[ Upstream commit 3172fb9866 ]
There may be concurrency between perf_cgroup_switch and
perf_cgroup_event_disable. Consider the following scenario: after a new
perf cgroup event is created on CPU0, the new event may not trigger
a reprogramming, causing ctx->is_active to be 0. In this case, when CPU1
disables this perf event, it executes __perf_remove_from_context->
list _del_event->perf_cgroup_event_disable on CPU1, which causes a race
with perf_cgroup_switch running on CPU0.
The following describes the details of this concurrency scenario:
CPU0 CPU1
perf_cgroup_switch:
...
# cpuctx->cgrp is not NULL here
if (READ_ONCE(cpuctx->cgrp) == NULL)
return;
perf_remove_from_context:
...
raw_spin_lock_irq(&ctx->lock);
...
# ctx->is_active == 0 because reprogramm is not
# tigger, so CPU1 can do __perf_remove_from_context
# for CPU0
__perf_remove_from_context:
perf_cgroup_event_disable:
...
if (--ctx->nr_cgroups)
...
# this warning will happened because CPU1 changed
# ctx.nr_cgroups to 0.
WARN_ON_ONCE(cpuctx->ctx.nr_cgroups == 0);
[peterz: use guard instead of goto unlock]
Fixes: db4a835601 ("perf/core: Set cgroup in CPU contexts for new cgroup events")
Signed-off-by: Luo Gengkun <luogengkun@huaweicloud.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20250604033924.3914647-3-luogengkun@huaweicloud.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 61988e36dc ]
While chasing down a missing perf_cgroup_event_disable() elsewhere,
Leo Yan found that both perf_put_aux_event() and
perf_remove_sibling_event() were also missing one.
Specifically, the rule is that events that switch to OFF,ERROR need to
call perf_cgroup_event_disable().
Unify the disable paths to ensure this.
Fixes: ab43762ef0 ("perf: Allow normal events to output AUX data")
Fixes: 9f0c4fa111 ("perf/core: Add a new PERF_EV_CAP_SIBLING event capability")
Reported-by: Leo Yan <leo.yan@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20250605123343.GD35970@noisy.programming.kicks-ass.net
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4f6fc78212 ]
Baisheng Gao reported an ARM64 crash, which Mark decoded as being a
synchronous external abort -- most likely due to trying to access
MMIO in bad ways.
The crash further shows perf trying to do a user stack sample while in
exit_mmap()'s tlb_finish_mmu() -- i.e. while tearing down the address
space it is trying to access.
It turns out that we stop perf after we tear down the userspace mm; a
receipie for disaster, since perf likes to access userspace for
various reasons.
Flip this order by moving up where we stop perf in do_exit().
Additionally, harden PERF_SAMPLE_CALLCHAIN and PERF_SAMPLE_STACK_USER
to abort when the current task does not have an mm (exit_mm() makes
sure to set current->mm = NULL; before commencing with the actual
teardown). Such that CPU wide events don't trip on this same problem.
Fixes: c5ebcedb56 ("perf: Add ability to attach user stack dump to sample")
Reported-by: Baisheng Gao <baisheng.gao@unisoc.com>
Suggested-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20250605110815.GQ39944@noisy.programming.kicks-ass.net
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 261dce3d64 ]
Now when isolcpus is enabled via the cmdline, wq_isolated_cpumask does
not include these isolated CPUs, even wq_unbound_cpumask has already
excluded them. It is only when we successfully configure an isolate cpuset
partition that wq_isolated_cpumask gets overwritten by
workqueue_unbound_exclude_cpumask(), including both the cmdline-specified
isolated CPUs and the isolated CPUs within the cpuset partitions.
Fix this issue by initializing wq_isolated_cpumask properly in
workqueue_init_early().
Fixes: fe28f631fa ("workqueue: Add workqueue_unbound_exclude_cpumask() to exclude CPUs from wq_unbound_cpumask")
Signed-off-by: Chuyi Zhou <zhouchuyi@bytedance.com>
Reviewed-by: Waiman Long <longman@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 33796b9187 upstream.
During task_group creation, sched_create_group() calls
scx_group_set_weight() with CGROUP_WEIGHT_DFL to initialize the sched_ext
portion. This is premature and ends up calling ops.cgroup_set_weight() with
an incorrect @cgrp before ops.cgroup_init() is called.
sched_create_group() should just initialize SCX related fields in the new
task_group. Fix it by factoring out scx_tg_init() from sched_init() and
making sched_create_group() call that function instead of
scx_group_set_weight().
v2: Retain CONFIG_EXT_GROUP_SCHED ifdef in sched_init() as removing it leads
to build failures on !CONFIG_GROUP_SCHED configs.
Signed-off-by: Tejun Heo <tj@kernel.org>
Fixes: 8195136669 ("sched_ext: Add cgroup support")
Cc: stable@vger.kernel.org # v6.12+
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a9d0aab5eb upstream.
When faultable trace events were added, a trace event may no longer use
normal RCU to synchronize but instead used synchronize_rcu_tasks_trace().
This synchronization takes a much longer time to synchronize.
The filter logic would free the filters by calling
tracepoint_synchronize_unregister() after it unhooked the filter strings
and before freeing them. With this function now calling
synchronize_rcu_tasks_trace() this increased the time to free a filter
tremendously. On a PREEMPT_RT system, it was even more noticeable.
# time trace-cmd record -p function sleep 1
[..]
real 2m29.052s
user 0m0.244s
sys 0m20.136s
As trace-cmd would clear out all the filters before recording, it could
take up to 2 minutes to do a recording of "sleep 1".
To find out where the issues was:
~# trace-cmd sqlhist -e -n sched_stack select start.prev_state as state, end.next_comm as comm, TIMESTAMP_DELTA_USECS as delta, start.STACKTRACE as stack from sched_switch as start join sched_switch as end on start.prev_pid = end.next_pid
Which will produce the following commands (and -e will also execute them):
echo 's:sched_stack s64 state; char comm[16]; u64 delta; unsigned long stack[];' >> /sys/kernel/tracing/dynamic_events
echo 'hist:keys=prev_pid:__arg_18057_2=prev_state,__arg_18057_4=common_timestamp.usecs,__arg_18057_7=common_stacktrace' >> /sys/kernel/tracing/events/sched/sched_switch/trigger
echo 'hist:keys=next_pid:__state_18057_1=$__arg_18057_2,__comm_18057_3=next_comm,__delta_18057_5=common_timestamp.usecs-$__arg_18057_4,__stack_18057_6=$__arg_18057_7:onmatch(sched.sched_switch).trace(sched_stack,$__state_18057_1,$__comm_18057_3,$__delta_18057_5,$__stack_18057_6)' >> /sys/kernel/tracing/events/sched/sched_switch/trigger
The above creates a synthetic event that creates a stack trace when a task
schedules out and records it with the time it scheduled back in. Basically
the time a task is off the CPU. It also records the state of the task when
it left the CPU (running, blocked, sleeping, etc). It also saves the comm
of the task as "comm" (needed for the next command).
