From nobody Mon Feb 9 19:42:29 2026 Delivered-To: importer@patchew.org Received-SPF: none (zoho.com: 192.237.175.120 is neither permitted nor denied by domain of lists.xenproject.org) client-ip=192.237.175.120; envelope-from=xen-devel-bounces@lists.xenproject.org; helo=lists.xenproject.org; Authentication-Results: mx.zohomail.com; spf=none (zoho.com: 192.237.175.120 is neither permitted nor denied by domain of lists.xenproject.org) smtp.mailfrom=xen-devel-bounces@lists.xenproject.org ARC-Seal: i=1; a=rsa-sha256; t=1559039671; cv=none; d=zoho.com; s=zohoarc; b=mn+mE2+XZlyGCUTTTirlHi3VtmaaDFpjFDRkKIvuCrd+4Dr4b1DltTSy92n6yh6HDOgw0HkVIo123Le80krmhnkAp0/NnFHfuGodK3RI/kZ7l07B6PTObKY5KS6EG+6jOoRt8kivyZbzXdDjEJa/JWtBcZ1wxWjq3mtDwdAu+Mc= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zoho.com; s=zohoarc; t=1559039671; h=Content-Type:Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To:ARC-Authentication-Results; bh=qNlPyF9E9YW04vjTy/i2JFelZmomYowldsxpapSmDUM=; b=POF23mXMvZUfH1o0WAovlikDEvExsMNoLyDN/SG1vRdzO8FqDMwvsRV7jh+Ry8HlvVHm0UNc/4MM1Wkzt1zV9HiV70T2jZQ/h/bMfzgtYNcT2r40HGiX1gFvlK6OH7iby5eMwOCqbHw+UH7NyzhpVoqoMpW9RMtrufvsmJDxLKw= ARC-Authentication-Results: i=1; mx.zoho.com; spf=none (zoho.com: 192.237.175.120 is neither permitted nor denied by domain of lists.xenproject.org) smtp.mailfrom=xen-devel-bounces@lists.xenproject.org Return-Path: Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) by mx.zohomail.com with SMTPS id 1559039671087765.1968736528539; Tue, 28 May 2019 03:34:31 -0700 (PDT) Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.89) (envelope-from ) id 1hVZQ4-0004Zs-9j; Tue, 28 May 2019 10:33:24 +0000 Received: from us1-rack-dfw2.inumbo.com ([104.130.134.6]) by lists.xenproject.org with esmtp (Exim 4.89) (envelope-from ) id 1hVZQ2-0004Xi-7R for xen-devel@lists.xenproject.org; Tue, 28 May 2019 10:33:22 +0000 Received: from mx1.suse.de (unknown [195.135.220.15]) by us1-rack-dfw2.inumbo.com (Halon) with ESMTPS id 030aa49b-8134-11e9-8980-bc764e045a96; Tue, 28 May 2019 10:33:20 +0000 (UTC) Received: from relay2.suse.de (unknown [195.135.220.254]) by mx1.suse.de (Postfix) with ESMTP id 35921B028; Tue, 28 May 2019 10:33:18 +0000 (UTC) X-Inumbo-ID: 030aa49b-8134-11e9-8980-bc764e045a96 X-Virus-Scanned: by amavisd-new at test-mx.suse.de From: Juergen Gross To: xen-devel@lists.xenproject.org Date: Tue, 28 May 2019 12:32:21 +0200 Message-Id: <20190528103313.1343-9-jgross@suse.com> X-Mailer: git-send-email 2.16.4 In-Reply-To: <20190528103313.1343-1-jgross@suse.com> References: <20190528103313.1343-1-jgross@suse.com> Subject: [Xen-devel] [PATCH 08/60] xen/sched: introduce struct sched_resource X-BeenThere: xen-devel@lists.xenproject.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Cc: Juergen Gross , Tim Deegan , Stefano Stabellini , Wei Liu , Konrad Rzeszutek Wilk , George Dunlap , Andrew Cooper , Ian Jackson , Dario Faggioli , Julien Grall , Meng Xu , Jan Beulich MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Errors-To: xen-devel-bounces@lists.xenproject.org Sender: "Xen-devel" Add a scheduling abstraction layer between physical processors and the schedulers by introducing a struct sched_resource. Each scheduler unit running is active on such a scheduler resource. For the time being there is one struct sched_resource per cpu, but in future there might be one for each core or socket only. Signed-off-by: Juergen Gross --- V1: add accessor functions --- xen/common/sched_credit.c | 2 ++ xen/common/sched_credit2.c | 7 +++++++ xen/common/sched_null.c | 3 +++ xen/common/sched_rt.c | 2 ++ xen/common/schedule.c | 15 +++++++++++++++ xen/include/xen/sched-if.h | 18 ++++++++++++++++++ xen/include/xen/sched.h | 3 +++ 7 files changed, 50 insertions(+) diff --git a/xen/common/sched_credit.c b/xen/common/sched_credit.c index 4cfef189aa..0cd139ec29 100644 --- a/xen/common/sched_credit.c +++ b/xen/common/sched_credit.c @@ -1031,6 +1031,7 @@ csched_unit_insert(const struct scheduler *ops, struc= t sched_unit *unit) lock =3D vcpu_schedule_lock_irq(vc); =20 vc->processor =3D csched_cpu_pick(ops, unit); + unit->res =3D get_sched_res(vc->processor); =20 spin_unlock_irq(lock); =20 @@ -1667,6 +1668,7 @@ csched_runq_steal(int peer_cpu, int cpu, int pri, int= balance_step) WARN_ON(vc->is_urgent); runq_remove(speer); vc->processor =3D cpu; + vc->sched_unit->res =3D get_sched_res(cpu); /* * speer will start executing directly on cpu, without having = to * go through runq_insert(). So we must update the runnable co= unt diff --git a/xen/common/sched_credit2.c b/xen/common/sched_credit2.c index 86b44dc6cf..c92a090905 100644 --- a/xen/common/sched_credit2.c +++ b/xen/common/sched_credit2.c @@ -2519,6 +2519,7 @@ static void migrate(const struct scheduler *ops, &trqd->active); svc->vcpu->processor =3D cpumask_cycle(trqd->pick_bias, cpumask_scratch_cpu(cpu)); + svc->vcpu->sched_unit->res =3D get_sched_res(svc->vcpu->processor); trqd->pick_bias =3D svc->vcpu->processor; ASSERT(svc->vcpu->processor < nr_cpu_ids); =20 @@ -2774,6 +2775,7 @@ csched2_unit_migrate( } _runq_deassign(svc); vc->processor =3D new_cpu; + unit->res =3D get_sched_res(new_cpu); return; } =20 @@ -2794,7 +2796,10 @@ csched2_unit_migrate( if ( trqd !=3D svc->rqd ) migrate(ops, svc, trqd, now); else + { vc->processor =3D new_cpu; + unit->res =3D get_sched_res(new_cpu); + } } =20 static int @@ -3119,6 +3124,7 @@ csched2_unit_insert(const struct scheduler *ops, stru= ct sched_unit *unit) lock =3D vcpu_schedule_lock_irq(vc); =20 vc->processor =3D csched2_cpu_pick(ops, unit); + unit->res =3D get_sched_res(vc->processor); =20 spin_unlock_irq(lock); =20 @@ -3596,6 +3602,7 @@ csched2_schedule( { snext->credit +=3D CSCHED2_MIGRATE_COMPENSATION; snext->vcpu->processor =3D cpu; + snext->vcpu->sched_unit->res =3D get_sched_res(cpu); SCHED_STAT_CRANK(migrated); ret.migrated =3D 1; } diff --git a/xen/common/sched_null.c b/xen/common/sched_null.c index b622c4d7dc..7c01d0272d 100644 --- a/xen/common/sched_null.c +++ b/xen/common/sched_null.c @@ -343,6 +343,7 @@ static void vcpu_assign(struct null_private *prv, struc= t vcpu *v, { per_cpu(npc, cpu).vcpu =3D v; v->processor =3D cpu; + v->sched_unit->res =3D get_sched_res(cpu); cpumask_clear_cpu(cpu, &prv->cpus_free); =20 dprintk(XENLOG_G_INFO, "%d <-- %pv\n", cpu, v); @@ -421,6 +422,7 @@ static void null_unit_insert(const struct scheduler *op= s, retry: =20 cpu =3D v->processor =3D pick_cpu(prv, v); + unit->res =3D get_sched_res(cpu); =20 spin_unlock(lock); =20 @@ -667,6 +669,7 @@ static void null_unit_migrate(const struct scheduler *o= ps, * by this, will be fixed-up during resume. */ v->processor =3D new_cpu; + unit->res =3D get_sched_res(new_cpu); } =20 #ifndef NDEBUG diff --git a/xen/common/sched_rt.c b/xen/common/sched_rt.c index 700ee9353e..715a7cd218 100644 --- a/xen/common/sched_rt.c +++ b/xen/common/sched_rt.c @@ -894,6 +894,7 @@ rt_unit_insert(const struct scheduler *ops, struct sche= d_unit *unit) =20 /* This is safe because vc isn't yet being scheduled */ vc->processor =3D rt_cpu_pick(ops, unit); + unit->res =3D get_sched_res(vc->processor); =20 lock =3D vcpu_schedule_lock_irq(vc); =20 @@ -1124,6 +1125,7 @@ rt_schedule(const struct scheduler *ops, s_time_t now= , bool_t tasklet_work_sched if ( snext->vcpu->processor !=3D cpu ) { snext->vcpu->processor =3D cpu; + snext->vcpu->sched_unit->res =3D get_sched_res(cpu); ret.migrated =3D 1; } ret.time =3D snext->cur_budget; /* invoke the scheduler next time = */ diff --git a/xen/common/schedule.c b/xen/common/schedule.c index 49d25489ef..32ce248f24 100644 --- a/xen/common/schedule.c +++ b/xen/common/schedule.c @@ -63,6 +63,7 @@ static void poll_timer_fn(void *data); /* This is global for now so that private implementations can reach it */ DEFINE_PER_CPU(struct schedule_data, schedule_data); DEFINE_PER_CPU(struct scheduler *, scheduler); +DEFINE_PER_CPU(struct sched_resource *, sched_res); =20 /* Scratch space for cpumasks. */ DEFINE_PER_CPU(cpumask_t, cpumask_scratch); @@ -305,6 +306,7 @@ int sched_init_vcpu(struct vcpu *v, unsigned int proces= sor) if ( (unit =3D sched_alloc_unit(v)) =3D=3D NULL ) return 1; =20 + unit->res =3D get_sched_res(processor); /* Initialise the per-vcpu timers. */ init_timer(&v->periodic_timer, vcpu_periodic_timer_fn, v, v->processor); @@ -419,6 +421,7 @@ int sched_move_domain(struct domain *d, struct cpupool = *c) sched_set_affinity(v, &cpumask_all, &cpumask_all); =20 v->processor =3D new_p; + v->sched_unit->res =3D get_sched_res(new_p); /* * With v->processor modified we must not * - make any further changes assuming we hold the scheduler lock, @@ -787,9 +790,11 @@ void restore_vcpu_affinity(struct domain *d) } =20 v->processor =3D cpumask_any(cpumask_scratch_cpu(cpu)); + v->sched_unit->res =3D get_sched_res(v->processor); =20 lock =3D vcpu_schedule_lock_irq(v); v->processor =3D sched_pick_cpu(vcpu_scheduler(v), v->sched_unit); + v->sched_unit->res =3D get_sched_res(v->processor); spin_unlock_irq(lock); =20 if ( old_cpu !=3D v->processor ) @@ -1618,6 +1623,13 @@ static int cpu_schedule_up(unsigned int cpu) { struct schedule_data *sd =3D &per_cpu(schedule_data, cpu); void *sched_priv; + struct sched_resource *res; + + res =3D xzalloc(struct sched_resource); + if ( res =3D=3D NULL ) + return -ENOMEM; + res->processor =3D cpu; + set_sched_res(cpu, res); =20 per_cpu(scheduler, cpu) =3D &ops; spin_lock_init(&sd->_lock); @@ -1682,6 +1694,9 @@ static void cpu_schedule_down(unsigned int cpu) sd->sched_priv =3D NULL; =20 kill_timer(&sd->s_timer); + + set_sched_res(cpu, NULL); + xfree(sd); } =20 static int cpu_schedule_callback( diff --git a/xen/include/xen/sched-if.h b/xen/include/xen/sched-if.h index 8ddbeb4ffd..bc7f223aad 100644 --- a/xen/include/xen/sched-if.h +++ b/xen/include/xen/sched-if.h @@ -44,9 +44,24 @@ struct schedule_data { =20 #define curr_on_cpu(c) (per_cpu(schedule_data, c).curr) =20 +struct sched_resource { + unsigned int processor; +}; + DECLARE_PER_CPU(struct schedule_data, schedule_data); DECLARE_PER_CPU(struct scheduler *, scheduler); DECLARE_PER_CPU(struct cpupool *, cpupool); +DECLARE_PER_CPU(struct sched_resource *, sched_res); + +static inline struct sched_resource *get_sched_res(unsigned int cpu) +{ + return per_cpu(sched_res, cpu); +} + +static inline void set_sched_res(unsigned int cpu, struct sched_resource *= res) +{ + per_cpu(sched_res, cpu) =3D res; +} =20 /* * Scratch space, for avoiding having too many cpumask_t on the stack. @@ -334,7 +349,10 @@ static inline void sched_migrate(const struct schedule= r *s, if ( s->migrate ) s->migrate(s, unit, cpu); else + { unit->vcpu->processor =3D cpu; + unit->res =3D per_cpu(sched_res, cpu); + } } =20 static inline int sched_pick_cpu(const struct scheduler *s, diff --git a/xen/include/xen/sched.h b/xen/include/xen/sched.h index 4da1ab201d..d3a1a31c86 100644 --- a/xen/include/xen/sched.h +++ b/xen/include/xen/sched.h @@ -276,10 +276,13 @@ struct vcpu struct arch_vcpu arch; }; =20 +struct sched_resource; + struct sched_unit { struct vcpu *vcpu; void *priv; /* scheduler private data */ struct sched_unit *next_in_list; + struct sched_resource *res; }; =20 #define for_each_sched_unit(d, e) \ --=20 2.16.4 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel