated by task 316 on cpu 7 at 97.680481s (0.021813s ago):
krealloc_node_align_noprof+0x19c/0x340
lkdtm_KREALLOC_SHRINK_OVERFLOW+0x8c/0xc0 [lkdtm]
lkdtm_do_action+0x3a/0x60 [lkdtm]
...
==================================================================
Fix it by moving the old size calculation to the top of __do_krealloc()
and bounding all copy lengths by the new allocation size.
Fixes: 2cd8231796b5 ("mm/slub: allow to set node and align in k[v]realloc") Cc: stable@vger.kernel.org Reported-by: https://sashiko.dev/#/patchset/20260415143735.2974230-1-elver%40google.com Signed-off-by: Marco Elver <elver@google.com> Link: https://patch.msgid.link/20260416132837.3787694-1-elver@google.com Reviewed-by: Harry Yoo (Oracle) <harry@kernel.org> Signed-off-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Starting one background YNL notification listener per peer back-to-back
can intermittently stall the test setup before the listeners even reach
the Python main function.
This was reproducible in a reduced test.sh setup-only loop: a single
listener stayed stable across repeated runs, while starting listeners
for all peers could hang early in the listener launch phase. Adding a
short delay between listener launches makes the listeners start cleanly
and eliminates the reproduced hangs in repeated normal and slow-runner
tests.
Serialize listener startup with a small sleep between setup_listener
calls.
Fixes: 77de28cd7cf1 ("selftests: ovpn: add notification parsing and matching") Signed-off-by: Ralf Lici <ralf@mandelbit.com> Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Ralf Lici [Mon, 23 Mar 2026 14:12:32 +0000 (15:12 +0100)]
selftests: ovpn: align command flow with TAP
Current tests do not properly adhere to the TAP infrastructure
therefore they do not properly report failures leading to hangs of
the CI machinery.
Restructure ovpn selftests into using the TAP infrastructure: split each
test in stages, execute stage bodies with fail-fast semantics, and emit
KTAP pass/fail for each stage.
Centralize behavior control in common.sh and makes the scripts use
dedicated wrappers for required-success, expected-failure, and non-fatal
commands. Also add the OVPN_VERBOSE mode that exposes captured command
output for debugging.
This way tests won't hang anymore in case of failure when executed
within the CI machinery.
This change also makes default OVPN_CLI and YNL resolution
independent from the caller CWD by anchoring both to COMMON_DIR, so
behavior is stable across direct execution and run_tests-style
execution.
Fixes: 959bc330a439 ("testing/selftests: add test tool and scripts for ovpn module") Signed-off-by: Ralf Lici <ralf@mandelbit.com> Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Ralf Lici [Fri, 20 Mar 2026 16:29:38 +0000 (17:29 +0100)]
selftests: ovpn: add prefix to helpers and shared variables
Current naming for shared variables, helpers and netnamespaces is
a bit unfortunate as it doesn't come with a clean prefix.
This showed to be problematic in case of name clashes with external
scripts or in case of abrupt test termination (hanging netns' weren't
easily reconducible to ovpn).
Rename common helper entry points and all shared globals in the ovpn
selftests to ovpn_ or OVPN_ names so test scripts and wrappers use a
single explicit prefix. Also rename the temporary network namespaces
created by the tests from peerN to ovpn_peerN. This makes leaked
namespaces easier to identify.
This is a mechanical refactor only, behavior is unchanged.
Fixes: 959bc330a439 ("testing/selftests: add test tool and scripts for ovpn module") Signed-off-by: Ralf Lici <ralf@mandelbit.com> Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Ralf Lici [Tue, 24 Mar 2026 14:54:18 +0000 (15:54 +0100)]
selftests: ovpn: flatten slurped notification JSON before filtering
Notification comparison uses jq -s, which slurps all inputs into an
array. Some inputs can be arrays themselves, and applying the .msg.peer
filter directly on those entries triggers jq type errors.
Expand any array-valued JSON items returned by jq -s before selecting
.msg.peer, so the filter handles both normal notification objects and []
entries without type errors.
Fixes: 77de28cd7cf1 ("selftests: ovpn: add notification parsing and matching") Signed-off-by: Ralf Lici <ralf@mandelbit.com> Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Ralf Lici [Tue, 24 Mar 2026 07:48:57 +0000 (08:48 +0100)]
selftests: ovpn: add nftables config dependencies for test-mark
test-mark.sh installs nftables rules in an inet/filter output chain and
verifies packet drops via nft counters. In vmksft this can fail when the
nftables core is not enabled by the ovpn selftest config.
Add the missing kernel options required by this test:
- CONFIG_NETFILTER
- CONFIG_NF_TABLES
- CONFIG_NF_TABLES_INET
Fixes: 7b80d8a33500 ("selftests: ovpn: add test for the FW mark feature") Reported-by: Jakub Kicinski <kuba@kernel.org> Closes: https://lore.kernel.org/all/20260319124114.42f91f72@kernel.org/ Signed-off-by: Ralf Lici <ralf@mandelbit.com> Signed-off-by: Antonio Quartulli <antonio@openvpn.net>
Giovanni Cabiddu [Thu, 16 Apr 2026 17:07:00 +0000 (18:07 +0100)]
crypto: acomp - fix wrong pointer stored by acomp_save_req()
acomp_save_req() stores &req->chain in req->base.data. When
acomp_reqchain_done() is invoked on asynchronous completion, it receives
&req->chain as the data argument but casts it directly to struct
acomp_req. Since data points to the chain member, all subsequent field
accesses are at a wrong offset, resulting in memory corruption.
The issue occurs when an asynchronous hardware implementation, such as
the QAT driver, completes a request that uses the DMA virtual address
interface (e.g. acomp_request_set_src_dma()). This combination causes
crypto_acomp_compress() to enter the acomp_do_req_chain() path, which
sets acomp_reqchain_done() as the completion callback via
acomp_save_req().
With KASAN enabled, this manifests as a general protection fault in
acomp_reqchain_done():
general protection fault, probably for non-canonical address 0xe000040000000000
KASAN: probably user-memory-access in range [0x0000400000000000-0x0000400000000007]
RIP: 0010:acomp_reqchain_done+0x15b/0x4e0
Call Trace:
<IRQ>
qat_comp_alg_callback+0x5d/0xa0 [intel_qat]
adf_ring_response_handler+0x376/0x8b0 [intel_qat]
adf_response_handler+0x60/0x170 [intel_qat]
tasklet_action_common+0x223/0x820
handle_softirqs+0x1ab/0x640
</IRQ>
Fix this by storing the request itself in req->base.data instead of
&req->chain, so that acomp_reqchain_done() receives the correct pointer.
Simplify acomp_restore_req() accordingly to access req->chain directly.
Fixes: 64929fe8c0a4 ("crypto: acomp - Remove request chaining") Cc: stable@vger.kernel.org Signed-off-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
ipv6: xfrm6: release dst on error in xfrm6_rcv_encap()
xfrm6_rcv_encap() performs an IPv6 route lookup when the skb does not
already have a dst attached. ip6_route_input_lookup() returns a
referenced dst entry even when the lookup resolves to an error route.
If dst->error is set, xfrm6_rcv_encap() drops the skb without attaching
the dst to the skb and without releasing the reference returned by the
lookup. Repeated packets hitting this path therefore leak dst entries.
Release the dst before jumping to the drop path.
Fixes: 0146dca70b87 ("xfrm: add support for UDPv6 encapsulation of ESP") Cc: stable@kernel.org Reported-by: Yifan Wu <yifanwucs@gmail.com> Reported-by: Juefei Pu <tomapufckgml@gmail.com> Co-developed-by: Yuan Tan <yuantan098@gmail.com> Signed-off-by: Yuan Tan <yuantan098@gmail.com> Suggested-by: Xin Liu <bird@lzu.edu.cn> Tested-by: Ruide Cao <caoruide123@gmail.com> Signed-off-by: Yilin Zhu <zylzyl2333@gmail.com> Signed-off-by: Ren Wei <n05ec@lzu.edu.cn> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Guangshuo Li [Wed, 15 Apr 2026 19:31:38 +0000 (03:31 +0800)]
ALSA: pcmtest: fix reference leak on failed device registration
When platform_device_register() fails in mod_init(), the embedded struct
device in pcmtst_pdev has already been initialized by
device_initialize(), but the failure path returns the error without
dropping the device reference for the current platform device:
Cássio Gabriel [Thu, 16 Apr 2026 13:24:40 +0000 (10:24 -0300)]
ALSA: 6fire: Fix input volume change detection
usb6fire_control_input_vol_put() stores the analog capture volume
as a signed offset in rt->input_vol[] (-15..+15), but it compares
the cached value against the user-visible mixer value (0..30)
before subtracting 15.
This mixes two domains in the change detection path. Since the
runtime is zero-initialized, the visible default is 15; writing 0
right after probe is ignored, while writing 15 is reported as a
change even though the cached value remains 0.
Normalize the user value before comparing it with the cached offset.
ALSA: usb-audio: Add quirk entries for NexiGo N930W webcam
The NexiGo N930W 60fps webcam (USB ID 3443:930d) hits the same
'cannot get freq at ep 0x84' error in snd-usb-audio as its sibling
N930AF (1bcf:2283). Without QUIRK_FLAG_GET_SAMPLE_RATE the ADC clock
is never configured and the microphone streams only zero samples.
Testing on Linux 6.17 with QUIRK_FLAG_GET_SAMPLE_RATE |
QUIRK_FLAG_MIC_RES_16 (via quirk_alias=3443930d:1bcf2283) confirmed
the microphone captures real audio after a cold USB re-enumeration.
Adding a native quirk_flags_table entry avoids the alias workaround.
Michal Wajdeczko [Thu, 16 Apr 2026 13:18:31 +0000 (15:18 +0200)]
drm/xe/pf: Fix VF's scheduling priority reporting
When preparing number of impacted VFs parameter for the reporting
helper function, we wrongly ended with adding +1 (representing PF)
twice, since local variable total_vfs was already adjusted. This
resulted in printing a message that was referring to an invalid VF:
Merge tag 'for-linus-iommufd' of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd
Pull iommufd updates from Jason Gunthorpe:
"Several fixes:
- Add missing static const
- Correct type 1 emulation for VFIO_CHECK_EXTENSION when no-iommu is
turned on
- Fix selftest memory leak and syzkaller splat
- Fix missed -EFAULT in fault reporting write() fops
- Fix a race where map/unmap with the internal IOVA allocator can
unmap things it should not"
* tag 'for-linus-iommufd' of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd:
iommufd: Fix a race with concurrent allocation and unmap
iommufd/selftest: Remove MOCK_IOMMUPT_AMDV1 format
iommufd: Fix return value of iommufd_fault_fops_write()
iommufd: update outdated comment for renamed iommufd_hw_pagetable_alloc()
iommufd/selftest: Fix page leaks in mock_viommu_{init,destroy}
iommufd: vfio compatibility extension check for noiommu mode
iommufd: Constify struct dma_buf_attach_ops
Paulo Alcantara [Fri, 17 Apr 2026 00:15:50 +0000 (21:15 -0300)]
smb: client: fix dir separator in SMB1 UNIX mounts
When calling cifs_mount_get_tcon() with SMB1 UNIX mounts,
@cifs_sb->mnt_cifs_flags needs to be read or updated only after
calling reset_cifs_unix_caps(), otherwise it might end up with missing
CIFS_MOUNT_POSIXACL and CIFS_MOUNT_POSIX_PATHS bits.
This fixes the wrong dir separator used in paths caused by the missing
CIFS_MOUNT_POSIX_PATHS bit in cifs_sb_info::mnt_cifs_flags.
Reported-by: "Kris Karas (Bug Reporting)" <bugs-a21@moonlit-rail.com> Closes: https://lore.kernel.org/r/f758f4ff-4d54-4244-931d-38f469c3ff14@moonlit-rail.com Fixes: 4fc3a433c139 ("smb: client: use atomic_t for mnt_cifs_flags") Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.org> Cc: David Howells <dhowells@redhat.com> Cc: linux-cifs@vger.kernel.org Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
Merge tag 'for-linus-fwctl' of git://git.kernel.org/pub/scm/linux/kernel/git/fwctl/fwctl
Pull fwctl updates from Jason Gunthorpe:
- New fwctl driver for Broadcom RDMA NICs
- Bug fix for non-modular builds
* tag 'for-linus-fwctl' of git://git.kernel.org/pub/scm/linux/kernel/git/fwctl/fwctl:
fwctl: Fix class init ordering to avoid NULL pointer dereference on device removal
fwctl/bnxt_fwctl: Add documentation entries
fwctl/bnxt_fwctl: Add bnxt fwctl device
fwctl/bnxt_en: Create an aux device for fwctl
fwctl/bnxt_en: Refactor aux bus functions to be more generic
fwctl/bnxt_en: Move common definitions to include/linux/bnxt/
Merge tag 'soc-arm-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Pull SoC ARM code updates from Arnd Bergmann:
"These are again very minimal updates:
- A workaround for firmware on Google Nexus 10
- A fix for early debugging on OMAP1
- A rework for Microchip SoC configuration
- Cleanups on OMAP2 an R-Car-Gen2"
* tag 'soc-arm-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
ARM: omap2: dead code cleanup in kconfig for ARCH_OMAP4
ARM: OMAP1: Fix DEBUG_LL and earlyprintk on OMAP16XX
arm64: Kconfig: provide a top-level switch for Microchip platforms
ARM: shmobile: rcar-gen2: Use of_phandle_args_equal() helper
ARM: omap: fix all kernel-doc warnings
ARM: omap2: Replace scnprintf with strscpy in omap3_cpuinfo
ARM: samsung: exynos5250: Allow CPU1 to boot
Merge tag 'soc-defconfig-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Pull SoC defconfig updates from Arnd Bergmann:
"As usual, we enable a number of additional device drivers as loadable
modules, to support the added platforms. The largest change this time
is for OMAP2/3, which were not that well supported in the generic
arm32 defconfig.
The Tegra SoC platforms are now enabled by default in Kconfig when
ARCH_TEGRA is enabled, which means the defconfig change is done at the
same time as the Kconfig change here"
* tag 'soc-defconfig-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (25 commits)
arch/arm: Drop CONFIG_FIRMWARE_EDID from defconfig files
arm64: defconfig: Enable DP83TG720 PHY driver
arm64: tegra: defconfig: Drop redundant ARCH_TEGRA_foo_SOC
ARM: tegra: defconfig: Drop redundant ARCH_TEGRA_foo_SOC
arm64: defconfig: enable pci-pwrctrl-generic as module
arm64: defconfig: Enable Lontium LT8713sx driver
arm64: defconfig: Enable Qualcomm Eliza SoC display clock controller
arm64: defconfig: enable IPQ5210 RDP504 base configs
arm64: defconfig: Enable Milos LPASS LPI pinctrl driver
arm64: defconfig: Enable Kaanapali clock controllers
arm64: defconfig: Enable configs for Arduino VENTUNO Q
arm64: defconfig: Enable Qualcomm Eliza basic resource providers
arm64: defconfig: Enable S5KJN1 camera sensor
arm64: defconfig: Enable configurations for Toradex Aquila AM69
arm64: defconfig: remove SENSORS_SA67MCU
arm64: defconfig: Enable Qualcomm WCD937x headphone codec as module
arm64: defconfig: Enable QCOMTEE module for QTEE-enabled Qualcomm SoCs
ARM: shmobile: defconfig: Refresh for v7.0-rc1
arm: multi_v7_defconfig: Enable more OMAP 3/4 related configs
ARM: multi_v7_defconfig: omap2plus_defconfig: Enable ITE IT66121 driver
...
Merge tag 'soc-drivers-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Pull SoC driver updates from Arnd Bergmann:
"The driver updates again are all over the place with many minor fixes
going into platform specific code. The most notable changes are:
- Support for Microchip pic64gx system controllers
- Work on cleaning up devicetree bindings for SoC drivers, and
converting them into the new format
- Lots of smaller changes for Qualcomm SoC drivers, including support
for a number of newly supported chips
- reset controller API cleanups and a new driver for Cix Sky1
- Reworks of the Tegra PMC and CBB drivers, along with a change to
how individual Tegra SoCs get selected in Kconfig and BPMP firmware
driver updates including a refresh of the ABI header to match the
version used by firmware
- STM32 updates to the firewall bus driver and support for the debug
bus through OP-TEE
- SCMI firmware driver improvements for reliability, in particular
for dealing with broken firmware interrupts
- Memory driver updates for Tegra, and a patch to remove the unused
Baikal T1 driver"
* tag 'soc-drivers-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (193 commits)
firmware: arm_ffa: Use the correct buffer size during RXTX_MAP
firmware: qcom: scm: Allow QSEECOM on Lenovo IdeaCentre Mini X
clk: spear: fix resource leak in clk_register_vco_pll()
reset: rzv2h-usb2phy: Add support for VBUS mux controller registration
reset: rzv2h-usb2phy: Convert to regmap API
dt-bindings: reset: renesas,rzv2h-usb2phy: Document RZ/G3E USB2PHY reset
dt-bindings: reset: renesas,rzv2h-usb2phy: Add '#mux-state-cells' property
soc: microchip: add mpfs gpio interrupt mux driver
dt-bindings: soc: microchip: document PolarFire SoC's gpio interrupt mux
gpio: mpfs: Add interrupt support
soc: qcom: ubwc: add helpers to get programmable values
soc: qcom: ubwc: add helper to get min_acc length
firmware: qcom: scm: Register gunyah watchdog device
soc: qcom: socinfo: Add SoC ID for SA8650P
dt-bindings: arm: qcom,ids: Add SoC ID for SA8650P
firmware: qcom: scm: Allow QSEECOM on Mahua CRD
soc: qcom: wcnss: simplify allocation of req
soc: qcom: pd-mapper: Add support for Eliza
soc: qcom: aoss: compare against normalized cooling state
soc: qcom: llcc: fix v1 SB syndrome register offset
...
Merge tag 'soc-dt-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Pull SoC devicetree updates from Arnd Bergmann:
"A number of SoC platforms are adding modernized variants of their
already supported chips time, with a total of 12 new SoCs, and two
older SoC getting removed:
- Qualcomm Glymur is a compute SoC using 18 Oryon-2 CPU cores
- Qualcomm Mahua is a variant of Glymur with only 12 CPU cores, but
largely identical.
- Qualcomm Eliza is an embeded platform for mobile phone (SM7750) and
IOT (QC7790S/M) workloads
- Qualcomm IPQ5210 is a wireless networking SoC using Cortex-A53
cores
- Qualcomm apq8084 and ipq806x had only rudimentary support but no
actual products using them, so they are now gone.
- Axis ARTPEC-9 is a follow-up to the ARTPEC-8 embedded SoC, using
the Samsung SoC platform but now with Cortex-A55 cores
- ARM Zena is a virtual platform in FVP using Cortex-A720AE cores,
with additional versions planned to be merged in the future.
- ARM corstone-1000-a320 is a reference platform for IOT, using
low-end Cortex-A320 cores
- Microchip LAN9691 is an updated 64-bit variant of the arm32 lan966x
series of networking SoCs
- Microchip PIC64GX is an embedded RISC-V chip using SIFIVE U54 CPU
cores
- Rockchip RV1103B is the low-end 32-bit single-core vision processor
- Renesas RZ/G3L (r9a08g046) is an industrial embedded chip using
Cortex-A55 cores, similar to the G3E and G3S variants we already
supported.
- NXP S32N79 is an automotive SoC using Cortex-A78AE cores, a
significant upgrade from the older S32V and S32G series
These all come with at least one reference board or an initial product
using these, in total there are 67 newly added boards. The ones for
already supported SoCs are:
- Two more Aspeed BMC based boards
- Three older tablets based on 32-bit OMAP4 and Exynos5 SoCs
- One Set-top-box based on Allwinner H6
- 22 additional industrial/embedded boards using 64-bit NXP i.MX8M or
i.MX9 SoCs
- 20 Qualcomm SoC based machines across all possible markets:
workstation, gaming, laptop, phone, networking, reference, ...
- Three more Rockchips rk35xx based boards
- Four variants of the Toradex Verdin using TI AM62
Other notable bits are:
- A cleanup for the 32-bit Tegra paz00 board moved the last board
specific code on Tegra into equivalent dts syntax.
- There continues to be a significant number of fixes for static
checking of dtc syntax, but it feels like this is slowing down,
hopefully getting into a state where most known issues are
addressed
- Additional hardware support for many existing boards across SoC
families, notably Qualcomm, Broadcom, i.MX2, i.MX6, Rockchips,
STM32, Mediatek, Tegra, TI and Microchip"
* tag 'soc-dt-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (841 commits)
arm64: dts: ti: k3: Use memory-region-names for r5f
ARM: dts: imx: Add DT overlays for DH i.MX6 DHCOM SoM and boards
ARM: dts: imx6sx: remove fallback compatible string fsl,imx28-lcdif
ARM: dts: imx25: rename node name tcq to touchscreen
ARM: dts: imx: b850v3: Disable unused usdhc4
ARM: dts: imx: b850v3: Define GPIO line names
ARM: dts: imx: b850v3: Use alphabetical sorting
ARM: dts: imx: bx50v3: Configure phy-mode to eliminate a warning
ARM: dts: imx: bx50v3: Configure switch PHY max-speed to 100Mbps
ARM: dts: imx7ulp: Add CPU clock and OPP table support
ARM: dts: imx7-mba7: Deassert BOOT_EN after boot
ARM: dts: tqma7: add boot phase properties
ARM: dts: imx7s: add boot phase properties
ARM: dts: tqma6ul[l]: correct spelling of TQ-Systems
ARM: dts: mba6ulx: add boot phase properties
ARM: dts: imx6ul[l]-tqma6ul[l]: add boot phase properties
ARM: dts: imx6ul/imx6ull: add boot phase properties
ARM: dts: imx6qdl-mba6: add boot phase properties
ARM: dts: imx6qdl-tqma6: add boot phase properties
ARM: dts: imx6qdl: add boot phase properties
...
Merge tag 'mm-nonmm-stable-2026-04-15-04-20' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Pull non-MM updates from Andrew Morton:
- "pid: make sub-init creation retryable" (Oleg Nesterov)
Make creation of init in a new namespace more robust by clearing away
some historical cruft which is no longer needed. Also some
documentation fixups
- "selftests/fchmodat2: Error handling and general" (Mark Brown)
Fix and a cleanup for the fchmodat2() syscall selftest
- "lib: polynomial: Move to math/ and clean up" (Andy Shevchenko)
- "hung_task: Provide runtime reset interface for hung task detector"
(Aaron Tomlin)
Give administrators the ability to zero out
/proc/sys/kernel/hung_task_detect_count
- "tools/getdelays: use the static UAPI headers from
tools/include/uapi" (Thomas Weißschuh)
Teach getdelays to use the in-kernel UAPI headers rather than the
system-provided ones
- "watchdog/hardlockup: Improvements to hardlockup" (Mayank Rungta)
Several cleanups and fixups to the hardlockup detector code and its
documentation
- "lib/bch: fix undefined behavior from signed left-shifts" (Josh Law)
A couple of small/theoretical fixes in the bch code
- "ocfs2/dlm: fix two bugs in dlm_match_regions()" (Junrui Luo)
- "cleanup the RAID5 XOR library" (Christoph Hellwig)
A quite far-reaching cleanup to this code. I can't do better than to
quote Christoph:
"The XOR library used for the RAID5 parity is a bit of a mess right
now. The main file sits in crypto/ despite not being cryptography
and not using the crypto API, with the generic implementations
sitting in include/asm-generic and the arch implementations
sitting in an asm/ header in theory. The latter doesn't work for
many cases, so architectures often build the code directly into
the core kernel, or create another module for the architecture
code.
Change this to a single module in lib/ that also contains the
architecture optimizations, similar to the library work Eric
Biggers has done for the CRC and crypto libraries later. After
that it changes to better calling conventions that allow for
smarter architecture implementations (although none is contained
here yet), and uses static_call to avoid indirection function call
overhead"
- "lib/list_sort: Clean up list_sort() scheduling workarounds"
(Kuan-Wei Chiu)
Clean up this library code by removing a hacky thing which was added
for UBIFS, which UBIFS doesn't actually need
- "Fix bugs in extract_iter_to_sg()" (Christian Ehrhardt)
Fix a few bugs in the scatterlist code, add in-kernel tests for the
now-fixed bugs and fix a leak in the test itself
- "kdump: Enable LUKS-encrypted dump target support in ARM64 and
PowerPC" (Coiby Xu)
Enable support of the LUKS-encrypted device dump target on arm64 and
powerpc
- "ocfs2: consolidate extent list validation into block read callbacks"
(Joseph Qi)
Cleanup, simplify, and make more robust ocfs2's validation of extent
list fields (Kernel test robot loves mounting corrupted fs images!)
* tag 'mm-nonmm-stable-2026-04-15-04-20' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (127 commits)
ocfs2: validate group add input before caching
ocfs2: validate bg_bits during freefrag scan
ocfs2: fix listxattr handling when the buffer is full
doc: watchdog: fix typos etc
update Sean's email address
ocfs2: use get_random_u32() where appropriate
ocfs2: split transactions in dio completion to avoid credit exhaustion
ocfs2: remove redundant l_next_free_rec check in __ocfs2_find_path()
ocfs2: validate extent block list fields during block read
ocfs2: remove empty extent list check in ocfs2_dx_dir_lookup_rec()
ocfs2: validate dx_root extent list fields during block read
ocfs2: fix use-after-free in ocfs2_fault() when VM_FAULT_RETRY
ocfs2: handle invalid dinode in ocfs2_group_extend
.get_maintainer.ignore: add Askar
ocfs2: validate bg_list extent bounds in discontig groups
checkpatch: exclude forward declarations of const structs
tools/accounting: handle truncated taskstats netlink messages
taskstats: set version in TGID exit notifications
ocfs2/heartbeat: fix slot mapping rollback leaks on error paths
arm64,ppc64le/kdump: pass dm-crypt keys to kdump kernel
...
====================
vsock/virtio: fix MSG_PEEK calculation on bytes to copy
`virtio_transport_stream_do_peek`, when calculating the number of bytes to
copy, didn't consider the `offset`, caused by partial reads that happened
before.
This might cause out-of-bounds read that lead to an EFAULT.
More details in the commits.
Commit 1 introduces the fix
Commit 2 introduces some preliminary work for adding a test and fixes a
problem in existing tests.
Commit 3 introduces a test that checks for this bug to avoid future
regressions.
For disclosure: this bug was found initially by claude opus 4.6, I then analyzed
it and worked on the fix and the test.
====================
Luigi Leonardi [Wed, 15 Apr 2026 15:09:30 +0000 (17:09 +0200)]
vsock/test: add MSG_PEEK after partial recv test
Add a test that verifies MSG_PEEK works correctly after a partial
recv().
