From: Mike Galbraith <efault@gmx.de>
To: Yong Zhang <yong.zhang0@gmail.com>
Cc: bharata@linux.vnet.ibm.com,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Ingo Molnar <mingo@elte.hu>,
linux-kernel@vger.kernel.org
Subject: Re: [patch] Re: autogroup: sched_setscheduler() fails
Date: Thu, 13 Jan 2011 04:54:50 +0100 [thread overview]
Message-ID: <1294890890.8089.39.camel@marge.simson.net> (raw)
In-Reply-To: <AANLkTimtwt+OGohMxGRC7L4Dm-h4wvCWYi-=13efqFtr@mail.gmail.com>
(mailer switched me to work account again. grrrr, switch yet again:)
On Wed, 2011-01-12 at 14:32 +0800, Yong Zhang wrote:
> Setting rt_runtime both in this patch and the previous one will make tune
> other's rt_runtime/rt_period trouble, I guess.
Yeah, as in making it impossible to allocate even 1 usec :) We need
zero bandwidth though, so tell __sched_setscheduler() that autogroups
having zero allocated is perfectly fine.
sched, autogroup: fix CONFIG_RT_GROUP_SCHED sched_setscheduler() failure.
If CONFIG_RT_GROUP_SCHED is set, __sched_setscheduler() fails due to autogroup
not allocating rt_runtime. Free unused/unusable rt_se and rt_rq, redirect RT
tasks to the root task group, and tell __sched_setscheduler() that it's ok.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Reported-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
---
kernel/sched.c | 3 ++-
kernel/sched_autogroup.c | 27 +++++++++++++++++++++++++++
kernel/sched_autogroup.h | 4 ++++
3 files changed, 33 insertions(+), 1 deletion(-)
Index: linux-2.6/kernel/sched_autogroup.c
===================================================================
--- linux-2.6.orig/kernel/sched_autogroup.c
+++ linux-2.6/kernel/sched_autogroup.c
@@ -27,6 +27,11 @@ static inline void autogroup_destroy(str
{
struct autogroup *ag = container_of(kref, struct autogroup, kref);
+#ifdef CONFIG_RT_GROUP_SCHED
+ /* We've redirected RT tasks to the root task group... */
+ ag->tg->rt_se = NULL;
+ ag->tg->rt_rq = NULL;
+#endif
sched_destroy_group(ag->tg);
}
@@ -55,6 +60,10 @@ static inline struct autogroup *autogrou
return ag;
}
+#ifdef CONFIG_RT_GROUP_SCHED
+static void free_rt_sched_group(struct task_group *tg);
+#endif
+
static inline struct autogroup *autogroup_create(void)
{
struct autogroup *ag = kzalloc(sizeof(*ag), GFP_KERNEL);
@@ -72,6 +81,19 @@ static inline struct autogroup *autogrou
init_rwsem(&ag->lock);
ag->id = atomic_inc_return(&autogroup_seq_nr);
ag->tg = tg;
+#ifdef CONFIG_RT_GROUP_SCHED
+ /*
+ * Autogroup RT tasks are redirected to the root task group
+ * so we don't have to move tasks around upon policy change,
+ * or flail around trying to allocate bandwidth on the fly.
+ * A bandwidth exception in __sched_setscheduler() allows
+ * the policy change to proceed. Thereafter, task_group()
+ * returns &root_task_group, so zero bandwidth is required.
+ */
+ free_rt_sched_group(tg);
+ tg->rt_se = root_task_group.rt_se;
+ tg->rt_rq = root_task_group.rt_rq;
+#endif
tg->autogroup = ag;
return ag;
@@ -106,6 +128,11 @@ task_wants_autogroup(struct task_struct
return true;
}
+static inline bool task_group_is_autogroup(struct task_group *tg)
+{
+ return tg != &root_task_group && tg->autogroup;
+}
+
static inline struct task_group *
autogroup_task_group(struct task_struct *p, struct task_group *tg)
{
Index: linux-2.6/kernel/sched_autogroup.h
===================================================================
--- linux-2.6.orig/kernel/sched_autogroup.h
+++ linux-2.6/kernel/sched_autogroup.h
@@ -15,6 +15,10 @@ autogroup_task_group(struct task_struct
static inline void autogroup_init(struct task_struct *init_task) { }
static inline void autogroup_free(struct task_group *tg) { }
+static inline bool task_group_is_autogroup(struct task_group *tg)
+{
+ return 0;
+}
static inline struct task_group *
autogroup_task_group(struct task_struct *p, struct task_group *tg)
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -4874,7 +4874,8 @@ recheck:
* assigned.
*/
if (rt_bandwidth_enabled() && rt_policy(policy) &&
- task_group(p)->rt_bandwidth.rt_runtime == 0) {
+ task_group(p)->rt_bandwidth.rt_runtime == 0 &&
+ !task_group_is_autogroup(task_group(p))) {
__task_rq_unlock(rq);
raw_spin_unlock_irqrestore(&p->pi_lock, flags);
return -EPERM;
next prev parent reply other threads:[~2011-01-13 3:54 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-10 9:16 Bharata B Rao
2011-01-10 10:29 ` [patch] " Mike Galbraith
2011-01-10 10:59 ` Peter Zijlstra
2011-01-10 16:42 ` Mike Galbraith
2011-01-11 17:10 ` Bharata B Rao
2011-01-11 18:48 ` Mike Galbraith
2011-01-12 3:37 ` Bharata B Rao
2011-01-12 5:40 ` Yong Zhang
2011-01-12 6:35 ` Bharata B Rao
2011-01-12 7:24 ` Mike Galbraith
2011-01-12 8:06 ` Bharata B Rao
2011-01-12 8:47 ` Mike Galbraith
2011-01-12 9:26 ` Bharata B Rao
2011-01-12 6:17 ` Mike Galbraith
2011-01-12 6:42 ` Bharata B Rao
2011-01-12 5:40 ` Mike Galbraith
2011-01-12 6:32 ` Yong Zhang
2011-01-12 8:55 ` Mike Galbraith
2011-01-13 3:54 ` Mike Galbraith [this message]
2011-01-13 5:59 ` Yong Zhang
2011-01-13 6:02 ` Yong Zhang
2011-01-13 6:13 ` Mike Galbraith
2011-01-13 8:46 ` Bharata B Rao
2011-01-17 13:16 ` Peter Zijlstra
2011-02-15 15:46 ` torbenh
2011-02-15 16:43 ` Mike Galbraith
2011-02-18 11:09 ` torbenh
2011-02-18 12:50 ` Mike Galbraith
2011-02-18 13:40 ` torbenh
2011-02-22 12:24 ` torbenh
2011-02-22 14:47 ` Mike Galbraith
2011-02-28 17:53 ` torbenh
2011-02-28 18:29 ` Mike Galbraith
2011-02-28 19:10 ` torbenh
2011-03-01 4:02 ` Mike Galbraith
2011-03-01 4:21 ` Mike Galbraith
2011-03-01 15:59 ` torbenh
2011-01-18 19:05 ` [tip:sched/urgent] sched, autogroup: Fix CONFIG_RT_GROUP_SCHED sched_setscheduler() failure tip-bot for Mike Galbraith
2011-01-12 5:43 ` [patch] Re: autogroup: sched_setscheduler() fails Yong Zhang
2011-01-12 6:25 ` Mike Galbraith
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1294890890.8089.39.camel@marge.simson.net \
--to=efault@gmx.de \
--cc=a.p.zijlstra@chello.nl \
--cc=bharata@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=yong.zhang0@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome