Commit 9c63e84db2 ("sched/core: Disable SD_PREFER_SIBLING on asymmetric
CPU capacity domains") removed the SD_PREFER_SIBLING from the domains with
asymmetric capacity. This was done to avoid spreading tasks to sibling
scheduling groups with less capacity, but this does not happen: checks for
capacity in update_sd_pick_busiest(), sched_balance_find_src_group(), and
sched_balance_find_src_rq() prevent migrations from high- to low-capacity
CPUs if the busiest group is not overloaded.
The cluster topology is a notable example: some systems have scheduling
domains spanning CPUs of asymmetric capacity, grouped into two or more
equal-capacity clusters sharing an L2 cache. When CONFIG_SCHED_CLUSTER is
enabled, SD_PREFER_SIBLING is needed in the domain to spread load across
these clusters.
CPUs with spare capacity, big or small, have always helped overloaded
groups. Once the overloading condition disappears, misfit load will still
be used to move high-utilization tasks to bigger CPUs if they have spare
capacity.
Adding the SD_PREFER_SIBLING flag shifts load balancing in shared-LLC
domains from equalizing the number of idle CPUs to equalizing the number
of running tasks. This enables migrations among clusters from newly-idle
load balance, where the outgoing task is already dequeued but the CPU
has not yet transitioned to idle.
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: Andrea Righi <arighi@nvidia.com>
Link: https://patch.msgid.link/20260720-rneri-fix-cas-clusters-v6-6-bb500bf4afd4@linux.intel.com
sched_balance_find_src_rq() avoids selecting a runqueue with a single
running task as busiest if doing so results in migrating the task to a
CPU with less than ~5% of extra capacity. It also unintentionally
prevents migrations between CPUs of identical capacity.
When CONFIG_SCHED_CLUSTER is enabled, load should be balanced across
clusters of CPUs with the same capacity. Allowing migration between CPUs
of identical capacity is necessary to meet this goal.
Use get_actual_cpu_capacity() to reflect architectural capacity as well
as diminished capacity due to hardware or cpufreq pressure. Guard this
check with the sched_cluster_active static key so that systems without
cluster topology are unaffected.
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: Andrea Righi <arighi@nvidia.com>
Link: https://patch.msgid.link/20260720-rneri-fix-cas-clusters-v6-5-bb500bf4afd4@linux.intel.com
In domains with asymmetric capacity, identifying misfit load in a
scheduling group is not useful when the destination CPU cannot help (i.e.,
its capacity exceeds the group's maximum CPU capacity by less than ~5%). In
such cases, it also prevents load balance among clusters of equal capacity
when CONFIG_SCHED_CLUSTER is enabled. This happens because
update_sd_pick_busiest() skips candidate groups of type misfit_task if the
destination CPU has similar capacity.
Skipping misfit load accounting in this situation allows the group to be
classified as has_spare or fully_busy and lets load balancing proceed. Keep
marking scheduling groups as overloaded when misfit tasks are present. The
sg_overloaded flag propagates to the root domain and allows bigger CPUs in
it to help via newly idle balance.
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: Andrea Righi <arighi@nvidia.com>
Link: https://patch.msgid.link/20260720-rneri-fix-cas-clusters-v6-4-bb500bf4afd4@linux.intel.com
update_sd_pick_busiest() may incorrectly select a fully_busy group as the
busiest group when its per-CPU capacity exceeds that of the destination
CPU. This happens because the type of busiest group is initialized to
group_has_spare and allows the fully_busy group to win the type comparison.
update_sd_pick_busiest() should not choose a candidate scheduling group
with at most one runnable task if its per-CPU capacity is greater than that
of the destination CPU. Such a check already exists, but it is done too
late: after the type comparison, preventing a subsequent fully_busy group
of equal per-CPU capacity from being correctly selected.
Move this check to occur before comparing group types.
Fixes: 0b0695f2b3 ("sched/fair: Rework load_balance()")
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: Andrea Righi <arighi@nvidia.com>
Link: https://patch.msgid.link/20260720-rneri-fix-cas-clusters-v6-3-bb500bf4afd4@linux.intel.com
The argument sg_overloaded of update_sg_lb_stats() is only consumed when
balancing at the root domain. It only makes sense to update it in such a
case. Commit 3229adbe78 ("sched/fair: Do not compute overloaded status
unnecessarily during lb") updated the logic accordingly but missed the case
in which the root domain has the SD_ASYM_CPUCAPACITY flag. Fix this.
Fixes: 3229adbe78 ("sched/fair: Do not compute overloaded status unnecessarily during lb")
Reported-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: Andrea Righi <arighi@nvidia.com>
Link: https://patch.msgid.link/20260720-rneri-fix-cas-clusters-v6-2-bb500bf4afd4@linux.intel.com
When picking a busiest CPU with only one running task, the function
sched_balance_find_src_rq() skips candidate CPUs if the destination CPU has
less than ~5% extra capacity. This condition only holds if all the SMT
siblings of a CPU are idle.
SMT siblings share the computing resources of a physical core and this
results in reduced capacity if more than one sibling is busy.
Skipping a CPU as described would prevent the load balancer from pulling
tasks from a scheduling group previously and correctly identified as
group_smt_balance (i.e., one with more than one task running).
Do not skip a candidate CPU of similar capacity if it has busy SMT
siblings.
Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Andrea Righi <arighi@nvidia.com>
Link: https://patch.msgid.link/20260720-rneri-fix-cas-clusters-v6-1-bb500bf4afd4@linux.intel.com
find_new_ilb() selects the first idle housekeeping CPU without
considering whether another thread is running on the same physical core.
On an SMT system, the idle load balancer can therefore activate both
siblings even when another housekeeping CPU has an entirely idle core.
On most SMT systems, this is not problematic because the idle load
balancer is a short-lived activity and the transient wakeup of a sibling
has negligible performance impact.
However, this can be particularly costly on NVIDIA Olympus cores used in
Vera. Briefly activating an otherwise idle sibling can reduce the
performance available to the other sibling and this effect does not
necessarily end once the activated sibling becomes idle: after the ILB
finishes and its CPU enters WFI, full single-thread performance is
restored only after the sibling has remained idle for a qualification
interval (10 Ki cycles on the tested Vera system). Repeated short
sibling wakeups can therefore sustain the interference even with little
actual overlap.
Prevent this by preferring an idle housekeeping CPU whose entire SMT
core is idle. Retain the first idle CPU as a fallback when no fully idle
core is available, so NOHZ balancing continues to make forward progress.
Once a partially busy core has been examined, skip its remaining SMT
siblings to avoid repeating the core-idle check on wide SMT systems.
Tests performed using an ad hoc GEMM benchmark running one CPU-intensive
task per SMT core within its CPU affinity mask improved from
approximately 6.2 TFLOP/s to 9.4 TFLOP/s.
Note that this preference may wake a fully idle physical core instead of
using an idle sibling of an active core, potentially increasing ILB
wakeup latency or energy consumption on some architectures. It may also
scan additional CPUs before selecting the one to run the ILB. The
selection falls back to the first idle CPU when no fully idle SMT core
is available. Non-SMT systems continue to select the first idle
housekeeping CPU.
