| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Wed, 4 Sep 2019 21:44:23 -0400
From: Julien Desfossez <jdesfossez@...italocean.com>
To: Tim Chen <tim.c.chen@...ux.intel.com>
Cc: Dario Faggioli <dfaggioli@...e.com>,
"Li, Aubrey" <aubrey.li@...ux.intel.com>,
Aaron Lu <aaron.lu@...ux.alibaba.com>,
Aubrey Li <aubrey.intel@...il.com>,
Subhra Mazumdar <subhra.mazumdar@...cle.com>,
Vineeth Remanan Pillai <vpillai@...italocean.com>,
Nishanth Aravamudan <naravamudan@...italocean.com>,
Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...nel.org>,
Thomas Gleixner <tglx@...utronix.de>,
Paul Turner <pjt@...gle.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Linux List Kernel Mailing <linux-kernel@...r.kernel.org>,
Frédéric Weisbecker <fweisbec@...il.com>,
Kees Cook <keescook@...omium.org>,
Greg Kerr <kerrnel@...gle.com>, Phil Auld <pauld@...hat.com>,
Valentin Schneider <valentin.schneider@....com>,
Mel Gorman <mgorman@...hsingularity.net>,
Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>,
Paolo Bonzini <pbonzini@...hat.com>
Subject: Re: [RFC PATCH v3 00/16] Core scheduling v3
> 1) Unfairness between the sibling threads
> -----------------------------------------
> One sibling thread could be suppressing and force idling
> the sibling thread over proportionally. Resulting in
> the force idled CPU not getting run and stall tasks on
> suppressed CPU.
>
> Status:
> i) Aaron has proposed a patchset here based on using one
> rq as a base reference for vruntime for task priority
> comparison between siblings.
>
> https://lore.kernel.org/lkml/20190725143248.GC992@aaronlu/
> It works well on fairness but has some initialization issues
>
> ii) Tim has proposed a patchset here to account for forced
> idle time in rq's min_vruntime
> https://lore.kernel.org/lkml/f96350c1-25a9-0564-ff46-6658e96d726c@linux.intel.com/
> It improves over v3 with simpler logic compared to
> Aaron's patch, but does not work as well on fairness
>
> iii) Tim has proposed yet another patch to maintain fairness
> of forced idle time between CPU threads per Peter's suggestion.
> https://lore.kernel.org/lkml/21933a50-f796-3d28-664c-030cb7c98431@linux.intel.com/
> Its performance has yet to be tested.
>
> 2) Not rescheduling forced idled CPU
> ------------------------------------
> The forced idled CPU does not get a chance to re-schedule
> itself, and will stall for a long time even though it
> has eligible tasks to run.
>
> Status:
> i) Aaron proposed a patch to fix this to check if there
> are runnable tasks when scheduling tick comes in.
> https://lore.kernel.org/lkml/20190725143344.GD992@aaronlu/
>
> ii) Vineeth has patches to this issue and also issue 1, based
> on scheduling in a new "forced idle task" when getting forced
> idle, but has yet to post the patches.
We finished writing and debugging the PoC for the coresched_idle task
and here are the results and the code.
Those patches are applied on top of Aaron's patches:
- sched: Fix incorrect rq tagged as forced idle
- wrapper for cfs_rq->min_vruntime
https://lore.kernel.org/lkml/20190725143127.GB992@aaronlu/
- core vruntime comparison
https://lore.kernel.org/lkml/20190725143248.GC992@aaronlu/
For the testing, we used the same strategy as described in
https://lore.kernel.org/lkml/20190802153715.GA18075@sinkpad/
No tag
------
Test Average Stdev
Alone 1306.90 0.94
nosmt 649.95 1.44
Aaron's full patchset: 828.15 32.45
Aaron's first 2 patches: 832.12 36.53
Tim's first patchset: 852.50 4.11
Tim's second patchset: 855.11 9.89
coresched_idle 985.67 0.83
Sysbench mem untagged, sysbench cpu tagged
------------------------------------------
Test Average Stdev
Alone 1306.90 0.94
nosmt 649.95 1.44
Aaron's full patchset: 586.06 1.77
Tim's first patchset: 852.50 4.11
Tim's second patchset: 663.88 44.43
coresched_idle 653.58 0.49
Sysbench mem tagged, sysbench cpu untagged
------------------------------------------
Test Average Stdev
Alone 1306.90 0.94
nosmt 649.95 1.44
Aaron's full patchset: 583.77 3.52
Tim's first patchset: 564.04 58.05
Tim's second patchset: 524.72 55.24
coresched_idle 653.30 0.81
Both sysbench tagged
--------------------
Test Average Stdev
Alone 1306.90 0.94
nosmt 649.95 1.44
Aaron's full patchset: 582.15 3.75
Tim's first patchset: 679.43 70.07
Tim's second patchset: 563.10 34.58
coresched_idle 653.12 1.68
As we can see from this stress-test, with the coresched_idle thread
being a real process, the fairness is more consistent (low stdev). Also,
the performance remains the same regardless of the tagging, and even
always slightly better than nosmt.
Thanks,
Julien
From: vpillai <vpillai@...italocean.com>
Date: Wed, 4 Sep 2019 17:41:38 +0000
Subject: [RFC PATCH 1/2] coresched_idle thread
---
kernel/sched/core.c | 46 ++++++++++++++++++++++++++++++++++++++++++++
kernel/sched/sched.h | 1 +
2 files changed, 47 insertions(+)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index f7839bf96e8b..fe560739c247 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3639,6 +3639,51 @@ static inline bool cookie_match(struct task_struct *a, struct task_struct *b)
return a->core_cookie == b->core_cookie;
}
+static int coresched_idle_worker(void *data)
+{
+ struct rq *rq = (struct rq *)data;
+
+ /*
+ * Transition to parked state and dequeue from runqueue.
+ * pick_task() will select us if needed without enqueueing.
+ */
+ set_special_state(TASK_PARKED);
+ schedule();
+
+ while (true) {
+ if (kthread_should_stop())
+ break;
+
+ play_idle(1);
+ }
+
+ return 0;
+}
+
+static void coresched_idle_worker_init(struct rq *rq)
+{
+
+ // XXX core_idle_task needs lock protection?
+ if (!rq->core_idle_task) {
+ rq->core_idle_task = kthread_create_on_cpu(coresched_idle_worker,
+ (void *)rq, cpu_of(rq), "coresched_idle");
+ if (rq->core_idle_task) {
+ wake_up_process(rq->core_idle_task);
+ }
+
+ }
+
+ return;
+}
+
+static void coresched_idle_worker_fini(struct rq *rq)
+{
+ if (rq->core_idle_task) {
+ kthread_stop(rq->core_idle_task);
+ rq->core_idle_task = NULL;
+ }
+}
+
// XXX fairness/fwd progress conditions
/*
* Returns
@@ -6774,6 +6819,7 @@ void __init sched_init(void)
atomic_set(&rq->nr_iowait, 0);
#ifdef CONFIG_SCHED_CORE
+ rq->core_idle_task = NULL;
rq->core = NULL;
rq->core_pick = NULL;
rq->core_enabled = 0;
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index e91c188a452c..c3ae0af55b05 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -965,6 +965,7 @@ struct rq {
unsigned int core_sched_seq;
struct rb_root core_tree;
bool core_forceidle;
+ struct task_struct *core_idle_task;
/* shared state */
unsigned int core_task_seq;
--
2.17.1
From: vpillai <vpillai@...italocean.com>
Date: Wed, 4 Sep 2019 18:22:55 +0000
Subject: [RFC PATCH 2/2] Use coresched_idle to force idle a sibling
Currently we use idle thread to force idle on a sibling. Lets
use the new coresched_idle thread that scheduler sees a valid
task during force idle.
---
kernel/sched/core.c | 66 ++++++++++++++++++++++++++++++++++++++-------
1 file changed, 56 insertions(+), 10 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index fe560739c247..e35d69a81adb 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -244,23 +244,33 @@ static int __sched_core_stopper(void *data)
static DEFINE_MUTEX(sched_core_mutex);
static int sched_core_count;
+static void coresched_idle_worker_init(struct rq *rq);
+static void coresched_idle_worker_fini(struct rq *rq);
static void __sched_core_enable(void)
{
+ int cpu;
+
// XXX verify there are no cookie tasks (yet)
static_branch_enable(&__sched_core_enabled);
stop_machine(__sched_core_stopper, (void *)true, NULL);
+ for_each_online_cpu(cpu)
+ coresched_idle_worker_init(cpu_rq(cpu));
printk("core sched enabled\n");
}
static void __sched_core_disable(void)
{
+ int cpu;
+
// XXX verify there are no cookie tasks (left)
stop_machine(__sched_core_stopper, (void *)false, NULL);
static_branch_disable(&__sched_core_enabled);
+ for_each_online_cpu(cpu)
+ coresched_idle_worker_fini(cpu_rq(cpu));
printk("core sched disabled\n");
}
@@ -3626,14 +3636,25 @@ __pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
#ifdef CONFIG_SCHED_CORE
+static inline bool is_force_idle_task(struct task_struct *p)
+{
+ BUG_ON(task_rq(p)->core_idle_task == NULL);
+ return task_rq(p)->core_idle_task == p;
+}
+
+static inline bool is_core_idle_task(struct task_struct *p)
+{
+ return is_idle_task(p) || is_force_idle_task(p);
+}
+
static inline bool cookie_equals(struct task_struct *a, unsigned long cookie)
{
- return is_idle_task(a) || (a->core_cookie == cookie);
+ return is_core_idle_task(a) || (a->core_cookie == cookie);
}
static inline bool cookie_match(struct task_struct *a, struct task_struct *b)
{
- if (is_idle_task(a) || is_idle_task(b))
+ if (is_core_idle_task(a) || is_core_idle_task(b))
return true;
return a->core_cookie == b->core_cookie;
@@ -3641,8 +3662,6 @@ static inline bool cookie_match(struct task_struct *a, struct task_struct *b)
static int coresched_idle_worker(void *data)
{
- struct rq *rq = (struct rq *)data;
-
/*
* Transition to parked state and dequeue from runqueue.
* pick_task() will select us if needed without enqueueing.
@@ -3666,7 +3685,7 @@ static void coresched_idle_worker_init(struct rq *rq)
// XXX core_idle_task needs lock protection?
if (!rq->core_idle_task) {
rq->core_idle_task = kthread_create_on_cpu(coresched_idle_worker,
- (void *)rq, cpu_of(rq), "coresched_idle");
+ NULL, cpu_of(rq), "coresched_idle");
if (rq->core_idle_task) {
wake_up_process(rq->core_idle_task);
}
@@ -3684,6 +3703,14 @@ static void coresched_idle_worker_fini(struct rq *rq)
}
}
+static inline struct task_struct *core_idle_task(struct rq *rq)
+{
+ BUG_ON(rq->core_idle_task == NULL);
+
+ return rq->core_idle_task;
+
+}
+
// XXX fairness/fwd progress conditions
/*
* Returns
@@ -3709,7 +3736,7 @@ pick_task(struct rq *rq, const struct sched_class *class, struct task_struct *ma
*/
if (max && class_pick->core_cookie &&
prio_less(class_pick, max))
- return idle_sched_class.pick_task(rq);
+ return core_idle_task(rq);
return class_pick;
}
@@ -3853,7 +3880,7 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
goto done;
}
- if (!is_idle_task(p))
+ if (!is_force_idle_task(p))
occ++;
rq_i->core_pick = p;
@@ -3906,7 +3933,6 @@ next_class:;
rq->core->core_pick_seq = rq->core->core_task_seq;
next = rq->core_pick;
rq->core_sched_seq = rq->core->core_pick_seq;
- trace_printk("picked: %s/%d %lx\n", next->comm, next->pid, next->core_cookie);
/*
* Reschedule siblings
@@ -3924,13 +3950,24 @@ next_class:;
WARN_ON_ONCE(!rq_i->core_pick);
- if (is_idle_task(rq_i->core_pick) && rq_i->nr_running)
+ if (is_core_idle_task(rq_i->core_pick) && rq_i->nr_running) {
+ /*
+ * Matching logic can sometimes select idle_task when
+ * iterating the sched_classes. If that selection is
+ * actually a forced idle case, we need to update the
+ * core_pick to coresched_idle.
+ */
+ if (is_idle_task(rq_i->core_pick))
+ rq_i->core_pick = core_idle_task(rq_i);
rq_i->core_forceidle = true;
+ }
rq_i->core_pick->core_occupation = occ;
- if (i == cpu)
+ if (i == cpu) {
+ next = rq_i->core_pick;
continue;
+ }
if (rq_i->curr != rq_i->core_pick) {
trace_printk("IPI(%d)\n", i);
@@ -3947,6 +3984,7 @@ next_class:;
WARN_ON_ONCE(1);
}
}
+ trace_printk("picked: %s/%d %lx\n", next->comm, next->pid, next->core_cookie);
done:
set_next_task(rq, next);
@@ -4200,6 +4238,12 @@ static void __sched notrace __schedule(bool preempt)
* is a RELEASE barrier),
*/
++*switch_count;
+#ifdef CONFIG_SCHED_CORE
+ if (next == rq->core_idle_task)
+ next->state = TASK_RUNNING;
+ else if (prev == rq->core_idle_task)
+ prev->state = TASK_PARKED;
+#endif
trace_sched_switch(preempt, prev, next);
@@ -6479,6 +6523,7 @@ int sched_cpu_activate(unsigned int cpu)
#ifdef CONFIG_SCHED_CORE
if (static_branch_unlikely(&__sched_core_enabled)) {
rq->core_enabled = true;
+ coresched_idle_worker_init(rq);
}
#endif
}
@@ -6535,6 +6580,7 @@ int sched_cpu_deactivate(unsigned int cpu)
struct rq *rq = cpu_rq(cpu);
if (static_branch_unlikely(&__sched_core_enabled)) {
rq->core_enabled = false;
+ coresched_idle_worker_fini(rq);
}
#endif
static_branch_dec_cpuslocked(&sched_smt_present);
--
2.17.1
Powered by blists - more mailing lists