~# echo 'hist:keys=state,stack.stacktrace:vals=delta:sort=state,delta if comm == "trace-cmd" && state & 3' > /sys/kernel/tracing/events/synthetic/sched_stack/trigger
The above creates a histogram with buckets per state, per stack, and the
value of the total time it was off the CPU for that stack trace. It filters
on tasks with "comm == trace-cmd" and only the sleeping and blocked states
(1 - sleeping, 2 - blocked).
~# trace-cmd record -p function sleep 1
~# cat /sys/kernel/tracing/events/synthetic/sched_stack/hist | tail -18
{ state: 2, stack.stacktrace __schedule+0x1545/0x3700
schedule+0xe2/0x390
schedule_timeout+0x175/0x200
wait_for_completion_state+0x294/0x440
__wait_rcu_gp+0x247/0x4f0
synchronize_rcu_tasks_generic+0x151/0x230
apply_subsystem_event_filter+0xa2b/0x1300
subsystem_filter_write+0x67/0xc0
vfs_write+0x1e2/0xeb0
ksys_write+0xff/0x1d0
do_syscall_64+0x7b/0x420
entry_SYSCALL_64_after_hwframe+0x76/0x7e
} hitcount: 237 delta: 99756288 <<--------------- Delta is 99 seconds!
Totals:
Hits: 525
Entries: 21
Dropped: 0
This shows that this particular trace waited for 99 seconds on
synchronize_rcu_tasks() in apply_subsystem_event_filter().
In fact, there's a lot of places in the filter code that spends a lot of
time waiting for synchronize_rcu_tasks_trace() in order to free the
filters.
Add helper functions that will use call_rcu*() variants to asynchronously
free the filters. This brings the timings back to normal:
# time trace-cmd record -p function sleep 1
[..]
real 0m14.681s
user 0m0.335s
sys 0m28.616s
And the histogram also shows this:
~# cat /sys/kernel/tracing/events/synthetic/sched_stack/hist | tail -21
{ state: 2, stack.stacktrace __schedule+0x1545/0x3700
schedule+0xe2/0x390
schedule_timeout+0x175/0x200
wait_for_completion_state+0x294/0x440
__wait_rcu_gp+0x247/0x4f0
synchronize_rcu_normal+0x3db/0x5c0
tracing_reset_online_cpus+0x8f/0x1e0
tracing_open+0x335/0x440
do_dentry_open+0x4c6/0x17a0
vfs_open+0x82/0x360
path_openat+0x1a36/0x2990
do_filp_open+0x1c5/0x420
do_sys_openat2+0xed/0x180
__x64_sys_openat+0x108/0x1d0
do_syscall_64+0x7b/0x420
} hitcount: 2 delta: 77044
Totals:
Hits: 55
Entries: 28
Dropped: 0
Where the total waiting time of synchronize_rcu_tasks_trace() is 77
milliseconds.
Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: "Paul E. McKenney" <paulmck@kernel.org>
Cc: Jan Kiszka <jan.kiszka@siemens.com>
Cc: Andreas Ziegler <ziegler.andreas@siemens.com>
Cc: Felix MOESSBAUER <felix.moessbauer@siemens.com>
Link: https://lore.kernel.org/20250606201936.1e3d09a9@batman.local.home
Reported-by: "Flot, Julien" <julien.flot@siemens.com>
Tested-by: Julien Flot <julien.flot@siemens.com>
Fixes: a363d27cdb ("tracing: Allow system call tracepoints to handle page faults")
Closes: https://lore.kernel.org/all/240017f656631c7dd4017aa93d91f41f653788ea.camel@siemens.com/
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 00d872dd54 ]
The trace_adjust_address() will take a given address and examine the
persistent ring buffer to see if the address matches a module that is
listed there. If it does not, it will just adjust the value to the core
kernel delta. But if the address was for something that was not part of
the core kernel text or data it should not be adjusted.
Check the result of the adjustment and only return the adjustment if it
lands in the current kernel text or data. If not, return the original
address.
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://lore.kernel.org/20250506102300.0ba2f9e0@gandalf.local.home
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 94bde253d3 ]
There is currently some confusion in the s390x JIT regarding whether
orig_call can be NULL and what that means. Originally the NULL value
was used to distinguish the struct_ops case, but this was superseded by
BPF_TRAMP_F_INDIRECT (see commit 0c970ed2f8 ("s390/bpf: Fix indirect
trampoline generation").
The remaining reason to have this check is that NULL can actually be
passed to the arch_bpf_trampoline_size() call - but not to the
respective arch_prepare_bpf_trampoline()! call - by
bpf_struct_ops_prepare_trampoline().
Remove this asymmetry by passing stub_func to both functions, so that
JITs may rely on orig_call never being NULL.
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://lore.kernel.org/r/20250512221911.61314-2-iii@linux.ibm.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 08d7becc1a ]
Right now, if the clocksource watchdog detects a clocksource skew, it might
perform a per CPU check, for example in the TSC case on x86. In other
words: supposing TSC is detected as unstable by the clocksource watchdog
running at CPU1, as part of marking TSC unstable the kernel will also run a
check of TSC readings on some CPUs to be sure it is synced between them
all.
But that check happens only on some CPUs, not all of them; this choice is
based on the parameter "verify_n_cpus" and in some random cpumask
calculation. So, the watchdog runs such per CPU checks on up to
"verify_n_cpus" random CPUs among all online CPUs, with the risk of
repeating CPUs (that aren't double checked) in the cpumask random
calculation.
But if "verify_n_cpus" > num_online_cpus(), it should skip the random
calculation and just go ahead and check the clocksource sync between
all online CPUs, without the risk of skipping some CPUs due to
duplicity in the random cpumask calculation.