This is to test a bug that was present in the
`virtio_transport_stream_do_peek()` when computing the number of bytes to
copy: After a partial read, the peek function didn't take into
consideration the number of bytes that were already read. So peeking the
whole buffer would cause an out-of-bounds read, that resulted in a -EFAULT.
This test does exactly this: do a partial recv on a buffer, then try to
peek the whole buffer content. The test re-uses
`test_stream_msg_peek_client()` to also cover this scenario.
Luigi Leonardi [Wed, 15 Apr 2026 15:09:29 +0000 (17:09 +0200)]
vsock/test: fix MSG_PEEK handling in recv_buf()
`recv_buf` does not handle the MSG_PEEK flag correctly: it keeps calling
`recv` until all requested bytes are available or an error occurs.
The problem is how it calculates the number of bytes read: MSG_PEEK
doesn't consume any bytes and will re-read the same bytes from the buffer
head, so summing the return value every time is wrong.
Moreover, MSG_PEEK doesn't consume the bytes in the buffer, so if more
bytes are requested than are available, the loop will never terminate,
because `recv` will never return EOF. For this reason, we need to compare
the number of bytes read with the number of bytes expected.
Add a check: if the MSG_PEEK flag is present, update the byte counter and
break out of the loop only after at least the expected number of bytes
have been received; otherwise, retry after a short delay to avoid
consuming too many CPU cycles.
This allows us to simplify the `test_stream_credit_update_test` by
reusing `recv_buf`, like some other tests already do.
Suggested-by: Stefano Garzarella <sgarzare@redhat.com> Signed-off-by: Luigi Leonardi <leonardi@redhat.com> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Link: https://patch.msgid.link/20260415-fix_peek-v4-2-8207e872759e@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Luigi Leonardi [Wed, 15 Apr 2026 15:09:28 +0000 (17:09 +0200)]
vsock/virtio: fix MSG_PEEK ignoring skb offset when calculating bytes to copy
`virtio_transport_stream_do_peek()` does not account for the skb offset
when computing the number of bytes to copy.
This means that, after a partial recv() that advances the offset, a peek
requesting more bytes than are available in the sk_buff causes
`skb_copy_datagram_iter()` to go past the valid payload, resulting in
a -EFAULT.
The dequeue path already handles this correctly.
Apply the same logic to the peek path.
Fixes: 0df7cd3c13e4 ("vsock/virtio/vhost: read data from non-linear skb") Reviewed-by: Stefano Garzarella <sgarzare@redhat.com> Acked-by: Arseniy Krasnov <avkrasnov@salutedevices.com> Signed-off-by: Luigi Leonardi <leonardi@redhat.com> Link: https://patch.msgid.link/20260415-fix_peek-v4-1-8207e872759e@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================
net: enetc: fix command BD ring issues
Currently, the implementation of command BD ring has two issues, one is
that the driver may obtain wrong consumer index of the ring, because the
driver does not mask out the SBE bit of the CIR value, so a wrong index
will be obtained when a SBE error ouccrs. The other one is that the DMA
buffer may be used after free. If netc_xmit_ntmp_cmd() times out and
returns an error, the pending command is not explicitly aborted, while
ntmp_free_data_mem() unconditionally frees the DMA buffer. If the buffer
has already been reallocated elsewhere, this may lead to silent memory
corruption. Because the hardware eventually processes the pending command
and perform a DMA write of the response to the physical address of the
freed buffer. So this patch set is to fix these two issues.
====================
The AI-generated review reported a potential DMA use-after-free issue
[1]. If netc_xmit_ntmp_cmd() times out and returns an error, the pending
command is not explicitly aborted, while ntmp_free_data_mem()
unconditionally frees the DMA buffer. If the buffer has already been
reallocated elsewhere, this may lead to silent memory corruption. Because
the hardware eventually processes the pending command and perform a DMA
write of the response to the physical address of the freed buffer.
To resolve this issue, this patch does the following modifications:
1. Convert cbdr->ring_lock from a spinlock to a mutex
The lock was originally a spinlock in case NTMP operations might be
invoked from atomic context. After downstream support for all NTMP
tables, no such usage has materialized. A mutex lock is now required
because the driver now needs to reclaim used BDs and release associated
DMA memory within the lock's context, while dma_free_coherent() might
sleep.
The hardware write-back overwrites the addr and len fields of the BD,
so the driver cannot rely on the hardware BD to free the associated DMA
memory. The driver now maintains a software shadow BD storing the DMA
buffer pointer, DMA address, and size. And netc_xmit_ntmp_cmd() only
reclaims older BDs when the number of used BDs reaches
NETC_CBDR_CLEAN_WORK (16). The software BD enables correct DMA memory
release. With this, struct ntmp_dma_buf and ntmp_free_data_mem() are no
longer needed and are removed.
3. Require callers to hold ring_lock across netc_xmit_ntmp_cmd()
netc_xmit_ntmp_cmd() releases the ring_lock before the caller finishes
consuming the response. At this point, if a concurrent thread submits
a new command, it may trigger ntmp_clean_cbdr() and free the DMA buffer
while it is still in use. Move ring_lock ownership to the caller to
ensure the response buffer cannot be reclaimed prematurely. So the
helpers ntmp_select_and_lock_cbdr() and ntmp_unlock_cbdr() are added.
These changes eliminate the DMA use-after-free condition and ensure safe
and consistent BD reclamation and DMA buffer lifecycle management.
net: enetc: correct the command BD ring consumer index
The command BD ring cousumer index register has the consumer index as
the lower 10 bits, and the bit 31 is SBE, which indicates whether a
system bus error occurred during execution of the CBD command. So if a
system bus error occurs, reading the register will get the SBE bit set.
However, the current implementation directly uses the register value as
the consumer index without masking it. Therefore, if a system bus error
occurs, an incorrect consumer index will be obtained, causing errors in
the processing of the command BD ring. Thus, we need to mask out the
other bits to obtain the correct consumer index.
In addition, this patch adds a check for the SBE bit after the polling
loop and returns an error if the bit is set.
net: pse-pd: fix out-of-bounds bitmap access in pse_isr() on 32-bit
In pse_isr(), notifs_mask was declared as a single unsigned long on the
stack (32 bits on 32-bit architectures). For PSE controllers with more
than 32 ports, this causes two problems:
- map_event callbacks could wrote bit positions >= 32 via
*notifs_mask |= BIT(i), which is undefined behaviour on a 32-bit
unsigned long and corrupts adjacent stack memory.
- for_each_set_bit(i, ¬ifs_mask, pcdev->nr_lines) treats
¬ifs_mask as a multi-word bitmap and reads beyond the single
unsigned long when nr_lines > BITS_PER_LONG.
Fix this by moving notifs_mask out of the stack and into struct pse_irq
as a dynamically allocated bitmap. It is sized with
BITS_TO_LONGS(pcdev->nr_lines) words in devm_pse_irq_helper(), so it
is always wide enough regardless of the host word size.
[Jakub]: No upstream driver currently supports >=32 ports.
Merge tag 'v7.1-rc1-part2-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Pull smb client updates from Steve French:
- Fix integer underflow in encrypted read
- Four debug patches, adding a few tracepoints
- Minor update to MAINTAINERS file (preferred server URL for cifs)
- Remove the BUG_ON() calls in d_mark_tmpfile_name
* tag 'v7.1-rc1-part2-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
MAINTAINERS: change git.samba.org to https
smb: client: fix integer underflow in receive_encrypted_read()
smb: client: add tracepoints for deferred handle caching
smb: client: add oplock level to smb3_open_done tracepoint
smb: client: add tracepoint for local lock conflicts
smb: client: add tracepoints for lock operations
vfs: get rid of BUG_ON() in d_mark_tmpfile_name()
net: dsa: remove redundant netdev_lock_ops() from conduit ethtool ops
DSA replaces the conduit (master) device's ethtool_ops with its own
wrappers that aggregate stats from both the conduit and DSA switch
ports. Taking the lock again inside the DSA wrappers causes a deadlock.
Stumbled upon this when booting qemu with fbnic and CONFIG_NET_DSA_LOOP=y
(which looks like some kind of testing device that auto-populates the ports
of eth0). `ethtool -i` is enough to deadlock. This means we have basically zero
coverage for DSA stuff with real ops locked devs.
Remove the redundant netdev_lock_ops()/netdev_unlock_ops() calls from
the DSA conduit ethtool wrappers.
Fixes: 2bcf4772e45a ("net: ethtool: try to protect all callback with netdev instance lock") Signed-off-by: Stanislav Fomichev <sdf@fomichev.me> Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com> Link: https://patch.msgid.link/20260414231035.1917035-1-sdf@fomichev.me Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net/sched: taprio: fix use-after-free in advance_sched() on schedule switch
In advance_sched(), when should_change_schedules() returns true,
switch_schedules() is called to promote the admin schedule to oper.
switch_schedules() queues the old oper schedule for RCU freeing via
call_rcu(), but 'next' still points into an entry of the old oper
schedule. The subsequent 'next->end_time = end_time' and
rcu_assign_pointer(q->current_entry, next) are use-after-free.
Fix this by selecting 'next' from the new oper schedule immediately
after switch_schedules(), and using its pre-calculated end_time.
setup_first_end_time() sets the first entry's end_time to
base_time + interval when the schedule is installed, so the value
is already correct.
The deleted 'end_time = sched_base_time(admin)' assignment was also
harmful independently: it would overwrite the new first entry's
pre-calculated end_time with just base_time.
Fixes: a3d43c0d56f1 ("taprio: Add support adding an admin schedule") Reported-by: Junxi Qian <qjx1298677004@gmail.com> Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net: mdio: MDIO_PIC64HPSC should depend on ARCH_MICROCHIP
The PIC64-HPSC/HX MDIO interface is only present on Microchip
PIC64-HPSC/HX SoCs. Hence add a dependency on ARCH_MICROCHIP, to
prevent asking the user about this driver when configuring a kernel
without Microchip SoC support.
Lorenzo Bianconi [Tue, 14 Apr 2026 14:08:52 +0000 (16:08 +0200)]
net: airoha: Wait for NPU PPE configuration to complete in airoha_ppe_offload_setup()
In order to properly enable flowtable hw offloading, poll
REG_PPE_FLOW_CFG register in airoha_ppe_offload_setup routine and
wait for NPU PPE configuration triggered by ppe_init callback to complete
before running airoha_ppe_hw_init().
Dave Airlie [Thu, 16 Apr 2026 21:32:30 +0000 (07:32 +1000)]
Merge tag 'topic/pipe-reorder-2026-04-15' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-next
drm/i915/display: change pipe allocation order for discrete platforms
This is a topic pull request for changing the pipe allocation order for
discrete platforms from the usual A,B,C,D to A,C,B,D. The goal is to
help pipe joiner configurations that reserve the adjacent pipe as the
secondary pipe without the user space knowing. More details in the
relevant commit message. The CRTC iteration is also changed to remain in
pipe order.
Jiri Olsa [Thu, 16 Apr 2026 10:00:34 +0000 (12:00 +0200)]
libbpf: Prevent double close and leak of btf objects
Sashiko found possible double close of btf object fd [1],
which happens when strdup in load_module_btfs fails at which
point the obj->btf_module_cnt is already incremented.
The error path close btf fd and so does later cleanup code in
bpf_object_post_load_cleanup function.
Also libbpf_ensure_mem failure leaves btf object not assigned
and it's leaked.
Replacing the err_out label with break to make the error path
less confusing as suggested by Alan.
Incrementing obj->btf_module_cnt only if there's no failure
and releasing btf object in error path.
====================
bpf: allow UTF-8 literals in bpf_bprintf_prepare()
bpf_bprintf_prepare() currently rejects any non-ASCII byte in format
strings, so helpers such as bpf_trace_printk() fail to emit UTF-8
literal text even when those bytes are not part of a format specifier.
Keep plain text permissive while continuing to parse '%' sequences as
ASCII-only. Patch 1 updates snprintf_negative() at the same time so the
selftests stay consistent during bisection. Patch 2 then extends
trace_printk coverage for both the valid UTF-8 literal case and the
invalid non-ASCII-after-'%' case.
Changes in v3:
- drop Suggested-by trailers and move review credit into this changelog
- update test_snprintf_negative() in patch 1/2 so plain non-ASCII text is
accepted while non-ASCII after '%' is still rejected, keeping
./test_progs -t snprintf aligned with the new behavior.
- clarify the trace_printk negative case with an explicit invalid format
string and comment
- address Paul Chaignon's review feedback and keep the negative coverage
requested earlier by Alan Maguire
Changes in v2:
- split the core change and selftest updates into two patches
- drop unnecessary isspace()/ispunct() casts
- add comments to clarify plain-text vs format-specifier handling
- add a negative selftest for non-ASCII bytes inside '%' sequences
Testing:
- Reproduced on x86_64 without the core fix: ASCII trace output works,
while UTF-8 literal text in bpf_trace_printk() is rejected and
produces no trace output
- Verified with tools/testing/selftests/bpf: ./test_progs -t trace_printk
- Verified with tools/testing/selftests/bpf: ./test_progs -t snprintf
====================
Extend trace_printk coverage to verify that UTF-8 literal text is
emitted successfully and that '%' parsing still rejects non-ASCII
bytes once format parsing starts.
Use an explicitly invalid format string for the negative case so the
ASCII-only parser expectation is visible from the test code itself.
bpf: allow UTF-8 literals in bpf_bprintf_prepare()
bpf_bprintf_prepare() only needs ASCII parsing for conversion
specifiers. Plain text can safely carry bytes >= 0x80, so allow
UTF-8 literals outside '%' sequences while keeping ASCII control
bytes rejected and format specifiers ASCII-only.
This keeps existing parsing rules for format directives unchanged,
while allowing helpers such as bpf_trace_printk() to emit UTF-8
literal text.
Update test_snprintf_negative() in the same commit so selftests keep
matching the new plain-text vs format-specifier split during bisection.
====================
bpf: Fix NULL deref when storing scalar into kptr slot
map_kptr_match_type() accesses reg->btf before confirming the register
is PTR_TO_BTF_ID. A scalar store into a kptr slot has no btf, causing
a NULL pointer dereference. Guard base_type() first.
bpf: Fix NULL deref in map_kptr_match_type for scalar regs
Commit ab6c637ad027 ("bpf: Fix a bpf_kptr_xchg() issue with local
kptr") refactored map_kptr_match_type() to branch on btf_is_kernel()
before checking base_type(). A scalar register stored into a kptr
slot has no btf, so the btf_is_kernel(reg->btf) call dereferences
NULL.
Move the base_type() != PTR_TO_BTF_ID guard before any reg->btf
access.
Fixes: ab6c637ad027 ("bpf: Fix a bpf_kptr_xchg() issue with local kptr") Reported-by: Hiker Cl <clhiker365@gmail.com> Closes: https://bugzilla.kernel.org/show_bug.cgi?id=221372 Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com> Acked-by: Paul Chaignon <paul.chaignon@gmail.com> Link: https://lore.kernel.org/r/20260416-kptr_crash-v1-1-5589356584b4@meta.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Tao Jiang [Wed, 15 Apr 2026 17:27:15 +0000 (01:27 +0800)]
nvme-pci: add quirk for Memblaze Pblaze5 (0x1c5f:0x0555)
The Memblaze Pblaze5 NVMe device (PCI ID 0x1c5f:0x0555)
is detected as a controller on recent kernels (tested on 5.15.85
and 6.8.4), but no namespace is exposed.
Tools like lsblk and fdisk do not report any block device.
dmesg shows:
nvme nvme0: missing or invalid SUBNQN field.
The device works correctly on older kernels (e.g. 4.19), suggesting
a compatibility issue with newer namespace handling.
This indicates the device does not properly support the
Namespace Descriptor List feature.
Applying NVME_QUIRK_NO_NS_DESC_LIST allows the namespace to be
discovered correctly.
Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com> Signed-off-by: Tao Jiang <tanroame.kyle@gmail.com> Signed-off-by: Keith Busch <kbusch@kernel.org>
projects / thirdparty / linux.git / log
