[PATCH v4] Introduce irqchip infrastructure

[PATCH v4] Introduce irqchip infrastructure

[Thomas Petazzoni]

Hello,

Here is a set of patches that introduces a small irqchip
infrastructure that allows the registration of irqchip drivers without
having each of those drivers to expose a public API and header in
include/linux/irqchip/, and moves a number of existing irqchip
controller to this infrastructure.

This new version (detailed changelog below) provides a number of
improvements: usage of a special link section to avoid the central
registration in irqchip.{c,h} of all irqchip drivers, and addition of
the GIC, VIC and Versatile FPGA IRQ drivers move, thanks to Rob
Herring and Linus Walleij.

Remaining issues to solve:

 * Russell would like arch/arm/include/asm/hardware/vic.h and
   arch/arm/include/asm/hardware/gic.h to move elsewhere, since the
   corresponding code is no longer in arch/arm/. I can move them in
   <linux/irqchip>, but that will cause a fairly large change as there
   are a big number of users of those headers in arch/arm.

 * The arch/arm/include/asm/hardware/vic.h has a few offset macros
   that should move into the .c file of the irqchip driver, but some
   of those offsets are bizarrely used in some other pieces of code in
   arch/arm/.

 * How to get rid entirely of the headers in <linux/irqchip/>. The
   remaining problematic functions are <foo>_handle_irq() and
   <foo>_irq_init() that are used by non-DT platforms. And the special
   case of gic_cascade_irq().

 * Move all the users of gic_of_init() to the irqchip mechanism
   (Exynos, i.MX 6, MSM, OMAP, SH Mobile, socfpga, spear13xx, tegra,
   ux500, vexpress) so that gic_of_init() no longer has to be
   exported.

Of course, I don't expect any of this to go in 3.8, it is 3.9 material
if we manage to reach a consensus on the remaining issues to solve
(and the other issues that will certainly show up or be raised by
other people).

This series has been built and boot tested for BCM2835 and Armada
XP. It has been built tested on Versatile with OF and without OF,
spear3xx, spear6xx and a multi_v6_v7 configuration with all possible
architectures enabled (therefore including picoxcell).

Changes since v3:

 * Switch to a link-trick based registration of the irqchip drivers,
   using the IRQCHIP_DECLARE() macro.

 * Integrate Rob Herring's patches to move the GIC and VIC irq
   controller drivers in drivers/irqchip/.

 * Migrate picoxcell, spear3xx and spear6xx to using the irqchip
   platform, so that no direct user of vic_of_init() remains.

 * Don't export vic_of_init() anymore: all users go through
   irqchip_init() now.

 * Integrate Linus Walleij patch that moves the Versatile FPGA irq
   controller driver to drivers/irqchip.

 * A set_handle_irq() ARM function was added to properly set the
   handle_arch_irq pointer to the right handler of the parent IRQ
   controller.

Changes since v2:

 * Fixed the entry in the MAINTAINERS file. Noticed by Rob Herring.

 * Simplified the Kconfig logic by making the IRQCHIP option enabled
   by default as soon as OF_IRQ is enabled. The individual ARCH_<foo>
   Kconfig options no longer have to select IRQCHIP. Also, the option
   was renamed from USE_IRQCHIP to just IRQCHIP. Suggested by Rob
   Herring.

 * Added Reviewed-by tags given by Rob Herring.

Changes since v1:

 * Add a new patch mentionning the drivers/irqchip in the list of
   directories part of the IRQ subsystem maintained by Thomas Gleixner
   in the MAINTAINERS file. Requested by Arnd Bergmann.

 * Reduce the amount of code movement in the irq-bcm2835.c and
   irq-armada-370-xp.c files by using one forward declaration for the
   IRQ handling entry point. Requested by Stephen Warren.

 * Rename the armctrl_of_init() function to bcm2835_irqchip_init() as
   requested by Stephen Warren.

 * Added the formal Acked-by and Reviewed-by received from Stephen
   Warren.

Thanks,

Thomas

[PATCH 01/16] irqchip: add basic infrastructure

[Thomas Petazzoni]

With the recent creation of the drivers/irqchip/ directory, it is
desirable to move irq controller drivers here. At the moment, the only
driver here is irq-bcm2835, the driver for the irq controller found in
the ARM BCM2835 SoC, present in Rasberry Pi systems. This irq
controller driver was exporting its initialization function and its
irq handling function through a header file in
<linux/irqchip/bcm2835.h>.

When proposing to also move another irq controller driver in
drivers/irqchip, Rob Herring raised the very valid point that moving
things to drivers/irqchip was good in order to remove more stuff from
arch/arm, but if it means adding gazillions of headers files in
include/linux/irqchip/, it would not be very nice.

So, upon the suggestion of Rob Herring and Arnd Bergmann, this commit
introduces a small infrastructure that defines a central
irqchip_init() function in drivers/irqchip/irqchip.c, which is meant
to be called as the ->init_irq() callback of ARM platforms. This
function calls of_irq_init() with an array that will progressively
contain the compatible strings of each irq controller driver, and also
a reference to the initialization functions of such drivers. The
drivers/irqchip/irqchip.h header file, currently empty, is added to
allow irq controller drivers to expose their initialization function
to the main irqchip.c file. Note that the irq controller driver
initialization function is responsible for setting the global
handle_arch_irq() variable, so that ARM platforms no longer have to
define the ->handle_irq field in their DT_MACHINE structure.

A global header, <linux/irqchip.h> is also added to expose the single
irqchip_init() function to the reset of the kernel.

A further commit moves the BCM2835 irq controller driver to this new
small infrastructure, therefore removing the include/linux/irqchip/
directory.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Stephen Warren <swarren@wwwdotorg.org>
Reviewed-by: Rob Herring <rob.herring@calxeda.com>
---
 drivers/irqchip/Kconfig           |    4 +++-
 drivers/irqchip/Makefile          |    1 +
 drivers/irqchip/irqchip.c         |   30 ++++++++++++++++++++++++++++++
 drivers/irqchip/irqchip.h         |   29 +++++++++++++++++++++++++++++
 include/asm-generic/vmlinux.lds.h |   12 +++++++++++-
 include/linux/irqchip.h           |   16 ++++++++++++++++
 6 files changed, 90 insertions(+), 2 deletions(-)
 create mode 100644 drivers/irqchip/irqchip.c
 create mode 100644 drivers/irqchip/irqchip.h
 create mode 100644 include/linux/irqchip.h

diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
index 1bb8bf6..88b0929 100644
--- a/drivers/irqchip/Kconfig
+++ b/drivers/irqchip/Kconfig
@@ -1 +1,3 @@
-# empty
+config IRQCHIP
+	def_bool y
+	depends on OF_IRQ
diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index 054321d..6b5a6e0 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -1 +1,2 @@
+obj-$(CONFIG_IRQCHIP) += irqchip.o
 obj-$(CONFIG_ARCH_BCM2835) += irq-bcm2835.o
diff --git a/drivers/irqchip/irqchip.c b/drivers/irqchip/irqchip.c
new file mode 100644
index 0000000..f496afc
--- /dev/null
+++ b/drivers/irqchip/irqchip.c
@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2012 Thomas Petazzoni
+ *
+ * Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/init.h>
+#include <linux/of_irq.h>
+
+#include "irqchip.h"
+
+/*
+ * This special of_device_id is the sentinel at the end of the
+ * of_device_id[] array of all irqchips. It is automatically placed at
+ * the end of the array by the linker, thanks to being part of a
+ * special section.
+ */
+static const struct of_device_id
+irqchip_of_match_end __used __section(__irqchip_of_end);
+
+extern struct of_device_id __irqchip_begin[];
+
+void __init irqchip_init(void)
+{
+	of_irq_init(__irqchip_begin);
+}
diff --git a/drivers/irqchip/irqchip.h b/drivers/irqchip/irqchip.h
new file mode 100644
index 0000000..e445ba2
--- /dev/null
+++ b/drivers/irqchip/irqchip.h
@@ -0,0 +1,29 @@
+/*
+ * Copyright (C) 2012 Thomas Petazzoni
+ *
+ * Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef _IRQCHIP_H
+#define _IRQCHIP_H
+
+/*
+ * This macro must be used by the different irqchip drivers to declare
+ * the association between their DT compatible string and their
+ * initialization function.
+ *
+ * @name: name that must be unique accross all IRQCHIP_DECLARE of the
+ * same file.
+ * @compstr: compatible string of the irqchip driver
+ * @fn: initialization function
+ */
+#define IRQCHIP_DECLARE(name,compstr,fn)				\
+	static const struct of_device_id irqchip_of_match_##name	\
+	__used __section(__irqchip_of_table)				\
+	= { .compatible = compstr, .data = fn }
+
+#endif
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index d1ea7ce..c80c599 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -149,6 +149,15 @@
 #define TRACE_SYSCALLS()
 #endif
 
+#ifdef CONFIG_IRQCHIP
+#define IRQCHIP_OF_MATCH_TABLE()					\
+	. = ALIGN(8);							\
+	VMLINUX_SYMBOL(__irqchip_begin) = .;				\
+	*(__irqchip_of_table)		  				\
+	*(__irqchip_of_end)
+#else
+#define IRQCHIP_OF_MATCH_TABLE()
+#endif
 
 #define KERNEL_DTB()							\
 	STRUCT_ALIGN();							\
@@ -493,7 +502,8 @@
 	DEV_DISCARD(init.rodata)					\
 	CPU_DISCARD(init.rodata)					\
 	MEM_DISCARD(init.rodata)					\
-	KERNEL_DTB()
+	KERNEL_DTB()							\
+	IRQCHIP_OF_MATCH_TABLE()
 
 #define INIT_TEXT							\
 	*(.init.text)							\
diff --git a/include/linux/irqchip.h b/include/linux/irqchip.h
new file mode 100644
index 0000000..e0006f1
--- /dev/null
+++ b/include/linux/irqchip.h
@@ -0,0 +1,16 @@
+/*
+ * Copyright (C) 2012 Thomas Petazzoni
+ *
+ * Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef _LINUX_IRQCHIP_H
+#define _LINUX_IRQCHIP_H
+
+void irqchip_init(void);
+
+#endif
-- 
1.7.9.5

[PATCH 01/16] irqchip: add basic infrastructure

[Stephen Warren]

On 11/20/2012 03:00 PM, Thomas Petazzoni wrote:
> With the recent creation of the drivers/irqchip/ directory, it is
> desirable to move irq controller drivers here. At the moment, the only
> driver here is irq-bcm2835, the driver for the irq controller found in
> the ARM BCM2835 SoC, present in Rasberry Pi systems. This irq
> controller driver was exporting its initialization function and its
> irq handling function through a header file in
> <linux/irqchip/bcm2835.h>.
...
> Reviewed-by: Stephen Warren <swarren@wwwdotorg.org>

That still stands as:

Reviewed-by: Stephen Warren <swarren@nvidia.com>

... although I think in this case the implementation changed enough it
probably shouldn't have been kept in v4.


> diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig

> +config IRQCHIP

I wonder if it should be IRQCHIP_OF, since it's OF-specific?

> +	def_bool y
> +	depends on OF_IRQ

For the drivers/clocksource patch I created, I required the
architecture/machine config option to select it rather than making it a
def_bool. Would that be better? I suppose if it's going to be selected
in a lot of places anyway, and since the code will just be dropped if it
isn't used, then making it default to on is reasonable though.

> diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h

> @@ -493,7 +502,8 @@
>  	DEV_DISCARD(init.rodata)					\
>  	CPU_DISCARD(init.rodata)					\
>  	MEM_DISCARD(init.rodata)					\
> -	KERNEL_DTB()
> +	KERNEL_DTB()							\
> +	IRQCHIP_OF_MATCH_TABLE()

Does it make sense to put that before KERNEL_DTB()? I did in my
drivers/clocksource patch just in case anything depended on KERNEL_DTB
being last along the lines of APPENDED_DTB. That said, now that I think
about it, nothing really should depend on the order...

[PATCH 01/16] irqchip: add basic infrastructure

[Thomas Petazzoni]

Stephen,

On Tue, 20 Nov 2012 15:40:48 -0700, Stephen Warren wrote:

> > Reviewed-by: Stephen Warren <swarren@wwwdotorg.org>
> 
> That still stands as:
> 
> Reviewed-by: Stephen Warren <swarren@nvidia.com>
> 
> ... although I think in this case the implementation changed enough it
> probably shouldn't have been kept in v4.

Indeed, my apologies. I was seeing some interest in this irqchip thing
in the recent days, and I wanted to show some of the progress but have
been quite busy with the mvebu development. So I was quite certainly a
bit too quick.

> > diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
> 
> > +config IRQCHIP
> 
> I wonder if it should be IRQCHIP_OF, since it's OF-specific?
> 
> > +	def_bool y
> > +	depends on OF_IRQ
> 
> For the drivers/clocksource patch I created, I required the
> architecture/machine config option to select it rather than making it a
> def_bool. Would that be better? I suppose if it's going to be selected
> in a lot of places anyway, and since the code will just be dropped if it
> isn't used, then making it default to on is reasonable though.

Well, Rob Herring suggesting this def_bool y originally, I think it
makes sense as all new DT platforms will most likely this mechanism for
their IRQ driver. But if people feel like having a per-platform
'select IRQCHIP', I'm fine as well.

> > diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
> 
> > @@ -493,7 +502,8 @@
> >  	DEV_DISCARD(init.rodata)					\
> >  	CPU_DISCARD(init.rodata)					\
> >  	MEM_DISCARD(init.rodata)					\
> > -	KERNEL_DTB()
> > +	KERNEL_DTB()							\
> > +	IRQCHIP_OF_MATCH_TABLE()
> 
> Does it make sense to put that before KERNEL_DTB()? I did in my
> drivers/clocksource patch just in case anything depended on KERNEL_DTB
> being last along the lines of APPENDED_DTB. That said, now that I think
> about it, nothing really should depend on the order...

I haven't thought about this, and I actually booted an APPENDED_DTB
kernel with this IRQCHIP_OF_MATCH_TABLE thing on Armada XP, and it
worked. I admit I haven't looked in details at what this KERNEL_DTB()
thing was.

Thanks!

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[PATCH 02/16] arm: add set_handle_irq() to register the parent IRQ controller handler function

[Thomas Petazzoni]

In order to allow irqchip drivers to register their IRQ handling
function as the parent IRQ controller handler function, we provide a
convenience function. This will avoid poking directly into the global
handle_arch_irq variable, and ensures that the user gets warned if
we're trying to register a second IRQ controller as the parent one.

Suggested by Arnd Bergmann.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 arch/arm/include/asm/mach/irq.h |    1 +
 arch/arm/kernel/irq.c           |   10 ++++++++++
 2 files changed, 11 insertions(+)

diff --git a/arch/arm/include/asm/mach/irq.h b/arch/arm/include/asm/mach/irq.h
index 15cb035..18c8830 100644
--- a/arch/arm/include/asm/mach/irq.h
+++ b/arch/arm/include/asm/mach/irq.h
@@ -22,6 +22,7 @@ extern int show_fiq_list(struct seq_file *, int);
 
 #ifdef CONFIG_MULTI_IRQ_HANDLER
 extern void (*handle_arch_irq)(struct pt_regs *);
+extern void set_handle_irq(void (*handle_irq)(struct pt_regs *));
 #endif
 
 /*
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index 8961650..f135dab 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -117,6 +117,16 @@ void __init init_IRQ(void)
 	machine_desc->init_irq();
 }
 
+#ifdef CONFIG_MULTI_IRQ_HANDLER
+void __init set_handle_irq(void (*handle_irq)(struct pt_regs *))
+{
+	if (WARN_ON(handle_arch_irq))
+		return;
+
+	handle_arch_irq = handle_irq;
+}
+#endif
+
 #ifdef CONFIG_SPARSE_IRQ
 int __init arch_probe_nr_irqs(void)
 {
-- 
1.7.9.5

[PATCH 02/16] arm: add set_handle_irq() to register the parent IRQ controller handler function

[Rob Herring]

On 11/20/2012 04:00 PM, Thomas Petazzoni wrote:
> In order to allow irqchip drivers to register their IRQ handling
> function as the parent IRQ controller handler function, we provide a
> convenience function. This will avoid poking directly into the global
> handle_arch_irq variable, and ensures that the user gets warned if
> we're trying to register a second IRQ controller as the parent one.
> 
> Suggested by Arnd Bergmann.
> 
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
> ---
>  arch/arm/include/asm/mach/irq.h |    1 +
>  arch/arm/kernel/irq.c           |   10 ++++++++++
>  2 files changed, 11 insertions(+)
> 
> diff --git a/arch/arm/include/asm/mach/irq.h b/arch/arm/include/asm/mach/irq.h
> index 15cb035..18c8830 100644
> --- a/arch/arm/include/asm/mach/irq.h
> +++ b/arch/arm/include/asm/mach/irq.h
> @@ -22,6 +22,7 @@ extern int show_fiq_list(struct seq_file *, int);
>  
>  #ifdef CONFIG_MULTI_IRQ_HANDLER
>  extern void (*handle_arch_irq)(struct pt_regs *);
> +extern void set_handle_irq(void (*handle_irq)(struct pt_regs *));
>  #endif
>  
>  /*
> diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
> index 8961650..f135dab 100644
> --- a/arch/arm/kernel/irq.c
> +++ b/arch/arm/kernel/irq.c
> @@ -117,6 +117,16 @@ void __init init_IRQ(void)
>  	machine_desc->init_irq();
>  }
>  
> +#ifdef CONFIG_MULTI_IRQ_HANDLER
> +void __init set_handle_irq(void (*handle_irq)(struct pt_regs *))
> +{
> +	if (WARN_ON(handle_arch_irq))
> +		return;

I wonder if we should just return silently without warning and rely on
the top level controller being initialized 1st. For the DT case, this is
guaranteed. In the non-DT case, we don't really know if we are the
top-level controller at least in a consistent way. AFAIK, it happens
that the 1st instance of the VIC or GIC is always the top level and we
can use that (i.e. gic_nr == 0). However, theoretically you could have a
VIC connected to a top-level GIC or even vice-versa.

Rob

> +
> +	handle_arch_irq = handle_irq;
> +}
> +#endif
> +
>  #ifdef CONFIG_SPARSE_IRQ
>  int __init arch_probe_nr_irqs(void)
>  {
> 

[PATCH 03/16] arm: bcm2835: convert to the irqchip infrastructure

[Thomas Petazzoni]

Register the irq controller driver in the main
drivers/irqchip/irqchip.c file, and make sure that the initialization
function of the driver sets handle_arch_irq() appropriately. This
requires a bit of movement in the driver since the
bcm2835_handle_irq() must move before the armctrl_of_init() function
to avoid a forward declaration.

On the arch/arm side, use irqchip_init() as the ->init_irq() callback,
and remove the definition of ->handle_irq() since this is now done by
the irq controller driver.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
Reviewed-by: Rob Herring <rob.herring@calxeda.com>
---
 arch/arm/mach-bcm2835/bcm2835.c |    5 ++---
 drivers/irqchip/irq-bcm2835.c   |   27 ++++++++++++++-------------
 include/linux/irqchip/bcm2835.h |   29 -----------------------------
 3 files changed, 16 insertions(+), 45 deletions(-)
 delete mode 100644 include/linux/irqchip/bcm2835.h

diff --git a/arch/arm/mach-bcm2835/bcm2835.c b/arch/arm/mach-bcm2835/bcm2835.c
index f6fea49..ab1bccd 100644
--- a/arch/arm/mach-bcm2835/bcm2835.c
+++ b/arch/arm/mach-bcm2835/bcm2835.c
@@ -13,10 +13,10 @@
  */
 
 #include <linux/init.h>
-#include <linux/irqchip/bcm2835.h>
 #include <linux/of_platform.h>
 #include <linux/bcm2835_timer.h>
 #include <linux/clk/bcm2835.h>
+#include <linux/irqchip.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
@@ -56,8 +56,7 @@ static const char * const bcm2835_compat[] = {
 
 DT_MACHINE_START(BCM2835, "BCM2835")
 	.map_io = bcm2835_map_io,
-	.init_irq = bcm2835_init_irq,
-	.handle_irq = bcm2835_handle_irq,
+	.init_irq = irqchip_init,
 	.init_machine = bcm2835_init,
 	.timer = &bcm2835_timer,
 	.dt_compat = bcm2835_compat
diff --git a/drivers/irqchip/irq-bcm2835.c b/drivers/irqchip/irq-bcm2835.c
index dc670cc..c2ca9b0 100644
--- a/drivers/irqchip/irq-bcm2835.c
+++ b/drivers/irqchip/irq-bcm2835.c
@@ -49,9 +49,11 @@
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 #include <linux/irqdomain.h>
-#include <linux/irqchip/bcm2835.h>
 
 #include <asm/exception.h>
+#include <asm/mach/irq.h>
+
+#include "irqchip.h"
 
 /* Put the bank and irq (32 bits) into the hwirq */
 #define MAKE_HWIRQ(b, n)	((b << 5) | (n))
@@ -135,8 +137,10 @@ static struct irq_domain_ops armctrl_ops = {
 	.xlate = armctrl_xlate
 };
 
-static int __init armctrl_of_init(struct device_node *node,
-	struct device_node *parent)
+static asmlinkage void bcm2835_handle_irq(struct pt_regs *regs);
+
+static int __init bcm2835_irqchip_init(struct device_node *node,
+				       struct device_node *parent)
 {
 	void __iomem *base;
 	int irq, b, i;
@@ -164,16 +168,10 @@ static int __init armctrl_of_init(struct device_node *node,
 			set_irq_flags(irq, IRQF_VALID | IRQF_PROBE);
 		}
 	}
-	return 0;
-}
 
-static struct of_device_id irq_of_match[] __initconst = {
-	{ .compatible = "brcm,bcm2835-armctrl-ic", .data = armctrl_of_init }
-};
+	set_handle_irq(bcm2835_handle_irq);
 
-void __init bcm2835_init_irq(void)
-{
-	of_irq_init(irq_of_match);
+	return 0;
 }
 
 /*
@@ -199,8 +197,8 @@ static void armctrl_handle_shortcut(int bank, struct pt_regs *regs,
 	handle_IRQ(irq_linear_revmap(intc.domain, irq), regs);
 }
 
-asmlinkage void __exception_irq_entry bcm2835_handle_irq(
-	struct pt_regs *regs)
+static asmlinkage void __exception_irq_entry
+bcm2835_handle_irq(struct pt_regs *regs)
 {
 	u32 stat, irq;
 
@@ -221,3 +219,6 @@ asmlinkage void __exception_irq_entry bcm2835_handle_irq(
 		}
 	}
 }
+
+IRQCHIP_DECLARE(bcm2835_armctrl_ic, "brcm,bcm2835-armctrl-ic",
+		bcm2835_irqchip_init);
diff --git a/include/linux/irqchip/bcm2835.h b/include/linux/irqchip/bcm2835.h
deleted file mode 100644
index 48a859b..0000000
--- a/include/linux/irqchip/bcm2835.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*
- * Copyright (C) 2010 Broadcom
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
- */
-
-#ifndef __LINUX_IRQCHIP_BCM2835_H_
-#define __LINUX_IRQCHIP_BCM2835_H_
-
-#include <asm/exception.h>
-
-extern void bcm2835_init_irq(void);
-
-extern asmlinkage void __exception_irq_entry bcm2835_handle_irq(
-	struct pt_regs *regs);
-
-#endif
-- 
1.7.9.5

[PATCH 04/16] arm: mvebu: move irq controller driver to drivers/irqchip

[Thomas Petazzoni]

