[ 3.5.y.z extended stable ] Patch "genirq: Always force thread affinity" has been added to staging queue

[Herton Ronaldo Krzesinski]

This is a note to let you know that I have just added a patch titled

    genirq: Always force thread affinity

to the linux-3.5.y-queue branch of the 3.5.y.z extended stable tree 
which can be found at:

 http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.5.y-queue

If you, or anyone else, feels it should not be added to this tree, please 
reply to this email.

For more information about the 3.5.y.z tree, see
https://wiki.ubuntu.com/Kernel/Dev/ExtendedStable

Thanks.
-Herton

------

>From 6591b78ff58eeb5f0e82220dd5953fcca2bd46dd Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx at linutronix.de>
Date: Sat, 3 Nov 2012 11:52:09 +0100
Subject: [PATCH] genirq: Always force thread affinity

commit 04aa530ec04f61875b99c12721162e2964e3318c upstream.

Sankara reported that the genirq core code fails to adjust the
affinity of an interrupt thread in several cases:

 1) On request/setup_irq() the call to setup_affinity() happens before
    the new action is registered, so the new thread is not notified.

 2) For secondary shared interrupts nothing notifies the new thread to
    change its affinity.

 3) Interrupts which have the IRQ_NO_BALANCE flag set are not moving
    the thread either.

Fix this by setting the thread affinity flag right on thread creation
time. This ensures that under all circumstances the thread moves to
the right place. Requires a check in irq_thread_check_affinity for an
existing affinity mask (CONFIG_CPU_MASK_OFFSTACK=y)

Reported-and-tested-by: Sankara Muthukrishnan <sankara.m at gmail.com>
Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1209041738200.2754@ionos
Signed-off-by: Thomas Gleixner <tglx at linutronix.de>
Signed-off-by: Herton Ronaldo Krzesinski <herton.krzesinski at canonical.com>
---
 kernel/irq/manage.c |   23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 5e42eb1..b4c6385 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -716,6 +716,7 @@ static void
 irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action)
 {
 	cpumask_var_t mask;
+	bool valid = true;

 	if (!test_and_clear_bit(IRQTF_AFFINITY, &action->thread_flags))
 		return;
@@ -730,10 +731,18 @@ irq_thread_check_affinity(struct irq_desc *desc, struct irqaction *action)
 	}

 	raw_spin_lock_irq(&desc->lock);
-	cpumask_copy(mask, desc->irq_data.affinity);
+	/*
+	 * This code is triggered unconditionally. Check the affinity
+	 * mask pointer. For CPU_MASK_OFFSTACK=n this is optimized out.
+	 */
+	if (desc->irq_data.affinity)
+		cpumask_copy(mask, desc->irq_data.affinity);
+	else
+		valid = false;
 	raw_spin_unlock_irq(&desc->lock);

-	set_cpus_allowed_ptr(current, mask);
+	if (valid)
+		set_cpus_allowed_ptr(current, mask);
 	free_cpumask_var(mask);
 }
 #else
@@ -936,6 +945,16 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new)
 		 */
 		get_task_struct(t);
 		new->thread = t;
+		/*
+		 * Tell the thread to set its affinity. This is
+		 * important for shared interrupt handlers as we do
+		 * not invoke setup_affinity() for the secondary
+		 * handlers as everything is already set up. Even for
+		 * interrupts marked with IRQF_NO_BALANCE this is
+		 * correct as we want the thread to move to the cpu(s)
+		 * on which the requesting code placed the interrupt.
+		 */
+		set_bit(IRQTF_AFFINITY, &new->thread_flags);
 	}

 	if (!alloc_cpumask_var(&mask, GFP_KERNEL)) {
--
1.7.9.5