The first parameter to kcpustat_field() is a pointer to the cpu kcpustat to
be fetched from. This parameter is error prone because a copy to a kcpustat
could be passed by accident instead of the original one. Also the kcpustat
structure can already be retrieved with the help of the mandatory CPU
argument.
Remove the needless parameter.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Tested-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://patch.msgid.link/20260508131647.43868-4-frederic@kernel.org
Offline handling happens from within the inner idle loop, after the
beginning of dyntick cputime accounting, nohz idle load balancing and
TIF_NEED_RESCHED polling.
This is not necessary and even buggy because:
* There is no dyntick handling to do. And calling tick_nohz_idle_enter()
messes up with the struct tick_sched reset that was performed on
tick_sched_timer_dying().
* There is no nohz idle balancing to do.
* Polling on TIF_RESCHED is irrelevant at this stage, there are no more
tasks allowed to run.
* No need to check if need_resched() before offline handling since
stop_machine is done and all per-cpu kthread should be done with
their job.
Therefore move the offline handling at the beginning of the idle loop.
This will also ease the idle cputime unification later by not elapsing
idle time while offline through the call to:
tick_nohz_idle_enter() -> tick_nohz_start_idle()
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Tested-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Reviewed-by: Rafael J. Wysocki (Intel) <rafael@kernel.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Link: https://patch.msgid.link/20260508131647.43868-3-frederic@kernel.org
assign_cfs_rq_runtime() during update_curr() sets the resched indicator
and relies on check_cfs_rq_runtime() during pick_next_task() /
put_prev_entity() to throttle the hierarchy once current task is
preempted / blocks.
Per-task throttle, on the other hand, uses throttle_cfs_rq() to simply
propagate the throttle signals, and then relies on task work to
individually throttle the runnable tasks on their way out to the
userspace.
Remove check_cfs_rq_runtime() and unify throttling into
account_cfs_rq_runtime() which only sets the cfs_rq->throttled,
cfs_rq->throttle_count indicators via throttle_cfs_rq() and optionally
adds the task work to the current task (donor) it is on the throttled
hierarchy.
throttle_cfs_rq() requests for sched_cfs_bandwidth_slice() worth of
bandwidth for the current hierarchy that enable it to continue running
uninterrupted when selected. For the rest, it requests a bare minimum of
"1" to ensure some bandwidth is available and pass the
"runtime_remaining > 0" checks once selected.
For SCHED_PROXY_EXEC, a mutex holder cannot exit to userspace without
dropping it first and the mutex_unlock() ensures proxy is stopped before
the mutex handoff which preserves the current semantics for running a
throttled task until it exits to the userspace even if it acts as a
donor.
[ prateek: rebased on tip, comments, commit message. ]
Reviewed-By: Benjamin Segall <bsegall@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Aaron Lu <ziqianlu@bytedance.com>
Link: https://patch.msgid.link/20260602071005.11942-1-kprateek.nayak@amd.com
An update_curr() during the enqueue of throttled task will start
throttling the hierarchy from subsequent commit. This can lead to
tg_throttle_down() seeing non-empty throttled_limbo_list for the cfs_rq
attaching the task from throttled_limbo_list one by one. For example:
R
|
A
/ \
*B C
|
rq->curr
*B is throttled with tasks on hte limbo list. When the tasks are
unthrottled via tg_unthrottle_up() and entity of group B is placed onto
A, update_curr() is called to catch up the vruntime and it may throttle
group A causing the subsequent tg_throttle_down() to see the pending
task's on B's limbo list.
tg_unthrottle_up()
/* --cfs_rq->throttle_count == 0 */
list_for_each_entry_safe(p, cfs_rq->throttled_limbo_list)
enqueue_task_fair()
enqueue_entity(se /* B->se */)
update_curr(cfs_rq /* A->gcfs_rq */)
account_cfs_rq_runtime(cfs_rq)
throttle_cfs_rq(cfs_rq /* A->gcfs_rq */ )
tg_throttle_down()
/* Reaches B->cfs_rq with throttle_count == 0 */
!!! !list_empty(&cfs_rq->throttled_limbo_list)) !!!
Move the tasks from throttled_limbo_list onto a local list before
starting the unthrottle to prevent the splat described above. If the
hierarchy is throttled again in middle of an unthrottle, put the pending
tasks back onto the limbo list to prevent running them unnecessarily.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Benjamin Segall <bsegall@google.com>
Tested-by: Aaron Lu <ziqianlu@bytedance.com>
Link: https://patch.msgid.link/20260602052531.11450-2-kprateek.nayak@amd.com
Subsequent commits will allow update_curr() to throttle the hierarchy
when the runtime accounting exceeds allocated quota. Call update_curr()
before the unthrottle event, and in tg_unthrottle_up() to catch up on
any remaining runtime and stabilize the "runtime_remaining" and
"throttle_count" for that cfs_rq.
Doing an update_curr() early ensures the cfs_rq is not throttled right
back up again when the unthrottle is in progress.
Since all callers of unthrottle_cfs_rq(), except two, already update the
rq_clock and call rq_clock_start_loop_update(), move the
update_rq_clock() from unthrottle_cfs_rq() to the callers that don't
update the rq_clock.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Benjamin Segall <bsegall@google.com>
Tested-by: Aaron Lu <ziqianlu@bytedance.com>
Link: https://patch.msgid.link/20260602052531.11450-1-kprateek.nayak@amd.com
When distribute_cfs_runtime() encounters a local cfs_rq, it adds it to a
local list and unthrottles it at the end, when it is done unthrottling
other cfs_rq(s) on cfs_b->throttled_cfs_rq until the bandwidth runs out.
Instead of using a local list, reuse the local CPU's
rq->throttled_csd_list and the __cfsb_csd_unthrottle() path for
unthrottle.
If this is the first cfs_rq to be queued on the "throttled_csd_list", it
prevents the need for a remote CPUs to interrupt this local CPU if they
themselves are performing async unthrottle.
If this is not the first cfs_rq on the list, there is an async unthrottle
operation pending on this local CPU and the unthrottle can be batched
together.
No functional changes intended.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Benjamin Segall <bsegall@google.com>
Tested-by: Aaron Lu <ziqianlu@bytedance.com>
Link: https://patch.msgid.link/20260602050005.11160-3-kprateek.nayak@amd.com
Routine conversion of rcu_read_lock(), spin_lock*, and rq_lock usage
within the cfs bandwidth controller to use class guards.
Only notable changes are:
- Checking for "cfs_rq->runtime_remaining <= 0" instead of the inverse
to spot a throttle and break early. This also saves the need
for extra indentation in the unthrottle case.
- Reordering of list_del_rcu() against throttled_clock indicator update
in unthrottle_cfs_rq(). Both are done with "cfs_b->lock" held after
the "cfs_rq->throttled" is cleared which make the reordering safe
against concurrent list modifications.
No functional changes intended.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Tested-by: Aaron Lu <ziqianlu@bytedance.com>
Link: https://patch.msgid.link/20260602050005.11160-2-kprateek.nayak@amd.com
To remove the cfs_rq pointer array in task_group, allocate the combined
cfs_rq and sched_entity using the per-cpu allocator.
This patch implements the following:
- Changes task_group->cfs_rq from 'struct cfs_rq **' to
'struct cfs_rq __percpu *'.
- Updates memory allocation in alloc_fair_sched_group() and
free_fair_sched_group() to use alloc_percpu() and free_percpu()
respectively.
- Uses the inline accessor tg_cfs_rq(tg, cpu) with per_cpu_ptr() to retrieve
the pointer to cfs_rq for the given task group and CPU.
- Replaces direct accesses tg->cfs_rq[cpu] with calls to the new tg_cfs_rq(tg,
cpu) helper.
- Handles the root_task_group: since struct rq is already a per-cpu variable
(runqueues), its embedded cfs_rq (rq->cfs) is also per-cpu. Therefore, we
assign root_task_group.cfs_rq = &runqueues.cfs.
- Cleanup the code in initializing the root task group.
This change places each CPU's cfs_rq and sched_entity in its local per-cpu
memory area to remove the per-task_group pointer arrays.
Signed-off-by: Zecheng Li <zecheng@google.com>
Signed-off-by: Zecheng Li <zli94@ncsu.edu>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Reviewed-by: Josh Don <joshdon@google.com>
Link: https://patch.msgid.link/20260522141623.600235-4-zli94@ncsu.edu
Now that struct sched_entity is co-located with struct cfs_rq for non-root task
groups, the task_group->se pointer array is redundant. The associated
sched_entity can be loaded directly from the cfs_rq.
This patch performs the access conversion with the helpers:
- is_root_task_group(tg): checks if a task group is the root task group. It
compares the task group's address with the global root_task_group variable.
- tg_se(tg, cpu): retrieves the cfs_rq and returns the address of the
co-located se. This function checks if tg is the root task group to ensure
behaving the same of previous tg->se[cpu]. Replaces all accesses that use
the tg->se[cpu] pointer array with calls to the new tg_se(tg, cpu) accessor.
- cfs_rq_se(cfs_rq): simplifies access paths like cfs_rq->tg->se[...] to use
the co-located sched_entity. This function also checks if tg is the root
task group to ensure same behavior.
Since tg_se is not in very hot code paths, and the branch is a register
comparison with an immediate value (`&root_task_group`), the performance impact
is expected to be negligible.
Signed-off-by: Zecheng Li <zecheng@google.com>
Signed-off-by: Zecheng Li <zli94@ncsu.edu>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Reviewed-by: Josh Don <joshdon@google.com>
Link: https://patch.msgid.link/20260522141623.600235-3-zli94@ncsu.edu
Improve data locality and reduce pointer chasing by allocating struct
cfs_rq and struct sched_entity together for non-root task groups. This
is achieved by introducing a new combined struct cfs_tg_state that
holds both objects in a single allocation.
This patch:
- Introduces struct cfs_tg_state that embeds cfs_rq, sched_entity, and
sched_statistics together in a single structure.
- Updates __schedstats_from_se() in stats.h to use cfs_tg_state for accessing
sched_statistics from a group sched_entity.
- Modifies alloc_fair_sched_group() and free_fair_sched_group() to allocate
and free the new struct as a single unit.
- Modifies the per-CPU pointers in task_group->se and task_group->cfs_rq to
point to the members in the new combined structure.
Signed-off-by: Zecheng Li <zecheng@google.com>
Signed-off-by: Zecheng Li <zli94@ncsu.edu>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Reviewed-by: Josh Don <joshdon@google.com>
Link: https://patch.msgid.link/20260522141623.600235-2-zli94@ncsu.edu
Commit ed4fb6d7ef ("hrtimer: Use and report correct timerslack values
for realtime tasks") sets timer_slack_ns to 0 for RT tasks in
__setscheduler_params(). However, when an RT task with SCHED_RESET_ON_FORK
creates child threads, the children inherit timer_slack_ns=0 from the
parent. sched_fork() resets the child's policy to SCHED_NORMAL but does
not restore timer_slack_ns, leaving the child permanently running with
zero slack.
Fix this by restoring timer_slack_ns from default_timer_slack_ns in
sched_fork() when resetting from RT/DL to NORMAL policy, matching the
existing behavior in __setscheduler_params().
Note: this fix alone requires a correct default_timer_slack_ns to be
effective. See the following patch for that fix.
Fixes: ed4fb6d7ef ("hrtimer: Use and report correct timerslack values for realtime tasks")
Reported-by: Qiaoting.Lin <linqiaoting@xiaomi.com>
Signed-off-by: Guanyou.Chen <chenguanyou@xiaomi.com>
Signed-off-by: Chunhui.Li <chunhui.li@mediatek.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260522131000.1664983-2-chenguanyou@xiaomi.com
Now that the proxy path uses ->is_blocked, use the '->is_blocked &&
!->blocked_on' state instead of PROXY_WAKING. Notably, this is where a
blocked_on relation is broken but the donor task might still need a return
migration.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260526113322.596522894%40infradead.org
Rather than gate the proxy paths with p->blocked_on, use p->is_blocked.
This opens up the state: '->is_blocked && !->blocked_on' for future use.
Notably, only proxy and delayed tasks can be ->on_rq && ->is_blocked, and it is
guaranteed that sched_class::pick_task() will never return a delayed task.
Therefore any task returned from pick_next_task() that has ->is_blocked set,
must be a proxy task.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260526113322.477954312%40infradead.org
Current code will 'unconditionally' return migrate on PROXY_WAKING, even if the
task is (still) on the original CPU.
Check task_cpu(p) against p->waking_cpu, which per proxy_set_task_cpu()
preserves the original CPU the task was on. If they do not mis-match, there is
no need to go through the more expensive wakeup path.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260527082916.GP3126523%40noisy.programming.kicks-ass.net
Upon entry to try_to_block_task(), p->is_blocked should be false. After all,
the prior wakeup would have made it so per ttwu_do_wakeup().
Ensure this is the case, rather than clearing it in the path that doesn't set
it.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: John Stultz <jstultz@google.com>
Link: https://patch.msgid.link/20260526113322.364017314%40infradead.org
The reason for the clause in try_to_wake_up() is, per its comment, that
find_proxy_task()'s proxy_deactivate() is not always called with a cleared
p->blocked_on.
However, that seems silly and easily cured. Make sure to always call
proxy_deactivate() with a cleared p->blocked_on such that we might remove this
clause from the common wake-up path.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: John Stultz <jstultz@google.com>
Link: https://patch.msgid.link/20260526113322.244729903%40infradead.org
Add link to the task this task is proxying for, and use it so
the mutex owner can do an intelligent hand-off of the mutex to
the task that the owner is running on behalf.
[jstultz: This patch was split out from larger proxy patch]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Juri Lelli <juri.lelli@redhat.com>
Signed-off-by: Valentin Schneider <valentin.schneider@arm.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>
Link: https://patch.msgid.link/20260512025635.2840817-8-jstultz@google.com
Add a new is_blocked flag to the task struct. This flag is set
by try_to_block_task() and cleared by ttwu_do_wakeup() and
tracks if the task is blocked.
Traditionally this would mirror !p->on_rq, however due things
like DELAY_DEQUEUE and PROXY_EXEC, this can diverge, so its
useful to manage separately.
Additionally with this, we might be able to get rid of the
p->se.sched_delayed (ab)use in the core code (eventually).
Taken whole cloth from Peter's email:
https://lore.kernel.org/lkml/20260501132143.GC1026330@noisy.programming.kicks-ass.net/
With a few additional p->is_blocked = 0 in a few cases where
we return current if blocked_on gets zeroed or there is
no owner. This may hint that these current special cases
might be dropped eventually.
This change also helps resolve wait-queue stalls seen with
proxy-execution. See previous patch attempts for details:
https://lore.kernel.org/lkml/20260430215103.2978955-2-jstultz@google.com/
Reported-by: Vineeth Pillai <vineethrp@google.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260512025635.2840817-7-jstultz@google.com
This patch adds logic so try_to_wake_up() will notice if we are
waking a task where blocked_on == PROXY_WAKING, and if necessary
dequeue the task so the wakeup will naturally return-migrate the
donor task back to a cpu it can run on.
This helps performance as we do the dequeue and wakeup under the
locks normally taken in the try_to_wake_up() and avoids having
to do proxy_force_return() from __schedule(), which has to
re-take similar locks and then force a pick again loop.
This was split out from the larger proxy patch, and
significantly reworked.
Credits for the original patch go to:
Peter Zijlstra (Intel) <peterz@infradead.org>
Juri Lelli <juri.lelli@redhat.com>
Valentin Schneider <valentin.schneider@arm.com>
Connor O'Brien <connoro@google.com>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260512025635.2840817-6-jstultz@google.com
Pull most of the logic out of try_to_block_task() and put it
into block_task() directly, so that we can call block_task() and
not have to worry about the failing cases in try_to_block_task()
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260512025635.2840817-5-jstultz@google.com
The DL scheduler keeps the current task in the rbtree, since
the deadline value isn't usually chagned while the task is
runnable.
This results in set_next_task() and put_prev_task() being
simpler, but unfortunately this causes complexity elsewhere.
Specifically when update_curr_dl() updates the deadline, it has
to dequeue and then enqueue the task.
From put_prev_task_dl(), we first call update_curr_dl(), and
then call enqueue_pushable_dl_task().
However, with Proxy Exec this goes awry. Since when a mutex is
released, we might wake the waiting rq->donor. This will cause
put_prev_task() to be called on the donor to take it off the
cpu for return migration.
At that point, from put_prev_task_dl() the update_curr_dl()
logic will dequeue & enqueue the task, and the enqueue function
will call enqueue_pushable_dl_task() (since the task_current()
check won't prevent it). Then back up the callstack in
put_prev_task_dl() we'll end up calling
enqueue_pushable_dl_task() again, tripping the
!RB_EMPTY_NODE(&p->pushable_dl_tasks) warning.
So to avoid this, use Peter's suggested[1] approach, and add a
dl_rq->curr pointer that is set/cleared from set_next_task()/
put_prev_task(), which effectively tracks the rq->donor. We can
then use this to avoid adding the active donor to the pushable
list from enqueue_task_dl().
[1]: https://lore.kernel.org/lkml/20260304095123.GP606826@noisy.programming.kicks-ass.net/
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260512025635.2840817-4-jstultz@google.com
As part of an improvement to handling pushable deadline tasks,
Peter suggested this cleanup[1], to use helper values for
dl_entity and dl_rq in the enqueue_task_dl() and
put_prev_task_dl() functions. There should be no functional
change from this patch.
To make sure this cleanup change doesn't obscure later logic
changes, I've split it into its own patch.
[1]: https://lore.kernel.org/lkml/20260304095123.GP606826@noisy.programming.kicks-ass.net/
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260512025635.2840817-3-jstultz@google.com
Historically, the prev value from __schedule() was the rq->curr.
This prev value is passed down through numerous functions, and
used in the class scheduler implementations. The fact that
prev was on_cpu until the end of __schedule(), meant it was
stable across the rq lock drops that the class->balance()
implementations often do.
However, with proxy-exec, the prev passed to functions called
by __schedule() is rq->donor, which may not be the same as
rq->curr and may not be on_cpu, this makes the prev value
potentially unstable across rq lock drops.
A recently found issue with proxy-exec, is when we begin doing
return migration from try_to_wake_up(), its possible we may be
waking up the rq->donor. When we do this, we proxy_resched_idle()
to put_prev_set_next() setting the rq->donor to rq->idle, allowing
the rq->donor to be return migrated and allowed to run.
This however runs into trouble, as on another cpu we might be in
the middle of calling __schedule(). Conceptually the rq lock is
held for the majority of the time, but in calling prev_balance()
its possible the class->balance() handler call may briefly drop the rq lock.
This opens a window for try_to_wake_up() to wake and return migrate the
rq->donor before the class logic reacquires the rq lock.
Unfortunately prev_balance() pass in a prev argument, to which we pass
rq->donor. However this prev value can now become stale and incorrect across a
rq lock drop.
So, to correct this, rework the prev_balance() call so that it does not take a
"prev" argument.
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260512025635.2840817-2-jstultz@google.com
BPF arena (kernel/bpf/arena.c) is compiled only on MMU && 64BIT, while
SCHED_CLASS_EXT depends on BPF_SYSCALL && BPF_JIT && DEBUG_INFO_BTF with no
64BIT requirement. On a 32-bit arch with a BPF JIT, SCX builds while the
arena helpers are absent, so the cid-form code's unconditional calls to
bpf_prog_arena() and bpf_arena_map_kern_vm_start() fail to link:
build_policy.o: undefined reference to `bpf_prog_arena'
build_policy.o: undefined reference to `bpf_arena_map_kern_vm_start'
Guard the three call sites with the same MMU && 64BIT condition that gates
arena.o. A cid-form scheduler needs a BPF arena, which isn't available on
such builds, so it can't run there regardless. cpu-form schedulers don't
touch the arena and are unaffected.
This is a quick workaround to get past the build errors. A fuller fix may
make the whole cid-form path conditional on the same condition, or drop
32-bit support outright.
Fixes: 0e2819cba9 ("sched_ext: Require an arena for cid-form schedulers")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202605310454.U9iByL2n-lkp@intel.com/
Closes: https://lore.kernel.org/oe-kbuild-all/202605310926.APXMc0RJ-lkp@intel.com/
Signed-off-by: Tejun Heo <tj@kernel.org>
update_tg_load_avg() is called once per leaf cfs_rq from the
__update_blocked_fair() walk that runs inside the NOHZ idle-balance
softirq, and again from update_load_avg() with UPDATE_TG. Its first
operation after the trivial early-outs is unconditionally:
now = sched_clock_cpu(cpu_of(rq_of(cfs_rq)));
if (now - cfs_rq->last_update_tg_load_avg < NSEC_PER_MSEC)
return;
Jakub ran into a system where nohz_idle_balance() was taking 75%
of a CPU (which is handling network traffic and doing many irq_exit_cpu
calls), with 35% of that CPU spent in update_load_avg, and 17% of the
CPU in sched_clock_cpu(), reading the TSC.
In a quick synthetic test, it looks like this patch reduces the
CPU use of sched_balance_update_blocked_averages by about 20%.
Switch the rate-limit to read rq_clock(rq_of(cfs_rq)) instead.
This eliminates the rdtsc, and uses a fairly fresh timestamp,
because all callers of update_tg_load_avg() and clear_tg_load_avg()
hold rq->lock and have called update_rq_clock(rq) within microseconds:
caller pre-state
__update_blocked_fair encloser did update_rq_clock(rq)
update_load_avg's three UPDATE_TG sites under rq->lock after enqueue/dequeue/update_curr
attach_/detach_entity_cfs_rq preceded by update_load_avg(...)
clear_tg_load_avg via offline path rq_clock_start_loop_update(rq) upfront
so rq->clock is fresh at every call. Since cfs_rqs are per-CPU
per-task_group, cfs_rq->last_update_tg_load_avg is always compared
against the same rq's clock; no cross-rq drift.
Signed-off-by: Rik van Riel <riel@surriel.com>
Assisted-by: Claude (Anthropic)
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260527110250.6a91718d@fangorn
Commit cd959a3562 ("sched_ext: Add a DL server for sched_ext tasks")
introduced an ext_server deadline server to protect sched_ext tasks from
fair/RT starvation, mirroring the existing fair_server.
Currently, both servers reserve their 50ms/1000ms bandwidth at boot,
regardless of whether a BPF scheduler is loaded. Unused bandwidth is
still reclaimed at runtime by other classes, but the static reservation
prevents the RT class from implicitly using that headroom when one of
the two classes is guaranteed to be empty.
A sysadmin can work around this by writing
/sys/kernel/debug/sched/{fair,ext}_server/cpu*/runtime, but that
requires manual action and not all systems expose debugfs.
A better approach is to make server bandwidth reservations dynamic: only
the scheduling policy that is currently active should register its
reservation, while the inactive one should not artificially hold
capacity (keeping both reservations only when the BPF scheduler is
running in partial mode):
+---------------------------------------------+-------------+------------+
| BPF scheduler state | fair server | ext server |
+---------------------------------------------+-------------+------------+
| not loaded (default boot) | reserved | none |
| loaded full mode (!SCX_OPS_SWITCH_PARTIAL) | none | reserved |
| loaded partial mode (SCX_OPS_SWITCH_PARTIAL)| reserved | reserved |
+---------------------------------------------+-------------+------------+
To achieve this, introduce an "attached/detached" state for each
deadline server, so the kernel can decide whether a server's bandwidth
should be accounted in global bandwidth tracking.
At boot, the system starts with only the fair server contributing to
bandwidth accounting. When a BPF scheduler is enabled, the ext server is
attached and may replace or complement the fair server depending on
whether full or partial mode is used. When sched_ext is disabled, the
system restores the previous deadline bandwidth values and behavior.
The transition logic ensures that switching between scheduling modes is
consistent and reversible, without losing runtime configuration or
requiring manual intervention.
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://patch.msgid.link/20260526164420.638711-2-arighi@nvidia.com
Writing runtime or period via the per-CPU dl_server debugfs files
(/sys/kernel/debug/sched/{fair,ext}_server/cpu*/{runtime,period}) on an
offline CPU can trigger two distinct kernel issues:
1) Divide-by-zero in dl_server_apply_params():
Oops: divide error: 0000 [#1] SMP NOPTI
RIP: 0010:dl_server_apply_params+0x239/0x3a0
Call Trace:
sched_server_write_common.isra.0+0x21a/0x3c0
full_proxy_write+0x78/0xd0
vfs_write+0xe7/0x6e0
Both __dl_sub() and __dl_add() divide by cpus internally, which can be
0 once the CPU has been removed from any active root-domain span (this
has been latent since the debugfs interface was introduced).
2) WARN_ON_ONCE in dl_server_start():
WARNING: kernel/sched/deadline.c:1805 at dl_server_start+0x232/0x270
Commit ee6e44dfe6 ("sched/deadline: Stop dl_server before CPU goes
offline") added this check to catch enqueueing the server on an
offline rq.
There's no meaningful semantics for re-configuring the per-CPU dl_server
bandwidth while the CPU is offline, so simply reject the write with
-EBUSY so userspace gets a clear error.
Closes: https://lore.kernel.org/all/20260526092228.3B6891F00A3A@smtp.kernel.org/
Fixes: d741f297bc ("sched/fair: Fair server interface")
Reported-by: Sashiko <sashiko-bot@kernel.org>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Juri Lelli <juri.lelli@redhat.com>
Tested-by: abaci-kreproducer <abaci@linux.alibaba.com>
Link: https://patch.msgid.link/20260526100502.575774-1-arighi@nvidia.com
Venkat Reported a boot kernel panic next-20260522. Git bisect pointed to
b5ea300a17 ("sched/cache: Make LLC id continuous")
Stacktrace points to llc_mask being null.
NIP [c000000000e58504] _find_first_bit+0x44/0x130
LR [c000000000e58500] _find_first_bit+0x40/0x130
Call Trace:
build_sched_domains+0xad8/0xe50
sched_init_smp+0xa8/0x164
kernel_init_freeable+0x250/0x370
ret_from_kernel_user_thread+0x14/0x1c
On powerpc, cpu_coregroup_mask is available only when the underlying
hardware support coregroup. In shared LPAR, QEMU guest or power9 etc
coregroup isn't supported. In such cases llc_mask was being referenced
when it was null leading to panic.
On powerpc, LLC is at SMT core level. So assumption that coregroup(MC)
domain point to LLC is wrong. Provide a way for archs to say where its
LLC is if it not at MC domain.
Fixes: b5ea300a17 ("sched/cache: Make LLC id continuous")
Closes: https://lore.kernel.org/all/51154de7-3700-4cb4-82f2-1b3a8fa427f7@linux.ibm.com/
Reported-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Co-developed-by: Chen, Yu C <yu.c.chen@intel.com>
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Tested-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
Link: https://patch.msgid.link/20260529075712.1181039-1-sshegde@linux.ibm.com
|| is a boolean operator, any nonzero (error) return short-circuits
to 1 rather than the actual errno. The caller in scx_init() logs and
propagates this value, so the wrong code reaches upper layers.
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
The reason for pick_next_task_fair() is the put/set optimization that
avoids touching the common ancestors. However, it is possible to
implement this in the put_prev_task() and set_next_task() calls as
used in put_prev_set_next_task().
Notably, put_prev_set_next_task() is the only site that:
- calls put_prev_task() with a .next argument;
- calls set_next_task() with .first = true.
This means that put_prev_task() can determine the common hierarchy and
stop there, and then set_next_task() can terminate where put_prev_task
stopped.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260511120628.057634261@infradead.org
With commit 50653216e4 ("sched: Add support to pick functions to
take rf") removing the balance callback, the pick_task() callback is
in charge of newidle balancing.
This means pick_task_fair() should do so too. This hasn't been a
problem in practise because pick_next_task_fair() is used. However,
since we'll be removing that one shortly, make sure pick_next_task()
is up to scratch.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260511120627.944705718@infradead.org
Robots figured out you can read and write this concurrently and got
'upset'. Gemini even noted sched_dynamic_show() can generate
'confusing' output if it observed different values during the
printing.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260511120627.176946327@infradead.org
Commit c9d93a73ce ("sched/fair: Drop redundant RCU read lock in NOHZ
kick path") removed the rcu_read_lock()/unlock() pair from
set_cpu_sd_state_busy() and set_cpu_sd_state_idle() on the assumption
that all callers run in a safe context for rcu_dereference_all(): IRQs
disabled or cpus_write_lock() held.
That assumption is wrong for the CPU hotplug teardown path. When CPUs
are taken offline, set_cpu_sd_state_busy() is invoked via:
cpuhp/N kthread
cpuhp_thread_fun()
cpuhp_invoke_callback()
sched_cpu_deactivate()
nohz_balance_exit_idle()
set_cpu_sd_state_busy()
rcu_dereference_all(per_cpu(sd_llc, cpu))
The cpuhp kthread holds cpu_hotplug_lock (percpu-rwsem) but runs with
preemption and IRQs enabled. As a result, lockdep correctly reports a
suspicious RCU usage on CPU offline, e.g.:
# echo 0 > /sys/devices/system/cpu/cpu1/online
=============================
WARNING: suspicious RCU usage
-----------------------------
kernel/sched/fair.c:12793 suspicious rcu_dereference_check() usage!
...
2 locks held by cpuhp/1/20:
#0: (cpu_hotplug_lock){++++}-{0:0}, at: cpuhp_thread_fun+0x42/0x1ae
#1: (cpuhp_state-down){+.+.}-{0:0}, at: cpuhp_thread_fun+0x72/0x1ae
Call Trace:
lockdep_rcu_suspicious
nohz_balance_exit_idle
sched_cpu_deactivate
cpuhp_invoke_callback
cpuhp_thread_fun
smpboot_thread_fn
Fix this by adding RCU read lock coverage to the one caller that lacks
it: nohz_balance_exit_idle() in the CPU hotplug teardown.
The other callers (nohz_balancer_kick() and nohz_balance_enter_idle())
genuinely run with IRQs disabled, so they remain unchanged.
Fixes: c9d93a73ce ("sched/fair: Drop redundant RCU read lock in NOHZ kick path")
Closes: https://lore.kernel.org/all/38fe0a1d-1a48-435a-910a-c278024d9ac9@samsung.com/
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260522092523.2046095-1-arighi@nvidia.com
ops_cid.set_cmask() expects a cmask. The kernel couldn't write into the
arena, so it translated cpumask -> cmask in kernel memory and passed the
result as a trusted pointer. The BPF cmask helpers all operate on arena
cmasks though, so the BPF side had to word-by-word probe-read the kernel
cmask into an arena cmask via cmask_copy_from_kernel() before any helper
could touch it. It works, but is clumsy.
With direct kernel-side arena access now in place, build the cmask in the
arena. The kernel writes to it through the kern_va side of the dual mapping.
BPF directly dereferences it via an __arena pointer like any other arena
struct.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Build a per-scheduler sub-allocator on top of pages claimed from the BPF
arena registered in the previous patch. Subsequent kernel-managed
arena-resident structures (e.g. per-CPU set_cmask cmask) carve their storage
from this pool.
scx_arena_pool_init() creates a gen_pool. scx_arena_alloc() returns the
kernel VA. On exhaustion, the pool grows by claiming more pages via
bpf_arena_alloc_pages_sleepable(). Chunks are added at the kernel-side
mapping address. Callers translate to the BPF-arena form themselves if
needed.
Allocations sleep (GFP_KERNEL) - they may grow the pool through vzalloc and
arena page allocation. All current consumers run from the enable path (after
ops.init() and the kernel-side arena auto-discovery, before validate_ops()),
where sleeping is fine.
scx_arena_pool_destroy() walks each chunk, returns outstanding ranges to the
gen_pool with gen_pool_free() and then calls gen_pool_destroy(). The
underlying arena pages are released when the arena map itself is torn down,
so the pool destroy doesn't free them explicitly.
v2: Switch scx_arena_alloc() to a loop. (Andrea)
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Upcoming patches will let the kernel place arena-resident scratch shared
with the BPF program (e.g. per-CPU set_cmask cmask) so the BPF side can
dereference it directly via __arena pointers, replacing the current
cmask_copy_from_kernel() probe-read loop. That requires each cid-form
scheduler to expose its arena to the kernel. Kernel- side accesses are
recovered by the per-arena scratch-page mechanism.
bpf_scx_reg_cid() walks the struct_ops member progs via
bpf_struct_ops_for_each_prog() and reads each prog's arena via
bpf_prog_arena(). The verifier enforces one arena per program, so each
member prog contributes at most one arena. All non-NULL contributions must
match and at least one member prog must use an arena. The map ref is held on
scx_sched and dropped on sched destroy. cpu-form schedulers (bpf_scx_reg)
are unchanged - no arena requirement.
Signed-off-by: Tejun Heo <tj@kernel.org>
- Spurious WARN in ops_dequeue() racing with concurrent dispatch.
- Self-deadlock between scheduler disable and a concurrent sub-sched
enable.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCahCHGQ4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGdKOAP9C6xYbZSXuPJSugQg7Ogq7GTcMf0EtK7CGVb9x
0pVJigEA1E17Vqf1WTWTp2DOsoPV1adS51wcoTGvklJc0eFRrw0=
=i40E
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-7.1-rc4-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- Spurious WARN in ops_dequeue() racing with concurrent dispatch
- Self-deadlock between scheduler disable and a concurrent sub-sched
enable
* tag 'sched_ext-for-7.1-rc4-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: Fix spurious WARN on stale ops_state in ops_dequeue()
sched_ext: Fix deadlock between scx_root_disable() and concurrent forks
ops_dequeue() can race with finish_dispatch() and spuriously trigger the
"queued task must be in BPF scheduler's custody" warning.
ops_dequeue() snapshots p->scx.ops_state via atomic_long_read_acquire()
and then, in the SCX_OPSS_QUEUED arm, asserts that SCX_TASK_IN_CUSTODY
is set. The two reads are not atomic w.r.t. a concurrent
finish_dispatch() running on another CPU:
CPU 1 CPU 2
===== =====
dequeue_task_scx()
ops_dequeue()
opss = read_acquire(ops_state)
= SCX_OPSS_QUEUED
finish_dispatch()
cmpxchg ops_state:
SCX_OPSS_QUEUED -> SCX_OPSS_DISPATCHING [succeeds]
dispatch_enqueue(SCX_DSQ_GLOBAL,
SCX_ENQ_CLEAR_OPSS)
call_task_dequeue()
p->scx.flags &= ~SCX_TASK_IN_CUSTODY
WARN_ON_ONCE(!(p->scx.flags &
SCX_TASK_IN_CUSTODY))
/* opss is stale: QUEUED,
* but task already claimed */
set_release(ops_state, SCX_OPSS_NONE)
The race has been observed via two distinct call chains: the most common
goes through sched_setaffinity(), a rarer variant through
sched_change_begin().
For SCX_DSQ_GLOBAL / SCX_DSQ_BYPASS, dispatch_enqueue() clears
SCX_TASK_IN_CUSTODY before clearing ops_state to SCX_OPSS_NONE
(intentional, to avoid concurrent non-atomic RMW of p->scx.flags against
ops_dequeue()). The window between those two writes is exactly what
ops_dequeue() observes as "QUEUED without custody".
The observed state is not actually inconsistent, it just means CPU 1 has
already claimed the task and the QUEUED value held by CPU 2 is stale.
Re-read ops_state in that case; the next read is guaranteed to return
SCX_OPSS_DISPATCHING or SCX_OPSS_NONE, both of which exit the switch
cleanly. The retry is bounded: once IN_CUSTODY is cleared, ops_state has
already advanced past QUEUED for this dispatch cycle, and a fresh QUEUED
would require re-enqueue under p's rq lock, which CPU 2 holds.
Changes in v2:
- Use READ_ONCE() for p->scx.flags to ensure fresh reads and prevent
compiler reordering in the lockless path
- Add cpu_relax() to reduce power consumption and improve performance
during the spin-wait
- Use unlikely() to optimize branch prediction for the common case
- Expand the in-code comment to document the race condition and
bounded retry guarantee
Fixes: ebf1ccff79 ("sched_ext: Fix ops.dequeue() semantics")
Suggested-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Samuele Mariotti <smariotti@disroot.org>
Signed-off-by: Paolo Valente <paolo.valente@unimore.it>
Signed-off-by: Tejun Heo <tj@kernel.org>
Sub-sched cap code and other upcoming consumers need bulk cmask ops, both
mutating (and/or/copy/andnot) and predicate (subset/intersects/empty).
cmask_walk_op2() walks the intersection of two ranges word by word;
cmask_walk_op1() walks one range. Both are __always_inline and dispatched on
a compile-time-constant op enum, so each public entry collapses to a
specialized loop with the inner switch reduced to one arm.
Two-cmask ops only touch bits in the intersection of the two ranges; bits
outside are left unchanged. scx_cmask_or_racy() and scx_cmask_copy_racy()
mirror the locking forms but read @src word-by-word through data_race();
callers handle ordering with concurrent writers themselves.
v2: Add scx_cmask_empty().
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_cmask carries @base and @nr_cids but not the bits[] allocation size, so
helpers reshaping the active range have no way to check it fits and later
kfuncs taking caller-provided storage can't validate it.
Add @alloc_words (u64 word count) annotated with __counted_by, and split the
bit-range API into three helpers:
- SCX_CMASK_DEFINE() / __SCX_CMASK_DEFINE() define an on-stack cmask, the
latter taking an explicit capacity for oversized storage.
SCX_CMASK_DEFINE_SHARD() is a thin wrapper that always reserves
SCX_CID_SHARD_MAX_CPUS bits of storage.
- scx_cmask_init() / __scx_cmask_init() initialize a cmask, with the same
tight-vs-explicit split.
- scx_cmask_reframe() reshapes the active range without resizing storage.
The BPF mirror (cmask_init / __cmask_init / cmask_reframe) gets the same
shape.
Add scx_cmask_clear() and scx_cmask_fill() to zero and set the
active-range bits respectively. scx_cpumask_to_cmask() uses
scx_cmask_clear(); scx_cmask_init() would otherwise re-write @alloc_words
on every call.
A later patch uses @alloc_words in scx_cmask_ref_shard() to refuse output
storage that can't hold the requested shard.
v2: Init per-CPU scx_set_cmask_scratch (was zero-init, emitted empty
cmasks). Add nr_cids/alloc_cids check in BPF __cmask_init().
(sashiko AI)
Widen SCX_CMASK_NR_WORDS()/CMASK_NR_WORDS() to compute in u64 so that
@nr_cids near U32_MAX no longer wraps to a small value and bypasses
the bounds check in cmask_reframe(). (Andrea)
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
struct scx_cmask is a base-windowed bitmap over cid space. Each bit
represents one cid, so the count of active bits is the count of cids. The
sibling struct scx_cid_shard already uses nr_cids. Rename as a prep so the
following patches that grow the cmask API can use the consistent name.
v2: Also rename src->nr_bits / dst->nr_bits in
cmask_copy_from_kernel(). (sashiko AI)
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Recent optimizations of sd->shared assignment moved to allocating a
single instance of per-CPU sched_domain_shared objects per s_data.
Recent optimizations to select_idle_capacity() moved the sd->shared
assignments to "sd_asym" domain when ASYM_CPUCAPACITY is detected but
cache-aware scheduling mandates the presence of "sd_llc_shared" to
compute and cache per-LLC statistics.
Use an "alloc_flags" union in sched_domain_shared to claim a
sched_domain_shared object per sched_domain. Allocation starts searching
for an available / matching sched_domain_shared instance from the first
CPU of sched_domain_span(sd) (sd can be sd_llc, or sd_asym). If the
shared object is claimed by another domain, the instance corresponding
to next CPU in the domain span is explored until a matching / available
instance is found.
In case of a single CPU in sched_domain_span(), the domain will be
degenerated and a temporary overlap of ->shared objects across different
domains is acceptable.
"alloc_flags" forms a union with "nr_idle_scan" and the stale flags are
left as is when the sd->shared is published. The expectation is for the
first load balancing instance to correct the value just like the current
behavior, except the initial value is no longer 0.
Originally-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: Andrea Righi <arighi@nvidia.com>
RT migration is done aggressively. When a CPU schedules out a high
priority RT task for a lower priority task, it will look to see if there's
any RT tasks that are waiting to run on another CPU that is of higher
priority than the task this CPU is about to run. If it finds one, it will
pull that task over to the CPU and allow it to run there instead.