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Mete Durlu <meted@linux.ibm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260804151324.918020-1-arighi@nvidia.com
No caller checks the return value from stop_one_cpu_nowait(). All
callers require the callback to run and arrange for the target CPU's
stopper to remain enabled while queuing the work. In particular, commit
f0498d2a54 ("sched: Fix stop_one_cpu_nowait() vs hotplug") added
preemption protection to the scheduler callers so that queuing must
succeed once the target CPU has been observed online.
Therefore, a failure is an unrecoverable violation rather than a condition
individual callers can recover from. Diagnose it with WARN_ON_ONCE() in
stop_one_cpu_nowait(). A check in the common helper covers current and
future callers consistently, while individual checks would duplicate
the same non-recoverable handling at every call site.
Make the function return void because there is no longer a meaningful
result for callers to consume.
On UP, warn if the supplied CPU is not the current CPU because the work
cannot be scheduled in that case.
Signed-off-by: Yury Norov <ynorov@nvidia.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Bradley Morgan <include@grrlz.net>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://patch.msgid.link/20260729022355.325058-1-ynorov@nvidia.com
Load balancing can migrate delayed dequeue tasks to even the load between
CPUs.
sched_balance_rq()
-> detach_task()
-> deactivate_task(DEQUEUE_NOCLOCK)
-> set_task_cpu(dst_cpu)
-> attach_task()
-> activate_task(ENQUEUE_NOCLOCK)
-> wakeup_preempt()
A delayed task with shorter slice can be dequeued during pick_next_entity()
but then jump to preempt because eligible.
Always jump to update for delayed dequeue task even with shorter slice.
Fixes: ba0d3bf5f9 ("sched/eevdf: Cancel slice protection if short slice task is eligible")
Reported-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260701164920.1571352-1-vincent.guittot@linaro.org
A few comments still describe the pre-EEVDF CFS world:
- sysctl_sched_base_slice is documented as "Minimal preemption
granularity for CPU-bound tasks". That was the wording of the old
sysctl_sched_min_granularity, renamed in commit e4ec3318a1
("sched/debug: Rename sysctl_sched_min_granularity to
sysctl_sched_base_slice"). Under EEVDF it is the default base time
slice / request size (r_i) used to compute the virtual deadline, as
documented in update_deadline().
- Two comments still mention sched_slice(), which was removed when the
fair class committed to EEVDF in commit 5e963f2bd4 ("sched/fair:
Commit to EEVDF"). The dequeue-path comment should simply refer to
the task's slice (se->slice); the forced-idle comment describes the
slice accounting now performed by __entity_slice_used(), which is the
function actually used right below it.
No functional changes.
[ mingo: Ported to a more recent scheduler base ]
Signed-off-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260629030200.3165589-1-zhanxusheng@xiaomi.com
nr_migrations_cold, nr_wakeups_passive and nr_wakeups_idle are not
being updated anywhere. So remove them.
These are per process stats. So updating sched stats version isn't
necessary.
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260625124648.802832-2-sshegde@linux.ibm.com
psi_account_irqtime() reads irq_time_read() into a per-rq cumulative
counter and only bails out when the delta vs. the previously accounted
amount is negative. A delta of exactly zero is treated as "do the
work": psi_write_begin() is taken, cpu_clock(cpu) is read (which on
x86 ends up in native_sched_clock() / rdtsc) and the cgroup ancestor
chain is walked to add zero to every group's PSI_IRQ_FULL bucket.
The zero-delta case is common in practice -- it fires every time a
context switch crosses a PSI group boundary on a CPU that hasn't
serviced an interrupt between the two switches.
Measured on a 176-thread AMD EPYC 9D64 server running a compute
intensive production workload, instrumented with bpftrace over a 30s
window (irq_time_read() read directly from the per-CPU cpu_irqtime so
that delta == 0 and delta < 0 could be separated):
@total 17,229,311 (100.0%)
@ret_curr_swapper 7,864,195 ( 45.6%) curr->pid == 0
@ret_samegrp 323,299 ( 1.9%) same cgroup as prev
@reached_delta 9,041,817 ( 52.5%)
@delta_positive 6,358,192 ( 36.9%) real work
@delta_zero 2,683,625 ( 15.6%) work wasted (this patch)
@delta_negative (0) ( 0.0%) monotonic clock
So 15.6 % of all psi_account_irqtime() calls - and 29.7 % of the
calls that get past the early returns - hit the delta == 0 case;
delta < 0 did not occur once in the 30 s window. Under the current
code each of those ~89 k calls per second performs the full seqcount
write + cpu_clock() read + cgroup-chain walk just to add 0 to every
group's PSI_IRQ_FULL counter.
Extend the early-return to also cover delta == 0. rq->psi_irq_time
does not need updating in that case (it would store the same value
back) and no PSI bucket would change. The existing behaviour for
delta > 0 is untouched.
Signed-off-by: Usama Arif <usama.arif@linux.dev>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Link: https://patch.msgid.link/20260617175219.2494857-2-usama.arif@linux.dev
Because active_balance is initialized to 0, and need_active_balance()
is a pre-condition for setting it to 1, the condition '!active_balance
|| need_active_balance()' is a truism and can be removed.
Signed-off-by: Xin Zhao <jackzxcui1989@163.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260617072151.1173416-3-jackzxcui1989@163.com
Active load balancing relies on migration threads, which temporarily preempt
tasks on the source runqueue (src_rq). This preemption can negatively impact
overall system performance. The active balancing logic includes a check to
verify whether the current task (curr) on src_rq can actually run on the
destination runqueue (dst_rq). We have observed that when curr is a CFS task
and its on_rq flag is 0, the active balancing failure rate is exceptionally
high. The following table summarizes test data collected over 300 seconds on an
18-CPU platform under a specific fillback task scenario:
fair: busiest->curr->sched_class == &fair_sched_class
on_rq: busiest->curr->on_rq
total: active balance count triggered of correspondent type
fail: fail to migrate one task in active_load_balance_cpu_stop()
fair && !on_rq !fair && !on_rq
domain total fail total fail
cpu0 0x00003 0 0 0 0
cpu0 0x3ffff 33 33 1 1
cpu1 0x00003 0 0 0 0
cpu1 0x3ffff 42 42 0 0
cpu2 0x0003c 4 4 0 0
cpu2 0x3ffff 12 12 0 0
cpu3 0x0003c 3 3 0 0
cpu3 0x3ffff 8 7 0 0
cpu4 0x0003c 2 2 0 0
cpu4 0x3ffff 5 4 0 0
cpu5 0x0003c 4 4 0 0
cpu5 0x3ffff 8 8 0 0
cpu6 0x003c0 60 60 0 0
cpu6 0x3ffff 28 27 0 0
cpu7 0x003c0 194 184 0 0
cpu7 0x3ffff 35 35 1 1
cpu8 0x003c0 240 228 0 0
cpu8 0x3ffff 28 28 0 0
cpu9 0x003c0 0 0 0 0
cpu9 0x3ffff 10 10 0 0
cpu10 0x03c00 52 50 0 0
cpu10 0x3ffff 0 0 0 0
cpu11 0x03c00 70 68 0 0
cpu11 0x3ffff 1 1 0 0
cpu12 0x03c00 73 72 0 0
cpu12 0x3ffff 0 0 0 0
cpu13 0x03c00 79 76 0 0
cpu13 0x3ffff 0 0 0 0
cpu14 0x3c000 0 0 0 0
cpu14 0x3ffff 57 55 1 0
cpu15 0x3c000 53 52 1 0
cpu15 0x3ffff 30 29 0 0
cpu16 0x3c000 344 341 10 6
cpu16 0x3ffff 103 100 2 1
cpu17 0x3c000 183 179 2 2
cpu17 0x3ffff 78 77 0 0
sum 1839 1791 18 11
In __schedule(), before curr is updated to next, pick_next_task() invokes
sched_balance_rq(). This function temporarily unlocks and relocks the runqueue,
creating a window where other CPUs may observe rq->curr->on_rq as 0.
We can safely skip active balancing when src_rq->curr->on_rq == 0, as other
eligible tasks have likely already been evaluated.
We retain the affinity check on dst_rq to trigger active balancing, since such
tasks are often woken by (or wake up) tasks on src_rq that share similar
affinity constraints. Furthermore, detach_tasks() releases the runqueue lock;
any tasks awakened during this window may preempt the previous CFS task. My
testing (data not shown) indicates that active balancing succeeds in 98.4% of
cases where !fair && on_rq.
This scenario does not require a stop-work callback, but would necessitate an
additional detach/attach path. As Valentin and Vincent have already discussed,
this addition does not appear justified at this time (see [1]).
Since can_migrate_task() already checks on_cpu during the cfs_tasks traversal,
adding an on_rq check will have negligible performance overhead due to cache
locality.
There are two reasons for not combining the on_rq check with the
cpumask_test_cpu() check:
- Avoiding new scenarios that would skip the logic for resetting
balance_interval to min_interval.
- The existing check for whether the busiest CPU recently triggered active
load balancing already filters more cases than the on_rq check.
[1]: https://lore.kernel.org/lkml/20190815145107.5318-5-valentin.schneider@arm.com/
Signed-off-by: Xin Zhao <jackzxcui1989@163.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Link: https://patch.msgid.link/20260617072151.1173416-2-jackzxcui1989@163.com
When a task with a shorter slice is enqueued, we protect the running
task which has a longer slice until it becomes ineligible instead of a
full slice in order to speedup the switch to other tasks until the task
with the shortest slice is scheduled. This helps to the task to not wait
too many full slices before running.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260624151229.1710703-7-vincent.guittot@linaro.org
Even if p will not preempt current, it modifies the avg_vruntime and
possibly the min slice. Make sure to update the slice protection with the
updated figures. As an example, Batch and Sched Idle tasks can otherwise
get a larger lag than their slice and finaly delay the scheduling of a
normal task, which deadline will be a later.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260624151229.1710703-6-vincent.guittot@linaro.org
If a short slice task will not be the next to be picked but is eligible,
we cancel the slice protection to speedup the time when the short slice
task will be the next to run.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260624151229.1710703-5-vincent.guittot@linaro.org
Even if resched is already set, we might want to update or even cancel
the slice protection and ensure that the newly waking task will be the
next one to run.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260624151229.1710703-4-vincent.guittot@linaro.org
Take into account the lag of current task when updating the slice
protection in order to ensure that the absolute value of lags will remain
in the range [0 : slice+tick]
A task that already has a negative lag will see its protection reduced
whereas a task with positive lag will keep a full slice protection.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260624151229.1710703-3-vincent.guittot@linaro.org
If a shorter slice task can preempt current at wakeup, we make sure that
the decision will not be overwritten in between by setting the task as the
next buddy. This still implies that the waking task remains eligible when
the scheduler will actually pick the next task to run.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260624151229.1710703-2-vincent.guittot@linaro.org
On a tangential note, I just noticed set_preempt_buddy() has two unused
parameters. Seems to have been like that since it was introduced in
commit e837456fdc ("sched/fair: Reimplement NEXT_BUDDY to align with
EEVDF goals").
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Change fair/cgroup to a single runqueue.
Infamously fair/cgroup isn't working for a number of people; typically
the complaint is latencies and/or overhead. The latency issue is due
to the intermediate entries that represent a combination of tasks and
thereby obfuscate the runnability of tasks.
The approach here is to leave the cgroup hierarchy as is; including
the intermediate enqueue/dequeue but move the actual EEVDF runqueue
outside. This means things like the shares_weight approximation are
fully preserved.
That is, given a hierarchy like:
R
|
se--G1
/ \
G2--se se--G3
/ \ |
T1--se se--T2 se--T3
This is fully maintained for load tracking, however the EEVDF parts of
cfs_rq/se go unused for the intermediates and are instead connected
like:
_R_
/ | \
T1 T2 T3
Since the effective weight of the entities is determined by the
hierarchy, this gets recomputed on enqueue,set_next_task and tick.
Notably, the effective weight (se->h_load) is computed from the
hierarchical fraction: se->load / cfs_rq->load.
Since EEVDF is now exclusively operating on rq->cfs, it needs to
consider cfs_rq->h_nr_queued rather than cfs_rq->nr_queued. Similarly,
only tasks can get delayed, simplifying some of the cgroup cleanup.
One place where additional information was required was
set_next_task() / put_prev_task(), where we need to track 'current'
both in the hierarchical sense (cfs_rq->h_curr) and in the flat sense
(cfs_rq->curr).
As a result of only having a single level to pick from, much of the
complications in pick_next_task() and preemption go away.
Since many of the hierarchical operations are still there, this won't
immediately fix the performance issues, but hopefully it will fix some
of the latency issues.
TODO: split struct cfs_rq / struct sched_entity
TODO: try and get rid of h_curr
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260605124052.227463677%40infradead.org
For all the reasons described in the preceding patches, the way cgroup
weight is computed is problematic. However, changing it is bound to
also lead to trouble. Esp. since people might have taken to inflating
the weight value where they can.
Since things are configurable, change the default and hope this serves
more people than it hurts, esp. in the longer run.
Specifically, this prepares for a flattened runqueue, where the hierarchical
weight becomes far more important (F_g^d terms), so getting rid of small F_g is
imperative.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260605124052.080482755%40infradead.org
Since we are exploring this space; include a scheme that scales by total number
of runnable tasks. This results in:
F_g_n' = M * F_g_n
This will obviously have: avg(F_g_n') > 1, (it will be ~M/N in fact).
And while that sounds odd, it actually has a fairly straight foward meaning for
"cpu.weight": average weight per member task.
This is an entirely valid and workable option, it is however wildly different
from the traditional meaning.
Included for completeness (and curiosity).
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260605124051.921991975%40infradead.org
Improve upon the previous scheme ("max") by no longer assuming maximal
concurrency. Instead scale by: 'min(nr_tasks, nr_cpus)'. This handles
the low concurrency cases more gracefully:
F_g_n' = min(M, N) * F_g_n
Notably this is the first mode where:
avg(F_g_n) = 1
In the single task case it reduces to ("smp") and then it nicely scales up
until it hits N, where it behaves like ("max").
This is no longer clipped at nice -20. Strictly speaking it isn't different
from the normal SMP scenario where all tasks are extremely unbalanced. There
are no unnatural inflations in this scheme.
The meaning of "cpu.weight" would be: weight per active CPU.
NOTE: Compute the group wide number of tasks by extending the tg->load_avg
computation with tg->runnable_avg, since cfs_rq->runnable_avg is based on
cfs_rq->h_nr_running.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260605124051.740585993%40infradead.org
In order to avoid the average CPU fraction avg(F_g_n) becoming tiny '1/N',
assume each cgroup is maximally concurrent and distrubute 'N*weight', such
that:
F_g_n' = N * F_g_n
Giving:
avg(F_g_n') = N*avg(F_g_n) ~ N * 1/N = 1
And while this sounds like it solves things, remember what that ~ meant. There
is the corner case when a cgroup is minimally loaded, eg a single runnable
task, therefore limit the CPU fraction to that of a nice -20 task to avoid
getting too much load.
This last bit is what makes it different from a previous proposal to allow
raising cpu.weight to '100 * N', that would not limit the mininal concurrency
case and results in a very large F_g_n. And just like F_g_n << 1 is
problematic, so is F_g_n >> 1 for the exact same reasons (it would drown the
kthreads, but it also risks overflowing the load values).
So while this might appear to be a better scheme than the current default
scheme, it doesn't really handle less than maximal concurrency nicely -- it
clips and introduces artificially large weights. So where the traditional SMP
mode works well when nr_tasks << nr_cpus, MAX doesn't work well in that regime
and vice-versa.
The meaning of "cpu.weight" would be: weight per allowed CPU.
Included for completeness (and infrastructure).
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260605124051.589618504%40infradead.org
Instead of calculating the proportional fraction of the group weight for each
CPU, just give each CPU the full measure, ignoring these pesky SMP problems.
This makes the SMP cgroup fraction (F_g_n) equal to 1, and ensures a single
task in a cgroup competes on equal footing to a task in a level above.
However, as already explored, this is not a very good policy because it gets
the SMP weight distribution wrong. Included for completeness.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260605124051.450303977%40infradead.org
The effective task weight (W_t') for a task in cgroup g on CPU n is given by:
W_t
W_t' = W_g * F_g_n * ----------
\Sum W_t_n
Where W_g is the group's weight (cpu.weight), F_g_n is the fraction of the
group weight for CPU n and W_t/W is the relative weight of this task against
all other tasks in the same group on the same CPU.
Furthermore, this makes:
\Sum W_t_n
F_g_n = ----------
\Sum W_t
The fraction of weight inside the group of CPU n against the whole group.
The problem is with F_g_n, the primary goal of this fraction is to make sure
that the relative weight of tasks, when distributed over CPUs is maintained.
For example, consider 4 (equal weight) tasks and 2 CPUs with a 1:3
distribution, then if F_g_n would simply be 1 (no weight re-distribution) the
effective relative weights (W_t') of the tasks in our group would be:
CPU0 CPU1
W_g W_g/3
W_g/3
W_g/3
IOW, the lucky task on CPU0 would get an equal amount of weight as all 3 tasks
on CPU1 combined. However, with the weight redistribution, this becomes:
CPU0 CPU1
W_g/4 W_g/4
W_g/4
W_g/4
All tasks are equal weight (as intended). However, as is already evident from
this example, the more CPUs you add, the smaller F_g_n becomes, which creates a
disparity against tasks not in our group.
Specifically:
avg(F_g_n) ~ 1/N
This leads to a weight mismatch in the hierarchy. IOW tasks cannot compete
fairly across hierarchy levels.
*Notably*, what is meant by avg(F_g_n) being proportional to 1/N is that when
there are at least N runnable tasks, the average of this fraction tends to 1/N.
For a hierarchy of depth d, this gets even worse, since that gets terms on the
order of:
avg(F_g_n)^d ~ 1/(N^d)
Given fixed point arithmetic, this also leads to numerical trouble.
However, the meaning of "cpu.weight" is simple and intiutive: the total weight
of the cgroup. But as explored above, there is deception in this simplicity.
Prepare to add a few alternative methods for distributing weight.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260605124051.338602724%40infradead.org
A divide-by-zero crash is observed when running hackbench:
[14697.488452] CPU: 112 UID: 0 PID: 124791 Comm: hackbench Not tainted 7.1.0-rc2+
[14697.492627] RIP: 0010:propagate_entity_load_avg+0x35f/0x3e0
[14697.506799] <TASK>
[14697.507411] __dequeue_task+0x2b4/0xc70
[14697.508677] dequeue_task_fair+0x36/0x370
[14697.509047] dequeue_task+0x101/0x2f0
[14697.509426] __schedule+0x1b1/0x1a00
[14697.510868] anon_pipe_read+0x3da/0x450
[14697.511400] vfs_read+0x361/0x390
[14697.512053] __x64_sys_read+0x19/0x30
The divide-by-zero happens here:
if (scale_load_down(gcfs_rq->load.weight)) {
load_sum = div_u64(gcfs_rq->avg.load_sum,
scale_load_down(gcfs_rq->load.weight));
}
gcfs_rq->load.weight is an insane large value and is truncated
to the lower 32 bits by div_u64, which happen to be 0.
Using AI for investigation, the cause is a u32 overflow in
update_tg_cfs_runnable(), and flat pickup became a victim when using
tg_tasks():
u32 new_sum, divider;
...
new_sum = se->avg.runnable_avg * divider; <-- boom
The following sequence shows how this triggers the crash:
propagate_entity_load_avg()
update_tg_cfs_runnable() # u32 overflow corrupts runnable_sum
__update_load_avg_cfs_rq()
___update_load_avg() # computes insane runnable_avg
update_tg_load_avg() # propagates to tg->runnable_avg
update_cfs_group()
calc_concur_shares()
tg_tasks() # long-to-int truncation, negative nr
reweight_entity() # corrupted se->load.weight
update_load_add() # corrupted cfs_rq->load.weight
propagate_entity_load_avg()
update_tg_cfs_load()
div_u64() # divide-by-zero
Fix by widening new_sum from u32 to u64 (no need to force tg_tasks()
to return unsigned long after this fix)
Fixes: 95246d1ec8 ("sched/pelt: Relax the sync of runnable_sum with runnable_avg")
Assisted-by: Claude:claude-opus-4.6
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/a22eea2b-4c4a-4623-9a44-d7b18c0c91c8@intel.com
The way wakeup_preempt() works since commit 704069649b ("sched/core: Rework
sched_class::wakeup_preempt() and rq_modified_*()") is that it will call
rq->next_class->wakeup_preempt(rq, p) when p is of an equal or higher class,
and raise ->next_class when higher.
This means that:
running idle task
wakeup fair-A
(next_class == idle)
if (sched_class_above(fair, idle)) {
wakeup_preempt_idle(fair-A);
resched_curr(rq);
next_class = fair;
}
wakeup fair-B
(next_class == fair)
if (fair == fair)
wakeup_preempt_fair(fair-B);
(but current is idle)
All wakeup_preempt_$class() methods, except for wakeup_preempt_scx() (for whoem
this was build) ignore cross-class wakeups by testing if @p is of the right
class, but per the above case, it also should check current.
This is mostly harmless in the current form, but will lead to trouble with
later patches.
Fixes: 704069649b ("sched/core: Rework sched_class::wakeup_preempt() and rq_modified_*()")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260626074605.GB2568396%40noisy.programming.kicks-ass.net
device removal, along with documentation fixes and minor AMD driver
cleanups.
-----BEGIN PGP SIGNATURE-----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=V3cD
-----END PGP SIGNATURE-----
Merge tag 'ntb-7.2' of https://github.com/jonmason/ntb
Pull NTB updates from Jon Mason:
"An EPF bug fix to prevent an invalid unmap during device removal,
along with documentation fixes and minor AMD driver cleanups"
* tag 'ntb-7.2' of https://github.com/jonmason/ntb:
ntb: amd: Use named initializer for pci_device_id::driver_data
NTB: fix kernel-doc warnings in ntb.h
NTB: epf: Avoid pci_iounmap() with offset when PEER_SPAD and CONFIG share BAR
ntb_hw_amd: Fix incorrect debug message in link disable path
- Updates to Synaptics RMI4 driver to fix potential OOB accesses in
F30 and F3A keymap handling
- A workaround in Synaptics RMI4 to tolerate buggy firmware on some
touchpads (e.g. ThinkPad T14 Gen 1) that report incomplete register
descriptor structures, preventing probe failures
- A revert of an incorrect register descriptor address calculation in
Synaptics RMI4 driver
- A fix for a regression in HP GSC PS/2 (gscps2) driver where the
receive buffer write index was not advanced, leaving keyboard and
mouse unusable.
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQST2eWILY88ieB2DOtAj56VGEWXnAUCakCsxwAKCRBAj56VGEWX
nCICAQDSgfrAi+4SqTb92EjtdQO+ypluS42mKO75LTitJcS8dAEA1iKmdss8mGww
c4ai4W+UFxori7IqgkoQ7LfaPTOWlA0=
=QP4c
-----END PGP SIGNATURE-----
Merge tag 'input-for-v7.2-rc0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Pull more input updates from Dmitry Torokhov:
- Updates to Synaptics RMI4 driver to fix potential OOB accesses in F30
and F3A keymap handling
- A workaround in Synaptics RMI4 to tolerate buggy firmware on some
touchpads (e.g. ThinkPad T14 Gen 1) that report incomplete register
descriptor structures, preventing probe failures
- A revert of an incorrect register descriptor address calculation in
Synaptics RMI4 driver
- A fix for a regression in HP GSC PS/2 (gscps2) driver where the
receive buffer write index was not advanced, leaving keyboard and
mouse unusable.
* tag 'input-for-v7.2-rc0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: gscps2 - advance receive buffer write index
Input: rmi4 - tolerate short register descriptor structure
Revert "Input: rmi4 - fix register descriptor address calculation"
Input: synaptics-rmi4 - bound the F30 keymap to the GPIO/LED count
Input: synaptics-rmi4 - bound the F3A keymap to the GPIO count
Two more fixes that I managed to put into the public branch merged into
next before my first PR but missed to include them in it. The first
change is a relevant change that fixes misconfigurations due to a
variable overflow. The 2nd is only cosmetic but very obviously an
improvement.
-----BEGIN PGP SIGNATURE-----
iQEzBAABCgAdFiEEP4GsaTp6HlmJrf7Tj4D7WH0S/k4FAmpAE2gACgkQj4D7WH0S
/k7tnAgAsjkBQWoav1MF8kyVzew+iIuVvBrtUPimzeMs2gzwC1czi/TPKCvZxGND
XaPESmf51lC9AZuFd/2Wfvb2GydxA/wmsbVB0Q6ZBwLtS6/L4yiv3DpYN2to9yWN
QvVHeVCBdSbMHOQXdG0iFbhMiLyUX5YwCwyZT2cuVUHb4gvNLuDuKbgSXX9Odh1R
/p9C0afNMbdxuj2yRy+S8CM5Rl5v4yfBw6cswKX6w3uA+LnvksWuC8og7tEfFa/F
Vdwy4csWeKHGm7jEP9o6iSlWYy+DMKf5Llop+yFLZV6Db6AXMlEpNOXJsGxEv95y
qMTQFCGhepNWqVuUDDOOeUs60Asjpg==
=0WY9
-----END PGP SIGNATURE-----
Merge tag 'pwm/for-7.2-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux
Pull pwm fixes from Uwe Kleine-König:
"Two more fixes that I managed to put into the public branch merged
into next before my first pull request but missed to include them in
it.
The first change is a relevant change that fixes misconfigurations due
to a variable overflow. The second is only cosmetic but very obviously
an improvement"
* tag 'pwm/for-7.2-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux:
pwm: rzg2l-gpt: Add missing newlines to dev_err_probe() messages
pwm: rzg2l-gpt: Fix period_ticks type from u32 to u64
Fixes:
- fbcon: fix NULL pointer dereference for a console without vc_data [Ian Bridges]
- fbdev: fix fb_new_modelist to prevent null-ptr-deref in fb_videomode_to_var [Ian Bridges]
Fixes in failure paths:
- pm2fb: unwind write-cache setting on probe failure [Haoxiang Li]
- goldfishfb: fail pan display on base-update timeout [Pengpeng Hou]
- viafb: return error on DMA copy time-out [Pengpeng Hou]
- fbcon: fix out-of-bounds read in error path of fbcon_do_set_font() [Mingyu Wang]
- fbdev: fix modelist use-after-free in store_modes() [Ian Bridges]
Code cleanup:
- vga16fb: clean up platform_device_id table [Uwe Kleine-König]
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQS86RI+GtKfB8BJu973ErUQojoPXwUCakAKnwAKCRD3ErUQojoP
X27+AQCPvMJUMScjO1dlGKkPnnIFiOWCU6AW1aM2m6T+qWrCTQEAnNi5CCI3JiLi
6NCcyPm7/4R6tSXDm+UcWJ5tyzaNvww=
=EBHb
-----END PGP SIGNATURE-----
Merge tag 'fbdev-for-7.2-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/linux-fbdev
Pull more fbdev updates from Helge Deller:
"Fixes for generic fbdev & fbcon code for the handling of modelists
and preventing a potential NULL ptr dereference in the console code.
Fix missed cleanups in the error path of various fbdev drivers.
And Uwe Kleine-König contributed a cleanup patch to use named
initializers in the vga16fb driver"
* tag 'fbdev-for-7.2-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/linux-fbdev:
fbdev: Fix fb_new_modelist to prevent null-ptr-deref in fb_videomode_to_var
fbcon: fix NULL pointer dereference for a console without vc_data
fbdev: fix use-after-free in store_modes()
fbdev: viafb: return an error when DMA copy times out
fbdev: goldfishfb: fail pan display on base-update timeout
fbdev: fbcon: fix out-of-bounds read in err_out of fbcon_do_set_font()
fbdev: pm2fb: unwind WC setup on probe failure
fbdev: vga16fb: Drop unused assignment of platform_device_id driver data
A collection of small bug fixes accumulated over the last week.
Most are device-specific fixes while there are a few core fixes as
well.
Here are the highlights:
ALSA Core:
- A fix for an uninitialised heap leak in ALSA sequencer core
- A fix for error handling/resource leak in compress-offload API
USB-audio:
- A teardown-ordering fix in USB MIDI 2.0 to prevent use-after-free
- Bounds and length checks for packet data in Native Instruments caiaq
/ Traktor Kontrol input parsers
- Avoidance of expensive kobject path lookups in DualSense controller
matches
- Robustness/memory leak fixes for Qualcomm USB offload driver
- Focusrite Control Protocol (FCP) NULL-pointer dereference fix and a
new device quirk (ISA C8X)
- Device-specific quirks for Yamaha CDS3000 and SC13A
HD-Audio:
- A bunch of quirks and mute/mic-mute LED fixups for various laptops
(Acer, Clevo, Lenovo, HP)
ASoC & SoundWire:
- Avoid failing card registration if the device_link creation fails
- A workaround for SoundWire randconfig build failures by making
helper functions static inline
- Corrected MCLK reference validation for CS530x codecs
- Clean up of untested, problematic guard() macro replacements in
Rockchip SAI driver
- Fix for eDMA maxburst misalignment with channel count in Freescale
ASRC
- Miscellaneous hardware-specific fixes (qcom, rt5650, tlv320aic3x,
tas2781/3)
Others:
- Bounds and length checks for packet data in Apple iSight
-----BEGIN PGP SIGNATURE-----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=hJ5w
-----END PGP SIGNATURE-----
Merge tag 'sound-fix-7.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"A collection of small bug fixes accumulated over the last week.
Most are device-specific fixes while there are a few core fixes as
well.
Here are the highlights:
ALSA Core:
- A fix for an uninitialised heap leak in ALSA sequencer core
- A fix for error handling/resource leak in compress-offload API
USB-audio:
- A teardown-ordering fix in USB MIDI 2.0 to prevent use-after-free
- Bounds and length checks for packet data in Native Instruments
caiaq / Traktor Kontrol input parsers
- Avoidance of expensive kobject path lookups in DualSense controller
matches
- Robustness/memory leak fixes for Qualcomm USB offload driver
- Focusrite Control Protocol (FCP) NULL-pointer dereference fix and a
new device quirk (ISA C8X)
- Device-specific quirks for Yamaha CDS3000 and SC13A
HD-Audio:
- A bunch of quirks and mute/mic-mute LED fixups for various laptops
(Acer, Clevo, Lenovo, HP)
ASoC & SoundWire:
- Avoid failing card registration if the device_link creation fails
- A workaround for SoundWire randconfig build failures by making
helper functions static inline
- Corrected MCLK reference validation for CS530x codecs
- Clean up of untested, problematic guard() macro replacements in
Rockchip SAI driver
- Fix for eDMA maxburst misalignment with channel count in Freescale
ASRC
- Miscellaneous hardware-specific fixes (qcom, rt5650, tlv320aic3x,
tas2781/3)
Others:
- Bounds and length checks for packet data in Apple iSight"
* tag 'sound-fix-7.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (46 commits)
ALSA: FCP: Fix NULL pointer dereference in interface lookup
ALSA: hda/realtek: Update Acer Nitro ANV15-41 quirk to enable mute LED
ASoC: fsl_asrc_dma: fix eDMA maxburst misalignment with channel count
ASoC: codecs: pcm512x: only print info once on no sclk
ASoC: tas2781: Update default register address to TAS2563
ALSA: firewire: isight: bound the sample count to the packet payload
ALSA: usb-audio: qcom: Free QMI handle
ALSA: hda: Add Lenovo Legion 7i 16IAX7 17AA3874 quirk
ALSA: usb-audio: avoid kobject path lookup in DualSense match
ALSA: hda/realtek: Add quirk for Acer Nitro ANV15-41
ASoC: soc-core: Don't fail if device_link could not be created
ASoC: rockchip: rockchip_sai: #include <linux/platform_device.h> explicitly
ALSA: seq: Fix uninitialised heap leak in snd_seq_event_dup()
ASoC: rt5575: Use __le32 for SPI burst write address
ASoC: tas2783: Update loaded firmware names to linux-firmware 20260519
ASoC: SDCA: Validate written enum value in ge_put_enum_double()
ASoC: realtek: Add back local call to sdw_show_ping_status()
ASoC: ti: Add back local call to sdw_show_ping_status()
ASoC: max98373: Add back local call to sdw_show_ping_status()
ASoC: es9356: Add back local call to sdw_show_ping_status()
...
Subsystem:
- add rtc_read_next_alarm() to read next expiring timer
Drivers:
- ds1307: handle OSF for ds1337/ds1339/ds3231, add clock provider for ds1307,
fix wday for rx8130
- m41t93: DT support, alarm, clock provider, watchdog support
- mv: add suspend/resume support for wakeup
- pcap: remove driver
- renesas-rtca3: many fixes
-----BEGIN PGP SIGNATURE-----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=SCOZ
-----END PGP SIGNATURE-----
Merge tag 'rtc-7.2' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux
Pull RTC updates from Alexandre Belloni:
"Most of the work and improvements are for features of the m41t93.
The ds1307 also gets support for OSF (Oscillator Stop Flag) for
new variants.
The pcap driver is being removed as the Motorola EZX support was
removed a while ago.
Subsystem:
- add rtc_read_next_alarm() to read next expiring timer
Drivers:
- ds1307: handle OSF for ds1337/ds1339/ds3231, add clock provider for
ds1307, fix wday for rx8130
- m41t93: DT support, alarm, clock provider, watchdog support
- mv: add suspend/resume support for wakeup
- pcap: remove driver
- renesas-rtca3: many fixes"
* tag 'rtc-7.2' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux: (36 commits)
rtc: ds1307: update reference to removed CONFIG_RTC_DRV_DS1307_HWMON
platform/x86: amd-pmc: Fix S0i3 wakeup with alarmtimer
rtc: s35390a: fix typo in comment
rtc: cmos: unregister HPET IRQ handler on probe failure
rtc: ds1307: Fix off-by-one issue with wday for rx8130
dt-bindings: rtc: ds1307: Add epson,rx8901
rtc: bq32000: add delay between RTC reads
rtc: m41t93: Add watchdog support
rtc: m41t93: Add square wave clock provider support
rtc: m41t93: Add alarm support
rtc: m41t93: migrate to regmap api for register access
rtc: m41t93: add device tree support
dt-bindings: rtc: Add ST m41t93
rtc: ds1307: add support for clock provider in ds1307
rtc: mv: add suspend/resume support for wakeup
rtc: aspeed: add AST2700 compatible
dt-bindings: rtc: add ASPEED AST2700 compatible
rtc: interface: fix typos in rtc_handle_legacy_irq() documentation
rtc: msc313: fix NULL deref in shared IRQ handler at probe
rtc: remove unused pcap driver
...
- Fix a bug where in a specific edge case, file contents en/decryption
could be done with the wrong data unit size.
- Fix the data structure used for keeping track of users that have added
an fscrypt key to be a simple list instead of a 'struct key' keyring.
This fixes issues such as a lockdep report found by syzbot and
possible unintended interactions with the keyctl() system calls.
-----BEGIN PGP SIGNATURE-----
iIoEABYIADIWIQSacvsUNc7UX4ntmEPzXCl4vpKOKwUCaj8bBhQcZWJpZ2dlcnNA
a2VybmVsLm9yZwAKCRDzXCl4vpKOKyhIAP47lR+H783gopiz10Z7dwLQr2EHMfZ1
NcU7Zfq++AzZRQD/Tvv9/daqSh0OJTDNkBskSgVj6z7sRlQVNHY0wrD+QwE=
=LWES
-----END PGP SIGNATURE-----
Merge tag 'fscrypt-for-linus' of git://git.kernel.org/pub/scm/fs/fscrypt/linux
Pull fscrypt fixes from Eric Biggers:
- Fix a bug where in a specific edge case, file contents en/decryption
could be done with the wrong data unit size
- Fix the data structure used for keeping track of users that have
added an fscrypt key to be a simple list instead of a 'struct key'
keyring
This fixes issues such as a lockdep report found by syzbot and
possible unintended interactions with the keyctl() system calls
* tag 'fscrypt-for-linus' of git://git.kernel.org/pub/scm/fs/fscrypt/linux:
fscrypt: Replace mk_users keyring with simple list
fscrypt: Fix key setup in edge case with multiple data unit sizes
Commit 44f9200699 ("Input: gscps2 - use guard notation when
acquiring spinlock") moved the receive loop into gscps2_read_data()
and gscps2_report_data().
While moving the code, it preserved the writes to
buffer[ps2port->append], but omitted the following producer index
update from the original loop:
ps2port->append = (ps2port->append + 1) & BUFFER_SIZE;
As a result, append never advances. Since gscps2_report_data() only
reports bytes while act != append, the receive buffer always appears
empty and no keyboard or mouse data reaches the serio core.
Restore the omitted index update.
Fixes: 44f9200699 ("Input: gscps2 - use guard notation when acquiring spinlock")
Cc: stable@vger.kernel.org # 6.13+
Signed-off-by: Xu Rao <raoxu@uniontech.com>
Link: https://patch.msgid.link/460B5655BA580C60+20260624094739.850306-1-raoxu@uniontech.com
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Some touchpads (e.g. ThinkPad T14 Gen 1) have buggy firmware that reports
a register descriptor structure size that is too small for the number of
registers it claims to have in the presence map. The remaining bytes in
the structure are 0, which with the new strict bounds checking causes the
parser to fail with -EIO, aborting the device probe.
Tolerate such short reads by dropping the remaining (unparseable or
0-size) registers from the list instead of failing the probe,
preventing the driver from trying to use them.
Fixes: 0adb483fbf ("Input: rmi4 - refactor register descriptor parsing")
Reported-by: Barry K. Nathan <barryn@pobox.com>
Tested-by: Barry K. Nathan <barryn@pobox.com>
Cc: stable@vger.kernel.org
Assisted-by: Antigravity:gemini-3.5-flash
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
The register descriptor presence register is a packet register, which
means its bytes share a single RMI address. It does not occupy
consecutive addresses, and the register structure that follows it
is located at the next RMI address (presence_address + 1), not
(presence_address + presence_size).
Revert the incorrect address calculation introduced in commit
a98518e724.
Reported-by: "Barry K. Nathan" <barryn@pobox.com>
Tested-by: "Barry K. Nathan" <barryn@pobox.com>
Cc: stable@vger.kernel.org
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
The current list initialisation depends on the well hidden two zeros in
the PCI_VDEVICE macro. Instead use a named initialisation that is more
robust and easier to understand.
Signed-off-by: Uwe Kleine-König (The Capable Hub) <u.kleine-koenig@baylibre.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Correct a function name and function parameter name to avoid
kernel-doc warnings:
Warning: include/linux/ntb.h:575 expecting prototype for
ntb_default_port_count(). Prototype was for ntb_default_peer_port_count()
instead
Warning: include/linux/ntb.h:590 function parameter 'pidx' not
described in 'ntb_default_peer_port_number'
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
When BAR_PEER_SPAD and BAR_CONFIG share one PCI BAR, the module teardown
path ends up calling pci_iounmap() on the same iomem with some offset,
which is unnecessary and triggers a kernel warning like the following:
Trying to vunmap() nonexistent vm area (0000000069a5ffe8)
WARNING: mm/vmalloc.c:3470 at vunmap+0x58/0x68, CPU#5: modprobe/2937
[...]
Call trace:
vunmap+0x58/0x68 (P)
iounmap+0x34/0x48
pci_iounmap+0x2c/0x40
ntb_epf_pci_remove+0x44/0x80 [ntb_hw_epf]
pci_device_remove+0x48/0xf8
device_remove+0x50/0x88
device_release_driver_internal+0x1c8/0x228
driver_detach+0x50/0xb0
bus_remove_driver+0x74/0x100
driver_unregister+0x34/0x68
pci_unregister_driver+0x34/0xa0
ntb_epf_pci_driver_exit+0x14/0xfe0 [ntb_hw_epf]
[...]
Fix it by unmapping only when PEER_SPAD and CONFIG use difference bars.
Cc: stable@vger.kernel.org
Fixes: e75d5ae8ab ("NTB: epf: Allow more flexibility in the memory BAR map method")
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Signed-off-by: Koichiro Den <den@valinux.co.jp>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
amd_ntb_link_disable() prints "Enabling Link" which is misleading.
Update the message to reflect that the link is being disabled.
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
sysfb
- drm/sysfb truncation and alignment fixes.
edid:
- fix edid OOB read in tile parsing
- increase displayid topology id to correct size.
nouveau:
- fix error handling paths in nouveau
amdxdna:
- get_bo_info fix.
ivpu:
- fix leak when error handling in ivpu.
-----BEGIN PGP SIGNATURE-----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=eLQl
-----END PGP SIGNATURE-----
Merge tag 'drm-fixes-2026-06-27' of https://gitlab.freedesktop.org/drm/kernel
Pull drm fixes from Dave Airlie:
"These are just the fixes from our fixes branch, all pretty small and
scattered.
sysfb:
- drm/sysfb truncation and alignment fixes
edid:
- fix edid OOB read in tile parsing
- increase displayid topology id to correct size
nouveau:
- fix error handling paths in nouveau
amdxdna:
- get_bo_info fix
ivpu:
- fix leak when error handling in ivpu"
* tag 'drm-fixes-2026-06-27' of https://gitlab.freedesktop.org/drm/kernel:
drm/sysfb: Avoid truncating maximum stride
drm/sysfb: Return errno code from drm_sysfb_get_visible_size()
drm/sysfb: Avoid possible truncation with calculating visible size
drm/sysfb: Do not page-align visible size of the framebuffer
drm/edid: fix OOB read in drm_parse_tiled_block()
drm/nouveau: fix reversed error cleanup order in ucopy functions
drm/nouveau/acr: fix missing nvkm_done() in error path of nvkm_acr_oneinit()
accel/amdxdna: Use caller client for debug BO sync
drm/displayid: fix Tiled Display Topology ID size
accel/ivpu: fix HWS command queue leak on registration failure
i915:
- Fix corrupted display output on GLK, #16209
- Add missing Spectre mitigation for parallel submit IOCTL
- MTL+ fix for DP resume
- clear CRTC blobs after dropping refs
- fix sharpness filter on DP MST
xe:
- Set TTM beneficial order to 9 in Xe
- Several error path cleanups
- Fix TDR for unstarted jobs on kernel queues
- Several TLB invalidation fixes related to suspending LR queues
- Some small RAS fixes
- Multi-queue suspend fix for LR queues
- Revert inclusion of NVL_S firmware
amdgpu:
- devcoredump fixes
- SMU15 fix
- Various irq put/get imbalance cleanup fixes
- 8K panel fix
- DCN3.5 fix
- lockdep fix
- Cleaner shader sysfs IB overflow fix
- Async flip fixes
- GET_MAPPING_INFO fix
- CP_GFX_SHADOW fix
- Ctx pstate handling fix
- GTT bo move handling fixes
- Old UVD BO placement fixes
- GC9 mode2 reset fix
- IH6.1 version fix
- Soft IH ring fix
amdkfd:
- Fix doorbell/mmio double unpin on free
- CRIU fixes
- SMI event fixes
- Sysfs teardown fix
- Various boundary checking fixes
- Various error checking fixes
- SVM fix
-----BEGIN PGP SIGNATURE-----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=qcB5
-----END PGP SIGNATURE-----
Merge tag 'drm-next-2026-06-27' of https://gitlab.freedesktop.org/drm/kernel
Pull drm merge window fixes from Dave Airlie:
"This is the merge window fixes from our next tree, i915/xe and amdgpu
make up all of it.
I've got a separate fixes pull from our fixes branch arriving after
this.
i915:
- Fix corrupted display output on GLK, #16209
- Add missing Spectre mitigation for parallel submit IOCTL
- MTL+ fix for DP resume
- clear CRTC blobs after dropping refs
- fix sharpness filter on DP MST
xe:
- Set TTM beneficial order to 9 in Xe
- Several error path cleanups
- Fix TDR for unstarted jobs on kernel queues
- Several TLB invalidation fixes related to suspending LR queues
- Some small RAS fixes
- Multi-queue suspend fix for LR queues
- Revert inclusion of NVL_S firmware
amdgpu:
- devcoredump fixes
- SMU15 fix
- Various irq put/get imbalance cleanup fixes
- 8K panel fix
- DCN3.5 fix
- lockdep fix
- Cleaner shader sysfs IB overflow fix
- Async flip fixes
- GET_MAPPING_INFO fix
- CP_GFX_SHADOW fix
- Ctx pstate handling fix
- GTT bo move handling fixes
- Old UVD BO placement fixes
- GC9 mode2 reset fix
- IH6.1 version fix
- Soft IH ring fix
amdkfd:
- Fix doorbell/mmio double unpin on free
- CRIU fixes
- SMI event fixes
- Sysfs teardown fix
- Various boundary checking fixes
- Various error checking fixes
- SVM fix"
* tag 'drm-next-2026-06-27' of https://gitlab.freedesktop.org/drm/kernel: (52 commits)
drm/i915/cdclk: Fix up CDCLK_FREQ_DECIMAL without a full PLL re-enable
drm/i915/gem: Add missing nospec on parallel submit slot
drm/amdgpu: Use system unbound workqueue for soft IH ring
amdgpu/ih6.1: Fix minor version
drm/amdkfd: Use exclusive bounds for SVM split alignment checks
drm/amdgpu/gfx9: Fix Ring and IB test fail after mode2
drm/amdgpu/uvd: Fix forcing MSG, FB BOs into VCPU segment when it isn't at 0 (v2)
drm/amdgpu/uvd: Place VCPU BO only in VRAM for UVD 4.x and older
drm/amdgpu: Fix amdgpu_bo_move() when old_mem and new_mem are both GTT
drm/amdgpu: Respect placement requirements in amdgpu_gtt_mgr functions
drm/amdgpu: Fix context pstate override handling
drm/amdkfd: Use memdup_array_user to copy data from/to user space at kfd ioctls
drm/amdkfd: check find_first_zero_bit before __set_bit on kfd->doorbell_bitmap
drm/amdkfd: Let driver decide buffer size at AMDKFD_IOC_GET_DMABUF_INFO ioctl
drm/amdgpu: fix recursive ww_mutex acquire in amdgpu_devcoredump_format
drm/amdgpu: convert amdgpu_vm_lock_by_pasid() to drm_exec
drm/amdgpu: Don't use UTS_RELEASE directly
drm/amdkfd: Fix NULL deref during sysfs teardown
drm/amdgpu: validate CP_GFX_SHADOW chunk size in CS pass1
drm/amdgpu: check amdgpu_vm_bo_find() result in GET_MAPPING_INFO
...
operators to get out of tricky livelock situations involving caps and
file locks without evicting the problematic client instance on the MDS
side or rebooting the client node both of which can be disruptive.
-----BEGIN PGP SIGNATURE-----
iQFHBAABCgAxFiEEydHwtzie9C7TfviiSn/eOAIR84sFAmo+rNUTHGlkcnlvbW92
QGdtYWlsLmNvbQAKCRBKf944AhHzi++dB/0VE3YM+kdZER9Y1CoPgxB32CSD5Yoz
cscBQCGtFM/0GHj2rMnvyeQ9+uHlQqzo/CMYURUBJ4ciHPLaZUyuJcp8R7bGNgk2
FZ0+9yBqJjjeSJST16MKhW7/NgzuIhg8Y3c+sDrLDThYkbmEM90MYLDpopqwjLhc
RL1mURFBQxuBLQiYujKXit1iXqYapJSRmthJAYN4pJ69w97zF+rO/qfvUGlRxYb3
0Hb4GzD5IyWmsRGwldzc7/CXDLjTjp/N+tu1BLuCQK68U3MAw8GV39/aSocFjxN8
rtq8TkgFb1dfSjwjD8pxei5pjBaoUjABnTg8wPDFoxEMKrVpabMeUTsb
=xeGm
-----END PGP SIGNATURE-----
Merge tag 'ceph-for-7.2-rc1' of https://github.com/ceph/ceph-client
Pull ceph updates from Ilya Dryomov:
"This adds support for manual client session reset in CephFS, allowing
operators to get out of tricky livelock situations involving caps and
file locks without evicting the problematic client instance on the MDS
side or rebooting the client node both of which can be disruptive"
* tag 'ceph-for-7.2-rc1' of https://github.com/ceph/ceph-client:
ceph: add manual reset debugfs control and tracepoints
ceph: add client reset state machine and session teardown
ceph: add diagnostic timeout loop to wait_caps_flush()
ceph: harden send_mds_reconnect and handle active-MDS peer reset
ceph: use proper endian conversion for flock_len in reconnect
ceph: convert inode flags to named bit positions and atomic bitops
rbd: switch to dynamic root device
