public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Add SA_PERCPU_IRQ flag support
@ 2006-03-17  0:31 Dimitri Sivanich
  2006-03-17  2:09 ` Christoph Hellwig
  2006-03-17 23:26 ` Andrew Morton
  0 siblings, 2 replies; 12+ messages in thread
From: Dimitri Sivanich @ 2006-03-17  0:31 UTC (permalink / raw)
  To: Linus Torvalds
  Cc: linux-ia64, linux-kernel, Christoph Hellwig, Christoph Lameter,
	Jes Sorensen

The generic request_irq/setup_irq code should support the SA_PERCPU_IRQ flag.

Not sure why it was left out, but this patch adds that support.

Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>

Index: linux/kernel/irq/manage.c
===================================================================
--- linux.orig/kernel/irq/manage.c	2006-03-16 14:05:27.957927445 -0600
+++ linux/kernel/irq/manage.c	2006-03-16 14:06:02.283661867 -0600
@@ -201,6 +201,9 @@ int setup_irq(unsigned int irq, struct i
 	 * The following block of code has to be executed atomically
 	 */
 	spin_lock_irqsave(&desc->lock,flags);
+	if (new->flags & SA_PERCPU_IRQ) {
+		desc->status |= IRQ_PER_CPU;
+	}
 	p = &desc->action;
 	if ((old = *p) != NULL) {
 		/* Can't share interrupts unless both agree to */

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-17  0:31 Dimitri Sivanich
@ 2006-03-17  2:09 ` Christoph Hellwig
  2006-03-17 23:26 ` Andrew Morton
  1 sibling, 0 replies; 12+ messages in thread
From: Christoph Hellwig @ 2006-03-17  2:09 UTC (permalink / raw)
  To: Dimitri Sivanich
  Cc: Linus Torvalds, linux-ia64, linux-kernel, Christoph Hellwig,
	Christoph Lameter, Jes Sorensen

On Thu, Mar 16, 2006 at 06:31:14PM -0600, Dimitri Sivanich wrote:
> The generic request_irq/setup_irq code should support the SA_PERCPU_IRQ flag.
> 
> Not sure why it was left out, but this patch adds that support.
> 
> Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
> 
> Index: linux/kernel/irq/manage.c
> ===================================================================
> --- linux.orig/kernel/irq/manage.c	2006-03-16 14:05:27.957927445 -0600
> +++ linux/kernel/irq/manage.c	2006-03-16 14:06:02.283661867 -0600
> @@ -201,6 +201,9 @@ int setup_irq(unsigned int irq, struct i
>  	 * The following block of code has to be executed atomically
>  	 */
>  	spin_lock_irqsave(&desc->lock,flags);
> +	if (new->flags & SA_PERCPU_IRQ) {
> +		desc->status |= IRQ_PER_CPU;
> +	}

looks good, although the braces aren't really needed for single-line
conditionals.


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-17  0:31 Dimitri Sivanich
  2006-03-17  2:09 ` Christoph Hellwig
@ 2006-03-17 23:26 ` Andrew Morton
  2006-03-18  9:49   ` Andrew Morton
  1 sibling, 1 reply; 12+ messages in thread
From: Andrew Morton @ 2006-03-17 23:26 UTC (permalink / raw)
  To: Dimitri Sivanich; +Cc: torvalds, linux-ia64, linux-kernel, hch, clameter, jes

Dimitri Sivanich <sivanich@sgi.com> wrote:
>
> he generic request_irq/setup_irq code should support the SA_PERCPU_IRQ flag.
> 
> Not sure why it was left out, but this patch adds that support.
> 
> Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
> 
> Index: linux/kernel/irq/manage.c
> ===================================================================
> --- linux.orig/kernel/irq/manage.c	2006-03-16 14:05:27.957927445 -0600
> +++ linux/kernel/irq/manage.c	2006-03-16 14:06:02.283661867 -0600
> @@ -201,6 +201,9 @@ int setup_irq(unsigned int irq, struct i
>  	 * The following block of code has to be executed atomically
>  	 */
>  	spin_lock_irqsave(&desc->lock,flags);
> +	if (new->flags & SA_PERCPU_IRQ) {
> +		desc->status |= IRQ_PER_CPU;
> +	}
>  	p = &desc->action;
>  	if ((old = *p) != NULL) {
>  		/* Can't share interrupts unless both agree to */

Yes, we're presently going in and needlessly taking desc->lock on every
cpu-local interrupt.  However it only appears to affect mmtimer.c at
present.

Shouldn't we also have a check in there, to make sure that all the other
handlers off this IRQ agree about the per-cpuness of this interrupt?

Your patch broke the x86 build, btw..


From: Dimitri Sivanich <sivanich@sgi.com>

The generic request_irq/setup_irq code should support the SA_PERCPU_IRQ flag.

Also, check that all handlers agree to do per-cpu IRQs.

Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---

 kernel/irq/manage.c |   23 +++++++++++++++++++----
 1 files changed, 19 insertions(+), 4 deletions(-)

diff -puN kernel/irq/manage.c~generic-irq-add-sa_percpu_irq-flag-support kernel/irq/manage.c
--- 25/kernel/irq/manage.c~generic-irq-add-sa_percpu_irq-flag-support	Fri Mar 17 15:06:44 2006
+++ 25-akpm/kernel/irq/manage.c	Fri Mar 17 15:23:02 2006
@@ -204,10 +204,14 @@ int setup_irq(unsigned int irq, struct i
 	p = &desc->action;
 	if ((old = *p) != NULL) {
 		/* Can't share interrupts unless both agree to */
-		if (!(old->flags & new->flags & SA_SHIRQ)) {
-			spin_unlock_irqrestore(&desc->lock,flags);
-			return -EBUSY;
-		}
+		if (!(old->flags & new->flags & SA_SHIRQ))
+			goto mismatch;
+
+#ifdef ARCH_HAS_IRQ_PER_CPU
+		/* All handlers must agree on per-cpuness */
+		if ((old->flags & IRQ_PER_CPU) != (new->flags & IRQ_PER_CPU))
+			goto mismatch;
+#endif
 
 		/* add new interrupt at end of irq queue */
 		do {
@@ -219,6 +223,11 @@ int setup_irq(unsigned int irq, struct i
 
 	*p = new;
 
+#ifdef ARCH_HAS_IRQ_PER_CPU
+	if (new->flags & SA_PERCPU_IRQ)
+		desc->status |= IRQ_PER_CPU;
+#endif
+
 	if (!shared) {
 		desc->depth = 0;
 		desc->status &= ~(IRQ_DISABLED | IRQ_AUTODETECT |
@@ -236,6 +245,12 @@ int setup_irq(unsigned int irq, struct i
 	register_handler_proc(irq, new);
 
 	return 0;
+
+mismatch:
+	spin_unlock_irqrestore(&desc->lock, flags);
+	printk(KERN_ERR "%s: irq handler mismatch\n", __FUNCTION__);
+	dump_stack();
+	return -EBUSY;
 }
 
 #ifdef CONFIG_DEBUG_SHIRQ
_


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-17 23:26 ` Andrew Morton
@ 2006-03-18  9:49   ` Andrew Morton
  2006-03-20 14:17     ` Dimitri Sivanich
  0 siblings, 1 reply; 12+ messages in thread
From: Andrew Morton @ 2006-03-18  9:49 UTC (permalink / raw)
  To: sivanich, torvalds, linux-ia64, linux-kernel, hch, clameter, jes

Andrew Morton <akpm@osdl.org> wrote:
>
> +#ifdef ARCH_HAS_IRQ_PER_CPU
>  +	if (new->flags & SA_PERCPU_IRQ)
>  +		desc->status |= IRQ_PER_CPU;
>  +#endif

OK, five architectures define ARCH_HAS_IRQ_PER_CPU but only one of them
defines SA_PERCPU_IRQ.    Giving up.

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-18  9:49   ` Andrew Morton
@ 2006-03-20 14:17     ` Dimitri Sivanich
  2006-03-20 14:22       ` Matthew Wilcox
  0 siblings, 1 reply; 12+ messages in thread
From: Dimitri Sivanich @ 2006-03-20 14:17 UTC (permalink / raw)
  To: Andrew Morton; +Cc: torvalds, linux-ia64, linux-kernel, hch, clameter, jes

On Sat, Mar 18, 2006 at 01:49:00AM -0800, Andrew Morton wrote:
> Andrew Morton <akpm@osdl.org> wrote:
> >
> > +#ifdef ARCH_HAS_IRQ_PER_CPU
> >  +	if (new->flags & SA_PERCPU_IRQ)
> >  +		desc->status |= IRQ_PER_CPU;
> >  +#endif
> 
> OK, five architectures define ARCH_HAS_IRQ_PER_CPU but only one of them
> defines SA_PERCPU_IRQ.    Giving up.

Could we do the following (at least for now)?:

+#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
+	if (new->flags & SA_PERCPU_IRQ)
+		desc->status |= IRQ_PER_CPU;
+#endif

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-20 14:17     ` Dimitri Sivanich
@ 2006-03-20 14:22       ` Matthew Wilcox
  2006-03-20 14:25         ` Dimitri Sivanich
  0 siblings, 1 reply; 12+ messages in thread
From: Matthew Wilcox @ 2006-03-20 14:22 UTC (permalink / raw)
  To: Dimitri Sivanich
  Cc: Andrew Morton, torvalds, linux-ia64, linux-kernel, hch, clameter,
	jes

On Mon, Mar 20, 2006 at 08:17:47AM -0600, Dimitri Sivanich wrote:
> On Sat, Mar 18, 2006 at 01:49:00AM -0800, Andrew Morton wrote:
> > Andrew Morton <akpm@osdl.org> wrote:
> > >
> > > +#ifdef ARCH_HAS_IRQ_PER_CPU
> > >  +	if (new->flags & SA_PERCPU_IRQ)
> > >  +		desc->status |= IRQ_PER_CPU;
> > >  +#endif
> > 
> > OK, five architectures define ARCH_HAS_IRQ_PER_CPU but only one of them
> > defines SA_PERCPU_IRQ.    Giving up.
> 
> Could we do the following (at least for now)?:
> 
> +#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
> +	if (new->flags & SA_PERCPU_IRQ)
> +		desc->status |= IRQ_PER_CPU;
> +#endif

Why not just

#ifdef SA_PERCPU_IRQ
	if (new->flags & SA_PERCPU_IRQ)
		desc->status |= IRQ_PER_CPU;
#endif

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-20 14:22       ` Matthew Wilcox
@ 2006-03-20 14:25         ` Dimitri Sivanich
  0 siblings, 0 replies; 12+ messages in thread
From: Dimitri Sivanich @ 2006-03-20 14:25 UTC (permalink / raw)
  To: Matthew Wilcox
  Cc: Andrew Morton, torvalds, linux-ia64, linux-kernel, hch, clameter,
	jes

On Mon, Mar 20, 2006 at 07:22:07AM -0700, Matthew Wilcox wrote:
> On Mon, Mar 20, 2006 at 08:17:47AM -0600, Dimitri Sivanich wrote:
> > On Sat, Mar 18, 2006 at 01:49:00AM -0800, Andrew Morton wrote:
> > > Andrew Morton <akpm@osdl.org> wrote:
> > > >
> > > > +#ifdef ARCH_HAS_IRQ_PER_CPU
> > > >  +	if (new->flags & SA_PERCPU_IRQ)
> > > >  +		desc->status |= IRQ_PER_CPU;
> > > >  +#endif
> > > 
> > > OK, five architectures define ARCH_HAS_IRQ_PER_CPU but only one of them
> > > defines SA_PERCPU_IRQ.    Giving up.
> > 
> > Could we do the following (at least for now)?:
> > 
> > +#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
> > +	if (new->flags & SA_PERCPU_IRQ)
> > +		desc->status |= IRQ_PER_CPU;
> > +#endif
> 
> Why not just
> 
> #ifdef SA_PERCPU_IRQ
> 	if (new->flags & SA_PERCPU_IRQ)
> 		desc->status |= IRQ_PER_CPU;
> #endif

Fine with me. I was simply maintaining the ARCH_HAS_IRQ_PER_CPU convention.

^ permalink raw reply	[flat|nested] 12+ messages in thread

* [PATCH] Add SA_PERCPU_IRQ flag support
@ 2006-03-21 21:38 Dimitri Sivanich
  2006-03-21 23:37 ` Andrew Morton
  0 siblings, 1 reply; 12+ messages in thread
From: Dimitri Sivanich @ 2006-03-21 21:38 UTC (permalink / raw)
  To: Andrew Morton; +Cc: linux-kernel, hch, clameter, jes

The generic request_irq/setup_irq code should support the SA_PERCPU_IRQ flag.

This patch was posted previously, but this one should build on all arch's.

Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>

Index: linux-2.6.15/kernel/irq/manage.c
===================================================================
--- linux-2.6.15.orig/kernel/irq/manage.c	2006-03-20 13:11:01.766522017 -0600
+++ linux-2.6.15/kernel/irq/manage.c	2006-03-21 15:26:12.305876769 -0600
@@ -206,6 +206,10 @@ int setup_irq(unsigned int irq, struct i
 	 * The following block of code has to be executed atomically
 	 */
 	spin_lock_irqsave(&desc->lock,flags);
+#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
+	if (new->flags & SA_PERCPU_IRQ)
+		desc->status |= IRQ_PER_CPU;
+#endif
 	p = &desc->action;
 	if ((old = *p) != NULL) {
 		/* Can't share interrupts unless both agree to */


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-21 21:38 [PATCH] Add SA_PERCPU_IRQ flag support Dimitri Sivanich
@ 2006-03-21 23:37 ` Andrew Morton
  2006-03-21 23:45   ` Christoph Lameter
  2006-03-22  0:52   ` Dimitri Sivanich
  0 siblings, 2 replies; 12+ messages in thread
From: Andrew Morton @ 2006-03-21 23:37 UTC (permalink / raw)
  To: Dimitri Sivanich; +Cc: linux-kernel, hch, clameter, jes

Dimitri Sivanich <sivanich@sgi.com> wrote:
>
> The generic request_irq/setup_irq code should support the SA_PERCPU_IRQ flag.
> 
> This patch was posted previously, but this one should build on all arch's.
> 
> Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
> 
> Index: linux-2.6.15/kernel/irq/manage.c
> ===================================================================
> --- linux-2.6.15.orig/kernel/irq/manage.c	2006-03-20 13:11:01.766522017 -0600
> +++ linux-2.6.15/kernel/irq/manage.c	2006-03-21 15:26:12.305876769 -0600
> @@ -206,6 +206,10 @@ int setup_irq(unsigned int irq, struct i
>  	 * The following block of code has to be executed atomically
>  	 */
>  	spin_lock_irqsave(&desc->lock,flags);
> +#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
> +	if (new->flags & SA_PERCPU_IRQ)
> +		desc->status |= IRQ_PER_CPU;
> +#endif
>  	p = &desc->action;
>  	if ((old = *p) != NULL) {
>  		/* Can't share interrupts unless both agree to */

hm.  Last time around I pointed out that we should be checking that all
handlers for this IRQ agree about the percpuness.  What happened to
that?

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-21 23:37 ` Andrew Morton
@ 2006-03-21 23:45   ` Christoph Lameter
  2006-03-22  2:00     ` Benjamin Herrenschmidt
  2006-03-22  0:52   ` Dimitri Sivanich
  1 sibling, 1 reply; 12+ messages in thread
From: Christoph Lameter @ 2006-03-21 23:45 UTC (permalink / raw)
  To: Andrew Morton; +Cc: Dimitri Sivanich, linux-kernel, hch, jes

On Tue, 21 Mar 2006, Andrew Morton wrote:

> hm.  Last time around I pointed out that we should be checking that all
> handlers for this IRQ agree about the percpuness.  What happened to
> that?

Maybe simply add this patch on top?

Index: linux-2.6.16-rc6-mm2/kernel/irq/manage.c
===================================================================
--- linux-2.6.16-rc6-mm2.orig/kernel/irq/manage.c	2006-03-21 15:41:26.000000000 -0800
+++ linux-2.6.16-rc6-mm2/kernel/irq/manage.c	2006-03-21 15:43:15.000000000 -0800
@@ -204,6 +204,8 @@ int setup_irq(unsigned int irq, struct i
 #if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
 	if (new->flags & SA_PERCPU_IRQ)
 		desc->status |= IRQ_PER_CPU;
+	else
+		BUG_ON(desc->status & IRQ_PER_CPU);
 #endif
 	p = &desc->action;
 	if ((old = *p) != NULL) {

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-21 23:37 ` Andrew Morton
  2006-03-21 23:45   ` Christoph Lameter
@ 2006-03-22  0:52   ` Dimitri Sivanich
  1 sibling, 0 replies; 12+ messages in thread
From: Dimitri Sivanich @ 2006-03-22  0:52 UTC (permalink / raw)
  To: Andrew Morton; +Cc: linux-kernel, hch, clameter, jes

On Tue, Mar 21, 2006 at 03:37:47PM -0800, Andrew Morton wrote:
> hm.  Last time around I pointed out that we should be checking that all
> handlers for this IRQ agree about the percpuness.  What happened to
> that?

I believe this version has what you are looking for.

Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>

Index: linux-2.6.15/kernel/irq/manage.c
===================================================================
--- linux-2.6.15.orig/kernel/irq/manage.c	2006-03-20 13:11:01.766522017 -0600
+++ linux-2.6.15/kernel/irq/manage.c	2006-03-21 18:44:52.297990979 -0600
@@ -209,10 +209,14 @@ int setup_irq(unsigned int irq, struct i
 	p = &desc->action;
 	if ((old = *p) != NULL) {
 		/* Can't share interrupts unless both agree to */
-		if (!(old->flags & new->flags & SA_SHIRQ)) {
-			spin_unlock_irqrestore(&desc->lock,flags);
-			return -EBUSY;
-		}
+		if (!(old->flags & new->flags & SA_SHIRQ))
+			goto mismatch;
+
+#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
+		/* All handlers must agree on per-cpuness */
+		if ((old->flags & IRQ_PER_CPU) != (new->flags & IRQ_PER_CPU))
+			goto mismatch;
+#endif
 
 		/* add new interrupt at end of irq queue */
 		do {
@@ -223,7 +227,10 @@ int setup_irq(unsigned int irq, struct i
 	}
 
 	*p = new;
-
+#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
+	if (new->flags & SA_PERCPU_IRQ)
+		desc->status |= IRQ_PER_CPU;
+#endif
 	if (!shared) {
 		desc->depth = 0;
 		desc->status &= ~(IRQ_DISABLED | IRQ_AUTODETECT |
@@ -241,6 +248,12 @@ int setup_irq(unsigned int irq, struct i
 	register_handler_proc(irq, new);
 
 	return 0;
+
+mismatch:
+	spin_unlock_irqrestore(&desc->lock, flags);
+	printk(KERN_ERR "%s: irq handler mismatch\n", __FUNCTION__);
+	dump_stack();
+	return -EBUSY;
 }
 
 /*

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [PATCH] Add SA_PERCPU_IRQ flag support
  2006-03-21 23:45   ` Christoph Lameter
@ 2006-03-22  2:00     ` Benjamin Herrenschmidt
  0 siblings, 0 replies; 12+ messages in thread
From: Benjamin Herrenschmidt @ 2006-03-22  2:00 UTC (permalink / raw)
  To: Christoph Lameter; +Cc: Andrew Morton, Dimitri Sivanich, linux-kernel, hch, jes


> +	else
> +		BUG_ON(desc->status & IRQ_PER_CPU);

I see that you are a member of the school of brutalism ! :)

Ben.



^ permalink raw reply	[flat|nested] 12+ messages in thread

end of thread, other threads:[~2006-03-22  2:02 UTC | newest]

Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-03-21 21:38 [PATCH] Add SA_PERCPU_IRQ flag support Dimitri Sivanich
2006-03-21 23:37 ` Andrew Morton
2006-03-21 23:45   ` Christoph Lameter
2006-03-22  2:00     ` Benjamin Herrenschmidt
2006-03-22  0:52   ` Dimitri Sivanich
  -- strict thread matches above, loose matches on Subject: below --
2006-03-17  0:31 Dimitri Sivanich
2006-03-17  2:09 ` Christoph Hellwig
2006-03-17 23:26 ` Andrew Morton
2006-03-18  9:49   ` Andrew Morton
2006-03-20 14:17     ` Dimitri Sivanich
2006-03-20 14:22       ` Matthew Wilcox
2006-03-20 14:25         ` Dimitri Sivanich

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox