| Age | Commit message (Collapse) | Author |
|---|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler updates from Ingo Molnar:
"Core & fair scheduler changes:
- Tweak wait_task_inactive() to force dequeue sched_delayed tasks
(John Stultz)
- Adhere to place_entity() constraints (Peter Zijlstra)
- Allow decaying util_est when util_avg > CPU capacity (Pierre
Gondois)
- Fix up wake_up_sync() vs DELAYED_DEQUEUE (Xuewen Yan)
Energy management:
- Introduce sched_update_asym_prefer_cpu() (K Prateek Nayak)
- cpufreq/amd-pstate: Update asym_prefer_cpu when core rankings
change (K Prateek Nayak)
- Align uclamp and util_est and call before freq update (Xuewen Yan)
CPU isolation:
- Make use of more than one housekeeping CPU (Phil Auld)
RT scheduler:
- Fix race in push_rt_task() (Harshit Agarwal)
- Add kernel cmdline option for rt_group_sched (Michal Koutný)
Scheduler topology support:
- Improve topology_span_sane speed (Steve Wahl)
Scheduler debugging:
- Move and extend the sched_process_exit() tracepoint (Andrii
Nakryiko)
- Add RT_GROUP WARN checks for non-root task_groups (Michal Koutný)
- Fix trace_sched_switch(.prev_state) (Peter Zijlstra)
- Untangle cond_resched() and live-patching (Peter Zijlstra)
Fixes and cleanups:
- Misc fixes and cleanups (K Prateek Nayak, Michal Koutný, Peter
Zijlstra, Xuewen Yan)"
* tag 'sched-core-2025-05-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (26 commits)
sched/uclamp: Align uclamp and util_est and call before freq update
sched/util_est: Simplify condition for util_est_{en,de}queue()
sched/fair: Fixup wake_up_sync() vs DELAYED_DEQUEUE
sched,livepatch: Untangle cond_resched() and live-patching
sched/core: Tweak wait_task_inactive() to force dequeue sched_delayed tasks
sched/fair: Adhere to place_entity() constraints
sched/debug: Print the local group's asym_prefer_cpu
cpufreq/amd-pstate: Update asym_prefer_cpu when core rankings change
sched/topology: Introduce sched_update_asym_prefer_cpu()
sched/fair: Use READ_ONCE() to read sg->asym_prefer_cpu
sched/isolation: Make use of more than one housekeeping cpu
sched/rt: Fix race in push_rt_task
sched: Add annotations to RT_GROUP_SCHED fields
sched: Add RT_GROUP WARN checks for non-root task_groups
sched: Do not construct nor expose RT_GROUP_SCHED structures if disabled
sched: Bypass bandwitdh checks with runtime disabled RT_GROUP_SCHED
sched: Skip non-root task_groups with disabled RT_GROUP_SCHED
sched: Add commadline option for RT_GROUP_SCHED toggling
sched: Always initialize rt_rq's task_group
sched: Remove unneeed macro wrap
...
|
|
Merge cpufreq updates for 6.16-rc1:
- Refactor cpufreq_online(), add and use cpufreq policy locking guards,
use __free() in policy reference counting, and clean up core cpufreq
code on top of that (Rafael Wysocki).
- Fix boost handling on CPU suspend/resume and sysfs updates (Viresh
Kumar).
- Fix des_perf clamping with max_perf in amd_pstate_update() (Dhananjay
Ugwekar).
- Add offline, online and suspend callbacks to the amd-pstate driver,
rename and use the existing amd_pstate_epp callbacks in it (Dhananjay
Ugwekar).
- Add support for the "Requested CPU Min frequency" BIOS option to the
amd-pstate driver (Dhananjay Ugwekar).
- Reset amd-pstate driver mode after running selftests (Swapnil
Sapkal).
- Add helper for governor checks to the schedutil cpufreq governor and
move cpufreq-specific EAS checks to cpufreq (Rafael Wysocki).
- Populate the cpu_capacity sysfs entries from the intel_pstate driver
after registering asym capacity support (Ricardo Neri).
- Add support for enabling Energy-aware scheduling (EAS) to the
intel_pstate driver when operating in the passive mode on a hybrid
platform (Rafael Wysocki).
- Avoid shadowing ret in amd_pstate_ut_check_driver() (Nathan
Chancellor).
