* [PATCH v2] sched_ext: pass the initial cpu.idle state in scx_cgroup_init_args @ 2026-08-24 14:28 Tao Cui 2026-08-24 15:07 ` Andrea Righi 2026-08-24 16:54 ` Tejun Heo 0 siblings, 2 replies; 5+ messages in thread From: Tao Cui @ 2026-08-24 14:28 UTC (permalink / raw) To: tj, void, arighi Cc: changwoo, suzhidao, sched-ext, linux-kernel, bpf, cui.tao, Tao Cui From: Tao Cui <cuitao@kylinos.cn> scx_cgroup_init_args carries the initial weight and bandwidth control parameters of a cgroup to ops.cgroup_init(), but not its cpu.idle state. A cgroup that was already configured idle before the scheduler was loaded (or before it was onlined under it) is presented as non-idle, and the BPF scheduler only learns about it if cpu.idle is written again later. Add the idle state to scx_cgroup_init_args and fill it in all four places that build the args: scx_tg_online() for cgroups onlined under the scheduler, scx_cgroup_init() for cgroups that already exist when the scheduler is loaded, and the sub-scheduler handover paths scx_cgroup_claim_subtree() and scx_cgroup_return_subtree(). Verified in a VM with a probe scheduler printing the init args: a cgroup configured cpu.idle=1 before loading shows idle=1 in ops.cgroup_init(), the default shows 0, and later cpu.idle writes still come through ops.cgroup_set_idle(). The sub-scheduler paths are compile-tested only. Signed-off-by: Tao Cui <cuitao@kylinos.cn> --- v1 -> v2: Also fill .idle in the sub-scheduler handover paths in sub.c, scx_cgroup_claim_subtree() and scx_cgroup_return_subtree(), missed in v1 and pointed out by Andrea Righi and the sashiko AI review bot. v1: https://lore.kernel.org/r/20260824133954.561956-1-cui.tao@linux.dev kernel/sched/ext/ext.c | 4 +++- kernel/sched/ext/internal.h | 3 +++ kernel/sched/ext/sub.c | 2 ++ 3 files changed, 8 insertions(+), 1 deletion(-) diff --git a/kernel/sched/ext/ext.c b/kernel/sched/ext/ext.c index b646711a45fe..a7218314dcfb 100644 --- a/kernel/sched/ext/ext.c +++ b/kernel/sched/ext/ext.c @@ -4764,7 +4764,8 @@ int scx_tg_online(struct task_group *tg) { .weight = tg->scx.weight, .bw_period_us = tg->scx.bw_period_us, .bw_quota_us = tg->scx.bw_quota_us, - .bw_burst_us = tg->scx.bw_burst_us }; + .bw_burst_us = tg->scx.bw_burst_us, + .idle = tg->scx.idle }; ret = SCX_CALL_OP_RET(sch, cgroup_init, NULL, tg->css.cgroup, &args); @@ -5185,6 +5186,7 @@ static int scx_cgroup_init(struct scx_sched *sch) .bw_period_us = tg->scx.bw_period_us, .bw_quota_us = tg->scx.bw_quota_us, .bw_burst_us = tg->scx.bw_burst_us, + .idle = tg->scx.idle, }; ret = SCX_CALL_OP_RET(sch, cgroup_init, NULL, css->cgroup, &args); diff --git a/kernel/sched/ext/internal.h b/kernel/sched/ext/internal.h index 53e136a47924..aa149a9c29f7 100644 --- a/kernel/sched/ext/internal.h +++ b/kernel/sched/ext/internal.h @@ -259,6 +259,9 @@ struct scx_cgroup_init_args { u64 bw_period_us; u64 bw_quota_us; u64 bw_burst_us; + + /* whether the cgroup is configured idle via cpu.idle */ + bool idle; }; enum scx_cpu_preempt_reason { diff --git a/kernel/sched/ext/sub.c b/kernel/sched/ext/sub.c index 0554448835bd..9701bd1d8ea0 100644 --- a/kernel/sched/ext/sub.c +++ b/kernel/sched/ext/sub.c @@ -1361,6 +1361,7 @@ static s32 scx_cgroup_claim_subtree(struct scx_sched *sch) .bw_period_us = tg->scx.bw_period_us, .bw_quota_us = tg->scx.bw_quota_us, .bw_burst_us = tg->scx.bw_burst_us, + .idle = tg->scx.idle, }; if (tg->scx.sched != parent || @@ -1464,6 +1465,7 @@ static void scx_cgroup_return_subtree(struct scx_sched *sch) .bw_period_us = tg->scx.bw_period_us, .bw_quota_us = tg->scx.bw_quota_us, .bw_burst_us = tg->scx.bw_burst_us, + .idle = tg->scx.idle, }; /* the first pass must have transferred everything */ -- 2.43.0 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH v2] sched_ext: pass the initial cpu.idle state in scx_cgroup_init_args 2026-08-24 14:28 [PATCH v2] sched_ext: pass the initial cpu.idle state in scx_cgroup_init_args Tao Cui @ 2026-08-24 15:07 ` Andrea Righi 2026-08-25 2:20 ` Tao Cui 2026-08-24 16:54 ` Tejun Heo 1 sibling, 1 reply; 5+ messages in thread From: Andrea Righi @ 2026-08-24 15:07 UTC (permalink / raw) To: Tao Cui; +Cc: tj, void, changwoo, suzhidao, sched-ext, linux-kernel, bpf, Tao Cui Hi Tao, On Mon, Aug 24, 2026 at 10:28:16PM +0800, Tao Cui wrote: > From: Tao Cui <cuitao@kylinos.cn> > > scx_cgroup_init_args carries the initial weight and bandwidth control > parameters of a cgroup to ops.cgroup_init(), but not its cpu.idle > state. A cgroup that was already configured idle before the scheduler > was loaded (or before it was onlined under it) is presented as > non-idle, and the BPF scheduler only learns about it if cpu.idle is > written again later. > > Add the idle state to scx_cgroup_init_args and fill it in all four > places that build the args: scx_tg_online() for cgroups onlined under > the scheduler, scx_cgroup_init() for cgroups that already exist when > the scheduler is loaded, and the sub-scheduler handover paths > scx_cgroup_claim_subtree() and scx_cgroup_return_subtree(). > > Verified in a VM with a probe scheduler printing the init args: a > cgroup configured cpu.idle=1 before loading shows idle=1 in > ops.cgroup_init(), the default shows 0, and later cpu.idle writes > still come through ops.cgroup_set_idle(). The sub-scheduler paths > are compile-tested only. > > Signed-off-by: Tao Cui <cuitao@kylinos.cn> We should probably add: Fixes: 347ed2d566da ("sched/ext: Implement cgroup_set_idle() callback" With that: Reviewed-by: Andrea Righi <arighi@nvidia.com> Thanks, -Andrea > --- > v1 -> v2: Also fill .idle in the sub-scheduler handover paths in > sub.c, scx_cgroup_claim_subtree() and scx_cgroup_return_subtree(), > missed in v1 and pointed out by Andrea Righi and the sashiko AI > review bot. > > v1: https://lore.kernel.org/r/20260824133954.561956-1-cui.tao@linux.dev > > kernel/sched/ext/ext.c | 4 +++- > kernel/sched/ext/internal.h | 3 +++ > kernel/sched/ext/sub.c | 2 ++ > 3 files changed, 8 insertions(+), 1 deletion(-) > > diff --git a/kernel/sched/ext/ext.c b/kernel/sched/ext/ext.c > index b646711a45fe..a7218314dcfb 100644 > --- a/kernel/sched/ext/ext.c > +++ b/kernel/sched/ext/ext.c > @@ -4764,7 +4764,8 @@ int scx_tg_online(struct task_group *tg) > { .weight = tg->scx.weight, > .bw_period_us = tg->scx.bw_period_us, > .bw_quota_us = tg->scx.bw_quota_us, > - .bw_burst_us = tg->scx.bw_burst_us }; > + .bw_burst_us = tg->scx.bw_burst_us, > + .idle = tg->scx.idle }; > > ret = SCX_CALL_OP_RET(sch, cgroup_init, > NULL, tg->css.cgroup, &args); > @@ -5185,6 +5186,7 @@ static int scx_cgroup_init(struct scx_sched *sch) > .bw_period_us = tg->scx.bw_period_us, > .bw_quota_us = tg->scx.bw_quota_us, > .bw_burst_us = tg->scx.bw_burst_us, > + .idle = tg->scx.idle, > }; > > ret = SCX_CALL_OP_RET(sch, cgroup_init, NULL, css->cgroup, &args); > diff --git a/kernel/sched/ext/internal.h b/kernel/sched/ext/internal.h > index 53e136a47924..aa149a9c29f7 100644 > --- a/kernel/sched/ext/internal.h > +++ b/kernel/sched/ext/internal.h > @@ -259,6 +259,9 @@ struct scx_cgroup_init_args { > u64 bw_period_us; > u64 bw_quota_us; > u64 bw_burst_us; > + > + /* whether the cgroup is configured idle via cpu.idle */ > + bool idle; > }; > > enum scx_cpu_preempt_reason { > diff --git a/kernel/sched/ext/sub.c b/kernel/sched/ext/sub.c > index 0554448835bd..9701bd1d8ea0 100644 > --- a/kernel/sched/ext/sub.c > +++ b/kernel/sched/ext/sub.c > @@ -1361,6 +1361,7 @@ static s32 scx_cgroup_claim_subtree(struct scx_sched *sch) > .bw_period_us = tg->scx.bw_period_us, > .bw_quota_us = tg->scx.bw_quota_us, > .bw_burst_us = tg->scx.bw_burst_us, > + .idle = tg->scx.idle, > }; > > if (tg->scx.sched != parent || > @@ -1464,6 +1465,7 @@ static void scx_cgroup_return_subtree(struct scx_sched *sch) > .bw_period_us = tg->scx.bw_period_us, > .bw_quota_us = tg->scx.bw_quota_us, > .bw_burst_us = tg->scx.bw_burst_us, > + .idle = tg->scx.idle, > }; > > /* the first pass must have transferred everything */ > -- > 2.43.0 > ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] sched_ext: pass the initial cpu.idle state in scx_cgroup_init_args 2026-08-24 15:07 ` Andrea Righi @ 2026-08-25 2:20 ` Tao Cui 0 siblings, 0 replies; 5+ messages in thread From: Tao Cui @ 2026-08-25 2:20 UTC (permalink / raw) To: Andrea Righi Cc: cui.tao, tj, void, changwoo, suzhidao, sched-ext, linux-kernel, bpf, Tao Cui Hello, Andrea, 在 2026/8/24 23:07, Andrea Righi 写道: > Hi Tao, > > On Mon, Aug 24, 2026 at 10:28:16PM +0800, Tao Cui wrote: >> From: Tao Cui <cuitao@kylinos.cn> >> >> scx_cgroup_init_args carries the initial weight and bandwidth control >> parameters of a cgroup to ops.cgroup_init(), but not its cpu.idle >> state. A cgroup that was already configured idle before the scheduler >> was loaded (or before it was onlined under it) is presented as >> non-idle, and the BPF scheduler only learns about it if cpu.idle is >> written again later. >> >> Add the idle state to scx_cgroup_init_args and fill it in all four >> places that build the args: scx_tg_online() for cgroups onlined under >> the scheduler, scx_cgroup_init() for cgroups that already exist when >> the scheduler is loaded, and the sub-scheduler handover paths >> scx_cgroup_claim_subtree() and scx_cgroup_return_subtree(). >> >> Verified in a VM with a probe scheduler printing the init args: a >> cgroup configured cpu.idle=1 before loading shows idle=1 in >> ops.cgroup_init(), the default shows 0, and later cpu.idle writes >> still come through ops.cgroup_set_idle(). The sub-scheduler paths >> are compile-tested only. >> >> Signed-off-by: Tao Cui <cuitao@kylinos.cn> > > We should probably add: > > Fixes: 347ed2d566da ("sched/ext: Implement cgroup_set_idle() callback" > > With that: > > Reviewed-by: Andrea Righi <arighi@nvidia.com> > Thank you for the review, and for catching the sub.c handover paths earlier. v3 adds the Fixes: tag you suggested and carries your Reviewed-by. The field is also renamed to sched_idle per Tejun's review; the rename is mechanical, so I kept the tag, but I'm happy to wait for a re-review if you prefer. Thanks, Tao > Thanks, > -Andrea > >> --- >> v1 -> v2: Also fill .idle in the sub-scheduler handover paths in >> sub.c, scx_cgroup_claim_subtree() and scx_cgroup_return_subtree(), >> missed in v1 and pointed out by Andrea Righi and the sashiko AI >> review bot. >> >> v1: https://lore.kernel.org/r/20260824133954.561956-1-cui.tao@linux.dev >> >> kernel/sched/ext/ext.c | 4 +++- >> kernel/sched/ext/internal.h | 3 +++ >> kernel/sched/ext/sub.c | 2 ++ >> 3 files changed, 8 insertions(+), 1 deletion(-) >> >> diff --git a/kernel/sched/ext/ext.c b/kernel/sched/ext/ext.c >> index b646711a45fe..a7218314dcfb 100644 >> --- a/kernel/sched/ext/ext.c >> +++ b/kernel/sched/ext/ext.c >> @@ -4764,7 +4764,8 @@ int scx_tg_online(struct task_group *tg) >> { .weight = tg->scx.weight, >> .bw_period_us = tg->scx.bw_period_us, >> .bw_quota_us = tg->scx.bw_quota_us, >> - .bw_burst_us = tg->scx.bw_burst_us }; >> + .bw_burst_us = tg->scx.bw_burst_us, >> + .idle = tg->scx.idle }; >> >> ret = SCX_CALL_OP_RET(sch, cgroup_init, >> NULL, tg->css.cgroup, &args); >> @@ -5185,6 +5186,7 @@ static int scx_cgroup_init(struct scx_sched *sch) >> .bw_period_us = tg->scx.bw_period_us, >> .bw_quota_us = tg->scx.bw_quota_us, >> .bw_burst_us = tg->scx.bw_burst_us, >> + .idle = tg->scx.idle, >> }; >> >> ret = SCX_CALL_OP_RET(sch, cgroup_init, NULL, css->cgroup, &args); >> diff --git a/kernel/sched/ext/internal.h b/kernel/sched/ext/internal.h >> index 53e136a47924..aa149a9c29f7 100644 >> --- a/kernel/sched/ext/internal.h >> +++ b/kernel/sched/ext/internal.h >> @@ -259,6 +259,9 @@ struct scx_cgroup_init_args { >> u64 bw_period_us; >> u64 bw_quota_us; >> u64 bw_burst_us; >> + >> + /* whether the cgroup is configured idle via cpu.idle */ >> + bool idle; >> }; >> >> enum scx_cpu_preempt_reason { >> diff --git a/kernel/sched/ext/sub.c b/kernel/sched/ext/sub.c >> index 0554448835bd..9701bd1d8ea0 100644 >> --- a/kernel/sched/ext/sub.c >> +++ b/kernel/sched/ext/sub.c >> @@ -1361,6 +1361,7 @@ static s32 scx_cgroup_claim_subtree(struct scx_sched *sch) >> .bw_period_us = tg->scx.bw_period_us, >> .bw_quota_us = tg->scx.bw_quota_us, >> .bw_burst_us = tg->scx.bw_burst_us, >> + .idle = tg->scx.idle, >> }; >> >> if (tg->scx.sched != parent || >> @@ -1464,6 +1465,7 @@ static void scx_cgroup_return_subtree(struct scx_sched *sch) >> .bw_period_us = tg->scx.bw_period_us, >> .bw_quota_us = tg->scx.bw_quota_us, >> .bw_burst_us = tg->scx.bw_burst_us, >> + .idle = tg->scx.idle, >> }; >> >> /* the first pass must have transferred everything */ >> -- >> 2.43.0 >> ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] sched_ext: pass the initial cpu.idle state in scx_cgroup_init_args 2026-08-24 14:28 [PATCH v2] sched_ext: pass the initial cpu.idle state in scx_cgroup_init_args Tao Cui 2026-08-24 15:07 ` Andrea Righi @ 2026-08-24 16:54 ` Tejun Heo 2026-08-25 2:26 ` Tao Cui 1 sibling, 1 reply; 5+ messages in thread From: Tejun Heo @ 2026-08-24 16:54 UTC (permalink / raw) To: Tao Cui Cc: void, arighi, changwoo, suzhidao, emil, sched-ext, linux-kernel, bpf, cuitao Hello, Tao. On Mon, Aug 24, 2026 at 10:28:16PM +0800, Tao Cui wrote: > + > + /* whether the cgroup is configured idle via cpu.idle */ > + bool idle; In sched_ext, a bare "idle" reads as CPU idle state (ops.update_idle(), idle cpumasks, scx_bpf_pick_idle_cpu()). cpu.idle is the cgroup analog of the SCHED_IDLE policy, so please name the field sched_idle instead. Renaming tg->scx.idle to match would be nice too. Thanks. -- tejun ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2] sched_ext: pass the initial cpu.idle state in scx_cgroup_init_args 2026-08-24 16:54 ` Tejun Heo @ 2026-08-25 2:26 ` Tao Cui 0 siblings, 0 replies; 5+ messages in thread From: Tao Cui @ 2026-08-25 2:26 UTC (permalink / raw) To: Tejun Heo Cc: cui.tao, void, arighi, changwoo, suzhidao, emil, sched-ext, linux-kernel, bpf, cuitao Hello, Tejun, 在 2026/8/25 00:54, Tejun Heo 写道: > Hello, Tao. > > On Mon, Aug 24, 2026 at 10:28:16PM +0800, Tao Cui wrote: >> + >> + /* whether the cgroup is configured idle via cpu.idle */ >> + bool idle; > > In sched_ext, a bare "idle" reads as CPU idle state (ops.update_idle(), > idle cpumasks, scx_bpf_pick_idle_cpu()). cpu.idle is the cgroup analog > of the SCHED_IDLE policy, so please name the field sched_idle instead. > Renaming tg->scx.idle to match would be nice too. > Thanks for the review. Both are renamed in v3: the new scx_cgroup_init_args field, and tg->scx.idle -> tg->scx.sched_idle as a separate second patch in the series, so 1/2 stays minimal for stable backport. Thanks, Tao > Thanks. > ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2026-08-25 2:27 UTC | newest] Thread overview: 5+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2026-08-24 14:28 [PATCH v2] sched_ext: pass the initial cpu.idle state in scx_cgroup_init_args Tao Cui 2026-08-24 15:07 ` Andrea Righi 2026-08-25 2:20 ` Tao Cui 2026-08-24 16:54 ` Tejun Heo 2026-08-25 2:26 ` Tao Cui
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox