find_new_ilb() returns nr_cpu_ids on failure - which is the usual
cpumask bitops return pattern, but is weird & unnecessary in this
context: not only is it a global variable, it it is a +1 out of
bounds CPU index and also has different signedness ...
Its only user, kick_ilb(), then checks the return against nr_cpu_ids
to decide to return. There's no other use.
So instead of this, use a standard -1 return on failure to find an
idle CPU, as the argument is signed already.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Link: https://lore.kernel.org/r/20231006102518.2452758-4-mingo@kernel.org
The following commit:
9b3c4ab304 ("sched,rcu: Rework try_invoke_on_locked_down_task()")
... renamed try_invoke_on_locked_down_task() to task_call_func(),
but forgot to update the comment in try_to_wake_up().
But it turns out that the smp_rmb() doesn't live in task_call_func()
either, it was moved to __task_needs_rq_lock() in:
91dabf33ae ("sched: Fix race in task_call_func()")
Fix that now.
Also fix the s/smb/smp typo while at it.
Reported-by: Zhang Qiao <zhangqiao22@huawei.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230731085759.11443-1-zhangqiao22@huawei.com
When cpufreq's policy is 'single', there is a scenario that will
cause sg_policy's next_freq to be unable to update.
When the CPU's util is always max, the cpufreq will be max,
and then if we change the policy's scaling_max_freq to be a
lower freq, indeed, the sg_policy's next_freq need change to
be the lower freq, however, because the cpu_is_busy, the next_freq
would keep the max_freq.
For example:
The cpu7 is a single CPU:
unisoc:/sys/devices/system/cpu/cpufreq/policy7 # while true;do done& [1] 4737
unisoc:/sys/devices/system/cpu/cpufreq/policy7 # taskset -p 80 4737
pid 4737's current affinity mask: ff
pid 4737's new affinity mask: 80
unisoc:/sys/devices/system/cpu/cpufreq/policy7 # cat scaling_max_freq
2301000
unisoc:/sys/devices/system/cpu/cpufreq/policy7 # cat scaling_cur_freq
2301000
unisoc:/sys/devices/system/cpu/cpufreq/policy7 # echo 2171000 > scaling_max_freq
unisoc:/sys/devices/system/cpu/cpufreq/policy7 # cat scaling_max_freq
2171000
At this time, the sg_policy's next_freq would stay at 2301000, which
is wrong.
To fix this, add a check for the ->need_freq_update flag.
[ mingo: Clarified the changelog. ]
Co-developed-by: Guohua Yan <guohua.yan@unisoc.com>
Signed-off-by: Xuewen Yan <xuewen.yan@unisoc.com>
Signed-off-by: Guohua Yan <guohua.yan@unisoc.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: "Rafael J. Wysocki" <rafael@kernel.org>
Link: https://lore.kernel.org/r/20230719130527.8074-1-xuewen.yan@unisoc.com
The expectation is that placing a task at avg_vruntime() makes it
eligible. Turns out there is a corner case where this is not the case.
Specifically, avg_vruntime() relies on the fact that integer division
is a flooring function (eg. it discards the remainder). By this
property the value returned is slightly left of the true average.
However! when the average is a negative (relative to min_vruntime) the
effect is flipped and it becomes a ceil, with the result that the
returned value is just right of the average and thus not eligible.
Fixes: af4cf40470 ("sched/fair: Add cfs_rq::avg_vruntime")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
The validation of the value written to sched_rt_period_us was broken
because:
- the sysclt_sched_rt_period is declared as unsigned int
- parsed by proc_do_intvec()
- the range is asserted after the value parsed by proc_do_intvec()
Because of this negative values written to the file were written into a
unsigned integer that were later on interpreted as large positive
integers which did passed the check:
if (sysclt_sched_rt_period <= 0)
return EINVAL;
This commit fixes the parsing by setting explicit range for both
perid_us and runtime_us into the sched_rt_sysctls table and processes
the values with proc_dointvec_minmax() instead.
Alternatively if we wanted to use full range of unsigned int for the
period value we would have to split the proc_handler and use
proc_douintvec() for it however even the
Documentation/scheduller/sched-rt-group.rst describes the range as 1 to
INT_MAX.
As far as I can tell the only problem this causes is that the sysctl
file allows writing negative values which when read back may confuse
userspace.