- Drop redundant cpus_read_lock() from store_local_boost() in the
cpufreq core (Seyediman Seyedarab).
- Replace sscanf() with kstrtouint() in the cpufreq code and use a
symbol instead of a raw number in it (Bowen Yu).
- Add support for autonomous CPU performance state selection to the
CPPC cpufreq driver (Lifeng Zheng).
* pm-cpufreq: (31 commits)
cpufreq: CPPC: Add support for autonomous selection
cpufreq: Update sscanf() to kstrtouint()
cpufreq: Replace magic number
cpufreq: drop redundant cpus_read_lock() from store_local_boost()
cpufreq/amd-pstate: Avoid shadowing ret in amd_pstate_ut_check_driver()
cpufreq: intel_pstate: Document hybrid processor support
cpufreq: intel_pstate: EAS: Increase cost for CPUs using L3 cache
cpufreq: intel_pstate: EAS support for hybrid platforms
cpufreq: Drop policy locking from cpufreq_policy_is_good_for_eas()
cpufreq: intel_pstate: Populate the cpu_capacity sysfs entries
arch_topology: Relocate cpu_scale to topology.[h|c]
cpufreq/sched: Move cpufreq-specific EAS checks to cpufreq
cpufreq/sched: schedutil: Add helper for governor checks
amd-pstate-ut: Reset amd-pstate driver mode after running selftests
cpufreq/amd-pstate: Add support for the "Requested CPU Min frequency" BIOS option
cpufreq/amd-pstate: Add offline, online and suspend callbacks for amd_pstate_driver
cpufreq: Force sync policy boost with global boost on sysfs update
cpufreq: Preserve policy's boost state after resume
cpufreq: Introduce policy_set_boost()
cpufreq: Don't unnecessarily call set_boost()
...
|
|
BPF schedulers that use both builtin CPU idle mechanism and
ops.update_idle() may want to use the latter to create interlocking between
ops.enqueue() and CPU idle transitions so that either ops.enqueue() sees the
idle bit or ops.update_idle() sees the task queued somewhere. This can
prevent race conditions where CPUs go idle while tasks are waiting in DSQs.
For such interlocking to work, ops.update_idle() must be called after
builtin CPU masks are updated. Relocate the invocation. Currently, there are
no ordering requirements on transitions from idle and this relocation isn't
expected to make meaningful differences in that direction.
This also makes the ops.update_idle() behavior semantically consistent:
any action performed in this callback should be able to override the
builtin idle state, not the other way around.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-and-tested-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
There is no reason to restrict scx_bpf_select_cpu_dfl() invocations to
ops.select_cpu() while allowing scx_bpf_select_cpu_and() to be used from
multiple contexts, as both provide equivalent functionality, with the
latter simply accepting an additional "allowed" cpumask.
Therefore, unify the two APIs, enabling both kfuncs to be used from
ops.select_cpu(), ops.enqueue(), and unlocked contexts (e.g., via BPF
test_run).
This allows schedulers to implement a consistent idle CPU selection
policy and helps reduce code duplication.
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
The commit dfa0a574cbc47 ("sched/uclamg: Handle delayed dequeue")
has add the sched_delayed check to prevent double uclamp_dec/inc.
However, it put the uclamp_rq_inc() after enqueue_task().
This may lead to the following issues:
When a task with uclamp goes through enqueue_task() and could trigger
cpufreq update, its uclamp won't even be considered in the cpufreq
update. It is only after enqueue will the uclamp be added to rq
buckets, and cpufreq will only pick it up at the next update.
This could cause a delay in frequency updating. It may affect
the performance(uclamp_min > 0) or power(uclamp_max < 1024).
So, just like util_est, put the uclamp_rq_inc() before enqueue_task().
And as for the sched_delayed_task, same as util_est, using the
sched_delayed flag to prevent inc the sched_delayed_task's uclamp,
using the ENQUEUE_DELAYED flag to allow inc the sched_delayed_task's uclamp
which is being woken up.
Signed-off-by: Xuewen Yan <xuewen.yan@unisoc.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20250417043457.10632-3-xuewen.yan@unisoc.com
|
|
To prevent double enqueue/dequeue of the util-est for sched_delayed tasks,
commit 729288bc6856 ("kernel/sched: Fix util_est accounting for DELAY_DEQUEUE")
added the corresponding check. This check excludes double en/dequeue during
task migration and priority changes.
In fact, these conditions can be simplified.
For util_est_dequeue, we know that sched_delayed flag is set in dequeue_entity.
When the task is sleeping, we need to call util_est_dequeue to subtract
util-est from the cfs_rq. At this point, sched_delayed has not yet been set.
If we find that sched_delayed is already set, it indicates that this task
has already called dequeue_task_fair once. In this case, there is no need to
call util_est_dequeue again. Therefore, simply checking the sched_delayed flag
should be sufficient to prevent unnecessary util_est updates during the dequeue.
For util_est_enqueue, our goal is to add the util_est to the cfs_rq
when task enqueue. However, we don't want to add the util_est of a
sched_delayed task to the cfs_rq because the task is sleeping.
Therefore, we can exclude the util_est_enqueue for sched_delayed tasks
by checking the sched_delayed flag. However, when waking up a delayed task,
the sched_delayed flag is cleared after util_est_enqueue. As a result,
if we only check the sched_delayed flag, we would miss the util_est_enqueue.
Since waking up a sched_delayed task calls enqueue_task with the ENQUEUE_DELAYED flag,
we can determine whether to call util_est_enqueue by checking if the
enqueue_flag contains ENQUEUE_DELAYED.
Signed-off-by: Xuewen Yan <xuewen.yan@unisoc.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://lore.kernel.org/r/20250417043457.10632-2-xuewen.yan@unisoc.com
|
|
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: 152e11f6df29 ("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
|
|
Allow scx_bpf_select_cpu_and() to be used from an unlocked context, in
addition to ops.enqueue() or ops.select_cpu().
This enables schedulers, including user-space ones, to implement a
consistent idle CPU selection policy and helps reduce code duplication.
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
Validate locking correctness when accessing p->nr_cpus_allowed and
p->cpus_ptr inside scx_bpf_select_cpu_and(): if the rq lock is held,
access is safe; otherwise, require that p->pi_lock is held.
This allows to catch potential unsafe calls to scx_bpf_select_cpu_and().
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
Relocate the scx_kf_allowed_if_unlocked(), so it can be used from other
source files (e.g., ext_idle.c).
No functional change.
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
Naked scx_root dereferences are being used as temporary markers to indicate
that they need to be updated to point to the right scheduler instance.
Explain the situation.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Andrea Righi <arighi@nvidia.com>
|
|
In preparation of hierarchical scheduling support, add @sch to scx_exit()
and friends:
- scx_exit/error() updated to take explicit @sch instead of assuming
scx_root.
- scx_kf_exit/error() added. These are to be used from kfuncs, don't take
@sch and internally determine the scx_sched instance to abort. Currently,
it's always scx_root but once multiple scheduler support is in place, it
will be the scx_sched instance that invoked the kfunc. This simplifies
many callsites and defers scx_sched lookup until error is triggered.
- @sch is propagated to ops_cpu_valid() and ops_sanitize_err(). The CPU
validity conditions in ops_cpu_valid() are factored into __cpu_valid() to
implement kf_cpu_valid() which is the counterpart to scx_kf_exit/error().
- All users are converted. Most conversions are straightforward.
check_rq_for_timeouts() and scx_softlockup() are updated to use explicit
rcu_dereference*(scx_root) for safety as they may execute asynchronous to
the exit path. scx_tick() is also updated to use rcu_dereference(). While
not strictly necessary due to the preceding scx_enabled() test and IRQ
disabled context, this removes the subtlety at no noticeable cost.
No behavior changes intended.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
|
|
__scx_exit() is the base exit implementation and there are three wrappers on
top of it - scx_exit(), __scx_error() and scx_error(). This is more
confusing than helpful especially given that there are only a couple users
of scx_exit() and __scx_error(). To simplify the situation:
- Make __scx_exit() take va_list and rename it to scx_vexit(). This is to
ease implementing more complex extensions on top.
- Make scx_exit() a varargs wrapper around __scx_exit(). scx_exit() now
takes both @kind and @exit_code.
- Convert existing scx_exit() and __scx_error() users to use the new
scx_exit().
- scx_error() remains unchanged.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
|
|
In preparation of hierarchical scheduling support, make SCX_CALL_OP*() take
explicit @sch instead of assuming scx_root. As scx_root is still the only
scheduler instance, this patch doesn't make any functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
|
|
- Always cache scx_root into local variable sch before using.