Normally, this pulling is done by looking at the RT overloaded mask (rto)
which contains all the CPUs in the scheduler domain with RT tasks that are
waiting to run due to a higher priority RT task currently running on their
CPU. The CPU that is about to schedule a lower priority task will grab the
rq lock of the overloaded CPU and move the RT task from that CPU's runqueue
to the local one and schedule the higher priority RT task.
This caused issues when a lot of CPUs would schedule a lower priority task
at the same time. They would all try to grab the same runqueue lock of
the CPU with the overloaded RT tasks. Only the first CPU that got in will
get that task. All the others would wait until they got the runqueue lock
and see there's nothing to pull and do nothing. On systems with lots of
CPUs, this caused a large latency (up to 500us) which is beyond what
PREEMPT_RT is to allow.
The solution to that was to create an RT_PUSH_IPI logic. When any CPU
wanted to pull a task, instead of grabbing the runqueue lock of the
overloaded CPU, it would start by sending an IPI to the overloaded CPU,
and that IPI handler would have the CPU with the waiting RT task do a push
instead. Then that handler would send an IPI to the next CPU with
overloaded RT tasks, and so on. Note, after the first CPU starts this
process, if another CPU wanted to do a pull, it would see that the process
has already begun and would only increment a counter to have the IPIs
continue again.
The RT_PUSH_IPI solved the latency problem with PREEMPT_RT but could cause
a new issue with non PREEMPT_RT. Namely, softirqs run in a threaded
context on PREEMPT_RT but they can run in an interrupt context in non-RT.
If an IPI lands on a CPU that has just woken up multiple RT tasks and the
current CPU is running a non RT or a low priority RT task, instead of
doing a push, it would simply do a schedule on that CPU. But if a softirq
was also executing on this CPU, the schedule would need to wait until the
softirq finished. Until then, the CPU would still be considered overloaded
as there are RT tasks still waiting to run on it.
A live lock occurred on a workload that was doing heavy networking traffic
on a large machine where the softirqs would run 500us out of 750us. And it
would also be waking up RT tasks, causing the RT pull logic to be
constantly executed.
When a softirq triggered on a CPU with RT tasks queued but not running
yet, and the other CPUs would see this CPU as being overloaded, they would
send an IPI over to it. The CPU would notice that the waiting RT tasks are
of higher priority than the currently running task and simply schedule
that CPU instead. But because the softirq was executing, before it could
schedule, it would receive another IPI to do the same. The amount of IPIs
would slow down the currently running softirq so much that before it could
return back to task context, it would execute another softirq never
allowing the CPU to schedule. This live locked that CPU.
As RT_PUSH_IPI was created to help PREEMPT_RT, make it default off if
PREEMPT_RT is not enabled.
Fixes: b6366f048e ("sched/rt: Use IPI to trigger RT task push migration instead of pulling")
Closes: https://lore.kernel.org/all/20260506235716.2530720-1-tj@kernel.org/
Reported-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260515103740.25ccbed8@gandalf.local.home
K Prateek noticed we weren't setting the rq->next_class in
proxy_resched_idle(), when I was debugging an issue seen with
CONFIG_SCHED_PROXY_EXEC and some of Peter's new patches, and
suggested this fix.
So set rq->next_class when we temporarily switch the donor to
idle, so we don't accidentally call wakeup_preempt_fair()
with idle as the donor.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260514234732.3170197-1-jstultz@google.com
Add to select_idle_capacity() the same SIS_UTIL-controlled idle-scan
mechanism, already used by select_idle_cpu(): when sched_feat(SIS_UTIL)
is enabled and the LLC domain has sched_domain_shared data, derive the
per-attempt scan limit from sd->shared->nr_idle_scan.
That bounds the walk on large LLCs: once nr_idle_scan is exhausted,
return the best CPU seen so far. The early exit is gated on
!has_idle_core so an active idle-core search (SMT with idle cores
reported by test_idle_cores()) isn't cut short before it gets a chance
to find one.
Co-developed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260509180955.1840064-6-arighi@nvidia.com
When SD_ASYM_CPUCAPACITY load balancing considers pulling a misfit task,
capacity_of(dst_cpu) can overstate available compute if the SMT sibling is
busy: the core does not deliver its full nominal capacity.
If SMT is active and dst_cpu is not on a fully idle core, skip this
destination so we do not migrate a misfit expecting a capacity upgrade we
cannot actually provide.
Reported-by: Felix Abecassis <fabecassis@nvidia.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260509180955.1840064-5-arighi@nvidia.com
On systems with asymmetric CPU capacity (e.g., ACPI/CPPC reporting
different per-core frequencies), the wakeup path uses
select_idle_capacity() and prioritizes idle CPUs with higher capacity
for better task placement. However, when those CPUs belong to SMT cores,
their effective capacity can be much lower than the nominal capacity
when the sibling thread is busy: SMT siblings compete for shared
resources, so a "high capacity" CPU that is idle but whose sibling is
busy does not deliver its full capacity. This effective capacity
reduction cannot be modeled by the static capacity value alone.
Introduce SMT awareness in the asym-capacity idle selection policy: when
SMT is active, always prefer fully-idle SMT cores over partially-idle
ones.
Prioritizing fully-idle SMT cores yields better task placement because
the effective capacity of partially-idle SMT cores is reduced; always
preferring them when available leads to more accurate capacity usage on
task wakeup.
On an SMT system with asymmetric CPU capacities (NVIDIA Vera Rubin),
SMT-aware idle selection has been shown to improve throughput by around
15-18% over NO_ASYM mainline and by around 60% over ASYM mainline, for
CPU-bound workloads (NVBLAS) running an amount of tasks equal to the
amount of SMT cores.
Reported-by: Felix Abecassis <fabecassis@nvidia.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260511142502.3873984-1-arighi@nvidia.com
On asymmetric CPU capacity systems, the wakeup path uses
select_idle_capacity(), which scans the span of sd_asym_cpucapacity
rather than sd_llc.
The has_idle_cores hint however lives on sd_llc->shared, so the
wakeup-time read of has_idle_cores operates on an LLC-scoped blob while
the actual scan/decision spans the asym domain; nr_busy_cpus also lives
in the same shared sched_domain data, but it's never used in the asym
CPU capacity scenario.
Therefore, move the sched_domain_shared object to sd_asym_cpucapacity
whenever the CPU has a SD_ASYM_CPUCAPACITY_FULL ancestor and that
ancestor is non-overlapping (i.e., not built from SD_NUMA). In that case
the scope of has_idle_cores matches the scope of the wakeup scan.
Fall back to attaching the shared object to sd_llc in three cases:
1) plain symmetric systems (no SD_ASYM_CPUCAPACITY_FULL anywhere);
2) CPUs in an exclusive cpuset that carves out a symmetric capacity
island: has_asym is system-wide but those CPUs have no
SD_ASYM_CPUCAPACITY_FULL ancestor in their hierarchy and follow
the symmetric LLC path in select_idle_sibling();
3) exotic topologies where SD_ASYM_CPUCAPACITY_FULL lands on an
SD_NUMA-built domain. init_sched_domain_shared() keys the shared
blob off cpumask_first(span), which on overlapping NUMA domains
would alias unrelated spans onto the same blob. Keep the shared
object on the LLC there; select_idle_capacity() gracefully skips
the has_idle_cores preference when sd->shared is NULL.
While at it, also rename the per-CPU sd_llc_shared to sd_balance_shared,
as it is no longer strictly tied to the LLC.
Co-developed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260516055850.1345932-1-arighi@nvidia.com
nohz_balancer_kick() is reached from sched_balance_trigger(), which is
called from sched_tick(). sched_tick() runs with IRQs disabled, so the
additional rcu_read_lock/unlock() used around sched_domain accesses in
this path is redundant. Rely on the existing IRQ-disabled context (and
the rcu_dereference_all() checking) instead.
The same applies to set_cpu_sd_state_idle(), called from the idle entry
path with IRQs disabled, and to set_cpu_sd_state_busy(), reachable via
nohz_balance_exit_idle() from two contexts: nohz_balancer_kick() (IRQs
disabled, as above) and sched_cpu_deactivate() (the CPUHP_AP_ACTIVE
teardown, which runs under cpus_write_lock(), so it cannot race with
sched-domain rebuilds). In both cases the rcu_dereference_all()
validation is sufficient.
No functional change intended.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260509180955.1840064-2-arighi@nvidia.com
There is a use of sched_smt_active() and explicit use of sched_smt_present.
Remove the explicit usage for better code maintenance and readability.
Note that this differs slightly for update_idle_core. It used to call
static_branch_unlikely earlier and now it will call static_branch_likely.
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Link: https://patch.msgid.link/20260515172456.542799-5-sshegde@linux.ibm.com
For fastpaths such as wakeup and load balance even minimal code additions
can add up. is_core_idle is accessed during load balance.
Other callsites of is_core_idle make sched_smt_active() check first.
Make the same check in should_we_balance.
Rest of access to cpu_smt_mask isn't in fastpath.
Note: Remove the stale comment above is_core_idle. Enqueue methods
of fair aren't close to it anymore.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Link: https://patch.msgid.link/20260515172456.542799-4-sshegde@linux.ibm.com
Now, that cpu_smt_mask is defined as cpumask_of(cpu) for
CONFIG_SCHED_SMT=n, it is possible to get rid of the ifdeffery.
Effectively,
- This makes sched_smt_present is defined always
- cpumask_weight(cpumask_of(cpu)) == 1. So sched_smt_present_inc/dec
will never enable the sched_smt_present. Which is expected.
- Paths that were compile-time eliminated become runtime guarded
using static keys.
- Defines set_idle_cores, test_idle_cores, etc which could likely benefit
the CONFIG_SCHED_SMT=n systems to use the same optimizations within the
LLC at wakeups.
- This will expose sched_smt_present symbol for CONFIG_SCHED_SMT=n.
Likely not a concern.
- There is a bloat of code CONFIG_SCHED_SMT=n. (NR_CPUS=2048)
add/remove: 24/18 grow/shrink: 26/28 up/down: 6396/-3188 (3208)
Total: Before=30629880, After=30633088, chg +0.01%
- No code bloat for CONFIG_SCHED_SMT=y, which is expected.
- Add comments around stop_core_cpuslocked on why ifdefs are not
removed.
- This leaves the remaining uses of CONFIG_SCHED_SMT mainly for
topology building bits which has a policy based decision.
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Phil Auld <pauld@redhat.com>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Acked-by: Tejun Heo <tj@kernel.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260515172456.542799-3-sshegde@linux.ibm.com
util_est_update() must be called after updating util_avg during the dequeue
of a task and only when the task is not delayed dequeue.
Move util_est_update() in update_load_avg().
Fixes: b55945c500 ("sched: Fix pick_next_task_fair() vs try_to_wake_up() race")
Closes: https://lore.kernel.org/all/20260512124653.305275-1-qyousef@layalina.io/
Reported-by: Qais Yousef <qyousef@layalina.io>
Reviewed-and-tested-by: Qais Yousef <qyousef@layalina.io>
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260518102345.268452-1-vincent.guittot@linaro.org
Commit 77baa5bafc ("sched/cputime: Fix mul_u64_u64_div_u64() precision
for cputime") added a clamp in cputime_adjust():
if (unlikely(stime > rtime))
stime = rtime;
The justification was that mul_u64_u64_div_u64() could over-approximate
on some architectures (notably arm64 and the old 32-bit fallback), so
the mathematically impossible stime > rtime was nevertheless reachable
and would underflow utime = rtime - stime.
That premise no longer holds. Commit b29a62d87c ("mul_u64_u64_div_u64:
make it precise always") replaced the fallback implementation with an
exact 128-bit long division, and the x86_64 inline asm already produced
exact results. The helper now returns the mathematically correct
floor(a*b/d) on every architecture, so stime <= rtime is guaranteed by
stime <= stime + utime and the clamp is dead code.
Remove it along with its stale comment.
Signed-off-by: Nicolas Pitre <nico@fluxnic.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260514202629.673539-1-nico@fluxnic.net
Enqueue and replenish non-deferred deadline servers when their runtime is
exhausted and the replenishment timer could not be started because it is
too close to the wake-up instant.
Signed-off-by: Yuri Andriaccio <yurand2000@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260430213835.62217-2-yurand2000@gmail.com
Set the default total bandwidth for SCHED_DEADLINE tasks and servers
to ONE. FIFO/RR tasks are already throttled by fair-servers and
ext-servers, and the sysctl_sched_rt_runtime parameter now only
defines the total bw that is allowed to deadline entities.
Signed-off-by: Yuri Andriaccio <yurand2000@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260430213835.62217-22-yurand2000@gmail.com
On fork without CLONE_VM, the child gets a new mm,
the parent's preferred_llc value is stale for the
child.
Fix this by resetting the task's preferred_llc to -1.
This bug was reported by sashiko.
Fixes: 47d8696b95 ("sched/cache: Assign preferred LLC ID to processes")
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/0ec7309d0e24ede97656754d1505b7490403d966.1778703694.git.tim.c.chen@linux.intel.com
sched_cache_present is a global static key, but build_sched_domains()
is called per partition from the "Build new domains" loop in
partition_sched_domains_locked(). Each call unconditionally sets the
key based solely on the has_multi_llcs local variable for that partition.
The call to the last partition set the value even when there
are previous partitions with multiple LLCs.
If partition A (multi-LLC) is built first, the key is enabled. Then
when partition B (single-LLC) is built, the key is disabled. The
multi-LLC partition A is still active but the key is now off.
Fix it by doing a similar thing as sched_energy_present: check the
multi-LLCs during the iteration over all the partitions rather than
checking it on a single partition.
This bug was reported by sashiko.
Fixes: d59f4fd1d3 ("sched/cache: Enable cache aware scheduling for multi LLCs NUMA node")
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/c541af2547d54509fbfd3b3a1e8072e2e5c7ff68.1778703694.git.tim.c.chen@linux.intel.com
If there are multiple LLCs in the system, cache aware scheduling
should be enabled. However, there is a corner case where, if there
is a single NUMA node and a single LLC per node, cache aware
scheduling will be turned on in the current implementation -
because at this moment, the parent domain has not yet been
degenerated, and it is possible that the current domain has the
same cpu span as its parent. There is no need to turn cache aware
scheduling on in this scenario.
Fix it by iterating the parent domains to find a domain that is
a superset of the current sd_llc, so that later, after the duplicated
parent domains have been degenerated, cache aware scheduling will
take effect.
For example, the expected behavior would be:
2 sockets, 1 LLC per socket: MC span=0-3, PKG span=0-7, has_multi_llcs=true
1 socket, 2 LLCs per socket: MC span=0-3, PKG span=0-7, has_multi_llcs=true
2 sockets, 2 LLCs per socket: MC span=0-3, PKG span=0-7, has_multi_llcs=true
1 socket, 1 LLC per socket: MC span=0-3, PKG span=0-3, has_multi_llcs=false
This bug was reported by sashiko.
Fixes: d59f4fd1d3 ("sched/cache: Enable cache aware scheduling for multi LLCs NUMA node")
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/6328a8a7f40925cec2a712d81ee58128a4c4444a.1778703694.git.tim.c.chen@linux.intel.com
sched_cache_active_set_unlocked() checks hardware support without
locks:
static void sched_cache_active_set(bool locked)
{
/* hardware does not support */
if (!static_branch_likely(&sched_cache_present)) {
_sched_cache_active_set(false, locked);
return;
}
...
If build_sched_domains() runs concurrently during CPU hotplug,
it can disable sched_cache_present under sched_domains_mutex
and the CPU hotplug lock. If a debugfs write thread evaluates
sched_cache_present as true right before that, and then blocks
or gets preempted, it might proceed to enable sched_cache_active
after the hardware support has been marked as absent. Make it
safer by acquiring cpus_read_lock() and sched_domains_mutex_lock()
when the user changes sched_cache_active via debugfs.
This bug was reported by sashiko.
Fixes: 067a313581 ("sched/cache: Allow the user space to turn on and off cache aware scheduling")
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/9afddf439687f04bb56b46625bd9f153eb8abad5.1778703694.git.tim.c.chen@linux.intel.com
The currently running task cur may not be a CFS task, such as
an RT or Deadline task. For non-CFS tasks, the task_util(cur)
utilization average is not maintained, so this might pass a
stale or meaningless value to can_migrate_llc().
Check if the task is CFS before getting its task_util().
This bug was reported by sashiko.
Fixes: 714059f79f ("sched/cache: Handle moving single tasks to/from their preferred LLC")
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/f9161133cf040d286dca11344a112c5ef2a5253d.1778703694.git.tim.c.chen@linux.intel.com
There is a race condition that, after a task is enqueued
on a runqueue, task_llc(p) may change due to CPU hotplug,
because the llc_id is dynamically allocated and adjusted
at runtime.
Therefore, checking task_llc(p) to determine whether the
task is being dequeued from its preferred LLC is unreliable
and can cause inconsistent values.
To fix this problem, record whether p is enqueued on its
preferred LLC, in order to pair with account_llc_dequeue()
to maintain a consistent nr_pref_llc_running per runqueue.
This bug was reported by sashiko, and the solution was once
suggested by Prateek.
Fixes: 46afe3af7e ("sched/cache: Track LLC-preferred tasks per runqueue")
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/0c8c6a1571d66792a4d2ff0103ba3cc13e059046.1778703694.git.tim.c.chen@linux.intel.com
mm->sc_stat.cpu is written by task_cache_work() and could be read
locklessly by several functions on other CPUs. Use READ_ONCE and
WRITE_ONCE on mm->sc_stat.cpu access and write to prevent inconsistent
values from compiler optimizations when there are multiple accesses.
For example in get_pref_llc(), if the writer updated the field between
two compiler-generated loads, the validation (e.g., cpu != -1) and
subsequent use (e.g., llc_id(cpu)) could operate on different values,
allowing a negative CPU ID to be used as an index.
Leave plain write in mm_init_sched(), where the mm is not
yet visible to other CPUs.
This bug was reported by sashiko.
Fixes: 47d8696b95 ("sched/cache: Assign preferred LLC ID to processes")
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/63ea494f12efcf265d7134400a06cd75d7f2c310.1778703694.git.tim.c.chen@linux.intel.com
A concurrent task exit might cause a NULL pointer dereference
in account_mm_sched(). Use the locally cached mm pointer instead,
since the active_mm reference guarantees the structure remains
allocated. Meanwhile, skip the kernel thread because it has
nothing to do with cache aware scheduling.
This bug was reported by sashiko and Vern.
Fixes: df0d984759 ("sched/cache: Introduce infrastructure for cache-aware load balancing")
Reported-by: Vern Hao <haoxing990@gmail.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/09cf7ee3-6e27-4505-9692-4b4a4707c8b2@gmail.com/
Link: https://patch.msgid.link/066d8cfa45d4822bf4367e788c50377c66bbcc82.1778703694.git.tim.c.chen@linux.intel.com
rcu_dereference_all() should be used to access the
sd_llc domain under RCU protection.
This bug was reported by sashiko.
Fixes: df0d984759 ("sched/cache: Introduce infrastructure for cache-aware load balancing")
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/2dc49455e861215d8059a1c877953f0b95990038.1778703694.git.tim.c.chen@linux.intel.com
Introduce a set of debugfs knobs to control how aggressively the
cache aware scheduling does the task aggregation.
(1) aggr_tolerance
With sched_cache enabled, the scheduler uses a process's footprint
as a proxy for its LLC footprint to determine if aggregating tasks
on the preferred LLC could cause cache contention. If the footprint
exceeds the LLC size, aggregation is skipped. Since the kernel
cannot efficiently track per-task cache usage (resctrl is
user-space only), userspace can provide a more accurate hint.
Introduce /sys/kernel/debug/sched/llc_balancing/aggr_tolerance to
let users control how strictly footprint limits aggregation. Values
range from 0 to 100:
- 0: Cache-aware scheduling is disabled.
- 1: Strict; tasks with footprint larger than LLC size are skipped.
- >=100: Aggressive; tasks are aggregated regardless of footprint.
For example, with a 32MB L3 cache:
- aggr_tolerance=1 -> tasks with footprint > 32MB are skipped.
- aggr_tolerance=99 -> tasks with footprint > 784GB are skipped
(784GB = (1 + (99 - 1) * 256) * 32MB).
Similarly, /sys/kernel/debug/sched/llc_balancing/aggr_tolerance also
controls how strictly the number of active threads is considered when
doing cache aware load balance. The number of SMTs is also considered.
High SMT counts reduce the aggregation capacity, preventing excessive
task aggregation on SMT-heavy systems like Power10/Power11.
Yangyu suggested introducing separate aggregation controls for the
number of active threads and memory footprint checks. Since there are
plans to add per-process/task group controls, fine-grained tunables are
deferred to that implementation.
(2) epoch_period, epoch_affinity_timeout,
imb_pct, overaggr_pct are also turned into tunables.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Suggested-by: Madadi Vineeth Reddy <vineethr@linux.ibm.com>
Suggested-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Suggested-by: Tingyin Duan <tingyin.duan@gmail.com>
Suggested-by: Jianyong Wu <jianyong.wu@outlook.com>
Suggested-by: Yangyu Chen <cyy@cyyself.name>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Tingyin Duan <tingyin.duan@gmail.com>
Link: https://patch.msgid.link/1c62cc060ba2b33d7b1f0ed98b3390128edbae93.1778703694.git.tim.c.chen@linux.intel.com
Prateek and Tingyin reported that memory-intensive workloads (such as
stream) can saturate memory bandwidth and caches on the preferred LLC
when sched_cache aggregates too many threads.
To mitigate this, estimate a process's memory footprint by comparing
its NUMA balancing fault statistics to the size of the LLC. If the
footprint exceeds the LLC size, skip cache-aware scheduling.
Note that footprint is only an approximation of the memory footprint,
since the kernel lacks suitable metrics to estimate the real working
set. If a user-provided hint is available in the future, it would be
more accurate. A later patch will allow users to provide a hint to
adjust this threshold.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Suggested-by: Vern Hao <vernhao@tencent.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Tingyin Duan <tingyin.duan@gmail.com>
Link: https://patch.msgid.link/95cf64a385bcc12f18dcebe9d59e8d3ba8bb318f.1778703694.git.tim.c.chen@linux.intel.com
Cache aware scheduling needs to know the LLC size that a process
can use, so as to avoid memory-intensive tasks from being
over-aggregated on a single LLC.
Introduce a preparation patch to add get_effective_llc_bytes() to
get the LLC size that a CPU can use. The function can be further
enhanced by subtracting the LLC cache ways reserved by resctrl
(CAT in Intel RDT, etc).
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Tingyin Duan <tingyin.duan@gmail.com>
Link: https://patch.msgid.link/37afee09ff608034da0ce149e72d33b6f4698edf.1778703694.git.tim.c.chen@linux.intel.com
For a single thread, the current wakeup path tends to place it
on the same LLC where it was previously running with cache-hot
data. There is no need to enable cache-aware scheduling for
single-threaded processes for the following reasons:
1. Cache-aware scheduling primarily benefits multi-threaded
processes where threads share data. Single-threaded processes
typically have no inter-thread data sharing and thus gain little.
2. Enabling it incurs the additional overhead of tracking the
thread's residency in the LLCs.
3. Bypassing single-threaded processes avoids excessive
concentration of such tasks on a single LLC.
Nevertheless, this check can be omitted if users explicitly
provide hints for such single-threaded workloads where different
processes have shared memory, e.g., via prctl() or other interfaces
to be added in the future.
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Tingyin Duan <tingyin.duan@gmail.com>
Link: https://patch.msgid.link/8a59a13aa58fdb48e410ecb2aabd97fe3ea5d256.1778703694.git.tim.c.chen@linux.intel.com
A performance regression was observed by Prateek when running hackbench
with many threads per process (high fd count). To avoid this, processes
with a large number of active threads are excluded from cache-aware
scheduling.
With sched_cache enabled, record the number of active threads in each
process during the periodic task_cache_work(). While iterating over
CPUs, if the currently running task belongs to the same process as the
task that launched task_cache_work(), increment the active thread count.
If the number of active threads within the process exceeds the number
of Cores (divided by the SMT number) in the LLC, do not enable
cache-aware scheduling. However, on systems with a smaller number of
CPUs within 1 LLC, like Power10/Power11 with SMT4 and an LLC size of 4,
this check effectively disables cache-aware scheduling for any process.
One possible solution suggested by Peter is to use an LLC-mask instead
of a single LLC value for preference. Once there are a 'few' LLCs as
preference, this constraint becomes a little easier. It could be an
enhancement in the future.
For users who wish to perform task aggregation regardless, a debugfs knob
is provided for tuning in a subsequent change.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Suggested-by: Aaron Lu <ziqianlu@bytedance.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Tingyin Duan <tingyin.duan@gmail.com>
Link: https://patch.msgid.link/d076cd21a8e6c6341d1e2d927e118db770ebb650.1778703694.git.tim.c.chen@linux.intel.com
Scanning online CPUs to calculate the occupancy might be
time-consuming. Only allow 1 thread of the process to scan
the CPUs at the same time, which is similar to what
NUMA balance does in task_numa_work().
Signed-off-by: Jianyong Wu <wujianyong@hygon.cn>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/5672b52e588b855b01e5a1a17822f7c6c7237a3d.1778703694.git.tim.c.chen@linux.intel.com
Pull to receive:
39e25a2100 ("sched_ext: Drop NONE early return in scx_disable_and_exit_task()")
b273b75b8d ("sched_ext: INIT_LIST_HEAD() &sch->all in scx_alloc_and_add_sched()")
cceb874eee ("sched_ext: Defer sub_kset base put to scx_sched_free_rcu_work")
6ae315d379 ("sched_ext: Use HK_TYPE_DOMAIN_BOOT to detect isolcpus= domain isolation")
515e3996a4 ("sched_ext: Fix deadlock between scx_root_disable() and concurrent forks")
to prepare for-7.2 for further sub-sched changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_root_disable() enters SCX_DISABLING before it grabs scx_enable_mutex to
clear __scx_switched_all and scx_switching_all. task_should_scx() short-circuits on DISABLING,
so forks in that window land on fair while next_active_class() still skips
fair - the new tasks stall.
This can deadlock the disable path itself: scx_alloc_and_add_sched() runs
under scx_enable_mutex and creates a helper kthread; if that new kthread is
one of the stalled fair tasks, the mutex holder waits forever and
scx_root_disable() can never make progress. Only sub-sched support exposes
this, since sub-sched enables are the only path where
scx_alloc_and_add_sched() can race the root's disable.
Move the DISABLING check after @scx_switching_all. @scx_switching_all
serves as a proxy for __scx_switched_all, so while it's set, forks keep
going to scx. Once cleared, DISABLING applies normally.
v2: Reword in-source comment and description. (Andrea)
Fixes: 337ec00b1d ("sched_ext: Implement cgroup sub-sched enabling and disabling")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Bulk is hardening of the new sub-scheduler infrastructure.
- UAFs and lifecycle bugs on the sub-sched attach/detach paths: parent
sub_kset freed under a racing child, list_del_rcu on an uninitialized
list head, ops->priv stomped by concurrent attach/detach, and a UAF in
the init-failure error path.
- Task state-machine reorg closing concurrent enable-vs-dead races: a
task exiting during the unlocked init window could trip NULL ops
derefs or skip exit_task() cleanup.
- A scx_link_sched() self-deadlock on scx_sched_lock.
- isolcpus: stop dereferencing the now-RCU-protected HK_TYPE_DOMAIN
cpumask without RCU, and stop rejecting BPF schedulers when only
cpuset isolated partitions are active.
- PREEMPT_RT: disable irq_work runs in hardirq context so dumps show the
failing task rather than the irq_work kthread.
- Assorted !CONFIG_EXT_SUB_SCHED, randconfig, and selftest build fixes.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCagTk1g4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGT6TAP0ZbRHz9ViligecZXIHjEvZQjEV4sn1NLpGi4og
V0Ol2AD/RzqHQZo5+HpMz4hPrcZdkAWcr74cLrNTJ2WQjOk4RgE=
=6Mbx
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-7.1-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
"The bulk of this is hardening of the new sub-scheduler infrastructure.
- UAFs and lifecycle bugs on the sub-sched attach/detach paths:
parent sub_kset freed under a racing child, list_del_rcu on an
uninitialized list head, ops->priv stomped by concurrent
attach/detach, and a UAF in the init-failure error path
- Task state-machine reorg closing concurrent enable-vs-dead races: a
task exiting during the unlocked init window could trip NULL ops
derefs or skip exit_task() cleanup
- A scx_link_sched() self-deadlock on scx_sched_lock
- isolcpus: stop dereferencing the now-RCU-protected HK_TYPE_DOMAIN
cpumask without RCU, and stop rejecting BPF schedulers when only
cpuset isolated partitions are active
- PREEMPT_RT: disable irq_work runs in hardirq context so dumps show
the failing task rather than the irq_work kthread
- Assorted !CONFIG_EXT_SUB_SCHED, randconfig, and selftest build
fixes"
* tag 'sched_ext-for-7.1-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: Use HK_TYPE_DOMAIN_BOOT to detect isolcpus= domain isolation
sched_ext: Defer sub_kset base put to scx_sched_free_rcu_work
sched_ext: INIT_LIST_HEAD() &sch->all in scx_alloc_and_add_sched()
sched_ext: Drop NONE early return in scx_disable_and_exit_task()
sched_ext: Avoid UAF in scx_root_enable_workfn() init failure path
sched_ext: Clear ops->priv on scx_alloc_and_add_sched() error paths
sched_ext: Fix ops->priv clobber on concurrent attach/detach
selftests/sched_ext: Fix build error in dequeue selftest
sched_ext: Handle SCX_TASK_NONE in disable/switched_from paths
sched_ext: Close sub-sched init race with post-init DEAD recheck
sched_ext: Close root-enable vs sched_ext_dead() race with SCX_TASK_INIT_BEGIN
sched_ext: Replace SCX_TASK_OFF_TASKS flag with SCX_TASK_DEAD state
sched_ext: Inline scx_init_task() and move RESET_RUNNABLE_AT into scx_set_task_state()
sched_ext: Cleanups in preparation for the SCX_TASK_INIT_BEGIN/DEAD work
sched_ext: Use IRQ_WORK_INIT_HARD() to initialize sch->disable_irq_work
sched_ext: Fix !CONFIG_EXT_SUB_SCHED build warnings
sched_ext: Drop unused scx_find_sub_sched() stub
sched_ext: Move scx_error() out of scx_link_sched()'s lock region
- cpuset fixes:
- Partition invalidation could return CPUs still in use by sibling
partitions, producing overlapping effective_cpus.
- cpuset_can_attach() over-reserved DL bandwidth on moves that stayed
within the same root domain.
- Pending DL migration state leaked into later attaches when a later
can_attach() check failed.
- Reorder PF_EXITING and __GFP_HARDWALL checks so dying tasks can
allocate from any node and exit quickly.
- dmem: propagate -ENOMEM instead of spinning forever when the fallback
pool allocation also fails.
- selftests/cgroup: percpu test error-path leak, bogus numeric
comparison of cpuset strings, and a zero-length read() that silently
passed OOM-kill tests.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCagTkzw4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGR+AAQCcYEGJ+yNAzzrTcY8xy7333rorMckSmZt18jzv
1KSqEQD+KjindGNcWP/meQBPnEjcBjix6i961mgnQ99e/UD2HQ4=
=4pT3
-----END PGP SIGNATURE-----
Merge tag 'cgroup-for-7.1-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup
Pull cgroup fixes from Tejun Heo:
- cpuset fixes:
- Partition invalidation could return CPUs still in use by sibling
partitions, producing overlapping effective_cpus
- cpuset_can_attach() over-reserved DL bandwidth on moves that
stayed within the same root domain
- Pending DL migration state leaked into later attaches when a
later can_attach() check failed
- Reorder PF_EXITING and __GFP_HARDWALL checks so dying tasks can
allocate from any node and exit quickly
- dmem: propagate -ENOMEM instead of spinning forever when the fallback
pool allocation also fails
- selftests/cgroup: percpu test error-path leak, bogus numeric
comparison of cpuset strings, and a zero-length read() that silently
passed OOM-kill tests
* tag 'cgroup-for-7.1-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cgroup/cpuset: Return only actually allocated CPUs during partition invalidation
selftests/cgroup: Fix error path leaks in test_percpu_basic
cgroup/cpuset: Reserve DL bandwidth only for root-domain moves
cgroup/cpuset: Reset DL migration state on can_attach() failure
selftests/cgroup: Fix string comparison in write_test
selftests/cgroup: Fix cg_read_strcmp() empty string comparison
cgroup/dmem: Return -ENOMEM on failed pool preallocation
cgroup/cpuset: move PF_EXITING check before __GFP_HARDWALL in cpuset_current_node_allowed()
scx_enable() refuses to attach a BPF scheduler when isolcpus=domain is
in effect by comparing housekeeping_cpumask(HK_TYPE_DOMAIN) against
cpu_possible_mask.
Since commit 27c3a5967f ("sched/isolation: Convert housekeeping
cpumasks to rcu pointers"), HK_TYPE_DOMAIN's cpumask is RCU protected
and dereferencing it requires either RCU read lock, the cpu_hotplug
write lock, or the cpuset lock; scx_enable() holds none of these, so
booting with isolcpus=domain and attaching any BPF scheduler triggers
the following lockdep splat:
=============================
WARNING: suspicious RCU usage
-----------------------------
kernel/sched/isolation.c:60 suspicious rcu_dereference_check() usage!
1 lock held by scx_flash/281:
#0: ffffffff8379fce0 (update_mutex){+.+.}-{4:4}, at:
bpf_struct_ops_link_create+0x134/0x1c0
Call Trace:
dump_stack_lvl+0x6f/0xb0
lockdep_rcu_suspicious.cold+0x37/0x70
housekeeping_cpumask+0xcd/0xe0
scx_enable.isra.0+0x17/0x120
bpf_scx_reg+0x5e/0x80
bpf_struct_ops_link_create+0x151/0x1c0
__sys_bpf+0x1e4b/0x33c0
__x64_sys_bpf+0x21/0x30
do_syscall_64+0x117/0xf80
entry_SYSCALL_64_after_hwframe+0x77/0x7f
In addition, commit 03ff735101 ("cpuset: Update HK_TYPE_DOMAIN cpumask
from cpuset") made HK_TYPE_DOMAIN include cpuset isolated partitions as
well, which means the current check also rejects BPF schedulers when a
cpuset partition is active. That contradicts the original intent of
commit 9f391f94a1 ("sched_ext: Disallow loading BPF scheduler if
isolcpus= domain isolation is in effect"), which explicitly noted that
cpuset partitions are honored through per-task cpumasks and should not
be rejected.
Switch to housekeeping_enabled(HK_TYPE_DOMAIN_BOOT), which reads only
the housekeeping flag bit (no RCU dereference) and reflects exactly the
boot-time isolcpus= configuration that the error message refers to.
Fixes: 27c3a5967f ("sched/isolation: Convert housekeeping cpumasks to rcu pointers")
Cc: stable@vger.kernel.org # v7.0+
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Frederic Weisbecker <frederic@kernel.org>
scx_sub_enable_workfn() pins parent->kobj before dropping scx_sched_lock,
but that does not pin parent->sub_kset. Concurrent disable can
kset_unregister and free sub_kset before scx_alloc_and_add_sched()
dereferences it.
Split sub_kset teardown: kobject_del() at disable keeps sysfs removal; defer
kobject_put() to scx_sched_free_rcu_work so the memory survives. A racing
child sees state_in_sysfs=0 with valid memory, sysfs_create_dir() fails, and
the existing exit_kind gate in scx_link_sched() turns it away with -ENOENT.
Fixes: 411d3ef1a7 ("sched_ext: Unregister sub_kset on scheduler disable")
Signed-off-by: Tejun Heo <tj@kernel.org>
On scx_link_sched() error paths (parent disabled, hash insert failure),
&sch->all is never added to scx_sched_all. The cleanup path runs
scx_unlink_sched() unconditionally, which calls list_del_rcu(&sch->all) on a
list_head that was never initialized triggering a corruption warning.
Initialize &sch->all.
Fixes: 54be8de423 ("sched_ext: Factor out scx_link_sched() and scx_unlink_sched()")
Signed-off-by: Tejun Heo <tj@kernel.org>
d3e73a0808 ("sched_ext: Handle SCX_TASK_NONE in disable/switched_from
paths") skipped the trailing scx_set_task_sched(p, NULL) on NONE tasks.
After scx_fail_parent() parks a task at NONE/sched=parent and the parent
is later freed via queue_rcu_work() during root_disable, the preserved
p->scx.sched dangles - print_scx_info() from sched_show_task() reads
sch->ops.name from freed memory.
Drop the early return. __scx_disable_and_exit_task() already short-
circuits on NONE and the SUB_INIT block was cleared by
scx_fail_parent()'s earlier call, so clearing p->scx.sched is the only
work left - and the one thing the path actually needs.
v2: Extend the SUB_INIT block comment to note that the flag is only
set on the sub-enable path, so it's always clear on the NONE
re-entry (Andrea).
Fixes: d3e73a0808 ("sched_ext: Handle SCX_TASK_NONE in disable/switched_from paths")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Pull to receive:
9a415cc537 ("sched_ext: Avoid UAF in scx_root_enable_workfn() init failure path")
Conflicts with for-7.2's scx_task_iter_relock() rework. The fix moves
put_task_struct(p) past scx_error(); for-7.2 still has it at the old
position. Resolved by dropping the old one.
Signed-off-by: Tejun Heo <tj@kernel.org>
In scx_root_enable_workfn(), put_task_struct(p) is called before scx_error()
dereferences p->comm and p->pid. If the iterator's reference is the last
drop, the task is freed synchronously and the deref becomes a UAF.
Move put_task_struct() past scx_error().
Reported-by: Sashiko <sashiko-bot@kernel.org>
Closes: https://lore.kernel.org/all/20260511214031.AF5E9C2BCB0@smtp.kernel.org/
Fixes: f0e1a0643a ("sched_ext: Implement BPF extensible scheduler class")
Cc: stable@vger.kernel.org # v6.12+
Signed-off-by: Tejun Heo <tj@kernel.org>
Mark !CONFIG_EXT_SUB_SCHED dummy stubs static inline to avoid
-Wunused-function in configs without callers. No functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
cpuset_can_attach() currently adds the bandwidth of all migrating
SCHED_DEADLINE tasks to sum_migrate_dl_bw. If the source and destination
cpuset effective CPU masks do not overlap, the whole sum is then
reserved in the destination root domain.
set_cpus_allowed_dl(), however, subtracts bandwidth from the source
root domain only when the affinity change really moves the task between
root domains. A DL task can move between cpusets that are still in the
same root domain, so including that task in sum_migrate_dl_bw can reserve
destination bandwidth without a matching source-side subtraction.
Share the root-domain move test with set_cpus_allowed_dl(). Keep
nr_migrate_dl_tasks counting all migrating deadline tasks for cpuset DL
task accounting, but add to sum_migrate_dl_bw only for tasks that need a
root-domain bandwidth move. Keep using the destination cpuset effective
CPU mask and leave the broader can_attach()/attach() transaction model
unchanged.
Fixes: 2ef269ef1a ("cgroup/cpuset: Free DL BW in case can_attach() fails")
Cc: stable@vger.kernel.org # v6.10+
Signed-off-by: Guopeng Zhang <zhangguopeng@kylinos.cn>
Reviewed-by: Waiman Long <longman@redhat.com>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Tested-by: Juri Lelli <juri.lelli@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
The tryget_task_struct() calls in scx_sub_disable(),
scx_root_enable_workfn() and scx_sub_enable_workfn() can never fail at
the points they're invoked:
- scx_root_enable_workfn() iterates over scx_tasks under scx_tasks_lock
and rq lock. sched_ext_dead() removes tasks from scx_tasks under the
same scx_tasks_lock before put_task_struct_rcu_user() runs in
finish_task_switch(). So any task observed in scx_tasks must have
usage > 0; put_task_struct_rcu_user() hasn't been called and the
delayed_put_task_struct() callback that decrements usage cannot have
been queued.
- scx_sub_disable() and scx_sub_enable_workfn() iterate via
css_task_iter, which takes a reference on each task in
css_task_iter_next() and holds it until the next iter_next() call, so
usage > 0 is guaranteed by the iter itself.
The actual filter for dead tasks is the SCX_TASK_DEAD check inside
scx_task_iter_next_locked(), not tryget; tryget only fails on zero
usage, a state that can't be reached for tasks visible to these iters.
