Reimplement NEXT_BUDDY preemption to take into account the deadline and
eligibility of the wakee with respect to the waker. In the event
multiple buddies could be considered, the one with the earliest deadline
is selected.
Sync wakeups are treated differently to every other type of wakeup. The
WF_SYNC assumption is that the waker promises to sleep in the very near
future. This is violated in enough cases that WF_SYNC should be treated
as a suggestion instead of a contract. If a waker does go to sleep almost
immediately then the delay in wakeup is negligible. In other cases, it's
throttled based on the accumulated runtime of the waker so there is a
chance that some batched wakeups have been issued before preemption.
For all other wakeups, preemption happens if the wakee has a earlier
deadline than the waker and eligible to run.
While many workloads were tested, the two main targets were a modified
dbench4 benchmark and hackbench because the are on opposite ends of the
spectrum -- one prefers throughput by avoiding preemption and the other
relies on preemption.
First is the dbench throughput data even though it is a poor metric but
it is the default metric. The test machine is a 2-socket machine and the
backing filesystem is XFS as a lot of the IO work is dispatched to kernel
threads. It's important to note that these results are not representative
across all machines, especially Zen machines, as different bottlenecks
are exposed on different machines and filesystems.
dbench4 Throughput (misleading but traditional)
6.18-rc1 6.18-rc1
vanilla sched-preemptnext-v5
Hmean 1 1268.80 ( 0.00%) 1269.74 ( 0.07%)
Hmean 4 3971.74 ( 0.00%) 3950.59 ( -0.53%)
Hmean 7 5548.23 ( 0.00%) 5420.08 ( -2.31%)
Hmean 12 7310.86 ( 0.00%) 7165.57 ( -1.99%)
Hmean 21 8874.53 ( 0.00%) 9149.04 ( 3.09%)
Hmean 30 9361.93 ( 0.00%) 10530.04 ( 12.48%)
Hmean 48 9540.14 ( 0.00%) 11820.40 ( 23.90%)
Hmean 79 9208.74 ( 0.00%) 12193.79 ( 32.42%)
Hmean 110 8573.12 ( 0.00%) 11933.72 ( 39.20%)
Hmean 141 7791.33 ( 0.00%) 11273.90 ( 44.70%)
Hmean 160 7666.60 ( 0.00%) 10768.72 ( 40.46%)
As throughput is misleading, the benchmark is modified to use a short
loadfile report the completion time duration in milliseconds.
dbench4 Loadfile Execution Time
6.18-rc1 6.18-rc1
vanilla sched-preemptnext-v5
Amean 1 14.62 ( 0.00%) 14.69 ( -0.46%)
Amean 4 18.76 ( 0.00%) 18.85 ( -0.45%)
Amean 7 23.71 ( 0.00%) 24.38 ( -2.82%)
Amean 12 31.25 ( 0.00%) 31.87 ( -1.97%)
Amean 21 45.12 ( 0.00%) 43.69 ( 3.16%)
Amean 30 61.07 ( 0.00%) 54.33 ( 11.03%)
Amean 48 95.91 ( 0.00%) 77.22 ( 19.49%)
Amean 79 163.38 ( 0.00%) 123.08 ( 24.66%)
Amean 110 243.91 ( 0.00%) 175.11 ( 28.21%)
Amean 141 343.47 ( 0.00%) 239.10 ( 30.39%)
Amean 160 401.15 ( 0.00%) 283.73 ( 29.27%)
Stddev 1 0.52 ( 0.00%) 0.51 ( 2.45%)
Stddev 4 1.36 ( 0.00%) 1.30 ( 4.04%)
Stddev 7 1.88 ( 0.00%) 1.87 ( 0.72%)
Stddev 12 3.06 ( 0.00%) 2.45 ( 19.83%)
Stddev 21 5.78 ( 0.00%) 3.87 ( 33.06%)
Stddev 30 9.85 ( 0.00%) 5.25 ( 46.76%)
Stddev 48 22.31 ( 0.00%) 8.64 ( 61.27%)
Stddev 79 35.96 ( 0.00%) 18.07 ( 49.76%)
Stddev 110 59.04 ( 0.00%) 30.93 ( 47.61%)
Stddev 141 85.38 ( 0.00%) 40.93 ( 52.06%)
Stddev 160 96.38 ( 0.00%) 39.72 ( 58.79%)
That is still looking good and the variance is reduced quite a bit.
Finally, fairness is a concern so the next report tracks how many
milliseconds does it take for all clients to complete a workfile. This
one is tricky because dbench makes to effort to synchronise clients so
the durations at benchmark start time differ substantially from typical
runtimes. This problem could be mitigated by warming up the benchmark
for a number of minutes but it's a matter of opinion whether that
counts as an evasion of inconvenient results.
dbench4 All Clients Loadfile Execution Time
6.18-rc1 6.18-rc1
vanilla sched-preemptnext-v5
Amean 1 15.06 ( 0.00%) 15.07 ( -0.03%)
Amean 4 603.81 ( 0.00%) 524.29 ( 13.17%)
Amean 7 855.32 ( 0.00%) 1331.07 ( -55.62%)
Amean 12 1890.02 ( 0.00%) 2323.97 ( -22.96%)
Amean 21 3195.23 ( 0.00%) 2009.29 ( 37.12%)
Amean 30 13919.53 ( 0.00%) 4579.44 ( 67.10%)
Amean 48 25246.07 ( 0.00%) 5705.46 ( 77.40%)
Amean 79 29701.84 ( 0.00%) 15509.26 ( 47.78%)
Amean 110 22803.03 ( 0.00%) 23782.08 ( -4.29%)
Amean 141 36356.07 ( 0.00%) 25074.20 ( 31.03%)
Amean 160 17046.71 ( 0.00%) 13247.62 ( 22.29%)
Stddev 1 0.47 ( 0.00%) 0.49 ( -3.74%)
Stddev 4 395.24 ( 0.00%) 254.18 ( 35.69%)
Stddev 7 467.24 ( 0.00%) 764.42 ( -63.60%)
Stddev 12 1071.43 ( 0.00%) 1395.90 ( -30.28%)
Stddev 21 1694.50 ( 0.00%) 1204.89 ( 28.89%)
Stddev 30 7945.63 ( 0.00%) 2552.59 ( 67.87%)
Stddev 48 14339.51 ( 0.00%) 3227.55 ( 77.49%)
Stddev 79 16620.91 ( 0.00%) 8422.15 ( 49.33%)
Stddev 110 12912.15 ( 0.00%) 13560.95 ( -5.02%)
Stddev 141 20700.13 ( 0.00%) 14544.51 ( 29.74%)
Stddev 160 9079.16 ( 0.00%) 7400.69 ( 18.49%)
This is more of a mixed bag but it at least shows that fairness
is not crippled.
The hackbench results are more neutral but this is still important.
It's possible to boost the dbench figures by a large amount but only by
crippling the performance of a workload like hackbench. The WF_SYNC
behaviour is important for these workloads and is why the WF_SYNC
changes are not a separate patch.
hackbench-process-pipes
6.18-rc1 6.18-rc1
vanilla sched-preemptnext-v5
Amean 1 0.2657 ( 0.00%) 0.2150 ( 19.07%)
Amean 4 0.6107 ( 0.00%) 0.6060 ( 0.76%)
Amean 7 0.7923 ( 0.00%) 0.7440 ( 6.10%)
Amean 12 1.1500 ( 0.00%) 1.1263 ( 2.06%)
Amean 21 1.7950 ( 0.00%) 1.7987 ( -0.20%)
Amean 30 2.3207 ( 0.00%) 2.5053 ( -7.96%)
Amean 48 3.5023 ( 0.00%) 3.9197 ( -11.92%)
Amean 79 4.8093 ( 0.00%) 5.2247 ( -8.64%)
Amean 110 6.1160 ( 0.00%) 6.6650 ( -8.98%)
Amean 141 7.4763 ( 0.00%) 7.8973 ( -5.63%)
Amean 172 8.9560 ( 0.00%) 9.3593 ( -4.50%)
Amean 203 10.4783 ( 0.00%) 10.8347 ( -3.40%)
Amean 234 12.4977 ( 0.00%) 13.0177 ( -4.16%)
Amean 265 14.7003 ( 0.00%) 15.5630 ( -5.87%)
Amean 296 16.1007 ( 0.00%) 17.4023 ( -8.08%)
Processes using pipes are impacted but the variance (not presented) indicates
it's close to noise and the results are not always reproducible. If executed
across multiple reboots, it may show neutral or small gains so the worst
measured results are presented.
Hackbench using sockets is more reliably neutral as the wakeup
mechanisms are different between sockets and pipes.
hackbench-process-sockets
6.18-rc1 6.18-rc1
vanilla sched-preemptnext-v2
Amean 1 0.3073 ( 0.00%) 0.3263 ( -6.18%)
Amean 4 0.7863 ( 0.00%) 0.7930 ( -0.85%)
Amean 7 1.3670 ( 0.00%) 1.3537 ( 0.98%)
Amean 12 2.1337 ( 0.00%) 2.1903 ( -2.66%)
Amean 21 3.4683 ( 0.00%) 3.4940 ( -0.74%)
Amean 30 4.7247 ( 0.00%) 4.8853 ( -3.40%)
Amean 48 7.6097 ( 0.00%) 7.8197 ( -2.76%)
Amean 79 14.7957 ( 0.00%) 16.1000 ( -8.82%)
Amean 110 21.3413 ( 0.00%) 21.9997 ( -3.08%)
Amean 141 29.0503 ( 0.00%) 29.0353 ( 0.05%)
Amean 172 36.4660 ( 0.00%) 36.1433 ( 0.88%)
Amean 203 39.7177 ( 0.00%) 40.5910 ( -2.20%)
Amean 234 42.1120 ( 0.00%) 43.5527 ( -3.42%)
Amean 265 45.7830 ( 0.00%) 50.0560 ( -9.33%)
Amean 296 50.7043 ( 0.00%) 54.3657 ( -7.22%)
As schbench has been mentioned in numerous bugs recently, the results
are interesting. A test case that represents the default schbench
behaviour is
schbench Wakeup Latency (usec)
6.18.0-rc1 6.18.0-rc1
vanilla sched-preemptnext-v5
Amean Wakeup-50th-80 7.17 ( 0.00%) 6.00 ( 16.28%)
Amean Wakeup-90th-80 46.56 ( 0.00%) 19.78 ( 57.52%)
Amean Wakeup-99th-80 119.61 ( 0.00%) 89.94 ( 24.80%)
Amean Wakeup-99.9th-80 3193.78 ( 0.00%) 328.22 ( 89.72%)
schbench Requests Per Second (ops/sec)
6.18.0-rc1 6.18.0-rc1
vanilla sched-preemptnext-v5
Hmean RPS-20th-80 8900.91 ( 0.00%) 9176.78 ( 3.10%)
Hmean RPS-50th-80 8987.41 ( 0.00%) 9217.89 ( 2.56%)
Hmean RPS-90th-80 9123.73 ( 0.00%) 9273.25 ( 1.64%)
Hmean RPS-max-80 9193.50 ( 0.00%) 9301.47 ( 1.17%)
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20251112122521.1331238-3-mgorman@techsingularity.net
The NEXT_BUDDY feature reinforces wakeup preemption to encourage the last
wakee to be scheduled sooner on the assumption that the waker/wakee share
cache-hot data. In CFS, it was paired with LAST_BUDDY to switch back on
the assumption that the pair of tasks still share data but also relied
on START_DEBIT and the exact WAKEUP_PREEMPTION implementation to get
good results.
NEXT_BUDDY has been disabled since commit 0ec9fab3d1 ("sched: Improve
latencies and throughput") and LAST_BUDDY was removed in commit 5e963f2bd4
("sched/fair: Commit to EEVDF"). The reasoning is not clear but as vruntime
spread is mentioned so the expectation is that NEXT_BUDDY had an impact
on overall fairness. It was not noted why LAST_BUDDY was removed but it
is assumed that it's very difficult to reason what LAST_BUDDY's correct
and effective behaviour should be while still respecting EEVDFs goals.
Peter Zijlstra noted during review;
I think I was just struggling to make sense of things and figured
less is more and axed it.
I have vague memories trying to work through the dynamics of
a wakeup-stack and the EEVDF latency requirements and getting
a head-ache.
NEXT_BUDDY is easier to reason about given that it's a point-in-time
decision on the wakees deadline and eligibilty relative to the waker. Enable
NEXT_BUDDY as a preparation path to document that the decision to ignore
the current implementation is deliberate. While not presented, the results
were at best neutral and often much more variable.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20251112122521.1331238-2-mgorman@techsingularity.net
The NUMA sched domain sets the SD_SERIALIZE flag by default, allowing
only one NUMA load balancing operation to run system-wide at a time.
Currently, each sched group leader directly under NUMA domain attempts
to acquire the global sched_balance_running flag via cmpxchg() before
checking whether load balancing is due or whether it is the designated
load balancer for that NUMA domain. On systems with a large number
of cores, this causes significant cache contention on the shared
sched_balance_running flag.
This patch reduces unnecessary cmpxchg() operations by first checking
that the balancer is the designated leader for a NUMA domain from
should_we_balance(), and the balance interval has expired before
trying to acquire sched_balance_running to load balance a NUMA
domain.
On a 2-socket Granite Rapids system with sub-NUMA clustering enabled,
running an OLTP workload, 7.8% of total CPU cycles were previously spent
in sched_balance_domain() contending on sched_balance_running before
this change.
: 104 static __always_inline int arch_atomic_cmpxchg(atomic_t *v, int old, int new)
: 105 {
: 106 return arch_cmpxchg(&v->counter, old, new);
0.00 : ffffffff81326e6c: xor %eax,%eax
0.00 : ffffffff81326e6e: mov $0x1,%ecx
0.00 : ffffffff81326e73: lock cmpxchg %ecx,0x2394195(%rip) # ffffffff836bb010 <sched_balance_running>
: 110 sched_balance_domains():
: 12234 if (atomic_cmpxchg_acquire(&sched_balance_running, 0, 1))
99.39 : ffffffff81326e7b: test %eax,%eax
0.00 : ffffffff81326e7d: jne ffffffff81326e99 <sched_balance_domains+0x209>
: 12238 if (time_after_eq(jiffies, sd->last_balance + interval)) {
0.00 : ffffffff81326e7f: mov 0x14e2b3a(%rip),%rax # ffffffff828099c0 <jiffies_64>
0.00 : ffffffff81326e86: sub 0x48(%r14),%rax
0.00 : ffffffff81326e8a: cmp %rdx,%rax
After applying this fix, sched_balance_domain() is gone from the profile
and there is a 5% throughput improvement.
[peterz: made it so that redo retains the 'lock' and split out the
CPU_NEWLY_IDLE change to a separate patch]
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Reviewed-by: Srikar Dronamraju <srikar@linux.ibm.com>
Tested-by: Mohini Narkhede <mohini.narkhede@intel.com>
Tested-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://patch.msgid.link/6fed119b723c71552943bfe5798c93851b30a361.1762800251.git.tim.c.chen@linux.intel.com
cpumask_subset(a,b) -> cpumask_weight(a) should be same as cpumask_weight_and(a,b)
for_each_cpu_and(a,b) to count cpus could be replaced by cpumask_weight_and(a,b)
No Functional Change. It could save a few cycles since cpumask_weight_and
would be more efficient. Plus one less stack variable.
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://patch.msgid.link/20251014100342.978936-3-sshegde@linux.ibm.com
Gabriel reported that the dl_server doesn't stop as expected.
The problem was found to be the fact that idle time and fair runtime are
treated equally. Both will count towards dl_server runtime and push the
activation forwards when it is in the zero-laxity wait state.
Notably:
dl_server_update_idle()
update_curr_dl_se()
if (dl_defer && dl_throttled && dl_runtime_exceeded())
hrtimer_try_to_cancel(); // stop timer
replenish_dl_new_period()
deadline = now + dl_deadline; // fwd period
runtime = dl_runtime;
start_dl_timer(); // restart timer
And while we do want idle time accounted towards the *current* activation of
the dl_server -- after all, a fair task could've ran if we had any -- we don't
necessarily want idle time to cause or push forward an activation.
Introduce dl_defer_idle to make this distinction. It will be set once idle time
pushed the activation forward, once set idle time will only be allowed to
consume any runtime but not push the activation. This will then cause
dl_server_timer() to fire, which will stop the dl_server.
Any non-idle time accounting during this phase will clear dl_defer_idle, so
only a full period of idle will cause the dl_server to stop.
Reported-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20251101000057.GA2184199@noisy.programming.kicks-ass.net
The dl_server time accounting code is a little odd. The normal scheduler
pattern is to update curr before doing something, such that the old state is
fully accounted before changing state.
Notably, the dl_server_timer() needs to propagate the current time accounting
since the current task could be ran by dl_server and thus this can affect
dl_se->runtime. Similarly for dl_server_start().
And since the (deferred) dl_server wants idle time accounted, rework
sched_idle_class time accounting to be more like all the others.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20251020141130.GJ3245006@noisy.programming.kicks-ass.net
Since commit d4c64207b8 ("sched: Cleanup the sched_change NOCLOCK usage"),
update_rq_clock() is called in do_set_cpus_allowed() -> sched_change_begin()
to update the rq clock. This results in a duplicate call update_rq_clock()
in __set_cpus_allowed_ptr_locked().
While holding the rq lock and before calling do_set_cpus_allowed(),
there is nothing that depends on an updated rq_clock.
Therefore, remove the redundant update_rq_clock() in
__set_cpus_allowed_ptr_locked() to avoid the warning about double
rq clock updates.
Fixes: d4c64207b8 ("sched: Cleanup the sched_change NOCLOCK usage")
Signed-off-by: Hao Jia <jiahao1@lixiang.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20251029093655.31252-1-jiahao.kernel@gmail.com
Basically, from the constraint that the sum of lag is zero, you can
infer that the 0-lag point is the weighted average of the individual
vruntime, which is what we're trying to compute:
\Sum w_i * v_i
avg = --------------
\Sum w_i
Now, since vruntime takes the whole u64 (worse, it wraps), this
multiplication term in the numerator is not something we can compute;
instead we do the min_vruntime (v0 henceforth) thing like:
v_i = (v_i - v0) + v0
This does two things:
- it keeps the key: (v_i - v0) 'small';
- it creates a relative 0-point in the modular space.
If you do that subtitution and work it all out, you end up with:
\Sum w_i * (v_i - v0)
avg = --------------------- + v0
\Sum w_i
Since you cannot very well track a ratio like that (and not suffer
terrible numerical problems) we simpy track the numerator and
denominator individually and only perform the division when strictly
needed.
Notably, the numerator lives in cfs_rq->avg_vruntime and the denominator
lives in cfs_rq->avg_load.
The one extra 'funny' is that these numbers track the entities in the
tree, and current is typically outside of the tree, so avg_vruntime()
adds current when needed before doing the division.
(vruntime_eligible() elides the division by cross-wise multiplication)
Anyway, as mentioned above, we currently use the CFS era min_vruntime
for this purpose. However, this thing can only move forward, while the
above avg can in fact move backward (when a non-eligible task leaves,
the average becomes smaller), this can cause trouble when through
happenstance (or construction) these values drift far enough apart to
wreck the game.
Replace cfs_rq::min_vruntime with cfs_rq::zero_vruntime which is kept
near/at avg_vruntime, following its motion.
The down-side is that this requires computing the avg more often.
Fixes: 147f3efaa2 ("sched/fair: Implement an EEVDF-like scheduling policy")
Reported-by: Zicheng Qu <quzicheng@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20251106111741.GC4068168@noisy.programming.kicks-ass.net
Cc: stable@vger.kernel.org
Commit abfc01077d ("sched: Fix do_set_cpus_allowed() locking")
overlooked that __balance_push_cpu_stop() calls select_fallback_rq()
with rq->lock held. This makes that set_cpus_allowed_force() will
recursively take rq->lock and the machine locks up.
Run select_fallback_rq() earlier, without holding rq->lock. This opens
up a race window where a task could get migrated out from under us, but
that is harmless, we want the task migrated.
select_fallback_rq() itself will not be subject to concurrency as it
will be fully serialized by p->pi_lock, so there is no chance of
set_cpus_allowed_force() getting called with different arguments and
selecting different fallback CPUs for one task.
Fixes: abfc01077d ("sched: Fix do_set_cpus_allowed() locking")
Reported-by: Jan Polensky <japo@linux.ibm.com>
Reported-by: kernel test robot <oliver.sang@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Jan Polensky <japo@linux.ibm.com>
Closes: https://lore.kernel.org/oe-lkp/202510271206.24495a68-lkp@intel.com
Link: https://patch.msgid.link/20251027110133.GI3245006@noisy.programming.kicks-ass.net
Pull sched_ext fixes from Tejun Heo:
- Fix scx_kick_pseqs corruption when multiple schedulers are loaded
concurrently
- Allocate scx_kick_cpus_pnt_seqs lazily using kvzalloc() to handle
systems with large CPU counts
- Defer queue_balance_callback() until after ops.dispatch to fix
callback ordering issues
- Sync error_irq_work before freeing scx_sched to prevent
use-after-free
- Mark scx_bpf_dsq_move_set_[slice|vtime]() with KF_RCU for proper RCU
protection
- Fix flag check for deferred callbacks
* tag 'sched_ext-for-6.18-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: fix flag check for deferred callbacks
sched_ext: Fix scx_kick_pseqs corruption on concurrent scheduler loads
sched_ext: Allocate scx_kick_cpus_pnt_seqs lazily using kvzalloc()
sched_ext: defer queue_balance_callback() until after ops.dispatch
sched_ext: Sync error_irq_work before freeing scx_sched
sched_ext: Mark scx_bpf_dsq_move_set_[slice|vtime]() with KF_RCU
Pull char/misc driver fixes from Greg KH:
"Here are some small char/misc/android driver fixes for 6.18-rc3 for
reported issues. Included in here are:
- rust binder fixes for reported issues
- mei device id addition
- mei driver fixes
- comedi bugfix
- most usb driver bugfixes
- fastrpc memory leak fix
All of these have been in linux-next for a while with no reported
issues"
* tag 'char-misc-6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
most: usb: hdm_probe: Fix calling put_device() before device initialization
most: usb: Fix use-after-free in hdm_disconnect
binder: remove "invalid inc weak" check
mei: txe: fix initialization order
comedi: fix divide-by-zero in comedi_buf_munge()
mei: late_bind: Fix -Wincompatible-function-pointer-types-strict
misc: fastrpc: Fix dma_buf object leak in fastrpc_map_lookup
mei: me: add wildcat lake P DID
misc: amd-sbi: Clarify that this is a BMC driver
nvmem: rcar-efuse: add missing MODULE_DEVICE_TABLE
binder: Fix missing kernel-doc entries in binder.c
rust_binder: report freeze notification only when fully frozen
rust_binder: don't delete FreezeListener if there are pending duplicates
rust_binder: freeze_notif_done should resend if wrong state
rust_binder: remove warning about orphan mappings
rust_binder: clean `clippy::mem_replace_with_default` warning
Pull staging driver fixes from Greg KH:
"Here are some small staging driver fixes for the gpib subsystem to
resolve some reported issues. Included in here are:
- memory leak fixes
- error code fixes
- proper protocol fixes
All of these have been in linux-next for almost 2 weeks now with no
reported issues"
* tag 'staging-6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
staging: gpib: Fix device reference leak in fmh_gpib driver
staging: gpib: Return -EINTR on device clear
staging: gpib: Fix sending clear and trigger events
staging: gpib: Fix no EOI on 1 and 2 byte writes
Pull tty/serial driver fixes from Greg KH:
"Here are some small tty and serial driver fixes for reported issues.
Included in here are:
- sh-sci serial driver fixes
- 8250_dw and _mtk driver fixes
- sc16is7xx driver bugfix
- new 8250_exar device ids added
All of these have been in linux-next this past week with no reported
issues"
* tag 'tty-6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
serial: 8250_mtk: Enable baud clock and manage in runtime PM
serial: 8250_dw: handle reset control deassert error
dt-bindings: serial: sh-sci: Fix r8a78000 interrupts
serial: sc16is7xx: remove useless enable of enhanced features
serial: 8250_exar: add support for Advantech 2 port card with Device ID 0x0018
tty: serial: sh-sci: fix RSCI FIFO overrun handling
Pull USB driver fixes from Greg KH:
"Here are some small USB driver fixes and new device ids for 6.18-rc3.
Included in here are:
- new option serial driver device ids added
- dt bindings fixes for numerous platforms
- xhci bugfixes for many reported regressions
- usbio dependency bugfix
- dwc3 driver fix
- raw-gadget bugfix
All of these have been in linux-next this week with no reported issues"
* tag 'usb-6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
USB: serial: option: add Telit FN920C04 ECM compositions
USB: serial: option: add Quectel RG255C
tcpm: switch check for role_sw device with fw_node
usb/core/quirks: Add Huawei ME906S to wakeup quirk
usb: raw-gadget: do not limit transfer length
USB: serial: option: add UNISOC UIS7720
xhci: dbc: enable back DbC in resume if it was enabled before suspend
xhci: dbc: fix bogus 1024 byte prefix if ttyDBC read races with stall event
usb: xhci-pci: Fix USB2-only root hub registration
dt-bindings: usb: qcom,snps-dwc3: Fix bindings for X1E80100
usb: misc: Add x86 dependency for Intel USBIO driver
dt-bindings: usb: switch: split out ports definition
usb: dwc3: Don't call clk_bulk_disable_unprepare() twice
dt-bindings: usb: dwc3-imx8mp: dma-range is required only for imx8mp
Pull x86 fixes from Borislav Petkov:
- Remove dead code leftovers after a recent mitigations cleanup which
fail a Clang build
- Make sure a Retbleed mitigation message is printed only when
necessary
- Correct the last Zen1 microcode revision for which Entrysign sha256
check is needed
- Fix a NULL ptr deref when mounting the resctrl fs on a system which
supports assignable counters but where L3 total and local bandwidth
monitoring has been disabled at boot
* tag 'x86_urgent_for_v6.18_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/bugs: Remove dead code which might prevent from building
x86/bugs: Qualify RETBLEED_INTEL_MSG
x86/microcode: Fix Entrysign revision check for Zen1/Naples
x86,fs/resctrl: Fix NULL pointer dereference with events force-disabled in mbm_event mode
Pull irq fixes from Borislav Petkov:
- Restore the original buslock locking in a couple of places in the irq
core subsystem after a rework
* tag 'irq_urgent_for_v6.18_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
genirq/manage: Add buslock back in to enable_irq()
genirq/manage: Add buslock back in to __disable_irq_nosync()
genirq/chip: Add buslock back in to irq_set_handler()
Pull objtool fixes from Borislav Petkov:
- Fix x32 build due to wrong format specifier on that sub-arch
- Add one more Rust noreturn function to objtool's list
* tag 'objtool_urgent_for_v6.18_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
objtool: Fix failure when being compiled on x32 system
objtool/rust: add one more `noreturn` Rust function
Pull scheduler fix from Borislav Petkov:
- Make sure a CFS runqueue on a throttled hierarchy has its PELT clock
throttled otherwise task movement and manipulation would lead to
dangling cfs_rq references and an eventual crash
* tag 'sched_urgent_for_v6.18_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Start a cfs_rq on throttled hierarchy with PELT clock throttled
Pull timer fix from Borislav Petkov:
- Do not create more than eight (max supported) AUX clocks sysfs
hierarchies
* tag 'timers_urgent_for_v6.18_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
timekeeping: Fix aux clocks sysfs initialization loop bound
Pull driver core fixes from Danilo Krummrich:
- In Device::parent(), do not make any assumptions on the device
context of the parent device
- Check visibility before changing ownership of a sysfs attribute
group
- In topology_parse_cpu_capacity(), replace an incorrect usage of
PTR_ERR_OR_ZERO() with IS_ERR_OR_NULL()
- In devcoredump, fix a circular locking dependency between
struct devcd_entry::mutex and kernfs
- Do not warn about a pending fw_devlink sync state
* tag 'driver-core-6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/driver-core/driver-core:
arch_topology: Fix incorrect error check in topology_parse_cpu_capacity()
rust: device: fix device context of Device::parent()
sysfs: check visibility before changing group attribute ownership
devcoredump: Fix circular locking dependency with devcd->mutex.
driver core: fw_devlink: Don't warn about sync_state() pending
Pull firewire fixes from Takashi Sakamoto:
"A small collection of FireWire fixes. This includes corrections to
sparse and API documentation"
* tag 'firewire-fixes-6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394:
firewire: init_ohci1394_dma: add missing function parameter documentation
firewire: core: fix __must_hold() annotation
Pull RISC-V fixes from Paul Walmsley:
- Close a race during boot between userspace vDSO usage and some
late-initialized vDSO data
- Improve performance on systems with non-CPU-cache-coherent
DMA-capable peripherals by enabling write combining on
pgprot_dmacoherent() allocations
- Add human-readable detail for RISC-V IPI tracing
- Provide more information to zsmalloc on 64-bit RISC-V to improve
allocation
- Silence useless boot messages about CPUs that have been disabled in
DT
- Resolve some compiler and smatch warnings and remove a redundant
macro
* tag 'riscv-for-linus-6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
riscv: hwprobe: avoid uninitialized variable use in hwprobe_arch_id()
riscv: cpufeature: avoid uninitialized variable in has_thead_homogeneous_vlenb()
riscv: hwprobe: Fix stale vDSO data for late-initialized keys at boot
riscv: add a forward declaration for cpuinfo_op
RISC-V: Don't print details of CPUs disabled in DT
riscv: Remove the PER_CPU_OFFSET_SHIFT macro
riscv: mm: Define MAX_POSSIBLE_PHYSMEM_BITS for zsmalloc
riscv: Register IPI IRQs with unique names
ACPI: RIMT: Fix unused function warnings when CONFIG_IOMMU_API is disabled
RISC-V: Define pgprot_dmacoherent() for non-coherent devices
Pull xfs fixes from Carlos Maiolino:
"The main highlight here is a fix for a bug brought in by the removal
of attr2 mount option, where some installations might actually have
'attr2' explicitly configured in fstab preventing system to boot by
not being able to remount the rootfs as RW.
Besides that there are a couple fix to the zonefs implementation,
changing XFS_ONLINE_SCRUB_STATS to depend on DEBUG_FS (was select
before), and some other minor changes"
* tag 'xfs-fixes-6.18-rc3' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: fix locking in xchk_nlinks_collect_dir
xfs: loudly complain about defunct mount options
xfs: always warn about deprecated mount options
xfs: don't set bt_nr_sectors to a negative number
xfs: don't use __GFP_NOFAIL in xfs_init_fs_context
xfs: cache open zone in inode->i_private
xfs: avoid busy loops in GCD
xfs: XFS_ONLINE_SCRUB_STATS should depend on DEBUG_FS
xfs: do not tightly pack-write large files
xfs: Improve CONFIG_XFS_RT Kconfig help
Pull smb server fixes from Steve French:
"smbdirect (RDMA) fixes in order avoid potential submission queue
overflows:
- free transport teardown fix
- credit related fixes (five server related, one client related)"
* tag 'v6.18-rc2-smb-server-fixes' of git://git.samba.org/ksmbd:
smb: server: let free_transport() wait for SMBDIRECT_SOCKET_DISCONNECTED
smb: client: make use of smbdirect_socket.send_io.lcredits.*
smb: server: make use of smbdirect_socket.send_io.lcredits.*
smb: server: simplify sibling_list handling in smb_direct_flush_send_list/send_done
smb: server: smb_direct_disconnect_rdma_connection() already wakes all waiters on error
smb: smbdirect: introduce smbdirect_socket.send_io.lcredits.*
smb: server: allocate enough space for RW WRs and ib_drain_qp()
Clang is not happy with set but unused variable (this is visible
with `make W=1` build:
kernel/sched/sched.h:3744:18: error: variable 'cpumask' set but not used [-Werror,-Wunused-but-set-variable]
It seems like the variable was never used along with the assignment
that does not have side effects as far as I can see. Remove those
altogether.
Fixes: 223baf9d17 ("sched: Fix performance regression introduced by mm_cid")
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Eric Biggers <ebiggers@kernel.org>
Reviewed-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull drm fixes from Simona Vetter:
"Very quiet, all just small stuff and nothing scary pending to my
knowledge:
- drm_panic: bunch of size calculation fixes
- pantor: fix kernel panic on partial gpu va unmap
- rockchip: hdmi hotplug setup fix
- amdgpu: dp mst, dc/display fixes
- i915: fix panic structure leak
- xe: madvise uapi fix, wq alloc error, vma flag handling fix"
* tag 'drm-fixes-2025-10-24' of https://gitlab.freedesktop.org/drm/kernel:
drm/xe: Check return value of GGTT workqueue allocation
drm/amd/display: use GFP_NOWAIT for allocation in interrupt handler
drm/amd/display: increase max link count and fix link->enc NULL pointer access
drm/amd/display: Fix NULL pointer dereference
drm/panic: Fix 24bit pixel crossing page boundaries
drm/panic: Fix divide by 0 if the screen width < font width
drm/panic: Fix kmsg text drawing rectangle
drm/panic: Fix qr_code, ensure vmargin is positive
drm/panic: Fix overlap between qr code and logo
drm/panic: Fix drawing the logo on a small narrow screen
drm/xe/uapi: Hide the madvise autoreset behind a VM_BIND flag
drm/xe: Retain vma flags when recreating and splitting vmas for madvise
drm/i915/panic: fix panic structure allocation memory leak
drm/panthor: Fix kernel panic on partial unmap of a GPU VA region
drm/rockchip: dw_hdmi: use correct SCLIN mask for RK3228
Pull pci fixes from Bjorn Helgaas:
- Add DWC custom pci_ops for the root bus instead of overwriting the
DBI base address, which broke drivers that rely on the DBI address
for iATU programming; fixes an FU740 probe regression (Krishna
Chaitanya Chundru)
- Revert qcom ECAM enablement, which is rendered unnecessary by the DWC
custom pci_ops (Krishna Chaitanya Chundru)
- Fix longstanding MIPS Malta resource registration issues to avoid
exposing them when the next commit fixes the boot failure (Maciej W.
Rozycki)
- Use pcibios_align_resource() on MIPS Malta to fix boot failure caused
by using the generic pci_enable_resources() (Ilpo Järvinen)
- Enable only ASPM L0s and L1, not L1 PM Substates, for devicetree
platforms because we lack information required to configure L1
Substates; fixes regressions on powerpc and rockchip. A qcom
regression (L1 Substates no longer enabled) remains and will be
addressed next (Bjorn Helgaas)
* tag 'pci-v6.18-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/pci/pci:
PCI/ASPM: Enable only L0s and L1 for devicetree platforms
MIPS: Malta: Use pcibios_align_resource() to block io range
MIPS: Malta: Fix PCI southbridge legacy resource reservations
MIPS: Malta: Fix keyboard resource preventing i8042 driver from registering
Revert "PCI: qcom: Prepare for the DWC ECAM enablement"
PCI: dwc: Use custom pci_ops for root bus DBI vs ECAM config access
Add missing kernel-doc parameter descriptions for five functions
in init_ohci1394_dma.c to fix documentation warnings when building
with W=1.
This patch addresses the following warnings:
- init_ohci1394_wait_for_busresets: missing @ohci description
- init_ohci1394_enable_physical_dma: missing @ohci description
- init_ohci1394_reset_and_init_dma: missing @ohci description
- init_ohci1394_controller: missing @num, @slot, @func descriptions
- setup_ohci1394_dma: missing @opt description
Tested with GCC 13.2.0 and W=1 flag. All documentation warnings
for these functions have been resolved.
Signed-off-by: Nirbhay Sharma <nirbhay.lkd@gmail.com>
Link: https://lore.kernel.org/r/20251024203219.101990-2-nirbhay.lkd@gmail.com
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Pull crypto library fix from Eric Biggers:
"Avoid some false-positive KMSAN warnings by restoring the dependency
of the architecture-optimized Poly1305 code on !KMSAN"
* tag 'libcrypto-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux:
lib/crypto: poly1305: Restore dependency of arch code on !KMSAN
Pull smb client fixes from Steve French:
- add missing tracepoints
- smbdirect (RDMA) fix
- fix potential issue with credits underflow
- rename fix
- improvement to calc_signature and additional cleanup patch
* tag '6.18-rc2-smb-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
cifs: #include cifsglob.h before trace.h to allow structs in tracepoints
cifs: Call the calc_signature functions directly
smb: client: get rid of d_drop() in cifs_do_rename()
cifs: Fix TCP_Server_Info::credits to be signed
cifs: Add a couple of missing smb3_rw_credits tracepoints
smb: client: allocate enough space for MR WRs and ib_drain_qp()
Pull block fixes from Jens Axboe:
- Fix dma alignment for PI
- Fix selinux bogosity with nbd, where sendmsg would get rejected
* tag 'block-6.18-20251023' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
block: require LBA dma_alignment when using PI
nbd: override creds to kernel when calling sock_{send,recv}msg()
Pull io_uring fixes from Jens Axboe:
- Add MAINTAINERS entry for zcrx, mostly so that netdev gets
automatically CC'ed by default on any changes there too.
- Fix for the SQPOLL busy vs work time accounting.
It was using getrusage(), which was both broken from a thread point
of view (we only care about the SQPOLL thread itself), and vastly
overkill as only the systime was used. On top of that, also be a bit
smarter in when it's queried. It used excessive CPU before this
change. Marked for stable as well.
- Fix provided ring buffer auto commit for uring_cmd.
- Fix a few style issues and sparse annotation for a lock.
* tag 'io_uring-6.18-20251023' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
io_uring: fix buffer auto-commit for multishot uring_cmd
io_uring: correct __must_hold annotation in io_install_fixed_file
io_uring zcrx: add MAINTAINERS entry
io_uring: Fix code indentation error
io_uring/sqpoll: be smarter on when to update the stime usage
io_uring/sqpoll: switch away from getrusage() for CPU accounting
io_uring: fix incorrect unlikely() usage in io_waitid_prep()
Pull slab fixes from Vlastimil Babka:
- Two fixes for race conditions in obj_exts allocation (Hao Ge)
- Fix for slab accounting imbalance due to deferred slab decativation
(Vlastimil Babka)
* tag 'slab-for-6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab:
slab: Fix obj_ext mistakenly considered NULL due to race condition
slab: fix slab accounting imbalance due to defer_deactivate_slab()
slab: Avoid race on slab->obj_exts in alloc_slab_obj_exts
Pull devicetree fixes from Rob Herring:
- Fix handling of GICv5 ITS MSI properties on platforms with
'msi-parent' as well as a of_node refcounting fix.
This is also preparation for further refactoring in 6.19 to use
common DT parsing of MSI properties.
* tag 'devicetree-fixes-for-6.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
of/irq: Export of_msi_xlate() for module usage
of/irq: Fix OF node refcount in of_msi_get_domain()
of/irq: Add msi-parent check to of_msi_xlate()
Pull SoC fixes from Arnd Bergmann:
"The main change this time is an update to the MAINTAINERS file,
listing Krzysztof Kozlowski, Alexandre Belloni, and Linus Walleij as
additional maintainers for the SoC tree, in order to go back to a
group maintainership. Drew Fustini joins as an additional reviewer for
the SoC tree.
Thanks to all of you for volunteering to help out.
On the actual bugfixes, we have a few correctness changes for firmware
drivers (qtee, arm-ffa, scmi) and two devicetree fixes for Raspberry
Pi"
* tag 'soc-fixes-6.18-2' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:
soc: officially expand maintainership team
firmware: arm_scmi: Fix premature SCMI_XFER_FLAG_IS_RAW clearing in raw mode
firmware: arm_scmi: Skip RAW initialization on failure
include: trace: Fix inflight count helper on failed initialization
firmware: arm_scmi: Account for failed debug initialization
ARM: dts: broadcom: rpi: Switch to V3D firmware clock
arm64: dts: broadcom: bcm2712: Define VGIC interrupt
firmware: arm_ffa: Add support for IMPDEF value in the memory access descriptor
tee: QCOMTEE should depend on ARCH_QCOM
tee: qcom: return -EFAULT instead of -EINVAL if copy_from_user() fails
tee: qcom: prevent potential off by one read
Pull spi fixes from Mark Brown:
"A moderately large collection of device specific changes here, mostly
fixes but also including a few new quirks and device IDs. This is all
fairly routine even for the affected devices"
* tag 'spi-fix-v6.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: dt-bindings: spi-rockchip: Add RK3506 compatible
spi: intel-pci: Add support for Intel Wildcat Lake SPI serial flash
spi: intel-pci: Add support for Arrow Lake-H SPI serial flash
spi: intel: Add support for 128M component density
spi: airoha: fix reading/writing of flashes with more than one plane per lun
spi: airoha: switch back to non-dma mode in the case of error
spi: airoha: add support of dual/quad wires spi modes to exec_op() handler
spi: airoha: return an error for continuous mode dirmap creation cases
spi: amlogic: fix spifc build error
spi: cadence-quadspi: Fix pm_runtime unbalance on dma EPROBE_DEFER
spi: spi-nxp-fspi: limit the clock rate for different sample clock source selection
spi: spi-nxp-fspi: add extra delay after dll locked
spi: spi-nxp-fspi: re-config the clock rate when operation require new clock rate
spi: dw-mmio: add error handling for reset_control_deassert()
spi: rockchip-sfc: Fix DMA-API usage
spi: dt-bindings: cadence: add soc-specific compatible strings for zynqmp and versal-net
Pull gpio fixes from Bartosz Golaszewski:
- fix regressions in regmap cache initialization in gpio-104-idio-16
and gpio-pci-idio-16
- configure first 16 GPIO lines of the IDIO-16 as fixed outputs
- fix duplicated IRQ mapping that can lead to an RCU stall in gpio-ljca
- fix printf formatters passed to dev_err() and make failure to set
debounce period non fatal
* tag 'gpio-fixes-for-v6.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
gpio: ljca: Fix duplicated IRQ mapping
gpiolib: acpi: Use %pe when passing an error pointer to dev_err()
gpiolib: acpi: Make set debounce errors non fatal
gpio: idio-16: Define fixed direction of the GPIO lines
gpio: regmap: add the .fixed_direction_output configuration parameter
gpio: pci-idio-16: Define maximum valid register address offset
gpio: 104-idio-16: Define maximum valid register address offset
Since Olof moved on from the soc tree maintenance, Arnd has mainly taken
care of the day-to-day activities around the SoC tree by himself, which
is generally not a good setup.
Krzysztof, Linus and Alexandre have volunteered to become co-maintainers
of the SoC tree, with the plan of taking turns to do merges and reviews
to spread the workload. In addition, Drew joins as another reviewer.
Acked-by: Krzysztof Kozlowski <krzk@kernel.org>
Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Drew Fustini <fustini@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Mel Gorman
Phil Auld
Peter Zijlstra
Tim Chen
Shrikanth Hegde
Hao Jia
Fernand Sieber
Linus Torvalds
Andy Shevchenko
Nirbhay Sharma
Arnd Bergmann
Lorenzo Pieralisi