- Don't use scx_root if cached sch is available.
- Wrap !sch test with unlikely().
- Pass @scx into scx_cgroup_init/exit().
No behavior changes intended.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
|
|
With the goal of deprecating / removing VOLUNTARY preempt, live-patch
needs to stop relying on cond_resched() to make forward progress.
Instead, rely on schedule() with TASK_FREEZABLE set. Just like
live-patching, the freezer needs to be able to stop tasks in a safe /
known state.
[bigeasy: use likely() in __klp_sched_try_switch() and update comments]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Petr Mladek <pmladek@suse.com>
Tested-by: Petr Mladek <pmladek@suse.com>
Tested-by: Miroslav Benes <mbenes@suse.cz>
Acked-by: Miroslav Benes <mbenes@suse.cz>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Link: https://lore.kernel.org/r/20250509113659.wkP_HJ5z@linutronix.de
|
|
|
|
cpuset memory pinning
Unlike sched_skip_vma_numa tracepoint which tracks skipped VMAs, this
tracks the task subjected to cpuset.mems pinning and prints out its
allowed memory node mask.
Link: https://lkml.kernel.org/r/20250424024523.2298272-3-libo.chen@oracle.com
Signed-off-by: Libo Chen <libo.chen@oracle.com>
Cc: "Chen, Tim C" <tim.c.chen@intel.com>
Cc: Chen Yu <yu.c.chen@intel.com>
Cc: Chris Hyser <chris.hyser@oracle.com>
Cc: Daniel Jordan <daniel.m.jordan@oracle.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Madadi Vineeth Reddy <vineethr@linux.ibm.com>
Cc: Mel Gorman <mgorman <mgorman@suse.de>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Raghavendra K T <raghavendra.kt@amd.com>
Cc: Srikanth Aithal <sraithal@amd.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "sched/numa: Skip VMA scanning on memory pinned to one NUMA
node via cpuset.mems", v5.
This patch (of 2):
When the memory of the current task is pinned to one NUMA node by cgroup,
there is no point in continuing the rest of VMA scanning and hinting page
faults as they will just be overhead. With this change, there will be no
more unnecessary PTE updates or page faults in this scenario.
We have seen up to a 6x improvement on a typical java workload running on
VMs with memory and CPU pinned to one NUMA node via cpuset in a two-socket
AARCH64 system. With the same pinning, on a 18-cores-per-socket Intel
platform, we have seen 20% improvment in a microbench that creates a
30-vCPU selftest KVM guest with 4GB memory, where each vCPU reads 4KB
pages in a fixed number of loops.
Link: https://lkml.kernel.org/r/20250424024523.2298272-1-libo.chen@oracle.com
Link: https://lkml.kernel.org/r/20250424024523.2298272-2-libo.chen@oracle.com
Signed-off-by: Libo Chen <libo.chen@oracle.com>
Tested-by: Chen Yu <yu.c.chen@intel.com>
Tested-by: Srikanth Aithal <sraithal@amd.com>
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Cc: "Chen, Tim C" <tim.c.chen@intel.com>
Cc: Chris Hyser <chris.hyser@oracle.com>
Cc: Daniel Jordan <daniel.m.jordan@oracle.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: K Prateek Nayak <kprateek.nayak@amd.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Madadi Vineeth Reddy <vineethr@linux.ibm.com>
Cc: Mel Gorman <mgorman <mgorman@suse.de>
Cc: Michal Koutný <mkoutny@suse.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Raghavendra K T <raghavendra.kt@amd.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
"A little bit invasive for rc6 but they're important fixes, pass tests
fine and won't break anything outside sched_ext:
- scx_bpf_cpuperf_set() calls internal functions that require the rq
to be locked. It assumed that the BPF caller has rq locked but
that's not always true. Fix it by tracking whether rq is currently
held by the CPU and grabbing it if necessary
- bpf_iter_scx_dsq_new() was leaving the DSQ iterator in an
uninitialized state after an error. However, next() and destroy()
can be called on an iterator which failed initialization and thus
they always need to be initialized even after an init error. Fix by
always initializing the iterator
- Remove duplicate BTF_ID_FLAGS() entries"
* tag 'sched_ext-for-6.15-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: bpf_iter_scx_dsq_new() should always initialize iterator
sched_ext: Fix rq lock state in hotplug ops
sched_ext: Remove duplicate BTF_ID_FLAGS definitions
sched_ext: Fix missing rq lock in scx_bpf_cpuperf_set()
sched_ext: Track currently locked rq
|
|
task_storage_{get,delete} has been moved to bpf_base_func_proto.
Suggested-by: Andrii Nakryiko <andrii.nakryiko@gmail.com>
Signed-off-by: Feng Yang <yangfeng@kylinos.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/bpf/20250506061434.94277-3-yangfeng59949@163.com
|
|
Doing cpufreq-specific EAS checks that require accessing policy
internals directly from sched_is_eas_possible() is a bit unfortunate,
so introduce cpufreq_ready_for_eas() in cpufreq, move those checks
into that new function and make sched_is_eas_possible() call it.
While at it, address a possible race between the EAS governor check
and governor change by doing the former under the policy rwsem.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/2317800.iZASKD2KPV@rjwysocki.net
|
|
Add a helper for checking if schedutil is the current governor for
a given cpufreq policy and use it in sched_is_eas_possible() to avoid
accessing cpufreq policy internals directly from there.
No intentional functional impact.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/3365956.44csPzL39Z@rjwysocki.net
|
|
To receive 428dc9fc0873 ("sched_ext: bpf_iter_scx_dsq_new() should always
initialize iterator") which conflicts with cdf5a6faa8cf ("sched_ext: Move
dsq_hash into scx_sched"). The conflict is a simple context conflict which
can be resolved by taking changes from both changes in the right order.
|
|
BPF programs may call next() and destroy() on BPF iterators even after new()
returns an error value (e.g. bpf_for_each() macro ignores error returns from
new()). bpf_iter_scx_dsq_new() could leave the iterator in an uninitialized
state after an error return causing bpf_iter_scx_dsq_next() to dereference
garbage data. Make bpf_iter_scx_dsq_new() always clear $kit->dsq so that
next() and destroy() become noops.
Signed-off-by: Tejun Heo <tj@kernel.org>
Fixes: 650ba21b131e ("sched_ext: Implement DSQ iterator")
Cc: stable@vger.kernel.org # v6.12+
Acked-by: Andrea Righi <arighi@nvidia.com>
|
|
A sched_ext scheduler may trigger __scx_exit() from a BPF timer
callback, where scx_root may not be safely dereferenced.
This can lead to a NULL pointer dereference as shown below (triggered by
scx_tickless):
BUG: kernel NULL pointer dereference, address: 0000000000000330
...
CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 6.14.0-virtme #1 PREEMPT(full)
RIP: 0010:__scx_exit+0x2b/0x190
...
Call Trace:
<IRQ>
scx_bpf_get_idle_smtmask+0x59/0x80
bpf_prog_8320d4217989178c_dispatch_all_cpus+0x35/0x1b6
...
bpf_prog_97f847d871513f95_sched_timerfn+0x4c/0x264
bpf_timer_cb+0x7a/0x140
__hrtimer_run_queues+0x1f9/0x3a0
hrtimer_run_softirq+0x8c/0xd0
handle_softirqs+0xd3/0x3d0
__irq_exit_rcu+0x9a/0xc0
irq_exit_rcu+0xe/0x20
Fix this by checking for a valid scx_root and adding proper RCU
protection.
Fixes: 48e1267773866 ("sched_ext: Introduce scx_sched")
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
Using a DSQ iterators from a timer callback can trigger the following
lockdep splat when accessing scx_root:
=============================
WARNING: suspicious RCU usage
6.14.0-virtme #1 Not tainted
-----------------------------
kernel/sched/ext.c:6907 suspicious rcu_dereference_check() usage!
other info that might help us debug this:
rcu_scheduler_active = 2, debug_locks = 1
no locks held by swapper/0/0.