Commit b7d4b28db7 ("sched_ext: Use SCX_TASK_READY test instead of
tryget_task_struct() during class switch") removed an analogous tryget
in the class-switch loop. Convert the remaining tryget calls to plain
get_task_struct() and update the comment in scx_root_enable_workfn()
that suggested tasks could be observed with zero @usage waiting for an
RCU grace period.
Link: https://lore.kernel.org/all/agCLBxHEUqWIepx8@google.com
Suggested-by: Alice Ryhl <aliceryhl@google.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_alloc_and_add_sched() can fail after @sch has been assigned to
ops->priv. In those cases @sch is torn down (either via kfree() through
the err_free_* chain or via kobject_put() -> scx_kobj_release() -> RCU
work), but @ops->priv is left pointing at the about-to-be-freed pointer.
With the recent -EBUSY gate in scx_root_enable_workfn() and
scx_sub_enable_workfn() that rejects an attach when @ops->priv is still
non-NULL, see commit bbf30b383c ("sched_ext: Fix ops->priv clobber on
concurrent attach/detach"), a dangling @ops->priv permanently locks the
kdata out: every future attach attempt sees a stale binding and returns
-EBUSY even though no scheduler is actually attached.
Clear @ops->priv on the post-assign failure paths so that the kdata
returns to its pre-attach state when the function returns ERR_PTR().
Fixes: bbf30b383c ("sched_ext: Fix ops->priv clobber on concurrent attach/detach")
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Under heavy concurrent attach/detach operations, scx_claim_exit() can
trigger a NULL pointer dereference. This can be reproduced running the
reload_loop kselftests inside a virtme-ng session:
$ vng -v -- ./tools/testing/selftests/sched_ext/runner -t reload_loop
...
BUG: kernel NULL pointer dereference, address: 0000000000000400
RIP: 0010:scx_claim_exit+0x3b/0x120
Call Trace:
<TASK>
bpf_scx_unreg+0x45/0xb0
bpf_struct_ops_map_link_dealloc+0x39/0x50
bpf_link_release+0x18/0x20
__fput+0x10b/0x2e0
__x64_sys_close+0x47/0xa0
The underlying race (diagnosed by Tejun Heo) is a stomp of @ops->priv,
not a missing NULL check:
T2 unreg(K) T1 reg(K)
----------- ---------
sch = ops->priv = sch_b800
scx_disable; flush_disable_work
[scx_root_disable: scx_root=NULL,
mutex_unlock, state=DISABLED]
mutex_lock; state ok
scx_alloc_and_add_sched:
ops->priv = sch_a800
scx_root = sch_a800; init=0
state=ENABLED; mutex_unlock
[flush returns]
RCU_INIT_POINTER(ops->priv, NULL) <-- clobbers sch_a800
kobject_put(sch_b800)
T1 acquires scx_enable_mutex inside scx_root_disable()'s mutex_unlock
window and starts a fresh attach on the same kdata, assigning sch_a800
to @ops->priv. T2 then continues out of scx_disable()/flush_disable_work
and clobbers @ops->priv to NULL, leaking sch_a800; the bpf_link is gone
but state stays SCX_ENABLED, so all future attaches fail with -EBUSY
permanently. The next bpf_scx_unreg() on that kdata then reads NULL
@ops->priv and dereferences it in scx_claim_exit().
Make @ops->priv the lifecycle binding: in scx_root_enable_workfn() and
scx_sub_enable_workfn(), after the existing state check and still under
scx_enable_mutex, refuse with -EBUSY if @ops->priv is non-NULL. This
rejects an attempt to reuse a kdata that is still bound to a previous
scheduler instance, closing the race without changing the unreg side.
Fixes: 105dcd005b ("sched_ext: Introduce scx_prog_sched()")
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Conflict between:
[1] 41e3312861 ("sched_ext: add p->scx.tid and SCX_OPS_TID_TO_TASK lookup")
[2] c941d7391f ("sched_ext: Close root-enable vs sched_ext_dead() race with SCX_TASK_INIT_BEGIN")
in scx_root_enable_workfn()'s post-init block. [1] added a tid hash
insertion under a scoped_guard() for scx_tasks_lock; [2] wraps the same
region in task_rq_lock() for a DEAD recheck. A naive merge would invert the
iter's outer/inner order.
[3] f25ad1e3cb ("sched_ext: Add scx_task_iter_relock() and use it in scx_root_enable_workfn()")
was added to for-7.2 for a clean resolution: scx_task_iter_relock(iter, p)
takes both scx_tasks_lock and @p's rq lock in iter order.
Resolved by routing both sides through [3]'s dual-lock helper: the post-init
region runs under a single scx_task_iter_relock() acquisition, with [2]'s
state machine and [1]'s hash insert in sequence inside it.
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_root_enable_workfn()'s post-init block re-acquires scx_tasks_lock
briefly via a scoped_guard() for the tid hash insertion. c941d7391f
("sched_ext: Close root-enable vs sched_ext_dead() race with
SCX_TASK_INIT_BEGIN") on for-7.1-fixes adds a post-init DEAD recheck that
holds the task's rq lock across the state-machine updates in the same
region. A naive merge would acquire scx_tasks_lock while the rq lock is
held, inverting the iter's outer/inner order (scx_tasks_lock then rq lock).
Add scx_task_iter_relock(iter, p), the counterpart to
scx_task_iter_unlock(), that re-acquires scx_tasks_lock and, if @p is
non-NULL, @p's rq lock. The locks are tracked in @iter so subsequent
iteration releases them.
Use it in scx_root_enable_workfn()'s post-init block and drop the
now-redundant scoped_guard on the hash insertion. The post-init region now
runs with both scx_tasks_lock and the task's rq lock held across the init
failure check, the state-machine updates and the hash insert.
v2: Move scx_task_iter_relock() earlier to ease the for-7.1-fixes merge.
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_fail_parent() leaves cgroup tasks at (state=NONE, sched=parent,
sched_class=ext) until the parent itself is torn down by the scx_error() it
raised. When the later root_disable iterates them, two paths trip on NONE.
scx_disable_and_exit_task() re-enters the wrapper at NONE: the inner switch
returns early but the trailing scx_set_task_sched(p, NULL) clobbers the
parent sched left by scx_fail_parent(), and scx_set_task_state(p, NONE)
wastes a write on an already-NONE task. switched_from_scx() then calls
scx_disable_task(), which WARNs on non-ENABLED state and writes state=READY,
producing a NONE -> READY transition the validation matrix rejects.
Treat NONE as "nothing to do" in both paths. Add a NONE early-return at the
top of scx_disable_and_exit_task() and a parallel NONE check in
switched_from_scx() next to task_dead_and_done().
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_sub_enable_workfn()'s init pass and scx_sub_disable() migration both
drop the rq lock to call __scx_init_task() against the other sched. A
TASK_DEAD @p can fall through sched_ext_dead() in that window.
sched_ext_dead() runs ops.exit_task() on the sched @p was attached to, not
on the sched whose init just completed, so the new allocation leaks.
Reuse the DEAD signal set by sched_ext_dead(). After __scx_init_task()
returns, take task_rq_lock(p) and check for DEAD; on hit, call
scx_sub_init_cancel_task() against the sub sched the init ran for and drop
@p; on miss, proceed as before.
Reported-by: zhidao su <suzhidao@xiaomi.com>
Link: https://lore.kernel.org/all/20260429133155.3825247-1-suzhidao@xiaomi.com/
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_root_enable_workfn() drops the iter rq lock for ops.init_task() and a
TASK_DEAD @p can fall through sched_ext_dead() in that window. The race hits
when sched_ext_dead() observes SCX_TASK_INIT (the intermediate state before
@p->scx.sched is published) and dereferences NULL via SCX_HAS_OP(NULL,
exit_task), or observes SCX_TASK_NONE during the unlocked init window and
skips cleanup so exit_task() never runs.
Add SCX_TASK_INIT_BEGIN. The enable path writes NONE -> INIT_BEGIN under the
iter rq lock, then takes the rq lock again after init to walk INIT_BEGIN ->
INIT -> READY. sched_ext_dead() that wins the rq-lock race observes
INIT_BEGIN and sets DEAD without calling into ops; the post-init recheck
unwinds via scx_sub_init_cancel_task().
scx_fork() runs single-threaded against sched_ext_dead() (the task is not on
scx_tasks until scx_post_fork() adds it) so its INIT_BEGIN -> INIT walk
needs no rq-lock pairing; it rolls back to NONE on ops.init_task() failure.
The validation matrix grows the INIT_BEGIN row and the INIT_BEGIN -> DEAD
edge; INIT now requires INIT_BEGIN as the predecessor. scx_sub_disable()'s
migration writes INIT_BEGIN as a synthetic predecessor to satisfy the
tightened verification.
The sub-sched paths still race with sched_ext_dead() during the unlocked
init window. This will be fixed by the next patch.
Reported-by: zhidao su <suzhidao@xiaomi.com>
Link: https://lore.kernel.org/all/20260429133155.3825247-1-suzhidao@xiaomi.com/
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
SCX_TASK_OFF_TASKS marked tasks already through sched_ext_dead() so cgroup
task iteration would skip them. This can be expressed better with a task
state. Replace the flag with SCX_TASK_DEAD.
scx_disable_and_exit_task() resets state to NONE on its way out, so
sched_ext_dead() now sets DEAD after the wrapper returns. The validation
matrix grows NONE -> DEAD, warns on DEAD -> NONE, and tightens READY's
predecessor to INIT or ENABLED so the new DEAD value cannot silently
transition to READY.
Prepares for the following enable vs dead race fix.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Prepare for the SCX_TASK_INIT_BEGIN/DEAD work that follows by collapsing the
scx_init_task() helper. Move the SCX_TASK_RESET_RUNNABLE_AT setting into
scx_set_task_state() on the INIT transition (it was set unconditionally at
every INIT site through the scx_init_task() helper), inline scx_init_task()
into scx_fork() and scx_root_enable_workfn(), and drop the helper.
As a side effect, scx_sub_disable() migration sequence now also sets
RESET_RUNNABLE_AT (it previously wrote INIT directly without going through
scx_init_task()). The flag triggers a runnable_at reset on the next
set_task_runnable(), which is harmless on a task that has just been moved
between scheds.
On root-enable, p->scx.flags is written without the task's rq lock. The task
isn't visible to scx yet, and a follow-up patch restores the lock-held
write.
v2: Note p->scx.flags rq-lock relaxation on root-enable path. (Andrea)
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Cleanups in preparation for the state-machine work that follows:
- Convert three sub-sched call sites that open-code
rcu_assign_pointer(p->scx.sched, ...) to scx_set_task_sched().
- Move scx_get_task_state()/scx_set_task_state() above the SCX task iter
section so scx_task_iter_next_locked() can use them without a forward
declaration.
No functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
ca1d48a86f ("sched_ext: Use offsetofend on both sides of the ops_cid
layout assert") replaced sizeof() with offsetofend() to dodge 32-bit
PPC trailing padding, but the resulting check is tautological: with
CID_OFFSET_MATCH(priv, priv) already enforcing offsetof(priv) equality
and @priv being the same type in both structs, the two offsetofends
are equal by construction. The original protection - catching a stray
field added past @priv in sched_ext_ops_cid - is gone.
Anchor on a zero-size __end[] marker appended after @priv. Its offset
sits flush after @priv regardless of trailing struct padding; if a
field is inserted past @priv, __end shifts and the assert fires.
Closes: https://lore.kernel.org/all/20260508215211.0C03AC2BCB0@smtp.kernel.org/
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
- Fix spurious failures in rseq self-tests (Mark Brown)
- Fix rseq rseq::cpu_id_start ABI regression due to
TCMalloc's creative use of the supposedly read-only field.
The fix is to introduce a new ABI variant based on
a new (larger) rseq area registration size, to keep
the TCMalloc use of rseq backwards compatible on new kernels.
(Thomas Gleixner)
- Fix wakeup_preempt_fair() for not waking up task (Vincent Guittot)
- Fix s64 mult overflow in vruntime_eligible() (Zhan Xusheng)
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=2QJp
-----END PGP SIGNATURE-----
Merge tag 'sched-urgent-2026-05-09' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
- Fix spurious failures in rseq self-tests (Mark Brown)
- Fix rseq rseq::cpu_id_start ABI regression due to TCMalloc's creative
use of the supposedly read-only field
The fix is to introduce a new ABI variant based on a new (larger)
rseq area registration size, to keep the TCMalloc use of rseq
backwards compatible on new kernels (Thomas Gleixner)
- Fix wakeup_preempt_fair() for not waking up task (Vincent Guittot)
- Fix s64 mult overflow in vruntime_eligible() (Zhan Xusheng)
* tag 'sched-urgent-2026-05-09' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Fix wakeup_preempt_fair() for not waking up task
sched/fair: Fix overflow in vruntime_eligible()
selftests/rseq: Expand for optimized RSEQ ABI v2
rseq: Reenable performance optimizations conditionally
rseq: Implement read only ABI enforcement for optimized RSEQ V2 mode
selftests/rseq: Validate legacy behavior
selftests/rseq: Make registration flexible for legacy and optimized mode
selftests/rseq: Skip tests if time slice extensions are not available
rseq: Revert to historical performance killing behaviour
rseq: Don't advertise time slice extensions if disabled
rseq: Protect rseq_reset() against interrupts
rseq: Set rseq::cpu_id_start to 0 on unregistration
selftests/rseq: Don't run tests with runner scripts outside of the scripts
sizeof() includes trailing struct pad, offsetofend() doesn't. On
32-bit PPC, sched_ext_ops_cid tail-pads 4 bytes past @priv and the
assert trips. Use offsetofend() on both sides.
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202605081637.DbH4SZ1E-lkp@intel.com/
Fixes: 7e655ed7b9 ("sched_ext: Add bpf_sched_ext_ops_cid struct_ops type")
Signed-off-by: Tejun Heo <tj@kernel.org>
For built with PREEMPT_RT kernels, the scx_disable_irq_workfn() is
called from per-cpu irq_work kthreads context, this means that
when call the scx_dump_state() in the scx_disable_irq_workfn() to
output current->comm/pid, it always output current irq_work kthread's
comm/pid. this commit therefore use the IRQ_WORK_INIT_HARD() to
initialize sch->disable_irq_work to make scx_disable_irq_workfn() is
called from hardirq context.
Fixes: f4a6c506d1 ("sched_ext: Always bounce scx_disable() through irq_work")
Signed-off-by: Zqiang <qiang.zhang@linux.dev>
Signed-off-by: Tejun Heo <tj@kernel.org>
W=1 with CONFIG_EXT_SUB_SCHED=n flags 'err_msg' uninitialized and
'err_free_lb_resched' unused. Initialize err_msg and gate the label.
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_find_sub_sched()'s only caller, scx_bpf_sub_dispatch(), is gated on
CONFIG_EXT_SUB_SCHED. When CONFIG_EXT_SUB_SCHED=n the caller compiles out
and the stub becomes dead code, tripping -Wunused-function on randconfigs.
Drop the stub.
Fixes: 25037af712 ("sched_ext: Add rhashtable lookup for sub-schedulers")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/all/202605080556.42PXw8U9-lkp@intel.com/
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_link_sched() holds scx_sched_lock. The scx_error() calls inside take the
same lock through scx_claim_exit() and deadlock. Move them out of the guard.
Fixes: 6b4576b097 ("sched_ext: Reject sub-sched attachment to a disabled parent")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Make sure to only call pick_next_entity() on an non-empty cfs_rq.
The assumption that p is always enqueued and not delayed, is only true for
wakeup. If p was moved while delayed, pick_next_entity() will dequeue it and
the cfs might become empty. Test if there are still queued tasks before trying
again to determine if p could be the next one to be picked.
There are at least 2 cases:
When cfs becomes idle, it tries to pull tasks but if those pulled tasks are
delayed, they will be dequeued when attached to cfs. attach_tasks() ->
attach_task() -> wakeup_preempt(rq, p, 0);
A misfit task running on cfs A triggers a load balance to be pulled on a better
cpu, the load balance on cfs B starts an active load balance to pulled the
running misfit task. If there is a delayed dequeue task on cfs A, it can be
pulled instead of the previously running misfit task. attach_one_task() ->
attach_task() -> wakeup_preempt(rq, p, 0);
Fixes: ac8e69e693 ("sched/fair: Fix wakeup_preempt_fair() vs delayed dequeue")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260503104503.1732682-1-vincent.guittot@linaro.org
Zhan Xusheng reported running into sporadic a s64 mult overflow in
vruntime_eligible().
When constructing a worst case scenario:
If you have cgroups, then you can have an entity of weight 2 (per
calc_group_shares()), and its vlag should then be bounded by: (slice+TICK_NSEC)
* NICE_0_LOAD, which is around 44 bits as per the comment on entity_key().
The other extreme is 100*NICE_0_LOAD, thus you get:
{key, weight}[] := {
puny: { (slice + TICK_NSEC) * NICE_0_LOAD, 2 },
max: { 0, 100*NICE_0_LOAD },
}
The avg_vruntime() would end up being very close to 0 (which is
zero_vruntime), so no real help making that more accurate.
vruntime_eligible(puny) ends up with:
avg = 2 * puny.key (+ 0)
load = 2 + 100 * NICE_0_LOAD
avg >= puny.key * load
And that is: (slice + TICK_NSEC) * NICE_0_LOAD * NICE_0_LOAD * 100, which will
overflow s64.
Zhan suggested using __builtin_mul_overflow(), however after staring at
compiler output for various architectures using godbolt, it seems that using an
__int128 multiplication often results in better code.
Specifically, a number of architectures already compute the __int128 product to
determine the overflow. Eg. arm64 already has the 'smulh' instruction used. By
explicitly doing an __int128 multiply, it will emit the 'mul; smulh' pattern,
which modern cores can fuse (armv8-a clang-22.1.0). x86_64 has less branches
(no OF handling).
Since Linux has ARCH_SUPPORTS_INT128 to gate __int128 usage, also provide the
__builtin_mul_overflow() variant as a fallback.
[peterz: Changelog and __int128 bits]
Fixes: 556146ce5e ("sched/fair: Avoid overflow in enqueue_entity()")
Reported-by: Zhan Xusheng <zhanxusheng1024@gmail.com>
Closes: https://patch.msgid.link/20260415145742.10359-1-zhanxusheng%40xiaomi.com
Signed-off-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260505103155.GN3102924%40noisy.programming.kicks-ass.net
- Fix idle CPU selection returning prev_cpu outside the task's cpus_ptr
when the BPF caller's allowed mask was wider. Stable backport.
- Two opposite-direction gaps in scx_task_iter's cgroup-scoped mode
versus the global mode:
- Tasks past exit_signals() are filtered by the cgroup walk but kept by
global. Sub-scheduler enable abort leaked __scx_init_task() state.
Add a CSS_TASK_ITER_WITH_DEAD flag to cgroup's task iterator
(scx_task_iter is its only user) and use it.
- Tasks past sched_ext_dead() are still returned, tripping
WARN_ON_ONCE() in callers or making them touch torn-down state. Mark
and skip under the per-task rq lock.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCafphXA4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGbI/AP4nRHDusUuYDSFBLyHODvLZXfMM3Nb0yzS7euQJ
qvx6OQEA1p5AyRWA2apFvKjjQrl1dOb5vUlro1Fj8VF51X7Spwc=
=olGB
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-7.1-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- Fix idle CPU selection returning prev_cpu outside the task's cpus_ptr
when the BPF caller's allowed mask was wider. Stable backport.
- Two opposite-direction gaps in scx_task_iter's cgroup-scoped mode
versus the global mode:
- Tasks past exit_signals() are filtered by the cgroup walk but kept
by global. Sub-scheduler enable abort leaked __scx_init_task()
state. Add a CSS_TASK_ITER_WITH_DEAD flag to cgroup's task
iterator (scx_task_iter is its only user) and use it.
- Tasks past sched_ext_dead() are still returned, tripping
WARN_ON_ONCE() in callers or making them touch torn-down state.
Mark and skip under the per-task rq lock.
* tag 'sched_ext-for-7.1-rc2-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: idle: Recheck prev_cpu after narrowing allowed mask
sched_ext: Skip past-sched_ext_dead() tasks in scx_task_iter_next_locked()
cgroup, sched_ext: Include exiting tasks in cgroup iter
The recent RSEQ optimization work broke the TCMalloc abuse of the RSEQ ABI
as it not longer unconditionally updates the CPU, node, mm_cid fields,
which are documented as read only for user space. Due to the observed
behavior of the kernel it was possible for TCMalloc to overwrite the
cpu_id_start field for their own purposes and rely on the kernel to update
it unconditionally after each context switch and before signal delivery.
The RSEQ ABI only guarantees that these fields are updated when the data
changes, i.e. the task is migrated or the MMCID of the task changes due to
switching from or to per CPU ownership mode.
The optimization work eliminated the unconditional updates and reduced them
to the documented ABI guarantees, which results in a massive performance
win for syscall, scheduling heavy work loads, which in turn breaks the
TCMalloc expectations.
There have been several options discussed to restore the TCMalloc
functionality while preserving the optimization benefits. They all end up
in a series of hard to maintain workarounds, which in the worst case
introduce overhead for everyone, e.g. in the scheduler.
The requirements of TCMalloc and the optimization work are diametral and
the required work arounds are a maintainence burden. They end up as fragile
constructs, which are blocking further optimization work and are pretty
much guaranteed to cause more subtle issues down the road.
The optimization work heavily depends on the generic entry code, which is
not used by all architectures yet. So the rework preserved the original
mechanism moslty unmodified to keep the support for architectures, which
handle rseq in their own exit to user space loop. That code is currently
optimized out by the compiler on architectures which use the generic entry
code.
This allows to revert back to the original behaviour by replacing the
compile time constant conditions with a runtime condition where required,
which disables the optimization and the dependend time slice extension
feature until the run-time condition can be enabled in the RSEQ
registration code on a per task basis again.
The following changes are required to restore the original behavior, which
makes TCMalloc work again:
1) Replace the compile time constant conditionals with runtime
conditionals where appropriate to prevent the compiler from optimizing
the legacy mode out
2) Enforce unconditional update of IDs on context switch for the
non-optimized v1 mode
3) Enforce update of IDs in the pre signal delivery path for the
non-optimized v1 mode
4) Enforce update of IDs in the membarrier(RSEQ) IPI for the
non-optimized v1 mode
5) Make time slice and future extensions depend on optimized v2 mode
This brings back the full performance problems, but preserves the v2
optimization code and for generic entry code using architectures also the
TIF_RSEQ optimization which avoids a full evaluation of the exit to user
mode loop in many cases.
Fixes: 566d8015f7 ("rseq: Avoid CPU/MM CID updates when no event pending")
Reported-by: Mathias Stearn <mathias@mongodb.com>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Tested-by: Dmitry Vyukov <dvyukov@google.com>
Closes: https://lore.kernel.org/CAHnCjA25b+nO2n5CeifknSKHssJpPrjnf+dtr7UgzRw4Zgu=oA@mail.gmail.com
Link: https://patch.msgid.link/20260428224427.517051752%40kernel.org
Cc: stable@vger.kernel.org
The wrapper functions __trace_set_current_state() and
__trace_set_need_resched() allow the tracepoints to be called from code
outside sched/core.c, those calls are already guarded by a
tracepoint_enabled(<tp>) so there is no need to repeat this check once
again inside the call using trace_<tp>().
Use the new trace_call__<tp>() API to directly call the tracepoint
without check. Those helper functions must be called after the
appropriate check.
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260429094227.34087-1-gmonaco@redhat.com
Currently, the membarrier system call uses a single global mutex
(`membarrier_ipi_mutex`) to serialize expedited commands. This causes
significant contention on large systems when multiple threads invoke
membarrier concurrently, even if they target different CPUs.
This contention becomes critical when combined with CFS bandwidth
throttling/unthrottling, during which interrupts can be disabled for
relatively long periods on target CPUs. If membarrier is waiting for a
response from such a CPU, it holds the global mutex, blocking all other
membarrier calls on the system. This cascade effect can lead to hard
lockups when thousands of threads stall waiting for the mutex.
Optimize `MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ` when a specific CPU is
targeted by introducing per-CPU mutexes. Broadcast commands and commands
without a specific CPU target continue to use the global mutex.
This prevents the cascade lockup scenario. As measured by the stress test
introduced in the subsequent patch, on an AMD Turin machine with 384 CPUs
(2 NUMA nodes with SMT=2), this optimization yields 200x more
throughput.
Signed-off-by: Aniket Gattani <aniketgattani@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260503212205.3714217-2-aniketgattani@google.com
scx_vexit() forwards (fmt, args) to vscnprintf(); bstr_format() and
__bstr_format() forward fmt to bstr_printf(); the BPF kfunc wrappers
scx_bpf_exit_bstr(), scx_bpf_error_bstr() and scx_bpf_dump_bstr() in
turn forward fmt to those formatters. None of them have __printf(),
so clang -Wmissing-format-attribute fires on the forwarded calls and
C-side callers don't get format-string checking.
Annotate the six functions with __printf(N, 0) matching the fmt
parameter position in each.
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/all/202605041112.Y6OG7v9r-lkp@intel.com/
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_select_cpu_dfl() narrows @allowed to @cpus_allowed & @p->cpus_ptr
when the BPF caller supplies a @cpus_allowed that differs from
@p->cpus_ptr and @p doesn't have full affinity. However,
@is_prev_allowed was computed against the original (wider)
@cpus_allowed, so the prev_cpu fast paths could pick a @prev_cpu that
is in @cpus_allowed but not in @p->cpus_ptr, violating the intended
invariant that the returned CPU is always usable by @p. The kernel
masks this via the SCX_EV_SELECT_CPU_FALLBACK fallback, but the
behavior contradicts the documented contract.
Move the @is_prev_allowed evaluation past the narrowing block so it
tests against the final @allowed mask.
Fixes: ee9a4e9279 ("sched_ext: idle: Properly handle invalid prev_cpu during idle selection")
Cc: stable@vger.kernel.org # v6.16+
Assisted-by: Claude <noreply@anthropic.com>
Signed-off-by: David Carlier <devnexen@gmail.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
sysrq_handle_sched_ext_reset() is called from __handle_sysrq(), which
already holds rcu_read_lock() while invoking the sysrq handler. Remove
the redundant rcu_read_lock/unlock() pair.
Signed-off-by: Zqiang <qiang.zhang@linux.dev>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_task_iter's cgroup-scoped mode can return tasks whose
sched_ext_dead() has already completed: cgroup_task_dead() removes
from cset->tasks after sched_ext_dead() in finish_task_switch() and is
irq-work deferred on PREEMPT_RT. The global mode is fine -
sched_ext_dead() removes from scx_tasks via list_del_init() first.
Callers (sub-sched enable prep/abort/apply, scx_sub_disable(),
scx_fail_parent()) assume returned tasks are still on @sch and trip
WARN_ON_ONCE() or operate on torn-down state otherwise.
Set %SCX_TASK_OFF_TASKS in sched_ext_dead() under @p's rq lock and
have scx_task_iter_next_locked() skip flagged tasks under the same
lock. Setter and reader serialize on the per-task rq lock - no race.
Signed-off-by: Tejun Heo <tj@kernel.org>
a72f73c4dd ("cgroup: Don't expose dead tasks in cgroup") made
css_task_iter_advance() skip exiting tasks so cgroup.procs stays consistent
with waitpid() visibility. Unfortunately, this broke scx_task_iter.
scx_task_iter walks either scx_tasks (global) or a cgroup subtree via
css_task_iter() and the two modes are expected to cover the same set of
tasks. After the above change the cgroup-scoped mode silently skips tasks
past exit_signals() that are still on scx_tasks.
scx_sub_enable_workfn()'s abort path is one of the symptoms: an exiting
SCX_TASK_SUB_INIT task can race past the cgroup iter leaking
__scx_init_task() state. Other iterations share the same gap.
Add CSS_TASK_ITER_WITH_DEAD to opt out of the skip and use it from
scx_task_iter().
Fixes: b0e4c2f8a0 ("sched_ext: Implement cgroup subtree iteration for scx_task_iter")
Reported-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
- Fix the delayed dequeue negative lag increase fix in the
fair scheduler (Peter Zijlstra)
- Fix wakeup_preempt_fair() to do proper delayed dequeue
(Vincent Guittot)
- Clear sched_entity::rel_deadline when initializing
forked entities, which bug can cause all tasks to be
EEVDF-ineligible, causing a NULL pointer dereference
crash in pick_next_entity() (Zicheng Qu)
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----
iQJFBAABCgAvFiEEBpT5eoXrXCwVQwEKEnMQ0APhK1gFAmn2+tsRHG1pbmdvQGtl
cm5lbC5vcmcACgkQEnMQ0APhK1gDdg//WX8E7dyiwRaFNsPMgEvi4Q7pBoXAb5Zi
/iYNlpw1/QRG9KF59175CC3cLOVLJ3dA+79EZPS/mQSuukdTxJg6sbPTkneULV3D
l8DkjH0uDS7mZBqlLDC+Xjqv1y7Y01V6qN9Si8hpR5rP3D0heWdspAGS5aSZ/8Dp
h/VSYl2R615Z32NaS0Xys2hy4M0/I30Uuw4fJScvIYeAMb4s5+6RHEQmuuo25A3O
HQg9Ljqi5NQaHwzvHTYjCenh/NENOd/tu/kZzFgW57HJqSXGM7KBqcjaK68q4vCl
LgBlsfux7RTbnrEAIhGYBSoDss+tBbMKm5qaYZNENJpLS8ptm4J3iKgUJ0W2e3dW
Zp6IjVkj0E+qC65WnFENXsiDr+/fZ9d71/xq2L4z6SxQNv1mtX2+f/HUYyKU5jCc
I4NDEBLGbdRVNuPW7esECVIEVRYFR1cPdZigLW8M7JEnr0p0skF1zgnnMtVuK6Ep
qONYldUIHWdsx67yOSdhykSyq6Qfaew/UKuG1ivlN0BDL4I/AWf+4BVMHJrigeok
xKD8DRWH6s7fSicfM2aJMmj6nRSR2Zz5a9T3lE4LxSvDm41JEnatqpb2Xhjri+I0
21slsm4AZmh1xR1kj7sTKxdHJn0E+lNN/XSZP6OcCoNqlr2XEGMRxWpVo/WqJnO6
HVG9/VoSP1w=
=z+8E
-----END PGP SIGNATURE-----
Merge tag 'sched-urgent-2026-05-03' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
- Fix the delayed dequeue negative lag increase fix in the
fair scheduler (Peter Zijlstra)
- Fix wakeup_preempt_fair() to do proper delayed dequeue
(Vincent Guittot)
- Clear sched_entity::rel_deadline when initializing
forked entities, which bug can cause all tasks to be
EEVDF-ineligible, causing a NULL pointer dereference
crash in pick_next_entity() (Zicheng Qu)
* tag 'sched-urgent-2026-05-03' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Clear rel_deadline when initializing forked entities
sched/fair: Fix wakeup_preempt_fair() vs delayed dequeue
sched/fair: Fix the negative lag increase fix
Sub-scheduler support is tied to the cid-form struct_ops: sub_attach /
sub_detach will communicate allocation via cmask, and the hierarchy assumes
all participants share a single topological cid space. A cpu-form root that
accepts sub-scheds would need cpu <-> cid translation on every cross-sched
interaction, defeating the purpose.
Enforce this at validate_ops():
- A sub-scheduler (scx_parent(sch) non-NULL) must be cid-form.
- A root that exposes sub_attach / sub_detach must be cid-form.
scx_qmap, which is currently the only scheduler demoing sub-sched support,
was converted to cid-form in the preceding patch, so this doesn't cause
breakage.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
cid and cpu are both small s32s, trivially confused when a cid-form
scheduler calls a cpu-keyed kfunc. Reject cid-form programs that
reference any kfunc in the new scx_kfunc_ids_cpu_only at verifier load
time.
The reverse direction is intentionally permissive: cpu-form schedulers
can freely call cid-form kfuncs to ease a gradual cpumask -> cid
migration.
The check sits in scx_kfunc_context_filter() right after the SCX
struct_ops gate and before the any/idle allow and per-op allow-list
checks, so it catches cpu-only kfuncs regardless of which set they
belong to (any, idle, or select_cpu).
v2: Sync per-entry kfunc flags with their primary declarations (Zhao).
pahole intersects flags across BTF_ID_FLAGS() occurrences, so
omitting them drops the flags globally.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Zhao Mengmeng <zhaomengmeng@kylinos.cn>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
cpumask is awkward from BPF and unusable from arena; cid/cmask work in
both. Sub-sched enqueue will need cmask. Without a full cid interface,
schedulers end up mixing forms - a subtle-bug factory.
Add sched_ext_ops_cid, which mirrors sched_ext_ops with cid/cmask
replacing cpu/cpumask in the topology-carrying callbacks.
cpu_acquire/cpu_release are deprecated and absent; a prior patch
moved them past @priv so the cid-form can omit them without
disturbing shared-field offsets.
The two structs share byte-identical layout up to @priv, so the
existing bpf_scx init/check hooks, has_op bitmap, and
scx_kf_allow_flags[] are offset-indexed and apply to both.
BUILD_BUG_ON in scx_init() pins the shared-field and renamed-callback
offsets so any future drift trips at boot.
The kernel<->BPF boundary translates between cpu and cid:
- A static key, enabled on cid-form sched load, gates the translation
so cpu-form schedulers pay nothing.
- dispatch, update_idle, cpu_online/offline and dump_cpu translate
the cpu arg at the callsite.
- select_cpu also translates the returned cid back to a cpu.
- set_cpumask is wrapped to synthesize a cmask in a per-cpu scratch
before calling the cid-form callback.
All scheds in a hierarchy share one form. The static key drives the
hot-path branch.
v2: Use struct_size() for the set_cmask_scratch percpu alloc. Move
cid-shard fields and assertions into the later cid-shard patch.
v3: Drop `static` on scx_set_cmask_scratch; add extern in ext_internal.h.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
cpumask is awkward from BPF and unusable from arena; cid/cmask work in
both. Sub-sched enqueue will need cmask. Without full cid coverage a
scheduler has to mix cid and cpu forms, which is a subtle-bug factory.
Close the gap with a cid-native interface.
Pair every cpu-form kfunc that takes a cpu id with a cid-form
equivalent (kick, task placement, cpuperf query/set, per-cpu current
task, nr-cpu-ids). Add two cid-natives with no cpu-form sibling:
scx_bpf_this_cid() (cid of the running cpu, scx equivalent of
bpf_get_smp_processor_id) and scx_bpf_nr_online_cids().
scx_bpf_cpu_rq is deprecated; no cid-form counterpart. NUMA node info
is reachable via scx_bpf_cid_topo() on the BPF side.
Each cid-form wrapper is a thin cid -> cpu translation that delegates
to the cpu path, registered in the same context sets so usage
constraints match.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Sub-scheduler code built on cids needs bitmaps scoped to a slice of cid
space (e.g. the idle cids of a shard). A cpumask sized for NR_CPUS wastes
most of its bits for a small window and is awkward in BPF.
scx_cmask covers [base, base + nr_bits). bits[] is aligned to the global
64-cid grid: bits[0] spans [base & ~63, (base & ~63) + 64). Any two
cmasks therefore address bits[] against the same global windows, so
cross-cmask word ops reduce to
dest->bits[i] OP= operand->bits[i - delta]
with no bit-shifting, at the cost of up to one extra storage word for
head misalignment. This alignment guarantee is the reason binary ops
can stay word-level; every mutating helper preserves it.
Kernel side in ext_cid.[hc]; BPF side in tools/sched_ext/include/scx/
cid.bpf.h. BPF side drops the scx_ prefix (redundant in BPF code) and
adds the extra helpers that basic idle-cpu selection needs.
No callers yet.
v2: Narrow to helpers that will be used in the planned changes;
set/bit/find/zero ops will be added as usage develops.
v3: cmask_copy_from_kernel: validate src->base == 0 via probe-read;
bit-level nr_bits check instead of round-up word count. (Sashiko)
v4: Bump CMASK_CAS_TRIES to 1<<23 so abort fires only after seconds
of real spinning, not on plausible contention. Switch
__builtin_ctzll() to the ctzll() wrapper for clang compat
(Changwoo).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The auto-probed cid mapping reflects the kernel's view of topology
(node -> LLC -> core), but a BPF scheduler may want a different layout -
to align cid slices with its own partitioning, or to work around how the
kernel reports a particular machine.
Add scx_bpf_cid_override(), callable from ops.init() of the root
scheduler. It validates the caller-supplied cpu->cid array and replaces
the in-place mapping; topo info is invalidated. A compat.bpf.h wrapper
silently no-ops on kernels that lack the kfunc.
A new SCX_KF_ALLOW_INIT bit in the kfunc context filter restricts the
kfunc to ops.init() at verifier load time.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Raw cpu numbers are clumsy for sharding and cross-sched communication,
especially from BPF. The space is sparse, numerical closeness doesn't
track topological closeness (x86 hyperthreading often scatters SMT
siblings), and a range of cpu ids doesn't describe anything meaningful.
Sub-sched support makes this acute: cpu allocation, revocation, and
state constantly flow across sub-scheds. Passing whole cpumasks scales
poorly (every op scans 4K bits) and cpumasks are awkward in BPF.
cids assign every cpu a dense, topology-ordered id. CPUs sharing a core,
LLC, or NUMA node occupy contiguous cid ranges, so a topology unit
becomes a (start, length) slice. Communication passes slices; BPF can
process a u64 word of cids at a time.
Build the mapping once at root enable by walking online cpus node -> LLC
-> core. Possible-but-not-online cpus tail the space with no-topo cids.
Expose kfuncs to map cpu <-> cid in either direction and to query each
cid's topology metadata.
v2: Use kzalloc_objs()/kmalloc_objs() for the three allocs in
scx_cid_arrays_alloc() (Cheng-Yang Chou).
v3: scx_cid_init() failure path now drops cpus_read_lock();
BUILD_BUG_ON tightened to match BPF cmask helpers' NR_CPUS<=8192.
(Sashiko)
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Pass struct scx_enable_cmd to scx_enable() rather than unpacking @ops
at every call site and re-packing into a fresh cmd inside. bpf_scx_reg()
now builds the cmd on its stack and hands it in; scx_enable() just
wires up the kthread work and waits.
Relocate struct scx_enable_cmd above scx_alloc_and_add_sched() so
upcoming patches that also want the cmd can see it.
No behavior change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
cpu_acquire and cpu_release are deprecated and slated for removal. Move
their declarations to the end of struct sched_ext_ops so an upcoming
cid-form struct (sched_ext_ops_cid) can omit them entirely without
disturbing the offsets of the shared fields.
Switch the two SCX_HAS_OP() callers for these ops to direct field checks
since the relocated ops sit outside the SCX_OPI_END range covered by the
has_op bitmap.
scx_kf_allow_flags[] auto-sizes to the highest used SCX_OP_IDX, so
SCX_OP_IDX(cpu_release) moving to a higher index just enlarges the
sparse array; the lookup logic is unchanged.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Callers that already know the cpu is valid shouldn't have to pay for a
redundant check. scx_kick_cpu() is called from the in-kernel balance loop
break-out path with the current cpu (trivially valid) and from
scx_bpf_kick_cpu() with a BPF-supplied cpu that does need validation. Move
the check out of scx_kick_cpu() into scx_bpf_kick_cpu() so the backend is
reusable by callers that have already validated.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Things shared across multiple .c files belong in a header. scx_exit() /
scx_error() (and their scx_vexit() / scx_verror() siblings) are already
called from ext_idle.c and the upcoming ext_cid.c, and it was only
build_policy.c's textual inclusion of ext.c that made the references
resolve. Move the whole family to ext_internal.h.
Pure visibility change.
v4: Rebased over the exit_cpu plumbing. scx_exit() and scx_verror()
are now macros wrapping raw_smp_processor_id(); move both macros
plus the underlying __scx_exit() / scx_vexit() declarations to
the header.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Rename the static ext.c helper and declare it in ext_internal.h so
ext_idle.c and the upcoming cid code can call it directly instead of
relying on build_policy.c textual inclusion.
Pure rename and visibility change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Introduce kernel/sched/ext_types.h as the early-def header for the
sched_ext compilation unit. Included from kernel/sched/build_policy.c
before ext_internal.h so every later header and source in the unit
sees its content without re-inclusion. Later patches add their types
here (struct scx_cid_topo, scx_cmask, scx_cid_shard, etc.) so the
subsystem has one place to stash types shared across the TU.
Move enum scx_consts (SCX_DSP_DFL_MAX_BATCH, SCX_WATCHDOG_MAX_TIMEOUT,
SCX_SUB_MAX_DEPTH, etc.) here as the initial content. Ops-facing
content stays in ext_internal.h.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
When sched_ext is disabled by an error, the CPU that triggered the exit
is the most relevant piece of information for diagnosing the problem.
However, if there are many CPUs, the dump can get truncated and that
CPU's information may not appear in the output.
Add an exit_cpu field to scx_exit_info and thread it through scx_vexit()
/ __scx_exit(). For the watchdog stall path, populate it from cpu_of(rq)
in check_rq_for_timeouts(). For all other exit paths, define a scx_exit()
macro that wraps __scx_exit() with raw_smp_processor_id(), so the CPU
that initiated the exit is captured automatically, with no call-site
changes needed.
In scx_dump_state(), report the exit CPU in the dump header ("on cpu N")
and dump that CPU first, skipping it in the per-CPU loop, so the most
relevant CPU is never truncated out of the dump. The SysRq-D path
initializes exit_cpu to -1 so debug dumps not tied to an exit don't
arbitrarily promote CPU 0.
Signed-off-by: Changwoo Min <changwoo@igalia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Factor out the per-CPU state dump logic from the for_each_possible_cpu
loop in scx_dump_state() into a new scx_dump_cpu() helper to improve
readability. No functional change.
Signed-off-by: Changwoo Min <changwoo@igalia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
The merge window pulled in the cgroup sub-scheduler infrastructure, and
new AI reviews are accelerating bug reporting and fixing - hence the
larger than usual fixes batch.
- Use-after-frees during scheduler load/unload. The disable path
could free the BPF scheduler while deferred irq_work / kthread work
was still in flight; cgroup setter callbacks read the active
scheduler outside the rwsem that synchronizes against teardown.
Fixed both, and reused the disable drain in the enable error paths
so the BPF JIT page can't be freed under live callbacks.
- Several BPF op invocations didn't tell the framework which runqueue
was already locked, so helper kfuncs that re-acquire the runqueue
by CPU could deadlock on the held lock. Fixed at the affected
callsites, including recursive parent-into-child dispatch.
- The hardlockup notifier ran from NMI but eventually took a
non-NMI-safe lock. Bounced through irq_work.
- A handful of bugs in the new sub-scheduler hierarchy: helper
kfuncs hard-coded the root instead of resolving the caller's
scheduler; the enable error path tried to disable per-task state
that had never been initialized, and leaked cpus_read_lock on the
way out; a sysfs object was leaked on every load/unload; the
dispatch fast-path used the root scheduler instead of the task's;
and a couple of CONFIG #ifdef guards were misclassified.
- Verifier-time hardening: BPF programs of unrelated struct_ops
types (e.g. tcp_congestion_ops) could call sched_ext kfuncs - a
semantic bug and, once sub-sched was enabled, a KASAN
out-of-bounds read. Now rejected at load. Plus a few NULL and
cross-task argument checks on sched_ext kfuncs, and a selftest
covering the new deny.
- rhashtable (Herbert): restored the insecure_elasticity toggle and
bounced the deferred-resize kick through irq_work to break a
lock-order cycle observable from raw-spinlock callers. sched_ext's
scheduler-instance hash is the first user of both.
- The bypass-mode load balancer used file-scope cpumasks; with
multiple scheduler instances now possible, those raced. Moved
per-instance, plus a follow-up to skip tasks whose recorded CPU is
stale relative to the new owning runqueue.
- Smaller fixes: a dispatch queue's first-task tracking misbehaved
when a parked iterator cursor sat in the list; the runqueue's
next-class wasn't promoted on local-queue enqueue, leaving an SCX
task behind RT in edge cases; the reference qmap scheduler stopped
erroring on legitimate cross-scheduler task-storage misses.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCafEN/A4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGaydAQCxWrUqnZXhxF4LnztjxTF2tgv7p8P7TbpS6aU6
etqRpAEA9RFmIXs7XrhwCm0n2BwSgjvrNxnWfPhWvuH0uN0GTAA=
=wLna
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-7.1-rc1-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
"The merge window pulled in the cgroup sub-scheduler infrastructure,
and new AI reviews are accelerating bug reporting and fixing - hence
the larger than usual fixes batch:
- Use-after-frees during scheduler load/unload:
- The disable path could free the BPF scheduler while deferred
irq_work / kthread work was still in flight
- cgroup setter callbacks read the active scheduler outside the
rwsem that synchronizes against teardown
Fix both, and reuse the disable drain in the enable error paths so
the BPF JIT page can't be freed under live callbacks.
- Several BPF op invocations didn't tell the framework which runqueue
was already locked, so helper kfuncs that re-acquire the runqueue
by CPU could deadlock on the held lock
Fix the affected callsites, including recursive parent-into-child
dispatch.
- The hardlockup notifier ran from NMI but eventually took a
non-NMI-safe lock. Bounce it through irq_work.
- A handful of bugs in the new sub-scheduler hierarchy:
- helper kfuncs hard-coded the root instead of resolving the
caller's scheduler
- the enable error path tried to disable per-task state that had
never been initialized, and leaked cpus_read_lock on the way
out
- a sysfs object was leaked on every load/unload
- the dispatch fast-path used the root scheduler instead of the
task's
- a couple of CONFIG #ifdef guards were misclassified
- Verifier-time hardening: BPF programs of unrelated struct_ops types
(e.g. tcp_congestion_ops) could call sched_ext kfuncs - a semantic
bug and, once sub-sched was enabled, a KASAN out-of-bounds read.
Now rejected at load. Plus a few NULL and cross-task argument
checks on sched_ext kfuncs, and a selftest covering the new deny.
- rhashtable (Herbert): restore the insecure_elasticity toggle and
bounce the deferred-resize kick through irq_work to break a
lock-order cycle observable from raw-spinlock callers. sched_ext's
scheduler-instance hash is the first user of both.
- The bypass-mode load balancer used file-scope cpumasks; with
multiple scheduler instances now possible, those raced. Move to
per-instance cpumasks, plus a follow-up to skip tasks whose
recorded CPU is stale relative to the new owning runqueue.
- Smaller fixes:
- a dispatch queue's first-task tracking misbehaved when a parked
iterator cursor sat in the list
- the runqueue's next-class wasn't promoted on local-queue
enqueue, leaving an SCX task behind RT in edge cases
- the reference qmap scheduler stopped erroring on legitimate
cross-scheduler task-storage misses"
* tag 'sched_ext-for-7.1-rc1-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext: (26 commits)
sched_ext: Fix scx_flush_disable_work() UAF race
sched_ext: Call wakeup_preempt() in local_dsq_post_enq()
sched_ext: Release cpus_read_lock on scx_link_sched() failure in root enable
sched_ext: Reject NULL-sch callers in scx_bpf_task_set_slice/dsq_vtime
sched_ext: Refuse cross-task select_cpu_from_kfunc calls
sched_ext: Align cgroup #ifdef guards with SUB_SCHED vs GROUP_SCHED
sched_ext: Make bypass LB cpumasks per-scheduler
sched_ext: Pass held rq to SCX_CALL_OP() for core_sched_before
sched_ext: Pass held rq to SCX_CALL_OP() for dump_cpu/dump_task
sched_ext: Save and restore scx_locked_rq across SCX_CALL_OP
sched_ext: Use dsq->first_task instead of list_empty() in dispatch_enqueue() FIFO-tail
sched_ext: Resolve caller's scheduler in scx_bpf_destroy_dsq() / scx_bpf_dsq_nr_queued()
sched_ext: Read scx_root under scx_cgroup_ops_rwsem in cgroup setters
sched_ext: Don't disable tasks in scx_sub_enable_workfn() abort path
sched_ext: Skip tasks with stale task_rq in bypass_lb_cpu()
sched_ext: Guard scx_dsq_move() against NULL kit->dsq after failed iter_new
sched_ext: Unregister sub_kset on scheduler disable
sched_ext: Defer scx_hardlockup() out of NMI
sched_ext: sync disable_irq_work in bpf_scx_unreg()
sched_ext: Fix local_dsq_post_enq() to use task's scheduler in sub-sched
...
scx_flush_disable_work() calls irq_work_sync() followed by
kthread_flush_work() to ensure that the disable kthread work has
fully completed before bpf_scx_unreg() frees the SCX scheduler.
However, a concurrent scx_vexit() (e.g., triggered by a watchdog stall)
creates a race window between scx_claim_exit() and irq_work_queue():
CPU A (scx_vexit (watchdog)) CPU B (bpf_scx_unreg)
---- ----
scx_claim_exit()
atomic_try_cmpxchg(NONE->kind)
stack_trace_save()
vscnprintf()
scx_disable()
scx_claim_exit() -> FAIL
scx_flush_disable_work()
irq_work_sync() // no-op: not queued yet
kthread_flush_work() // no-op: not queued yet
kobject_put(&sch->kobj) -> free %sch
irq_work_queue() -> UAF on %sch
scx_disable_irq_workfn()
kthread_queue_work() -> UAF
The root cause is that CPU B's scx_flush_disable_work() returns after
syncing an irq_work that has not yet been queued, while CPU A is still
executing the code between scx_claim_exit() and irq_work_queue().
Loop until exit_kind reaches SCX_EXIT_DONE or SCX_EXIT_NONE, draining
disable_irq_work and disable_work in each pass. This ensures that any
work queued after the previous check is caught, while also correctly
handling cases where no disable was triggered (e.g., the
scx_sub_enable_workfn() abort path).
Fixes: 510a270554 ("sched_ext: sync disable_irq_work in bpf_scx_unreg()")
Reported-by: https://sashiko.dev/#/patchset/20260424100221.32407-1-icheng%40nvidia.com
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Move <linux/btf_ids.h> from ext.c and "ext_idle.h" from ext.c (plus its
self-include in ext_idle.c) into build_policy.c. Subsequent patches add
their headers the same way for consistency.
No functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
There are several edge cases (see linked thread) where an IMMED task
can be left lingering on a local DSQ if an RT task swoops in at the
wrong time. All of these edge cases are due to rq->next_class being idle
even after dispatching a task to rq's local DSQ. We should bump
rq->next_class to &ext_sched_class as soon as we've inserted a task into
the local DSQ.
To optimize the common case of rq->next_class == &ext_sched_class,
only call wakeup_preempt() if rq->next_class is below EXT. If next_class
is EXT or above, wakeup_preempt() is a no-op anyway.
This lets us also simplify the preempt_curr() logic a bit since
wakeup_preempt() will call preempt_curr() for us if next_class is
below EXT.
Link: https://lore.kernel.org/all/DHZPHUFXB4N3.2RY28MUEWBNYK@google.com/
Signed-off-by: Kuba Piecuch <jpiecuch@google.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
A yield-triggered crash can happen when a newly forked sched_entity
enters the fair class with se->rel_deadline unexpectedly set.
The failing sequence is:
1. A task is forked while se->rel_deadline is still set.
2. __sched_fork() initializes vruntime, vlag and other sched_entity
state, but does not clear rel_deadline.
3. On the first enqueue, enqueue_entity() calls place_entity().
4. Because se->rel_deadline is set, place_entity() treats se->deadline
as a relative deadline and converts it to an absolute deadline by
adding the current vruntime.
5. However, the forked entity's deadline is not a valid inherited
relative deadline for this new scheduling instance, so the conversion
produces an abnormally large deadline.
6. If the task later calls sched_yield(), yield_task_fair() advances
se->vruntime to se->deadline.
7. The inflated vruntime is then used by the following enqueue path,
where the vruntime-derived key can overflow when multiplied by the
entity weight.
8. This corrupts cfs_rq->sum_w_vruntime, breaks EEVDF eligibility
calculation, and can eventually make all entities appear ineligible.
pick_next_entity() may then return NULL unexpectedly, leading to a
later NULL dereference.
A captured trace shows the effect clearly. Before yield, the entity's
vruntime was around:
9834017729983308
After yield_task_fair() executed:
se->vruntime = se->deadline
the vruntime jumped to:
19668035460670230
and the deadline was later advanced further to:
19668035463470230
This shows that the deadline had already become abnormally large before
yield_task_fair() copied it into vruntime.
rel_deadline is only meaningful when se->deadline really carries a
relative deadline that still needs to be placed against vruntime. A
freshly forked sched_entity should not inherit or retain this state.
Clear se->rel_deadline in __sched_fork(), together with the other
sched_entity runtime state, so that the first enqueue does not interpret
the new entity's deadline as a stale relative deadline.
Fixes: 82e9d0456e ("sched/fair: Avoid re-setting virtual deadline on 'migrations'")
Analyzed-by: Hui Tang <tanghui20@huawei.com>
Analyzed-by: Zhang Qiao <zhangqiao22@huawei.com>
Signed-off-by: Zicheng Qu <quzicheng@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260424071113.1199600-1-quzicheng@huawei.com
Similar to how pick_next_entity() must dequeue delayed entities, so too must
wakeup_preempt_fair(). Any delayed task being found means it is eligible and
hence past the 0-lag point, ready for removal.
Worse, by not removing delayed entities from consideration, it can skew the
preemption decision, with the end result that a short slice wakeup will not
result in a preemption.
tip/sched/core tip/sched/core +this patch
cyclictest slice (ms) (default)2.8 8 8
hackbench slice (ms) (default)2.8 20 20
Total Samples | 22559 22595 22683
Average (us) | 157 64( 59%) 59( 8%)
Median (P50) (us) | 57 57( 0%) 58(- 2%)
90th Percentile (us) | 64 60( 6%) 60( 0%)
99th Percentile (us) | 2407 67( 97%) 67( 0%)
99.9th Percentile (us) | 3400 2288( 33%) 727( 68%)
Maximum (us) | 5037 9252(-84%) 7461( 19%)
Fixes: f12e148892 ("sched/fair: Prepare pick_next_task() for delayed dequeue")
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260422093400.319251-1-vincent.guittot@linaro.org
Vincent reported that my rework of his original patch lost a little
something.
Specifically it got the return value wrong; it should not compare
against the old se->vlag, but rather against the current value. Since
the thing that matters is if the effective vruntime of an entity is
affected and the thing needs repositioning or not.
Fixes: 059258b0d4 ("sched/fair: Prevent negative lag increase during delayed dequeue")
Reported-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://patch.msgid.link/20260423094107.GT3102624%40noisy.programming.kicks-ass.net
Pull to receive:
c0e8ddc76d ("sched_ext: Align cgroup #ifdef guards with SUB_SCHED vs GROUP_SCHED")
which conflicts with:
41e3312861 ("sched_ext: add p->scx.tid and SCX_OPS_TID_TO_TASK lookup")
It's a simple context conflict. Take changes from both.
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_root_enable_workfn() takes cpus_read_lock() before
scx_link_sched(sch), but the `if (ret) goto err_disable` on failure
skips the matching cpus_read_unlock() - all other err_disable gotos
along this path drop the lock first.
scx_link_sched() only returns non-zero on the sub-sched path
(parent != NULL), so the leak path is unreachable via the root
caller today. Still, the unwind is out of line with the surrounding
paths.
Drop cpus_read_lock() before goto err_disable.
v2: Correct Fixes: tag (Andrea Righi).
Fixes: 25037af712 ("sched_ext: Add rhashtable lookup for sub-schedulers")
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_prog_sched(aux) returns NULL for TRACING / SYSCALL BPF progs that
have no struct_ops association when the root scheduler has sub_attach
set. scx_bpf_task_set_slice() and scx_bpf_task_set_dsq_vtime() pass
that NULL into scx_task_on_sched(sch, p), which under
CONFIG_EXT_SUB_SCHED is rcu_access_pointer(p->scx.sched) == sch. For
any non-scx task p->scx.sched is NULL, so NULL == NULL returns true
and the authority gate is bypassed - a privileged but
non-struct_ops-associated prog can poke p->scx.slice /
p->scx.dsq_vtime on arbitrary tasks.
Reject !sch up front so the gate only admits callers with a resolved
scheduler.
Fixes: 245d09c594 ("sched_ext: Enforce scheduler ownership when updating slice and dsq_vtime")
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
select_cpu_from_kfunc() skipped pi_lock for @p when called from
ops.select_cpu() or another rq-locked SCX op, assuming the held lock
protects @p. scx_bpf_select_cpu_dfl() / __scx_bpf_select_cpu_and() accept an
arbitrary KF_RCU task_struct, so a caller in e.g. ops.select_cpu(p1) or
ops.enqueue(p1) can pass some other p2 - the held pi_lock / rq lock is p1's,
not p2's - and reading p2->cpus_ptr / nr_cpus_allowed races with
set_cpus_allowed_ptr() and migrate_disable_switch() on another CPU.
Abort the scheduler on cross-task calls in both branches: for
ops.select_cpu() use scx_kf_arg_task_ok() to verify @p is the wake-up
task recorded in current->scx.kf_tasks[] by SCX_CALL_OP_TASK_RET();
for other rq-locked SCX ops compare task_rq(p) against scx_locked_rq().
v2: Switch the in_select_cpu cross-task check from direct_dispatch_task
comparison to scx_kf_arg_task_ok(). The former spuriously rejects when
ops.select_cpu() calls scx_bpf_dsq_insert() first, then calls
scx_bpf_select_cpu_*() on the same task. (Andrea Righi)
Fixes: 0022b32850 ("sched_ext: Decouple kfunc unlocked-context check from kf_mask")
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Andrea Righi <arighi@nvidia.com>
Two EXT_GROUP_SCHED/SUB_SCHED guards are misclassified:
- scx_root_enable_workfn()'s cgroup_get(cgrp) and the err_put_cgrp unwind
in scx_alloc_and_add_sched() are under `#if GROUP || SUB`, but the
matching cgroup_put() in scx_sched_free_rcu_work() is inside `#ifdef SUB`
only (via sch->cgrp, stored only under SUB). GROUP-only would leak a
reference on every root-sched enable.
- sch_cgroup() / set_cgroup_sched() live under `#if GROUP || SUB` but touch
SUB-only fields (sch->cgrp, cgroup->scx_sched). GROUP-only wouldn't
compile.
GROUP needs CGROUP_SCHED; SUB needs only CGROUPS. CGROUPS=y/CGROUP_SCHED=n
gives the reachable GROUP=n, SUB=y combination; GROUP=y, SUB=n isn't
reachable today (SUB is def_bool y under CGROUPS). Neither miscategorization
triggers a real bug in any reachable config, but keep the guards honest:
- Narrow cgroup_get and err_put_cgrp to `#ifdef SUB` (matches the free-side
put).
- Move sch_cgroup() and set_cgroup_sched() to a separate `#ifdef SUB` block
with no-op stubs for the !SUB case; keep root_cgroup() and scx_cgroup_{
lock,unlock}() under `#if GROUP || SUB` since those only need cgroup core.
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_bypass_lb_{donee,resched}_cpumask were file-scope statics shared by all
scheduler instances. With CONFIG_EXT_SUB_SCHED, multiple sched instances
each arm their own bypass_lb_timer; concurrent bypass_lb_node() calls RMW
the global cpumasks with no lock, corrupting donee/resched decisions.
Move the cpumasks into struct scx_sched, allocate them alongside the timer
in scx_alloc_and_add_sched(), free them in scx_sched_free_rcu_work().
Fixes: 95d1df610c ("sched_ext: Implement load balancer for bypass mode")
Cc: stable@vger.kernel.org # v6.19+
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_prio_less() runs from core-sched's pick_next_task() path with rq
locked but invokes ops.core_sched_before() with NULL locked_rq, leaving
scx_locked_rq_state NULL. If the BPF callback calls a kfunc that
re-acquires rq based on scx_locked_rq() - e.g. scx_bpf_cpuperf_set(cpu)
- it re-acquires the already-held rq.
Pass task_rq(a).
Fixes: 7b0888b7cc ("sched_ext: Implement core-sched support")
Cc: stable@vger.kernel.org # v6.12+
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_dump_state() walks CPUs with rq_lock_irqsave() held and invokes
ops.dump_cpu / ops.dump_task with NULL locked_rq, leaving
scx_locked_rq_state NULL. If the BPF callback calls a kfunc that
re-acquires rq based on scx_locked_rq() - e.g. scx_bpf_cpuperf_set(cpu)
- it re-acquires the already-held rq.
Pass the held rq to SCX_CALL_OP(). Thread it into scx_dump_task() too.
The pre-loop ops.dump call runs before rq_lock_irqsave() so keeps
rq=NULL.
Fixes: 07814a9439 ("sched_ext: Print debug dump after an error exit")
Cc: stable@vger.kernel.org # v6.12+
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
SCX_CALL_OP{,_RET}() unconditionally clears scx_locked_rq_state to NULL on
exit. Correct at the top level, but ops can recurse via
scx_bpf_sub_dispatch(): a parent's ops.dispatch calls the helper, which
invokes the child's ops.dispatch under another SCX_CALL_OP. When the inner
call returns, the NULL clobbers the outer's state. The parent's BPF then
calls kfuncs like scx_bpf_cpuperf_set() which read scx_locked_rq()==NULL and
re-acquire the already-held rq.
Snapshot scx_locked_rq_state on entry and restore on exit. Rename the rq
parameter to locked_rq across all SCX_CALL_OP* macros so the snapshot local
can be typed as 'struct rq *' without colliding with the parameter token in
the expansion. SCX_CALL_OP_TASK{,_RET}() and SCX_CALL_OP_2TASKS_RET() funnel
through the two base macros and inherit the fix.
Fixes: 4f8b122848 ("sched_ext: Add basic building blocks for nested sub-scheduler dispatching")
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
dispatch_enqueue()'s FIFO-tail path used list_empty(&dsq->list) to decide
whether to set dsq->first_task on enqueue. dsq->list can contain parked BPF
iterator cursors (SCX_DSQ_LNODE_ITER_CURSOR), so list_empty() is not a
reliable "no real task" check. If the last real task is unlinked while a
cursor is parked, first_task becomes NULL; the next FIFO-tail enqueue then
sees list_empty() == false and skips the first_task update, leaving
scx_bpf_dsq_peek() returning NULL for a non-empty DSQ.
Test dsq->first_task directly, which already tracks only real tasks and is
maintained under dsq->lock.
Fixes: 44f5c8ec5b ("sched_ext: Add lockless peek operation for DSQs")
Cc: stable@vger.kernel.org # v6.19+
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Cc: Ryan Newton <newton@meta.com>
scx_bpf_create_dsq() resolves the calling scheduler via scx_prog_sched(aux)
and inserts the new DSQ into that scheduler's dsq_hash. Its inverse
scx_bpf_destroy_dsq() and the query helper scx_bpf_dsq_nr_queued() were
hard-coded to rcu_dereference(scx_root), so a sub-scheduler could only
destroy or query DSQs in the root scheduler's hash - never its own. If the
root had a DSQ with the same id, the sub-sched silently destroyed it and the
root aborted on the next dispatch ("invalid DSQ ID 0x0..").
Take a const struct bpf_prog_aux *aux via KF_IMPLICIT_ARGS and resolve the
scheduler with scx_prog_sched(aux), matching scx_bpf_create_dsq().
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_group_set_{weight,idle,bandwidth}() cache scx_root before acquiring
scx_cgroup_ops_rwsem, so the pointer can be stale by the time the op runs.
If the loaded scheduler is disabled and freed (via RCU work) and another is
enabled between the naked load and the rwsem acquire, the reader sees
scx_cgroup_enabled=true (the new scheduler's) but dereferences the freed one
- UAF on SCX_HAS_OP(sch, ...) / SCX_CALL_OP(sch, ...).
scx_cgroup_enabled is toggled only under scx_cgroup_ops_rwsem write
(scx_cgroup_{init,exit}), so reading scx_root inside the rwsem read section
correlates @sch with the enabled snapshot.
Fixes: a5bd6ba30b ("sched_ext: Use cgroup_lock/unlock() to synchronize against cgroup operations")
Cc: stable@vger.kernel.org # v6.18+
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_sub_enable_workfn()'s prep loop calls __scx_init_task(sch, p, false)
without transitioning task state, then sets SCX_TASK_SUB_INIT. If prep fails
partway, the abort path runs __scx_disable_and_exit_task(sch, p) on the
marked tasks. Task state is still the parent's ENABLED, so that dispatches
to the SCX_TASK_ENABLED arm and calls scx_disable_task(sch, p) - i.e.
child->ops.disable() - for tasks on which child->ops.enable() never ran. A
BPF sub-scheduler allocating per-task state in enable/freeing in disable
would operate on uninitialized state.
The dying-task branch in scx_disable_and_exit_task() has the same problem,
and scx_enabling_sub_sched was cleared before the abort cleanup loop - a
task exiting during cleanup tripped the WARN and skipped both ops.exit_task
and the SCX_TASK_SUB_INIT clear, leaking per-task resources and leaving the
task stuck.
Introduce scx_sub_init_cancel_task() that calls ops.exit_task with
cancelled=true - matching what the top-level init path does when init_task
itself returns -errno. Use it in the abort loop and in the dying-task
branch. scx_enabling_sub_sched now stays set until the abort loop finishes
clearing SUB_INIT, so concurrent exits hitting the dying-task branch can
still find @sch. That branch also clears SCX_TASK_SUB_INIT unconditionally
when seen, leaving the task unmarked even if the WARN fires.
Fixes: 337ec00b1d ("sched_ext: Implement cgroup sub-sched enabling and disabling")
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
bypass_lb_cpu() transfers tasks between per-CPU bypass DSQs without
migrating them - task_cpu() only updates when the donee later consumes the
task via move_remote_task_to_local_dsq(). If the LB timer fires again before
consumption and the new DSQ becomes a donor, @p is still on the previous CPU
and task_rq(@p) != donor_rq. @p can't be moved without its own rq locked.
Skip such tasks.
Fixes: 95d1df610c ("sched_ext: Implement load balancer for bypass mode")
Cc: stable@vger.kernel.org # v6.19+
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
bpf_iter_scx_dsq_new() clears kit->dsq on failure and
bpf_iter_scx_dsq_{next,destroy}() guard against that. scx_dsq_move() doesn't -
it dereferences kit->dsq immediately, so a BPF program that calls
scx_bpf_dsq_move[_vtime]() after a failed iter_new oopses the kernel.
Return false if kit->dsq is NULL.
Fixes: 4c30f5ce4f ("sched_ext: Implement scx_bpf_dispatch[_vtime]_from_dsq()")
Cc: stable@vger.kernel.org # v6.12+
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
When ops.sub_attach is set, scx_alloc_and_add_sched() creates sub_kset as a
child of &sch->kobj, which pins the parent with its own reference. The
disable paths never call kset_unregister(), so the final kobject_put() in
bpf_scx_unreg() leaves a stale reference and scx_kobj_release() never runs,
leaking the whole struct scx_sched on every load/unload cycle.
Unregister sub_kset in scx_root_disable() and scx_sub_disable() before
kobject_del(&sch->kobj).
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Reported-by: Chris Mason <clm@meta.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_hardlockup() runs from NMI and eventually calls scx_claim_exit(),
which takes scx_sched_lock. scx_sched_lock isn't NMI-safe and grabbing
it from NMI context can lead to deadlocks.
The hardlockup handler is best-effort recovery and the disable path it
triggers runs off of irq_work anyway. Move the handle_lockup() call into
an irq_work so it runs in IRQ context.
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
When unregistered my self-written scx scheduler, the following panic
occurs.
[ 229.923133] Kernel text patching generated an invalid instruction at 0xffff80009bc2c1f8!
[ 229.923146] Internal error: Oops - BRK: 00000000f2000100 [#1] SMP
[ 230.077871] CPU: 48 UID: 0 PID: 1760 Comm: kworker/u583:7 Not tainted 7.0.0+ #3 PREEMPT(full)
[ 230.086677] Hardware name: NVIDIA GB200 NVL/P3809-BMC, BIOS 02.05.12 20251107
[ 230.093972] Workqueue: events_unbound bpf_map_free_deferred
[ 230.099675] Sched_ext: invariant_0.1.0_aarch64_unknown_linux_gnu_debug (disabling), task: runnable_at=-174ms
[ 230.116843] pc : 0xffff80009bc2c1f8
[ 230.120406] lr : dequeue_task_scx+0x270/0x2d0
[ 230.217749] Call trace:
[ 230.228515] 0xffff80009bc2c1f8 (P)
[ 230.232077] dequeue_task+0x84/0x188
[ 230.235728] sched_change_begin+0x1dc/0x250
[ 230.240000] __set_cpus_allowed_ptr_locked+0x17c/0x240
[ 230.245250] __set_cpus_allowed_ptr+0x74/0xf0
[ 230.249701] ___migrate_enable+0x4c/0xa0
[ 230.253707] bpf_map_free_deferred+0x1a4/0x1b0
[ 230.258246] process_one_work+0x184/0x540
[ 230.262342] worker_thread+0x19c/0x348
[ 230.266170] kthread+0x13c/0x150
[ 230.269465] ret_from_fork+0x10/0x20
[ 230.281393] Code: d4202000 d4202000 d4202000 d4202000 (d4202000)
[ 230.287621] ---[ end trace 0000000000000000 ]---
[ 231.160046] Kernel panic - not syncing: Oops - BRK: Fatal exception in interrupt
The root cause is that the JIT page backing ops->quiescent() is freed
before all callers of that function have stopped.
The expected ordering during teardown is:
bitmap_zero(sch->has_op) + synchronize_rcu()
-> guarantees no CPU will ever call sch->ops.* again
-> only THEN free the BPF struct_ops JIT page
bpf_scx_unreg() is supposed to enforce the order, but after
commit f4a6c506d1 ("sched_ext: Always bounce scx_disable() through
irq_work"), disable_work is no longer queued directly, causing
kthread_flush_work() to be a noop. Thus, the caller drops the struct_ops
map too early and poisoned with AARCH64_BREAK_FAULT before
disable_workfn ever execute.
So the subsequent dequeue_task() still sees SCX_HAS_OP(sch, quiescent)
as true and calls ops.quiescent, which hit on the poisoned page and BRK
panic.
Add a helper scx_flush_disable_work() so the future use cases that want
to flush disable_work can use it.
Also amend the call for scx_root_enable_workfn() and
scx_sub_enable_workfn() which have similar pattern in the error path.
Fixes: f4a6c506d1 ("sched_ext: Always bounce scx_disable() through irq_work")
Signed-off-by: Richard Cheng <icheng@nvidia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
local_dsq_post_enq() calls call_task_dequeue() with scx_root instead of
the scheduler instance actually managing the task. When
CONFIG_EXT_SUB_SCHED is enabled, tasks may be managed by a sub-scheduler
whose ops.dequeue() callback differs from root's. Using scx_root causes
the wrong scheduler's ops.dequeue() to be consulted: sub-sched tasks
dispatched to a local DSQ via scx_bpf_dsq_move_to_local() will have
SCX_TASK_IN_CUSTODY cleared but the sub-scheduler's ops.dequeue() is
never invoked, violating the custody exit semantics.
Fix by adding a 'struct scx_sched *sch' parameter to local_dsq_post_enq()
and move_local_task_to_local_dsq(), and propagating the correct scheduler
from their callers dispatch_enqueue(), move_task_between_dsqs(), and
consume_dispatch_q().
This is consistent with dispatch_enqueue()'s non-local path which already
passes 'sch' directly to call_task_dequeue() for global/bypass DSQs.
Fixes: ebf1ccff79 ("sched_ext: Fix ops.dequeue() semantics")
Signed-off-by: zhidao su <suzhidao@xiaomi.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Pull to receive:
2d2b026c3e ("sched_ext: Deny SCX kfuncs to non-SCX struct_ops programs")
which modifies scx_kfunc_context_filter() to avoid conflicts with planned
changes in for-7.2.
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_kfunc_context_filter() currently allows non-SCX struct_ops programs
(e.g. tcp_congestion_ops) to call SCX unlocked kfuncs. This is wrong
for two reasons:
- It is semantically incorrect: a TCP congestion control program has no
business calling SCX kfuncs such as scx_bpf_kick_cpu().
- With CONFIG_EXT_SUB_SCHED=y, kfuncs like scx_bpf_kick_cpu() call
scx_prog_sched(aux), which invokes bpf_prog_get_assoc_struct_ops(aux)
and casts the result to struct sched_ext_ops * before reading ops->priv.
For a non-SCX struct_ops program the returned pointer is the kdata of
that struct_ops type, which is far smaller than sched_ext_ops, making
the read an out-of-bounds access (confirmed with KASAN).
Extend the filter to cover scx_kfunc_set_any and scx_kfunc_set_idle as
well, and deny all SCX kfuncs for any struct_ops program that is not the
SCX struct_ops. This addresses both issues: the semantic contract is
enforced at the verifier level, and the runtime out-of-bounds access
becomes unreachable.
Fixes: d1d3c1c6ae ("sched_ext: Add verifier-time kfunc context filter")
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
BPF schedulers that can't hold task_struct pointers (arena-backed ones in
particular) key tasks by pid. During exit, pid is released before the
task finishes passing through scheduler callbacks, so a dying task
becomes invisible to the BPF side mid-schedule. scx_qmap hits this: an
exiting task's dispatch callback can't recover its queue entry, stalling
dispatch until SCX_EXIT_ERROR_STALL.
Add a unique non-zero u64 p->scx.tid assigned at fork that survives the
full task lifetime including exit. scx_bpf_tid_to_task() looks up the
task; unlike bpf_task_from_pid(), it handles exiting tasks.
The lookup costs an rhashtable insert/remove under scx_tasks_lock, so
root schedulers opt in via SCX_OPS_TID_TO_TASK. Sub-schedulers that set
the flag to declare a dependency are rejected at attach if root didn't
opt in.
scx_qmap converted: keys tasks by tid and enables SCX_OPS_ENQ_EXITING.
Pre-patch it stalls within seconds under a non-leader-exec workload;
with the patch it runs cleanly.
v3: Warn on rhashtable_lookup_insert_fast() failure via new
scx_tid_hash_insert() helper (Cheng-Yang Chou).
v2: Guard scx_root deref in scx_bpf_tid_to_task() error path. The kfunc
is registered via scx_kfunc_set_any and reachable from tracing and
syscall programs when no scheduler is attached (Cheng-Yang Chou).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Pull to receive 73bd122778 ("rhashtable: Restore insecure_elasticity
toggle") as a dependency for upcoming patches that use
.insecure_elasticity = true on their rhashtables.
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_sched_hash is inserted into under scx_sched_lock (raw_spinlock_irq)
in scx_link_sched(). rhashtable's sync grow path calls get_random_u32()
and does a GFP_ATOMIC allocation; both acquire regular spinlocks, which
is unsafe under raw_spinlock_t. Set insecure_elasticity to skip the
sync grow.
v2:
- Dropped dsq_hash changes. Insertion is not under raw_spin_lock.
- Switched from no_sync_grow flag to insecure_elasticity.
Fixes: 25037af712 ("sched_ext: Add rhashtable lookup for sub-schedulers")
Signed-off-by: Tejun Heo <tj@kernel.org>
Take scx_qmap for example, when sub scheduler is attached, there is
'BPF sub-scheduler "qmap" enabled' message, but when detached, the log
is missing. Add a new function to do the log thing, it can be used by
both root scheduler and sub scheduler.
Signed-off-by: Zhao Mengmeng <zhaomengmeng@kylinos.cn>
Reviewed-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
- Fix printf format warning for bprintf
sunrpc uses a trace_printk() that triggers a printf warning during the
compile. Move the __printf() attribute around for when debugging is not
enabled the warning will go away.
- Remove redundant check for EVENT_FILE_FL_FREED in event_filter_write()
The FREED flag is checked in the call to event_file_file() and then
checked again right afterward, which is unneeded.
- Clean up event_file_file() and event_file_data() helpers
These helper functions played a different role in the past, but now with
eventfs, the READ_ONCE() isn't needed. Simplify the code a bit and also
add a warning to event_file_data() if the file or its data is not present.
- Remove updating file->private_data in tracing open
All access to the file private data is handled by the helper functions,
which do not use file->private_data. Stop updating it on open.
- Show ENUM names in function arguments via BTF in function tracing
When showing the function arguments when func-args option is set for
function tracing, if one of the arguments is found to be an enum, show the
name of the enum instead of its number.
- Add new trace_call__##name() API for tracepoints
Tracepoints are enabled via static_branch() blocks, where when not
enabled, there's only a nop that is in the code where the execution will
just skip over it. When tracing is enabled, the nop is converted to a
direct jump to the tracepoint code. Sometimes more calculations are
required to be performed to update the parameters of the tracepoint. In
this case, trace_##name##_enabled() is called which is a static_branch()
that gets enabled only when the tracepoint is enabled. This allows the
extra calculations to also be skipped by the nop:
if (trace_foo_enabled()) {
x = bar();
trace_foo(x);
}
Where the x=bar() is only performed when foo is enabled. The problem with
this approach is that there's now two static_branch() calls. One for
checking if the tracepoint is enabled, and then again to know if the
tracepoint should be called. The second one is redundant.
Introduce trace_call__foo() that will call the foo() tracepoint directly
without doing a static_branch():
if (trace_foo_enabled()) {
x = bar();
trace_call__foo();
}
- Update various locations to use the new trace_call__##name() API
- Move snapshot code out of trace.c
Cleaning up trace.c to not be a "dump all", move the snapshot code out of
it and into a new trace_snapshot.c file.
- Clean up some "%*.s" to "%*s"
- Allow boot kernel command line options to be called multiple times
Have options like:
ftrace_filter=foo ftrace_filter=bar ftrace_filter=zoo
Equal to:
ftrace_filter=foo,bar,zoo
- Fix ipi_raise event CPU field to be a CPU field
The ipi_raise target_cpus field is defined as a __bitmask(). There is now a
__cpumask() field definition. Update the field to use that.
- Have hist_field_name() use a snprintf() and not a series of strcat()
It's safer to use snprintf() that a series of strcat().
- Fix tracepoint regfunc balancing
A tracepoint can define a "reg" and "unreg" function that gets called
before the tracepoint is enabled, and after it is disabled respectively.
But on error, after the "reg" func is called and the tracepoint is not
enabled, the "unreg" function is not called to tear down what the "reg"
function performed.
- Fix output that shows what histograms are enabled
Event variables are displayed incorrectly in the histogram output.
Instead of "sched.sched_wakeup.$var", it is showing
"$sched.sched_wakeup.var" where the '$' is in the incorrect location.
- Some other simple cleanups.
-----BEGIN PGP SIGNATURE-----
iIoEABYKADIWIQRRSw7ePDh/lE+zeZMp5XQQmuv6qgUCaeCpvxQccm9zdGVkdEBn
b29kbWlzLm9yZwAKCRAp5XQQmuv6qt2WAP44m85BbAjBqJe4WR103eOXV+bREBta
dRoReKJOMe519gEAp0rK/HoCvHgHhIGe3gaGdIsNhnaxoFyNWMG/wokoLAY=
=Hg6+
-----END PGP SIGNATURE-----
Merge tag 'trace-v7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull tracing updates from Steven Rostedt:
- Fix printf format warning for bprintf
sunrpc uses a trace_printk() that triggers a printf warning during
the compile. Move the __printf() attribute around for when debugging
is not enabled the warning will go away
- Remove redundant check for EVENT_FILE_FL_FREED in
event_filter_write()
The FREED flag is checked in the call to event_file_file() and then
checked again right afterward, which is unneeded
- Clean up event_file_file() and event_file_data() helpers
These helper functions played a different role in the past, but now
with eventfs, the READ_ONCE() isn't needed. Simplify the code a bit
and also add a warning to event_file_data() if the file or its data
is not present
- Remove updating file->private_data in tracing open
All access to the file private data is handled by the helper
functions, which do not use file->private_data. Stop updating it on
open
- Show ENUM names in function arguments via BTF in function tracing
When showing the function arguments when func-args option is set for
function tracing, if one of the arguments is found to be an enum,
show the name of the enum instead of its number
- Add new trace_call__##name() API for tracepoints
Tracepoints are enabled via static_branch() blocks, where when not
enabled, there's only a nop that is in the code where the execution
will just skip over it. When tracing is enabled, the nop is converted
to a direct jump to the tracepoint code. Sometimes more calculations
are required to be performed to update the parameters of the
tracepoint. In this case, trace_##name##_enabled() is called which is
a static_branch() that gets enabled only when the tracepoint is
enabled. This allows the extra calculations to also be skipped by the
nop:
if (trace_foo_enabled()) {
x = bar();
trace_foo(x);
}
Where the x=bar() is only performed when foo is enabled. The problem
with this approach is that there's now two static_branch() calls. One
for checking if the tracepoint is enabled, and then again to know if
the tracepoint should be called. The second one is redundant
Introduce trace_call__foo() that will call the foo() tracepoint
directly without doing a static_branch():
if (trace_foo_enabled()) {
x = bar();
trace_call__foo();
}
- Update various locations to use the new trace_call__##name() API
- Move snapshot code out of trace.c
Cleaning up trace.c to not be a "dump all", move the snapshot code
out of it and into a new trace_snapshot.c file
- Clean up some "%*.s" to "%*s"
- Allow boot kernel command line options to be called multiple times
Have options like:
ftrace_filter=foo ftrace_filter=bar ftrace_filter=zoo
Equal to:
ftrace_filter=foo,bar,zoo
- Fix ipi_raise event CPU field to be a CPU field
The ipi_raise target_cpus field is defined as a __bitmask(). There is
now a __cpumask() field definition. Update the field to use that
- Have hist_field_name() use a snprintf() and not a series of strcat()
It's safer to use snprintf() that a series of strcat()
- Fix tracepoint regfunc balancing
A tracepoint can define a "reg" and "unreg" function that gets called
before the tracepoint is enabled, and after it is disabled
respectively. But on error, after the "reg" func is called and the
tracepoint is not enabled, the "unreg" function is not called to tear
down what the "reg" function performed
- Fix output that shows what histograms are enabled
Event variables are displayed incorrectly in the histogram output
Instead of "sched.sched_wakeup.$var", it is showing
"$sched.sched_wakeup.var" where the '$' is in the incorrect location
- Some other simple cleanups
* tag 'trace-v7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (24 commits)
selftests/ftrace: Add test case for fully-qualified variable references
tracing: Fix fully-qualified variable reference printing in histograms
tracepoint: balance regfunc() on func_add() failure in tracepoint_add_func()
tracing: Rebuild full_name on each hist_field_name() call
tracing: Report ipi_raise target CPUs as cpumask
tracing: Remove duplicate latency_fsnotify() stub
tracing: Preserve repeated trace_trigger boot parameters
tracing: Append repeated boot-time tracing parameters
tracing: Remove spurious default precision from show_event_trigger/filter formats
cpufreq: Use trace_call__##name() at guarded tracepoint call sites
tracing: Remove tracing_alloc_snapshot() when snapshot isn't defined
tracing: Move snapshot code out of trace.c and into trace_snapshot.c
mm: damon: Use trace_call__##name() at guarded tracepoint call sites
btrfs: Use trace_call__##name() at guarded tracepoint call sites
spi: Use trace_call__##name() at guarded tracepoint call sites
i2c: Use trace_call__##name() at guarded tracepoint call sites
kernel: Use trace_call__##name() at guarded tracepoint call sites
tracepoint: Add trace_call__##name() API
tracing: trace_mmap.h: fix a kernel-doc warning
tracing: Pretty-print enum parameters in function arguments
...
- Refactor da_monitor header to share handlers across monitor types
No functional changes, only less code duplication.
- Add Hybrid Automata model class
Add a new model class that extends deterministic automata by adding
constraints on transitions and states. Those constraints can take into
account wall-clock time and as such allow RV monitor to make
assertions on real time. Add documentation and code generation
scripts.
- Add stall monitor as hybrid automaton example
Add a monitor that triggers a violation when a task is stalling as an
example of automaton working with real time variables.
- Convert the opid monitor to a hybrid automaton
The opid monitor can be heavily simplified if written as a hybrid
automaton: instead of tracking preempt and interrupt enable/disable
events, it can just run constraints on the preemption/interrupt
states when events like wakeup and need_resched verify.
- Add support for per-object monitors in DA/HA
Allow writing deterministic and hybrid automata monitors for generic
objects (e.g. any struct), by exploiting a hash table where objects
are saved. This allows to track more than just tasks in RV. For
instance it will be used to track deadline entities in deadline
monitors.
- Add deadline tracepoints and move some deadline utilities
Prepare the ground for deadline monitors by defining events and
exporting helpers.
- Add nomiss deadline monitor
Add first example of deadline monitor asserting all entities complete
before their deadline.
- Improve rvgen error handling
Introduce AutomataError exception class and better handle expected
exceptions while showing a backtrace for unexpected ones.
- Improve python code quality in rvgen
Refactor the rvgen generation scripts to align with python best
practices: use f-strings instead of %, use len() instead of __len__(),
remove semicolons, use context managers for file operations, fix
whitespace violations, extract magic strings into constants, remove
unused imports and methods.
- Fix small bugs in rvgen
The generator scripts presented some corner case bugs: logical error in
validating what a correct dot file looks like, fix an isinstance()
check, enforce a dot file has an initial state, fix type annotations
and typos in comments.
- rvgen refactoring
Refactor automata.py to use iterator-based parsing and handle required
arguments directly in argparse.
- Allow epoll in rtapp-sleep monitor
The epoll_wait call is now rt-friendly so it should be allowed in the
sleep monitor as a valid sleep method.
-----BEGIN PGP SIGNATURE-----
iIoEABYKADIWIQRRSw7ePDh/lE+zeZMp5XQQmuv6qgUCad4kUxQccm9zdGVkdEBn
b29kbWlzLm9yZwAKCRAp5XQQmuv6qiJgAPsGJ0TCTubdn2x6fpwLDKUcSsNsYrok
m8gLHeK0rmkKNQD/ajUW+RBsOufXAsnSzUoUcl7sw1TSiQJ085U+0+ZeDgM=
=WZqk
-----END PGP SIGNATURE-----
Merge tag 'trace-rv-v7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull runtime verification updates from Steven Rostedt:
- Refactor da_monitor header to share handlers across monitor types
No functional changes, only less code duplication.
- Add Hybrid Automata model class
Add a new model class that extends deterministic automata by adding
constraints on transitions and states. Those constraints can take
into account wall-clock time and as such allow RV monitor to make
assertions on real time. Add documentation and code generation
scripts.
- Add stall monitor as hybrid automaton example
Add a monitor that triggers a violation when a task is stalling as an
example of automaton working with real time variables.
- Convert the opid monitor to a hybrid automaton
The opid monitor can be heavily simplified if written as a hybrid
automaton: instead of tracking preempt and interrupt enable/disable
events, it can just run constraints on the preemption/interrupt
states when events like wakeup and need_resched verify.
- Add support for per-object monitors in DA/HA
Allow writing deterministic and hybrid automata monitors for generic
objects (e.g. any struct), by exploiting a hash table where objects
are saved. This allows to track more than just tasks in RV. For
instance it will be used to track deadline entities in deadline
monitors.
- Add deadline tracepoints and move some deadline utilities
Prepare the ground for deadline monitors by defining events and
exporting helpers.
- Add nomiss deadline monitor
Add first example of deadline monitor asserting all entities complete
before their deadline.
- Improve rvgen error handling
Introduce AutomataError exception class and better handle expected
exceptions while showing a backtrace for unexpected ones.
- Improve python code quality in rvgen
Refactor the rvgen generation scripts to align with python best
practices: use f-strings instead of %, use len() instead of
__len__(), remove semicolons, use context managers for file
operations, fix whitespace violations, extract magic strings into
constants, remove unused imports and methods.
- Fix small bugs in rvgen
The generator scripts presented some corner case bugs: logical error
in validating what a correct dot file looks like, fix an isinstance()
check, enforce a dot file has an initial state, fix type annotations
and typos in comments.
- rvgen refactoring
Refactor automata.py to use iterator-based parsing and handle
required arguments directly in argparse.
- Allow epoll in rtapp-sleep monitor
The epoll_wait call is now rt-friendly so it should be allowed in the
sleep monitor as a valid sleep method.
* tag 'trace-rv-v7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (32 commits)
rv: Allow epoll in rtapp-sleep monitor
rv/rvgen: fix _fill_states() return type annotation
rv/rvgen: fix unbound loop variable warning
rv/rvgen: enforce presence of initial state
rv/rvgen: extract node marker string to class constant
rv/rvgen: fix isinstance check in Variable.expand()
rv/rvgen: make monitor arguments required in rvgen
rv/rvgen: remove unused __get_main_name method
rv/rvgen: remove unused sys import from dot2c
rv/rvgen: refactor automata.py to use iterator-based parsing
rv/rvgen: use class constant for init marker
rv/rvgen: fix DOT file validation logic error
rv/rvgen: fix PEP 8 whitespace violations
rv/rvgen: fix typos in automata and generator docstring and comments
rv/rvgen: use context managers for file operations
rv/rvgen: remove unnecessary semicolons
rv/rvgen: replace __len__() calls with len()
rv/rvgen: replace % string formatting with f-strings
rv/rvgen: remove bare except clauses in generator
rv/rvgen: introduce AutomataError exception class
...
- Cgroup sub-scheduler groundwork. Multiple BPF schedulers can be
attached to cgroups and the dispatch path is made hierarchical. This
involves substantial restructuring of the core dispatch, bypass,
watchdog, and dump paths to be per-scheduler, along with new
infrastructure for scheduler ownership enforcement, lifecycle
management, and cgroup subtree iteration. The enqueue path is not yet
updated and will follow in a later cycle.
- scx_bpf_dsq_reenq() generalized to support any DSQ including remote
local DSQs and user DSQs. Built on top of this, SCX_ENQ_IMMED
guarantees that tasks dispatched to local DSQs either run immediately
or get reenqueued back through ops.enqueue(), giving schedulers tighter
control over queueing latency. Also useful for opportunistic CPU
sharing across sub-schedulers.
- ops.dequeue() was only invoked when the core knew a task was in BPF
data structures, missing scheduling property change events and skipping
callbacks for non-local DSQ dispatches from ops.select_cpu(). Fixed to
guarantee exactly one ops.dequeue() call when a task leaves BPF
scheduler custody.
- Kfunc access validation moved from runtime to BPF verifier time,
removing runtime mask enforcement.
- Idle SMT sibling prioritization in the idle CPU selection path.
- Documentation, selftest, and tooling updates. Misc bug fixes and
cleanups.
- Merges from tip/sched-core, cgroup/for-7.1, and for-7.0-fixes to
resolve dependencies and conflicts for the above changes.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCad0uaA4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGbktAQD2HrKdydyEfefz/n4mNpIXh/DFYX49NgKYcgUh
sKy4ngD/Sy7nAZS2zwM+36PN6jBV7+cfuoaiKPgCstPFeGsvPwU=
=fsgj
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext updates from Tejun Heo:
- cgroup sub-scheduler groundwork
Multiple BPF schedulers can be attached to cgroups and the dispatch
path is made hierarchical. This involves substantial restructuring of
the core dispatch, bypass, watchdog, and dump paths to be
per-scheduler, along with new infrastructure for scheduler ownership
enforcement, lifecycle management, and cgroup subtree iteration
The enqueue path is not yet updated and will follow in a later cycle
- scx_bpf_dsq_reenq() generalized to support any DSQ including remote
local DSQs and user DSQs
Built on top of this, SCX_ENQ_IMMED guarantees that tasks dispatched
to local DSQs either run immediately or get reenqueued back through
ops.enqueue(), giving schedulers tighter control over queueing
latency
Also useful for opportunistic CPU sharing across sub-schedulers
- ops.dequeue() was only invoked when the core knew a task was in BPF
data structures, missing scheduling property change events and
skipping callbacks for non-local DSQ dispatches from ops.select_cpu()
Fixed to guarantee exactly one ops.dequeue() call when a task leaves
BPF scheduler custody
- Kfunc access validation moved from runtime to BPF verifier time,
removing runtime mask enforcement
- Idle SMT sibling prioritization in the idle CPU selection path
- Documentation, selftest, and tooling updates. Misc bug fixes and
cleanups
* tag 'sched_ext-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext: (134 commits)
tools/sched_ext: Add explicit cast from void* in RESIZE_ARRAY()
sched_ext: Make string params of __ENUM_set() const
tools/sched_ext: Kick home CPU for stranded tasks in scx_qmap
sched_ext: Drop spurious warning on kick during scheduler disable
sched_ext: Warn on task-based SCX op recursion
sched_ext: Rename scx_kf_allowed_on_arg_tasks() to scx_kf_arg_task_ok()
sched_ext: Remove runtime kfunc mask enforcement
sched_ext: Add verifier-time kfunc context filter
sched_ext: Drop redundant rq-locked check from scx_bpf_task_cgroup()
sched_ext: Decouple kfunc unlocked-context check from kf_mask
sched_ext: Fix ops.cgroup_move() invocation kf_mask and rq tracking
sched_ext: Track @p's rq lock across set_cpus_allowed_scx -> ops.set_cpumask
sched_ext: Add select_cpu kfuncs to scx_kfunc_ids_unlocked
sched_ext: Drop TRACING access to select_cpu kfuncs
selftests/sched_ext: Fix wrong DSQ ID in peek_dsq error message
sched_ext: Documentation: improve accuracy of task lifecycle pseudo-code
selftests/sched_ext: Improve runner error reporting for invalid arguments
sched_ext: Documentation: Fix scx_bpf_move_to_local kfunc name
sched_ext: Documentation: Add ops.dequeue() to task lifecycle
tools/sched_ext: Fix off-by-one in scx_sdt payload zeroing
...
Fair scheduling updates:
- Skip SCHED_IDLE rq for SCHED_IDLE tasks (Christian Loehle)
- Remove superfluous rcu_read_lock() in the wakeup path (K Prateek Nayak)
- Simplify the entry condition for update_idle_cpu_scan() (K Prateek Nayak)
- Simplify SIS_UTIL handling in select_idle_cpu() (K Prateek Nayak)
- Avoid overflow in enqueue_entity() (K Prateek Nayak)
- Update overutilized detection (Vincent Guittot)
- Prevent negative lag increase during delayed dequeue (Vincent Guittot)
- Clear buddies for preempt_short (Vincent Guittot)
- Implement more complex proportional newidle balance (Peter Zijlstra)
- Increase weight bits for avg_vruntime (Peter Zijlstra)
- Use full weight to __calc_delta() (Peter Zijlstra)
RT and DL scheduling updates:
- Fix incorrect schedstats for rt and dl thread (Dengjun Su)
- Skip group schedulable check with rt_group_sched=0 (Michal Koutný)
- Move group schedulability check to sched_rt_global_validate()
(Michal Koutný)
- Add reporting of runtime left & abs deadline to sched_getattr()
for DEADLINE tasks (Tommaso Cucinotta)
Scheduling topology updates by K Prateek Nayak:
- Compute sd_weight considering cpuset partitions
- Extract "imb_numa_nr" calculation into a separate helper
- Allocate per-CPU sched_domain_shared in s_data
- Switch to assigning "sd->shared" from s_data
- Remove sched_domain_shared allocation with sd_data
Energy-aware scheduling updates:
- Filter false overloaded_group case for EAS (Vincent Guittot)
- PM: EM: Switch to rcu_dereference_all() in wakeup path
(Dietmar Eggemann)
Infrastructure updates:
- Replace use of system_unbound_wq with system_dfl_wq (Marco Crivellari)
Proxy scheduling updates by John Stultz:
- Make class_schedulers avoid pushing current, and get rid of proxy_tag_curr()
- Minimise repeated sched_proxy_exec() checking
- Fix potentially missing balancing with Proxy Exec
- Fix and improve task::blocked_on et al handling
- Add assert_balance_callbacks_empty() helper
- Add logic to zap balancing callbacks if we pick again
- Move attach_one_task() and attach_task() helpers to sched.h
- Handle blocked-waiter migration (and return migration)
- Add K Prateek Nayak to scheduler reviewers for proxy execution
Misc cleanups and fixes by John Stultz, Joseph Salisbury,
Peter Zijlstra, K Prateek Nayak, Michal Koutný, Randy Dunlap,
Shrikanth Hegde, Vincent Guittot, Zhan Xusheng, Xie Yuanbin
and Vincent Guittot.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=NjyS
-----END PGP SIGNATURE-----
Merge tag 'sched-core-2026-04-13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler updates from Ingo Molnar:
"Fair scheduling updates:
- Skip SCHED_IDLE rq for SCHED_IDLE tasks (Christian Loehle)
- Remove superfluous rcu_read_lock() in the wakeup path (K Prateek Nayak)
- Simplify the entry condition for update_idle_cpu_scan() (K Prateek Nayak)
- Simplify SIS_UTIL handling in select_idle_cpu() (K Prateek Nayak)
- Avoid overflow in enqueue_entity() (K Prateek Nayak)
- Update overutilized detection (Vincent Guittot)
- Prevent negative lag increase during delayed dequeue (Vincent Guittot)
- Clear buddies for preempt_short (Vincent Guittot)
- Implement more complex proportional newidle balance (Peter Zijlstra)
- Increase weight bits for avg_vruntime (Peter Zijlstra)
- Use full weight to __calc_delta() (Peter Zijlstra)
RT and DL scheduling updates:
- Fix incorrect schedstats for rt and dl thread (Dengjun Su)
- Skip group schedulable check with rt_group_sched=0 (Michal Koutný)
- Move group schedulability check to sched_rt_global_validate()
(Michal Koutný)
- Add reporting of runtime left & abs deadline to sched_getattr()
for DEADLINE tasks (Tommaso Cucinotta)
Scheduling topology updates by K Prateek Nayak:
- Compute sd_weight considering cpuset partitions
- Extract "imb_numa_nr" calculation into a separate helper
- Allocate per-CPU sched_domain_shared in s_data
- Switch to assigning "sd->shared" from s_data
- Remove sched_domain_shared allocation with sd_data
Energy-aware scheduling updates:
- Filter false overloaded_group case for EAS (Vincent Guittot)
- PM: EM: Switch to rcu_dereference_all() in wakeup path
(Dietmar Eggemann)
Infrastructure updates:
- Replace use of system_unbound_wq with system_dfl_wq (Marco Crivellari)
Proxy scheduling updates by John Stultz:
- Make class_schedulers avoid pushing current, and get rid of proxy_tag_curr()
- Minimise repeated sched_proxy_exec() checking
- Fix potentially missing balancing with Proxy Exec
- Fix and improve task::blocked_on et al handling
- Add assert_balance_callbacks_empty() helper
- Add logic to zap balancing callbacks if we pick again
- Move attach_one_task() and attach_task() helpers to sched.h
- Handle blocked-waiter migration (and return migration)
- Add K Prateek Nayak to scheduler reviewers for proxy execution
Misc cleanups and fixes by John Stultz, Joseph Salisbury, Peter
Zijlstra, K Prateek Nayak, Michal Koutný, Randy Dunlap, Shrikanth
Hegde, Vincent Guittot, Zhan Xusheng, Xie Yuanbin and Vincent Guittot"
* tag 'sched-core-2026-04-13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (46 commits)
sched/eevdf: Clear buddies for preempt_short
sched/rt: Cleanup global RT bandwidth functions
sched/rt: Move group schedulability check to sched_rt_global_validate()
sched/rt: Skip group schedulable check with rt_group_sched=0
sched/fair: Avoid overflow in enqueue_entity()
sched: Use u64 for bandwidth ratio calculations
sched/fair: Prevent negative lag increase during delayed dequeue
sched/fair: Use sched_energy_enabled()
sched: Handle blocked-waiter migration (and return migration)
sched: Move attach_one_task and attach_task helpers to sched.h
sched: Add logic to zap balance callbacks if we pick again
sched: Add assert_balance_callbacks_empty helper
sched/locking: Add special p->blocked_on==PROXY_WAKING value for proxy return-migration
sched: Fix modifying donor->blocked on without proper locking
locking: Add task::blocked_lock to serialize blocked_on state
sched: Fix potentially missing balancing with Proxy Exec
sched: Minimise repeated sched_proxy_exec() checking
sched: Make class_schedulers avoid pushing current, and get rid of proxy_tag_curr()
MAINTAINERS: Add K Prateek Nayak to scheduler reviewers
sched/core: Get this cpu once in ttwu_queue_cond()
...
- A rework of the hrtimer subsystem to reduce the overhead for frequently
armed timers, especially the hrtick scheduler timer.
- Better timer locality decision
- Simplification of the evaluation of the first expiry time by
keeping track of the neighbor timers in the RB-tree by providing a
RB-tree variant with neighbor links. That avoids walking the
RB-tree on removal to find the next expiry time, but even more
important allows to quickly evaluate whether a timer which is
rearmed changes the position in the RB-tree with the modified
expiry time or not. If not, the dequeue/enqueue sequence which both
can end up in rebalancing can be completely avoided.
- Deferred reprogramming of the underlying clock event device. This
optimizes for the situation where a hrtimer callback sets the need
resched bit. In that case the code attempts to defer the
re-programming of the clock event device up to the point where the
scheduler has picked the next task and has the next hrtick timer
armed. In case that there is no immediate reschedule or soft
interrupts have to be handled before reaching the reschedule point
in the interrupt entry code the clock event is reprogrammed in one
of those code paths to prevent that the timer becomes stale.
- Support for clocksource coupled clockevents
The TSC deadline timer is coupled to the TSC. The next event is
programmed in TSC time. Currently this is done by converting the
CLOCK_MONOTONIC based expiry value into a relative timeout,
converting it into TSC ticks, reading the TSC adding the delta
ticks and writing the deadline MSR.
As the timekeeping core has the conversion factors for the TSC
already, the whole back and forth conversion can be completely
avoided. The timekeeping core calculates the reverse conversion
factors from nanoseconds to TSC ticks and utilizes the base
timestamps of TSC and CLOCK_MONOTONIC which are updated once per
tick. This allows a direct conversion into the TSC deadline value
without reading the time and as a bonus keeps the deadline
conversion in sync with the TSC conversion factors, which are
updated by adjtimex() on systems with NTP/PTP enabled.
- Allow inlining of the clocksource read and clockevent write
functions when they are tiny enough, e.g. on x86 RDTSC and WRMSR.
With all those enhancements in place a hrtick enabled scheduler
provides the same performance as without hrtick. But also other hrtimer
users obviously benefit from these optimizations.
- Robustness improvements and cleanups of historical sins in the hrtimer
and timekeeping code.
- Rewrite of the clocksource watchdog.
The clocksource watchdog code has over time reached the state of an
impenetrable maze of duct tape and staples. The original design, which was
made in the context of systems far smaller than today, is based on the
assumption that the to be monitored clocksource (TSC) can be trivially
compared against a known to be stable clocksource (HPET/ACPI-PM timer).
Over the years this rather naive approach turned out to have major
flaws. Long delays between the watchdog invocations can cause wrap
arounds of the reference clocksource. The access to the reference
clocksource degrades on large multi-sockets systems dure to
interconnect congestion. This has been addressed with various
heuristics which degraded the accuracy of the watchdog to the point
that it fails to detect actual TSC problems on older hardware which
exposes slow inter CPU drifts due to firmware manipulating the TSC to
hide SMI time.
The rewrite addresses this by:
- Restricting the validation against the reference clocksource to the
boot CPU which is usually closest to the legacy block which
contains the reference clocksource (HPET/ACPI-PM).
- Do a round robin validation betwen the boot CPU and the other CPUs
based only on the TSC with an algorithm similar to the TSC
synchronization code during CPU hotplug.
- Being more leniant versus remote timeouts
- The usual tiny fixes, cleanups and enhancements all over the place
-----BEGIN PGP SIGNATURE-----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=2qc8
-----END PGP SIGNATURE-----
Merge tag 'timers-core-2026-04-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull timer core updates from Thomas Gleixner:
- A rework of the hrtimer subsystem to reduce the overhead for
frequently armed timers, especially the hrtick scheduler timer:
- Better timer locality decision
- Simplification of the evaluation of the first expiry time by
keeping track of the neighbor timers in the RB-tree by providing
a RB-tree variant with neighbor links. That avoids walking the
RB-tree on removal to find the next expiry time, but even more
important allows to quickly evaluate whether a timer which is
rearmed changes the position in the RB-tree with the modified
expiry time or not. If not, the dequeue/enqueue sequence which
both can end up in rebalancing can be completely avoided.
- Deferred reprogramming of the underlying clock event device. This
optimizes for the situation where a hrtimer callback sets the
need resched bit. In that case the code attempts to defer the
re-programming of the clock event device up to the point where
the scheduler has picked the next task and has the next hrtick
timer armed. In case that there is no immediate reschedule or
soft interrupts have to be handled before reaching the reschedule
point in the interrupt entry code the clock event is reprogrammed
in one of those code paths to prevent that the timer becomes
stale.
- Support for clocksource coupled clockevents
The TSC deadline timer is coupled to the TSC. The next event is
programmed in TSC time. Currently this is done by converting the
CLOCK_MONOTONIC based expiry value into a relative timeout,
converting it into TSC ticks, reading the TSC adding the delta
ticks and writing the deadline MSR.
As the timekeeping core has the conversion factors for the TSC
already, the whole back and forth conversion can be completely
avoided. The timekeeping core calculates the reverse conversion
factors from nanoseconds to TSC ticks and utilizes the base
timestamps of TSC and CLOCK_MONOTONIC which are updated once per
tick. This allows a direct conversion into the TSC deadline value
without reading the time and as a bonus keeps the deadline
conversion in sync with the TSC conversion factors, which are
updated by adjtimex() on systems with NTP/PTP enabled.
- Allow inlining of the clocksource read and clockevent write
functions when they are tiny enough, e.g. on x86 RDTSC and WRMSR.
With all those enhancements in place a hrtick enabled scheduler
provides the same performance as without hrtick. But also other
hrtimer users obviously benefit from these optimizations.
- Robustness improvements and cleanups of historical sins in the
hrtimer and timekeeping code.
- Rewrite of the clocksource watchdog.
The clocksource watchdog code has over time reached the state of an
impenetrable maze of duct tape and staples. The original design,
which was made in the context of systems far smaller than today, is
based on the assumption that the to be monitored clocksource (TSC)
can be trivially compared against a known to be stable clocksource
(HPET/ACPI-PM timer).
Over the years this rather naive approach turned out to have major
flaws. Long delays between the watchdog invocations can cause wrap
arounds of the reference clocksource. The access to the reference
clocksource degrades on large multi-sockets systems dure to
interconnect congestion. This has been addressed with various
heuristics which degraded the accuracy of the watchdog to the point
that it fails to detect actual TSC problems on older hardware which
exposes slow inter CPU drifts due to firmware manipulating the TSC to
hide SMI time.
The rewrite addresses this by:
- Restricting the validation against the reference clocksource to
the boot CPU which is usually closest to the legacy block which
contains the reference clocksource (HPET/ACPI-PM).
- Do a round robin validation betwen the boot CPU and the other
CPUs based only on the TSC with an algorithm similar to the TSC
synchronization code during CPU hotplug.
- Being more leniant versus remote timeouts
- The usual tiny fixes, cleanups and enhancements all over the place
* tag 'timers-core-2026-04-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (75 commits)
alarmtimer: Access timerqueue node under lock in suspend
hrtimer: Fix incorrect #endif comment for BITS_PER_LONG check
posix-timers: Fix stale function name in comment
timers: Get this_cpu once while clearing the idle state
clocksource: Rewrite watchdog code completely
clocksource: Don't use non-continuous clocksources as watchdog
x86/tsc: Handle CLOCK_SOURCE_VALID_FOR_HRES correctly
MIPS: Don't select CLOCKSOURCE_WATCHDOG
parisc: Remove unused clocksource flags
hrtimer: Add a helper to retrieve a hrtimer from its timerqueue node
hrtimer: Remove trailing comma after HRTIMER_MAX_CLOCK_BASES
hrtimer: Mark index and clockid of clock base as const
hrtimer: Drop unnecessary pointer indirection in hrtimer_expire_entry event
hrtimer: Drop spurious space in 'enum hrtimer_base_type'
hrtimer: Don't zero-initialize ret in hrtimer_nanosleep()
hrtimer: Remove hrtimer_get_expires_ns()
timekeeping: Mark offsets array as const
timekeeping/auxclock: Consistently use raw timekeeper for tk_setup_internals()
timer_list: Print offset as signed integer
tracing: Use explicit array size instead of sentinel elements in symbol printing
...
- Update qcom-hw DT bindings to include Eliza hardware (Abel Vesa)
- Update cpufreq-dt-platdev blocklist (Faruque Ansari)
- Minor updates to driver and dt-bindings for Tegra (Thierry Reding,
Rosen Penev)
- Add MAINTAINERS entry for CPPC driver (Viresh Kumar)
- Add support for new features: CPPC performance priority, Dynamic EPP,
Raw EPP, and new unit tests for them to amd-pstate (Gautham Shenoy,
Mario Limonciello)
- Fix sysfs files being present when HW missing and broken/outdated
documentation in the amd-pstate driver (Ninad Naik, Gautham Shenoy)
- Pass the policy to cpufreq_driver->adjust_perf() to avoid using
cpufreq_cpu_get() in the .adjust_perf() callback in amd-pstate which
leads to a scheduling-while-atomic bug (K Prateek Nayak)
- Clean up dead code in Kconfig for cpufreq (Julian Braha)
- Remove max_freq_req update for pre-existing cpufreq policy and add a
boost_freq_req QoS request to save the boost constraint instead of
overwriting the last scaling_max_freq constraint (Pierre Gondois)
- Embed cpufreq QoS freq_req objects in cpufreq policy so they all
are allocated in one go along with the policy to simplify lifetime
rules and avoid error handling issues (Viresh Kumar)
- Use DMI max speed when CPPC is unavailable in the acpi-cpufreq
scaling driver (Henry Tseng)
- Switch policy_is_shared() in cpufreq to using cpumask_nth() instead
of cpumask_weight() because the former is more efficient (Yury Norov)
- Use sysfs_emit() in sysfs show functions for cpufreq governor
attributes (Thorsten Blum)
- Update intel_pstate to stop returning an error when "off" is written
to its status sysfs attribute while the driver is already off (Fabio
De Francesco)
- Include current frequency in the debug message printed by
__cpufreq_driver_target() (Pengjie Zhang)
- Refine stopped tick handling in the menu cpuidle governor and
rearrange stopped tick handling in the teo cpuidle governor (Rafael
Wysocki)
- Add Panther Lake C-states table to the intel_idle driver (Artem
Bityutskiy)
- Clean up dead dependencies on CPU_IDLE in Kconfig (Julian Braha)
- Simplify cpuidle_register_device() with guard() (Huisong Li)
- Use performance level if available to distinguish between rates in
OPP debugfs (Manivannan Sadhasivam)
- Fix scoped_guard in dev_pm_opp_xlate_required_opp() (Viresh Kumar)
- Return -ENODATA if the snapshot image is not loaded (Alberto Garcia)
- Remove inclusion of crypto/hash.h from hibernate_64.c on x86 (Eric
Biggers)
- Clean up and rearrange the intel_rapl power capping driver to make
the respective interface drivers (TPMI, MSR, and MMOI) hold their
own settings and primitives and consolidate PL4 and PMU support
flags into rapl_defaults (Kuppuswamy Sathyanarayanan)
- Correct kernel-doc function parameter names in the power capping core
code (Randy Dunlap)
- Remove unneeded casting for HZ_PER_KHZ in devfreq (Andy Shevchenko)
- Use _visible attribute to replace create/remove_sysfs_files() in
devfreq (Pengjie Zhang)
- Add Tegra114 support to activity monitor device in tegra30-devfreq as
a preparation to upcoming EMC controller support (Svyatoslav Ryhel)
- Fix mistakes in cpupower man pages, add the boost and epp options to
the cpupower-frequency-info man page, and add the perf-bias option to
the cpupower-info man page (Roberto Ricci)
- Remove unnecessary extern declarations from getopt.h in arguments
parsing functions in cpufreq-set, cpuidle-info, cpuidle-set,
cpupower-info, and cpupower-set utilities (Kaushlendra Kumar)
-----BEGIN PGP SIGNATURE-----
iQFGBAABCAAwFiEEcM8Aw/RY0dgsiRUR7l+9nS/U47UFAmnY9TISHHJqd0Byand5
c29ja2kubmV0AAoJEO5fvZ0v1OO1G9gH/j5mEqfPpiwX6fQ/ZwOGdNOOPVA5w9j4
KPHSMwMD5lZkoaZfasp2vt27KY5SOoVVvRZ2DKkFJ3Jai4I3cUPZYypga2nre1ag
tgzX4vOjcw2r40Eda6ezWl1h4mca/xJJBX7xH2+hn1JY+Y1in37g50CqMIjKh96z
Uugkk6UZytL1XcF55PMhIUgDf6pDtRT5UOW9xOKOkUt8FVWTJ7ei3HaWyV5kDmVq
b5eQ42+OH7y6sWNnoKczFd8fStvh6J/avoJurBEvcOQhMcjaIaB48G19+KjDg73E
NjrVcgG20P2rltBvV2d0J1TKskZHkaP7XjIeWfkwjGZhee3FL7ssS/g=
=fRCO
-----END PGP SIGNATURE-----
Merge tag 'pm-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management updates from Rafael Wysocki:
"Once again, cpufreq is the most active development area, mostly
because of the new feature additions and documentation updates in the
amd-pstate driver, but there are also changes in the cpufreq core
related to boost support and other assorted updates elsewhere.
Next up are power capping changes due to the major cleanup of the
Intel RAPL driver.
On the cpuidle front, a new C-states table for Intel Panther Lake is
added to the intel_idle driver, the stopped tick handling in the menu
and teo governors is updated, and there are a couple of cleanups.
Apart from the above, support for Tegra114 is added to devfreq and
there are assorted cleanups of that code, there are also two updates
of the operating performance points (OPP) library, two minor updates
related to hibernation, and cpupower utility man pages updates and
cleanups.
Specifics:
- Update qcom-hw DT bindings to include Eliza hardware (Abel Vesa)
- Update cpufreq-dt-platdev blocklist (Faruque Ansari)
- Minor updates to driver and dt-bindings for Tegra (Thierry Reding,
Rosen Penev)
- Add MAINTAINERS entry for CPPC driver (Viresh Kumar)
- Add support for new features: CPPC performance priority, Dynamic
EPP, Raw EPP, and new unit tests for them to amd-pstate (Gautham
Shenoy, Mario Limonciello)
- Fix sysfs files being present when HW missing and broken/outdated
documentation in the amd-pstate driver (Ninad Naik, Gautham Shenoy)
- Pass the policy to cpufreq_driver->adjust_perf() to avoid using
cpufreq_cpu_get() in the .adjust_perf() callback in amd-pstate
which leads to a scheduling-while-atomic bug (K Prateek Nayak)
- Clean up dead code in Kconfig for cpufreq (Julian Braha)
- Remove max_freq_req update for pre-existing cpufreq policy and add
a boost_freq_req QoS request to save the boost constraint instead
of overwriting the last scaling_max_freq constraint (Pierre
Gondois)
- Embed cpufreq QoS freq_req objects in cpufreq policy so they all
are allocated in one go along with the policy to simplify lifetime
rules and avoid error handling issues (Viresh Kumar)
- Use DMI max speed when CPPC is unavailable in the acpi-cpufreq
scaling driver (Henry Tseng)
- Switch policy_is_shared() in cpufreq to using cpumask_nth() instead
of cpumask_weight() because the former is more efficient (Yury
Norov)
- Use sysfs_emit() in sysfs show functions for cpufreq governor
attributes (Thorsten Blum)
- Update intel_pstate to stop returning an error when "off" is
written to its status sysfs attribute while the driver is already
off (Fabio De Francesco)
- Include current frequency in the debug message printed by
__cpufreq_driver_target() (Pengjie Zhang)
- Refine stopped tick handling in the menu cpuidle governor and
rearrange stopped tick handling in the teo cpuidle governor (Rafael
Wysocki)
- Add Panther Lake C-states table to the intel_idle driver (Artem
Bityutskiy)
- Clean up dead dependencies on CPU_IDLE in Kconfig (Julian Braha)
- Simplify cpuidle_register_device() with guard() (Huisong Li)
- Use performance level if available to distinguish between rates in
OPP debugfs (Manivannan Sadhasivam)
- Fix scoped_guard in dev_pm_opp_xlate_required_opp() (Viresh Kumar)
- Return -ENODATA if the snapshot image is not loaded (Alberto
Garcia)
- Remove inclusion of crypto/hash.h from hibernate_64.c on x86 (Eric
Biggers)
- Clean up and rearrange the intel_rapl power capping driver to make
the respective interface drivers (TPMI, MSR, and MMOI) hold their
own settings and primitives and consolidate PL4 and PMU support
flags into rapl_defaults (Kuppuswamy Sathyanarayanan)
- Correct kernel-doc function parameter names in the power capping
core code (Randy Dunlap)
- Remove unneeded casting for HZ_PER_KHZ in devfreq (Andy Shevchenko)
- Use _visible attribute to replace create/remove_sysfs_files() in
devfreq (Pengjie Zhang)
- Add Tegra114 support to activity monitor device in tegra30-devfreq
as a preparation to upcoming EMC controller support (Svyatoslav
Ryhel)
- Fix mistakes in cpupower man pages, add the boost and epp options
to the cpupower-frequency-info man page, and add the perf-bias
option to the cpupower-info man page (Roberto Ricci)
- Remove unnecessary extern declarations from getopt.h in arguments
parsing functions in cpufreq-set, cpuidle-info, cpuidle-set,
cpupower-info, and cpupower-set utilities (Kaushlendra Kumar)"
* tag 'pm-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (74 commits)
cpufreq/amd-pstate: Add POWER_SUPPLY select for dynamic EPP
cpupower: remove extern declarations in cmd functions
cpuidle: Simplify cpuidle_register_device() with guard()
PM / devfreq: tegra30-devfreq: add support for Tegra114
PM / devfreq: use _visible attribute to replace create/remove_sysfs_files()
PM / devfreq: Remove unneeded casting for HZ_PER_KHZ
MAINTAINERS: amd-pstate: Step down as maintainer, add Prateek as reviewer
cpufreq: Pass the policy to cpufreq_driver->adjust_perf()
cpufreq/amd-pstate: Pass the policy to amd_pstate_update()
cpufreq/amd-pstate-ut: Add a unit test for raw EPP
cpufreq/amd-pstate: Add support for raw EPP writes
cpufreq/amd-pstate: Add support for platform profile class
cpufreq/amd-pstate: add kernel command line to override dynamic epp
cpufreq/amd-pstate: Add dynamic energy performance preference
Documentation: amd-pstate: fix dead links in the reference section
cpufreq/amd-pstate: Cache the max frequency in cpudata
Documentation/amd-pstate: Add documentation for amd_pstate_floor_{freq,count}
Documentation/amd-pstate: List amd_pstate_prefcore_ranking sysfs file
Documentation/amd-pstate: List amd_pstate_hw_prefcore sysfs file
amd-pstate-ut: Add a testcase to validate the visibility of driver attributes
...
kick_cpus_irq_workfn() warns when scx_kick_syncs is NULL, but this can
legitimately happen when a BPF timer or other kick source races with
free_kick_syncs() during scheduler disable. Drop the pr_warn_once() and
add a comment explaining the race.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Zhao Mengmeng <zhaomengmeng@kylinos.cn>
The kf_tasks[] design assumes task-based SCX ops don't nest - if they
did, kf_tasks[0] would get clobbered. The old scx_kf_allow() WARN_ONCE
caught invalid nesting via kf_mask, but that machinery is gone now.
Add a WARN_ON_ONCE(current->scx.kf_tasks[0]) at the top of each
SCX_CALL_OP_TASK*() macro. Checking kf_tasks[0] alone is sufficient: all
three variants (SCX_CALL_OP_TASK, SCX_CALL_OP_TASK_RET,
SCX_CALL_OP_2TASKS_RET) write to kf_tasks[0], so a non-NULL value at
entry to any of the three means re-entry from somewhere in the family.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Andrea Righi <arighi@nvidia.com>
The "kf_allowed" framing on this helper comes from the old runtime
scx_kf_allowed() gate, which has been removed. Rename it to describe what it
actually does in the new model.
Pure rename, no functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Andrea Righi <arighi@nvidia.com>
Now that scx_kfunc_context_filter enforces context-sensitive kfunc
restrictions at BPF load time, the per-task runtime enforcement via
scx_kf_mask is redundant. Remove it entirely:
- Delete enum scx_kf_mask, the kf_mask field on sched_ext_entity, and
the scx_kf_allow()/scx_kf_disallow()/scx_kf_allowed() helpers along
with the higher_bits()/highest_bit() helpers they used.
- Strip the @mask parameter (and the BUILD_BUG_ON checks) from the
SCX_CALL_OP[_RET]/SCX_CALL_OP_TASK[_RET]/SCX_CALL_OP_2TASKS_RET
macros and update every call site. Reflow call sites that were
wrapped only to fit the old 5-arg form and now collapse onto a single
line under ~100 cols.
- Remove the in-kfunc scx_kf_allowed() runtime checks from
scx_dsq_insert_preamble(), scx_dsq_move(), scx_bpf_dispatch_nr_slots(),
scx_bpf_dispatch_cancel(), scx_bpf_dsq_move_to_local___v2(),
scx_bpf_sub_dispatch(), scx_bpf_reenqueue_local(), and the per-call
guard inside select_cpu_from_kfunc().
scx_bpf_task_cgroup() and scx_kf_allowed_on_arg_tasks() were already
cleaned up in the "drop redundant rq-locked check" patch.
scx_kf_allowed_if_unlocked() was rewritten in the preceding "decouple"
patch. No further changes to those helpers here.
Co-developed-by: Juntong Deng <juntong.deng@outlook.com>
Signed-off-by: Juntong Deng <juntong.deng@outlook.com>
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Move enforcement of SCX context-sensitive kfunc restrictions from per-task
runtime kf_mask checks to BPF verifier-time filtering, using the BPF core's
struct_ops context information.
A shared .filter callback is attached to each context-sensitive BTF set
and consults a per-op allow table (scx_kf_allow_flags[]) indexed by SCX
ops member offset. Disallowed calls are now rejected at program load time
instead of at runtime.
The old model split reachability across two places: each SCX_CALL_OP*()
set bits naming its op context, and each kfunc's scx_kf_allowed() check
OR'd together the bits it accepted. A kfunc was callable when those two
masks overlapped. The new model transposes the result to the caller side -
each op's allow flags directly list the kfunc groups it may call. The old
bit assignments were:
Call-site bits:
ops.select_cpu = ENQUEUE | SELECT_CPU
ops.enqueue = ENQUEUE
ops.dispatch = DISPATCH
ops.cpu_release = CPU_RELEASE
Kfunc-group accepted bits:
enqueue group = ENQUEUE | DISPATCH
select_cpu group = SELECT_CPU | ENQUEUE
dispatch group = DISPATCH
cpu_release group = CPU_RELEASE
Intersecting them yields the reachability now expressed directly by
scx_kf_allow_flags[]:
ops.select_cpu -> SELECT_CPU | ENQUEUE
ops.enqueue -> SELECT_CPU | ENQUEUE
ops.dispatch -> ENQUEUE | DISPATCH
ops.cpu_release -> CPU_RELEASE
Unlocked ops carried no kf_mask bits and reached only unlocked kfuncs;
that maps directly to UNLOCKED in the new table.
Equivalence was checked by walking every (op, kfunc-group) combination
across SCX ops, SYSCALL, and non-SCX struct_ops callers against the old
scx_kf_allowed() runtime checks. With two intended exceptions (see below),
all combinations reach the same verdict; disallowed calls are now caught at
load time instead of firing scx_error() at runtime.
scx_bpf_dsq_move_set_slice() and scx_bpf_dsq_move_set_vtime() are
exceptions: they have no runtime check at all, but the new filter rejects
them from ops outside dispatch/unlocked. The affected cases are nonsensical
- the values these setters store are only read by
scx_bpf_dsq_move{,_vtime}(), which is itself restricted to
dispatch/unlocked, so a setter call from anywhere else was already dead
code.
Runtime scx_kf_mask enforcement is left in place by this patch and removed
in a follow-up.
Original-patch-by: Juntong Deng <juntong.deng@outlook.com>
Original-patch-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_kf_allowed_on_arg_tasks() runs both an scx_kf_allowed(__SCX_KF_RQ_LOCKED)
mask check and a kf_tasks[] check. After the preceding call-site fixes,
every SCX_CALL_OP_TASK*() invocation has kf_mask & __SCX_KF_RQ_LOCKED
non-zero, so the mask check is redundant whenever the kf_tasks[] check
passes. Drop it and simplify the helper to take only @sch and @p.
Fold the locking guarantee into the SCX_CALL_OP_TASK() comment block, which
scx_bpf_task_cgroup() now points to.
No functional change.
Extracted from a larger verifier-time kfunc context filter patch
originally written by Juntong Deng.
Original-patch-by: Juntong Deng <juntong.deng@outlook.com>
Cc: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_kf_allowed_if_unlocked() uses !current->scx.kf_mask as a proxy for "no
SCX-tracked lock held". kf_mask is removed in a follow-up patch, so its two
callers - select_cpu_from_kfunc() and scx_dsq_move() - need another basis.
Add a new bool scx_rq.in_select_cpu, set across the SCX_CALL_OP_TASK_RET
that invokes ops.select_cpu(), to capture the one case where SCX itself
holds no lock but try_to_wake_up() holds @p's pi_lock. Together with
scx_locked_rq(), it expresses the same accepted-context set.
select_cpu_from_kfunc() needs a runtime test because it has to take
different locking paths depending on context. Open-code as a three-way
branch. The unlocked branch takes raw_spin_lock_irqsave(&p->pi_lock)
directly - pi_lock alone is enough for the fields the kfunc reads, and is
lighter than task_rq_lock().
scx_dsq_move() doesn't really need a runtime test - its accepted contexts
could be enforced at verifier load time. But since the runtime state is
already there and using it keeps the upcoming load-time filter simpler, just
write it the same way: (scx_locked_rq() || in_select_cpu) &&
!kf_allowed(DISPATCH).
scx_kf_allowed_if_unlocked() is deleted with the conversions.
No semantic change.
v2: s/No functional change/No semantic change/ - the unlocked path now acquires
pi_lock instead of the heavier task_rq_lock() (Andrea Righi).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
sched_move_task() invokes ops.cgroup_move() inside task_rq_lock(tsk), so
@p's rq lock is held. The SCX_CALL_OP_TASK invocation mislabels this:
- kf_mask = SCX_KF_UNLOCKED (== 0), claiming no lock is held.
- rq = NULL, so update_locked_rq() doesn't run and scx_locked_rq()
returns NULL.
Switch to SCX_KF_REST and pass task_rq(p), matching ops.set_cpumask()
from set_cpus_allowed_scx().
Three effects:
- scx_bpf_task_cgroup() becomes callable (was rejected by
scx_kf_allowed(__SCX_KF_RQ_LOCKED)). Safe; rq lock is held.
- scx_bpf_dsq_move() is now rejected (was allowed via the unlocked
branch). Calling it while holding an unrelated task's rq lock is
risky; rejection is correct.
- scx_bpf_select_cpu_*() previously took the unlocked branch in
select_cpu_from_kfunc() and called task_rq_lock(p, &rf), which
would deadlock against the already-held pi_lock. Now it takes the
locked-rq branch and is rejected with -EPERM via the existing
kf_allowed(SCX_KF_SELECT_CPU | SCX_KF_ENQUEUE) check. Latent
deadlock fix.
No in-tree scheduler is known to call any of these from ops.cgroup_move().
v2: Add Fixes: tag (Andrea Righi).
Fixes: 18853ba782 ("sched_ext: Track currently locked rq")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
select_cpu_from_kfunc() has an extra scx_kf_allowed_if_unlocked() branch
that accepts calls from unlocked contexts and takes task_rq_lock() itself
- a "callable from unlocked" property encoded in the kfunc body rather
than in set membership. That's fine while the runtime check is the
authoritative gate, but the upcoming verifier-time filter uses set
membership as the source of truth and needs it to reflect every context
the kfunc may be called from.
Add the three select_cpu kfuncs to scx_kfunc_ids_unlocked so their full
set of callable contexts is captured by set membership. This follows the
existing dual-set convention used by scx_bpf_dsq_move{,_vtime} and
scx_bpf_dsq_move_set_{slice,vtime}, which are members of both
scx_kfunc_ids_dispatch and scx_kfunc_ids_unlocked.
While at it, add brief comments on each duplicate BTF_ID_FLAGS block
(including the pre-existing dsq_move ones) explaining the dual
membership.
No runtime behavior change: the runtime check in select_cpu_from_kfunc()
remains the authoritative gate until it is removed along with the rest
of the scx_kf_mask enforcement in a follow-up.
v2: Clarify dispatch-set comment to name scx_bpf_dsq_move*() explicitly so it
doesn't appear to cover scx_bpf_sub_dispatch() (Andrea Righi).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The select_cpu kfuncs - scx_bpf_select_cpu_dfl(), scx_bpf_select_cpu_and()
and __scx_bpf_select_cpu_and() - take task_rq_lock() internally. Exposing
them via scx_kfunc_set_idle to BPF_PROG_TYPE_TRACING is unsafe: arbitrary
tracing contexts (kprobes, tracepoints, fentry, LSM) may run with @p's
pi_lock state unknown.
Move them out of scx_kfunc_ids_idle into a new scx_kfunc_ids_select_cpu
set registered only for STRUCT_OPS and SYSCALL.
Extracted from a larger verifier-time kfunc context filter patch
originally written by Juntong Deng.
Original-patch-by: Juntong Deng <juntong.deng@outlook.com>
Cc: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Merge cpufreq updates for 7.1-rc1:
- Update qcom-hw DT bindings to include Eliza hardware (Abel Vesa)
- Update cpufreq-dt-platdev blocklist (Faruque Ansari)
- Minor updates to driver and dt-bindings for Tegra (Thierry Reding,
Rosen Penev)
- Add MAINTAINERS entry for CPPC driver (Viresh Kumar)
- Add support for new features: CPPC performance priority, Dynamic EPP,
Raw EPP, and new unit tests for them to amd-pstate (Gautham Shenoy,
Mario Limonciello)
- Fix sysfs files being present when HW missing and broken/outdated
documentation in the amd-pstate driver (Ninad Naik, Gautham Shenoy)
- Pass the policy to cpufreq_driver->adjust_perf() to avoid using
cpufreq_cpu_get() in the .adjust_perf() callback in amd-pstate which
leads to a scheduling-while-atomic bug (K Prateek Nayak)
- Clean up dead code in Kconfig for cpufreq (Julian Braha)
- Remove max_freq_req update for pre-existing cpufreq policy and add a
boost_freq_req QoS request to save the boost constraint instead of
overwriting the last scaling_max_freq constraint (Pierre Gondois)
- Embed cpufreq QoS freq_req objects in cpufreq policy so they all
are allocated in one go along with the policy to simplify lifetime
rules and avoid error handling issues (Viresh Kumar)
- Use DMI max speed when CPPC is unavailable in the acpi-cpufreq
scaling driver (Henry Tseng)
- Switch policy_is_shared() in cpufreq to using cpumask_nth() instead
of cpumask_weight() because the former is more efficient (Yury Norov)
- Use sysfs_emit() in sysfs show functions for cpufreq governor
attributes (Thorsten Blum)
- Update intel_pstate to stop returning an error when "off" is written
to its status sysfs attribute while the driver is already off (Fabio
De Francesco)
- Include current frequency in the debug message printed by
__cpufreq_driver_target() (Pengjie Zhang)
* pm-cpufreq: (38 commits)
cpufreq/amd-pstate: Add POWER_SUPPLY select for dynamic EPP
MAINTAINERS: amd-pstate: Step down as maintainer, add Prateek as reviewer
cpufreq: Pass the policy to cpufreq_driver->adjust_perf()
cpufreq/amd-pstate: Pass the policy to amd_pstate_update()
cpufreq/amd-pstate-ut: Add a unit test for raw EPP
cpufreq/amd-pstate: Add support for raw EPP writes
cpufreq/amd-pstate: Add support for platform profile class
cpufreq/amd-pstate: add kernel command line to override dynamic epp
cpufreq/amd-pstate: Add dynamic energy performance preference
Documentation: amd-pstate: fix dead links in the reference section
cpufreq/amd-pstate: Cache the max frequency in cpudata
Documentation/amd-pstate: Add documentation for amd_pstate_floor_{freq,count}
Documentation/amd-pstate: List amd_pstate_prefcore_ranking sysfs file
Documentation/amd-pstate: List amd_pstate_hw_prefcore sysfs file
amd-pstate-ut: Add a testcase to validate the visibility of driver attributes
amd-pstate-ut: Add module parameter to select testcases
amd-pstate: Introduce a tracepoint trace_amd_pstate_cppc_req2()
amd-pstate: Add sysfs support for floor_freq and floor_count
amd-pstate: Add support for CPPC_REQ2 and FLOOR_PERF
x86/cpufeatures: Add AMD CPPC Performance Priority feature.
...
Provide a debugfs directory llc_balancing, and a knob named
"enabled" under it to allow the user to turn off and on the
cache aware scheduling at runtime.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/0aa56f7fc48db2f8f700cd1aa34dedd0ec88351b.1775065312.git.tim.c.chen@linux.intel.com
Introduce sched_cache_present to enable cache aware scheduling for
multi LLCs NUMA node Cache-aware load balancing should only be
enabled if there are more than 1 LLCs within 1 NUMA node.
sched_cache_present is introduced to indicate whether this
platform supports this topology.
Test results:
The first test platform is a 2 socket Intel Sapphire Rapids with 30
cores per socket. The DRAM interleaving is enabled in the BIOS so it
essential has one NUMA node with two last level caches. There are 60
CPUs associated with each last level cache.
The second test platform is a AMD Genoa. There are 4 Nodes and 32 CPUs
per node. Each node has 2 CCXs and each CCX has 16 CPUs.
hackbench/schbench/netperf/stream/stress-ng/chacha20 were launched
on these two platforms.
[TL;DR]
Sappire Rapids:
hackbench shows significant improvement when the number of
different active threads is below the capacity of a LLC.
schbench shows limitted wakeup latency improvement.
ChaCha20-xiangshan(risc-v simulator) shows good throughput
improvement. No obvious difference was observed in
netperf/stream/stress-ng in Hmean.
Genoa:
Significant improvement is observed in hackbench when
the active number of threads is lower than the number
of CPUs within 1 LLC. On v2, Aaron reported improvement
of hackbench/redis when system is underloaded.
ChaCha20-xiangshan shows huge throughput improvement.
Phoronix has tested v1 and shows good improvements in 30+
cases[3]. No obvious difference was observed in
netperf/stream/stress-ng in Hmean.
Detail:
Due to length constraints, data without much difference with
baseline is not presented.
Sapphire Rapids:
[hackbench pipe]
================
case load baseline(std%) compare%( std%)
threads-pipe-10 1-groups 1.00 ( 1.22) +26.09 ( 1.10)
threads-pipe-10 2-groups 1.00 ( 4.90) +22.88 ( 0.18)
threads-pipe-10 4-groups 1.00 ( 2.07) +9.00 ( 3.49)
threads-pipe-10 8-groups 1.00 ( 8.13) +3.45 ( 3.62)
threads-pipe-16 1-groups 1.00 ( 2.11) +26.30 ( 0.08)
threads-pipe-16 2-groups 1.00 ( 15.13) -1.77 ( 11.89)
threads-pipe-16 4-groups 1.00 ( 4.37) +0.58 ( 7.99)
threads-pipe-16 8-groups 1.00 ( 2.88) +2.71 ( 3.50)
threads-pipe-2 1-groups 1.00 ( 9.40) +22.07 ( 0.71)
threads-pipe-2 2-groups 1.00 ( 9.99) +18.01 ( 0.95)
threads-pipe-2 4-groups 1.00 ( 3.98) +24.66 ( 0.96)
threads-pipe-2 8-groups 1.00 ( 7.00) +21.83 ( 0.23)
threads-pipe-20 1-groups 1.00 ( 1.03) +28.84 ( 0.21)
threads-pipe-20 2-groups 1.00 ( 4.42) +31.90 ( 3.15)
threads-pipe-20 4-groups 1.00 ( 9.97) +4.56 ( 1.69)
threads-pipe-20 8-groups 1.00 ( 1.87) +1.25 ( 0.74)
threads-pipe-4 1-groups 1.00 ( 4.48) +25.67 ( 0.78)
threads-pipe-4 2-groups 1.00 ( 9.14) +4.91 ( 2.08)
threads-pipe-4 4-groups 1.00 ( 7.68) +19.36 ( 1.53)
threads-pipe-4 8-groups 1.00 ( 10.79) +7.20 ( 12.20)
threads-pipe-8 1-groups 1.00 ( 4.69) +21.93 ( 0.03)
threads-pipe-8 2-groups 1.00 ( 1.16) +25.29 ( 0.65)
threads-pipe-8 4-groups 1.00 ( 2.23) -1.27 ( 3.62)
threads-pipe-8 8-groups 1.00 ( 4.65) -3.08 ( 2.75)
Note: The default number of fd in hackbench is changed from 20 to various
values to ensure that threads fit within a single LLC, especially on AMD
systems. Take "threads-pipe-8, 2-groups" for example, the number of fd
is 8, and 2 groups are created.
[schbench]
The 99th percentile wakeup latency shows some improvements when the
system is underload, while it does not bring much difference with
the increasing of system utilization.
99th Wakeup Latencies Base (mean std) Compare (mean std) Change
=========================================================================
thread=2 9.00(0.00) 9.00(1.73) 0.00%
thread=4 7.33(0.58) 6.33(0.58) +13.64%
thread=8 9.00(0.00) 7.67(1.15) +14.78%
thread=16 8.67(0.58) 8.67(1.53) 0.00%
thread=32 9.00(0.00) 7.00(0.00) +22.22%
thread=64 9.33(0.58) 9.67(0.58) -3.64%
thread=128 12.00(0.00) 12.00(0.00) 0.00%
[chacha20 on simulated risc-v]
baseline:
Host time spent: 67861ms
cache aware scheduling enabled:
Host time spent: 54441ms
Time reduced by 24%
Genoa:
[hackbench pipe]
The default number of fd is 20, which exceed the number of CPUs
in a LLC. So the fd is adjusted to 2, 4, 6, 8, 20 respectively.
Exclude the result with large run-to-run variance, 10% ~ 50%
improvement is observed when the system is underloaded:
[hackbench pipe]
================
case load baseline(std%) compare%( std%)
threads-pipe-2 1-groups 1.00 ( 2.89) +47.33 ( 1.20)
threads-pipe-2 2-groups 1.00 ( 3.88) +39.82 ( 0.61)
threads-pipe-2 4-groups 1.00 ( 8.76) +5.57 ( 13.10)
threads-pipe-20 1-groups 1.00 ( 4.61) +11.72 ( 1.06)
threads-pipe-20 2-groups 1.00 ( 6.18) +14.55 ( 1.47)
threads-pipe-20 4-groups 1.00 ( 2.99) +10.16 ( 4.49)
threads-pipe-4 1-groups 1.00 ( 4.23) +43.70 ( 2.14)
threads-pipe-4 2-groups 1.00 ( 3.68) +8.45 ( 4.04)
threads-pipe-4 4-groups 1.00 ( 17.72) +2.42 ( 1.14)
threads-pipe-6 1-groups 1.00 ( 3.10) +7.74 ( 3.83)
threads-pipe-6 2-groups 1.00 ( 3.42) +14.26 ( 4.53)
threads-pipe-6 4-groups 1.00 ( 10.34) +10.94 ( 7.12)
threads-pipe-8 1-groups 1.00 ( 4.21) +9.06 ( 4.43)
threads-pipe-8 2-groups 1.00 ( 1.88) +3.74 ( 0.58)
threads-pipe-8 4-groups 1.00 ( 2.78) +23.96 ( 1.18)
[chacha20 on simulated risc-v]
Host time spent: 54762ms
Host time spent: 28295ms
Time reduced by 48%
Suggested-by: Libo Chen <libchen@purestorage.com>
Suggested-by: Adam Li <adamli@os.amperecomputing.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/71972e12ab4f08aff422b31e34df09bdbd94de84.1775065312.git.tim.c.chen@linux.intel.com
During load balancing, make can_migrate_task()
consider a task's LLC preference.
Prevent a task from being moved out of its preferred LLC.
During the regular load balancing, if
the task cannot be migrated due to LLC locality, the
nr_balance_failed also should not be increased.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/53da65f3d59de31e1a1dc59a4093d8dd9d4dc206.1775065312.git.tim.c.chen@linux.intel.com
Cache aware scheduling mainly does two things:
1. Prevent task from migrating out of its preferred LLC if not
nessasary.
2. Migrating task to their preferred LLC if nessasary.
For 1:
In the generic load balance, if the busiest runqueue has only one task,
active balancing may be invoked to move it away. However, this migration
might break LLC locality.
Prevent regular load balance from migrating a task that
prefers the current LLC. The load level and imbalance do not warrant
breaking LLC preference per the can_migrate_llc() policy. Here, the
benefit of LLC locality outweighs the power efficiency gained from
migrating the only runnable task away.
Before migration, check whether the task is running on its preferred
LLC: Do not move a lone task to another LLC if it would move the task
away from its preferred LLC or cause excessive imbalance between LLCs.
For 2:
On the other hand, if the migration type is migrate_llc_task, it means
that there are tasks on the env->src_cpu that want to be migrated to
their preferred LLC, launch the active load balance anyway.
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/9b816d8c27fabf2a9c0e1f61a6b90afe8ec4ad52.1775065312.git.tim.c.chen@linux.intel.com
Introduce a new migration type, migrate_llc_task, to support
cache-aware load balancing.
After identifying the busiest sched_group (having the most tasks
preferring the destination LLC), mark migrations with this type.
During load balancing, each runqueue in the busiest sched_group is
examined, and the runqueue with the highest number of tasks preferring
the destination CPU is selected as the busiest runqueue.
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/b9df27c19cc5121ddb2a7d1be7f9d52fec1563dc.1775065312.git.tim.c.chen@linux.intel.com
During LLC load balancing, first check for tasks that prefer the
destination LLC and balance them to it before others.
Mark source sched groups containing tasks preferring non local LLCs
with the group_llc_balance flag. This ensures the load balancer later
pulls or pushes these tasks toward their preferred LLCs.
The priority of group_llc_balance is lower than that of group_overloaded
and higher than that of all other group types. This is because
group_llc_balance may exacerbate load imbalance, and if the LLC balancing
attempt fails, the nr_balance_failed mechanism will trigger other group
types to rebalance the load.
The load balancer selects the busiest sched_group and migrates tasks
to less busy groups to distribute load across CPUs.
With cache-aware scheduling enabled, the busiest sched_group is
the one with most tasks preferring the destination LLC. If
the group has the llc_balance flag set, cache aware load balancing is
triggered.
Introduce the helper function update_llc_busiest() to identify the
sched_group with the most tasks preferring the destination LLC.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Suggested-by: Madadi Vineeth Reddy <vineethr@linux.ibm.com>
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/baa458f45eab3f602af090c6d6af63dc864f5ec6.1775065312.git.tim.c.chen@linux.intel.com
There is no need to check the local group twice for both
group_asym_packing and group_smt_balance. Adjust the code
to facilitate future checks for group types (cache-aware
load balancing) as well.
No functional changes are expected.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/99a57865c8ae1847087a5c00e92d24351cf3e5a8.1775065312.git.tim.c.chen@linux.intel.com
During LLC load balancing, tabulate the number of tasks on each runqueue
that prefer the LLC contains the env->dst_cpu in a sched group.
For example, consider a system with 4 LLC sched groups (LLC0 to LLC3)
balancing towards LLC3. LLC0 has 3 tasks preferring LLC3, LLC1 has
2, and LLC2 has 1. LLC0, having the most tasks preferring LLC3, is
selected as the busiest source to pick tasks from.
Within a source LLC, the total number of tasks preferring a destination
LLC is computed by summing counts across all CPUs in that LLC. For
instance, if LLC0 has CPU0 with 2 tasks and CPU1 with 1 task preferring
LLC3, the total for LLC0 is 3.
These statistics allow the load balancer to choose tasks from source
sched groups that best match their preferred LLCs.
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/3d8502a33a753c4384b368f97f64ee70b1cea0db.1775065312.git.tim.c.chen@linux.intel.com
Calculate the number of tasks' LLC preferences for each runqueue.
This statistic is computed during task enqueue and dequeue
operations, and is used by the cache-aware load balancing.
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/d15a64436d3acd19c5c53344c5e9d3d0b79b3233.1775065312.git.tim.c.chen@linux.intel.com
The lowest level of sched domain for each CPU is assigned an
array where each element tracks the number of tasks preferring
a given LLC, indexed from 0 to max_lid. Since each CPU
has its dedicated sd, this implies that each CPU will have
a dedicated task LLC preference counter.
For example, sd->llc_counts[3] = 2 signifies that there
are 2 tasks on this runqueue which prefer to run within LLC3.
The load balancer can use this information to identify busy
runqueues and migrate tasks to their preferred LLC domains.
This array will be reallocated at runtime during sched domain
rebuild.
Introduce the buffer allocation mechanism, and the statistics
will be calculated in the subsequent patch.
Note: the LLC preference statistics of each CPU are reset on
sched domain rebuild and may under count temporarily, until the
CPU becomes idle and the count is cleared. This is a trade off
to avoid complex data synchronization across sched domain builds.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/42e79eceb8cd6be8a032401d481d101913bc5703.1775065312.git.tim.c.chen@linux.intel.com
For each runqueue, track the number of tasks with an LLC preference
and how many of them are running on their preferred LLC. This mirrors
nr_numa_running and nr_preferred_running for NUMA balancing, and will
be used by cache-aware load balancing in later patches.
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/459a37102f3d74a4e09ea58401d2094ac731d044.1775065312.git.tim.c.chen@linux.intel.com
With cache-aware scheduling enabled, each task is assigned a
preferred LLC ID. This allows quick identification of the LLC domain
where the task prefers to run, similar to numa_preferred_nid in
NUMA balancing.
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/f2ceecba5858680349ad4ce9303a2121f0bb7272.1775065312.git.tim.c.chen@linux.intel.com
Introduce an index mapping between CPUs and their LLCs. This provides
a roughly continuous per LLC index needed for cache-aware load balancing in
later patches.
The existing per_cpu llc_id usually points to the first CPU of the
LLC domain, which is sparse and unsuitable as an array index. Using
llc_id directly would waste memory.
With the new mapping, CPUs in the same LLC share an approximate
continuous id:
per_cpu(llc_id, CPU=0...15) = 0
per_cpu(llc_id, CPU=16...31) = 1
per_cpu(llc_id, CPU=32...47) = 2
...
Note that the LLC IDs are allocated via bitmask, so the IDs may be
reused during CPU offline->online transitions.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Originally-by: K Prateek Nayak <kprateek.nayak@amd.com>
Co-developed-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/047ef46339e4db497b54a89940a7ebedf27fcf28.1775065312.git.tim.c.chen@linux.intel.com
Cache-aware scheduling aggregates threads onto their preferred LLC,
mainly through load balancing. When the preferred LLC becomes
saturated, more threads are still placed there, increasing latency.
A mechanism is needed to limit aggregation so that the preferred LLC
does not become overloaded.
Introduce helper functions can_migrate_llc() and
can_migrate_llc_task() to enforce the LLC migration policy:
1. Aggregate a task to its preferred LLC if both source and
destination LLCs are not too busy, or if doing so will not
leave the preferred LLC much more imbalanced than the
non-preferred one (>20% utilization difference, a little
higher than the default imbalance_pct(17%) of the LLC domain
as hysteresis). Later this threshold will be turned into tunable
debugfs.
2. Allow moving a task from overloaded preferred LLC to a non
preferred LLC if this will not cause the non preferred LLC
to become too imbalanced to cause a later migration back.
3. If both LLCs are too busy, let the generic load balance to
spread the tasks.
Further (hysteresis)action could be taken in the future to prevent tasks
from being migrated into and out of the preferred LLC frequently (back and
forth): the threshold for migrating a task out of its preferred LLC should
be higher than that for migrating it into the LLC.
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/d19b52589cdceaee5e625980959f4d1982d6d7c9.1775065312.git.tim.c.chen@linux.intel.com
When a system becomes busy and a process's preferred LLC is
saturated with too many threads, tasks within that LLC migrate
frequently. These in LLC migrations introduce latency and degrade
performance. To avoid this, task aggregation should be suppressed
when the preferred LLC is overloaded, which requires a metric to
indicate LLC utilization.
Record per LLC utilization/cpu capacity during periodic load
balancing. These statistics will be used in later patches to decide
whether tasks should be aggregated into their preferred LLC.
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/a48151b3d57f2a42a5971aaead1b7f81e69229f4.1775065312.git.tim.c.chen@linux.intel.com
When NUMA balancing is enabled, the kernel currently iterates over all
online CPUs to aggregate process-wide occupancy data. On large systems,
this global scan introduces significant overhead.
To reduce scan latency, limit the search to a subset of relevant CPUs:
1. The task's preferred NUMA node.
2. The node where the task is currently running.
3. The node that contains the task's current preferred LLC..
While focusing solely on the preferred NUMA node is ideal, a
process-wide scan must remain flexible because the "preferred node"
is a per-task attribute. Different threads within the same process may
have different preferred nodes, causing the process-wide preference to
migrate. Maintaining a mask that covers both the preferred and active
running nodes ensures accuracy while significantly reducing the number of
CPUs inspected.
Future work may integrate numa_group to further refine task aggregation.
Suggested-by: Madadi Vineeth Reddy <vineethr@linux.ibm.com>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Co-developed-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/57ed5fcec9b242803fe4ea2ce6e7f3de6a6efc6b.1775065312.git.tim.c.chen@linux.intel.com
Adds infrastructure to enable cache-aware load balancing,
which improves cache locality by grouping tasks that share resources
within the same cache domain. This reduces cache misses and improves
overall data access efficiency.
In this initial implementation, threads belonging to the same process
are treated as entities that likely share working sets. The mechanism
tracks per-process CPU occupancy across cache domains and attempts to
migrate threads toward cache-hot domains where their process already
has active threads, thereby enhancing locality.
This provides a basic model for cache affinity. While the current code
targets the last-level cache (LLC), the approach could be extended to
other domain types such as clusters (L2) or node-internal groupings.
At present, the mechanism selects the CPU within an LLC that has the
highest recent runtime. Subsequent patches in this series will use this
information in the load-balancing path to guide task placement toward
preferred LLCs.
In the future, more advanced policies could be integrated through NUMA
balancing-for example, migrating a task to its preferred LLC when spare
capacity exists, or swapping tasks across LLCs to improve cache affinity.
Grouping of tasks could also be generalized from that of a process
to be that of a NUMA group, or be user configurable.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/6269a53221b9439b9ca00d18a9d1946fb64d8cff.1775065312.git.tim.c.chen@linux.intel.com
The sched_rt_global_constraints() function is a remnant that used to set
up global RT throttling but that is no more since commit 5f6bd380c7
("sched/rt: Remove default bandwidth control") and the function ended up
only doing schedulability check.
Move the check into the validation function where it fits better.
(The order of validations sched_dl_global_validate() and
sched_rt_global_validate() shouldn't matter.)
Signed-off-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260323-sched-rert_groups-v3-2-1e7d5ed6b249@suse.com
The warning from the commit 87f1fb77d8 ("sched: Add RT_GROUP WARN
checks for non-root task_groups") is wrong -- it assumes that only
task_groups with rt_rq are traversed, however, the schedulability check
would iterate all task_groups even when rt_group_sched=0 is disabled at
boot time but some non-root task_groups exist.
The schedulability check is supposed to validate:
a) that children don't overcommit its parent,
b) no RT task group overcommits global RT limit.
but with rt_group_sched=0 there is no (non-trivial) hierarchy of RT groups,
therefore skip the validation altogether. Otherwise, writes to the
global sched_rt_runtime_us knob will be rejected with incorrect
validation error.
This fix is immaterial with CONFIG_RT_GROUP_SCHED=n.
Fixes: 87f1fb77d8 ("sched: Add RT_GROUP WARN checks for non-root task_groups")
Signed-off-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260323-sched-rert_groups-v3-1-1e7d5ed6b249@suse.com
John noted that commit 1151354225 ("sched/deadline: Fix 'stuck' dl_server")
unfixed the issue from commit a3a70caf79 ("sched/deadline: Fix dl_server
behaviour").
The issue in commit 1151354225 was for wakeups of the server after the
deadline; in which case you *have* to start a new period. The case for
a3a70caf79 is wakeups before the deadline.
Now, because the server is effectively running a least-laxity policy, it means
that any wakeup during the runnable phase means dl_entity_overflow() will be
true. This means we need to adjust the runtime to allow it to still run until
the existing deadline expires.
Use the revised wakeup rule for dl_defer entities.
Fixes: 1151354225 ("sched/deadline: Fix 'stuck' dl_server")
Reported-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Tested-by: John Stultz <jstultz@google.com>
Link: https://patch.msgid.link/20260404102244.GB22575@noisy.programming.kicks-ass.net
Here is one scenario which was triggered when running:
stress-ng --yield=32 -t 10000000s&
while true; do perf bench sched messaging -p -t -l 100000 -g 16; done
on a 256CPUs machine after about an hour into the run:
__enqeue_entity: entity_key(-141245081754) weight(90891264) overflow_mul(5608800059305154560) vlag(57498) delayed?(0)
cfs_rq: zero_vruntime(3809707759657809) sum_w_vruntime(0) sum_weight(0) nr_queued(1)
cfs_rq->curr: entity_key(0) vruntime(3809707759657809) deadline(3809723966988476) weight(37)
The above comes from __enqueue_entity() after a place_entity(). Breaking
this down:
vlag_initial = 57498
vlag = (57498 * (37 + 90891264)) / 37 = 141,245,081,754
vruntime = 3809707759657809 - 141245081754 = 3,809,566,514,576,055
entity_key(se, cfs_rq) = -141,245,081,754
Now, multiplying the entity_key with its own weight results to
5,608,800,059,305,154,560 (same as what overflow_mul() suggests) but
in Python, without overflow, this would be: -1,2837,944,014,404,397,056
Avoid the overflow (without doing the division for avg_vruntime()), by moving
zero_vruntime to the new entity when it is heavier.
Fixes: 4823725d9d ("sched/fair: Increase weight bits for avg_vruntime")
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
[peterz: suggested 'weight > load' condition]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260407120052.GG3738010@noisy.programming.kicks-ass.net
to_ratio() computes BW_SHIFT-scaled bandwidth ratios from u64 period and
runtime values, but it returns unsigned long. tg_rt_schedulable() also
stores the current group limit and the accumulated child sum in unsigned
long.
On 32-bit builds, large bandwidth ratios can be truncated and the RT
group sum can wrap when enough siblings are present. That can let an
overcommitted RT hierarchy pass the schedulability check, and it also
narrows the helper result for other callers.
Return u64 from to_ratio() and use u64 for the RT group totals so
bandwidth ratios are preserved and compared at full width on both 32-bit
and 64-bit builds.
Fixes: b40b2e8eb5 ("sched: rt: multi level group constraints")
Assisted-by: Codex:GPT-5
Signed-off-by: Joseph Salisbury <joseph.salisbury@oracle.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260403210014.2713404-1-joseph.salisbury@oracle.com
- Fix stale direct dispatch state in ddsp_dsq_id which can cause
spurious warnings in mark_direct_dispatch() on task wakeup.
- Fix is_bpf_migration_disabled() false negative on non-PREEMPT_RCU
configs which can lead to incorrectly dispatching migration-disabled
tasks to remote CPUs.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCac//0w4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGdqUAP9kEuxvB+pxjheSKV0j7zvDHd+ksMxjQTRoBmyu
PE0hIgEA5gAax8ebef9MlyRVsm9Qh7v/AmovUHt75oeCnDk++Ag=
=hD7A
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-7.0-rc6-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
"These are late but both fix subtle yet critical problems and the blast
radius is limited strictly to sched_ext.
- Fix stale direct dispatch state in ddsp_dsq_id which can cause
spurious warnings in mark_direct_dispatch() on task wakeup
- Fix is_bpf_migration_disabled() false negative on non-PREEMPT_RCU
configs which can lead to incorrectly dispatching migration-
disabled tasks to remote CPUs"
* tag 'sched_ext-for-7.0-rc6-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: Fix stale direct dispatch state in ddsp_dsq_id
sched_ext: Fix is_bpf_migration_disabled() false negative on non-PREEMPT_RCU
Conflict in kernel/sched/ext.c between:
7e0ffb72de ("sched_ext: Fix stale direct dispatch state in
ddsp_dsq_id")
which clears ddsp state at individual call sites instead of
dispatch_enqueue(), and sub-sched related code reorg and API updates on
for-7.1. Resolved by applying the ddsp fix with for-7.1's signatures.
Signed-off-by: Tejun Heo <tj@kernel.org>
@p->scx.ddsp_dsq_id can be left set (non-SCX_DSQ_INVALID) triggering a
spurious warning in mark_direct_dispatch() when the next wakeup's
ops.select_cpu() calls scx_bpf_dsq_insert(), such as:
WARNING: kernel/sched/ext.c:1273 at scx_dsq_insert_commit+0xcd/0x140
The root cause is that ddsp_dsq_id was only cleared in dispatch_enqueue(),
which is not reached in all paths that consume or cancel a direct dispatch
verdict.
Fix it by clearing it at the right places:
- direct_dispatch(): cache the direct dispatch state in local variables
and clear it before dispatch_enqueue() on the synchronous path. For
the deferred path, the direct dispatch state must remain set until
process_ddsp_deferred_locals() consumes them.
- process_ddsp_deferred_locals(): cache the dispatch state in local
variables and clear it before calling dispatch_to_local_dsq(), which
may migrate the task to another rq.
- do_enqueue_task(): clear the dispatch state on the enqueue path
(local/global/bypass fallbacks), where the direct dispatch verdict is
ignored.
- dequeue_task_scx(): clear the dispatch state after dispatch_dequeue()
to handle both the deferred dispatch cancellation and the holding_cpu
race, covering all cases where a pending direct dispatch is
cancelled.
- scx_disable_task(): clear the direct dispatch state when
transitioning a task out of the current scheduler. Waking tasks may
have had the direct dispatch state set by the outgoing scheduler's
ops.select_cpu() and then been queued on a wake_list via
ttwu_queue_wakelist(), when SCX_OPS_ALLOW_QUEUED_WAKEUP is set. Such
tasks are not on the runqueue and are not iterated by scx_bypass(),
so their direct dispatch state won't be cleared. Without this clear,
any subsequent SCX scheduler that tries to direct dispatch the task
will trigger the WARN_ON_ONCE() in mark_direct_dispatch().
Fixes: 5b26f7b920 ("sched_ext: Allow SCX_DSQ_LOCAL_ON for direct dispatches")
Cc: stable@vger.kernel.org # v6.12+
Cc: Daniel Hodges <hodgesd@meta.com>
Cc: Patrick Somaru <patsomaru@meta.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Delayed dequeue feature aims to reduce the negative lag of a dequeued
task while sleeping but it can happens that newly enqueued tasks will
move backward the avg vruntime and increase its negative lag.
When the delayed dequeued task wakes up, it has more neg lag compared
to being dequeued immediately or to other tasks that have been
dequeued just before theses new enqueues.
Ensure that the negative lag of a delayed dequeued task doesn't
increase during its delayed dequeued phase while waiting for its neg
lag to diseappear. Similarly, we remove any positive lag that the
delayed dequeued task could have gain during thsi period.
Short slice tasks are particularly impacted in overloaded system.
Test on snapdragon rb5:
hackbench -T -p -l 16000000 -g 2 1> /dev/null &
cyclictest -t 1 -i 2777 -D 333 --policy=fair --mlock -h 20000 -q
The scheduling latency of cyclictest is:
tip/sched/core tip/sched/core +this patch
cyclictest slice (ms) (default)2.8 8 8
hackbench slice (ms) (default)2.8 20 20
Total Samples | 115632 119733 119806
Average (us) | 364 64(-82%) 61(- 5%)
Median (P50) (us) | 60 56(- 7%) 56( 0%)
90th Percentile (us) | 1166 62(-95%) 62( 0%)
99th Percentile (us) | 4192 73(-98%) 72(- 1%)
99.9th Percentile (us) | 8528 2707(-68%) 1300(-52%)
Maximum (us) | 17735 14273(-20%) 13525(- 5%)
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260331162352.551501-1-vincent.guittot@linaro.org
Use helper sched_energy_enabled() everywhere we want to test if EAS is
enabled instead of mixing sched_energy_enabled() and direct call to
static_branch_unlikely().
No functional change
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260327132013.2800517-1-vincent.guittot@linaro.org
Add logic to handle migrating a blocked waiter to a remote
cpu where the lock owner is runnable.
Additionally, as the blocked task may not be able to run
on the remote cpu, add logic to handle return migration once
the waiting task is given the mutex.
Because tasks may get migrated to where they cannot run, also
modify the scheduling classes to avoid sched class migrations on
mutex blocked tasks, leaving find_proxy_task() and related logic
to do the migrations and return migrations.
This was split out from the larger proxy patch, and
significantly reworked.
Credits for the original patch go to:
Peter Zijlstra (Intel) <peterz@infradead.org>
Juri Lelli <juri.lelli@redhat.com>
Valentin Schneider <valentin.schneider@arm.com>
Connor O'Brien <connoro@google.com>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260324191337.1841376-11-jstultz@google.com
The fair scheduler locally introduced attach_one_task() and
attach_task() helpers, but these could be generically useful so
move this code to sched.h so we can use them elsewhere.
One minor tweak made to utilize guard(rq_lock)(rq) to simplifiy
the function.
Suggested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260324191337.1841376-10-jstultz@google.com
With proxy-exec, a task is selected to run via pick_next_task(),
and then if it is a mutex blocked task, we call find_proxy_task()
to find a runnable owner. If the runnable owner is on another
cpu, we will need to migrate the selected donor task away, after
which we will pick_again can call pick_next_task() to choose
something else.
However, in the first call to pick_next_task(), we may have
had a balance_callback setup by the class scheduler. After we
pick again, its possible pick_next_task_fair() will be called
which calls sched_balance_newidle() and sched_balance_rq().
This will throw a warning:
[ 8.796467] rq->balance_callback && rq->balance_callback != &balance_push_callback
[ 8.796467] WARNING: CPU: 32 PID: 458 at kernel/sched/sched.h:1750 sched_balance_rq+0xe92/0x1250
...
[ 8.796467] Call Trace:
[ 8.796467] <TASK>
[ 8.796467] ? __warn.cold+0xb2/0x14e
[ 8.796467] ? sched_balance_rq+0xe92/0x1250
[ 8.796467] ? report_bug+0x107/0x1a0
[ 8.796467] ? handle_bug+0x54/0x90
[ 8.796467] ? exc_invalid_op+0x17/0x70
[ 8.796467] ? asm_exc_invalid_op+0x1a/0x20
[ 8.796467] ? sched_balance_rq+0xe92/0x1250
[ 8.796467] sched_balance_newidle+0x295/0x820
[ 8.796467] pick_next_task_fair+0x51/0x3f0
[ 8.796467] __schedule+0x23a/0x14b0
[ 8.796467] ? lock_release+0x16d/0x2e0
[ 8.796467] schedule+0x3d/0x150
[ 8.796467] worker_thread+0xb5/0x350
[ 8.796467] ? __pfx_worker_thread+0x10/0x10
[ 8.796467] kthread+0xee/0x120
[ 8.796467] ? __pfx_kthread+0x10/0x10
[ 8.796467] ret_from_fork+0x31/0x50
[ 8.796467] ? __pfx_kthread+0x10/0x10
[ 8.796467] ret_from_fork_asm+0x1a/0x30
[ 8.796467] </TASK>
This is because if a RT task was originally picked, it will
setup the rq->balance_callback with push_rt_tasks() via
set_next_task_rt().
Once the task is migrated away and we pick again, we haven't
processed any balance callbacks, so rq->balance_callback is not
in the same state as it was the first time pick_next_task was
called.
To handle this, add a zap_balance_callbacks() helper function
which cleans up the balance callbacks without running them. This
should be ok, as we are effectively undoing the state set in
the first call to pick_next_task(), and when we pick again,
the new callback can be configured for the donor task actually
selected.
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260324191337.1841376-9-jstultz@google.com
With proxy-exec utilizing pick-again logic, we can end up having
balance callbacks set by the preivous pick_next_task() call left
on the list.
So pull the warning out into a helper function, and make sure we
check it when we pick again.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260324191337.1841376-8-jstultz@google.com
As we add functionality to proxy execution, we may migrate a
donor task to a runqueue where it can't run due to cpu affinity.
Thus, we must be careful to ensure we return-migrate the task
back to a cpu in its cpumask when it becomes unblocked.
Peter helpfully provided the following example with pictures:
"Suppose we have a ww_mutex cycle:
,-+-* Mutex-1 <-.
Task-A ---' | | ,-- Task-B
`-> Mutex-2 *-+-'
Where Task-A holds Mutex-1 and tries to acquire Mutex-2, and
where Task-B holds Mutex-2 and tries to acquire Mutex-1.
Then the blocked_on->owner chain will go in circles.
Task-A -> Mutex-2
^ |
| v
Mutex-1 <- Task-B
We need two things:
- find_proxy_task() to stop iterating the circle;
- the woken task to 'unblock' and run, such that it can
back-off and re-try the transaction.
Now, the current code [without this patch] does:
__clear_task_blocked_on();
wake_q_add();
And surely clearing ->blocked_on is sufficient to break the
cycle.
Suppose it is Task-B that is made to back-off, then we have:
Task-A -> Mutex-2 -> Task-B (no further blocked_on)
and it would attempt to run Task-B. Or worse, it could directly
pick Task-B and run it, without ever getting into
find_proxy_task().
Now, here is a problem because Task-B might not be runnable on
the CPU it is currently on; and because !task_is_blocked() we
don't get into the proxy paths, so nobody is going to fix this
up.
Ideally we would have dequeued Task-B alongside of clearing
->blocked_on, but alas, [the lock ordering prevents us from
getting the task_rq_lock() and] spoils things."
Thus we need more than just a binary concept of the task being
blocked on a mutex or not.
So allow setting blocked_on to PROXY_WAKING as a special value
which specifies the task is no longer blocked, but needs to
be evaluated for return migration *before* it can be run.
This will then be used in a later patch to handle proxy
return-migration.
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260324191337.1841376-7-jstultz@google.com
Introduce an action enum in find_proxy_task() which allows
us to handle work needed to be done outside the mutex.wait_lock
and task.blocked_lock guard scopes.
This ensures proper locking when we clear the donor's blocked_on
pointer in proxy_deactivate(), and the switch statement will be
useful as we add more cases to handle later in this series.
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260324191337.1841376-6-jstultz@google.com
So far, we have been able to utilize the mutex::wait_lock
for serializing the blocked_on state, but when we move to
proxying across runqueues, we will need to add more state
and a way to serialize changes to this state in contexts
where we don't hold the mutex::wait_lock.
So introduce the task::blocked_lock, which nests under the
mutex::wait_lock in the locking order, and rework the locking
to use it.
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260324191337.1841376-5-jstultz@google.com
K Prateek pointed out that with Proxy Exec, we may have cases
where we context switch in __schedule(), while the donor remains
the same. This could cause balancing issues, since the
put_prev_set_next() logic short-cuts if (prev == next). With
proxy-exec prev is the previous donor, and next is the next
donor. Should the donor remain the same, but different tasks are
picked to actually run, the shortcut will have avoided enqueuing
the sched class balance callback.
So, if we are context switching, add logic to catch the
same-donor case, and trigger the put_prev/set_next calls to
ensure the balance callbacks get enqueued.
Closes: https://lore.kernel.org/lkml/20ea3670-c30a-433b-a07f-c4ff98ae2379@amd.com/
Reported-by: K Prateek Nayak <kprateek.nayak@amd.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260324191337.1841376-4-jstultz@google.com
Peter noted: Compilers are really bad (as in they utterly refuse)
optimizing (even when marked with __pure) the static branch
things, and will happily emit multiple identical in a row.
So pull out the one obvious sched_proxy_exec() branch in
__schedule() and remove some of the 'implicit' ones in that
path.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260324191337.1841376-3-jstultz@google.com
With proxy-execution, the scheduler selects the donor, but for
blocked donors, we end up running the lock owner.
This caused some complexity, because the class schedulers make
sure to remove the task they pick from their pushable task
lists, which prevents the donor from being migrated, but there
wasn't then anything to prevent rq->curr from being migrated
if rq->curr != rq->donor.
This was sort of hacked around by calling proxy_tag_curr() on
the rq->curr task if we were running something other then the
donor. proxy_tag_curr() did a dequeue/enqueue pair on the
rq->curr task, allowing the class schedulers to remove it from
their pushable list.
The dequeue/enqueue pair was wasteful, and additonally K Prateek
highlighted that we didn't properly undo things when we stopped
proxying, leaving the lock owner off the pushable list.
After some alternative approaches were considered, Peter
suggested just having the RT/DL classes just avoid migrating
when task_on_cpu().
So rework pick_next_pushable_dl_task() and the rt
pick_next_pushable_task() functions so that they skip over the
first pushable task if it is on_cpu.
Then just drop all of the proxy_tag_curr() logic.
Fixes: be39617e38 ("sched: Fix proxy/current (push,pull)ability")
Closes: https://lore.kernel.org/lkml/e735cae0-2cc9-4bae-b761-fcb082ed3e94@amd.com/
Reported-by: K Prateek Nayak <kprateek.nayak@amd.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260324191337.1841376-2-jstultz@google.com
Since commit 8e4f0b1ebc ("bpf: use rcu_read_lock_dont_migrate() for
trampoline.c"), the BPF prolog (__bpf_prog_enter) calls migrate_disable()
only when CONFIG_PREEMPT_RCU is enabled, via rcu_read_lock_dont_migrate().
Without CONFIG_PREEMPT_RCU, the prolog never touches migration_disabled,
so migration_disabled == 1 always means the task is truly
migration-disabled regardless of whether it is the current task.
The old unconditional p == current check was a false negative in this
case, potentially allowing a migration-disabled task to be dispatched to
a remote CPU and triggering scx_error in task_can_run_on_remote_rq().
Only apply the p == current disambiguation when CONFIG_PREEMPT_RCU is
enabled, where the ambiguity with the BPF prolog still exists.
Fixes: 8e4f0b1ebc ("bpf: use rcu_read_lock_dont_migrate() for trampoline.c")
Cc: stable@vger.kernel.org # v6.18+
Link: https://lore.kernel.org/lkml/20250821090609.42508-8-dongml2@chinatelecom.cn/
Signed-off-by: Changwoo Min <changwoo@igalia.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
In scx_set_task_state(), the default case was setting the
warn flag, but then returning immediately. This is problematic
because the only purpose of the warn flag is to trigger
WARN_ONCE, but the early return prevented it from ever firing,
leaving invalid task states undetected and untraced.
To fix this, a WARN_ONCE call is now added directly in the
default case.
The fix addresses two aspects:
- Guarantees the invalid task states are properly logged
and traced.
- Provides a distinct warning message
("sched_ext: Invalid task state") specifically for
states outside the defined scx_task_state enum values,
making it easier to distinguish from other transition
warnings.
This ensures proper detection and reporting of invalid states.
Signed-off-by: Samuele Mariotti <smariotti@disroot.org>
Signed-off-by: Paolo Valente <paolo.valente@unimore.it>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
cpufreq_cpu_get() can sleep on PREEMPT_RT in presence of concurrent
writer(s), however amd-pstate depends on fetching the cpudata via the
policy's driver data which necessitates grabbing the reference.
Since schedutil governor can call "cpufreq_driver->update_perf()"
during sched_tick/enqueue/dequeue with rq_lock held and IRQs disabled,
fetching the policy object using the cpufreq_cpu_get() helper in the
scheduler fast-path leads to "BUG: scheduling while atomic" on
PREEMPT_RT [1].
Pass the cached cpufreq policy object in sg_policy to the update_perf()
instead of just the CPU. The CPU can be inferred using "policy->cpu".
The lifetime of cpufreq_policy object outlasts that of the governor and
the cpufreq driver (allocated when the CPU is onlined and only reclaimed
when the CPU is offlined / the CPU device is removed) which makes it
safe to be referenced throughout the governor's lifetime.
Closes:https://lore.kernel.org/all/20250731092316.3191-1-spasswolf@web.de/ [1]
Fixes: 1d215f0319 ("cpufreq: amd-pstate: Add fast switch function for AMD P-State")
Reported-by: Bert Karwatzki <spasswolf@web.de>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Acked-by: Gary Guo <gary@garyguo.net> # Rust
Reviewed-by: Gautham R. Shenoy <gautham.shenoy@amd.com>
Reviewed-by: Zhongqiu Han <zhongqiu.han@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20260316081849.19368-3-kprateek.nayak@amd.com
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
The following fix in sched/urgent:
e08d007f9d ("sched/debug: Fix avg_vruntime() usage")
is in conflict with this pending commit in sched/core:
4823725d9d ("sched/fair: Increase weight bits for avg_vruntime")
Both modify the same variable definition and initialization blocks,
resolve it by merging the two.
Conflicts:
kernel/sched/debug.c
Signed-off-by: Ingo Molnar <mingo@kernel.org>
John reported that stress-ng-yield could make his machine unhappy and
managed to bisect it to commit b3d99f43c7 ("sched/fair: Fix
zero_vruntime tracking").
The commit in question changes avg_vruntime() from a function that is
a pure reader, to a function that updates variables. This turns an
unlocked sched/debug usage of this function from a minor mistake into
a data corruptor.
Fixes: af4cf40470 ("sched/fair: Add cfs_rq::avg_vruntime")
Fixes: b3d99f43c7 ("sched/fair: Fix zero_vruntime tracking")
Reported-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: John Stultz <jstultz@google.com>
Link: https://patch.msgid.link/20260401132355.196370805@infradead.org
John reported that stress-ng-yield could make his machine unhappy and
managed to bisect it to commit b3d99f43c7 ("sched/fair: Fix
zero_vruntime tracking").
The combination of yield and that commit was specific enough to
hypothesize the following scenario:
Suppose we have 2 runnable tasks, both doing yield. Then one will be
eligible and one will not be, because the average position must be in
between these two entities.
Therefore, the runnable task will be eligible, and be promoted a full
slice (all the tasks do is yield after all). This causes it to jump over
the other task and now the other task is eligible and current is no
longer. So we schedule.
Since we are runnable, there is no {de,en}queue. All we have is the
__{en,de}queue_entity() from {put_prev,set_next}_task(). But per the
fingered commit, those two no longer move zero_vruntime.
All that moves zero_vruntime are tick and full {de,en}queue.
This means, that if the two tasks playing leapfrog can reach the
critical speed to reach the overflow point inside one tick's worth of
time, we're up a creek.
Additionally, when multiple cgroups are involved, there is no guarantee
the tick will in fact hit every cgroup in a timely manner. Statistically
speaking it will, but that same statistics does not rule out the
possibility of one cgroup not getting a tick for a significant amount of
time -- however unlikely.
Therefore, just like with the yield() case, force an update at the end
of every slice. This ensures the update is never more than a single
slice behind and the whole thing is within 2 lag bounds as per the
comment on entity_key().
Fixes: b3d99f43c7 ("sched/fair: Fix zero_vruntime tracking")
Reported-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Tested-by: John Stultz <jstultz@google.com>
Link: https://patch.msgid.link/20260401132355.081530332@infradead.org
- Fix SCX_KICK_WAIT deadlock where multiple CPUs waiting for each other in
hardirq context form a cycle. Move the wait to a balance callback which
can drop the rq lock and process IPIs.
- Fix inconsistent NUMA node lookup in scx_select_cpu_dfl() where the
waker_node used cpu_to_node() while prev_cpu used
scx_cpu_node_if_enabled(), leading to undefined behavior when per-node
idle tracking is disabled.
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCacwiiQ4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGVILAP44s30JBpNyJ9JhAiCoTYzxzOXqqGbotnpQckMF
+7WoJAD/Z9dJO/Sw/AH0fX6WVJDmO0QsQvFXLXJBxWy7A5XVAA0=
=2DW5
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-7.0-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- Fix SCX_KICK_WAIT deadlock where multiple CPUs waiting for each other
in hardirq context form a cycle. Move the wait to a balance callback
which can drop the rq lock and process IPIs.
- Fix inconsistent NUMA node lookup in scx_select_cpu_dfl() where
the waker_node used cpu_to_node() while prev_cpu used
scx_cpu_node_if_enabled(), leading to undefined behavior when
per-node idle tracking is disabled.
* tag 'sched_ext-for-7.0-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
selftests/sched_ext: Add cyclic SCX_KICK_WAIT stress test
sched_ext: Fix SCX_KICK_WAIT deadlock by deferring wait to balance callback
sched_ext: Fix inconsistent NUMA node lookup in scx_select_cpu_dfl()
Some utility functions on sched_dl_entity can be useful outside of
deadline.c , for instance for modelling, without relying on raw
structure fields.
Move functions like dl_task_of and dl_is_implicit to deadline.h to make
them available outside.
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://lore.kernel.org/r/20260330111010.153663-12-gmonaco@redhat.com
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Add the following tracepoints:
* sched_dl_throttle(dl_se, cpu, type):
Called when a deadline entity is throttled
* sched_dl_replenish(dl_se, cpu, type):
Called when a deadline entity's runtime is replenished
* sched_dl_update(dl_se, cpu, type):
Called when a deadline entity updates without throttle or replenish
* sched_dl_server_start(dl_se, cpu, type):
Called when a deadline server is started
* sched_dl_server_stop(dl_se, cpu, type):
Called when a deadline server is stopped
Those tracepoints can be useful to validate the deadline scheduler with
RV and are not exported to tracefs.
Reviewed-by: Phil Auld <pauld@redhat.com>
Acked-by: Juri Lelli <juri.lelli@redhat.com>
Link: https://lore.kernel.org/r/20260330111010.153663-11-gmonaco@redhat.com
Signed-off-by: Gabriele Monaco <gmonaco@redhat.com>
Conflict in kernel/sched/ext.c init_sched_ext_class() between:
415cb193bb ("sched_ext: Fix SCX_KICK_WAIT deadlock by deferring wait
to balance callback")
which adds cpus_to_sync cpumask allocation, and:
84b1a0ea0b ("sched_ext: Implement scx_bpf_dsq_reenq() for user DSQs")
8c1b9453fd ("sched_ext: Convert deferred_reenq_locals from llist to
regular list")
which add deferred_reenq init code at the same location. Both are
independent additions. Include both.
Signed-off-by: Tejun Heo <tj@kernel.org>
SCX_KICK_WAIT busy-waits in kick_cpus_irq_workfn() using
smp_cond_load_acquire() until the target CPU's kick_sync advances. Because
the irq_work runs in hardirq context, the waiting CPU cannot reschedule and
its own kick_sync never advances. If multiple CPUs form a wait cycle, all
CPUs deadlock.
Replace the busy-wait in kick_cpus_irq_workfn() with resched_curr() to
force the CPU through do_pick_task_scx(), which queues a balance callback
to perform the wait. The balance callback drops the rq lock and enables
IRQs following the sched_core_balance() pattern, so the CPU can process
IPIs while waiting. The local CPU's kick_sync is advanced on entry to
do_pick_task_scx() and continuously during the wait, ensuring any CPU that
starts waiting for us sees the advancement and cannot form cyclic
dependencies.
Fixes: 90e55164da ("sched_ext: Implement SCX_KICK_WAIT")
Cc: stable@vger.kernel.org # v6.12+
Reported-by: Christian Loehle <christian.loehle@arm.com>
Link: https://lore.kernel.org/r/20260316100249.1651641-1-christian.loehle@arm.com
Signed-off-by: Tejun Heo <tj@kernel.org>
Tested-by: Christian Loehle <christian.loehle@arm.com>
Add a comment explaining the design intent behind rejecting built-in DSQs
(%SCX_DSQ_GLOBAL and %SCX_DSQ_LOCAL*) as sources. Local DSQs support
reenqueueing but the BPF scheduler cannot directly iterate or move tasks
from them. %SCX_DSQ_GLOBAL is similar but also doesn't support
reenqueueing because it maps to multiple per-node DSQs, making the scope
difficult to define.
Also annotate @dsq_id to make clear it must be a user-created DSQ.
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
025b1bd419 introduced SCX_EV_SUB_BYPASS_DISPATCH to track scheduling
of bypassed descendant tasks, and correctly increments it per-CPU and
displays it in sysfs and dump output. However, scx_read_events() which
aggregates per-CPU counters into a summary was not updated to include
this event, causing it to always read as zero in sysfs, in debug dumps,
and via the scx_bpf_events() kfunc.
Add the missing scx_agg_event() call for SCX_EV_SUB_BYPASS_DISPATCH.
Fixes: 025b1bd419 ("sched_ext: Implement hierarchical bypass mode")
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
When scx_bpf_dsq_move[_vtime]() is called on a task that belongs to a
different scheduler, scx_error() is invoked to flag the violation.
scx_error() schedules an asynchronous scheduler teardown via irq_work
and returns immediately, so execution falls through and the DSQ move
proceeds on a cross-scheduler task regardless, potentially corrupting
DSQ state.
Add the missing return false so the function exits right after
reporting the error, consistent with the other early-exit checks in
the same function (e.g. scx_vet_enq_flags() failure at the top).
Fixes: bb4d9fd551 ("sched_ext: scx_dsq_move() should validate the task belongs to the right scheduler")
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
The print_scx_info() always output scx_root structure's->ops.name,
but for built with CONFIG_EXT_SUB_SCHED=y kernels, the tasks may be
attach an sub scx_sched structure. this commit therefore use the
scx_task_sched_rcu() to correctly get scx_sched structure to output
ops.name, and drop state check.
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Zqiang <qiang.zhang@linux.dev>
Signed-off-by: Tejun Heo <tj@kernel.org>
Calling smp_processor_id() on:
- In CONFIG_DEBUG_PREEMPT=y, if preemption/irq is disabled, then it does
not print any warning.
- In CONFIG_DEBUG_PREEMPT=n, it doesn't do anything apart from getting
__smp_processor_id
So with both CONFIG_DEBUG_PREEMPT=y/n, in preemption disabled section
it is better to cache the value. It could save a few cycles. Though
tiny, repeated could add up to a small value.
ttwu_queue_cond is called with interrupt disabled. So preemption is
disabled. Hence cache the value once instead.
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Mukesh Kumar Chaurasiya (IBM) <mkchauras@gmail.com>
Link: https://patch.msgid.link/20260323193630.640311-3-sshegde@linux.ibm.com
Calling smp_processor_id() on:
- In CONFIG_DEBUG_PREEMPT=y, if preemption/irq is disabled, then it does
not print any warning.
- In CONFIG_DEBUG_PREEMPT=n, it doesn't do anything apart from getting
__smp_processor_id
So with both CONFIG_DEBUG_PREEMPT=y/n, in preemption disabled section
it is better to cache the value. It could save a few cycles. Though
tiny, repeated in loop could add up to a small value.
find_new_ilb is called in interrupt context. So preemption is disabled.
So Hoist the this_cpu out of loop
Signed-off-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Mukesh Kumar Chaurasiya (IBM) <mkchauras@gmail.com>
Reviewed-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://patch.msgid.link/20260323193630.640311-2-sshegde@linux.ibm.com
When scx_alloc_and_add_sched() creates the sub-scheduler kset, it sets
sch->kobj as the parent. Because sch->kobj.kset points to scx_kset,
registering this sub-kset triggers a KOBJ_ADD uevent. The uevent walk
finds scx_kset and calls scx_uevent() with the sub-kset's kobject.
scx_uevent() unconditionally uses container_of() to cast the incoming
kobject to struct scx_sched, producing a wild pointer when the kobject
belongs to the kset itself rather than a scheduler instance. Accessing
sch->ops.name through this pointer causes a KASAN slab-out-of-bounds
read:
BUG: KASAN: slab-out-of-bounds in string+0x3b6/0x4c0
Read of size 1 at addr ffff888004d04348 by task scx_enable_help/748
Call Trace:
string+0x3b6/0x4c0
vsnprintf+0x3ec/0x1550
add_uevent_var+0x160/0x3a0
scx_uevent+0x22/0x30
kobject_uevent_env+0x5dc/0x1730
kset_register+0x192/0x280
scx_alloc_and_add_sched+0x130d/0x1c60
...
Fix this by checking the kobject's ktype against scx_ktype before
performing the cast, and returning 0 for non-matching kobjects.
Tested with vng and scx_qmap without triggering any KASAN errors.
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
schedule_deferred() uses irq_work_queue() which always queues on the
calling CPU. The deferred work can run from any CPU correctly, and the
_locked() path already processes remote rqs from the calling CPU. However,
when falling through to the irq_work path, queuing on the target CPU is
preferable as the work can run sooner via IPI delivery rather than waiting
for the calling CPU to re-enable IRQs.
Currently, only reenqueue operations use this path - either BPF-initiated
reenqueue targeting a remote rq, or IMMED reenqueue when the target CPU is
busy running userspace (not in balance or wakeup, so the _locked() fast
paths aren't available). Use irq_work_queue_on() to target the owning CPU.
This improves IMMED reenqueue latency when tasks are dispatched to
remote local DSQs. Testing on a 24-CPU AMD Ryzen 3900X with scx_qmap
-I -F 50 (ALWAYS_ENQ_IMMED, every 50th enqueue forced to prev_cpu's
local DSQ) under heavy mixed load (2x CPU oversubscription, yield and
context-switch pressure, SCHED_FIFO bursts, periodic fork storms, mixed
nice levels, C-states disabled), measuring local DSQ residence time
(insert to remove) over 5 x 120s runs (~1.2M tasks per set):
>128us outliers: 71 -> 39 (-45%)
>256us outliers: 59 -> 36 (-39%)
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
When building with SCHED_CLASS_EXT=y but CGROUPS=n, clang reports errors
for undeclared cgroup_put() and cgroup_get() calls, and a warning for the
unused err_stop_helper label.
EXT_SUB_SCHED is def_bool y depending only on SCHED_CLASS_EXT, but it
fundamentally requires cgroups (cgroup_path, cgroup_get, cgroup_put,
cgroup_id, etc.). Add the missing CGROUPS dependency to EXT_SUB_SCHED in
init/Kconfig.
Guard cgroup_put() and cgroup_get() in the common paths with:
#if defined(CONFIG_EXT_GROUP_SCHED) || defined(CONFIG_EXT_SUB_SCHED)
Guard the err_stop_helper label with #ifdef CONFIG_EXT_SUB_SCHED since
all gotos targeting it are inside that same ifdef block.
Tested with both CGROUPS enabled and disabled.
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202603210903.IrKhPd6k-lkp@intel.com/
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Acked-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
In the WAKE_SYNC path of scx_select_cpu_dfl(), waker_node was computed
with cpu_to_node(), while node (for prev_cpu) was computed with
scx_cpu_node_if_enabled(). When scx_builtin_idle_per_node is disabled,
idle_cpumask(waker_node) is called with a real node ID even though
per-node idle tracking is disabled, resulting in undefined behavior.
Fix by using scx_cpu_node_if_enabled() for waker_node as well, ensuring
both variables are computed consistently.
Fixes: 48849271e6 ("sched_ext: idle: Per-node idle cpumasks")
Cc: stable@vger.kernel.org # v6.15+
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
In the default built-in idle CPU selection policy, when @prev_cpu is
busy and no fully idle core is available, try to place the task on its
SMT sibling if that sibling is idle, before searching any other idle CPU
in the same LLC.
Migration to the sibling is cheap and keeps the task on the same core,
preserving L1 cache and reducing wakeup latency.
On large SMT systems this appears to consistently boost throughput by
roughly 2-3% on CPU-bound workloads (running a number of tasks equal to
the number of SMT cores).
Cc: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
-----BEGIN PGP SIGNATURE-----
iQFSBAABCgA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAmm3G/UeHHRvcnZhbGRz
QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiGZJUH/R0vQ3Vha48QDEic
1NREwaHxAoTFi0i3y7OPPklqrP2V09D1qg4Q6fExYQVTQgV6F2DRjVbyPKrmr4ay
BA6aHrUdnFngYHpDlI1b1r7rJiAIN4WFHl7StO70bS+EB+UPsP9cfP3CKXUfKfqT
kyHXzUrd5QnjYmlb9rQw1E6rzsRamNtGUtZf7TwDidJYjtm3sPeDHUkjyRy4xkYd
UouIu6W7UXoicl38bJAgaWBY5BiYtjN6ktnY4/gcqDeqYd7mTM3Eb1B+OSXgFfip
F0OYfJhfWn+63WnPA+1I5jXWC1UrdVXTMK/NTYjhmGlfdmkLcWDlNGtu+qKZbpwj
fmF3Kyo=
=6nX1
-----END PGP SIGNATURE-----
Merge tag 'v7.0-rc4' into timers/core, to resolve conflict
Resolve conflict between this change in the upstream kernel:
4c652a4772 ("rseq: Mark rseq_arm_slice_extension_timer() __always_inline")
... and this pending change in timers/core:
0e98eb1481 ("entry: Prepare for deferred hrtimer rearming")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Use the "sd_llc" passed to select_idle_cpu() to obtain the
"sd_llc_shared" instead of dereferencing the per-CPU variable.
Since "sd->shared" is always reclaimed at the same time as "sd" via
call_rcu() and update_top_cache_domain() always ensures a valid
"sd->shared" assignment when "sd_llc" is present, "sd_llc->shared" can
always be dereferenced without needing an additional check.
While at it move the cpumask_and() operation after the SIS_UTIL bailout
check to avoid unnecessarily computing the cpumask.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-10-kprateek.nayak@amd.com
Only the topmost SD_SHARE_LLC domain has the "sd->shared" assigned.
Simply use "sd->shared" as an indicator for load balancing at the highest
SD_SHARE_LLC domain in update_idle_cpu_scan() instead of relying on
llc_size.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-9-kprateek.nayak@amd.com
select_task_rq_fair() is always called with p->pi_lock held and IRQs
disabled which makes it equivalent of an RCU read-side.
Since commit 71fedc41c2 ("sched/fair: Switch to
rcu_dereference_all()") switched to using rcu_dereference_all() in the
wakeup path, drop the explicit rcu_read_{lock,unlock}() in the fair
task's wakeup path.
Future plans to reuse select_task_rq_fair() /
find_energy_efficient_cpu() in the fair class' balance callback will do
so with IRQs disabled and will comply with the requirements of
rcu_dereference_all() which makes this safe keeping in mind future
development plans too.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-8-kprateek.nayak@amd.com
Similar to commit 71fedc41c2 ("sched/fair: Switch to
rcu_dereference_all()"), switch to checking for rcu_read_lock_any_held()
in idle_get_state() to allow removing superfluous rcu_read_lock()
regions in the fair task's wakeup path where the pi_lock is held and
IRQs are disabled.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-7-kprateek.nayak@amd.com
Now that "sd->shared" assignments are using the sched_domain_shared
objects allocated with s_data, remove the sd_data based allocations.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-6-kprateek.nayak@amd.com
Use the "sched_domain_shared" object allocated in s_data for
"sd->shared" assignments. Assign "sd->shared" for the topmost
SD_SHARE_LLC domain before degeneration and rely on the degeneration
path to correctly pass down the shared object to "sd_llc".
sd_degenerate_parent() ensures degenerating domains must have the same
sched_domain_span() which ensures 1:1 passing down of the shared object.
If the topmost SD_SHARE_LLC domain degenerates, the shared object is
freed from destroy_sched_domain() when the last reference is dropped.
claim_allocations() NULLs out the objects that have been assigned as
"sd->shared" and the unassigned ones are freed from the __sds_free()
path.
To keep all the claim_allocations() bits in one place,
claim_allocations() has been extended to accept "s_data" and iterate the
domains internally to free both "sched_domain_shared" and the
per-topology-level data for the particular CPU in one place.
Post cpu_attach_domain(), all reclaims of "sd->shared" are handled via
call_rcu() on the sched_domain object via destroy_sched_domains_rcu().
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-5-kprateek.nayak@amd.com
The "sched_domain_shared" object is allocated for every topology level
in __sdt_alloc() and is freed post sched domain rebuild if they aren't
assigned during sd_init().
"sd->shared" is only assigned for SD_SHARE_LLC domains and out of all
the assigned objects, only "sd_llc_shared" is ever used by the
scheduler.
Since only "sd_llc_shared" is ever used, and since SD_SHARE_LLC domains
never overlap, allocate only a single range of per-CPU
"sched_domain_shared" object with s_data instead of doing it per
topology level.
The subsequent commit uses the degeneration path to correctly assign the
"sd->shared" to the topmost SD_SHARE_LLC domain.
No functional changes are expected at this point.
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-4-kprateek.nayak@amd.com
Subsequent changes to assign "sd->shared" from "s_data" would
necessitate finding the topmost SD_SHARE_LLC to assign shared object to.
This is very similar to the "imb_numa_nr" computation loop except that
"imb_numa_nr" cares about the first domain without the SD_SHARE_LLC flag
(immediate parent of sd_llc) whereas the "sd->shared" assignment would
require sd_llc itself.
Extract the "imb_numa_nr" calculation into a helper
adjust_numa_imbalance() and use the current loop in the
build_sched_domains() to find the sd_llc.
While at it, guard the call behind CONFIG_NUMA's status since
"imb_numa_nr" only makes sense on NUMA enabled configs with SD_NUMA
domains.
No functional changes intended.
Suggested-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-3-kprateek.nayak@amd.com
The "sd_weight" used for calculating the load balancing interval, and
its limits, considers the span weight of the entire topology level
without accounting for cpuset partitions.
For example, consider a large system of 128CPUs divided into 8 * 16CPUs
partition which is typical when deploying virtual machines:
[ PKG Domain: 128CPUs ]
[Partition0: 16CPUs][Partition1: 16CPUs] ... [Partition7: 16CPUs]
Although each partition only contains 16CPUs, the load balancing
interval is set to a minimum of 128 jiffies considering the span of the
entire domain with 128CPUs which can lead to longer imbalances within
the partition although balancing within is cheaper with 16CPUs.
Compute the "sd_weight" after computing the "sd_span" considering the
cpu_map covered by the partition, and set the load balancing interval,
and its limits accordingly.
For the above example, the balancing intervals for the partitions PKG
domain changes as follows:
before after
balance_interval 128 16
min_interval 128 16
max_interval 256 32
Intervals are now proportional to the CPUs in the partitioned domain as
was intended by the original formula.
Fixes: cb83b629ba ("sched/numa: Rewrite the CONFIG_NUMA sched domain support")
Signed-off-by: K Prateek Nayak <kprateek.nayak@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Shrikanth Hegde <sshegde@linux.ibm.com>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Tested-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Link: https://patch.msgid.link/20260312044434.1974-2-kprateek.nayak@amd.com
There are two special cases in the idle loop that are handled
inconsistently even though they are analogous.
The first one is when a cpuidle driver is absent and the default CPU
idle time power management implemented by the architecture code is used.
In that case, the scheduler tick is stopped every time before invoking
default_idle_call().
The second one is when a cpuidle driver is present, but there is only
one idle state in its table. In that case, the scheduler tick is never
stopped at all.
Since each of these approaches has its drawbacks, reconcile them with
the help of one simple heuristic. Namely, stop the tick if the CPU has
been woken up by it in the previous iteration of the idle loop, or let
it tick otherwise.
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Reviewed-by: Qais Yousef <qyousef@layalina.io>
Reviewed-by: Aboorva Devarajan <aboorvad@linux.ibm.com>
Fixes: ed98c34919 ("sched: idle: Do not stop the tick before cpuidle_idle_call()")
[ rjw: Added Fixes tag, changelog edits ]
Link: https://patch.msgid.link/4741364.LvFx2qVVIh@rafael.j.wysocki
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
ancestors[] is a flexible array member that needs level + 1 slots to
hold all ancestors including self (indices 0..level), but kzalloc_flex()
only allocates `level` slots:
sch = kzalloc_flex(*sch, ancestors, level);
...
sch->ancestors[level] = sch; /* one past the end */
For the root scheduler (level = 0), zero slots are allocated and
ancestors[0] is written immediately past the end of the object.
KASAN reports:
BUG: KASAN: slab-out-of-bounds in scx_alloc_and_add_sched+0x1c17/0x1d10
Write of size 8 at addr ffff888066b56538 by task scx_enable_help/667
The buggy address is located 0 bytes to the right of
allocated 1336-byte region [ffff888066b56000, ffff888066b56538)
Fix by passing level + 1 to kzalloc_flex().
Tested with vng + scx_lavd, KASAN no longer triggers.
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
kobject_init_and_add() failure requires kobject_put() for proper cleanup, but
the error paths were using kfree(sch) possibly leaking the kobject name. The
kset_create_and_add() failure was already using kobject_put() correctly.
Switch the kobject_init_and_add() error paths to use kobject_put(). As the
release path puts the cgroup ref, make scx_alloc_and_add_sched() always
consume @cgrp via a new err_put_cgrp label at the bottom of the error chain
and update scx_sub_enable_workfn() accordingly.
Fixes: 17108735b4 ("sched_ext: Use dynamic allocation for scx_sched")
Reported-by: David Carlier <devnexen@gmail.com>
Link: https://lore.kernel.org/r/20260314134457.46216-1-devnexen@gmail.com
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The abort path in scx_sub_enable_workfn() fell through to out_put_cgrp,
double-putting the cgroup ref already owned by sch->cgrp. It also skipped
kthread_flush_work() needed to flush the disable path.
Relocate the abort block above err_unlock_and_disable so it falls through to
err_disable.
Fixes: 337ec00b1d ("sched_ext: Implement cgroup sub-sched enabling and disabling")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
- Fix CID hangs due to a race between concurrent forks
- Fix vfork()/CLONE_VM MMCID bug causing hangs
- Remove pointless preemption guard
- Fix CID task list walk performance regression on large systems
by removing the known-flaky and slow counting logic using
for_each_process_thread() in mm_cid_*fixup_tasks_to_cpus(),
and implementing a simple sched_mm_cid::node list instead
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=ce5G
-----END PGP SIGNATURE-----
Merge tag 'sched-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fixes from Ingo Molnar:
"More MM-CID fixes, mostly fixing hangs/races:
- Fix CID hangs due to a race between concurrent forks
- Fix vfork()/CLONE_VM MMCID bug causing hangs
- Remove pointless preemption guard
- Fix CID task list walk performance regression on large systems
by removing the known-flaky and slow counting logic using
for_each_process_thread() in mm_cid_*fixup_tasks_to_cpus(), and
implementing a simple sched_mm_cid::node list instead"
* tag 'sched-urgent-2026-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/mmcid: Avoid full tasklist walks
sched/mmcid: Remove pointless preempt guard
sched/mmcid: Handle vfork()/CLONE_VM correctly
sched/mmcid: Prevent CID stalls due to concurrent forks
Under CONFIG_EXT_SUB_SCHED, the kzalloc() and kstrdup() failure
paths jump to err_stop_helper without first setting ret. The
function then returns ERR_PTR(ret) with ret uninitialized, which
can produce ERR_PTR(0) (NULL), causing the caller's IS_ERR() check
to pass and leading to a NULL pointer dereference.
Set ret = -ENOMEM before each goto to fix the error path.
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
- Fix data races flagged by KCSAN: add missing READ_ONCE()/WRITE_ONCE()
annotations for lock-free accesses to module parameters and dsq->seq.
- Fix silent truncation of upper 32 enqueue flags (SCX_ENQ_PREEMPT and
above) when passed through the int sched_class interface.
- Documentation updates: scheduling class precedence, task ownership
state machine, example scheduler descriptions, config list cleanup.
- Selftest fix for format specifier and buffer length in
file_write_long().
-----BEGIN PGP SIGNATURE-----
iIQEABYKACwWIQTfIjM1kS57o3GsC/uxYfJx3gVYGQUCabRyHg4cdGpAa2VybmVs
Lm9yZwAKCRCxYfJx3gVYGZiWAQCmUOHiGAk73p9DDn6Zyrm+o/iQm/iOinchBeUs
ZiG0bgEAn15giAnLCA5Zs6cG7PemxBH1v7ctyzTjh1VsBds0rwo=
=zXix
-----END PGP SIGNATURE-----
Merge tag 'sched_ext-for-7.0-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext
Pull sched_ext fixes from Tejun Heo:
- Fix data races flagged by KCSAN: add missing READ_ONCE()/WRITE_ONCE()
annotations for lock-free accesses to module parameters and dsq->seq
- Fix silent truncation of upper 32 enqueue flags (SCX_ENQ_PREEMPT and
above) when passed through the int sched_class interface
- Documentation updates: scheduling class precedence, task ownership
state machine, example scheduler descriptions, config list cleanup
- Selftest fix for format specifier and buffer length in
file_write_long()
* tag 'sched_ext-for-7.0-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
sched_ext: Use WRITE_ONCE() for the write side of scx_enable helper pointer
sched_ext: Fix enqueue_task_scx() truncation of upper enqueue flags
sched_ext: Documentation: Update sched-ext.rst
sched_ext: Use READ_ONCE() for scx_slice_bypass_us in scx_bypass()
sched_ext: Documentation: Mention scheduling class precedence
sched_ext: Document task ownership state machine
sched_ext: Use READ_ONCE() for lock-free reads of module param variables
sched_ext/selftests: Fix format specifier and buffer length in file_write_long()
sched_ext: Use WRITE_ONCE() for the write side of dsq->seq update
schedule_dsq_reenq() always uses schedule_deferred() which falls back to
irq_work. However, callers like schedule_reenq_local() already hold the
target rq lock, and scx_bpf_dsq_reenq() may hold it via the ops callback.
Add a locked_rq parameter so schedule_dsq_reenq() can use
schedule_deferred_locked() when the target rq is already held. The locked
variant can use cheaper paths (balance callbacks, wakeup hooks) instead of
always bouncing through irq_work.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
SCX_ENQ_IMMED makes enqueue to local DSQs succeed only if the task can start
running immediately. Otherwise, the task is re-enqueued through ops.enqueue().
This provides tighter control but requires specifying the flag on every
insertion.
Add SCX_OPS_ALWAYS_ENQ_IMMED ops flag. When set, SCX_ENQ_IMMED is
automatically applied to all local DSQ enqueues including through
scx_bpf_dsq_move_to_local().
scx_qmap is updated with -I option to test the feature and -F option for
IMMED stress testing which forces every Nth enqueue to a busy local DSQ.
v2: - Cover scx_bpf_dsq_move_to_local() path (now has enq_flags via ___v2).
- scx_qmap: Remove sched_switch and cpu_release handlers (superseded by
kernel-side wakeup_preempt_scx()). Add -F for IMMED stress testing.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_bpf_dsq_move_to_local() moves a task from a non-local DSQ to the
current CPU's local DSQ. This is an indirect way of dispatching to a local
DSQ and should support enq_flags like direct dispatches do - e.g.
SCX_ENQ_HEAD for head-of-queue insertion and SCX_ENQ_IMMED for immediate
execution guarantees.
Add scx_bpf_dsq_move_to_local___v2() with an enq_flags parameter. The
original becomes a v1 compat wrapper passing 0. The compat macro is updated
to a three-level chain: v2 (7.1+) -> v1 (current) -> scx_bpf_consume
(pre-rename). All in-tree BPF schedulers are updated to pass 0.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add enq_flags parameter to consume_dispatch_q() and consume_remote_task(),
passing it through to move_{local,remote}_task_to_local_dsq(). All callers
pass 0.
No functional change. This prepares for SCX_ENQ_IMMED support on the consume
path.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add SCX_ENQ_IMMED enqueue flag for local DSQ insertions. Once a task is
dispatched with IMMED, it either gets on the CPU immediately and stays on it,
or gets reenqueued back to the BPF scheduler. It will never linger on a local
DSQ behind other tasks or on a CPU taken by a higher-priority class.
rq_is_open() uses rq->next_class to determine whether the rq is available,
and wakeup_preempt_scx() triggers reenqueue when a higher-priority class task
arrives. These capture all higher class preemptions. Combined with reenqueue
points in the dispatch path, all cases where an IMMED task would not execute
immediately are covered.
SCX_TASK_IMMED persists in p->scx.flags until the next fresh enqueue, so the
guarantee survives SAVE/RESTORE cycles. If preempted while running,
put_prev_task_scx() reenqueues through ops.enqueue() with
SCX_TASK_REENQ_PREEMPTED instead of silently placing the task back on the
local DSQ.
This enables tighter scheduling latency control by preventing tasks from
piling up on local DSQs. It also enables opportunistic CPU sharing across
sub-schedulers - without this, a sub-scheduler can stuff the local DSQ of a
shared CPU, making it difficult for others to use.
v2: - Rewrite is_curr_done() as rq_is_open() using rq->next_class and
implement wakeup_preempt_scx() to achieve complete coverage of all
cases where IMMED tasks could get stranded.
- Track IMMED persistently in p->scx.flags and reenqueue
preempted-while-running tasks through ops.enqueue().
- Bound deferred reenq cycles (SCX_REENQ_LOCAL_MAX_REPEAT).
- Misc renames, documentation.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add scx_vet_enq_flags() stub and call it from scx_dsq_insert_preamble() and
scx_dsq_move(). Pass dsq_id into preamble so the vetting function can
validate flag and DSQ combinations.
No functional change. This prepares for SCX_ENQ_IMMED which will populate the
vetting function.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Split task_should_reenq() into local_task_should_reenq() and
user_task_should_reenq(). The local variant takes reenq_flags by pointer.
No functional change. This prepares for SCX_ENQ_IMMED which will add
IMMED-specific logic to the local variant.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Replace the comma operator with separate statements when assigning
NUMA fault statistics. This improves readability and follows kernel
coding style.
Signed-off-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260309024247.10908-1-zhanxusheng@xiaomi.com
Chasing vfork()'ed tasks on a CID ownership mode switch requires a full
task list walk, which is obviously expensive on large systems.
Avoid that by keeping a list of tasks using a mm MMCID entity in mm::mm_cid
and walk this list instead. This removes the proven to be flaky counting
logic and avoids a full task list walk in the case of vfork()'ed tasks.
Fixes: fbd0e71dc3 ("sched/mmcid: Provide CID ownership mode fixup functions")
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260310202526.183824481@kernel.org
This is a leftover from the early versions of this function where it could
be invoked without mm::mm_cid::lock held.
Remove it and add lockdep asserts instead.
Fixes: 653fda7ae7 ("sched/mmcid: Switch over to the new mechanism")
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260310202526.116363613@kernel.org
Matthieu and Jiri reported stalls where a task endlessly loops in
mm_get_cid() when scheduling in.
It turned out that the logic which handles vfork()'ed tasks is broken. It
is invoked when the number of tasks associated to a process is smaller than
the number of MMCID users. It then walks the task list to find the
vfork()'ed task, but accounts all the already processed tasks as well.
If that double processing brings the number of to be handled tasks to 0,
the walk stops and the vfork()'ed task's CID is not fixed up. As a
consequence a subsequent schedule in fails to acquire a (transitional) CID
and the machine stalls.
Cure this by removing the accounting condition and make the fixup always
walk the full task list if it could not find the exact number of users in
the process' thread list.
Fixes: fbd0e71dc3 ("sched/mmcid: Provide CID ownership mode fixup functions")
Closes: https://lore.kernel.org/b24ffcb3-09d5-4e48-9070-0b69bc654281@kernel.org
Reported-by: Matthieu Baerts <matttbe@kernel.org>
Reported-by: Jiri Slaby <jirislaby@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260310202526.048657665@kernel.org
A newly forked task is accounted as MMCID user before the task is visible
in the process' thread list and the global task list. This creates the
following problem:
CPU1 CPU2
fork()
sched_mm_cid_fork(tnew1)
tnew1->mm.mm_cid_users++;
tnew1->mm_cid.cid = getcid()
-> preemption
fork()
sched_mm_cid_fork(tnew2)
tnew2->mm.mm_cid_users++;
// Reaches the per CPU threshold
mm_cid_fixup_tasks_to_cpus()
for_each_other(current, p)
....
As tnew1 is not visible yet, this fails to fix up the already allocated CID
of tnew1. As a consequence a subsequent schedule in might fail to acquire a
(transitional) CID and the machine stalls.
Move the invocation of sched_mm_cid_fork() after the new task becomes
visible in the thread and the task list to prevent this.
This also makes it symmetrical vs. exit() where the task is removed as CID
user before the task is removed from the thread and task lists.
Fixes: fbd0e71dc3 ("sched/mmcid: Provide CID ownership mode fixup functions")
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260310202525.969061974@kernel.org
scx_claim_exit() propagates exits to descendants under scx_sched_lock.
A sub-sched being attached concurrently could be missed if it links
after the propagation. Check the parent's exit_kind in scx_link_sched()
under scx_sched_lock to interlock against scx_claim_exit() - either the
parent sees the child in its iteration or the child sees the parent's
non-NONE exit_kind and fails attachment.
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
There are two sites that nest rq lock inside scx_sched_lock:
- scx_bypass() takes scx_sched_lock then rq lock per CPU to propagate
per-cpu bypass flags and re-enqueue tasks.
- sysrq_handle_sched_ext_dump() takes scx_sched_lock to iterate all
scheds, scx_dump_state() then takes rq lock per CPU for dump.
And scx_claim_exit() takes scx_sched_lock to propagate exits to
descendants. It can be reached from scx_tick(), BPF kfuncs, and many
other paths with rq lock already held, creating the reverse ordering:
rq lock -> scx_sched_lock vs. scx_sched_lock -> rq lock
Fix by flipping scx_bypass() to take rq lock first, and dropping
scx_sched_lock from sysrq_handle_sched_ext_dump() as scx_sched_all is
already RCU-traversable and scx_dump_lock now prevents dumping a dead
sched. This makes the consistent ordering rq lock -> scx_sched_lock.
Reported-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Link: https://lore.kernel.org/r/20260309163025.2240221-1-yphbchou0911@gmail.com
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_disable() directly called kthread_queue_work() which can acquire
worker->lock, pi_lock and rq->__lock. This made scx_disable() unsafe to
call while holding locks that conflict with this chain - in particular,
scx_claim_exit() calls scx_disable() for each descendant while holding
scx_sched_lock, which nests inside rq->__lock in scx_bypass().
The error path (scx_vexit()) was already bouncing through irq_work to
avoid this issue. Generalize the pattern to all scx_disable() calls by
always going through irq_work. irq_work_queue() is lockless and safe to
call from any context, and the actual kthread_queue_work() call happens
in the irq_work handler outside any locks.
Rename error_irq_work to disable_irq_work to reflect the broader usage.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add a dedicated scx_dump_lock and per-sched dump_disabled flag so that
debug dumping can be safely disabled during sched teardown without
relying on scx_sched_lock. This is a prep for the next patch which
decouples the sysrq dump path from scx_sched_lock to resolve a lock
ordering issue.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
sub_detach is the parent's op called to notify the parent that a child
is detaching. Test parent->ops.sub_detach instead of sch->ops.sub_detach.
Fixes: ebeca1f930 ("sched_ext: Introduce cgroup sub-sched support")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Prefer using IS_ERR_OR_NULL() over using IS_ERR() and a manual NULL
check.
Change generated with coccinelle.
Signed-off-by: Philipp Hahn <phahn-oss@avm.de>
Signed-off-by: Tejun Heo <tj@kernel.org>
If the cpuidle governor .select() callback is skipped because there
is only one idle state in the cpuidle driver, the .reflect() callback
should be skipped as well, at least for consistency (if not for
correctness), so do it.
Fixes: e5c9ffc6ae ("cpuidle: Skip governor when only one idle state is available")
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Reviewed-by: Christian Loehle <christian.loehle@arm.com>
Reviewed-by: Aboorva Devarajan <aboorvad@linux.ibm.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://patch.msgid.link/12857700.O9o76ZdvQC@rafael.j.wysocki
c2a57380df ("sched: Replace use of system_unbound_wq with system_dfl_wq")
converted system_unbound_wq usages in ext.c but missed the queue_rcu_work()
call in scx_kobj_release() which was added later by the dynamic scx_sched
allocation conversion. Apply the same conversion.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Marco Crivellari <marco.crivellari@suse.com>
Pull sched/core to resolve conflicts between:
c2a57380df ("sched: Replace use of system_unbound_wq with system_dfl_wq")
from the tip tree and commit:
cde94c032b ("sched_ext: Make watchdog sub-sched aware")
The latter moves around code modiefied by the former. Apply the changes in
the new locations.
Signed-off-by: Tejun Heo <tj@kernel.org>
While running scx_flatcg, dmesg prints "SCX_OPS_HAS_CGROUP_WEIGHT is
deprecated and a noop", in code, SCX_OPS_HAS_CGROUP_WEIGHT has been
marked as DEPRECATED, and will be removed on 6.18. Now it's time to do it.
Signed-off-by: Zhao Mengmeng <zhaomengmeng@kylinos.cn>
Signed-off-by: Tejun Heo <tj@kernel.org>
scx_enable() uses double-checked locking to lazily initialize a static
kthread_worker pointer. The fast path reads helper locklessly:
if (!READ_ONCE(helper)) { // lockless read -- no helper_mutex
The write side initializes helper under helper_mutex, but previously
used a plain assignment:
helper = kthread_run_worker(0, "scx_enable_helper");
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
plain write -- KCSAN data race with READ_ONCE() above
Since READ_ONCE() on the fast path and the plain write on the
initialization path access the same variable without a common lock,
they constitute a data race. KCSAN requires that all sides of a
lock-free access use READ_ONCE()/WRITE_ONCE() consistently.
Use a temporary variable to stage the result of kthread_run_worker(),
and only WRITE_ONCE() into helper after confirming the pointer is
valid. This avoids a window where a concurrent caller on the fast path
could observe an ERR pointer via READ_ONCE(helper) before the error
check completes.
Fixes: b06ccbabe2 ("sched_ext: Fix starvation of scx_enable() under fair-class saturation")
Signed-off-by: zhidao su <suzhidao@xiaomi.com>
Acked-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
This reverts commit 9adfcef334.
dsq->nr is protected by dsq->lock and reading while holding the lock doesn't
constitute a racy read.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: zhidao su <suzhidao@xiaomi.com>
PI and setscheduler() syscalls, resulting in kernel warnings
and misbehavior. Found during stress-testing.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=vP5l
-----END PGP SIGNATURE-----
Merge tag 'sched-urgent-2026-03-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler fix from Ingo Molnar:
"Fix a DL scheduler bug that may corrupt internal metrics during PI and
setscheduler() syscalls, resulting in kernel warnings and misbehavior.
Found during stress-testing"
* tag 'sched-urgent-2026-03-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/deadline: Fix missing ENQUEUE_REPLENISH during PI de-boosting
ffa7ae0724 ("sched_ext: Add reenq_flags plumbing to scx_bpf_dsq_reenq()")
introduced task_should_reenq() as a filter inside reenq_local(), requiring
SCX_REENQ_ANY to be set in order to match any task. scx_bpf_dsq_reenq()
handles this correctly by converting a bare reenq_flags=0 to SCX_REENQ_ANY,
but scx_bpf_reenqueue_local() was not updated and continued to call
reenq_local() with 0, causing it to silently reenqueue zero tasks.
Fix by passing SCX_REENQ_ANY directly.
Fixes: ffa7ae0724 ("sched_ext: Add reenq_flags plumbing to scx_bpf_dsq_reenq()")
Signed-off-by: Cheng-Yang Chou <yphbchou0911@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
SCX_ENQ_REENQ indicates that a task is being re-enqueued but doesn't tell the
BPF scheduler why. Add SCX_TASK_REENQ_REASON flags using bits 12-13 of
p->scx.flags to communicate the reason during ops.enqueue():
- NONE: Not being reenqueued
- KFUNC: Reenqueued by scx_bpf_dsq_reenq() and friends
More reasons will be added.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Task states (NONE, INIT, READY, ENABLED) were defined in a separate enum with
unshifted values and then shifted when stored in scx_entity.flags. Simplify by
defining them as pre-shifted values directly in scx_ent_flags and removing the
separate scx_task_state enum. This removes the need for shifting when
reading/writing state values.
scx_get_task_state() now returns the masked flags value directly.
scx_set_task_state() accepts the pre-shifted state value. scx_dump_task()
shifts down for display to maintain readable output.
No functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
schedule_dsq_reenq() always acquires deferred_reenq_lock to queue a reenqueue
request. Add a lockless fast-path to skip lock acquisition when the request is
already pending with the required flags set.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_bpf_dsq_reenq() currently only supports local DSQs. Extend it to support
user-defined DSQs by adding a deferred re-enqueue mechanism similar to the
local DSQ handling.
Add per-cpu deferred_reenq_user_node/flags to scx_dsq_pcpu and
deferred_reenq_users list to scx_rq. When scx_bpf_dsq_reenq() is called on a
user DSQ, the DSQ's per-cpu node is added to the current rq's deferred list.
process_deferred_reenq_users() then iterates the DSQ using the cursor helpers
and re-enqueues each task.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Factor out cursor-based DSQ iteration from bpf_iter_scx_dsq_next() into
nldsq_cursor_next_task() and the task-lost check from scx_dsq_move() into
nldsq_cursor_lost_task() to prepare for reuse.
As ->priv is only used to record dsq->seq for cursors, update
INIT_DSQ_LIST_CURSOR() to take the DSQ pointer and set ->priv from dsq->seq
so that users don't have to read it manually. Move scx_dsq_iter_flags enum
earlier so nldsq_cursor_next_task() can use SCX_DSQ_ITER_REV.
bypass_lb_cpu() now sets cursor.priv to dsq->seq but doesn't use it.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add per-CPU data structure to dispatch queues. Each DSQ now has a percpu
scx_dsq_pcpu which contains a back-pointer to the DSQ. This will be used by
future changes to implement per-CPU reenqueue tracking for user DSQs.
init_dsq() now allocates the percpu data and can fail, so it returns an
error code. All callers are updated to handle failures. exit_dsq() is added
to free the percpu data and is called from all DSQ cleanup paths.
In scx_bpf_create_dsq(), init_dsq() is called before rcu_read_lock() since
alloc_percpu() requires GFP_KERNEL context, and dsq->sched is set
afterwards.
v2: Fix err_free_pcpu to only exit_dsq() initialized bypass DSQs (Andrea
Righi).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add infrastructure to pass flags through the deferred reenqueue path.
reenq_local() now takes a reenq_flags parameter, and scx_sched_pcpu gains a
deferred_reenq_local_flags field to accumulate flags from multiple
scx_bpf_dsq_reenq() calls before processing. No flags are defined yet.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_bpf_reenqueue_local() can only trigger re-enqueue of the current CPU's
local DSQ. Introduce scx_bpf_dsq_reenq() which takes a DSQ ID and can target
any local DSQ including remote CPUs via SCX_DSQ_LOCAL_ON | cpu. This will be
expanded to support user DSQs by future changes.
scx_bpf_reenqueue_local() is reimplemented as a simple wrapper around
scx_bpf_dsq_reenq(SCX_DSQ_LOCAL, 0) and may be deprecated in the future.
Update compat.bpf.h with a compatibility shim and scx_qmap to test the new
functionality.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Wrap the deferred_reenq_local_node list_head into struct
scx_deferred_reenq_local. More fields will be added and this allows using a
shorthand pointer to access them.
No functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The deferred reenqueue local mechanism uses an llist (lockless list) for
collecting schedulers that need their local DSQs re-enqueued. Convert to a
regular list protected by a raw_spinlock.
The llist was used for its lockless properties, but the upcoming changes to
support remote reenqueue require more complex list operations that are
difficult to implement correctly with lockless data structures. A spinlock-
protected regular list provides the necessary flexibility.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Previously, both process_ddsp_deferred_locals() and reenq_local() required
forward declarations. Reorganize so that only run_deferred() needs to be
declared. Both callees are grouped right before run_deferred() for better
locality. This reduces forward declaration clutter and will ease adding more
to the run_deferred() path.
No functional changes.
v2: Also relocate process_ddsp_deferred_locals() next to run_deferred()
(Daniel Jordan).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Change find_global_dsq() to take a CPU number directly instead of a task
pointer. This prepares for callers where the CPU is available but the task is
not.
No functional changes.
v2: Rename tcpu to cpu in find_global_dsq() (Emil Tsalapatis).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Extract pnode allocation and deallocation logic into alloc_pnode() and
free_pnode() helpers. This simplifies scx_alloc_and_add_sched() and prepares
for adding more per-node initialization and cleanup in subsequent patches.
No functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Global DSQs are currently stored as an array of scx_dispatch_q pointers,
one per NUMA node. To allow adding more per-node data structures, wrap the
global DSQ in scx_sched_pnode and replace global_dsqs with pnode array.
NUMA-aware allocation is maintained. No functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Move scx_bpf_task_cgroup() kfunc definition and its BTF_ID entry to the end
of the kfunc section before __bpf_kfunc_end_defs() for cleaner code
organization.
No functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Reviewed-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
ops.quiescent() is invoked with the same deq_flags as ops.dequeue(), so
the BPF scheduler is able to distinguish sleep vs property changes in
both callbacks.
However, dequeue_task_scx() receives deq_flags as an int from the
sched_class interface, so SCX flags above bit 32 (%SCX_DEQ_SCHED_CHANGE)
are truncated. ops_dequeue() reconstructs the full u64 for ops.dequeue(),
but ops.quiescent() is still called with the original int and can never
see %SCX_DEQ_SCHED_CHANGE.
Fix this by constructing the full u64 deq_flags in dequeue_task_scx()
(renaming the int parameter to core_deq_flags) and passing the complete
flags to both ops_dequeue() and ops.quiescent().
Fixes: ebf1ccff79 ("sched_ext: Fix ops.dequeue() semantics")
Signed-off-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
enqueue_task_scx() takes int enq_flags from the sched_class interface.
SCX enqueue flags starting at bit 32 (SCX_ENQ_PREEMPT and above) are
silently truncated when passed through activate_task(). extra_enq_flags
was added as a workaround - storing high bits in rq->scx.extra_enq_flags
and OR-ing them back in enqueue_task_scx(). However, the OR target is
still the int parameter, so the high bits are lost anyway.
The current impact is limited as the only affected flag is SCX_ENQ_PREEMPT
which is informational to the BPF scheduler - its loss means the scheduler
doesn't know about preemption but doesn't cause incorrect behavior.
Fix by renaming the int parameter to core_enq_flags and introducing a
u64 enq_flags local that merges both sources. All downstream functions
already take u64 enq_flags.
Fixes: f0e1a0643a ("sched_ext: Implement BPF extensible scheduler class")
Cc: stable@vger.kernel.org # v6.12+
Acked-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
This is an early-stage partial implementation that demonstrates the core
building blocks for nested sub-scheduler dispatching. While significant
work remains in the enqueue path and other areas, this patch establishes
the fundamental mechanisms needed for hierarchical scheduler operation.
The key building blocks introduced include:
- Private stack support for ops.dispatch() to prevent stack overflow when
walking down nested schedulers during dispatch operations
- scx_bpf_sub_dispatch() kfunc that allows parent schedulers to trigger
dispatch operations on their direct child schedulers
- Proper parent-child relationship validation to ensure dispatch requests
are only made to legitimate child schedulers
- Updated scx_dispatch_sched() to handle both nested and non-nested
invocations with appropriate kf_mask handling
The qmap scheduler is updated to demonstrate the functionality by calling
scx_bpf_sub_dispatch() on registered child schedulers when it has no
tasks in its own queues.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add rhashtable-based lookup for sub-schedulers indexed by cgroup_id to
enable efficient scheduler discovery in preparation for multiple scheduler
support. The hash table allows quick lookup of the appropriate scheduler
instance when processing tasks from different cgroups.
This extends scx_link_sched() to register sub-schedulers in the hash table
and scx_unlink_sched() to remove them. A new scx_find_sub_sched() function
provides the lookup interface.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Factor out scx_link_sched() and scx_unlink_sched() functions to reduce
code duplication in the scheduler enable/disable paths.
No functional change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_bpf_reenqueue_local() currently re-enqueues all tasks on the local DSQ
regardless of which sub-scheduler owns them. With multiple sub-schedulers,
each should only re-enqueue tasks it owns or are owned by its descendants.
Replace the per-rq boolean flag with a lock-free linked list to track
per-scheduler reenqueue requests. Filter tasks in reenq_local() using
hierarchical ownership checks and block deferrals during bypass to prevent
use on dead schedulers.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add a back pointer from scx_sched_pcpu to scx_sched. This will be used by
the next patch to make scx_bpf_reenqueue_local() sub-sched aware.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The preceding changes implemented the framework to support cgroup
sub-scheds and updated scheduling paths and kfuncs so that they have
minimal but working support for sub-scheds. However, actual sub-sched
enabling/disabling hasn't been implemented yet and all tasks stayed on
scx_root.
Implement cgroup sub-sched enabling and disabling to actually activate
sub-scheds:
- Both enable and disable operations bypass only the tasks in the subtree
of the child being enabled or disabled to limit disruptions.
- When enabling, all candidate tasks are first initialized for the child
sched. Once that succeeds, the tasks are exited for the parent and then
switched over to the child. This adds a bit of complication but
guarantees that child scheduler failures are always contained.
- Disabling works the same way in the other direction. However, when the
parent may fail to initialize a task, disabling is propagated up to the
parent. While this means that a parent sched fail due to a child sched
event, the failure can only originate from the parent itself (its
ops.init_task()). The only effect a malfunctioning child can have on the
parent is attempting to move the tasks back to the parent.
After this change, although not all the necessary mechanisms are in place
yet, sub-scheds can take control of their tasks and schedule them.
v2: Fix missing scx_cgroup_unlock()/percpu_up_write() in abort path
(Cheng-Yang Chou).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Extend scx_dump_state() to support multiple schedulers and improve task
identification in dumps. The function now takes a specific scheduler to
dump and can optionally filter tasks by scheduler.
scx_dump_task() now displays which scheduler each task belongs to, using
"*" to mark tasks owned by the scheduler being dumped. Sub-schedulers
are identified with their level and cgroup ID.
The SysRq-D handler now iterates through all active schedulers under
scx_sched_lock and dumps each one separately. For SysRq-D dumps, only
tasks owned by each scheduler are dumped to avoid redundancy since all
schedulers are being dumped. Error-triggered dumps continue to dump all
tasks since only that specific scheduler is being dumped.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The scx_dump_state() function uses a regular spinlock to serialize
access. In a subsequent patch, this function will be called while
holding scx_sched_lock, which is a raw spinlock, creating a lock
nesting violation.
Convert the dump_lock to a raw spinlock and use the guard macro for
cleaner lock management.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Currently, the watchdog checks all tasks as if they are all on scx_root.
Move scx_watchdog_timeout inside scx_sched and make check_rq_for_timeouts()
use the timeout from the scx_sched associated with each task.
refresh_watchdog() is added, which determines the timer interval as half of
the shortest watchdog timeouts of all scheds and arms or disarms it as
necessary. Every scx_sched instance has equivalent or better detection
latency while sharing the same timer.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_dsp_ctx and scx_dsp_max_batch are global variables used in the dispatch
path. In prepration for multiple scheduler support, move the former into
scx_sched_pcpu and the latter into scx_sched. No user-visible behavior
changes intended.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The cgroup sub-sched support involves invasive changes to many areas of
sched_ext. The overall scaffolding is now in place and the next step is
implementing sub-sched enable/disable.
To enable partial testing and verification, update balance_one() to
dispatch from all scx_sched instances until it finds a task to run. This
should keep scheduling working when sub-scheds are enabled with tasks on
them. This will be replaced by BPF-driven hierarchical operation.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
When a sub-scheduler enters bypass mode, its tasks must be scheduled by an
ancestor to guarantee forward progress. Tasks from bypassing descendants are
queued in the bypass DSQs of the nearest non-bypassing ancestor, or the root
scheduler if all ancestors are bypassing. This requires coordination between
bypassing schedulers and their hosts.
Add bypass_enq_target_dsq() to find the correct bypass DSQ by walking up the
hierarchy until reaching a non-bypassing ancestor. When a sub-scheduler starts
bypassing, all its runnable tasks are re-enqueued after scx_bypassing() is set,
ensuring proper migration to ancestor bypass DSQs.
Update scx_dispatch_sched() to handle hosting bypassed descendants. When a
scheduler is not bypassing but has bypassing descendants, it must schedule both
its own tasks and bypassed descendant tasks. A simple policy is implemented
where every Nth dispatch attempt (SCX_BYPASS_HOST_NTH=2) consumes from the
bypass DSQ. A fallback consumption is also added at the end of dispatch to
ensure bypassed tasks make progress even when normal scheduling is idle.
Update enable_bypass_dsp() and disable_bypass_dsp() to increment
bypass_dsp_enable_depth on both the bypassing scheduler and its parent host,
ensuring both can detect that bypass dispatch is active through
bypass_dsp_enabled().
Add SCX_EV_SUB_BYPASS_DISPATCH event counter to track scheduling of bypassed
descendant tasks.
v2: Fix comment typos (Andrea).
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The bypass_depth field tracks nesting of bypass operations but is also used
to determine whether the bypass dispatch path should be active. With
hierarchical scheduling, child schedulers may need to activate their parent's
bypass dispatch path without affecting the parent's bypass_depth, requiring
separation of these concerns.
Add bypass_dsp_enable_depth and bypass_dsp_claim to independently control
bypass dispatch path activation. The new enable_bypass_dsp() and
disable_bypass_dsp() functions manage this state with proper claim semantics
to prevent races. The bypass dispatch path now only activates when
bypass_dsp_enabled() returns true, which checks the new enable_depth counter.
The disable operation is carefully ordered after all tasks are moved out of
bypass DSQs to ensure they are drained before the dispatch path is disabled.
During scheduler teardown, disable_bypass_dsp() is called explicitly to ensure
cleanup even if bypass mode was never entered normally.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The @prev parameter passed into ops.dispatch() is expected to be on the
same sched. Passing in @prev which isn't on the sched can spuriously
trigger failures that can kill the scheduler. Pass in @prev iff it's on
the same sched.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
In preparation of multiple scheduler support, factor out
scx_dispatch_sched() from balance_one(). The function boundary makes
remembering $prev_on_scx and $prev_on_rq less useful. Open code $prev_on_scx
in balance_one() and $prev_on_rq in both balance_one() and
scx_dispatch_sched().
No functional changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Bypass mode is used to simplify enable and disable paths and guarantee
forward progress when something goes wrong. When enabled, all tasks skip BPF
scheduling and fall back to simple in-kernel FIFO scheduling. While this
global behavior can be used as-is when dealing with sub-scheds, that would
allow any sub-sched instance to affect the whole system in a significantly
disruptive manner.
Make bypass state hierarchical by propagating it to descendants and updating
per-cpu flags accordingly. This allows an scx_sched to bypass if itself or
any of its ancestors are in bypass mode. However, this doesn't make the
actual bypass enqueue and dispatch paths hierarchical yet. That will be done
in later patches.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
In preparation of multiple scheduler support, make bypass state
per-scx_sched. Move scx_bypass_depth, bypass_timestamp and bypass_lb_timer
from globals into scx_sched. Move SCX_RQ_BYPASSING from rq to scx_sched_pcpu
as SCX_SCHED_PCPU_BYPASSING.
scx_bypass() now takes @sch and scx_rq_bypassing(rq) is replaced with
scx_bypassing(sch, cpu). All callers updated.
scx_bypassed_for_enable existed to balance the global scx_bypass_depth when
enable failed. Now that bypass_depth is per-scheduler, the counter is
destroyed along with the scheduler on enable failure. Remove
scx_bypassed_for_enable.
As all tasks currently use the root scheduler, there's no observable behavior
change.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
To support bypass mode for sub-schedulers, move bypass_dsq from struct scx_rq
to struct scx_sched_pcpu. Add bypass_dsq() helper. Move bypass_dsq
initialization from init_sched_ext_class() to scx_alloc_and_attach_sched().
bypass_lb_cpu() now takes a CPU number instead of rq pointer. All callers
updated. No behavior change as all tasks use the root scheduler.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The abort state was tracked in the global scx_aborting flag which was used to
break out of potential live-lock scenarios when an error occurs. With
hierarchical scheduling, each scheduler instance must track its own abort
state independently so that an aborting scheduler doesn't interfere with
others.
Move the aborting flag into struct scx_sched and update all access sites. The
early initialization check in scx_root_enable() that warned about residual
aborting state is no longer needed as each scheduler instance now starts with
a clean state.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
The default time slice was stored in the global scx_slice_dfl variable which
was dynamically modified when entering and exiting bypass mode. With
hierarchical scheduling, each scheduler instance needs its own default slice
configuration so that bypass operations on one scheduler don't affect others.
Move slice_dfl into struct scx_sched and update all access sites. The bypass
logic now modifies the root scheduler's slice_dfl. At task initialization in
init_scx_entity(), use the SCX_SLICE_DFL constant directly since the task may
not yet be associated with a specific scheduler.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Call ops.core_sched_before() iff both tasks belong to the same scx_sched.
Otherwise, use timestamp based ordering.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
- Add the @sch parameter to scx_init_task() and drop @tg as it can be
obtained from @p. Separate out __scx_init_task() which does everything
except for the task state transition.
- Add the @sch parameter to scx_enable_task(). Separate out
__scx_enable_task() which does everything except for the task state
transition.
- Add the @sch parameter to scx_disable_task().
- Rename scx_exit_task() to scx_disable_and_exit_task() and separate out
__scx_disable_and_exit_task() which does everything except for the task
state transition.
While some task state transitions are relocated, no meaningful behavior
changes are expected.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_bpf_dsq_move[_vtime]() calls scx_dsq_move() to move task from a DSQ to
another. However, @p doesn't necessarily have to come form the containing
iteration and can thus be a task which belongs to another scx_sched. Verify
that @p is on the same scx_sched as the DSQ being iterated.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
scx_bpf_task_set_slice() and scx_bpf_task_set_dsq_vtime() now verify that
the calling scheduler has authority over the task before allowing updates.
This prevents schedulers from modifying tasks that don't belong to them in
hierarchical scheduling configurations.
Direct writes to p->scx.slice and p->scx.dsq_vtime are deprecated and now
trigger warnings. They will be disallowed in a future release.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Add checks to enforce scheduling authority boundaries when multiple
schedulers are present:
1. In scx_dsq_insert_preamble() and the dispatch retry path, ignore attempts
to insert tasks that the scheduler doesn't own, counting them via
SCX_EV_INSERT_NOT_OWNED. As BPF schedulers are allowed to ignore
dequeues, such attempts can occur legitimately during sub-scheduler
enabling when tasks move between schedulers. The counter helps distinguish
normal cases from scheduler bugs.
2. For scx_bpf_dsq_insert_vtime() and scx_bpf_select_cpu_and(), error out
when sub-schedulers are attached. These functions lack the aux__prog
parameter needed to identify the calling scheduler, so they cannot be used
safely with multiple schedulers. BPF programs should use the arg-wrapped
versions (__scx_bpf_dsq_insert_vtime() and __scx_bpf_select_cpu_and())
instead.
These checks ensure that with multiple concurrent schedulers, scheduler
identity can be properly determined and unauthorized task operations are
prevented or tracked.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
In preparation for multiple scheduler support, introduce scx_prog_sched()
accessor which returns the scx_sched instance associated with a BPF program.
The association is determined via the special KF_IMPLICIT_ARGS kfunc
parameter, which provides access to bpf_prog_aux. This aux can be used to
retrieve the struct_ops (sched_ext_ops) that the program is associated with,
and from there, the corresponding scx_sched instance.
For compatibility, when ops.sub_attach is not implemented (older schedulers
without sub-scheduler support), unassociated programs fall back to scx_root.
A warning is logged once per scheduler for such programs.
As scx_root is still the only scheduler, this shouldn't introduce
user-visible behavior changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
In preparation of multiple scheduler support, add p->scx.sched which points
to the scx_sched instance that the task is scheduled by, which is currently
always scx_root. Add scx_task_sched[_rcu]() accessors which return the
associated scx_sched of the specified task and replace the raw scx_root
dereferences with it where applicable. scx_task_on_sched() is also added to
test whether a given task is on the specified sched.
As scx_root is still the only scheduler, this shouldn't introduce
user-visible behavior changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
A system often runs multiple workloads especially in multi-tenant server
environments where a system is split into partitions servicing separate
more-or-less independent workloads each requiring an application-specific
scheduler. To support such and other use cases, sched_ext is in the process
of growing multiple scheduler support.
When partitioning a system in terms of CPUs for such use cases, an
oft-taken approach is hard partitioning the system using cpuset. While it
would be possible to tie sched_ext multiple scheduler support to cpuset
partitions, such an approach would have fundamental limitations stemming
from the lack of dynamism and flexibility.
Users often don't care which specific CPUs are assigned to which workload
and want to take advantage of optimizations which are enabled by running
workloads on a larger machine - e.g. opportunistic over-commit, improving
latency critical workload characteristics while maintaining bandwidth
fairness, employing control mechanisms based on different criteria than
on-CPU time for e.g. flexible memory bandwidth isolation, packing similar
parts from different workloads on same L3s to improve cache efficiency,
and so on.
As this sort of dynamic behaviors are impossible or difficult to implement
with hard partitioning, sched_ext is implementing cgroup sub-sched support
where schedulers can be attached to the cgroup hierarchy and a parent
scheduler is responsible for controlling the CPUs that each child can use
at any given moment. This makes CPU distribution dynamically controlled by
BPF allowing high flexibility.
This patch adds the skeletal sched_ext cgroup sub-sched support:
- sched_ext_ops.sub_cgroup_id and .sub_attach/detach() are added. Non-zero
sub_cgroup_id indicates that the scheduler is to be attached to the
identified cgroup. A sub-sched is attached to the cgroup iff the nearest
ancestor scheduler implements .sub_attach() and grants the attachment. Max
nesting depth is limited by SCX_SUB_MAX_DEPTH.
- When a scheduler exits, all its descendant schedulers are exited
together. Also, cgroup.scx_sched added which points to the effective
scheduler instance for the cgroup. This is updated on scheduler
init/exit and inherited on cgroup online. When a cgroup is offlined, the
attached scheduler is automatically exited.
- Sub-sched support is gated on CONFIG_EXT_SUB_SCHED which is
automatically enabled if both SCX and cgroups are enabled. Sub-sched
support is not tied to the CPU controller but rather the cgroup
hierarchy itself. This is intentional as the support for cpu.weight and
cpu.max based resource control is orthogonal to sub-sched support. Note
that CONFIG_CGROUPS around cgroup subtree iteration support for
scx_task_iter is replaced with CONFIG_EXT_SUB_SCHED for consistency.
- This allows loading sub-scheds and most framework operations such as
propagating disable down the hierarchy work. However, sub-scheds are not
operational yet and all tasks stay with the root sched. This will serve
as the basis for building up full sub-sched support.
- DSQs point to the scx_sched they belong to.
- scx_qmap is updated to allow attachment of sub-scheds and also serving
as sub-scheds.
- scx_is_descendant() is added but not yet used in this patch. It is used by
later changes in the series and placed here as this is where the function
belongs.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
In preparation for multiple scheduler support, reorganize the enable and
disable paths to make scheduler instances explicit. Extract
scx_root_disable() from scx_disable_workfn(). Rename scx_enable_workfn()
to scx_root_enable_workfn(). Change scx_disable() to take @sch parameter
and only queue disable_work if scx_claim_exit() succeeds for consistency.
Move exit_kind validation into scx_claim_exit(). The sysrq handler now
prints a message when no scheduler is loaded.
These changes don't materially affect user-visible behavior.
v2: Keep scx_enable() name as-is and only rename the workfn to
scx_root_enable_workfn(). Change scx_enable() return type to s32.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
- Always trigger the warning if p->scx.disallow is set for fork inits. There
is no reason to set it during forks.
- Flip the positions of if/else arms to ease adding error conditions.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Make sched_cgroup_fork() pass @kargs to scx_fork(). This will be used to
determine @p's cgroup for cgroup sub-sched support.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
Cc: Peter Zijlstra <peterz@infradead.org>
For the planned cgroup sub-scheduler support, enable/disable operations are
going to be subtree specific and iterating all tasks in the system for those
operations can be unnecessarily expensive and disruptive.
cgroup already has mechanisms to perform subtree task iterations. Implement
cgroup subtree iteration for scx_task_iter:
- Add optional @cgrp to scx_task_iter_start() which enables cgroup subtree
iteration.
- Make scx_task_iter use css_next_descendant_pre() and css_task_iter to
iterate all tasks in the cgroup subtree.
- Update all existing callers to pass NULL to maintain current behavior.
The two iteration mechanisms are independent and duplicate. It's likely that
scx_tasks can be removed in favor of always using cgroup iteration if
CONFIG_SCHED_CLASS_EXT depends on CONFIG_CGROUPS.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>
