public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: Esben Nielsen <nielsen.esben@googlemail.com>
Cc: Esben Nielsen <nielsen.esben@gogglemail.com>,
	Steven Rostedt <rostedt@goodmis.org>, Ingo Molnar <mingo@elte.hu>,
	linux-kernel@vger.kernel.org
Subject: Re: Why can't I set the priority of softirq-hrt? (Re: 2.6.17-rt1)
Date: Wed, 21 Jun 2006 18:26:04 +0200	[thread overview]
Message-ID: <1150907165.25491.4.camel@localhost.localdomain> (raw)
In-Reply-To: <Pine.LNX.4.64.0606211638560.6572@localhost.localdomain>

On Wed, 2006-06-21 at 16:43 +0100, Esben Nielsen wrote:
> What about the patch below. It compiles and my UP labtop runs fine, but I 
> haven't otherwise tested it.  My labtop runs RTExec without hichups 
> until now.

NAK, it puts the burden into the lock path and also does a remove / add
which results in walking the chain twice.

Find an version against the code in -mm below. Not too much tested yet.

	tglx

Index: linux-2.6.17-rc6-mm/kernel/rtmutex.c
===================================================================
--- linux-2.6.17-rc6-mm.orig/kernel/rtmutex.c	2006-06-21 13:34:34.000000000 +0200
+++ linux-2.6.17-rc6-mm/kernel/rtmutex.c	2006-06-21 15:45:21.000000000 +0200
@@ -160,7 +160,8 @@
 static int rt_mutex_adjust_prio_chain(task_t *task,
 				      int deadlock_detect,
 				      struct rt_mutex *orig_lock,
-				      struct rt_mutex_waiter *orig_waiter)
+				      struct rt_mutex_waiter *orig_waiter,
+				      struct task_struct *top_task)
 {
 	struct rt_mutex *lock;
 	struct rt_mutex_waiter *waiter, *top_waiter = orig_waiter;
@@ -188,7 +189,7 @@
 			prev_max = max_lock_depth;
 			printk(KERN_WARNING "Maximum lock depth %d reached "
 			       "task: %s (%d)\n", max_lock_depth,
-			       current->comm, current->pid);
+			       top_task->comm, top_task->pid);
 		}
 		put_task_struct(task);
 
@@ -228,7 +229,7 @@
 	}
 
 	/* Deadlock detection */
-	if (lock == orig_lock || rt_mutex_owner(lock) == current) {
+	if (lock == orig_lock || rt_mutex_owner(lock) == top_task) {
 		debug_rt_mutex_deadlock(deadlock_detect, orig_waiter, lock);
 		spin_unlock(&lock->wait_lock);
 		ret = deadlock_detect ? -EDEADLK : 0;
@@ -448,7 +449,8 @@
 
 	spin_unlock(&lock->wait_lock);
 
-	res = rt_mutex_adjust_prio_chain(owner, detect_deadlock, lock, waiter);
+	res = rt_mutex_adjust_prio_chain(owner, detect_deadlock, lock, waiter,
+					 current);
 
 	spin_lock(&lock->wait_lock);
 
@@ -561,12 +563,35 @@
 
 	spin_unlock(&lock->wait_lock);
 
-	rt_mutex_adjust_prio_chain(owner, 0, lock, NULL);
+	rt_mutex_adjust_prio_chain(owner, 0, lock, NULL, current);
 
 	spin_lock(&lock->wait_lock);
 }
 
 /*
+ * Recheck the pi chain, in case we got a priority setting
+ *
+ * Called from sched_setscheduler
+ */
+void rt_mutex_adjust_pi(struct task_struct *task)
+{
+	struct rt_mutex_waiter *waiter = task->pi_blocked_on;
+	unsigned long flags;
+
+	spin_lock_irqsave(&task->pi_lock, flags);
+
+	if (!waiter || waiter->list_entry.prio == task->prio) {
+		spin_unlock_irqrestore(&task->pi_lock, flags);
+		return;
+	}
+
+	get_task_struct(task);
+	spin_unlock_irqrestore(&task->pi_lock, flags);
+
+	rt_mutex_adjust_prio_chain(task, 0, NULL, NULL, task);
+}
+
+/*
  * Slow path lock function:
  */
 static int __sched
@@ -633,6 +658,7 @@
 			if (unlikely(ret))
 				break;
 		}
+
 		spin_unlock(&lock->wait_lock);
 
 		debug_rt_mutex_print_deadlock(&waiter);
Index: linux-2.6.17-rc6-mm/kernel/sched.c
===================================================================
--- linux-2.6.17-rc6-mm.orig/kernel/sched.c	2006-06-21 13:34:34.000000000 +0200
+++ linux-2.6.17-rc6-mm/kernel/sched.c	2006-06-21 18:21:16.000000000 +0200
@@ -4028,8 +4028,7 @@
 int sched_setscheduler(struct task_struct *p, int policy,
 		       struct sched_param *param)
 {
-	int retval;
-	int oldprio, oldpolicy = -1;
+	int retval, oldprio, oldpolicy = -1;
 	prio_array_t *array;
 	unsigned long flags;
 	runqueue_t *rq;
@@ -4119,6 +4118,8 @@
 	__task_rq_unlock(rq);
 	spin_unlock_irqrestore(&p->pi_lock, flags);
 
+	rt_mutex_adjust_pi(p);
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(sched_setscheduler);
Index: linux-2.6.17-rc6-mm/include/linux/sched.h
===================================================================
--- linux-2.6.17-rc6-mm.orig/include/linux/sched.h	2006-06-21 13:37:21.000000000 +0200
+++ linux-2.6.17-rc6-mm/include/linux/sched.h	2006-06-21 15:43:23.000000000 +0200
@@ -1123,11 +1123,13 @@
 #ifdef CONFIG_RT_MUTEXES
 extern int rt_mutex_getprio(task_t *p);
 extern void rt_mutex_setprio(task_t *p, int prio);
+extern void rt_mutex_adjust_pi(task_t *p);
 #else
 static inline int rt_mutex_getprio(task_t *p)
 {
 	return p->normal_prio;
 }
+# define rt_mutex_adjust_pi(p)		do { } while (0)
 #endif
 
 extern void set_user_nice(task_t *p, long nice);



  parent reply	other threads:[~2006-06-21 16:24 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-06-18  7:06 2.6.17-rt1 Ingo Molnar
2006-06-18 16:13 ` 2.6.17-rt1 Michal Piotrowski
     [not found] ` <Pine.LNX.4.64.0606201656230.11643@localhost.localdomain>
2006-06-20 15:13   ` Why can't I set the priority of softirq-hrt? (Re: 2.6.17-rt1) Thomas Gleixner
2006-06-20 17:09     ` Esben Nielsen
2006-06-20 16:35       ` Thomas Gleixner
2006-06-20 21:16         ` Esben Nielsen
2006-06-20 20:35           ` Thomas Gleixner
2006-06-20 23:19             ` Esben Nielsen
2006-06-20 16:39       ` Steven Rostedt
2006-06-20 18:12         ` Esben Nielsen
2006-06-20 17:21           ` Thomas Gleixner
2006-06-20 21:26             ` Esben Nielsen
2006-06-20 20:51               ` Thomas Gleixner
2006-06-21  8:20               ` Steven Rostedt
2006-06-21 11:05                 ` Esben Nielsen
2006-06-21 15:43                   ` Esben Nielsen
2006-06-21 15:21                     ` Steven Rostedt
2006-06-21 16:37                       ` Esben Nielsen
2006-06-21 15:51                         ` Steven Rostedt
2006-06-21 17:14                           ` Esben Nielsen
2006-06-21 16:26                     ` Thomas Gleixner [this message]
2006-06-21 18:30                       ` Ingo Molnar
2006-06-22 10:28                         ` Esben Nielsen
2006-06-21 21:29                       ` Esben Nielsen
2006-06-21 20:33                         ` Thomas Gleixner
2006-06-21 23:35                           ` Esben Nielsen
2006-06-22  7:06                             ` Thomas Gleixner
2006-06-22 10:32                               ` Esben Nielsen
2006-06-22 13:33                               ` Steven Rostedt
2006-06-22 13:45                       ` Steven Rostedt
2006-06-22 14:20                         ` Thomas Gleixner
2006-06-22 14:23                           ` Steven Rostedt
2006-06-22 14:26                             ` Thomas Gleixner
2006-06-22 18:06                               ` Esben Nielsen
2006-06-22 18:05                                 ` Thomas Gleixner
2006-06-23 11:23                                   ` Esben Nielsen
2006-06-23 11:06                                     ` Steven Rostedt
2006-07-03 11:48                                     ` Esben Nielsen
2006-06-21  8:13           ` Steven Rostedt
2006-06-21 11:03             ` Esben Nielsen
2006-06-22  0:57 ` 2.6.17-rt1 Lee Revell
2006-06-22  2:51   ` More weird latency trace output (was Re: 2.6.17-rt1) Lee Revell
2006-06-23  1:24     ` Lee Revell
2006-06-24 22:15       ` Lee Revell
2006-06-24 22:12         ` Ingo Molnar
2006-06-24 22:31           ` Lee Revell
2006-06-24 23:49           ` Lee Revell
2006-06-23 20:56 ` 2.6.17-rt1 - mm_struct leak Vernon Mauery
2006-06-24  9:24   ` Mark Hounschell
2006-06-24  9:32     ` Mark Hounschell
2006-06-30 16:02   ` [PATCH -RT]Re: " Vernon Mauery

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=1150907165.25491.4.camel@localhost.localdomain \
    --to=tglx@linutronix.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=nielsen.esben@gogglemail.com \
    --cc=nielsen.esben@googlemail.com \
    --cc=rostedt@goodmis.org \
    /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