stack backtrace:
CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 6.14.0-virtme #1 PREEMPT(full)
Sched_ext: tickless (enabled+all)
Call Trace:
<IRQ>
dump_stack_lvl+0x6f/0xb0
lockdep_rcu_suspicious.cold+0x4e/0xa3
bpf_iter_scx_dsq_new+0xb1/0xd0
bpf_prog_63f4fd1bccc101e7_dispatch_cpu+0x3e/0x156
bpf_prog_8320d4217989178c_dispatch_all_cpus+0x153/0x1b6
bpf_prog_97f847d871513f95_sched_timerfn+0x4c/0x264
? hrtimer_run_softirq+0x4f/0xd0
bpf_timer_cb+0x7a/0x140
__hrtimer_run_queues+0x1f9/0x3a0
hrtimer_run_softirq+0x8c/0xd0
handle_softirqs+0xd3/0x3d0
__irq_exit_rcu+0x9a/0xc0
irq_exit_rcu+0xe/0x20
sysvec_apic_timer_interrupt+0x73/0x80
Add a proper dereference check to explicitly validate RCU-safe access to
scx_root from rcu_read_lock() contexts and also from contexts that hold
rcu_read_lock_bh(), such as timer callbacks.
Fixes: cdf5a6faa8cf0 ("sched_ext: Move dsq_hash into scx_sched")
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
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: 152e11f6df29 ("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
|
|
With the global states and disable machinery moved into scx_sched,
scx_disable_workfn() can only be scheduled and run for the specific
scheduler instance. This makes it impossible for scx_disable_workfn() to see
SCX_EXIT_NONE. Turn that condition into WARN_ON_ONCE().
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
Because disable can be triggered from any place and the scheduler cannot be
trusted, disable path uses an irq_work to bounce and a kthread_work which is
executed on an RT helper kthread to perform disable. These must be per
scheduler instance to guarantee forward progress. Move them into scx_sched.
- If an scx_sched is accessible, its helper kthread is always valid making
the `helper` check in schedule_scx_disable_work() unnecessary. As the
function becomes trivial after the removal of the test, inline it.
- scx_create_rt_helper() has only one user - creation of the disable helper
kthread. Inline it into scx_alloc_and_add_sched().
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
The event counters are going to become per scheduler instance. Move
event_stats_cpu into scx_sched.
- [__]scx_add_event() are updated to take @sch. While at it, add missing
parentheses around @cnt expansions.
- scx_read_events() is updated to take @sch.
- scx_bpf_events() accesses scx_root under RCU read lock.
v2: - Replace stale scx_bpf_get_event_stat() reference in a comment with
scx_bpf_events().
- Trivial goto label rename.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
In prepration of moving event_stats_cpu into scx_sched, factor out
scx_read_events() out of scx_bpf_events() and update the in-kernel users -
scx_attr_events_show() and scx_dump_state() - to use scx_read_events()
instead of scx_bpf_events(). No observable behavior changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
In prepration of moving event_stats_cpu into scx_sched, move scx_event_stats
definitions above scx_sched definition. No functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
Global DSQs are going to become per scheduler instance. Move global_dsqs
into scx_sched. find_global_dsq() already takes a task_struct pointer as an
argument and should later be able to determine the scx_sched to use from
that. For now, assume scx_root.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
User DSQs are going to become per scheduler instance. Move dsq_hash into
scx_sched. This shifts the code that assumes scx_root to be the only
scx_sched instance up the call stack but doesn't remove them yet.
v2: Add missing rcu_read_lock() in scx_bpf_destroy_dsq() as per Andrea.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
More will be moved into scx_sched. Factor out the allocation and kobject
addition path into scx_alloc_and_add_sched().
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
create_dsq() is only used by scx_bpf_create_dsq() and the separation gets in
the way of making dsq_hash per scx_sched. Inline it into
scx_bpf_create_dsq(). While at it, add unlikely() around
SCX_DSQ_FLAG_BUILTIN test.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
To prepare for supporting multiple schedulers, make scx_sched allocated
dynamically. scx_sched->kobj is now an embedded field and the kobj's
lifetime determines the lifetime of the containing scx_sched.
- Enable path is updated so that kobj init and addition are performed later.
- scx_sched freeing is initiated in scx_kobj_release() and also goes through
an rcu_work so that scx_root can be accessed from an unsynchronized path -
scx_disable().
- sched_ext_ops->priv is added and used to point to scx_sched instance
created for the ops instance. This is used by bpf_scx_unreg() to determine
the scx_sched instance to disable and put.
No behavior changes intended.
v2: Andrea reported kernel oops due to scx_bpf_unreg() trying to deref NULL
scx_root after scheduler init failure. sched_ext_ops->priv added so that
scx_bpf_unreg() can always find the scx_sched instance to unregister
even if it failed early during init.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
SCX_HAS_OP() tests scx_root->has_op bitmap. The bitmap is currently in a
statically allocated struct scx_sched and initialized while loading the BPF
scheduler and cleared while unloading, and thus can be tested anytime.
However, scx_root will be switched to dynamic allocation and thus won't
always be deferenceable.
Most usages of SCX_HAS_OP() are already protected by scx_enabled() either
directly or indirectly (e.g. through a task which is on SCX). However, there
are a couple places that could try to dereference NULL scx_root. Update them
so that scx_root is guaranteed to be valid before SCX_HAS_OP() is called.
- In handle_hotplug(), test whether scx_root is NULL before doing anything
else. This is safe because scx_root updates will be protected by
cpus_read_lock().
- In scx_tg_offline(), test scx_cgroup_enabled before invoking SCX_HAS_OP(),
which should guarnatee that scx_root won't turn NULL. This is also in line
with other cgroup operations. As the code path is synchronized against
scx_cgroup_init/exit() through scx_cgroup_rwsem, this shouldn't cause any
behavior differences.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
To support multiple scheduler instances, collect some of the global
variables that should be specific to a scheduler instance into the new
struct scx_sched. scx_root is the root scheduler instance and points to a
static instance of struct scx_sched. Except for an extra dereference through
the scx_root pointer, this patch makes no functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
|
|
To receive e38be1c7647c ("sched_ext: Fix rq lock state in hotplug ops") to
avoid conflicts with scx_sched related patches pending for for-6.16.
|
|
The ops.cpu_online() and ops.cpu_offline() callbacks incorrectly assume
that the rq involved in the operation is locked, which is not the case
during hotplug, triggering the following warning:
WARNING: CPU: 1 PID: 20 at kernel/sched/sched.h:1504 handle_hotplug+0x280/0x340
Fix by not tracking the target rq as locked in the context of
ops.cpu_online() and ops.cpu_offline().
Fixes: 18853ba782bef ("sched_ext: Track currently locked rq")
Reported-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Tested-by: Changwoo Min <changwoo@igalia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fix from Ingo Molnar:
"Fix sporadic crashes in dequeue_entities() due to ... bad math.
[ Arguably if pick_eevdf()/pick_next_entity() was less trusting of
complex math being correct it could have de-escalated a crash into
a warning, but that's for a different patch ]"
* tag 'sched-urgent-2025-04-26' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/eevdf: Fix se->slice being set to U64_MAX and resulting crash
|
|
There is a code path in dequeue_entities() that can set the slice of a
sched_entity to U64_MAX, which sometimes results in a crash.
The offending case is when dequeue_entities() is called to dequeue a
delayed group entity, and then the entity's parent's dequeue is delayed.
In that case:
1. In the if (entity_is_task(se)) else block at the beginning of
dequeue_entities(), slice is set to
cfs_rq_min_slice(group_cfs_rq(se)). If the entity was delayed, then
it has no queued tasks, so cfs_rq_min_slice() returns U64_MAX.
2. The first for_each_sched_entity() loop dequeues the entity.
3. If the entity was its parent's only child, then the next iteration
tries to dequeue the parent.
4. If the parent's dequeue needs to be delayed, then it breaks from the
first for_each_sched_entity() loop _without updating slice_.
5. The second for_each_sched_entity() loop sets the parent's ->slice to
the saved slice, which is still U64_MAX.
This throws off subsequent calculations with potentially catastrophic
results. A manifestation we saw in production was:
6. In update_entity_lag(), se->slice is used to calculate limit, which
ends up as a huge negative number.
7. limit is used in se->vlag = clamp(vlag, -limit, limit). Because limit
is negative, vlag > limit, so se->vlag is set to the same huge
negative number.
8. In place_entity(), se->vlag is scaled, which overflows and results in
another huge (positive or negative) number.
9. The adjusted lag is subtracted from se->vruntime, which increases or
decreases se->vruntime by a huge number.
10. pick_eevdf() calls entity_eligible()/vruntime_eligible(), which
incorrectly returns false because the vruntime is so far from the
other vruntimes on the queue, causing the
(vruntime - cfs_rq->min_vruntime) * load calulation to overflow.
11. Nothing appears to be eligible, so pick_eevdf() returns NULL.
12. pick_next_entity() tries to dereference the return value of
pick_eevdf() and crashes.
