public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
To: linux-kernel@vger.kernel.org, linux-s390@vger.kernel.org
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>,
	Jan Glauber <jang@linux.vnet.ibm.com>,
	Martin Schwidefsky <schwidefsky@de.ibm.com>
Subject: [patch 05/14] extend virtual timer interface by mod_virt_timer_periodic
Date: Tue, 14 Apr 2009 15:53:32 +0200	[thread overview]
Message-ID: <20090414135347.536476032@de.ibm.com> (raw)
In-Reply-To: 20090414135327.576111185@de.ibm.com

[-- Attachment #1: 005-mod-vtimer.diff --]
[-- Type: text/plain, Size: 4428 bytes --]

From: Jan Glauber <jang@linux.vnet.ibm.com>

In case mod_virt_timer is used to add a non pending timer the timer
is always added as a one-shot timer. If mod_virt_timer is used for
periodic timers they may therfore be degraded to one-shot timers.

Add mod_virt_timer_periodic to the interface to allow safe re-programming
of the interval value.

Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---

 arch/s390/include/asm/timer.h |    1 
 arch/s390/kernel/vtime.c      |   57 ++++++++++++++++++++++++------------------
 2 files changed, 34 insertions(+), 24 deletions(-)

Index: quilt-2.6/arch/s390/include/asm/timer.h
===================================================================
--- quilt-2.6.orig/arch/s390/include/asm/timer.h
+++ quilt-2.6/arch/s390/include/asm/timer.h
@@ -41,6 +41,7 @@ extern void init_virt_timer(struct vtime
 extern void add_virt_timer(void *new);
 extern void add_virt_timer_periodic(void *new);
 extern int mod_virt_timer(struct vtimer_list *timer, __u64 expires);
+extern int mod_virt_timer_periodic(struct vtimer_list *timer, __u64 expires);
 extern int del_virt_timer(struct vtimer_list *timer);
 
 extern void init_cpu_vtimer(void);
Index: quilt-2.6/arch/s390/kernel/vtime.c
===================================================================
--- quilt-2.6.orig/arch/s390/kernel/vtime.c
+++ quilt-2.6/arch/s390/kernel/vtime.c
@@ -425,17 +425,7 @@ void add_virt_timer_periodic(void *new)
 }
 EXPORT_SYMBOL(add_virt_timer_periodic);
 
-/*
- * If we change a pending timer the function must be called on the CPU
- * where the timer is running on, e.g. by smp_call_function_single()
- *
- * The original mod_timer adds the timer if it is not pending. For
- * compatibility we do the same. The timer will be added on the current
- * CPU as a oneshot timer.
- *
- * returns whether it has modified a pending timer (1) or not (0)
- */
-int mod_virt_timer(struct vtimer_list *timer, __u64 expires)
+int __mod_vtimer(struct vtimer_list *timer, __u64 expires, int periodic)
 {
 	struct vtimer_queue *vq;
 	unsigned long flags;
@@ -444,39 +434,35 @@ int mod_virt_timer(struct vtimer_list *t
 	BUG_ON(!timer->function);
 	BUG_ON(!expires || expires > VTIMER_MAX_SLICE);
 
-	/*
-	 * This is a common optimization triggered by the
-	 * networking code - if the timer is re-modified
-	 * to be the same thing then just return:
-	 */
 	if (timer->expires == expires && vtimer_pending(timer))
 		return 1;
 
 	cpu = get_cpu();
 	vq = &per_cpu(virt_cpu_timer, cpu);
 
-	/* check if we run on the right CPU */
-	BUG_ON(timer->cpu != cpu);
-
 	/* disable interrupts before test if timer is pending */
 	spin_lock_irqsave(&vq->lock, flags);
 
 	/* if timer isn't pending add it on the current CPU */
 	if (!vtimer_pending(timer)) {
 		spin_unlock_irqrestore(&vq->lock, flags);
-		/* we do not activate an interval timer with mod_virt_timer */
-		timer->interval = 0;
+
+		if (periodic)
+			timer->interval = expires;
+		else
+			timer->interval = 0;
 		timer->expires = expires;
 		timer->cpu = cpu;
 		internal_add_vtimer(timer);
 		return 0;
 	}
 
+	/* check if we run on the right CPU */
+	BUG_ON(timer->cpu != cpu);
+
 	list_del_init(&timer->entry);
 	timer->expires = expires;
-
-	/* also change the interval if we have an interval timer */
-	if (timer->interval)
+	if (periodic)
 		timer->interval = expires;
 
 	/* the timer can't expire anymore so we can release the lock */
@@ -484,9 +470,32 @@ int mod_virt_timer(struct vtimer_list *t
 	internal_add_vtimer(timer);
 	return 1;
 }
+
+/*
+ * If we change a pending timer the function must be called on the CPU
+ * where the timer is running on.
+ *
+ * returns whether it has modified a pending timer (1) or not (0)
+ */
+int mod_virt_timer(struct vtimer_list *timer, __u64 expires)
+{
+	return __mod_vtimer(timer, expires, 0);
+}
 EXPORT_SYMBOL(mod_virt_timer);
 
 /*
+ * If we change a pending timer the function must be called on the CPU
+ * where the timer is running on.
+ *
+ * returns whether it has modified a pending timer (1) or not (0)
+ */
+int mod_virt_timer_periodic(struct vtimer_list *timer, __u64 expires)
+{
+	return __mod_vtimer(timer, expires, 1);
+}
+EXPORT_SYMBOL(mod_virt_timer_periodic);
+
+/*
  * delete a virtual timer
  *
  * returns whether the deleted timer was pending (1) or not (0)

-- 
blue skies,
   Martin.

"Reality continues to ruin my life." - Calvin.


  parent reply	other threads:[~2009-04-14 13:55 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-14 13:53 [patch 00/14] s390 patches for 2.6.30-rc1 Martin Schwidefsky
2009-04-14 13:53 ` [patch 01/14] s390: move machine flags to lowcore Martin Schwidefsky
2009-04-14 13:53 ` [patch 02/14] wire up preadv/pwritev system calls Martin Schwidefsky
2009-04-14 13:53 ` [patch 03/14] call nmi_enter/nmi_exit on machine checks Martin Schwidefsky
2009-04-14 13:53 ` [patch 04/14] stp synchronization retry timer Martin Schwidefsky
2009-04-14 13:53 ` Martin Schwidefsky [this message]
2009-04-14 13:53 ` [patch 06/14] appldata: Use new mod_virt_timer_periodic() function Martin Schwidefsky
2009-04-14 13:53 ` [patch 07/14] qdio: remove dead timeout handler Martin Schwidefsky
2009-04-14 13:53 ` [patch 08/14] dasd: Use the new async framework for autoonlining Martin Schwidefsky
2009-04-14 13:53 ` [patch 09/14] dasd: fix idaw boundary checking for track based ccw Martin Schwidefsky
2009-04-14 13:53 ` [patch 10/14] smp: fix cpu_possible_map initialization Martin Schwidefsky
2009-04-15  2:06   ` Rusty Russell
2009-04-14 13:53 ` [patch 11/14] fix idle time accounting Martin Schwidefsky
2009-04-14 13:53 ` [patch 12/14] cpu hotplug and accounting values Martin Schwidefsky
2009-04-14 13:53 ` [patch 13/14] add read_persistent_clock Martin Schwidefsky
2009-04-14 13:53 ` [patch 14/14] boot cputime accounting Martin Schwidefsky

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=20090414135347.536476032@de.ibm.com \
    --to=schwidefsky@de.ibm.com \
    --cc=heiko.carstens@de.ibm.com \
    --cc=jang@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-s390@vger.kernel.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