There is also a LTP test being submitted for these sysctl files at:
http://patchwork.ozlabs.org/project/ltp/patch/20230901144433.2526-1-chrubis@suse.cz/
Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20231002115553.3007-2-chrubis@suse.cz
Pull misc fixes from Andrew Morton:
"Fourteen hotfixes, eleven of which are cc:stable. The remainder
pertain to issues which were introduced after 6.5"
* tag 'mm-hotfixes-stable-2023-10-01-08-34' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
Crash: add lock to serialize crash hotplug handling
selftests/mm: fix awk usage in charge_reserved_hugetlb.sh and hugetlb_reparenting_test.sh that may cause error
mm: mempolicy: keep VMA walk if both MPOL_MF_STRICT and MPOL_MF_MOVE are specified
mm/damon/vaddr-test: fix memory leak in damon_do_test_apply_three_regions()
mm, memcg: reconsider kmem.limit_in_bytes deprecation
mm: zswap: fix potential memory corruption on duplicate store
arm64: hugetlb: fix set_huge_pte_at() to work with all swap entries
mm: hugetlb: add huge page size param to set_huge_pte_at()
maple_tree: add MAS_UNDERFLOW and MAS_OVERFLOW states
maple_tree: add mas_is_active() to detect in-tree walks
nilfs2: fix potential use after free in nilfs_gccache_submit_read_data()
mm: abstract moving to the next PFN
mm: report success more often from filemap_map_folio_range()
fs: binfmt_elf_efpic: fix personality for ELF-FDPIC
Pull scheduler fix from Ingo Molnar:
"Fix a RT tasks related lockup/live-lock during CPU offlining"
* tag 'sched-urgent-2023-10-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/rt: Fix live lock between select_fallback_rq() and RT push
Pull tracing fixes from Steven Rostedt:
- Make sure 32-bit applications using user events have aligned access
when running on a 64-bit kernel.
- Add cond_resched in the loop that handles converting enums in
print_fmt string is trace events.
- Fix premature wake ups of polling processes in the tracing ring
buffer. When a task polls waiting for a percentage of the ring buffer
to be filled, the writer still will wake it up at every event. Add
the polling's percentage to the "shortest_full" list to tell the
writer when to wake it up.
- For eventfs dir lookups on dynamic events, an event system's only
event could be removed, leaving its dentry with no children. This is
totally legitimate. But in eventfs_release() it must not access the
children array, as it is only allocated when the dentry has children.
* tag 'trace-v6.6-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
eventfs: Test for dentries array allocated in eventfs_release()
tracing/user_events: Align set_bit() address for all archs
tracing: relax trace_event_eval_update() execution with cond_resched()
ring-buffer: Update "shortest_full" in polling
All architectures should use a long aligned address passed to set_bit().
User processes can pass either a 32-bit or 64-bit sized value to be
updated when tracing is enabled when on a 64-bit kernel. Both cases are
ensured to be naturally aligned, however, that is not enough. The
address must be long aligned without affecting checks on the value
within the user process which require different adjustments for the bit
for little and big endian CPUs.
Add a compat flag to user_event_enabler that indicates when a 32-bit
value is being used on a 64-bit kernel. Long align addresses and correct
the bit to be used by set_bit() to account for this alignment. Ensure
compat flags are copied during forks and used during deletion clears.
Link: https://lore.kernel.org/linux-trace-kernel/20230925230829.341-2-beaub@linux.microsoft.com
Link: https://lore.kernel.org/linux-trace-kernel/20230914131102.179100-1-cleger@rivosinc.com/
Cc: stable@vger.kernel.org
Fixes: 7235759084 ("tracing/user_events: Use remote writes for event enablement")
Reported-by: Clément Léger <cleger@rivosinc.com>
Suggested-by: Clément Léger <cleger@rivosinc.com>
Signed-off-by: Beau Belgrave <beaub@linux.microsoft.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
When kernel is compiled without preemption, the eval_map_work_func()
(which calls trace_event_eval_update()) will not be preempted up to its
complete execution. This can actually cause a problem since if another
CPU call stop_machine(), the call will have to wait for the
eval_map_work_func() function to finish executing in the workqueue
before being able to be scheduled. This problem was observe on a SMP
system at boot time, when the CPU calling the initcalls executed
clocksource_done_booting() which in the end calls stop_machine(). We
observed a 1 second delay because one CPU was executing
eval_map_work_func() and was not preempted by the stop_machine() task.
Adding a call to cond_resched() in trace_event_eval_update() allows
other tasks to be executed and thus continue working asynchronously
like before without blocking any pending task at boot time.
Link: https://lore.kernel.org/linux-trace-kernel/20230929191637.416931-1-cleger@rivosinc.com
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Clément Léger <cleger@rivosinc.com>
Tested-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
It was discovered that the ring buffer polling was incorrectly stating
that read would not block, but that's because polling did not take into
account that reads will block if the "buffer-percent" was set. Instead,
the ring buffer polling would say reads would not block if there was any
data in the ring buffer. This was incorrect behavior from a user space
point of view. This was fixed by commit 42fb0a1e84 by having the polling
code check if the ring buffer had more data than what the user specified
"buffer percent" had.
The problem now is that the polling code did not register itself to the
writer that it wanted to wait for a specific "full" value of the ring
buffer. The result was that the writer would wake the polling waiter
whenever there was a new event. The polling waiter would then wake up, see
that there's not enough data in the ring buffer to notify user space and
then go back to sleep. The next event would wake it up again.
Before the polling fix was added, the code would wake up around 100 times
for a hackbench 30 benchmark. After the "fix", due to the constant waking
of the writer, it would wake up over 11,0000 times! It would never leave
the kernel, so the user space behavior was still "correct", but this
definitely is not the desired effect.
To fix this, have the polling code add what it's waiting for to the
"shortest_full" variable, to tell the writer not to wake it up if the
buffer is not as full as it expects to be.
Note, after this fix, it appears that the waiter is now woken up around 2x
the times it was before (~200). This is a tremendous improvement from the
11,000 times, but I will need to spend some time to see why polling is
more aggressive in its wakeups than the read blocking code.
Link: https://lore.kernel.org/linux-trace-kernel/20230929180113.01c2cae3@rorschach.local.home
Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Fixes: 42fb0a1e84 ("tracing/ring-buffer: Have polling block on watermark")
Reported-by: Julia Lawall <julia.lawall@inria.fr>
Tested-by: Julia Lawall <julia.lawall@inria.fr>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Pull dma-mapping fixes from Christoph Hellwig:
- fix the narea calculation in swiotlb initialization (Ross Lagerwall)
- fix the check whether a device has used swiotlb (Petr Tesarik)
* tag 'dma-mapping-6.6-2023-09-30' of git://git.infradead.org/users/hch/dma-mapping:
swiotlb: fix the check whether a device has used software IO TLB
swiotlb: use the calculated number of areas
Eric reported that handling corresponding crash hotplug event can be
failed easily when many memory hotplug event are notified in a short
period. They failed because failing to take __kexec_lock.
=======
[ 78.714569] Fallback order for Node 0: 0
[ 78.714575] Built 1 zonelists, mobility grouping on. Total pages: 1817886
[ 78.717133] Policy zone: Normal
[ 78.724423] crash hp: kexec_trylock() failed, elfcorehdr may be inaccurate
[ 78.727207] crash hp: kexec_trylock() failed, elfcorehdr may be inaccurate
[ 80.056643] PEFILE: Unsigned PE binary
=======
The memory hotplug events are notified very quickly and very many, while
the handling of crash hotplug is much slower relatively. So the atomic
variable __kexec_lock and kexec_trylock() can't guarantee the
serialization of crash hotplug handling.
Here, add a new mutex lock __crash_hotplug_lock to serialize crash hotplug
handling specifically. This doesn't impact the usage of __kexec_lock.
Link: https://lkml.kernel.org/r/20230926120905.392903-1-bhe@redhat.com
Fixes: 2472627561 ("crash: add generic infrastructure for crash hotplug support")
Signed-off-by: Baoquan He <bhe@redhat.com>
Tested-by: Eric DeVolder <eric.devolder@oracle.com>
Reviewed-by: Eric DeVolder <eric.devolder@oracle.com>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Cc: Sourabh Jain <sourabhjain@linux.ibm.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
find_energy_efficient_cpu() bails out early if effective util of the
task is 0 as the delta at this point will be zero and there's nothing
for EAS to do. When uclamp is being used, this could lead to wrong
decisions when uclamp_max is set to 0. In this case the task is capped
to performance point 0, but it is actually running and consuming energy
and we can benefit from EAS energy calculations.
Rework the condition so that it bails out when both util and uclamp_min
are 0.
We can do that without needing to use uclamp_task_util(); remove it.
Fixes: d81304bc61 ("sched/uclamp: Cater for uclamp in find_energy_efficient_cpu()'s early exit condition")
Signed-off-by: Qais Yousef (Google) <qyousef@layalina.io>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230916232955.2099394-3-qyousef@layalina.io
When uclamp_max is being used, the util of the task could be higher than
the spare capacity of the CPU, but due to uclamp_max value we force-fit
it there.
The way the condition for checking for max_spare_cap in
find_energy_efficient_cpu() was constructed; it ignored any CPU that has
its spare_cap less than or _equal_ to max_spare_cap. Since we initialize
max_spare_cap to 0; this lead to never setting max_spare_cap_cpu and
hence ending up never performing compute_energy() for this cluster and
missing an opportunity for a better energy efficient placement to honour
uclamp_max setting.
max_spare_cap = 0;
cpu_cap = capacity_of(cpu) - cpu_util(p); // 0 if cpu_util(p) is high
...
util_fits_cpu(...); // will return true if uclamp_max forces it to fit
...
// this logic will fail to update max_spare_cap_cpu if cpu_cap is 0
if (cpu_cap > max_spare_cap) {
max_spare_cap = cpu_cap;
max_spare_cap_cpu = cpu;
}
prev_spare_cap suffers from a similar problem.
Fix the logic by converting the variables into long and treating -1
value as 'not populated' instead of 0 which is a viable and correct
spare capacity value. We need to be careful signed comparison is used
when comparing with cpu_cap in one of the conditions.
Fixes: 1d42509e47 ("sched/fair: Make EAS wakeup placement consider uclamp restrictions")
Signed-off-by: Qais Yousef (Google) <qyousef@layalina.io>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230916232955.2099394-2-qyousef@layalina.io
dl_rq->dl_nr_migratory is increased whenever a DL entity is enqueued and it has
nr_cpus_allowed > 1. Unlike the pushable_dl_tasks tree, dl_rq->dl_nr_migratory
includes a dl_rq's current task. This means a dl_rq can have a migratable
current, N non-migratable queued tasks, and be flagged as overloaded and have
its CPU set in the dlo_mask, despite having an empty pushable_tasks tree.
Make an dl_rq's overload logic be driven by {enqueue,dequeue}_pushable_dl_task(),
in other words make DL RQs only be flagged as overloaded if they have at
least one runnable-but-not-current migratable task.
o push_dl_task() is unaffected, as it is a no-op if there are no pushable
tasks.
o pull_dl_task() now no longer scans runqueues whose sole migratable task is
their current one, which it can't do anything about anyway.
It may also now pull tasks to a DL RQ with dl_nr_running > 1 if only its
current task is migratable.
Since dl_rq->dl_nr_migratory becomes unused, remove it.
RT had the exact same mechanism (rt_rq->rt_nr_migratory) which was dropped
in favour of relying on rt_rq->pushable_tasks, see:
612f769edd ("sched/rt: Make rt_rq->pushable_tasks updates drive rto_mask")
Signed-off-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://lore.kernel.org/r/20230928150251.463109-1-vschneid@redhat.com
During RCU-boost testing with the TREE03 rcutorture config, I found that
after a few hours, the machine locks up.
On tracing, I found that there is a live lock happening between 2 CPUs.
One CPU has an RT task running, while another CPU is being offlined
which also has an RT task running. During this offlining, all threads
are migrated. The migration thread is repeatedly scheduled to migrate
actively running tasks on the CPU being offlined. This results in a live
lock because select_fallback_rq() keeps picking the CPU that an RT task
is already running on only to get pushed back to the CPU being offlined.
It is anyway pointless to pick CPUs for pushing tasks to if they are
being offlined only to get migrated away to somewhere else. This could
also add unwanted latency to this task.
Fix these issues by not selecting CPUs in RT if they are not 'active'
for scheduling, using the cpu_active_mask. Other parts in core.c already
use cpu_active_mask to prevent tasks from being put on CPUs going
offline.
With this fix I ran the tests for days and could not reproduce the
hang. Without the patch, I hit it in a few hours.
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Paul E. McKenney <paulmck@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230923011409.3522762-1-joel@joelfernandes.org
When CONFIG_SWIOTLB_DYNAMIC=y, devices which do not use the software IO TLB
can avoid swiotlb lookup. A flag is added by commit 1395706a14 ("swiotlb:
search the software IO TLB only if the device makes use of it"), the flag
is correctly set, but it is then never checked. Add the actual check here.
Note that this code is an alternative to the default pool check, not an
additional check, because:
1. swiotlb_find_pool() also searches the default pool;
2. if dma_uses_io_tlb is false, the default swiotlb pool is not used.
Tested in a KVM guest against a QEMU RAM-backed SATA disk over virtio and
*not* using software IO TLB, this patch increases IOPS by approx 2% for
4-way parallel I/O.
The write memory barrier in swiotlb_dyn_alloc() is not needed, because a
newly allocated pool must always be observed by swiotlb_find_slots() before
an address from that pool is passed to is_swiotlb_buffer().
Correctness was verified using the following litmus test:
C swiotlb-new-pool
(*
* Result: Never
*
* Check that a newly allocated pool is always visible when the
* corresponding swiotlb buffer is visible.
*)
{
mem_pools = default;
}
P0(int **mem_pools, int *pool)
{
/* add_mem_pool() */
WRITE_ONCE(*pool, 999);
rcu_assign_pointer(*mem_pools, pool);
}
P1(int **mem_pools, int *flag, int *buf)
{
/* swiotlb_find_slots() */
int *r0;
int r1;
rcu_read_lock();
r0 = READ_ONCE(*mem_pools);
r1 = READ_ONCE(*r0);
rcu_read_unlock();
if (r1) {
WRITE_ONCE(*flag, 1);
smp_mb();
}
/* device driver (presumed) */
WRITE_ONCE(*buf, r1);
}
P2(int **mem_pools, int *flag, int *buf)
{
/* device driver (presumed) */
int r0 = READ_ONCE(*buf);
/* is_swiotlb_buffer() */
int r1;
int *r2;
int r3;
smp_rmb();
r1 = READ_ONCE(*flag);
if (r1) {
/* swiotlb_find_pool() */
rcu_read_lock();
r2 = READ_ONCE(*mem_pools);
r3 = READ_ONCE(*r2);
rcu_read_unlock();
}
}
exists (2:r0<>0 /\ 2:r3=0) (* Not found. *)
Fixes: 1395706a14 ("swiotlb: search the software IO TLB only if the device makes use of it")
Reported-by: Jonathan Corbet <corbet@lwn.net>
Closes: https://lore.kernel.org/linux-iommu/87a5uz3ob8.fsf@meer.lwn.net/
Signed-off-by: Petr Tesarik <petr@tesarici.cz>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Pull workqueue fixes from Tejun Heo:
- Remove double allocation of wq_update_pod_attrs_buf
- Fix missing allocation of pwq_release_worker when
wq_cpu_intensive_thresh_us is set to a custom value
* tag 'wq-for-6.6-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: Fix missed pwq_release_worker creation in wq_cpu_intensive_thresh_init()
workqueue: Removed double allocation of wq_update_pod_attrs_buf
Sebastian noted that the rto_push_work IRQ work can be queued for a CPU
that has an empty pushable_tasks list, which means nothing useful will be
done in the IPI other than queue the work for the next CPU on the rto_mask.
rto_push_irq_work_func() only operates on tasks in the pushable_tasks list,
but the conditions for that irq_work to be queued (and for a CPU to be
added to the rto_mask) rely on rq_rt->nr_migratory instead.
nr_migratory is increased whenever an RT task entity is enqueued and it has
nr_cpus_allowed > 1. Unlike the pushable_tasks list, nr_migratory includes a
rt_rq's current task. This means a rt_rq can have a migratible current, N
non-migratible queued tasks, and be flagged as overloaded / have its CPU
set in the rto_mask, despite having an empty pushable_tasks list.
Make an rt_rq's overload logic be driven by {enqueue,dequeue}_pushable_task().
Since rt_rq->{rt_nr_migratory,rt_nr_total} become unused, remove them.
Note that the case where the current task is pushed away to make way for a
migration-disabled task remains unchanged: the migration-disabled task has
to be in the pushable_tasks list in the first place, which means it has
nr_cpus_allowed > 1.
Reported-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/20230811112044.3302588-1-vschneid@redhat.com
Pull tracing fixes from Steven Rostedt:
- Fix the "bytes" output of the per_cpu stat file
The tracefs/per_cpu/cpu*/stats "bytes" was giving bogus values as the
accounting was not accurate. It is suppose to show how many used
bytes are still in the ring buffer, but even when the ring buffer was
empty it would still show there were bytes used.
- Fix a bug in eventfs where reading a dynamic event directory (open)
and then creating a dynamic event that goes into that diretory screws
up the accounting.
On close, the newly created event dentry will get a "dput" without
ever having a "dget" done for it. The fix is to allocate an array on
dir open to save what dentries were actually "dget" on, and what ones
to "dput" on close.
* tag 'trace-v6.6-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
eventfs: Remember what dentries were created on dir open
ring-buffer: Fix bytes info in per_cpu buffer stats
Simplify the conditional logic for checking worker flags
by splitting the original compound `if` statement into
separate `if` and `else if` clauses.
This modification not only retains the previous functionality,
but also reduces a single `if` check, improving code clarity
and potentially enhancing performance.
Signed-off-by: Wang Jinchao <wangjinchao@xfusion.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/ZOIMvURE99ZRAYEj@fedora
We've observed the following warning being hit in
distribute_cfs_runtime():
SCHED_WARN_ON(cfs_rq->runtime_remaining > 0)
We have the following race:
- CPU 0: running bandwidth distribution (distribute_cfs_runtime).
Inspects the local cfs_rq and makes its runtime_remaining positive.
However, we defer unthrottling the local cfs_rq until after
considering all remote cfs_rq's.
- CPU 1: starts running bandwidth distribution from the slack timer. When
it finds the cfs_rq for CPU 0 on the throttled list, it observers the
that the cfs_rq is throttled, yet is not on the CSD list, and has a
positive runtime_remaining, thus triggering the warning in
distribute_cfs_runtime.
To fix this, we can rework the local unthrottling logic to put the local
cfs_rq on a local list, so that any future bandwidth distributions will
realize that the cfs_rq is about to be unthrottled.
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230922230535.296350-2-joshdon@google.com
Pull misc fixes from Andrew Morton:
"13 hotfixes, 10 of which pertain to post-6.5 issues. The other three
are cc:stable"
* tag 'mm-hotfixes-stable-2023-09-23-10-31' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
proc: nommu: fix empty /proc/<pid>/maps
filemap: add filemap_map_order0_folio() to handle order0 folio
proc: nommu: /proc/<pid>/maps: release mmap read lock
mm: memcontrol: fix GFP_NOFS recursion in memory.high enforcement
pidfd: prevent a kernel-doc warning
argv_split: fix kernel-doc warnings
scatterlist: add missing function params to kernel-doc
selftests/proc: fixup proc-empty-vm test after KSM changes
revert "scripts/gdb/symbols: add specific ko module load command"
selftests: link libasan statically for tests with -fsanitize=address
task_work: add kerneldoc annotation for 'data' argument
mm: page_alloc: fix CMA and HIGHATOMIC landing on the wrong buddy list
sh: mm: re-add lost __ref to ioremap_prot() to fix modpost warning
The 'bytes' info in file 'per_cpu/cpu<X>/stats' means the number of
bytes in cpu buffer that have not been consumed. However, currently
after consuming data by reading file 'trace_pipe', the 'bytes' info
was not changed as expected.
# cat per_cpu/cpu0/stats
entries: 0
overrun: 0
commit overrun: 0
bytes: 568 <--- 'bytes' is problematical !!!
oldest event ts: 8651.371479
now ts: 8653.912224
dropped events: 0
read events: 8
The root cause is incorrect stat on cpu_buffer->read_bytes. To fix it:
1. When stat 'read_bytes', account consumed event in rb_advance_reader();
2. When stat 'entries_bytes', exclude the discarded padding event which
is smaller than minimum size because it is invisible to reader. Then
use rb_page_commit() instead of BUF_PAGE_SIZE at where accounting for
page-based read/remove/overrun.
Also correct the comments of ring_buffer_bytes_cpu() in this patch.
Link: https://lore.kernel.org/linux-trace-kernel/20230921125425.1708423-1-zhengyejian1@huawei.com
Cc: stable@vger.kernel.org
Fixes: c64e148a3b ("trace: Add ring buffer stats to measure rate of events")
Signed-off-by: Zheng Yejian <zhengyejian1@huawei.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Pull scheduler fix from Ingo Molnar:
"Fix a PF_IDLE initialization bug that generated warnings on tiny-RCU"
* tag 'sched-urgent-2023-09-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
kernel/sched: Modify initial boot task idle setup
Pull networking fixes from Paolo Abeni:
"Including fixes from netfilter and bpf.
Current release - regressions:
- bpf: adjust size_index according to the value of KMALLOC_MIN_SIZE
- netfilter: fix entries val in rule reset audit log
- eth: stmmac: fix incorrect rxq|txq_stats reference
Previous releases - regressions:
- ipv4: fix null-deref in ipv4_link_failure
- netfilter:
- fix several GC related issues
- fix race between IPSET_CMD_CREATE and IPSET_CMD_SWAP
- eth: team: fix null-ptr-deref when team device type is changed
- eth: i40e: fix VF VLAN offloading when port VLAN is configured
- eth: ionic: fix 16bit math issue when PAGE_SIZE >= 64KB
Previous releases - always broken:
- core: fix ETH_P_1588 flow dissector
- mptcp: fix several connection hang-up conditions
- bpf:
- avoid deadlock when using queue and stack maps from NMI
- add override check to kprobe multi link attach
- hsr: properly parse HSRv1 supervisor frames.
- eth: igc: fix infinite initialization loop with early XDP redirect
- eth: octeon_ep: fix tx dma unmap len values in SG
- eth: hns3: fix GRE checksum offload issue"
* tag 'net-6.6-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (87 commits)
sfc: handle error pointers returned by rhashtable_lookup_get_insert_fast()
igc: Expose tx-usecs coalesce setting to user
octeontx2-pf: Do xdp_do_flush() after redirects.
bnxt_en: Flush XDP for bnxt_poll_nitroa0()'s NAPI
net: ena: Flush XDP packets on error.
net/handshake: Fix memory leak in __sock_create() and sock_alloc_file()
net: hinic: Fix warning-hinic_set_vlan_fliter() warn: variable dereferenced before check 'hwdev'
netfilter: ipset: Fix race between IPSET_CMD_CREATE and IPSET_CMD_SWAP
netfilter: nf_tables: fix memleak when more than 255 elements expired
netfilter: nf_tables: disable toggling dormant table state more than once
vxlan: Add missing entries to vxlan_get_size()
net: rds: Fix possible NULL-pointer dereference
team: fix null-ptr-deref when team device type is changed
net: bridge: use DEV_STATS_INC()
net: hns3: add 5ms delay before clear firmware reset irq source
net: hns3: fix fail to delete tc flower rules during reset issue
net: hns3: only enable unicast promisc when mac table full
net: hns3: fix GRE checksum offload issue
net: hns3: add cmdq check for vf periodic service task
net: stmmac: fix incorrect rxq|txq_stats reference
...
Initial booting is setting the task flag to idle (PF_IDLE) by the call
path sched_init() -> init_idle(). Having the task idle and calling
call_rcu() in kernel/rcu/tiny.c means that TIF_NEED_RESCHED will be
set. Subsequent calls to any cond_resched() will enable IRQs,
potentially earlier than the IRQ setup has completed. Recent changes
have caused just this scenario and IRQs have been enabled early.
This causes a warning later in start_kernel() as interrupts are enabled
before they are fully set up.
Fix this issue by setting the PF_IDLE flag later in the boot sequence.
Although the boot task was marked as idle since (at least) d80e4fda576d,
I am not sure that it is wrong to do so. The forced context-switch on
idle task was introduced in the tiny_rcu update, so I'm going to claim
this fixes 5f6130fa52.
Fixes: 5f6130fa52 ("tiny_rcu: Directly force QS when call_rcu_[bh|sched]() on idle_task")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/linux-mm/CAMuHMdWpvpWoDa=Ox-do92czYRvkok6_x6pYUH+ZouMcJbXy+Q@mail.gmail.com/
The name is a bit opaque - make it clear that this is about wakeup
preemption.
Also rename the ->check_preempt_curr() methods similarly.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Other scheduling classes already postfix their similar methods
with the class name.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Currently, if the wq_cpu_intensive_thresh_us is set to specific
value, will cause the wq_cpu_intensive_thresh_init() early exit
and missed creation of pwq_release_worker. this commit therefore
create the pwq_release_worker in advance before checking the
wq_cpu_intensive_thresh_us.
Signed-off-by: Zqiang <qiang.zhang1211@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Fixes: 967b494e2f ("workqueue: Use a kthread_worker to release pool_workqueues")
First commit 2930155b2e ("workqueue: Initialize unbound CPU pods later in
the boot") added the initialization of wq_update_pod_attrs_buf to
workqueue_init_early(), and then latter on, commit 84193c0710
("workqueue: Generalize unbound CPU pods") added it as well. This appeared
in a kmemleak run where the second allocation made the first allocation
leak.
Fixes: 84193c0710 ("workqueue: Generalize unbound CPU pods")
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Tejun Heo <tj@kernel.org>
When using sysbench to benchmark Postgres in a single docker instance
with sysbench's nr_threads set to nr_cpu, it is observed there are times
update_cfs_group() and update_load_avg() shows noticeable overhead on
a 2sockets/112core/224cpu Intel Sapphire Rapids(SPR):
13.75% 13.74% [kernel.vmlinux] [k] update_cfs_group
10.63% 10.04% [kernel.vmlinux] [k] update_load_avg
Annotate shows the cycles are mostly spent on accessing tg->load_avg
with update_load_avg() being the write side and update_cfs_group() being
the read side. tg->load_avg is per task group and when different tasks
of the same taskgroup running on different CPUs frequently access
tg->load_avg, it can be heavily contended.
E.g. when running postgres_sysbench on a 2sockets/112cores/224cpus Intel
Sappire Rapids, during a 5s window, the wakeup number is 14millions and
migration number is 11millions and with each migration, the task's load
will transfer from src cfs_rq to target cfs_rq and each change involves
an update to tg->load_avg. Since the workload can trigger as many wakeups
and migrations, the access(both read and write) to tg->load_avg can be
unbound. As a result, the two mentioned functions showed noticeable
overhead. With netperf/nr_client=nr_cpu/UDP_RR, the problem is worse:
during a 5s window, wakeup number is 21millions and migration number is
14millions; update_cfs_group() costs ~25% and update_load_avg() costs ~16%.
Reduce the overhead by limiting updates to tg->load_avg to at most once
per ms. The update frequency is a tradeoff between tracking accuracy and
overhead. 1ms is chosen because PELT window is roughly 1ms and it
delivered good results for the tests that I've done. After this change,
the cost of accessing tg->load_avg is greatly reduced and performance
improved. Detailed test results below.
==============================
postgres_sysbench on SPR:
25%
base: 42382±19.8%
patch: 50174±9.5% (noise)
50%
base: 67626±1.3%
patch: 67365±3.1% (noise)
75%
base: 100216±1.2%
patch: 112470±0.1% +12.2%
100%
base: 93671±0.4%
patch: 113563±0.2% +21.2%
==============================
hackbench on ICL:
group=1
base: 114912±5.2%
patch: 117857±2.5% (noise)
group=4
base: 359902±1.6%
patch: 361685±2.7% (noise)
group=8
base: 461070±0.8%
patch: 491713±0.3% +6.6%
group=16
base: 309032±5.0%
patch: 378337±1.3% +22.4%
=============================
hackbench on SPR:
group=1
base: 100768±2.9%
patch: 103134±2.9% (noise)
group=4
base: 413830±12.5%
patch: 378660±16.6% (noise)
group=8
base: 436124±0.6%
patch: 490787±3.2% +12.5%
group=16
base: 457730±3.2%
patch: 680452±1.3% +48.8%
============================
netperf/udp_rr on ICL
25%
base: 114413±0.1%
patch: 115111±0.0% +0.6%
50%
base: 86803±0.5%
patch: 86611±0.0% (noise)
75%
base: 35959±5.3%
patch: 49801±0.6% +38.5%
100%
base: 61951±6.4%
patch: 70224±0.8% +13.4%
===========================
netperf/udp_rr on SPR
25%
base: 104954±1.3%
patch: 107312±2.8% (noise)
50%
base: 55394±4.6%
patch: 54940±7.4% (noise)
75%
base: 13779±3.1%
patch: 36105±1.1% +162%
100%
base: 9703±3.7%
patch: 28011±0.2% +189%
==============================================
netperf/tcp_stream on ICL (all in noise range)
25%
base: 43092±0.1%
patch: 42891±0.5%
50%
base: 19278±14.9%
patch: 22369±7.2%
75%
base: 16822±3.0%
patch: 17086±2.3%
100%
base: 18216±0.6%
patch: 18078±2.9%
===============================================
netperf/tcp_stream on SPR (all in noise range)
25%
base: 34491±0.3%
patch: 34886±0.5%
50%
base: 19278±14.9%
patch: 22369±7.2%
75%
base: 16822±3.0%
patch: 17086±2.3%
100%
base: 18216±0.6%
patch: 18078±2.9%
Reported-by: Nitin Tekchandani <nitin.tekchandani@intel.com>
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Reviewed-by: David Vernet <void@manifault.com>
Tested-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Tested-by: Swapnil Sapkal <Swapnil.Sapkal@amd.com>
Link: https://lkml.kernel.org/r/20230912065808.2530-2-aaron.lu@intel.com
After commit f5d39b0208 ("freezer,sched: Rewrite core freezer logic"),
tasks that transition directly from TASK_FREEZABLE to TASK_FROZEN are
always woken up on the thaw path. Prior to that commit, tasks could ask
freezer to consider them "frozen enough" via freezer_do_not_count(). The
commit replaced freezer_do_not_count() with a TASK_FREEZABLE state which
allows freezer to immediately mark the task as TASK_FROZEN without
waking up the task. This is efficient for the suspend path, but on the
thaw path, the task is always woken up even if the task didn't need to
wake up and goes back to its TASK_(UN)INTERRUPTIBLE state. Although
these tasks are capable of handling of the wakeup, we can observe a
power/perf impact from the extra wakeup.
We observed on Android many tasks wait in the TASK_FREEZABLE state
(particularly due to many of them being binder clients). We observed
nearly 4x the number of tasks and a corresponding linear increase in
latency and power consumption when thawing the system. The latency
increased from ~15ms to ~50ms.
Avoid the spurious wakeups by saving the state of TASK_FREEZABLE tasks.
If the task was running before entering TASK_FROZEN state
(__refrigerator()) or if the task received a wake up for the saved
state, then the task is woken on thaw. saved_state from PREEMPT_RT locks
can be re-used because freezer would not stomp on the rtlock wait flow:
TASK_RTLOCK_WAIT isn't considered freezable.
Reported-by: Prakash Viswalingam <quic_prakashv@quicinc.com>
Signed-off-by: Elliot Berman <quic_eberman@quicinc.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In preparation for freezer to also use saved_state, remove the
CONFIG_PREEMPT_RT compilation guard around saved_state.
On the arm64 platform I tested which did not have CONFIG_PREEMPT_RT,
there was no statistically significant deviation by applying this patch.
Test methodology:
perf bench sched message -g 40 -l 40
Signed-off-by: Elliot Berman <quic_eberman@quicinc.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler fixes from Ingo Molnar:
"Fix a performance regression on large SMT systems, an Intel SMT4
balancing bug, and a topology setup bug on (Intel) hybrid processors"
* tag 'sched-urgent-2023-09-17' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/sched: Restore the SD_ASYM_PACKING flag in the DIE domain
sched/fair: Fix SMT4 group_smt_balance handling
sched/fair: Optimize should_we_balance() for large SMT systems
