linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Yinghai Lu <yinghai@kernel.org>
To: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Suresh Siddha <suresh.b.siddha@intel.com>,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	Rusty Russell <rusty@rustcorp.com.au>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: [PATCH 4/8] irq: only update affinity in chip set_affinity() -v3
Date: Wed, 15 Apr 2009 18:41:44 -0700	[thread overview]
Message-ID: <49E68CD8.7060009@kernel.org> (raw)
In-Reply-To: <49E68C41.4020801@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:[~2009-04-16  1:43 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-16  1:39 [PATCH 1/8] x86/irq: remove NUMA_MIGRATIE_IRQ_DESC Yinghai Lu
2009-04-16  1:40 ` [PATCH 2/8] irq: correct CPUMASKS_OFFSTACK typo -v3 Yinghai Lu
2009-04-16  1:41 ` [PATCH 3/8] irq: make set_affinity to return status -v2 Yinghai Lu
2009-04-16  9:08   ` Ingo Molnar
2009-04-16  9:17     ` Joe Perches
2009-04-16  9:25       ` Ingo Molnar
2009-04-16  1:41 ` Yinghai Lu [this message]
2009-04-16 23:25   ` [PATCH 4/8] irq: only update affinity in chip set_affinity() -v3 Suresh Siddha
2009-04-17  1:58     ` Yinghai Lu
2009-04-16  1:42 ` [PATCH 5/8] irq: change irq_desc_alloc to take node instead cpu Yinghai Lu
2009-04-16  9:10   ` Ingo Molnar
2009-04-16  1:43 ` [PATCH 6/8] irq: make io_apic_set_pci_routing to take device Yinghai Lu
2009-04-16  9:16   ` Ingo Molnar
2009-04-16  1:44 ` [PATCH 7/8] x86/irq: make MSI irq_desc numa aware Yinghai Lu
2009-04-16  9:18   ` Ingo Molnar
2009-04-16  1:44 ` [PATCH 8/8] irq: make ht " Yinghai Lu
2009-04-16  9:21   ` Ingo Molnar
2009-04-16 23:34     ` Suresh Siddha
2009-04-16  9:03 ` [PATCH 1/8] x86/irq: remove NUMA_MIGRATIE_IRQ_DESC Ingo Molnar
2009-04-17  4:17   ` [PATCH 1/9] x86/irq: remove NUMA_MIGRATE_IRQ_DESC Yinghai Lu
2009-04-17  4:18     ` [PATCH 2/9] irq: correct CPUMASKS_OFFSTACK typo -v4 Yinghai Lu
2009-04-28  0:58       ` [PATCH 2/9] x86/irq: remove NUMA_MIGRATE_IRQ_DESC -v2 Yinghai Lu
2009-05-01 17:06         ` [tip:irq/numa] x86/irq: remove leftover code from NUMA_MIGRATE_IRQ_DESC tip-bot for Yinghai Lu
2009-04-17  4:19     ` [PATCH 3/9] irq: make set_affinity to return status -v2 Yinghai Lu
2009-04-28  0:59       ` Yinghai Lu
2009-05-01 17:06         ` [tip:irq/numa] irq: change ->set_affinity() to return status tip-bot for Yinghai Lu
2009-05-06  6:55         ` [tip:irq/numa] irq: change ->set_affinity() to return status, fix tip-bot for Ingo Molnar
2009-04-17  4:20     ` [PATCH 4/9] irq: only update affinity if set_affinity() sucessfully -v4 Yinghai Lu
2009-04-28  0:59       ` Yinghai Lu
2009-05-01 17:07         ` [tip:irq/numa] irq: only update affinity if ->set_affinity() is sucessfull tip-bot for Yinghai Lu
2009-04-17  4:21     ` [PATCH 5/9] x86/irq: change irq_desc_alloc to take node instead cpu Yinghai Lu
2009-04-28  1:00       ` [PATCH 5/9] x86/irq: change irq_desc_alloc to take node instead cpu -v2 Yinghai Lu
2009-05-01 17:07         ` [tip:irq/numa] x86/irq: change irq_desc_alloc() to take node instead of cpu tip-bot for Yinghai Lu
2009-04-17  4:23     ` [PATCH 6/9] irq: make io_apic_set_pci_routing to take device Yinghai Lu
2009-04-23  4:42       ` Len Brown
2009-04-26 21:53         ` [PATCH] x86/acpi: remove compress irq trick for 32bit Yinghai Lu
2009-04-26 21:55           ` [PATCH] x86/acpi: calling mp_config_acpi_gsi in mp_register_gsi Yinghai Lu
2009-04-26 23:25             ` Yinghai Lu
2009-04-27  1:12             ` [PATCH] x86/acpi: calling mp_config_acpi_gsi in mp_register_gsi -v2 Yinghai Lu
2009-04-27  3:30           ` [PATCH] x86/acpi: remove compress irq trick for 32bit Ingo Molnar
2009-05-14 15:41           ` Len Brown
2009-05-15 13:06             ` Ingo Molnar
2009-04-28  1:01       ` [PATCH 6/9] irq: make io_apic_set_pci_routing to take device Yinghai Lu
2009-05-01 17:07         ` [tip:irq/numa] irq: change ACPI GSI APIs to also take a device argument tip-bot for Yinghai Lu
2009-04-17  4:24     ` [PATCH 7/9] irq: make io_apic_set_pci_routing to use device Yinghai Lu
2009-04-28  1:01       ` Yinghai Lu
2009-05-01 17:07         ` [tip:irq/numa] irq: change io_apic_set_pci_routing() to use device parameter tip-bot for Yinghai Lu
2009-04-17  4:25     ` [PATCH 8/9] x86/irq: make MSI irq_desc numa aware Yinghai Lu
2009-04-27  3:26       ` [PATCH 8/9] x86/irq: make MSI irq_desc numa aware -v2 Yinghai Lu
2009-04-28  1:02       ` [PATCH 8/9] x86/irq: make MSI irq_desc numa aware -v3 Yinghai Lu
2009-04-28 10:14         ` Ingo Molnar
2009-04-28 14:17           ` Yinghai Lu
2009-04-28 14:32             ` Ingo Molnar
2009-04-28 19:07               ` Yinghai Lu
2009-04-29 12:00                 ` Ingo Molnar
2009-04-29 16:27                   ` [PATCH] x86/irq: using move_irq_desc Yinghai Lu
2009-04-29 20:41                     ` Ingo Molnar
2009-04-29 21:23                       ` Ingo Molnar
2009-04-29 21:55                         ` Yinghai Lu
2009-04-29 23:25                         ` Yinghai Lu
2009-04-30  7:29                           ` Ingo Molnar
2009-04-30  8:17                             ` [PATCH] x86/irq: using move_irq_desc -v3 Yinghai Lu
2009-05-01 17:08                               ` [tip:irq/numa] x86/irq: use move_irq_desc() in create_irq_nr() tip-bot for Yinghai Lu
2009-04-28 21:41               ` [PATCH] x86/irq: use move_irq_desc only when CONFIG_NUMA_IRQ_DESC is set Yinghai Lu
2009-05-01 17:07         ` [tip:irq/numa] x86/irq: change MSI irq_desc to be more numa aware tip-bot for Yinghai Lu
2009-04-17  4:25     ` [PATCH 9/9] irq: make ht irq_desc " Yinghai Lu
2009-04-28  1:02       ` Yinghai Lu
2009-04-28 10:19         ` Ingo Molnar
2009-04-28 16:28           ` Jesse Barnes
2009-04-28 16:28           ` Jesse Barnes
2009-05-01 17:07         ` [tip:irq/numa] irq: make ht irq_desc more " tip-bot for Yinghai Lu
2009-04-17 13:49     ` [PATCH 1/9] x86/irq: remove NUMA_MIGRATE_IRQ_DESC Ingo Molnar
2009-04-28  0:57     ` [PATCH 1/9] irq: correct CPUMASKS_OFFSTACK typo -v5 Yinghai Lu
2009-05-01 17:06       ` [tip:irq/numa] irq, cpumask: correct CPUMASKS_OFFSTACK typo and fix fallout tip-bot for Yinghai Lu
2009-04-17  1:00 ` [tip:irq/urgent] x86/irq: mark NUMA_MIGRATIE_IRQ_DESC broken tip-bot for Yinghai Lu
2009-04-17  1:21 ` [tip:irq/urgent] x86/irq: mark NUMA_MIGRATE_IRQ_DESC broken tip-bot for Yinghai Lu

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=49E68CD8.7060009@kernel.org \
    --to=yinghai@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=ebiederm@xmission.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=rusty@rustcorp.com.au \
    --cc=suresh.b.siddha@intel.com \
    --cc=tglx@linutronix.de \
    /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;
as well as URLs for NNTP newsgroup(s).