Tests in a 4 CPU laptop with TSC skew detected led to some cases of the per
CPU verification skipping some CPU even with verify_n_cpus=8, due to the
duplicity on random cpumask generation. Skipping the randomization when the
number of online CPUs is smaller than verify_n_cpus, solves that.
Suggested-by: Thadeu Lima de Souza Cascardo <cascardo@igalia.com>
Signed-off-by: Guilherme G. Piccoli <gpiccoli@igalia.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Paul E. McKenney <paulmck@kernel.org>
Link: https://lore.kernel.org/all/20250323173857.372390-1-gpiccoli@igalia.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit f914b52c37 upstream.
The following issue happens with a buggy module:
BUG: unable to handle page fault for address: ffffffffc05d0218
PGD 1bd66f067 P4D 1bd66f067 PUD 1bd671067 PMD 101808067 PTE 0
Oops: Oops: 0000 [#1] SMP KASAN PTI
Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
RIP: 0010:sized_strscpy+0x81/0x2f0
RSP: 0018:ffff88812d76fa08 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffffffffc0601010 RCX: dffffc0000000000
RDX: 0000000000000038 RSI: dffffc0000000000 RDI: ffff88812608da2d
RBP: 8080808080808080 R08: ffff88812608da2d R09: ffff88812608da68
R10: ffff88812608d82d R11: ffff88812608d810 R12: 0000000000000038
R13: ffff88812608da2d R14: ffffffffc05d0218 R15: fefefefefefefeff
FS: 00007fef552de740(0000) GS:ffff8884251c7000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffffffffc05d0218 CR3: 00000001146f0000 CR4: 00000000000006f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
ftrace_mod_get_kallsym+0x1ac/0x590
update_iter_mod+0x239/0x5b0
s_next+0x5b/0xa0
seq_read_iter+0x8c9/0x1070
seq_read+0x249/0x3b0
proc_reg_read+0x1b0/0x280
vfs_read+0x17f/0x920
ksys_read+0xf3/0x1c0
do_syscall_64+0x5f/0x2e0
entry_SYSCALL_64_after_hwframe+0x76/0x7e
The above issue may happen as follows:
(1) Add kprobe tracepoint;
(2) insmod test.ko;
(3) Module triggers ftrace disabled;
(4) rmmod test.ko;
(5) cat /proc/kallsyms; --> Will trigger UAF as test.ko already removed;
ftrace_mod_get_kallsym()
...
strscpy(module_name, mod_map->mod->name, MODULE_NAME_LEN);
...
The problem is when a module triggers an issue with ftrace and
sets ftrace_disable. The ftrace_disable is set when an anomaly is
discovered and to prevent any more damage, ftrace stops all text
modification. The issue that happened was that the ftrace_disable stops
more than just the text modification.
When a module is loaded, its init functions can also be traced. Because
kallsyms deletes the init functions after a module has loaded, ftrace
saves them when the module is loaded and function tracing is enabled. This
allows the output of the function trace to show the init function names
instead of just their raw memory addresses.
When a module is removed, ftrace_release_mod() is called, and if
ftrace_disable is set, it just returns without doing anything more. The
problem here is that it leaves the mod_list still around and if kallsyms
is called, it will call into this code and access the module memory that
has already been freed as it will return:
strscpy(module_name, mod_map->mod->name, MODULE_NAME_LEN);
Where the "mod" no longer exists and triggers a UAF bug.
Link: https://lore.kernel.org/all/20250523135452.626d8dcd@gandalf.local.home/
Cc: stable@vger.kernel.org
Fixes: aba4b5c22c ("ftrace: Save module init functions kallsyms symbols for tracing")
Link: https://lore.kernel.org/20250529111955.2349189-2-yebin@huaweicloud.com
Signed-off-by: Ye Bin <yebin10@huawei.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 7123dbbef8 upstream.
When updating `watchdog_thresh`, there is a race condition between writing
the new `watchdog_thresh` value and stopping the old watchdog timer. If
the old timer triggers during this window, it may falsely detect a
softlockup due to the old interval and the new `watchdog_thresh` value
being used. The problem can be described as follow:
# We asuume previous watchdog_thresh is 60, so the watchdog timer is
# coming every 24s.
echo 10 > /proc/sys/kernel/watchdog_thresh (User space)
|
+------>+ update watchdog_thresh (We are in kernel now)
|
| # using old interval and new `watchdog_thresh`
+------>+ watchdog hrtimer (irq context: detect softlockup)
|
|
+-------+
|
|
+ softlockup_stop_all
To fix this problem, introduce a shadow variable for `watchdog_thresh`.
The update to the actual `watchdog_thresh` is delayed until after the old
timer is stopped, preventing false positives.
The following testcase may help to understand this problem.
---------------------------------------------
echo RT_RUNTIME_SHARE > /sys/kernel/debug/sched/features
echo -1 > /proc/sys/kernel/sched_rt_runtime_us
echo 0 > /sys/kernel/debug/sched/fair_server/cpu3/runtime
echo 60 > /proc/sys/kernel/watchdog_thresh
taskset -c 3 chrt -r 99 /bin/bash -c "while true;do true; done" &
echo 10 > /proc/sys/kernel/watchdog_thresh &
---------------------------------------------
The test case above first removes the throttling restrictions for
real-time tasks. It then sets watchdog_thresh to 60 and executes a
real-time task ,a simple while(1) loop, on cpu3. Consequently, the final
command gets blocked because the presence of this real-time thread
prevents kworker:3 from being selected by the scheduler. This eventually
triggers a softlockup detection on cpu3 due to watchdog_timer_fn operating
with inconsistent variable - using both the old interval and the updated
watchdog_thresh simultaneously.
[nysal@linux.ibm.com: fix the SOFTLOCKUP_DETECTOR=n case]
Link: https://lkml.kernel.org/r/20250502111120.282690-1-nysal@linux.ibm.com
Link: https://lkml.kernel.org/r/20250421035021.3507649-1-luogengkun@huaweicloud.com
Signed-off-by: Luo Gengkun <luogengkun@huaweicloud.com>
Signed-off-by: Nysal Jan K.A. <nysal@linux.ibm.com>
Cc: Doug Anderson <dianders@chromium.org>
Cc: Joel Granados <joel.granados@kernel.org>
Cc: Song Liu <song@kernel.org>
Cc: Thomas Gleinxer <tglx@linutronix.de>
Cc: "Nysal Jan K.A." <nysal@linux.ibm.com>
Cc: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 37fb58a727 upstream.
An issue was found:
# cd /sys/fs/cgroup/freezer/
# mkdir test
# echo FROZEN > test/freezer.state
# cat test/freezer.state
FROZEN
# sleep 1000 &
[1] 863
# echo 863 > test/cgroup.procs
# cat test/freezer.state
FREEZING
When tasks are migrated to a frozen cgroup, the freezer fails to
immediately freeze the tasks, causing the cgroup to remain in the
"FREEZING".
The freeze_task() function is called before clearing the CGROUP_FROZEN
flag. This causes the freezing() check to incorrectly return false,
preventing __freeze_task() from being invoked for the migrated task.
To fix this issue, clear the CGROUP_FROZEN state before calling
freeze_task().
Fixes: f5d39b0208 ("freezer,sched: Rewrite core freezer logic")
Cc: stable@vger.kernel.org # v6.1+
Reported-by: Zhong Jiawei <zhongjiawei1@huawei.com>
Signed-off-by: Chen Ridong <chenridong@huawei.com>
Acked-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c70fc32f44 upstream.
Mike reports that commit 6d71a9c616 ("sched/fair: Fix EEVDF entity
placement bug causing scheduling lag") relies on commit 4423af84b2
("sched/fair: optimize the PLACE_LAG when se->vlag is zero") to not
trip a WARN in place_entity().
What happens is that the lag of the very last entity is 0 per
definition -- the average of one element matches the value of that
element. Therefore place_entity() will match the condition skipping
the lag adjustment:
if (sched_feat(PLACE_LAG) && cfs_rq->nr_queued && se->vlag) {
Without the 'se->vlag' condition -- it will attempt to adjust the zero
lag even though we're inserting into an empty tree.
Notably, we should have failed the 'cfs_rq->nr_queued' condition, but
don't because they didn't get updated.
Additionally, move update_load_add() after placement() as is
consistent with other place_entity() users -- this change is
non-functional, place_entity() does not use cfs_rq->load.
Fixes: 6d71a9c616 ("sched/fair: Fix EEVDF entity placement bug causing scheduling lag")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reported-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: "Peter Zijlstra (Intel)" <peterz@infradead.org>
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/c216eb4ef0e0e0029c600aefc69d56681cee5581.camel@gmx.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 690e47d140 upstream.
Overview
========
When a CPU chooses to call push_rt_task and picks a task to push to
another CPU's runqueue then it will call find_lock_lowest_rq method
which would take a double lock on both CPUs' runqueues. If one of the
locks aren't readily available, it may lead to dropping the current
runqueue lock and reacquiring both the locks at once. During this window
it is possible that the task is already migrated and is running on some
other CPU. These cases are already handled. However, if the task is
migrated and has already been executed and another CPU is now trying to
wake it up (ttwu) such that it is queued again on the runqeue
(on_rq is 1) and also if the task was run by the same CPU, then the
current checks will pass even though the task was migrated out and is no
longer in the pushable tasks list.
Crashes
=======
This bug resulted in quite a few flavors of crashes triggering kernel
panics with various crash signatures such as assert failures, page
faults, null pointer dereferences, and queue corruption errors all
coming from scheduler itself.
Some of the crashes:
-> kernel BUG at kernel/sched/rt.c:1616! BUG_ON(idx >= MAX_RT_PRIO)
Call Trace:
? __die_body+0x1a/0x60
? die+0x2a/0x50
? do_trap+0x85/0x100
? pick_next_task_rt+0x6e/0x1d0
? do_error_trap+0x64/0xa0
? pick_next_task_rt+0x6e/0x1d0
? exc_invalid_op+0x4c/0x60
? pick_next_task_rt+0x6e/0x1d0
? asm_exc_invalid_op+0x12/0x20
? pick_next_task_rt+0x6e/0x1d0
__schedule+0x5cb/0x790
? update_ts_time_stats+0x55/0x70
schedule_idle+0x1e/0x40
do_idle+0x15e/0x200
cpu_startup_entry+0x19/0x20
start_secondary+0x117/0x160
secondary_startup_64_no_verify+0xb0/0xbb
-> BUG: kernel NULL pointer dereference, address: 00000000000000c0
Call Trace:
? __die_body+0x1a/0x60
? no_context+0x183/0x350
? __warn+0x8a/0xe0
? exc_page_fault+0x3d6/0x520
? asm_exc_page_fault+0x1e/0x30
? pick_next_task_rt+0xb5/0x1d0
? pick_next_task_rt+0x8c/0x1d0
__schedule+0x583/0x7e0
? update_ts_time_stats+0x55/0x70
schedule_idle+0x1e/0x40
do_idle+0x15e/0x200
cpu_startup_entry+0x19/0x20
start_secondary+0x117/0x160
secondary_startup_64_no_verify+0xb0/0xbb
-> BUG: unable to handle page fault for address: ffff9464daea5900
kernel BUG at kernel/sched/rt.c:1861! BUG_ON(rq->cpu != task_cpu(p))
-> kernel BUG at kernel/sched/rt.c:1055! BUG_ON(!rq->nr_running)
Call Trace:
? __die_body+0x1a/0x60
? die+0x2a/0x50
? do_trap+0x85/0x100
? dequeue_top_rt_rq+0xa2/0xb0
? do_error_trap+0x64/0xa0
? dequeue_top_rt_rq+0xa2/0xb0
? exc_invalid_op+0x4c/0x60
? dequeue_top_rt_rq+0xa2/0xb0
? asm_exc_invalid_op+0x12/0x20
? dequeue_top_rt_rq+0xa2/0xb0
dequeue_rt_entity+0x1f/0x70
dequeue_task_rt+0x2d/0x70
__schedule+0x1a8/0x7e0
? blk_finish_plug+0x25/0x40
schedule+0x3c/0xb0
futex_wait_queue_me+0xb6/0x120
futex_wait+0xd9/0x240
do_futex+0x344/0xa90
? get_mm_exe_file+0x30/0x60
? audit_exe_compare+0x58/0x70
? audit_filter_rules.constprop.26+0x65e/0x1220
__x64_sys_futex+0x148/0x1f0
do_syscall_64+0x30/0x80
entry_SYSCALL_64_after_hwframe+0x62/0xc7
-> BUG: unable to handle page fault for address: ffff8cf3608bc2c0
Call Trace:
? __die_body+0x1a/0x60
? no_context+0x183/0x350
? spurious_kernel_fault+0x171/0x1c0
? exc_page_fault+0x3b6/0x520
? plist_check_list+0x15/0x40
? plist_check_list+0x2e/0x40
? asm_exc_page_fault+0x1e/0x30
? _cond_resched+0x15/0x30
? futex_wait_queue_me+0xc8/0x120
? futex_wait+0xd9/0x240
? try_to_wake_up+0x1b8/0x490
? futex_wake+0x78/0x160
? do_futex+0xcd/0xa90
? plist_check_list+0x15/0x40
? plist_check_list+0x2e/0x40
? plist_del+0x6a/0xd0
? plist_check_list+0x15/0x40
? plist_check_list+0x2e/0x40
? dequeue_pushable_task+0x20/0x70
? __schedule+0x382/0x7e0
? asm_sysvec_reschedule_ipi+0xa/0x20
? schedule+0x3c/0xb0
? exit_to_user_mode_prepare+0x9e/0x150
? irqentry_exit_to_user_mode+0x5/0x30
? asm_sysvec_reschedule_ipi+0x12/0x20
Above are some of the common examples of the crashes that were observed
due to this issue.
Details
=======
Let's look at the following scenario to understand this race.
1) CPU A enters push_rt_task
a) CPU A has chosen next_task = task p.
b) CPU A calls find_lock_lowest_rq(Task p, CPU Z’s rq).
c) CPU A identifies CPU X as a destination CPU (X < Z).
d) CPU A enters double_lock_balance(CPU Z’s rq, CPU X’s rq).
e) Since X is lower than Z, CPU A unlocks CPU Z’s rq. Someone else has
locked CPU X’s rq, and thus, CPU A must wait.
2) At CPU Z
a) Previous task has completed execution and thus, CPU Z enters
schedule, locks its own rq after CPU A releases it.
b) CPU Z dequeues previous task and begins executing task p.
c) CPU Z unlocks its rq.
d) Task p yields the CPU (ex. by doing IO or waiting to acquire a
lock) which triggers the schedule function on CPU Z.
e) CPU Z enters schedule again, locks its own rq, and dequeues task p.
f) As part of dequeue, it sets p.on_rq = 0 and unlocks its rq.
3) At CPU B
a) CPU B enters try_to_wake_up with input task p.
b) Since CPU Z dequeued task p, p.on_rq = 0, and CPU B updates
B.state = WAKING.
c) CPU B via select_task_rq determines CPU Y as the target CPU.
4) The race
a) CPU A acquires CPU X’s lock and relocks CPU Z.
b) CPU A reads task p.cpu = Z and incorrectly concludes task p is
still on CPU Z.
c) CPU A failed to notice task p had been dequeued from CPU Z while
CPU A was waiting for locks in double_lock_balance. If CPU A knew
that task p had been dequeued, it would return NULL forcing
push_rt_task to give up the task p's migration.
d) CPU B updates task p.cpu = Y and calls ttwu_queue.
e) CPU B locks Ys rq. CPU B enqueues task p onto Y and sets task
p.on_rq = 1.
f) CPU B unlocks CPU Y, triggering memory synchronization.
g) CPU A reads task p.on_rq = 1, cementing its assumption that task p
has not migrated.
h) CPU A decides to migrate p to CPU X.
This leads to A dequeuing p from Y's queue and various crashes down the
line.
Solution
========
The solution here is fairly simple. After obtaining the lock (at 4a),
the check is enhanced to make sure that the task is still at the head of
the pushable tasks list. If not, then it is anyway not suitable for
being pushed out.
Testing
=======
The fix is tested on a cluster of 3 nodes, where the panics due to this
are hit every couple of days. A fix similar to this was deployed on such
cluster and was stable for more than 30 days.
Co-developed-by: Jon Kohler <jon@nutanix.com>
Signed-off-by: Jon Kohler <jon@nutanix.com>
Co-developed-by: Gauri Patwardhan <gauri.patwardhan@nutanix.com>
Signed-off-by: Gauri Patwardhan <gauri.patwardhan@nutanix.com>
Co-developed-by: Rahul Chunduru <rahul.chunduru@nutanix.com>
Signed-off-by: Rahul Chunduru <rahul.chunduru@nutanix.com>
Signed-off-by: Harshit Agarwal <harshit@nutanix.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: "Steven Rostedt (Google)" <rostedt@goodmis.org>
Reviewed-by: Phil Auld <pauld@redhat.com>
Tested-by: Will Ton <william.ton@nutanix.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20250225180553.167995-1-harshit@nutanix.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4fc78a7c9c upstream.
When reading a memory mapped buffer the reader page is just swapped out
with the last page written in the write buffer. If the reader page is the
same as the commit buffer (the buffer that is currently being written to)
it was assumed that it should never have missed events. If it does, it
triggers a WARN_ON_ONCE().
But there just happens to be one scenario where this can legitimately
happen. That is on a commit_overrun. A commit overrun is when an interrupt
preempts an event being written to the buffer and then the interrupt adds
so many new events that it fills and wraps the buffer back to the commit.
Any new events would then be dropped and be reported as "missed_events".
In this case, the next page to read is the commit buffer and after the
swap of the reader page, the reader page will be the commit buffer, but
this time there will be missed events and this triggers the following
warning:
------------[ cut here ]------------
WARNING: CPU: 2 PID: 1127 at kernel/trace/ring_buffer.c:7357 ring_buffer_map_get_reader+0x49a/0x780
Modules linked in: kvm_intel kvm irqbypass
CPU: 2 UID: 0 PID: 1127 Comm: trace-cmd Not tainted 6.15.0-rc7-test-00004-g478bc2824b45-dirty #564 PREEMPT
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
RIP: 0010:ring_buffer_map_get_reader+0x49a/0x780
Code: 00 00 00 48 89 fe 48 c1 ee 03 80 3c 2e 00 0f 85 ec 01 00 00 4d 3b a6 a8 00 00 00 0f 85 8a fd ff ff 48 85 c0 0f 84 55 fe ff ff <0f> 0b e9 4e fe ff ff be 08 00 00 00 4c 89 54 24 58 48 89 54 24 50
RSP: 0018:ffff888121787dc0 EFLAGS: 00010002
RAX: 00000000000006a2 RBX: ffff888100062800 RCX: ffffffff8190cb49
RDX: ffff888126934c00 RSI: 1ffff11020200a15 RDI: ffff8881010050a8
RBP: dffffc0000000000 R08: 0000000000000000 R09: ffffed1024d26982
R10: ffff888126934c17 R11: ffff8881010050a8 R12: ffff888126934c00
R13: ffff8881010050b8 R14: ffff888101005000 R15: ffff888126930008
FS: 00007f95c8cd7540(0000) GS:ffff8882b576e000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f95c8de4dc0 CR3: 0000000128452002 CR4: 0000000000172ef0
Call Trace:
<TASK>
? __pfx_ring_buffer_map_get_reader+0x10/0x10
tracing_buffers_ioctl+0x283/0x370
__x64_sys_ioctl+0x134/0x190
do_syscall_64+0x79/0x1c0
entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x7f95c8de48db
Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00
RSP: 002b:00007ffe037ba110 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00007ffe037bb2b0 RCX: 00007f95c8de48db
RDX: 0000000000000000 RSI: 0000000000005220 RDI: 0000000000000006
RBP: 00007ffe037ba180 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 00007ffe037bb6f8 R14: 00007f95c9065000 R15: 00005575c7492c90
</TASK>
irq event stamp: 5080
hardirqs last enabled at (5079): [<ffffffff83e0adb0>] _raw_spin_unlock_irqrestore+0x50/0x70
hardirqs last disabled at (5080): [<ffffffff83e0aa83>] _raw_spin_lock_irqsave+0x63/0x70
softirqs last enabled at (4182): [<ffffffff81516122>] handle_softirqs+0x552/0x710
softirqs last disabled at (4159): [<ffffffff815163f7>] __irq_exit_rcu+0x107/0x210
---[ end trace 0000000000000000 ]---
The above was triggered by running on a kernel with both lockdep and KASAN
as well as kmemleak enabled and executing the following command:
# perf record -o perf-test.dat -a -- trace-cmd record --nosplice -e all -p function hackbench 50
With perf interjecting a lot of interrupts and trace-cmd enabling all
events as well as function tracing, with lockdep, KASAN and kmemleak
enabled, it could cause an interrupt preempting an event being written to
add enough events to wrap the buffer. trace-cmd was modified to have
--nosplice use mmap instead of reading the buffer.
The way to differentiate this case from the normal case of there only
being one page written to where the swap of the reader page received that
one page (which is the commit page), check if the tail page is on the
reader page. The difference between the commit page and the tail page is
that the tail page is where new writes go to, and the commit page holds
the first write that hasn't been committed yet. In the case of an
interrupt preempting the write of an event and filling the buffer, it
would move the tail page but not the commit page.
Have the warning only trigger if the tail page is also on the reader page,
and also print out the number of events dropped by a commit overrun as
that can not yet be safely added to the page so that the reader can see
there were events dropped.
Cc: stable@vger.kernel.org
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Vincent Donnefort <vdonnefort@google.com>
Link: https://lore.kernel.org/20250528121555.2066527e@gandalf.local.home
Fixes: fe832be05a ("ring-buffer: Have mmapped ring buffer keep track of missed events")
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f90fff1e15 upstream.
If an exiting non-autoreaping task has already passed exit_notify() and
calls handle_posix_cpu_timers() from IRQ, it can be reaped by its parent
or debugger right after unlock_task_sighand().
If a concurrent posix_cpu_timer_del() runs at that moment, it won't be
able to detect timer->it.cpu.firing != 0: cpu_timer_task_rcu() and/or
lock_task_sighand() will fail.
Add the tsk->exit_state check into run_posix_cpu_timers() to fix this.
This fix is not needed if CONFIG_POSIX_CPU_TIMERS_TASK_WORK=y, because
exit_task_work() is called before exit_notify(). But the check still
makes sense, task_work_add(&tsk->posix_cputimers_work.work) will fail
anyway in this case.
Cc: stable@vger.kernel.org
Reported-by: Benoît Sevens <bsevens@google.com>
Fixes: 0bdd2ed413 ("sched: run_posix_cpu_timers: Don't check ->exit_state, use lock_task_sighand()")
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 9960be72a5 ]
In the idle CPU selection logic, attempting cross-node searches adds
unnecessary complexity when CONFIG_NUMA is disabled.
Since there's no meaningful concept of nodes in this case, simplify the
logic by restricting the idle CPU search to the current node only.
Fixes: 48849271e6 ("sched_ext: idle: Per-node idle cpumasks")
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 86bc9c7424 ]
syzkaller reported an issue:
WARNING: CPU: 3 PID: 217 at kernel/bpf/core.c:2357 __bpf_prog_ret0_warn+0xa/0x20 kernel/bpf/core.c:2357
Modules linked in:
CPU: 3 UID: 0 PID: 217 Comm: kworker/u32:6 Not tainted 6.15.0-rc4-syzkaller-00040-g8bac8898fe39
RIP: 0010:__bpf_prog_ret0_warn+0xa/0x20 kernel/bpf/core.c:2357
Call Trace:
<TASK>
bpf_dispatcher_nop_func include/linux/bpf.h:1316 [inline]
__bpf_prog_run include/linux/filter.h:718 [inline]
bpf_prog_run include/linux/filter.h:725 [inline]
cls_bpf_classify+0x74a/0x1110 net/sched/cls_bpf.c:105
...
When creating bpf program, 'fp->jit_requested' depends on bpf_jit_enable.
This issue is triggered because of CONFIG_BPF_JIT_ALWAYS_ON is not set
and bpf_jit_enable is set to 1, causing the arch to attempt JIT the prog,
but jit failed due to FAULT_INJECTION. As a result, incorrectly
treats the program as valid, when the program runs it calls
`__bpf_prog_ret0_warn` and triggers the WARN_ON_ONCE(1).
Reported-by: syzbot+0903f6d7f285e41cdf10@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/bpf/6816e34e.a70a0220.254cdc.002c.GAE@google.com
Fixes: fa9dd599b4 ("bpf: get rid of pure_initcall dependency to enable jits")
Signed-off-by: KaFai Wan <mannkafai@gmail.com>
Link: https://lore.kernel.org/r/20250526133358.2594176-1-mannkafai@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4e2e6841ff ]
This reverts commit 4a8f635a60.
Althought get_pid_task() internally already calls rcu_read_lock() and
rcu_read_unlock(), the find_vpid() was not.
The documentation for find_vpid() clearly states:
"Must be called with the tasklist_lock or rcu_read_lock() held."
Add proper rcu_read_lock/unlock() to protect the find_vpid().
Fixes: 4a8f635a60 ("bpf: remove unnecessary rcu_read_{lock,unlock}() in multi-uprobe attach logic")
Reported-by: Xuewen Yan <xuewen.yan@unisoc.com>
Signed-off-by: Di Shen <di.shen@unisoc.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20250520054943.5002-1-xuewen.yan@unisoc.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f2947c4b7d ]
The function event_trigger_alloc() creates an event_trigger_data
descriptor and states that it needs to be freed via event_trigger_free().
This is incorrect, it needs to be freed by trigger_data_free() as
event_trigger_free() adds ref counting.
Rename event_trigger_alloc() to trigger_data_alloc() and state that it
needs to be freed via trigger_data_free(). This naming convention
was introducing bugs.
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Tom Zanussi <zanussi@kernel.org>
Link: https://lore.kernel.org/20250507145455.776436410@goodmis.org
Fixes: 86599dbe2c ("tracing: Add helper functions to simplify event_command.parse() callback handling")
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7ab0fc61ce ]
The histogram trigger has three somewhat large arrays on the kernel stack:
unsigned long entries[HIST_STACKTRACE_DEPTH];
u64 var_ref_vals[TRACING_MAP_VARS_MAX];
char compound_key[HIST_KEY_SIZE_MAX];
Checking the function event_hist_trigger() stack frame size, it currently
uses 816 bytes for its stack frame due to these variables!
Instead, allocate a per CPU structure that holds these arrays for each
context level (normal, softirq, irq and NMI). That is, each CPU will have
4 of these structures. This will be allocated when the first histogram
trigger is enabled and freed when the last is disabled. When the
histogram callback triggers, it will request this structure. The request
will disable preemption, get the per CPU structure at the index of the
per CPU variable, and increment that variable.
The callback will use the arrays in this structure to perform its work and
then release the structure. That in turn will simply decrement the per CPU
index and enable preemption.
Moving the variables from the kernel stack to the per CPU structure brings
the stack frame of event_hist_trigger() down to just 112 bytes.
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Tom Zanussi <zanussi@kernel.org>
Link: https://lore.kernel.org/20250407123851.74ea8d58@gandalf.local.home
Fixes: 067fe038e7 ("tracing: Add variable reference handling to hist triggers")
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 714070c4cb ]
In the current implementation if the program is dev-bound to a specific
device, it will not be possible to perform XDP_REDIRECT into a DEVMAP
or CPUMAP even if the program is running in the driver NAPI context and
it is not attached to any map entry. This seems in contrast with the
explanation available in bpf_prog_map_compatible routine.
Fix the issue introducing __bpf_prog_map_compatible utility routine in
order to avoid bpf_prog_is_dev_bound() check running bpf_check_tail_call()
at program load time (bpf_prog_select_runtime()).
Continue forbidding to attach a dev-bound program to XDP maps
(BPF_MAP_TYPE_PROG_ARRAY, BPF_MAP_TYPE_DEVMAP and BPF_MAP_TYPE_CPUMAP).
Fixes: 3d76a4d3d4 ("bpf: XDP metadata RX kfuncs")
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1b17d4525b ]
Commit cdb8c100d8 ("include/linux/suspend.h: Only show pm_pr_dbg
messages at suspend/resume") caused PM debug messages to only be
printed during system-wide suspend and resume in progress, but it
forgot about hibernation.
Address this by adding a check for hibernation in progress to
pm_debug_messages_should_print().
Fixes: cdb8c100d8 ("include/linux/suspend.h: Only show pm_pr_dbg messages at suspend/resume")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://patch.msgid.link/4998903.GXAFRqVoOG@rjwysocki.net
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 179c0c7044 ]
When the device is of a non-CPU type, table[i].performance won't be
initialized in the previous em_init_performance(), resulting in division
by zero when calculating costs in em_compute_costs().
Since the 'cost' algorithm is only used for EAS energy efficiency
calculations and is currently not utilized by other device drivers, we
should add the _is_cpu_device(dev) check to prevent this division-by-zero
issue.
Fixes: 1b600da510 ("PM: EM: Optimize em_cpu_energy() and remove division")
Signed-off-by: Yaxiong Tian <tianyaxiong@kylinos.cn>
Reviewed-by: Lukasz Luba <lukasz.luba@arm.com>
Link: https://patch.msgid.link/tencent_7F99ED4767C1AF7889D0D8AD50F34859CE06@qq.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit aa3ee4f0b7 ]
Delayed dequeued feature keeps a sleeping task enqueued until its
lag has elapsed. As a result, it stays also visible in rq->nr_running.
So when in wake_affine_idle(), we should use the real running-tasks
in rq to check whether we should place the wake-up task to
current cpu.
On the other hand, add a helper function to return the nr-delayed.
Fixes: 152e11f6df ("sched/fair: Implement delayed dequeue")
Signed-off-by: Xuewen Yan <xuewen.yan@unisoc.com>
Reviewed-and-tested-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20250303105241.17251-2-xuewen.yan@unisoc.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b7ca5743a2 ]
It was reported that in 6.12, smpboot_create_threads() was
taking much longer then in 6.6.
I narrowed down the call path to:
smpboot_create_threads()
-> kthread_create_on_cpu()
-> kthread_bind()
-> __kthread_bind_mask()
->wait_task_inactive()
Where in wait_task_inactive() we were regularly hitting the
queued case, which sets a 1 tick timeout, which when called
multiple times in a row, accumulates quickly into a long
delay.
I noticed disabling the DELAY_DEQUEUE sched feature recovered
the performance, and it seems the newly create tasks are usually
sched_delayed and left on the runqueue.
So in wait_task_inactive() when we see the task
p->se.sched_delayed, manually dequeue the sched_delayed task
with DEQUEUE_DELAYED, so we don't have to constantly wait a
tick.
Fixes: 152e11f6df ("sched/fair: Implement delayed dequeue")
Reported-by: peter-yc.chang@mediatek.com
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://lkml.kernel.org/r/20250429150736.3778580-1-jstultz@google.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f51972e6f8 ]
According to the throttling mechanism, the pmu interrupts number can not
exceed the max_samples_per_tick in one tick. But this mechanism is
ineffective when max_samples_per_tick=1, because the throttling check is
skipped during the first interrupt and only performed when the second
interrupt arrives.
Perhaps this bug may cause little influence in one tick, but if in a
larger time scale, the problem can not be underestimated.
When max_samples_per_tick = 1:
Allowed-interrupts-per-second max-samples-per-second default-HZ ARCH
200 100 100 X86
500 250 250 ARM64
...
Obviously, the pmu interrupt number far exceed the user's expect.
Fixes: e050e3f0a7 ("perf: Fix broken interrupt rate throttling")
Signed-off-by: Qing Wang <wangqing7171@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20250405141635.243786-3-wangqing7171@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8feb053d53 ]
Gabriele noted that in case of signal_pending_state(), the tracepoint
sees a stale task-state.
Fixes: fa2c3254d7 ("sched/tracing: Don't re-read p->state when emitting sched_switch event")
Reported-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 2fbdb6d8e0 upstream.
On arm32, size_t is defined to be unsigned int, while PAGE_SIZE is
unsigned long. This hence triggers a compilation warning as min()
asserts the type of two operands to be equal. Casting PAGE_SIZE to size_t
solves this issue and works on other target architectures as well.
Compilation warning details:
kernel/trace/trace.c: In function 'tracing_splice_read_pipe':
./include/linux/minmax.h:20:28: warning: comparison of distinct pointer types lacks a cast
(!!(sizeof((typeof(x) *)1 == (typeof(y) *)1)))
^
./include/linux/minmax.h:26:4: note: in expansion of macro '__typecheck'
(__typecheck(x, y) && __no_side_effects(x, y))
^~~~~~~~~~~
...
kernel/trace/trace.c:6771:8: note: in expansion of macro 'min'
min((size_t)trace_seq_used(&iter->seq),
^~~
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/20250526013731.1198030-1-pantaixi@huaweicloud.com
Fixes: f5178c41bb ("tracing: Fix oob write in trace_seq_to_buffer()")
Reviewed-by: Jeongjun Park <aha310510@gmail.com>
Signed-off-by: Pan Taixi <pantaixi@huaweicloud.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull hotfixes from Andrew Morton:
"22 hotfixes.
13 are cc:stable and the remainder address post-6.14 issues or aren't
considered necessary for -stable kernels. 19 are for MM"
* tag 'mm-hotfixes-stable-2025-05-25-00-58' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (22 commits)
mailmap: add Jarkko's employer email address
mm: fix copy_vma() error handling for hugetlb mappings
memcg: always call cond_resched() after fn()
mm/hugetlb: fix kernel NULL pointer dereference when replacing free hugetlb folios
mm: vmalloc: only zero-init on vrealloc shrink
mm: vmalloc: actually use the in-place vrealloc region
alloc_tag: allocate percpu counters for module tags dynamically
module: release codetag section when module load fails
mm/cma: make detection of highmem_start more robust
MAINTAINERS: add mm memory policy section
MAINTAINERS: add mm ksm section
kasan: avoid sleepable page allocation from atomic context
highmem: add folio_test_partial_kmap()
MAINTAINERS: add hung-task detector section
taskstats: fix struct taskstats breaks backward compatibility since version 15
mm/truncate: fix out-of-bounds when doing a right-aligned split
MAINTAINERS: add mm reclaim section
MAINTAINERS: update page allocator section
mm: fix VM_UFFD_MINOR == VM_SHADOW_STACK on USERFAULTFD=y && ARM64_GCS=y
mm: mmap: map MAP_STACK to VM_NOHUGEPAGE only if THP is enabled
...
Pull crypto fixes from Herbert Xu:
"This fixes a regression in padata as well as an ancient double-free
bug in af_alg"
* tag 'v6.15-p7' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: algif_hash - fix double free in hash_accept
padata: do not leak refcount in reorder_work
A recent patch that addressed a UAF introduced a reference count leak:
the parallel_data refcount is incremented unconditionally, regardless
of the return value of queue_work(). If the work item is already queued,
the incremented refcount is never decremented.
Fix this by checking the return value of queue_work() and decrementing
the refcount when necessary.
Resolves:
Unreferenced object 0xffff9d9f421e3d80 (size 192):
comm "cryptomgr_probe", pid 157, jiffies 4294694003
hex dump (first 32 bytes):
80 8b cf 41 9f 9d ff ff b8 97 e0 89 ff ff ff ff ...A............
d0 97 e0 89 ff ff ff ff 19 00 00 00 1f 88 23 00 ..............#.
backtrace (crc 838fb36):
__kmalloc_cache_noprof+0x284/0x320
padata_alloc_pd+0x20/0x1e0
padata_alloc_shell+0x3b/0xa0
0xffffffffc040a54d
cryptomgr_probe+0x43/0xc0
kthread+0xf6/0x1f0
ret_from_fork+0x2f/0x50
ret_from_fork_asm+0x1a/0x30
Fixes: dd7d37ccf6 ("padata: avoid UAF for reorder_work")
Cc: <stable@vger.kernel.org>
Signed-off-by: Dominik Grzegorzek <dominik.grzegorzek@oracle.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Pull hotfixes from Andrew Morton:
"Nine singleton hotfixes, all MM. Four are cc:stable"
* tag 'mm-hotfixes-stable-2025-05-17-09-41' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
mm: userfaultfd: correct dirty flags set for both present and swap pte
zsmalloc: don't underflow size calculation in zs_obj_write()
mm/page_alloc: fix race condition in unaccepted memory handling
mm/page_alloc: ensure try_alloc_pages() plays well with unaccepted memory
MAINTAINERS: add mm GUP section
mm/codetag: move tag retrieval back upfront in __free_pages()
mm/memory: fix mapcount / refcount sanity check for mTHP reuse
kernel/fork: only call untrack_pfn_clear() on VMAs duplicated for fork()
mm: hugetlb: fix incorrect fallback for subpool
Luo Gengkun
Peter Zijlstra
Steven Rostedt
Chuyi Zhou
Tejun Heo
Shung-Hsi Yu
Ilya Leoshkevich
Hou Tao
Jiayuan Chen
Guilherme G. Piccoli
Ye Bin
Chen Ridong
Harshit Agarwal
Suren Baghdasaryan
Dmitry Antipov
Oleg Nesterov
Andrea Righi
Yonghong Song
KaFai Wan
Di Shen
Tao Chen
Miaoqian Lin
Lorenzo Bianconi
Rafael J. Wysocki
Zijun Hu
Yaxiong Tian
Xuewen Yan
Yongliang Gao
John Stultz
Qing Wang
Pan Taixi
Linus Torvalds
David Wang
Dominik Grzegorzek