Dumping the cfs_rq states from the core dumps with drgn showed tell-tale
huge vruntime ranges and bogus vlag values, and I also traced se->slice
being set to U64_MAX on live systems (which was usually "benign" since
the rest of the runqueue needed to be in a particular state to crash).
Fix it in dequeue_entities() by always setting slice from the first
non-empty cfs_rq.
Fixes: aef6987d8954 ("sched/eevdf: Propagate min_slice up the cgroup hierarchy")
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/f0c2d1072be229e1bdddc73c0703919a8b00c652.1745570998.git.osandov@fb.com
|
|
Some kfuncs specific to the idle CPU selection policy are registered in
both the scx_kfunc_ids_any and scx_kfunc_ids_idle blocks, even though
they should only be defined in the latter.
Remove the duplicates from scx_kfunc_ids_any.
Fixes: 337d1b354a297 ("sched_ext: Move built-in idle CPU selection policy to a separate file")
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
The ops.running() and ops.stopping() callbacks can be invoked from a CPU
other than the one the task is assigned to, particularly when a task
property is changed, as both scx_next_task_scx() and dequeue_task_scx() may
run on CPUs different from the task's target CPU.
This behavior can lead to confusion or incorrect assumptions if not
properly clarified, potentially resulting in bugs (see [1]).
Therefore, update the documentation to clarify this aspect and advise
users to use scx_bpf_task_cpu() to determine the actual CPU the task
will run on or was running on.
[1] https://github.com/sched-ext/scx/pull/1728
Cc: Jake Hillion <jake@hillion.co.uk>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
a11d6784d731 ("sched_ext: Fix missing rq lock in scx_bpf_cpuperf_set()")
added a call to scx_ops_error() which was renamed to scx_error() in
for-6.16. Fix it up.
|
|
scx_bpf_cpuperf_set() can be used to set a performance target level on
any CPU. However, it doesn't correctly acquire the corresponding rq
lock, which may lead to unsafe behavior and trigger the following
warning, due to the lockdep_assert_rq_held() check:
[ 51.713737] WARNING: CPU: 3 PID: 3899 at kernel/sched/sched.h:1512 scx_bpf_cpuperf_set+0x1a0/0x1e0
...
[ 51.713836] Call trace:
[ 51.713837] scx_bpf_cpuperf_set+0x1a0/0x1e0 (P)
[ 51.713839] bpf_prog_62d35beb9301601f_bpfland_init+0x168/0x440
[ 51.713841] bpf__sched_ext_ops_init+0x54/0x8c
[ 51.713843] scx_ops_enable.constprop.0+0x2c0/0x10f0
[ 51.713845] bpf_scx_reg+0x18/0x30
[ 51.713847] bpf_struct_ops_link_create+0x154/0x1b0
[ 51.713849] __sys_bpf+0x1934/0x22a0
Fix by properly acquiring the rq lock when possible or raising an error
if we try to operate on a CPU that is not the one currently locked.
Fixes: d86adb4fc0655 ("sched_ext: Add cpuperf support")
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
Some kfuncs provided by sched_ext may need to operate on a struct rq,
but they can be invoked from various contexts, specifically, different
scx callbacks.
While some of these callbacks are invoked with a particular rq already
locked, others are not. This makes it impossible for a kfunc to reliably
determine whether it's safe to access a given rq, triggering potential
bugs or unsafe behaviors, see for example [1].
To address this, track the currently locked rq whenever a sched_ext
callback is invoked via SCX_CALL_OP*().
This allows kfuncs that need to operate on an arbitrary rq to retrieve
the currently locked one and apply the appropriate action as needed.
[1] https://lore.kernel.org/lkml/20250325140021.73570-1-arighi@nvidia.com/
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Acked-by: Changwoo Min <changwoo@igalia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- Use kvzalloc() so that large exit_dump buffer allocations don't fail
easily
- Remove cpu.weight / cpu.idle unimplemented warnings which are more
annoying than helpful.
This makes SCX_OPS_HAS_CGROUP_WEIGHT unnecessary. Mark it for
deprecation
* tag 'sched_ext-for-6.15-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: Mark SCX_OPS_HAS_CGROUP_WEIGHT for deprecation
sched_ext: Remove cpu.weight / cpu.idle unimplemented warnings
sched_ext: Use kvzalloc for large exit_dump allocation
|
