All of lore.kernel.org
 help / color / mirror / Atom feed
From: Yinghai Lu <yinghai@kernel.org>
To: Ingo Molnar <mingo@elte.hu>,
	Rusty Russell <rusty@rustcorp.com.au>,
	Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	Andrew Morton <akpm@linux-foundation.org>
Cc: Gary Hade <garyhade@us.ibm.com>,
	lcm@us.ibm.com, "Pallipadi,
	Venkatesh" <venkatesh.pallipadi@intel.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: [PATCH 3/4] irq: only update affinity in chip set_affinity() -v3
Date: Tue, 14 Apr 2009 13:43:33 -0700	[thread overview]
Message-ID: <49E4F575.7060201@kernel.org> (raw)
In-Reply-To: <49E4F513.1060709@kernel.org>


Impact: keep affinity consistent

irq_set_affinity() and move_masked_irq() try to assign affinity
before calling chip set_affinity(). some archs are assigning again in set_affinity
again.

something like:
cpumask_cpy(desc->affinity, mask);
desc->chip->set_affinity(mask);

in the failing path, affinity should not be touched.

also set_extra_move_desc() ( called by set_affinity)  will rely on the old
affinity to decide if need to move irq_desc to different node when logical
flat apic mode is used.

So try remove those assignment, and make some missed arch to assign affinity
in their set_affinity.

v2: update after "irq, x86: Remove IRQ_DISABLED check in process context IRQ move"
v3: according to Ingo, change set_affinity() in irq_chip to return int.

Signed-off-by: Yinghai Lu <yinghai@kernel.org>


---
 kernel/irq/internals.h |    3 +++
 kernel/irq/manage.c    |   17 +++++++++++------
 kernel/irq/migration.c |   14 +++++++++-----
 3 files changed, 23 insertions(+), 11 deletions(-)

Index: linux-2.6/kernel/irq/internals.h
===================================================================
--- linux-2.6.orig/kernel/irq/internals.h
+++ linux-2.6/kernel/irq/internals.h
@@ -42,6 +42,9 @@ static inline void unregister_handler_pr
 
 extern int irq_select_affinity_usr(unsigned int irq);
 
+extern void
+irq_set_thread_affinity(struct irq_desc *desc, const struct cpumask *cpumask);
+
 /*
  * Debugging printout:
  */
Index: linux-2.6/kernel/irq/manage.c
===================================================================
--- linux-2.6.orig/kernel/irq/manage.c
+++ linux-2.6/kernel/irq/manage.c
@@ -80,7 +80,7 @@ int irq_can_set_affinity(unsigned int ir
 	return 1;
 }
 
-static void
+void
 irq_set_thread_affinity(struct irq_desc *desc, const struct cpumask *cpumask)
 {
 	struct irqaction *action = desc->action;
@@ -109,17 +109,22 @@ int irq_set_affinity(unsigned int irq, c
 	spin_lock_irqsave(&desc->lock, flags);
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
-	if (desc->status & IRQ_MOVE_PCNTXT)
-		desc->chip->set_affinity(irq, cpumask);
+	if (desc->status & IRQ_MOVE_PCNTXT) {
+		if (!desc->chip->set_affinity(irq, cpumask)) {
+			cpumask_copy(desc->affinity, cpumask);
+			irq_set_thread_affinity(desc, cpumask);
+		}
+	}
 	else {
 		desc->status |= IRQ_MOVE_PENDING;
 		cpumask_copy(desc->pending_mask, cpumask);
 	}
 #else
-	cpumask_copy(desc->affinity, cpumask);
-	desc->chip->set_affinity(irq, cpumask);
+	if (!desc->chip->set_affinity(irq, cpumask)) {
+		cpumask_copy(desc->affinity, cpumask);
+		irq_set_thread_affinity(desc, cpumask);
+	}
 #endif
-	irq_set_thread_affinity(desc, cpumask);
 	desc->status |= IRQ_AFFINITY_SET;
 	spin_unlock_irqrestore(&desc->lock, flags);
 	return 0;
Index: linux-2.6/kernel/irq/migration.c
===================================================================
--- linux-2.6.orig/kernel/irq/migration.c
+++ linux-2.6/kernel/irq/migration.c
@@ -1,5 +1,8 @@
 
 #include <linux/irq.h>
+#include <linux/interrupt.h>
+
+#include "internals.h"
 
 void move_masked_irq(int irq)
 {
@@ -39,11 +42,12 @@ void move_masked_irq(int irq)
 	 * masking the irqs.
 	 */
 	if (likely(cpumask_any_and(desc->pending_mask, cpu_online_mask)
-		   < nr_cpu_ids)) {
-		cpumask_and(desc->affinity,
-			    desc->pending_mask, cpu_online_mask);
-		desc->chip->set_affinity(irq, desc->affinity);
-	}
+		   < nr_cpu_ids))
+		if (!desc->chip->set_affinity(irq, desc->pending_mask)) {
+			cpumask_copy(desc->affinity, desc->pending_mask);
+			irq_set_thread_affinity(desc, desc->pending_mask);
+		}
+
 	cpumask_clear(desc->pending_mask);
 }
 

  parent reply	other threads:[~2009-04-14 20:45 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <200903231757.53709.rusty@rustcorp.com.au>
     [not found] ` <20090323165921.GA7559@us.ibm.com>
     [not found]   ` <200903241553.58209.rusty@rustcorp.com.au>
     [not found]     ` <20090402013108.GB7103@us.ibm.com>
     [not found]       ` <20090404003520.GA8847@us.ibm.com>
     [not found]         ` <20090410215515.GC7242@us.ibm.com>
     [not found]           ` <20090411065510.GA11799@elte.hu>
     [not found]             ` <20090413220321.GA11098@us.ibm.com>
     [not found]               ` <49E4146C.7060507@kernel.org>
     [not found]                 ` <49E4162A.3060701@kernel.org>
     [not found]                   ` <20090414131711.GA4403@elte.hu>
2009-04-14 20:41                     ` [PATCH 1/4] irq: correct CPUMASKS_OFFSTACK typo -v2 Yinghai Lu
2009-04-14 20:42                       ` [PATCH 2/4] irq: make set_affinity to return status Yinghai Lu
2009-04-15  3:27                         ` Rusty Russell
2009-04-15  5:44                         ` [PATCH 2/4] irq: make set_affinity to return status -v2 Yinghai Lu
2009-04-14 20:43                       ` Yinghai Lu [this message]
2009-04-14 20:44                       ` [PATCH 4/4] irq: move move_irq_desc calling to set_affinity directly -v4 Yinghai Lu
2009-04-14 22:03                         ` Suresh Siddha
2009-04-14 20:59                       ` [PATCH 1/4] irq: correct CPUMASKS_OFFSTACK typo -v2 Andrew Morton
2009-04-15 10:01                         ` Ingo Molnar
2009-04-15 19:17                           ` Andrew Morton

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=49E4F575.7060201@kernel.org \
    --to=yinghai@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=ebiederm@xmission.com \
    --cc=garyhade@us.ibm.com \
    --cc=hpa@zytor.com \
    --cc=lcm@us.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=rusty@rustcorp.com.au \
    --cc=tglx@linutronix.de \
    --cc=venkatesh.pallipadi@intel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.