Add the basic infrastructure to split the TIF_NEED_RESCHED bit in two.
Either bit will cause a resched on return-to-user, but only
TIF_NEED_RESCHED will drive IRQ preemption.
No behavioural change intended.
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lkml.kernel.org/r/20241007075055.219540785@infradead.org
Instead of solving the underlying problem of the double invocation of
__sched_fork() for idle tasks, sched-ext decided to hack around the issue
by partially clearing out the entity struct to preserve the already
enqueued node. A provided analysis and solution has been ignored for four
months.
Now that someone else has taken care of cleaning it up, remove the
disgusting hack and clear out the full structure. Remove the comment in the
structure declaration as well, as there is no requirement for @node being
the last element anymore.
Fixes: f0e1a0643a ("sched_ext: Implement BPF extensible scheduler class")
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/r/87ldy82wkc.ffs@tglx
Idle tasks are initialized via __sched_fork() twice:
fork_idle()
copy_process()
sched_fork()
__sched_fork()
init_idle()
__sched_fork()
Instead of cleaning this up, sched_ext hacked around it. Even when analyis
and solution were provided in a discussion, nobody cared to clean this up.
init_idle() is also invoked from sched_init() to initialize the boot CPU's
idle task, which requires the __sched_fork() invocation. But this can be
trivially solved by invoking __sched_fork() before init_idle() in
sched_init() and removing the __sched_fork() invocation from init_idle().
Do so and clean up the comments explaining this historical leftover.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20241028103142.359584747@linutronix.de
What psi needs to do on each enqueue and dequeue has gotten more
subtle, and the generic sched code trying to distill this into a bool
for the callbacks is awkward.
Pass the flags directly and let psi parse them. For that to work, the
#include "stats.h" (which has the psi callback implementations) needs
to be below the flag definitions in "sched.h". Move that section
further down, next to some of the other accounting stuff.
This also puts the ENQUEUE_SAVE/RESTORE branch behind the psi jump
label, slightly reducing overhead when PSI=y but runtime disabled.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20241014144358.GB1021@cmpxchg.org
Conflicts:
kernel/sched/ext.c
There's a context conflict between this upstream commit:
3fdb9ebcec sched_ext: Start schedulers with consistent p->scx.slice values
... and this fix in sched/urgent:
98442f0ccd sched: Fix delayed_dequeue vs switched_from_fair()
Resolve it.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
- More issues reported in the enable/disable paths on large machines with
many tasks due to scx_tasks_lock being held too long. Break up the task
iterations.
- Remove ops.select_cpu() dependency in bypass mode so that a misbehaving
implementation can't live-lock the machine by pushing all tasks to few
CPUs in bypass mode.
- Other misc fixes.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCZw7SMw4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGU6+AQDZKNsyRw0c8+rd7eASAtCzvzBU1KGf3RPrsbNF
pb3IGgEA4/uVBlVny10ZyeKaQ4A3L2G4ROyJwNEygUvhkvbfcgk=
=z7d9
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-6.12-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- More issues reported in the enable/disable paths on large machines
with many tasks due to scx_tasks_lock being held too long. Break up
the task iterations
- Remove ops.select_cpu() dependency in bypass mode so that a
misbehaving implementation can't live-lock the machine by pushing all
tasks to few CPUs in bypass mode
- Other misc fixes
* tag 'sched_ext-for-6.12-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: Remove unnecessary cpu_relax()
sched_ext: Don't hold scx_tasks_lock for too long
sched_ext: Move scx_tasks_lock handling into scx_task_iter helpers
sched_ext: bypass mode shouldn't depend on ops.select_cpu()
sched_ext: Move scx_buildin_idle_enabled check to scx_bpf_select_cpu_dfl()
sched_ext: Start schedulers with consistent p->scx.slice values
Revert "sched_ext: Use shorter slice while bypassing"
sched_ext: use correct function name in pick_task_scx() warning message
selftests: sched_ext: Add sched_ext as proper selftest target
As described in commit b07996c7ab ("sched_ext: Don't hold
scx_tasks_lock for too long"), we're doing a cond_resched() every 32
calls to scx_task_iter_next() to avoid RCU and other stalls. That commit
also added a cpu_relax() to the codepath where we drop and reacquire the
lock, but as Waiman described in [0], cpu_relax() should only be
necessary in busy loops to avoid pounding on a cacheline (or to allow a
hypertwin to more fully utilize a core).
Let's remove the unnecessary cpu_relax().
[0]: https://lore.kernel.org/all/35b3889b-904a-4d26-981f-c8aa1557a7c7@redhat.com/
Cc: Waiman Long <llong@redhat.com>
Signed-off-by: David Vernet <void@manifault.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Let's define the "scheduling context" as all the scheduler state
in task_struct for the task chosen to run, which we'll call the
donor task, and the "execution context" as all state required to
actually run the task.
Currently both are intertwined in task_struct. We want to
logically split these such that we can use the scheduling
context of the donor task selected to be scheduled, but use
the execution context of a different task to actually be run.
To this purpose, introduce rq->donor field to point to the
task_struct chosen from the runqueue by the scheduler, and will
be used for scheduler state, and preserve rq->curr to indicate
the execution context of the task that will actually be run.
This patch introduces the donor field as a union with curr, so it
doesn't cause the contexts to be split yet, but adds the logic to
handle everything separately.
[add additional comments and update more sched_class code to use
rq::proxy]
[jstultz: Rebased and resolved minor collisions, reworked to use
accessors, tweaked update_curr_common to use rq_proxy fixing rt
scheduling issues]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Juri Lelli <juri.lelli@redhat.com>
Signed-off-by: Connor O'Brien <connoro@google.com>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Metin Kaya <metin.kaya@arm.com>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: Metin Kaya <metin.kaya@arm.com>
Link: https://lore.kernel.org/r/20241009235352.1614323-8-jstultz@google.com
As we're going to re-use the deactivation logic,
split it into a helper.
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Metin Kaya <metin.kaya@arm.com>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: Metin Kaya <metin.kaya@arm.com>
Link: https://lore.kernel.org/r/20241009235352.1614323-7-jstultz@google.com
This patch consolidates rt and deadline pick_*_task functions to
a task_is_pushable() helper
This patch was broken out from a larger chain migration
patch originally by Connor O'Brien.
[jstultz: split out from larger chain migration patch,
renamed helper function]
Signed-off-by: Connor O'Brien <connoro@google.com>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Metin Kaya <metin.kaya@arm.com>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: Metin Kaya <metin.kaya@arm.com>
Link: https://lore.kernel.org/r/20241009235352.1614323-6-jstultz@google.com
Switch logic that deactivates, sets the task cpu,
and reactivates a task on a different rq to use a
helper that will be later extended to push entire
blocked task chains.
This patch was broken out from a larger chain migration
patch originally by Connor O'Brien.
[jstultz: split out from larger chain migration patch]
Signed-off-by: Connor O'Brien <connoro@google.com>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Metin Kaya <metin.kaya@arm.com>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: Metin Kaya <metin.kaya@arm.com>
Link: https://lore.kernel.org/r/20241009235352.1614323-5-jstultz@google.com
commit 223baf9d17 ("sched: Fix performance regression introduced by mm_cid")
introduced a per-mm/cpu current concurrency id (mm_cid), which keeps
a reference to the concurrency id allocated for each CPU. This reference
expires shortly after a 100ms delay.
These per-CPU references keep the per-mm-cid data cache-local in
situations where threads are running at least once on each CPU within
each 100ms window, thus keeping the per-cpu reference alive.
However, intermittent workloads behaving in bursts spaced by more than
100ms on each CPU exhibit bad cache locality and degraded performance
compared to purely per-cpu data indexing, because concurrency IDs are
allocated over various CPUs and cores, therefore losing cache locality
of the associated data.
Introduce the following changes to improve per-mm-cid cache locality:
- Add a "recent_cid" field to the per-mm/cpu mm_cid structure to keep
track of which mm_cid value was last used, and use it as a hint to
attempt re-allocating the same concurrency ID the next time this
mm/cpu needs to allocate a concurrency ID,
- Add a per-mm CPUs allowed mask, which keeps track of the union of
CPUs allowed for all threads belonging to this mm. This cpumask is
only set during the lifetime of the mm, never cleared, so it
represents the union of all the CPUs allowed since the beginning of
the mm lifetime (note that the mm_cpumask() is really arch-specific
and tailored to the TLB flush needs, and is thus _not_ a viable
approach for this),
- Add a per-mm nr_cpus_allowed to keep track of the weight of the
per-mm CPUs allowed mask (for fast access),
- Add a per-mm max_nr_cid to keep track of the highest number of
concurrency IDs allocated for the mm. This is used for expanding the
concurrency ID allocation within the upper bound defined by:
min(mm->nr_cpus_allowed, mm->mm_users)
When the next unused CID value reaches this threshold, stop trying
to expand the cid allocation and use the first available cid value
instead.
Spreading allocation to use all the cid values within the range
[ 0, min(mm->nr_cpus_allowed, mm->mm_users) - 1 ]
improves cache locality while preserving mm_cid compactness within the
expected user limits,
- In __mm_cid_try_get, only return cid values within the range
[ 0, mm->nr_cpus_allowed ] rather than [ 0, nr_cpu_ids ]. This
prevents allocating cids above the number of allowed cpus in
rare scenarios where cid allocation races with a concurrent
remote-clear of the per-mm/cpu cid. This improvement is made
possible by the addition of the per-mm CPUs allowed mask,
- In sched_mm_cid_migrate_to, use mm->nr_cpus_allowed rather than
t->nr_cpus_allowed. This criterion was really meant to compare
the number of mm->mm_users to the number of CPUs allowed for the
entire mm. Therefore, the prior comparison worked fine when all
threads shared the same CPUs allowed mask, but not so much in
scenarios where those threads have different masks (e.g. each
thread pinned to a single CPU). This improvement is made
possible by the addition of the per-mm CPUs allowed mask.
* Benchmarks
Each thread increments 16kB worth of 8-bit integers in bursts, with
a configurable delay between each thread's execution. Each thread run
one after the other (no threads run concurrently). The order of
thread execution in the sequence is random. The thread execution
sequence begins again after all threads have executed. The 16kB areas
are allocated with rseq_mempool and indexed by either cpu_id, mm_cid
(not cache-local), or cache-local mm_cid. Each thread is pinned to its
own core.
Testing configurations:
8-core/1-L3: Use 8 cores within a single L3
24-core/24-L3: Use 24 cores, 1 core per L3
192-core/24-L3: Use 192 cores (all cores in the system)
384-thread/24-L3: Use 384 HW threads (all HW threads in the system)
Intermittent workload delays between threads: 200ms, 10ms.
Hardware:
CPU(s): 384
On-line CPU(s) list: 0-383
Vendor ID: AuthenticAMD
Model name: AMD EPYC 9654 96-Core Processor
Thread(s) per core: 2
Core(s) per socket: 96
Socket(s): 2
Caches (sum of all):
L1d: 6 MiB (192 instances)
L1i: 6 MiB (192 instances)
L2: 192 MiB (192 instances)
L3: 768 MiB (24 instances)
Each result is an average of 5 test runs. The cache-local speedup
is calculated as: (cache-local mm_cid) / (mm_cid).
Intermittent workload delay: 200ms
per-cpu mm_cid cache-local mm_cid cache-local speedup
(ns) (ns) (ns)
8-core/1-L3 1374 19289 1336 14.4x
24-core/24-L3 2423 26721 1594 16.7x
192-core/24-L3 2291 15826 2153 7.3x
384-thread/24-L3 1874 13234 1907 6.9x
Intermittent workload delay: 10ms
per-cpu mm_cid cache-local mm_cid cache-local speedup
(ns) (ns) (ns)
8-core/1-L3 662 756 686 1.1x
24-core/24-L3 1378 3648 1035 3.5x
192-core/24-L3 1439 10833 1482 7.3x
384-thread/24-L3 1503 10570 1556 6.8x
[ This deprecates the prior "sched: NUMA-aware per-memory-map concurrency IDs"
patch series with a simpler and more general approach. ]
[ This patch applies on top of v6.12-rc1. ]
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Marco Elver <elver@google.com>
Link: https://lore.kernel.org/lkml/20240823185946.418340-1-mathieu.desnoyers@efficios.com/
The memory barrier rmb() in generic idle loop do_idle() function is not
needed, it doesn't order any load instruction, just remove it as needless
rmb() can cause performance impact.
The rmb() was introduced by the tglx/history.git commit f2f1b44c75c4
("[PATCH] Remove RCU abuse in cpu_idle()") to order the loads between
cpu_idle_map and pm_idle. It pairs with wmb() in function cpu_idle_wait().
And then with the removal of cpu_idle_state in function cpu_idle() and
wmb() in function cpu_idle_wait() in commit 783e391b7b ("x86: Simplify
cpu_idle_wait"), rmb() no longer has a reason to exist.
After that, commit d166991234 ("idle: Implement generic idle function")
implemented a generic idle function cpu_idle_loop() which resembles the
functionality found in arch/. And it retained the rmb() in generic idle
loop in file kernel/cpu/idle.c.
And at last, commit cf37b6b484 ("sched/idle: Move cpu/idle.c to
sched/idle.c") moved cpu/idle.c to sched/idle.c. And commit c1de45ca83
("sched/idle: Add support for tasks that inject idle") renamed function
cpu_idle_loop() to do_idle().
History Tree: https://git.kernel.org/pub/scm/linux/kernel/git/tglx/history.git
Signed-off-by: Zhongqiu Han <quic_zhonhan@quicinc.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20241009093745.9504-1-quic_zhonhan@quicinc.com
Sean noted that ever since commit 152e11f6df ("sched/fair: Implement
delayed dequeue") KVM's preemption notifiers have started
mis-classifying preemption vs blocking.
Notably p->on_rq is no longer sufficient to determine if a task is
runnable or blocked -- the aforementioned commit introduces tasks that
remain on the runqueue even through they will not run again, and
should be considered blocked for many cases.
Add the task_is_runnable() helper to classify things and audit all
external users of the p->on_rq state. Also add a few comments.
Fixes: 152e11f6df ("sched/fair: Implement delayed dequeue")
Reported-by: Sean Christopherson <seanjc@google.com>
Tested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/20241010091843.GK33184@noisy.programming.kicks-ass.net
Since sched_delayed tasks remain queued even after blocking, the load
balancer can migrate them between runqueues while PSI considers them
to be asleep. As a result, it misreads the migration requeue followed
by a wakeup as a double queue:
psi: inconsistent task state! task=... cpu=... psi_flags=4 clear=. set=4
First, call psi_enqueue() after p->sched_class->enqueue_task(). A
wakeup will clear p->se.sched_delayed while a migration will not, so
psi can use that flag to tell them apart.
Then teach psi to migrate any "sleep" state when delayed-dequeue tasks
are being migrated.
Delayed-dequeue tasks can be revived by ttwu_runnable(), which will
call down with a new ENQUEUE_DELAYED. Instead of further complicating
the wakeup conditional in enqueue_task(), identify migration contexts
instead and default to wakeup handling for all other cases.
It's not just the warning in dmesg, the task state corruption causes a
permanent CPU pressure indication, which messes with workload/machine
health monitoring.
Debugged-by-and-original-fix-by: K Prateek Nayak <kprateek.nayak@amd.com>
Fixes: 152e11f6df ("sched/fair: Implement delayed dequeue")
Closes: https://lore.kernel.org/lkml/20240830123458.3557-1-spasswolf@web.de/
Closes: https://lore.kernel.org/all/cd67fbcd-d659-4822-bb90-7e8fbb40a856@molgen.mpg.de/
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://lkml.kernel.org/r/20241010193712.GC181795@cmpxchg.org
psi_dequeue() in for blocked task expects psi_sched_switch() to clear
the TSK_.*RUNNING PSI flags and set the TSK_IOWAIT flags however
psi_sched_switch() uses "!task_on_rq_queued(prev)" to detect if the task
is blocked or still runnable which is no longer true with DELAY_DEQUEUE
since a blocking task can be left queued on the runqueue.
This can lead to PSI splats similar to:
psi: inconsistent task state! task=... cpu=... psi_flags=4 clear=0 set=4
when the task is requeued since the TSK_RUNNING flag was not cleared
when the task was blocked.
Explicitly communicate that the task was blocked to psi_sched_switch()
even if it was delayed and is still on the runqueue.
[ prateek: Broke off the relevant part from [1], commit message ]
Fixes: 152e11f6df ("sched/fair: Implement delayed dequeue")
Closes: https://lore.kernel.org/lkml/20240830123458.3557-1-spasswolf@web.de/
Closes: https://lore.kernel.org/all/cd67fbcd-d659-4822-bb90-7e8fbb40a856@molgen.mpg.de/
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Not-yet-signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Johannes Weiner <hannes@cmpxchg.org>
Link: https://lore.kernel.org/lkml/20241004123506.GR18071@noisy.programming.kicks-ass.net/ [1]
Commit 2e0199df25 ("sched/fair: Prepare exit/cleanup paths for delayed_dequeue")
and its follow up fixes try to deal with a rather unfortunate
situation where is task is enqueued in a new class, even though it
shouldn't have been. Mostly because the existing ->switched_to/from()
hooks are in the wrong place for this case.
This all led to Paul being able to trigger failures at something like
once per 10k CPU hours of RCU torture.
For now, do the ugly thing and move the code to the right place by
ignoring the switch hooks.
Note: Clean up the whole sched_class::switch*_{to,from}() thing.
Fixes: 2e0199df25 ("sched/fair: Prepare exit/cleanup paths for delayed_dequeue")
Reported-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20241003185037.GA5594@noisy.programming.kicks-ass.net
With KASAN and PREEMPT_RT enabled, calling task_work_add() in
task_tick_mm_cid() may cause the following splat.
[ 63.696416] BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:48
[ 63.696416] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 610, name: modprobe
[ 63.696416] preempt_count: 10001, expected: 0
[ 63.696416] RCU nest depth: 1, expected: 1
This problem is caused by the following call trace.
sched_tick() [ acquire rq->__lock ]
-> task_tick_mm_cid()
-> task_work_add()
-> __kasan_record_aux_stack()
-> kasan_save_stack()
-> stack_depot_save_flags()
-> alloc_pages_mpol_noprof()
-> __alloc_pages_noprof()
-> get_page_from_freelist()
-> rmqueue()
-> rmqueue_pcplist()
-> __rmqueue_pcplist()
-> rmqueue_bulk()
-> rt_spin_lock()
The rq lock is a raw_spinlock_t. We can't sleep while holding
it. IOW, we can't call alloc_pages() in stack_depot_save_flags().
The task_tick_mm_cid() function with its task_work_add() call was
introduced by commit 223baf9d17 ("sched: Fix performance regression
introduced by mm_cid") in v6.4 kernel.
Fortunately, there is a kasan_record_aux_stack_noalloc() variant that
calls stack_depot_save_flags() while not allowing it to allocate
new pages. To allow task_tick_mm_cid() to use task_work without
page allocation, a new TWAF_NO_ALLOC flag is added to enable calling
kasan_record_aux_stack_noalloc() instead of kasan_record_aux_stack()
if set. The task_tick_mm_cid() function is modified to add this new flag.
The possible downside is the missing stack trace in a KASAN report due
to new page allocation required when task_work_add_noallloc() is called
which should be rare.
Fixes: 223baf9d17 ("sched: Fix performance regression introduced by mm_cid")
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20241010014432.194742-1-longman@redhat.com
The deadline server code moved one of the start_hrtick_dl() calls
but dropped the dl specific hrtick_enabled check. This causes hrticks
to get armed even when sched_feat(HRTICK_DL) is false. Fix it.
Fixes: 63ba8422f8 ("sched/deadline: Introduce deadline servers")
Signed-off-by: Phil Auld <pauld@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://lore.kernel.org/r/20241004123729.460668-1-pauld@redhat.com
Iterating with scx_task_iter involves scx_tasks_lock and optionally the rq
lock of the task being iterated. Both locks can be released during iteration
and the iteration can be continued after re-grabbing scx_tasks_lock.
Currently, all lock handling is pushed to the caller which is a bit
cumbersome and makes it difficult to add lock-aware behaviors. Make the
scx_task_iter helpers handle scx_tasks_lock.
- scx_task_iter_init/scx_taks_iter_exit() now grabs and releases
scx_task_lock, respectively. Renamed to
scx_task_iter_start/scx_task_iter_stop() to more clearly indicate that
there are non-trivial side-effects.
- Add __ prefix to scx_task_iter_rq_unlock() to indicate that the function
is internal.
- Add scx_task_iter_unlock/relock(). The former drops both rq lock (if held)
and scx_tasks_lock and the latter re-locks only scx_tasks_lock.
This doesn't cause behavior changes and will be used to implement stall
avoidance.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Bypass mode was depending on ops.select_cpu() which can't be trusted as with
the rest of the BPF scheduler. Always enable and use scx_select_cpu_dfl() in
bypass mode.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Move the sanity check from the inner function scx_select_cpu_dfl() to the
exported kfunc scx_bpf_select_cpu_dfl(). This doesn't cause behavior
differences and will allow using scx_select_cpu_dfl() in bypass mode
regardless of scx_builtin_idle_enabled.
Signed-off-by: Tejun Heo <tj@kernel.org>
The disable path caps p->scx.slice to SCX_SLICE_DFL. As the field is already
being ignored at this stage during disable, the only effect this has is that
when the next BPF scheduler is loaded, it won't see unreasonable left-over
slices. Ultimately, this shouldn't matter but it's better to start in a
known state. Drop p->scx.slice capping from the disable path and instead
reset it to SCX_SLICE_DFL in the enable path.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
This reverts commit 6f34d8d382.
Slice length is ignored while bypassing and tasks are switched on every tick
and thus the patch does not make any difference. The perceived difference
was from test noise.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
pick_next_task_scx() was turned into pick_task_scx() since
commit 753e2836d1 ("sched_ext: Unify regular and core-sched pick
task paths"). Update the outdated message.
Signed-off-by: Honglei Wang <jameshongleiwang@126.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
- ops.enqueue() didn't have a way to tell whether select_task_rq_scx() and
thus ops.select() were skipped. Some schedulers were incorrectly using
SCX_ENQ_WAKEUP. Add SCX_ENQ_CPU_SELECTED and fix scx_qmap using it.
- Remove a spurious WARN_ON_ONCE() in scx_cgroup_exit().
- Fix error information clobbering during load.
- Add missing __weak markers to BPF helper declarations.
- Doc update.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCZwWKkA4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGelnAQDTA8GSIahTEHKM0c3yXE6K1/M56zo8Spp5OOA7
kXHR3AD/Y0RcXgaCvMI13aozmQWq756gyB6/qczN0+X3jx6wZwI=
=6xbe
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-6.12-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- ops.enqueue() didn't have a way to tell whether select_task_rq_scx()
and thus ops.select() were skipped. Some schedulers were incorrectly
using SCX_ENQ_WAKEUP. Add SCX_ENQ_CPU_SELECTED and fix scx_qmap using
it.
- Remove a spurious WARN_ON_ONCE() in scx_cgroup_exit()
- Fix error information clobbering during load
- Add missing __weak markers to BPF helper declarations
- Doc update
* tag 'sched_ext-for-6.12-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: Documentation: Update instructions for running example schedulers
sched_ext, scx_qmap: Add and use SCX_ENQ_CPU_SELECTED
sched/core: Add ENQUEUE_RQ_SELECTED to indicate whether ->select_task_rq() was called
sched/core: Make select_task_rq() take the pointer to wake_flags instead of value
sched_ext: scx_cgroup_exit() may be called without successful scx_cgroup_init()
sched_ext: Improve error reporting during loading
sched_ext: Add __weak markers to BPF helper function decalarations
scx_qmap and other schedulers in the SCX repo are using SCX_ENQ_WAKEUP to
tell whether ops.select_cpu() was called. This is incorrect as
ops.select_cpu() can be skipped in the wakeup path and leads to e.g.
incorrectly skipping direct dispatch for tasks that are bound to a single
CPU.
sched core has been updated to specify ENQUEUE_RQ_SELECTED if
->select_task_rq() was called. Map it to SCX_ENQ_CPU_SELECTED and update
scx_qmap to test it instead of SCX_ENQ_WAKEUP.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
Cc: Daniel Hodges <hodges.daniel.scott@gmail.com>
Cc: Changwoo Min <multics69@gmail.com>
Cc: Andrea Righi <andrea.righi@linux.dev>
Cc: Dan Schatzberg <schatzberg.dan@gmail.com>
During ttwu, ->select_task_rq() can be skipped if only one CPU is allowed or
migration is disabled. sched_ext schedulers may perform operations such as
direct dispatch from ->select_task_rq() path and it is useful for them to
know whether ->select_task_rq() was skipped in the ->enqueue_task() path.
Currently, sched_ext schedulers are using ENQUEUE_WAKEUP for this purpose
and end up assuming incorrectly that ->select_task_rq() was called for tasks
that are bound to a single CPU or migration disabled.
Make select_task_rq() indicate whether ->select_task_rq() was called by
setting WF_RQ_SELECTED in *wake_flags and make ttwu_do_activate() map that
to ENQUEUE_RQ_SELECTED for ->enqueue_task().
This will be used by sched_ext to fix ->select_task_rq() skip detection.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
This will be used to allow select_task_rq() to indicate whether
->select_task_rq() was called by modifying *wake_flags.
This makes try_to_wake_up() call all functions that take wake_flags with
WF_TTWU set. Previously, only select_task_rq() was. Using the same flags is
more consistent, and, as the flag is only tested by ->select_task_rq()
implementations, it doesn't cause any behavior differences.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David Vernet <void@manifault.com>
bit_wait_io_timeout has been unused since 2016's
commit 6290602709 ("mm: add PageWaiters indicating tasks are waiting for a page bit")
Remove it.
Signed-off-by: "Dr. David Alan Gilbert" <linux@treblig.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>
Link: https://lore.kernel.org/r/20241001234016.231696-1-linux@treblig.org
We do not have RESPECT_SLICE, we only have RUN_TO_PARITY.
Change RESPECT_SLICE to RUN_TO_PARITY, makes it more clear.
Signed-off-by: Huang Shijie <shijie@os.amperecomputing.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Christoph Lameter (Ampere) <cl@linux.com>
Link: https://lkml.kernel.org/r/20241001070456.10939-1-shijie@os.amperecomputing.com
When PLACE_LAG is enabled, from the relationship:
vl_i = (W + w_i)*vl'_i / W
we know that if vl'_i(se->vlag) is zero, the vl_i is zero too.
So if se->vlag is zero, there is no need to waste cycles to
do the calculation.
Signed-off-by: Huang Shijie <shijie@os.amperecomputing.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Christoph Lameter (Ampere) <cl@linux.com>
Link: https://lkml.kernel.org/r/20241001070021.10626-1-shijie@os.amperecomputing.com
The patch "5e963f2bd46 sched/fair: Commit to EEVDF"
removed the code following the DOUBLE_TICK:
-
- if (!sched_feat(EEVDF) && cfs_rq->nr_running > 1)
- check_preempt_tick(cfs_rq, curr);
The DOUBLE_TICK feature becomes dead code now, so remove it.
Signed-off-by: Huang Shijie <shijie@os.amperecomputing.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: "Christoph Lameter (Ampere)" <cl@linux.com>
Reviewed-by: Vishal Chourasia <vishalc@linux.ibm.com>
Link: https://lore.kernel.org/r/20241001065451.10356-1-shijie@os.amperecomputing.com
wake_up_var(), wait_var_event() and related interfaces are not
documented but have important ordering requirements. This patch adds
documentation and makes these requirements explicit.
The return values for those wait_var_event_* functions which return a
value are documented. Note that these are, perhaps surprisingly,
sometimes different from comparable wait_on_bit() functions.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20240925053405.3960701-4-neilb@suse.de
This patch revises the documention for wake_up_bit(),
clear_and_wake_up_bit(), and all the wait_on_bit() family of functions.
The new documentation places less emphasis on the pool of waitqueues
used (an implementation detail) and focuses instead on details of how
the functions behave.
The barriers included in the wait functions and clear_and_wake_up_bit()
and those required for wake_up_bit() are spelled out more clearly.
The error statuses returned are given explicitly.
The fact that the wait_on_bit_lock() function sets the bit is made more
obvious.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20240925053405.3960701-3-neilb@suse.de
wake_up_bit() currently allows a "void *". While this isn't strictly a
problem as the address is never dereferenced, it is inconsistent with
the corresponding wait_on_bit() which requires "unsigned long *" and
does dereference the pointer.
Any code that needs to wait for a change in something other than an
unsigned long would be better served by wake_up_var()/wait_var_event().
This patch changes all related "void *" to "unsigned long *".
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20240925053405.3960701-2-neilb@suse.de
568894edbe ("sched_ext: Add scx_cgroup_enabled to gate cgroup operations
and fix scx_tg_online()") assumed that scx_cgroup_exit() is only called
after scx_cgroup_init() finished successfully. This isn't true.
scx_cgroup_exit() can be called without scx_cgroup_init() being called at
all or after scx_cgroup_init() failed in the middle.
As init state is tracked per cgroup, scx_cgroup_exit() can be used safely to
clean up in all cases. Remove the incorrect WARN_ON_ONCE().
Signed-off-by: Tejun Heo <tj@kernel.org>
Fixes: 568894edbe ("sched_ext: Add scx_cgroup_enabled to gate cgroup operations and fix scx_tg_online()")
When the BPF scheduler fails, ops.exit() allows rich error reporting through
scx_exit_info. Use scx.exit() path consistently for all failures which can
be caused by the BPF scheduler:
- scx_ops_error() is called after ops.init() and ops.cgroup_init() failure
to record error information.
- ops.init_task() failure now uses scx_ops_error() instead of pr_err().
- The err_disable path updated to automatically trigger scx_ops_error() to
cover cases that the error message hasn't already been generated and
always return 0 indicating init success so that the error is reported
through ops.exit().
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: David Vernet <void@manifault.com>
Cc: Daniel Hodges <hodges.daniel.scott@gmail.com>
Cc: Changwoo Min <multics69@gmail.com>
Cc: Andrea Righi <andrea.righi@linux.dev>
Cc: Dan Schatzberg <schatzberg.dan@gmail.com>
Brandon reports sporadic, non-sensical spikes in cumulative pressure
time (total=) when reading cpu.pressure at a high rate. This is due to
a race condition between reader aggregation and tasks changing states.
While it affects all states and all resources captured by PSI, in
practice it most likely triggers with CPU pressure, since scheduling
events are so frequent compared to other resource events.
The race context is the live snooping of ongoing stalls during a
pressure read. The read aggregates per-cpu records for stalls that
have concluded, but will also incorporate ad-hoc the duration of any
active state that hasn't been recorded yet. This is important to get
timely measurements of ongoing stalls. Those ad-hoc samples are
calculated on-the-fly up to the current time on that CPU; since the
stall hasn't concluded, it's expected that this is the minimum amount
of stall time that will enter the per-cpu records once it does.
The problem is that the path that concludes the state uses a CPU clock
read that is not synchronized against aggregators; the clock is read
outside of the seqlock protection. This allows aggregators to race and
snoop a stall with a longer duration than will actually be recorded.
With the recorded stall time being less than the last snapshot
remembered by the aggregator, a subsequent sample will underflow and
observe a bogus delta value, resulting in an erratic jump in pressure.
Fix this by moving the clock read of the state change into the seqlock
protection. This ensures no aggregation can snoop live stalls past the
time that's recorded when the state concludes.
Reported-by: Brandon Duffany <brandon@buildbuddy.io>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=219194
Link: https://lore.kernel.org/lkml/20240827121851.GB438928@cmpxchg.org/
Fixes: df77430639 ("psi: Reduce calls to sched_clock() in psi")
Cc: stable@vger.kernel.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Chengming Zhou <chengming.zhou@linux.dev>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 85e511df3c ("sched/eevdf: Allow shorter slices to wakeup-preempt")
introduced a mechanism that a wakee with shorter slice could preempt
the current running task. It also lower the bar for the current task
to be preempted, by checking the rq->nr_running instead of cfs_rq->nr_running
when the current task has ran out of time slice. But there is a scenario
that is problematic. Say, if there is 1 cfs task and 1 rt task, before
85e511df3c, update_deadline() will not trigger a reschedule, and after
85e511df3c, since rq->nr_running is 2 and resched is true, a resched_curr()
would happen.
Some workloads (like the hackbench reported by lkp) do not like
over-scheduling. We can see that the preemption rate has been
increased by 2.2%:
1.654e+08 +2.2% 1.69e+08 hackbench.time.involuntary_context_switches
Restore its previous check criterion.
Fixes: 85e511df3c ("sched/eevdf: Allow shorter slices to wakeup-preempt")
Closes: https://lore.kernel.org/oe-lkp/202409231416.9403c2e9-oliver.sang@intel.com
Reported-by: kernel test robot <oliver.sang@intel.com>
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Honglei Wang <jameshongleiwang@126.com>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://lore.kernel.org/r/20240925085440.358138-1-yu.c.chen@intel.com
Meeting an unfinished DELAY_DEQUEUE treated entity in unthrottle_cfs_rq()
leads to a couple terminal scenarios. Finish it first, so ENQUEUE_WAKEUP
can proceed as it would have sans DELAY_DEQUEUE treatment.
Fixes: 152e11f6df ("sched/fair: Implement delayed dequeue")
Reported-by: Venkat Rao Bagalkote <venkat88@linux.vnet.ibm.com>
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Venkat Rao Bagalkote <venkat88@linux.vnet.ibm.com>
Link: https://lore.kernel.org/r/7515d2e64c989b9e3b828a9e21bcd959b99df06a.camel@gmx.de
The enable path uses three big locks - scx_fork_rwsem, scx_cgroup_rwsem and
cpus_read_lock. Currently, the locks are grabbed together which is prone to
locking order problems.
For example, currently, there is a possible deadlock involving
scx_fork_rwsem and cpus_read_lock. cpus_read_lock has to nest inside
scx_fork_rwsem due to locking order existing in other subsystems. However,
there exists a dependency in the other direction during hotplug if hotplug
needs to fork a new task, which happens in some cases. This leads to the
following deadlock:
scx_ops_enable() hotplug
percpu_down_write(&cpu_hotplug_lock)
percpu_down_write(&scx_fork_rwsem)
block on cpu_hotplug_lock
kthread_create() waits for kthreadd
kthreadd blocks on scx_fork_rwsem
Note that this doesn't trigger lockdep because the hotplug side dependency
bounces through kthreadd.
With the preceding scx_cgroup_enabled change, this can be solved by
decoupling cpus_read_lock, which is needed for static_key manipulations,
from the other two locks.
- Move the first block of static_key manipulations outside of scx_fork_rwsem
and scx_cgroup_rwsem. This is now safe with the preceding
scx_cgroup_enabled change.
- Drop scx_cgroup_rwsem and scx_fork_rwsem between the two task iteration
blocks so that __scx_ops_enabled static_key enabling is outside the two
rwsems.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-and-tested-by: Aboorva Devarajan <aboorvad@linux.ibm.com>
Link: http://lkml.kernel.org/r/8cd0ec0c4c7c1bc0119e61fbef0bee9d5e24022d.camel@linux.ibm.com
The disable path uses three big locks - scx_fork_rwsem, scx_cgroup_rwsem and
cpus_read_lock. Currently, the locks are grabbed together which is prone to
locking order problems. With the preceding scx_cgroup_enabled change, we can
decouple them:
- As cgroup disabling no longer requires modifying a static_key which
requires cpus_read_lock(), no need to grab cpus_read_lock() before
grabbing scx_cgroup_rwsem.
- cgroup can now be independently disabled before tasks are moved back to
the fair class.
Relocate scx_cgroup_exit() invocation before scx_fork_rwsem is grabbed, drop
now unnecessary cpus_read_lock() and move static_key operations out of
scx_fork_rwsem. This decouples all three locks in the disable path.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-and-tested-by: Aboorva Devarajan <aboorvad@linux.ibm.com>
Link: http://lkml.kernel.org/r/8cd0ec0c4c7c1bc0119e61fbef0bee9d5e24022d.camel@linux.ibm.com
If the BPF scheduler does not implement ops.cgroup_init(), scx_tg_online()
didn't set SCX_TG_INITED which meant that ops.cgroup_exit(), even if
implemented, won't be called from scx_tg_offline(). This is because
SCX_HAS_OP(cgroupt_init) is used to test both whether SCX cgroup operations
are enabled and ops.cgroup_init() exists.
Fix it by introducing a separate bool scx_cgroup_enabled to gate cgroup
operations and use SCX_HAS_OP(cgroup_init) only to test whether
ops.cgroup_init() exists. Make all cgroup operations consistently use
scx_cgroup_enabled to test whether cgroup operations are enabled.
scx_cgroup_enabled is added instead of using scx_enabled() to ease planned
locking updates.
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_ops_init_task() and the follow-up scx_ops_enable_task() in the fork path
were gated by scx_enabled() test and thus __scx_ops_enabled had to be turned
on before the first scx_ops_init_task() loop in scx_ops_enable(). However,
if an external entity causes sched_class switch before the loop is complete,
tasks which are not initialized could be switched to SCX.
The following can be reproduced by running a program which keeps toggling a
process between SCHED_OTHER and SCHED_EXT using sched_setscheduler(2).
sched_ext: Invalid task state transition 0 -> 3 for fish[1623]
WARNING: CPU: 1 PID: 1650 at kernel/sched/ext.c:3392 scx_ops_enable_task+0x1a1/0x200
...
Sched_ext: simple (enabling)
RIP: 0010:scx_ops_enable_task+0x1a1/0x200
...
switching_to_scx+0x13/0xa0
__sched_setscheduler+0x850/0xa50
do_sched_setscheduler+0x104/0x1c0
__x64_sys_sched_setscheduler+0x18/0x30
do_syscall_64+0x7b/0x140
entry_SYSCALL_64_after_hwframe+0x76/0x7e
Fix it by gating scx_ops_init_task() separately using
scx_ops_init_task_enabled. __scx_ops_enabled is now set after all tasks are
finished with scx_ops_init_task().
Signed-off-by: Tejun Heo <tj@kernel.org>
