[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[Santosh Shilimkar]

Commit 'ad3b6993' converted ARM smp_cross_call() to take IPI number
as a parameter to handle more event than SGI and do_IPI was suppose
to recover SGI number. But the do_IPI doesn't consider it and it's
getting detected as 'Unknown IPI message 0x1' with ipi numbers are
moved to starts from 'IPI_TIMER=2"

There can be 16 different SGI but only SGI1 is used as IPI so
only that one is handled in do_IPI as IPI_CPU_START.

Added IPI_CPU_START because it wasn't used and thought it's
appropriate. Not sure whether its the right one.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/include/asm/hardirq.h |    2 +-
 arch/arm/kernel/smp.c          |   11 ++++++++---
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/arch/arm/include/asm/hardirq.h b/arch/arm/include/asm/hardirq.h
index 89ad180..2635c8b 100644
--- a/arch/arm/include/asm/hardirq.h
+++ b/arch/arm/include/asm/hardirq.h
@@ -5,7 +5,7 @@
 #include <linux/threads.h>
 #include <asm/irq.h>
 
-#define NR_IPI	5
+#define NR_IPI	6
 
 typedef struct {
 	unsigned int __softirq_pending;
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 4539ebc..f089e35 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -48,6 +48,7 @@
 struct secondary_data secondary_data;
 
 enum ipi_msg_type {
+	IPI_CPU_START = 1,
 	IPI_TIMER = 2,
 	IPI_RESCHEDULE,
 	IPI_CALL_FUNC,
@@ -387,7 +388,8 @@ void arch_send_call_function_single_ipi(int cpu)
 }
 
 static const char *ipi_types[NR_IPI] = {
-#define S(x,s)	[x - IPI_TIMER] = s
+#define S(x, s)	[x - IPI_CPU_START] = s
+	S(IPI_CPU_START, "CPU start interrupts"),
 	S(IPI_TIMER, "Timer broadcast interrupts"),
 	S(IPI_RESCHEDULE, "Rescheduling interrupts"),
 	S(IPI_CALL_FUNC, "Function call interrupts"),
@@ -552,10 +554,13 @@ asmlinkage void __exception_irq_entry do_IPI(int ipinr, struct pt_regs *regs)
 	unsigned int cpu = smp_processor_id();
 	struct pt_regs *old_regs = set_irq_regs(regs);
 
-	if (ipinr >= IPI_TIMER && ipinr < IPI_TIMER + NR_IPI)
-		__inc_irq_stat(cpu, ipi_irqs[ipinr - IPI_TIMER]);
+	if (ipinr >= IPI_CPU_START && ipinr < IPI_TIMER + NR_IPI)
+		__inc_irq_stat(cpu, ipi_irqs[ipinr - IPI_CPU_START]);
 
 	switch (ipinr) {
+	case IPI_CPU_START:
+		/* Wake up from WFI/WFE using SGI */
+		break;
 	case IPI_TIMER:
 		ipi_timer();
 		break;
-- 
1.6.0.4

[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[Russell King - ARM Linux]

On Mon, Jan 17, 2011 at 05:26:35PM +0530, Santosh Shilimkar wrote:
> Commit 'ad3b6993' converted ARM smp_cross_call() to take IPI number
> as a parameter to handle more event than SGI and do_IPI was suppose
> to recover SGI number. But the do_IPI doesn't consider it and it's
> getting detected as 'Unknown IPI message 0x1' with ipi numbers are
> moved to starts from 'IPI_TIMER=2"
> 
> There can be 16 different SGI but only SGI1 is used as IPI so
> only that one is handled in do_IPI as IPI_CPU_START.
> 
> Added IPI_CPU_START because it wasn't used and thought it's
> appropriate. Not sure whether its the right one.

Background:
- we start at 2 because we don't know whether platforms have hard-wired
  SGI1 for startup.
- we don't trap SGI1 at all because we don't use it.
- there's nothing stopping platforms using another SGI to wakeup - we
  don't want to be adding more of this code.

I'd much rather see platforms deciding whether they need to use SGI1
or whether they can switch to SGI0 instead - and we make SGI0 be the
"ignored wakeup" SGI.  If everyone can do that, we get rid of the
silly offset entirely.

[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[Santosh Shilimkar]

> -----Original Message-----
> From: Russell King - ARM Linux [mailto:linux at arm.linux.org.uk]
> Sent: Monday, January 17, 2011 5:31 PM
> To: Santosh Shilimkar
> Cc: linux-arm-kernel at lists.infradead.org
> Subject: Re: [PATCH] ARM: smp: Fix Unknown IPI message 0x1
>
> On Mon, Jan 17, 2011 at 05:26:35PM +0530, Santosh Shilimkar wrote:
> > Commit 'ad3b6993' converted ARM smp_cross_call() to take IPI
> number
> > as a parameter to handle more event than SGI and do_IPI was
> suppose
> > to recover SGI number. But the do_IPI doesn't consider it and it's
> > getting detected as 'Unknown IPI message 0x1' with ipi numbers are
> > moved to starts from 'IPI_TIMER=2"
> >
> > There can be 16 different SGI but only SGI1 is used as IPI so
> > only that one is handled in do_IPI as IPI_CPU_START.
> >
> > Added IPI_CPU_START because it wasn't used and thought it's
> > appropriate. Not sure whether its the right one.
>
> Background:
> - we start at 2 because we don't know whether platforms have hard-
> wired
>   SGI1 for startup.
> - we don't trap SGI1 at all because we don't use it.
> - there's nothing stopping platforms using another SGI to wakeup -
> we
>   don't want to be adding more of this code.
>
OK.

> I'd much rather see platforms deciding whether they need to use SGI1
> or whether they can switch to SGI0 instead - and we make SGI0 be the
> "ignored wakeup" SGI.  If everyone can do that, we get rid of the
> silly offset entirely.

This should be ok. May be Catalin can comment on choice of SGI1 in
generic smp_cross_call().

I found below messge in boot which made me to make this patch.
'CPU1: Unknown IPI message 0x1'

Regards,
Santosh

[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[Russell King - ARM Linux]

On Mon, Jan 17, 2011 at 05:38:11PM +0530, Santosh Shilimkar wrote:
> This should be ok. May be Catalin can comment on choice of SGI1 in
> generic smp_cross_call().

It's something that came from ARMs early SMP work (in 2002 iirc), and
was carried forward.  What I don't know is whether there was any boot
protocol documentation produced which said "you will wait for any SGI"
or "you will wait for SGI1".

[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[Catalin Marinas]

On Mon, 2011-01-17 at 12:19 +0000, Russell King - ARM Linux wrote:
> On Mon, Jan 17, 2011 at 05:38:11PM +0530, Santosh Shilimkar wrote:
> > This should be ok. May be Catalin can comment on choice of SGI1 in
> > generic smp_cross_call().
> 
> It's something that came from ARMs early SMP work (in 2002 iirc), and
> was carried forward.  What I don't know is whether there was any boot
> protocol documentation produced which said "you will wait for any SGI"
> or "you will wait for SGI1".

At least on the ARM Ltd platforms, the boot monitor waits for any
interrupt (it is in WFI). I can't tell about other platforms.

-- 
Catalin

[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[Santosh Shilimkar]

> -----Original Message-----
> From: Catalin Marinas [mailto:catalin.marinas at arm.com]
> Sent: Monday, January 17, 2011 5:57 PM
> To: Russell King - ARM Linux
> Cc: Santosh Shilimkar; linux-arm-kernel at lists.infradead.org
> Subject: Re: [PATCH] ARM: smp: Fix Unknown IPI message 0x1
>
> On Mon, 2011-01-17 at 12:19 +0000, Russell King - ARM Linux wrote:
> > On Mon, Jan 17, 2011 at 05:38:11PM +0530, Santosh Shilimkar wrote:
> > > This should be ok. May be Catalin can comment on choice of SGI1
> in
> > > generic smp_cross_call().
> >
> > It's something that came from ARMs early SMP work (in 2002 iirc),
> and
> > was carried forward.  What I don't know is whether there was any
> boot
> > protocol documentation produced which said "you will wait for any
> SGI"
> > or "you will wait for SGI1".
>
> At least on the ARM Ltd platforms, the boot monitor waits for any
> interrupt (it is in WFI). I can't tell about other platforms.
>
Same for OMAP4. Any SGI will do.

Regards,
Santosh

[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[Russell King - ARM Linux]

On Mon, Jan 17, 2011 at 06:01:31PM +0530, Santosh Shilimkar wrote:
> > -----Original Message-----
> > From: Catalin Marinas [mailto:catalin.marinas at arm.com]
> > Sent: Monday, January 17, 2011 5:57 PM
> > To: Russell King - ARM Linux
> > Cc: Santosh Shilimkar; linux-arm-kernel at lists.infradead.org
> > Subject: Re: [PATCH] ARM: smp: Fix Unknown IPI message 0x1
> >
> > At least on the ARM Ltd platforms, the boot monitor waits for any
> > interrupt (it is in WFI). I can't tell about other platforms.
> >
> Same for OMAP4. Any SGI will do.

Ok, so we can convert OMAP, Realview, Versatile Express to use SGI0...
that's a start!

[PATCH] ARM: smp: Fix Unknown IPI message 0x1

[Jeff Ohlstein]

Russell King - ARM Linux wrote:
> On Mon, Jan 17, 2011 at 06:01:31PM +0530, Santosh Shilimkar wrote:
>   
>> Same for OMAP4. Any SGI will do.
>>     
>
> Ok, so we can convert OMAP, Realview, Versatile Express to use SGI0...
> that's a start!
>
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
>   

SGI0 works fine for msm as well.

Jeff

-- 
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.