Now that the drivers/irqchip/ directory has a minimal infrastructure
to support the addition of irq controller driver, we move the irq
controller driver for Armada 370 and Armada XP ARM SoCs from the
arch/arm/mach-mvebu/ directory to the drivers/irqchip/ directory and
update the irqchip infrastructure to take into account this new
driver.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Stephen Warren <swarren@wwwdotorg.org>
Reviewed-by: Rob Herring <rob.herring@calxeda.com>
Cc: Jason Cooper <jason@lakedaemon.net>
Cc: Andrew Lunn <andrew@lunn.ch>
Cc: Gregory Clement <gregory.clement@free-electrons.com>
---
 arch/arm/mach-mvebu/Makefile                       |    2 +-
 arch/arm/mach-mvebu/armada-370-xp.c                |    4 +--
 arch/arm/mach-mvebu/common.h                       |    3 ---
 drivers/irqchip/Makefile                           |    1 +
 .../irqchip}/irq-armada-370-xp.c                   |   28 +++++++++++---------
 5 files changed, 19 insertions(+), 19 deletions(-)
 rename {arch/arm/mach-mvebu => drivers/irqchip}/irq-armada-370-xp.c (87%)

diff --git a/arch/arm/mach-mvebu/Makefile b/arch/arm/mach-mvebu/Makefile
index 57f996b..7f4e9f4 100644
--- a/arch/arm/mach-mvebu/Makefile
+++ b/arch/arm/mach-mvebu/Makefile
@@ -2,4 +2,4 @@ ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/$(src)/include \
 	-I$(srctree)/arch/arm/plat-orion/include
 
 obj-y += system-controller.o
-obj-$(CONFIG_MACH_ARMADA_370_XP) += armada-370-xp.o irq-armada-370-xp.o addr-map.o
+obj-$(CONFIG_MACH_ARMADA_370_XP) += armada-370-xp.o addr-map.o
diff --git a/arch/arm/mach-mvebu/armada-370-xp.c b/arch/arm/mach-mvebu/armada-370-xp.c
index a2f5bbc..5e90401 100644
--- a/arch/arm/mach-mvebu/armada-370-xp.c
+++ b/arch/arm/mach-mvebu/armada-370-xp.c
@@ -16,6 +16,7 @@
 #include <linux/init.h>
 #include <linux/of_platform.h>
 #include <linux/io.h>
+#include <linux/irqchip.h>
 #include <linux/time-armada-370-xp.h>
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
@@ -54,8 +55,7 @@ static const char * const armada_370_xp_dt_compat[] = {
 DT_MACHINE_START(ARMADA_XP_DT, "Marvell Armada 370/XP (Device Tree)")
 	.init_machine	= armada_370_xp_dt_init,
 	.map_io		= armada_370_xp_map_io,
-	.init_irq	= armada_370_xp_init_irq,
-	.handle_irq     = armada_370_xp_handle_irq,
+	.init_irq	= irqchip_init,
 	.timer		= &armada_370_xp_timer,
 	.restart	= mvebu_restart,
 	.dt_compat	= armada_370_xp_dt_compat,
diff --git a/arch/arm/mach-mvebu/common.h b/arch/arm/mach-mvebu/common.h
index 02f89ea..f0eaa21 100644
--- a/arch/arm/mach-mvebu/common.h
+++ b/arch/arm/mach-mvebu/common.h
@@ -17,7 +17,4 @@
 
 void mvebu_restart(char mode, const char *cmd);
 
-void armada_370_xp_init_irq(void);
-void armada_370_xp_handle_irq(struct pt_regs *regs);
-
 #endif
diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index 6b5a6e0..5148ffd 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -1,2 +1,3 @@
 obj-$(CONFIG_IRQCHIP) += irqchip.o
 obj-$(CONFIG_ARCH_BCM2835) += irq-bcm2835.o
+obj-$(CONFIG_ARCH_MVEBU)   += irq-armada-370-xp.o
diff --git a/arch/arm/mach-mvebu/irq-armada-370-xp.c b/drivers/irqchip/irq-armada-370-xp.c
similarity index 87%
rename from arch/arm/mach-mvebu/irq-armada-370-xp.c
rename to drivers/irqchip/irq-armada-370-xp.c
index 5f5f939..09fad6c 100644
--- a/arch/arm/mach-mvebu/irq-armada-370-xp.c
+++ b/drivers/irqchip/irq-armada-370-xp.c
@@ -23,8 +23,11 @@
 #include <linux/of_irq.h>
 #include <linux/irqdomain.h>
 #include <asm/mach/arch.h>
+#include <asm/mach/irq.h>
 #include <asm/exception.h>
 
+#include "irqchip.h"
+
 /* Interrupt Controller Registers Map */
 #define ARMADA_370_XP_INT_SET_MASK_OFFS		(0x48)
 #define ARMADA_370_XP_INT_CLEAR_MASK_OFFS	(0x4C)
@@ -77,8 +80,12 @@ static struct irq_domain_ops armada_370_xp_mpic_irq_ops = {
 	.xlate = irq_domain_xlate_onecell,
 };
 
-static int __init armada_370_xp_mpic_of_init(struct device_node *node,
-					     struct device_node *parent)
+static asmlinkage void
+armada_370_xp_handle_irq(struct pt_regs *regs);
+
+static int __init
+armada_370_xp_mpic_of_init(struct device_node *node,
+			   struct device_node *parent)
 {
 	u32 control;
 
@@ -98,11 +105,14 @@ static int __init armada_370_xp_mpic_of_init(struct device_node *node,
 		panic("Unable to add Armada_370_Xp MPIC irq domain (DT)\n");
 
 	irq_set_default_host(armada_370_xp_mpic_domain);
+
+	set_handle_irq(armada_370_xp_handle_irq);
+
 	return 0;
 }
 
-asmlinkage void __exception_irq_entry armada_370_xp_handle_irq(struct pt_regs
-							       *regs)
+static asmlinkage void __exception_irq_entry
+armada_370_xp_handle_irq(struct pt_regs *regs)
 {
 	u32 irqstat, irqnr;
 
@@ -122,12 +132,4 @@ asmlinkage void __exception_irq_entry armada_370_xp_handle_irq(struct pt_regs
 	} while (1);
 }
 
-static const struct of_device_id mpic_of_match[] __initconst = {
-	{.compatible = "marvell,mpic", .data = armada_370_xp_mpic_of_init},
-	{},
-};
-
-void __init armada_370_xp_init_irq(void)
-{
-	of_irq_init(mpic_of_match);
-}
+IRQCHIP_DECLARE(mpic, "marvell,mpic", armada_370_xp_mpic_of_init);
-- 
1.7.9.5

[PATCH 05/16] irqchip: add to the directories part of the IRQ subsystem in MAINTAINERS

[Thomas Petazzoni]

Now that the drivers/irqchip/ directory is getting more code, it needs
a maintainer. The obvious maintainer for it is Thomas Gleixner, who is
maintaining the overall IRQ subsystem. So we add drivers/irqchip/ in
the list of directories that are part of the IRQ subsystem.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Rob Herring <rob.herring@calxeda.com>
---
 MAINTAINERS |    1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 59203e7..bebe2f0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4043,6 +4043,7 @@ M:	Thomas Gleixner <tglx@linutronix.de>
 S:	Maintained
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core
 F:	kernel/irq/
+F:	drivers/irqchip/
 
 IRQ DOMAINS (IRQ NUMBER MAPPING LIBRARY)
 M:	Benjamin Herrenschmidt <benh@kernel.crashing.org>
-- 
1.7.9.5

[PATCH 06/16] ARM: gic: move register definitions into .c file

[Thomas Petazzoni]

From: Rob Herring <rob.herring@calxeda.com>

There's only 1 user of gic registers outside of gic.c, so move the
register definitions into gic.c. For now, keep GIC_DIST_CNT as Tegra
uses it.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
---
 arch/arm/common/gic.c               |   21 ++++++++++++++++++++-
 arch/arm/include/asm/hardware/gic.h |   18 ------------------
 2 files changed, 20 insertions(+), 19 deletions(-)

diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index aa52699..55ea0d7 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -43,7 +43,26 @@
 #include <asm/exception.h>
 #include <asm/smp_plat.h>
 #include <asm/mach/irq.h>
-#include <asm/hardware/gic.h>
+
+#define GIC_CPU_CTRL			0x00
+#define GIC_CPU_PRIMASK			0x04
+#define GIC_CPU_BINPOINT		0x08
+#define GIC_CPU_INTACK			0x0c
+#define GIC_CPU_EOI			0x10
+#define GIC_CPU_RUNNINGPRI		0x14
+#define GIC_CPU_HIGHPRI			0x18
+
+#define GIC_DIST_CTRL			0x000
+#define GIC_DIST_CTR			0x004
+#define GIC_DIST_ENABLE_SET		0x100
+#define GIC_DIST_ENABLE_CLEAR		0x180
+#define GIC_DIST_PENDING_SET		0x200
+#define GIC_DIST_PENDING_CLEAR		0x280
+#define GIC_DIST_ACTIVE_BIT		0x300
+#define GIC_DIST_PRI			0x400
+#define GIC_DIST_TARGET			0x800
+#define GIC_DIST_CONFIG			0xc00
+#define GIC_DIST_SOFTINT		0xf00
 
 union gic_base {
 	void __iomem *common_base;
diff --git a/arch/arm/include/asm/hardware/gic.h b/arch/arm/include/asm/hardware/gic.h
index 4b1ce6c..8073dce 100644
--- a/arch/arm/include/asm/hardware/gic.h
+++ b/arch/arm/include/asm/hardware/gic.h
@@ -12,25 +12,7 @@
 
 #include <linux/compiler.h>
 
-#define GIC_CPU_CTRL			0x00
-#define GIC_CPU_PRIMASK			0x04
-#define GIC_CPU_BINPOINT		0x08
-#define GIC_CPU_INTACK			0x0c
-#define GIC_CPU_EOI			0x10
-#define GIC_CPU_RUNNINGPRI		0x14
-#define GIC_CPU_HIGHPRI			0x18
-
-#define GIC_DIST_CTRL			0x000
 #define GIC_DIST_CTR			0x004
-#define GIC_DIST_ENABLE_SET		0x100
-#define GIC_DIST_ENABLE_CLEAR		0x180
-#define GIC_DIST_PENDING_SET		0x200
-#define GIC_DIST_PENDING_CLEAR		0x280
-#define GIC_DIST_ACTIVE_BIT		0x300
-#define GIC_DIST_PRI			0x400
-#define GIC_DIST_TARGET			0x800
-#define GIC_DIST_CONFIG			0xc00
-#define GIC_DIST_SOFTINT		0xf00
 
 #ifndef __ASSEMBLY__
 #include <linux/irqdomain.h>
-- 
1.7.9.5

[PATCH 06/16] ARM: gic: move register definitions into .c file

[Rob Herring]

On 11/20/2012 04:00 PM, Thomas Petazzoni wrote:
> From: Rob Herring <rob.herring@calxeda.com>
> 
> There's only 1 user of gic registers outside of gic.c, so move the
> register definitions into gic.c. For now, keep GIC_DIST_CNT as Tegra
> uses it.
> 
> Signed-off-by: Rob Herring <rob.herring@calxeda.com>
> ---
>  arch/arm/common/gic.c               |   21 ++++++++++++++++++++-
>  arch/arm/include/asm/hardware/gic.h |   18 ------------------
>  2 files changed, 20 insertions(+), 19 deletions(-)

We need to drop this patch as Marc Zyngier pointed out to me that KVM
will need the distributor register definitions as part of the GIC
distributor emulation. It probably doesn't make sense to move only the
cpu interface registers.

Rob

> 
> diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
> index aa52699..55ea0d7 100644
> --- a/arch/arm/common/gic.c
> +++ b/arch/arm/common/gic.c
> @@ -43,7 +43,26 @@
>  #include <asm/exception.h>
>  #include <asm/smp_plat.h>
>  #include <asm/mach/irq.h>
> -#include <asm/hardware/gic.h>
> +
> +#define GIC_CPU_CTRL			0x00
> +#define GIC_CPU_PRIMASK			0x04
> +#define GIC_CPU_BINPOINT		0x08
> +#define GIC_CPU_INTACK			0x0c
> +#define GIC_CPU_EOI			0x10
> +#define GIC_CPU_RUNNINGPRI		0x14
> +#define GIC_CPU_HIGHPRI			0x18
> +
> +#define GIC_DIST_CTRL			0x000
> +#define GIC_DIST_CTR			0x004
> +#define GIC_DIST_ENABLE_SET		0x100
> +#define GIC_DIST_ENABLE_CLEAR		0x180
> +#define GIC_DIST_PENDING_SET		0x200
> +#define GIC_DIST_PENDING_CLEAR		0x280
> +#define GIC_DIST_ACTIVE_BIT		0x300
> +#define GIC_DIST_PRI			0x400
> +#define GIC_DIST_TARGET			0x800
> +#define GIC_DIST_CONFIG			0xc00
> +#define GIC_DIST_SOFTINT		0xf00
>  
>  union gic_base {
>  	void __iomem *common_base;
> diff --git a/arch/arm/include/asm/hardware/gic.h b/arch/arm/include/asm/hardware/gic.h
> index 4b1ce6c..8073dce 100644
> --- a/arch/arm/include/asm/hardware/gic.h
> +++ b/arch/arm/include/asm/hardware/gic.h
> @@ -12,25 +12,7 @@
>  
>  #include <linux/compiler.h>
>  
> -#define GIC_CPU_CTRL			0x00
> -#define GIC_CPU_PRIMASK			0x04
> -#define GIC_CPU_BINPOINT		0x08
> -#define GIC_CPU_INTACK			0x0c
> -#define GIC_CPU_EOI			0x10
> -#define GIC_CPU_RUNNINGPRI		0x14
> -#define GIC_CPU_HIGHPRI			0x18
> -
> -#define GIC_DIST_CTRL			0x000
>  #define GIC_DIST_CTR			0x004
> -#define GIC_DIST_ENABLE_SET		0x100
> -#define GIC_DIST_ENABLE_CLEAR		0x180
> -#define GIC_DIST_PENDING_SET		0x200
> -#define GIC_DIST_PENDING_CLEAR		0x280
> -#define GIC_DIST_ACTIVE_BIT		0x300
> -#define GIC_DIST_PRI			0x400
> -#define GIC_DIST_TARGET			0x800
> -#define GIC_DIST_CONFIG			0xc00
> -#define GIC_DIST_SOFTINT		0xf00
>  
>  #ifndef __ASSEMBLY__
>  #include <linux/irqdomain.h>
> 

[PATCH 07/16] ARM: gic: remove direct use of gic_raise_softirq

[Thomas Petazzoni]

From: Rob Herring <rob.herring@calxeda.com>

In preparation of moving gic code to drivers/irqchip, remove the direct
platform dependencies on gic_raise_softirq. Move the setup of
smp_cross_call into the gic code. Now that all platforms are using IPI#0
for core wakeup, create a common wakeup ipi function.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Kukjin Kim <kgene.kim@samsung.com>
Cc: Rob Herring <rob.herring@calxeda.com>
Cc: Sascha Hauer <kernel@pengutronix.de>
Cc: David Brown <davidb@codeaurora.org>
Cc: Daniel Walker <dwalker@fifo99.com>
Cc: Bryan Huntsman <bryanh@codeaurora.org>
Cc: Tony Lindgren <tony@atomide.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Magnus Damm <magnus.damm@gmail.com>
Cc: Viresh Kumar <viresh.linux@gmail.com>
Cc: Shiraz Hashim <shiraz.hashim@st.com>
Cc: Stephen Warren <swarren@wwwdotorg.org>
Cc: Srinidhi Kasagar <srinidhi.kasagar@stericsson.com>
Cc: Linus Walleij <linus.walleij@linaro.org>
---
 arch/arm/common/gic.c               |   45 +++++++++++++++++++----------------
 arch/arm/include/asm/hardware/gic.h |    1 -
 arch/arm/include/asm/smp.h          |    1 +
 arch/arm/kernel/smp.c               |    5 ++++
 arch/arm/mach-exynos/platsmp.c      |    4 +---
 arch/arm/mach-highbank/platsmp.c    |    4 +---
 arch/arm/mach-imx/platsmp.c         |    2 --
 arch/arm/mach-msm/platsmp.c         |    4 +---
 arch/arm/mach-omap2/omap-smp.c      |    4 +---
 arch/arm/mach-realview/platsmp.c    |    2 --
 arch/arm/mach-shmobile/platsmp.c    |    2 --
 arch/arm/mach-shmobile/smp-emev2.c  |    2 +-
 arch/arm/mach-spear13xx/platsmp.c   |    2 --
 arch/arm/mach-tegra/platsmp.c       |    2 --
 arch/arm/mach-ux500/platsmp.c       |    2 --
 arch/arm/mach-vexpress/ct-ca9x4.c   |    2 --
 arch/arm/mach-vexpress/platsmp.c    |    2 --
 arch/arm/plat-versatile/platsmp.c   |    2 +-
 18 files changed, 36 insertions(+), 52 deletions(-)

diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index 55ea0d7..1bee954 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -616,6 +616,27 @@ static void __init gic_pm_init(struct gic_chip_data *gic)
 }
 #endif
 
+#ifdef CONFIG_SMP
+static void gic_raise_softirq(const struct cpumask *mask, unsigned int irq)
+{
+	int cpu;
+	unsigned long map = 0;
+
+	/* Convert our logical CPU mask into a physical one. */
+	for_each_cpu(cpu, mask)
+		map |= 1 << cpu_logical_map(cpu);
+
+	/*
+	 * Ensure that stores to Normal memory are visible to the
+	 * other CPUs before issuing the IPI.
+	 */
+	dsb();
+
+	/* this always happens on GIC0 */
+	writel_relaxed(map << 16 | irq, gic_data_dist_base(&gic_data[0]) + GIC_DIST_SOFTINT);
+}
+#endif
+
 static int gic_irq_domain_map(struct irq_domain *d, unsigned int irq,
 				irq_hw_number_t hw)
 {
@@ -735,6 +756,9 @@ void __init gic_init_bases(unsigned int gic_nr, int irq_start,
 	if (WARN_ON(!gic->domain))
 		return;
 
+#ifdef CONFIG_SMP
+	set_smp_cross_call(gic_raise_softirq);
+#endif
 	gic_chip.flags |= gic_arch_extn.flags;
 	gic_dist_init(gic);
 	gic_cpu_init(gic);
@@ -748,27 +772,6 @@ void __cpuinit gic_secondary_init(unsigned int gic_nr)
 	gic_cpu_init(&gic_data[gic_nr]);
 }
 
-#ifdef CONFIG_SMP
-void gic_raise_softirq(const struct cpumask *mask, unsigned int irq)
-{
-	int cpu;
-	unsigned long map = 0;
-
-	/* Convert our logical CPU mask into a physical one. */
-	for_each_cpu(cpu, mask)
-		map |= 1 << cpu_logical_map(cpu);
-
-	/*
-	 * Ensure that stores to Normal memory are visible to the
-	 * other CPUs before issuing the IPI.
-	 */
-	dsb();
-
-	/* this always happens on GIC0 */
-	writel_relaxed(map << 16 | irq, gic_data_dist_base(&gic_data[0]) + GIC_DIST_SOFTINT);
-}
-#endif
-
 #ifdef CONFIG_OF
 static int gic_cnt __initdata = 0;
 
diff --git a/arch/arm/include/asm/hardware/gic.h b/arch/arm/include/asm/hardware/gic.h
index 8073dce..ed013df 100644
--- a/arch/arm/include/asm/hardware/gic.h
+++ b/arch/arm/include/asm/hardware/gic.h
@@ -26,7 +26,6 @@ int gic_of_init(struct device_node *node, struct device_node *parent);
 void gic_secondary_init(unsigned int);
 void gic_handle_irq(struct pt_regs *regs);
 void gic_cascade_irq(unsigned int gic_nr, unsigned int irq);
-void gic_raise_softirq(const struct cpumask *mask, unsigned int irq);
 
 static inline void gic_init(unsigned int nr, int start,
 			    void __iomem *dist , void __iomem *cpu)
diff --git a/arch/arm/include/asm/smp.h b/arch/arm/include/asm/smp.h
index 2e3be16..7933eb1 100644
--- a/arch/arm/include/asm/smp.h
+++ b/arch/arm/include/asm/smp.h
@@ -79,6 +79,7 @@ extern void cpu_die(void);
 
 extern void arch_send_call_function_single_ipi(int cpu);
 extern void arch_send_call_function_ipi_mask(const struct cpumask *mask);
+extern void arch_send_wakeup_ipi(int cpu);
 
 struct smp_operations {
 #ifdef CONFIG_SMP
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index fbc8b26..0b6a926 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -426,6 +426,11 @@ void arch_send_call_function_single_ipi(int cpu)
 	smp_cross_call(cpumask_of(cpu), IPI_CALL_FUNC_SINGLE);
 }
 
+void arch_send_wakeup_ipi(int cpu)
+{
+	smp_cross_call(cpumask_of(cpu), IPI_WAKEUP);
+}
+
 static const char *ipi_types[NR_IPI] = {
 #define S(x,s)	[x] = s
 	S(IPI_WAKEUP, "CPU wakeup interrupts"),
diff --git a/arch/arm/mach-exynos/platsmp.c b/arch/arm/mach-exynos/platsmp.c
index f93d820..41ee539 100644
--- a/arch/arm/mach-exynos/platsmp.c
+++ b/arch/arm/mach-exynos/platsmp.c
@@ -134,7 +134,7 @@ static int __cpuinit exynos_boot_secondary(unsigned int cpu, struct task_struct
 
 		__raw_writel(virt_to_phys(exynos4_secondary_startup),
 			CPU1_BOOT_REG);
-		gic_raise_softirq(cpumask_of(cpu), 0);
+		arch_send_wakeup_ipi(cpu);
 
 		if (pen_release == -1)
 			break;
@@ -175,8 +175,6 @@ static void __init exynos_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init exynos_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/mach-highbank/platsmp.c b/arch/arm/mach-highbank/platsmp.c
index fa9560e..c45fc34 100644
--- a/arch/arm/mach-highbank/platsmp.c
+++ b/arch/arm/mach-highbank/platsmp.c
@@ -32,7 +32,7 @@ static void __cpuinit highbank_secondary_init(unsigned int cpu)
 
 static int __cpuinit highbank_boot_secondary(unsigned int cpu, struct task_struct *idle)
 {
-	gic_raise_softirq(cpumask_of(cpu), 0);
+	arch_send_wakeup_ipi(cpu);
 	return 0;
 }
 
@@ -57,8 +57,6 @@ static void __init highbank_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init highbank_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/mach-imx/platsmp.c b/arch/arm/mach-imx/platsmp.c
index 2ac43e1..e7bde01 100644
--- a/arch/arm/mach-imx/platsmp.c
+++ b/arch/arm/mach-imx/platsmp.c
@@ -70,8 +70,6 @@ static void __init imx_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 void imx_smp_prepare(void)
diff --git a/arch/arm/mach-msm/platsmp.c b/arch/arm/mach-msm/platsmp.c
index 7ed69b69..c0eb0eb 100644
--- a/arch/arm/mach-msm/platsmp.c
+++ b/arch/arm/mach-msm/platsmp.c
@@ -115,7 +115,7 @@ static int __cpuinit msm_boot_secondary(unsigned int cpu, struct task_struct *id
 	 * the boot monitor to read the system wide flags register,
 	 * and branch to the address found there.
 	 */
-	gic_raise_softirq(cpumask_of(cpu), 0);
+	arch_send_wakeup_ipi(cpu);
 
 	timeout = jiffies + (1 * HZ);
 	while (time_before(jiffies, timeout)) {
@@ -153,8 +153,6 @@ static void __init msm_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-        set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init msm_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/mach-omap2/omap-smp.c b/arch/arm/mach-omap2/omap-smp.c
index 4d05fa8..e5f57fc 100644
--- a/arch/arm/mach-omap2/omap-smp.c
+++ b/arch/arm/mach-omap2/omap-smp.c
@@ -125,7 +125,7 @@ static int __cpuinit omap4_boot_secondary(unsigned int cpu, struct task_struct *
 		booted = true;
 	}
 
-	gic_raise_softirq(cpumask_of(cpu), 0);
+	arch_send_wakeup_ipi(cpu);
 
 	/*
 	 * Now the secondary core is starting up let it run its
@@ -192,8 +192,6 @@ static void __init omap4_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init omap4_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/mach-realview/platsmp.c b/arch/arm/mach-realview/platsmp.c
index 300f706..c9414d1 100644
--- a/arch/arm/mach-realview/platsmp.c
+++ b/arch/arm/mach-realview/platsmp.c
@@ -59,8 +59,6 @@ static void __init realview_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init realview_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/mach-shmobile/platsmp.c b/arch/arm/mach-shmobile/platsmp.c
index ed8d235..d393c52 100644
--- a/arch/arm/mach-shmobile/platsmp.c
+++ b/arch/arm/mach-shmobile/platsmp.c
@@ -26,6 +26,4 @@ void __init shmobile_smp_init_cpus(unsigned int ncores)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
diff --git a/arch/arm/mach-shmobile/smp-emev2.c b/arch/arm/mach-shmobile/smp-emev2.c
index f674562..e3880af 100644
--- a/arch/arm/mach-shmobile/smp-emev2.c
+++ b/arch/arm/mach-shmobile/smp-emev2.c
@@ -100,7 +100,7 @@ static int __cpuinit emev2_boot_secondary(unsigned int cpu, struct task_struct *
 	/* Tell ROM loader about our vector (in headsmp.S) */
 	emev2_set_boot_vector(__pa(shmobile_secondary_vector));
 
-	gic_raise_softirq(cpumask_of(cpu), 0);
+	arch_send_wakeup_ipi(cpu);
 	return 0;
 }
 
diff --git a/arch/arm/mach-spear13xx/platsmp.c b/arch/arm/mach-spear13xx/platsmp.c
index 2eaa3fa..27e3f69 100644
--- a/arch/arm/mach-spear13xx/platsmp.c
+++ b/arch/arm/mach-spear13xx/platsmp.c
@@ -104,8 +104,6 @@ static void __init spear13xx_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init spear13xx_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/mach-tegra/platsmp.c b/arch/arm/mach-tegra/platsmp.c
index 81cb265..076c140 100644
--- a/arch/arm/mach-tegra/platsmp.c
+++ b/arch/arm/mach-tegra/platsmp.c
@@ -160,8 +160,6 @@ static void __init tegra_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init tegra_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/mach-ux500/platsmp.c b/arch/arm/mach-ux500/platsmp.c
index 3db7782..774e527 100644
--- a/arch/arm/mach-ux500/platsmp.c
+++ b/arch/arm/mach-ux500/platsmp.c
@@ -155,8 +155,6 @@ static void __init ux500_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; i++)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init ux500_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/mach-vexpress/ct-ca9x4.c b/arch/arm/mach-vexpress/ct-ca9x4.c
index 4f471fa..3585449 100644
--- a/arch/arm/mach-vexpress/ct-ca9x4.c
+++ b/arch/arm/mach-vexpress/ct-ca9x4.c
@@ -195,8 +195,6 @@ static void __init ct_ca9x4_init_cpu_map(void)
 
 	for (i = 0; i < ncores; ++i)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init ct_ca9x4_smp_enable(unsigned int max_cpus)
diff --git a/arch/arm/mach-vexpress/platsmp.c b/arch/arm/mach-vexpress/platsmp.c
index 7db27c8..cd98c04 100644
--- a/arch/arm/mach-vexpress/platsmp.c
+++ b/arch/arm/mach-vexpress/platsmp.c
@@ -127,8 +127,6 @@ static void __init vexpress_dt_smp_init_cpus(void)
 
 	for (i = 0; i < ncores; ++i)
 		set_cpu_possible(i, true);
-
-	set_smp_cross_call(gic_raise_softirq);
 }
 
 static void __init vexpress_dt_smp_prepare_cpus(unsigned int max_cpus)
diff --git a/arch/arm/plat-versatile/platsmp.c b/arch/arm/plat-versatile/platsmp.c
index 04ca493..dcca151 100644
--- a/arch/arm/plat-versatile/platsmp.c
+++ b/arch/arm/plat-versatile/platsmp.c
@@ -79,7 +79,7 @@ int __cpuinit versatile_boot_secondary(unsigned int cpu, struct task_struct *idl
 	 * the boot monitor to read the system wide flags register,
 	 * and branch to the address found there.
 	 */
-	gic_raise_softirq(cpumask_of(cpu), 0);
+	arch_send_wakeup_ipi(cpu);
 
 	timeout = jiffies + (1 * HZ);
 	while (time_before(jiffies, timeout)) {
-- 
1.7.9.5

[PATCH 08/16] irqchip: Move ARM GIC to drivers/irqchip

[Thomas Petazzoni]

From: Rob Herring <rob.herring@calxeda.com>

Now that we have drivers/irqchip, move GIC irqchip to drivers/irqchip. This
is necessary to share the GIC with arm and arm64.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
---
 arch/arm/common/Kconfig                            |    8 --------
 arch/arm/common/Makefile                           |    1 -
 drivers/irqchip/Kconfig                            |    8 ++++++++
 drivers/irqchip/Makefile                           |    1 +
 arch/arm/common/gic.c => drivers/irqchip/irq-gic.c |    8 ++++++++
 5 files changed, 17 insertions(+), 9 deletions(-)
 rename arch/arm/common/gic.c => drivers/irqchip/irq-gic.c (99%)

diff --git a/arch/arm/common/Kconfig b/arch/arm/common/Kconfig
index 45ceeb0..7bf52b2 100644
--- a/arch/arm/common/Kconfig
+++ b/arch/arm/common/Kconfig
@@ -1,11 +1,3 @@
-config ARM_GIC
-	bool
-	select IRQ_DOMAIN
-	select MULTI_IRQ_HANDLER
-
-config GIC_NON_BANKED
-	bool
-
 config ARM_VIC
 	bool
 	select IRQ_DOMAIN
diff --git a/arch/arm/common/Makefile b/arch/arm/common/Makefile
index e8a4e58..4104b82 100644
--- a/arch/arm/common/Makefile
+++ b/arch/arm/common/Makefile
@@ -2,7 +2,6 @@
 # Makefile for the linux kernel.
 #
 
-obj-$(CONFIG_ARM_GIC)		+= gic.o
 obj-$(CONFIG_ARM_VIC)		+= vic.o
 obj-$(CONFIG_ICST)		+= icst.o
 obj-$(CONFIG_SA1111)		+= sa1111.o
diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
index 88b0929..2d7f350 100644
--- a/drivers/irqchip/Kconfig
+++ b/drivers/irqchip/Kconfig
@@ -1,3 +1,11 @@
 config IRQCHIP
 	def_bool y
 	depends on OF_IRQ
+
+config ARM_GIC
+	bool
+	select IRQ_DOMAIN
+	select MULTI_IRQ_HANDLER
+
+config GIC_NON_BANKED
+	bool
diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index 5148ffd..94118db 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -1,3 +1,4 @@
 obj-$(CONFIG_IRQCHIP) += irqchip.o
 obj-$(CONFIG_ARCH_BCM2835) += irq-bcm2835.o
 obj-$(CONFIG_ARCH_MVEBU)   += irq-armada-370-xp.o
+obj-$(CONFIG_ARM_GIC)   += irq-gic.o
diff --git a/arch/arm/common/gic.c b/drivers/irqchip/irq-gic.c
similarity index 99%
rename from arch/arm/common/gic.c
rename to drivers/irqchip/irq-gic.c
index 1bee954..1038e99 100644
--- a/arch/arm/common/gic.c
+++ b/drivers/irqchip/irq-gic.c
@@ -44,6 +44,8 @@
 #include <asm/smp_plat.h>
 #include <asm/mach/irq.h>
 
+#include "irqchip.h"
+
 #define GIC_CPU_CTRL			0x00
 #define GIC_CPU_PRIMASK			0x04
 #define GIC_CPU_BINPOINT		0x08
@@ -799,8 +801,14 @@ int __init gic_of_init(struct device_node *node, struct device_node *parent)
 	if (parent) {
 		irq = irq_of_parse_and_map(node, 0);
 		gic_cascade_irq(gic_cnt, irq);
+	} else {
+		set_handle_irq(gic_handle_irq);
 	}
+
 	gic_cnt++;
 	return 0;
 }
 #endif
+
+IRQCHIP_DECLARE(cortex_a15_gic, "arm,cortex-a15-gic", gic_of_init);
+IRQCHIP_DECLARE(cortex_a9_gic, "arm,cortex-a9-gic", gic_of_init);
-- 
1.7.9.5

[PATCH 09/16] irqchip: Move ARM VIC to drivers/irqchip

[Thomas Petazzoni]

From: Rob Herring <rob.herring@calxeda.com>

Now that we have drivers/irqchip, move VIC irqchip to drivers/irqchip.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
---
 arch/arm/common/Kconfig                            |   15 ----
 arch/arm/common/Makefile                           |    1 -
 drivers/irqchip/Kconfig                            |   15 ++++
 drivers/irqchip/Makefile                           |    1 +
 arch/arm/common/vic.c => drivers/irqchip/irq-vic.c |   75 +++++++++++---------
 5 files changed, 58 insertions(+), 49 deletions(-)
 rename arch/arm/common/vic.c => drivers/irqchip/irq-vic.c (98%)

diff --git a/arch/arm/common/Kconfig b/arch/arm/common/Kconfig
index 7bf52b2..9353184 100644
--- a/arch/arm/common/Kconfig
+++ b/arch/arm/common/Kconfig
@@ -1,18 +1,3 @@
-config ARM_VIC
-	bool
-	select IRQ_DOMAIN
-	select MULTI_IRQ_HANDLER
-
-config ARM_VIC_NR
-	int
-	default 4 if ARCH_S5PV210
-	default 3 if ARCH_S5PC100
-	default 2
-	depends on ARM_VIC
-	help
-	  The maximum number of VICs available in the system, for
-	  power management.
-
 config ICST
 	bool
 
diff --git a/arch/arm/common/Makefile b/arch/arm/common/Makefile
index 4104b82..dc8dd0d 100644
--- a/arch/arm/common/Makefile
+++ b/arch/arm/common/Makefile
@@ -2,7 +2,6 @@
 # Makefile for the linux kernel.
 #
 
-obj-$(CONFIG_ARM_VIC)		+= vic.o
 obj-$(CONFIG_ICST)		+= icst.o
 obj-$(CONFIG_SA1111)		+= sa1111.o
 obj-$(CONFIG_PCI_HOST_VIA82C505) += via82c505.o
diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
index 2d7f350..d80e5c7 100644
--- a/drivers/irqchip/Kconfig
+++ b/drivers/irqchip/Kconfig
@@ -9,3 +9,18 @@ config ARM_GIC
 
 config GIC_NON_BANKED
 	bool
+
+config ARM_VIC
+	bool
+	select IRQ_DOMAIN
+	select MULTI_IRQ_HANDLER
+
+config ARM_VIC_NR
+	int
+	default 4 if ARCH_S5PV210
+	default 3 if ARCH_S5PC100
+	default 2
+	depends on ARM_VIC
+	help
+	  The maximum number of VICs available in the system, for
+	  power management.
diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index 94118db..9c7f91e 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -2,3 +2,4 @@ obj-$(CONFIG_IRQCHIP) += irqchip.o
 obj-$(CONFIG_ARCH_BCM2835) += irq-bcm2835.o
 obj-$(CONFIG_ARCH_MVEBU)   += irq-armada-370-xp.o
 obj-$(CONFIG_ARM_GIC)   += irq-gic.o
+obj-$(CONFIG_ARM_VIC)   += irq-vic.o
diff --git a/arch/arm/common/vic.c b/drivers/irqchip/irq-vic.c
similarity index 98%
rename from arch/arm/common/vic.c
rename to drivers/irqchip/irq-vic.c
index e0d5388..98a0708 100644
--- a/arch/arm/common/vic.c
+++ b/drivers/irqchip/irq-vic.c
@@ -35,6 +35,8 @@
 #include <asm/mach/irq.h>
 #include <asm/hardware/vic.h>
 
+#include "irqchip.h"
+
 /**
  * struct vic_device - VIC PM device
  * @irq: The IRQ number for the base of the VIC.
@@ -390,6 +392,40 @@ void __init __vic_init(void __iomem *base, unsigned int irq_start,
 	vic_register(base, irq_start, vic_sources, resume_sources, node);
 }
 
+/*
+ * Handle each interrupt in a single VIC.  Returns non-zero if we've
+ * handled at least one interrupt.  This reads the status register
+ * before handling each interrupt, which is necessary given that
+ * handle_IRQ may briefly re-enable interrupts for soft IRQ handling.
+ */
+static int handle_one_vic(struct vic_device *vic, struct pt_regs *regs)
+{
+	u32 stat, irq;
+	int handled = 0;
+
+	while ((stat = readl_relaxed(vic->base + VIC_IRQ_STATUS))) {
+		irq = ffs(stat) - 1;
+		handle_IRQ(irq_find_mapping(vic->domain, irq), regs);
+		handled = 1;
+	}
+
+	return handled;
+}
+
+/*
+ * Keep iterating over all registered VIC's until there are no pending
+ * interrupts.
+ */
+asmlinkage void __exception_irq_entry vic_handle_irq(struct pt_regs *regs)
+{
+	int i, handled;
+
+	do {
+		for (i = 0, handled = 0; i < vic_id; ++i)
+			handled |= handle_one_vic(&vic_devices[i], regs);
+	} while (handled);
+}
+
 /**
  * vic_init() - initialise a vectored interrupt controller
  * @base: iomem base address
@@ -422,6 +458,8 @@ int __init vic_of_init(struct device_node *node, struct device_node *parent)
 
 	__vic_init(regs, irq_base, ~0, ~0, node);
 
+	set_handle_irq(vic_handle_irq);
+
 	return 0;
 
  out_unmap:
@@ -429,38 +467,9 @@ int __init vic_of_init(struct device_node *node, struct device_node *parent)
 
 	return -EIO;
 }
-#endif /* CONFIG OF */
-
-/*
- * Handle each interrupt in a single VIC.  Returns non-zero if we've
- * handled@least one interrupt.  This reads the status register
- * before handling each interrupt, which is necessary given that
- * handle_IRQ may briefly re-enable interrupts for soft IRQ handling.
- */
-static int handle_one_vic(struct vic_device *vic, struct pt_regs *regs)
-{
-	u32 stat, irq;
-	int handled = 0;
 
-	while ((stat = readl_relaxed(vic->base + VIC_IRQ_STATUS))) {
-		irq = ffs(stat) - 1;
-		handle_IRQ(irq_find_mapping(vic->domain, irq), regs);
-		handled = 1;
-	}
+IRQCHIP_DECLARE(versatile_vic, "arm,versatile-vic", vic_of_init);
+IRQCHIP_DECLARE(pl190_vic, "arm,pl190-vic", vic_of_init);
+IRQCHIP_DECLARE(pl192_vic, "arm,pl192-vic", vic_of_init);
 
-	return handled;
-}
-
-/*
- * Keep iterating over all registered VIC's until there are no pending
- * interrupts.
- */
-asmlinkage void __exception_irq_entry vic_handle_irq(struct pt_regs *regs)
-{
-	int i, handled;
-
-	do {
-		for (i = 0, handled = 0; i < vic_id; ++i)
-			handled |= handle_one_vic(&vic_devices[i], regs);
-	} while (handled);
-}
+#endif /* CONFIG OF */
-- 
1.7.9.5

[PATCH 09/16] irqchip: Move ARM VIC to drivers/irqchip

[Jamie Iles]

Hi Thomas,

On Tue, Nov 20, 2012 at 11:01:00PM +0100, Thomas Petazzoni wrote:
> +/*
> + * Keep iterating over all registered VIC's until there are no pending
> + * interrupts.
> + */
> +asmlinkage void __exception_irq_entry vic_handle_irq(struct pt_regs *regs)
> +{
> +	int i, handled;
> +
> +	do {
> +		for (i = 0, handled = 0; i < vic_id; ++i)
> +			handled |= handle_one_vic(&vic_devices[i], regs);
> +	} while (handled);
> +}

__exception_irq_entry is only defined for ARM, so I think we either need 
to add a Kconfig dependency on ARM or allow some other way to build on 
non-ARM platforms.

Jamie

[PATCH 10/16] ARM: highbank: use common irqchip_init

[Thomas Petazzoni]

From: Rob Herring <rob.herring@calxeda.com>

Now that we have common irqchip init, use it on highbank platform.

[Thomas Petazzoni: remove .handle_irq from DT_MACHINE_STRUCT, as it is
set by the the irq-gic driver].

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 arch/arm/mach-highbank/highbank.c |    9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/arch/arm/mach-highbank/highbank.c b/arch/arm/mach-highbank/highbank.c
index 40e36a5..45314ea 100644
--- a/arch/arm/mach-highbank/highbank.c
+++ b/arch/arm/mach-highbank/highbank.c
@@ -18,6 +18,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/io.h>
 #include <linux/irq.h>
+#include <linux/irqchip.h>
 #include <linux/irqdomain.h>
 #include <linux/of.h>
 #include <linux/of_irq.h>
@@ -82,11 +83,6 @@ void highbank_set_cpu_jump(int cpu, void *jump_addr)
 			  HB_JUMP_TABLE_PHYS(cpu) + 15);
 }
 
-const static struct of_device_id irq_match[] = {
-	{ .compatible = "arm,cortex-a9-gic", .data = gic_of_init, },
-	{}
-};
-
 #ifdef CONFIG_CACHE_L2X0
 static void highbank_l2x0_disable(void)
 {
@@ -97,7 +93,7 @@ static void highbank_l2x0_disable(void)
 
 static void __init highbank_init_irq(void)
 {
-	of_irq_init(irq_match);
+	irqchip_init();
 
 #ifdef CONFIG_CACHE_L2X0
 	/* Enable PL310 L2 Cache controller */
@@ -219,7 +215,6 @@ DT_MACHINE_START(HIGHBANK, "Highbank")
 	.map_io		= highbank_map_io,
 	.init_irq	= highbank_init_irq,
 	.timer		= &highbank_timer,
-	.handle_irq	= gic_handle_irq,
 	.init_machine	= highbank_init,
 	.dt_compat	= highbank_match,
 	.restart	= highbank_restart,
-- 
1.7.9.5

[PATCH 11/16] ARM: picoxcell: use common irqchip_init

[Thomas Petazzoni]

Now that we have a common irqchip infrastructure, use it on the
picoxcell platform:

 * Use irqchip_init() as the ->irq_init() machine function. The
   irqchip infrastructure already knows about the compatible strings
   for the VIC IRQ controller driver.

 * No longer needed to set ->handle_irq() in the machine structure, as
   it will be set by the VIC IRQ controller driver.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 arch/arm/mach-picoxcell/common.c |   14 ++------------
 1 file changed, 2 insertions(+), 12 deletions(-)

diff --git a/arch/arm/mach-picoxcell/common.c b/arch/arm/mach-picoxcell/common.c
index f6c0849..bf9f074 100644
--- a/arch/arm/mach-picoxcell/common.c
+++ b/arch/arm/mach-picoxcell/common.c
@@ -9,6 +9,7 @@
  */
 #include <linux/delay.h>
 #include <linux/irq.h>
+#include <linux/irqchip.h>
 #include <linux/irqdomain.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
@@ -70,16 +71,6 @@ static const char *picoxcell_dt_match[] = {
 	NULL
 };
 
-static const struct of_device_id vic_of_match[] __initconst = {
-	{ .compatible = "arm,pl192-vic", .data = vic_of_init, },
-	{ /* Sentinel */ }
-};
-
-static void __init picoxcell_init_irq(void)
-{
-	of_irq_init(vic_of_match);
-}
-
 static void picoxcell_wdt_restart(char mode, const char *cmd)
 {
 	/*
@@ -97,8 +88,7 @@ static void picoxcell_wdt_restart(char mode, const char *cmd)
 DT_MACHINE_START(PICOXCELL, "Picochip picoXcell")
 	.map_io		= picoxcell_map_io,
 	.nr_irqs	= NR_IRQS_LEGACY,
-	.init_irq	= picoxcell_init_irq,
-	.handle_irq	= vic_handle_irq,
+	.init_irq	= irqchip_init,
 	.timer		= &dw_apb_timer,
 	.init_machine	= picoxcell_init_machine,
 	.dt_compat	= picoxcell_dt_match,
-- 
1.7.9.5

[PATCH 11/16] ARM: picoxcell: use common irqchip_init

[Jamie Iles]

On Tue, Nov 20, 2012 at 11:01:02PM +0100, Thomas Petazzoni wrote:
> Now that we have a common irqchip infrastructure, use it on the
> picoxcell platform:
> 
>  * Use irqchip_init() as the ->irq_init() machine function. The
>    irqchip infrastructure already knows about the compatible strings
>    for the VIC IRQ controller driver.
> 
>  * No longer needed to set ->handle_irq() in the machine structure, as
>    it will be set by the VIC IRQ controller driver.
> 
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

Acked-by: Jamie Iles <jamie@jamieiles.com>

Thanks,

Jamie

[PATCH 12/16] ARM: spear3xx: use common irqchip_init

[Thomas Petazzoni]

Now that we have a common irqchip infrastructure, use it on the
spear3xx platform:

 * Use irqchip_init() as the ->irq_init() machine function. The
   irqchip infrastructure already knows about the compatible strings
   for the VIC IRQ controller driver.

 * No longer needed to set ->handle_irq() in the machine structure, as
   it will be set by the VIC IRQ controller driver.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 arch/arm/mach-spear3xx/include/mach/generic.h |    1 -
 arch/arm/mach-spear3xx/spear300.c             |    5 ++---
 arch/arm/mach-spear3xx/spear310.c             |    5 ++---
 arch/arm/mach-spear3xx/spear320.c             |    5 ++---
 arch/arm/mach-spear3xx/spear3xx.c             |   10 ----------
 5 files changed, 6 insertions(+), 20 deletions(-)

diff --git a/arch/arm/mach-spear3xx/include/mach/generic.h b/arch/arm/mach-spear3xx/include/mach/generic.h
index ce19113..c26b7a4 100644
--- a/arch/arm/mach-spear3xx/include/mach/generic.h
+++ b/arch/arm/mach-spear3xx/include/mach/generic.h
@@ -30,7 +30,6 @@ extern struct pl08x_platform_data pl080_plat_data;
 void __init spear_setup_of_timer(void);
 void __init spear3xx_clk_init(void);
 void __init spear3xx_map_io(void);
-void __init spear3xx_dt_init_irq(void);
 
 void spear_restart(char, const char *);
 
diff --git a/arch/arm/mach-spear3xx/spear300.c b/arch/arm/mach-spear3xx/spear300.c
index 6ec3005..a56922e 100644
--- a/arch/arm/mach-spear3xx/spear300.c
+++ b/arch/arm/mach-spear3xx/spear300.c
@@ -15,7 +15,7 @@
 
 #include <linux/amba/pl08x.h>
 #include <linux/of_platform.h>
-#include <asm/hardware/vic.h>
+#include <linux/irqchip.h>
 #include <asm/mach/arch.h>
 #include <plat/shirq.h>
 #include <mach/generic.h>
@@ -315,8 +315,7 @@ static void __init spear300_map_io(void)
 
 DT_MACHINE_START(SPEAR300_DT, "ST SPEAr300 SoC with Flattened Device Tree")
 	.map_io		=	spear300_map_io,
-	.init_irq	=	spear3xx_dt_init_irq,
-	.handle_irq	=	vic_handle_irq,
+	.init_irq	=	irqchip_init,
 	.timer		=	&spear3xx_timer,
 	.init_machine	=	spear300_dt_init,
 	.restart	=	spear_restart,
diff --git a/arch/arm/mach-spear3xx/spear310.c b/arch/arm/mach-spear3xx/spear310.c
index 1d0e435..3e9cb6c 100644
--- a/arch/arm/mach-spear3xx/spear310.c
+++ b/arch/arm/mach-spear3xx/spear310.c
@@ -15,8 +15,8 @@
 
 #include <linux/amba/pl08x.h>
 #include <linux/amba/serial.h>
+#include <linux/irqchip.h>
 #include <linux/of_platform.h>
-#include <asm/hardware/vic.h>
 #include <asm/mach/arch.h>
 #include <plat/shirq.h>
 #include <mach/generic.h>
@@ -456,8 +456,7 @@ static void __init spear310_map_io(void)
 
 DT_MACHINE_START(SPEAR310_DT, "ST SPEAr310 SoC with Flattened Device Tree")
 	.map_io		=	spear310_map_io,
-	.init_irq	=	spear3xx_dt_init_irq,
-	.handle_irq	=	vic_handle_irq,
+	.init_irq	=	irqchip_init,
 	.timer		=	&spear3xx_timer,
 	.init_machine	=	spear310_dt_init,
 	.restart	=	spear_restart,
diff --git a/arch/arm/mach-spear3xx/spear320.c b/arch/arm/mach-spear3xx/spear320.c
index fd823c6..675aa9c 100644
--- a/arch/arm/mach-spear3xx/spear320.c
+++ b/arch/arm/mach-spear3xx/spear320.c
@@ -13,11 +13,11 @@
 
 #define pr_fmt(fmt) "SPEAr320: " fmt
 
+#include <linux/irqchip.h>
 #include <linux/amba/pl022.h>
 #include <linux/amba/pl08x.h>
 #include <linux/amba/serial.h>
 #include <linux/of_platform.h>
-#include <asm/hardware/vic.h>
 #include <asm/mach/arch.h>
 #include <plat/shirq.h>
 #include <mach/generic.h>
@@ -471,8 +471,7 @@ static void __init spear320_map_io(void)
 
 DT_MACHINE_START(SPEAR320_DT, "ST SPEAr320 SoC with Flattened Device Tree")
 	.map_io		=	spear320_map_io,
-	.init_irq	=	spear3xx_dt_init_irq,
-	.handle_irq	=	vic_handle_irq,
+	.init_irq	=	irqchip_init,
 	.timer		=	&spear3xx_timer,
 	.init_machine	=	spear320_dt_init,
 	.restart	=	spear_restart,
diff --git a/arch/arm/mach-spear3xx/spear3xx.c b/arch/arm/mach-spear3xx/spear3xx.c
index 98144ba..4daecef 100644
--- a/arch/arm/mach-spear3xx/spear3xx.c
+++ b/arch/arm/mach-spear3xx/spear3xx.c
@@ -118,13 +118,3 @@ static void __init spear3xx_timer_init(void)
 struct sys_timer spear3xx_timer = {
 	.init = spear3xx_timer_init,
 };
-
-static const struct of_device_id vic_of_match[] __initconst = {
-	{ .compatible = "arm,pl190-vic", .data = vic_of_init, },
-	{ /* Sentinel */ }
-};
-
-void __init spear3xx_dt_init_irq(void)
-{
-	of_irq_init(vic_of_match);
-}
-- 
1.7.9.5

[PATCH 13/16] ARM: spear6xx: use common irqchip_init

[Thomas Petazzoni]

Now that we have a common irqchip infrastructure, use it on the
spear6xx platform:

 * Use irqchip_init() as the ->irq_init() machine function. The
   irqchip infrastructure already knows about the compatible strings
   for the VIC IRQ controller driver.

 * No longer needed to set ->handle_irq() in the machine structure, as
   it will be set by the VIC IRQ controller driver.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 arch/arm/mach-spear6xx/spear6xx.c |   14 ++------------
 1 file changed, 2 insertions(+), 12 deletions(-)

diff --git a/arch/arm/mach-spear6xx/spear6xx.c b/arch/arm/mach-spear6xx/spear6xx.c
index 5a5a52d..1892b64 100644
--- a/arch/arm/mach-spear6xx/spear6xx.c
+++ b/arch/arm/mach-spear6xx/spear6xx.c
@@ -16,6 +16,7 @@
 #include <linux/amba/pl08x.h>
 #include <linux/clk.h>
 #include <linux/err.h>
+#include <linux/irqchip.h>
 #include <linux/of.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
@@ -425,20 +426,9 @@ static const char *spear600_dt_board_compat[] = {
 	NULL
 };
 
-static const struct of_device_id vic_of_match[] __initconst = {
-	{ .compatible = "arm,pl190-vic", .data = vic_of_init, },
-	{ /* Sentinel */ }
-};
-
-static void __init spear6xx_dt_init_irq(void)
-{
-	of_irq_init(vic_of_match);
-}
-
 DT_MACHINE_START(SPEAR600_DT, "ST SPEAr600 (Flattened Device Tree)")
 	.map_io		=	spear6xx_map_io,
-	.init_irq	=	spear6xx_dt_init_irq,
-	.handle_irq	=	vic_handle_irq,
+	.init_irq	=	irqchip_init,
 	.timer		=	&spear6xx_timer,
 	.init_machine	=	spear600_dt_init,
 	.restart	=	spear_restart,
-- 
1.7.9.5

[PATCH 14/16] arm: vic: don't expose vic_of_init() anymore

[Thomas Petazzoni]

All users of vic_of_init() have been converted to the irqchip
infrastructure, and all future users should go through this new
infrastructure. Therefore, we no longer expose the vic_of_init()
function through the vic.h header.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 arch/arm/include/asm/hardware/vic.h |    1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/arm/include/asm/hardware/vic.h b/arch/arm/include/asm/hardware/vic.h
index e14af1a..3b2e881 100644
--- a/arch/arm/include/asm/hardware/vic.h
+++ b/arch/arm/include/asm/hardware/vic.h
@@ -50,7 +50,6 @@ struct pt_regs;
 void __vic_init(void __iomem *base, unsigned int irq_start, u32 vic_sources,
 		u32 resume_sources, struct device_node *node);
 void vic_init(void __iomem *base, unsigned int irq_start, u32 vic_sources, u32 resume_sources);
-int vic_of_init(struct device_node *node, struct device_node *parent);
 void vic_handle_irq(struct pt_regs *regs);
 
 #endif /* __ASSEMBLY__ */
-- 
1.7.9.5

[PATCH 15/16] irqchip: vic: mark vic_of_init() as static

[Thomas Petazzoni]

With all vic_of_init() users converted to the irqchip infrastructure,
there is no longer a reason for this function to be
exported. Therefore, mark it as static.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
 drivers/irqchip/irq-vic.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/irqchip/irq-vic.c b/drivers/irqchip/irq-vic.c
index 98a0708..c8c8cab 100644
--- a/drivers/irqchip/irq-vic.c
+++ b/drivers/irqchip/irq-vic.c
@@ -440,7 +440,8 @@ void __init vic_init(void __iomem *base, unsigned int irq_start,
 }
 
 #ifdef CONFIG_OF
-int __init vic_of_init(struct device_node *node, struct device_node *parent)
+static int __init vic_of_init(struct device_node *node,
+			      struct device_node *parent)
 {
 	void __iomem *regs;
 	int irq_base;
-- 
1.7.9.5

[PATCH 15/16] irqchip: vic: mark vic_of_init() as static

[Linus Walleij]

On Tue, Nov 20, 2012 at 11:01 PM, Thomas Petazzoni
<thomas.petazzoni@free-electrons.com> wrote:

> With all vic_of_init() users converted to the irqchip infrastructure,
> there is no longer a reason for this function to be
> exported. Therefore, mark it as static.
>
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

It may be that I have missed some patches, but are really all VIC
users converted in this patch series?

In arch/arm/mach-u300/core.c I have this:
        vic_init((void __iomem *) U300_INTCON0_VBASE, IRQ_U300_INTCON0_START,
                 mask[0], mask[0]);
        vic_init((void __iomem *) U300_INTCON1_VBASE, IRQ_U300_INTCON1_START,
                 mask[1], mask[1]);

arch/arm/mach-nomadik/cpu-8815.c:

        /* This modified VIC cell has two register blocks, at 0 and 0x20 */
        vic_init(io_p2v(NOMADIK_IC_BASE + 0x00), IRQ_VIC_START +  0, ~0, 0);
        vic_init(io_p2v(NOMADIK_IC_BASE + 0x20), IRQ_VIC_START + 32, ~0, 0);

And there are more still I think? I didn't see those converted so
I guess I must have missed some patches or something?

PS please make sure u300_defconfig and nhk8815_defconfig builds, or
I will be a sad puppy ...

Yours,
Linus Walleij

[PATCH 15/16] irqchip: vic: mark vic_of_init() as static

[Thomas Petazzoni]

Dear Linus Walleij,

On Wed, 21 Nov 2012 13:40:57 +0100, Linus Walleij wrote:
> On Tue, Nov 20, 2012 at 11:01 PM, Thomas Petazzoni
> <thomas.petazzoni@free-electrons.com> wrote:
> 
> > With all vic_of_init() users converted to the irqchip infrastructure,
> > there is no longer a reason for this function to be
> > exported. Therefore, mark it as static.
> >
> > Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
> 
> It may be that I have missed some patches, but are really all VIC
> users converted in this patch series?
> 
> In arch/arm/mach-u300/core.c I have this:
>         vic_init((void __iomem *) U300_INTCON0_VBASE, IRQ_U300_INTCON0_START,
>                  mask[0], mask[0]);
>         vic_init((void __iomem *) U300_INTCON1_VBASE, IRQ_U300_INTCON1_START,
>                  mask[1], mask[1]);
> 
> arch/arm/mach-nomadik/cpu-8815.c:
> 
>         /* This modified VIC cell has two register blocks, at 0 and 0x20 */
>         vic_init(io_p2v(NOMADIK_IC_BASE + 0x00), IRQ_VIC_START +  0, ~0, 0);
>         vic_init(io_p2v(NOMADIK_IC_BASE + 0x20), IRQ_VIC_START + 32, ~0, 0);
> 
> And there are more still I think? I didn't see those converted so
> I guess I must have missed some patches or something?

In this patch series, I'm making vic_of_init() static. vic_init() used
by non-DT platforms is kept as is, with no change, so I'm not sure why
you're worried about vic_init() being a problem.

> PS please make sure u300_defconfig and nhk8815_defconfig builds, or
> I will be a sad puppy ...

Indeed. The whole thing needs more testing, for sure. I guess I should
have labeled it as RFC, because it's really what it is.

Best regards,

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[PATCH 15/16] irqchip: vic: mark vic_of_init() as static

[Linus Walleij]

On Wed, Nov 21, 2012 at 1:53 PM, Thomas Petazzoni
<thomas.petazzoni@free-electrons.com> wrote:
> On Wed, 21 Nov 2012 13:40:57 +0100, Linus Walleij wrote:

>> It may be that I have missed some patches, but are really all VIC
>> users converted in this patch series?
>
> In this patch series, I'm making vic_of_init() static. vic_init() used
> by non-DT platforms is kept as is, with no change, so I'm not sure why
> you're worried about vic_init() being a problem.

Probably because I didn't parse the _of_ part :-/

Sorry about the fuzz!

But the idea is to use the same convenient mechanism for
the non-DT platforms later, so we get rid of all the incldue
headers I guess?

Yours,
Linus Walleij

[PATCH 15/16] irqchip: vic: mark vic_of_init() as static

[Thomas Petazzoni]

Linus,

On Wed, 21 Nov 2012 15:40:02 +0100, Linus Walleij wrote:
> On Wed, Nov 21, 2012 at 1:53 PM, Thomas Petazzoni
> <thomas.petazzoni@free-electrons.com> wrote:
> > On Wed, 21 Nov 2012 13:40:57 +0100, Linus Walleij wrote:
> 
> >> It may be that I have missed some patches, but are really all VIC
> >> users converted in this patch series?
> >
> > In this patch series, I'm making vic_of_init() static. vic_init() used
> > by non-DT platforms is kept as is, with no change, so I'm not sure why
> > you're worried about vic_init() being a problem.
> 
> Probably because I didn't parse the _of_ part :-/
> 
> Sorry about the fuzz!

No problem :)

> But the idea is to use the same convenient mechanism for
> the non-DT platforms later, so we get rid of all the incldue
> headers I guess?

Yes. I personally haven't yet thought of this part of the problem, but
I think Rob has a plan, and if not, we can certainly think of one
without too much problems.

Best regards,

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[PATCH 15/16] irqchip: vic: mark vic_of_init() as static

[Rob Herring]

On 11/21/2012 08:46 AM, Thomas Petazzoni wrote:
> Linus,
> 
> On Wed, 21 Nov 2012 15:40:02 +0100, Linus Walleij wrote:
>> On Wed, Nov 21, 2012 at 1:53 PM, Thomas Petazzoni
>> <thomas.petazzoni@free-electrons.com> wrote:
>>> On Wed, 21 Nov 2012 13:40:57 +0100, Linus Walleij wrote:
>>
>>>> It may be that I have missed some patches, but are really all VIC
>>>> users converted in this patch series?
>>>
>>> In this patch series, I'm making vic_of_init() static. vic_init() used
>>> by non-DT platforms is kept as is, with no change, so I'm not sure why
>>> you're worried about vic_init() being a problem.
>>
>> Probably because I didn't parse the _of_ part :-/
>>
>> Sorry about the fuzz!
> 
> No problem :)
> 
>> But the idea is to use the same convenient mechanism for
>> the non-DT platforms later, so we get rid of all the incldue
>> headers I guess?
> 
> Yes. I personally haven't yet thought of this part of the problem, but
> I think Rob has a plan, and if not, we can certainly think of one
> without too much problems.

While we can trim down the headers, I don't think we will get rid of
them in all cases. My only plan was everything gets converted to DT.
Then we can remove gic_init, gic_cascade_irq, and vic_init.

Rob

[PATCH 15/16] irqchip: vic: mark vic_of_init() as static

[Thomas Petazzoni]

Rob,

On Wed, 21 Nov 2012 09:07:59 -0600, Rob Herring wrote:

> > Yes. I personally haven't yet thought of this part of the problem, but
> > I think Rob has a plan, and if not, we can certainly think of one
> > without too much problems.
> 
> While we can trim down the headers, I don't think we will get rid of
> them in all cases. My only plan was everything gets converted to DT.
> Then we can remove gic_init, gic_cascade_irq, and vic_init.

Sounds like a good plan :-)

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[PATCH 16/16] ARM: plat-versatile: move FPGA irq driver to drivers/irqchip

[Thomas Petazzoni]

From: Linus Walleij <linus.walleij@linaro.org>

This moves the Versatile FPGA interrupt controller driver, used in
the Integrator/AP, Integrator/CP and some Versatile boards, out
of arch/arm/plat-versatile and down to drivers/irqchip where we
have consensus that such drivers belong. The header file is
consequently moved to <linux/platform_data/irq-versatile-fpga.h>.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
 arch/arm/Kconfig                                   |    4 +-
 arch/arm/mach-integrator/integrator_ap.c           |   12 +---
 arch/arm/mach-integrator/integrator_cp.c           |   11 +---
 arch/arm/mach-versatile/core.c                     |    2 +-
 arch/arm/plat-versatile/Kconfig                    |    9 ---
 arch/arm/plat-versatile/Makefile                   |    1 -
 drivers/irqchip/Kconfig                            |    9 +++
 drivers/irqchip/Makefile                           |    1 +
 .../irqchip/irq-versatile-fpga.c                   |   60 ++++++++++----------
 .../linux/irqchip/versatile-fpga.h                 |    2 -
 10 files changed, 49 insertions(+), 62 deletions(-)
 rename arch/arm/plat-versatile/fpga-irq.c => drivers/irqchip/irq-versatile-fpga.c (83%)
 rename arch/arm/plat-versatile/include/plat/fpga-irq.h => include/linux/irqchip/versatile-fpga.h (73%)

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index ade7e92..22b7ac7 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -284,8 +284,8 @@ config ARCH_INTEGRATOR
 	select MULTI_IRQ_HANDLER
 	select NEED_MACH_MEMORY_H
 	select PLAT_VERSATILE
-	select PLAT_VERSATILE_FPGA_IRQ
 	select SPARSE_IRQ
+	select VERSATILE_FPGA_IRQ
 	help
 	  Support for ARM's Integrator platform.
 
@@ -318,7 +318,7 @@ config ARCH_VERSATILE
 	select PLAT_VERSATILE
 	select PLAT_VERSATILE_CLCD
 	select PLAT_VERSATILE_CLOCK
-	select PLAT_VERSATILE_FPGA_IRQ
+	select VERSATILE_FPGA_IRQ
 	help
 	  This enables support for ARM Ltd Versatile board.
 
diff --git a/arch/arm/mach-integrator/integrator_ap.c b/arch/arm/mach-integrator/integrator_ap.c
index e6617c1..28649ac 100644
--- a/arch/arm/mach-integrator/integrator_ap.c
+++ b/arch/arm/mach-integrator/integrator_ap.c
@@ -20,6 +20,7 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
+#include <linux/irqchip.h>
 #include <linux/list.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
@@ -31,6 +32,7 @@
 #include <linux/clockchips.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
+#include <linux/irqchip/versatile-fpga.h>
 #include <linux/mtd/physmap.h>
 #include <linux/clk.h>
 #include <linux/platform_data/clk-integrator.h>
@@ -56,8 +58,6 @@
 #include <asm/mach/pci.h>
 #include <asm/mach/time.h>
 
-#include <plat/fpga-irq.h>
-
 #include "common.h"
 
 /* 
@@ -432,16 +432,11 @@ static struct sys_timer ap_of_timer = {
 	.init		= ap_init_timer_of,
 };
 
-static const struct of_device_id fpga_irq_of_match[] __initconst = {
-	{ .compatible = "arm,versatile-fpga-irq", .data = fpga_irq_of_init, },
-	{ /* Sentinel */ }
-};
-
 static void __init ap_init_irq_of(void)
 {
 	/* disable core module IRQs */
 	writel(0xffffffffU, VA_CMIC_BASE + IRQ_ENABLE_CLEAR);
-	of_irq_init(fpga_irq_of_match);
+	irqchip_init();
 	integrator_clk_init(false);
 }
 
@@ -502,7 +497,6 @@ DT_MACHINE_START(INTEGRATOR_AP_DT, "ARM Integrator/AP (Device Tree)")
 	.nr_irqs	= NR_IRQS_INTEGRATOR_AP,
 	.init_early	= ap_init_early,
 	.init_irq	= ap_init_irq_of,
-	.handle_irq	= fpga_handle_irq,
 	.timer		= &ap_of_timer,
 	.init_machine	= ap_init_of,
 	.restart	= integrator_restart,
diff --git a/arch/arm/mach-integrator/integrator_cp.c b/arch/arm/mach-integrator/integrator_cp.c
index 5b08e8e..64d1b57 100644
--- a/arch/arm/mach-integrator/integrator_cp.c
+++ b/arch/arm/mach-integrator/integrator_cp.c
@@ -10,6 +10,7 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
+#include <linux/irqchip.h>
 #include <linux/list.h>
 #include <linux/platform_device.h>
 #include <linux/dma-mapping.h>
@@ -20,6 +21,7 @@
 #include <linux/amba/clcd.h>
 #include <linux/amba/mmci.h>
 #include <linux/io.h>
+#include <linux/irqchip/versatile-fpga.h>
 #include <linux/gfp.h>
 #include <linux/mtd/physmap.h>
 #include <linux/platform_data/clk-integrator.h>
@@ -46,7 +48,6 @@
 #include <asm/hardware/timer-sp.h>
 
 #include <plat/clcd.h>
-#include <plat/fpga-irq.h>
 #include <plat/sched_clock.h>
 
 #include "common.h"
@@ -299,14 +300,9 @@ static struct sys_timer cp_of_timer = {
 	.init		= intcp_timer_init_of,
 };
 
-static const struct of_device_id fpga_irq_of_match[] __initconst = {
-	{ .compatible = "arm,versatile-fpga-irq", .data = fpga_irq_of_init, },
-	{ /* Sentinel */ }
-};
-
 static void __init intcp_init_irq_of(void)
 {
-	of_irq_init(fpga_irq_of_match);
+	irqchip_init();
 	integrator_clk_init(true);
 }
 
@@ -353,7 +349,6 @@ DT_MACHINE_START(INTEGRATOR_CP_DT, "ARM Integrator/CP (Device Tree)")
 	.nr_irqs	= NR_IRQS_INTEGRATOR_CP,
 	.init_early	= intcp_init_early,
 	.init_irq	= intcp_init_irq_of,
-	.handle_irq	= fpga_handle_irq,
 	.timer		= &cp_of_timer,
 	.init_machine	= intcp_init_of,
 	.restart	= integrator_restart,
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index 5b5c1ee..5d59294 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -32,6 +32,7 @@
 #include <linux/amba/mmci.h>
 #include <linux/amba/pl022.h>
 #include <linux/io.h>
+#include <linux/irqchip/versatile-fpga.h>
 #include <linux/gfp.h>
 #include <linux/clkdev.h>
 #include <linux/mtd/physmap.h>
@@ -51,7 +52,6 @@
 #include <asm/hardware/timer-sp.h>
 
 #include <plat/clcd.h>
-#include <plat/fpga-irq.h>
 #include <plat/sched_clock.h>
 
 #include "core.h"
diff --git a/arch/arm/plat-versatile/Kconfig b/arch/arm/plat-versatile/Kconfig
index 2a4ae8a..619f0fa 100644
--- a/arch/arm/plat-versatile/Kconfig
+++ b/arch/arm/plat-versatile/Kconfig
@@ -6,15 +6,6 @@ config PLAT_VERSATILE_CLOCK
 config PLAT_VERSATILE_CLCD
 	bool
 
-config PLAT_VERSATILE_FPGA_IRQ
-	bool
-	select IRQ_DOMAIN
-
-config PLAT_VERSATILE_FPGA_IRQ_NR
-       int
-       default 4
-       depends on PLAT_VERSATILE_FPGA_IRQ
-
 config PLAT_VERSATILE_LEDS
 	def_bool y if NEW_LEDS
 	depends on ARCH_REALVIEW || ARCH_VERSATILE
diff --git a/arch/arm/plat-versatile/Makefile b/arch/arm/plat-versatile/Makefile
index 74cfd94..f88d448 100644
--- a/arch/arm/plat-versatile/Makefile
+++ b/arch/arm/plat-versatile/Makefile
@@ -2,7 +2,6 @@ ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/$(src)/include
 
 obj-$(CONFIG_PLAT_VERSATILE_CLOCK) += clock.o
 obj-$(CONFIG_PLAT_VERSATILE_CLCD) += clcd.o
-obj-$(CONFIG_PLAT_VERSATILE_FPGA_IRQ) += fpga-irq.o
 obj-$(CONFIG_PLAT_VERSATILE_LEDS) += leds.o
 obj-$(CONFIG_PLAT_VERSATILE_SCHED_CLOCK) += sched-clock.o
 obj-$(CONFIG_SMP) += headsmp.o platsmp.o
diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
index d80e5c7..a774cd3 100644
--- a/drivers/irqchip/Kconfig
+++ b/drivers/irqchip/Kconfig
@@ -24,3 +24,12 @@ config ARM_VIC_NR
 	help
 	  The maximum number of VICs available in the system, for
 	  power management.
+
+config VERSATILE_FPGA_IRQ
+	bool
+	select IRQ_DOMAIN
+
+config VERSATILE_FPGA_IRQ_NR
+	int
+	default 4
+	depends on VERSATILE_FPGA_IRQ
diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile
index 9c7f91e..476a5e4 100644
--- a/drivers/irqchip/Makefile
+++ b/drivers/irqchip/Makefile
@@ -3,3 +3,4 @@ obj-$(CONFIG_ARCH_BCM2835) += irq-bcm2835.o
 obj-$(CONFIG_ARCH_MVEBU)   += irq-armada-370-xp.o
 obj-$(CONFIG_ARM_GIC)   += irq-gic.o
 obj-$(CONFIG_ARM_VIC)   += irq-vic.o
+obj-$(CONFIG_VERSATILE_FPGA_IRQ) += irq-versatile-fpga.o
diff --git a/arch/arm/plat-versatile/fpga-irq.c b/drivers/irqchip/irq-versatile-fpga.c
similarity index 83%
rename from arch/arm/plat-versatile/fpga-irq.c
rename to drivers/irqchip/irq-versatile-fpga.c
index 091ae10..3c48946 100644
--- a/arch/arm/plat-versatile/fpga-irq.c
+++ b/drivers/irqchip/irq-versatile-fpga.c
@@ -1,8 +1,10 @@
 /*
  *  Support for Versatile FPGA-based IRQ controllers
  */
+#include <linux/bitops.h>
 #include <linux/irq.h>
 #include <linux/io.h>
+#include <linux/irqchip/versatile-fpga.h>
 #include <linux/irqdomain.h>
 #include <linux/module.h>
 #include <linux/of.h>
@@ -10,7 +12,8 @@
 
 #include <asm/exception.h>
 #include <asm/mach/irq.h>
-#include <plat/fpga-irq.h>
+
+#include "irqchip.h"
 
 #define IRQ_STATUS		0x00
 #define IRQ_RAW_STATUS		0x04
@@ -41,7 +44,7 @@ struct fpga_irq_data {
 };
 
 /* we cannot allocate memory when the controllers are initially registered */
-static struct fpga_irq_data fpga_irq_devices[CONFIG_PLAT_VERSATILE_FPGA_IRQ_NR];
+static struct fpga_irq_data fpga_irq_devices[CONFIG_VERSATILE_FPGA_IRQ_NR];
 static int fpga_irq_id;
 
 static void fpga_irq_mask(struct irq_data *d)
@@ -117,13 +120,12 @@ static int fpga_irqdomain_map(struct irq_domain *d, unsigned int irq,
 	struct fpga_irq_data *f = d->host_data;
 
 	/* Skip invalid IRQs, only register handlers for the real ones */
-	if (!(f->valid & (1 << hwirq)))
+	if (!(f->valid & BIT(hwirq)))
 		return -ENOTSUPP;
 	irq_set_chip_data(irq, f);
 	irq_set_chip_and_handler(irq, &f->chip,
 				handle_level_irq);
 	set_irq_flags(irq, IRQF_VALID | IRQF_PROBE);
-	f->used_irqs++;
 	return 0;
 }
 
@@ -132,13 +134,15 @@ static struct irq_domain_ops fpga_irqdomain_ops = {
 	.xlate = irq_domain_xlate_onetwocell,
 };
 
-static __init struct fpga_irq_data *
-fpga_irq_prep_struct(void __iomem *base, const char *name, u32 valid) {
+void __init fpga_irq_init(void __iomem *base, const char *name, int irq_start,
+			  int parent_irq, u32 valid, struct device_node *node)
+{
 	struct fpga_irq_data *f;
+	int i;
 
 	if (fpga_irq_id >= ARRAY_SIZE(fpga_irq_devices)) {
-		printk(KERN_ERR "%s: too few FPGA IRQ controllers, increase CONFIG_PLAT_VERSATILE_FPGA_IRQ_NR\n", __func__);
-		return NULL;
+		pr_err("%s: too few FPGA IRQ controllers, increase CONFIG_PLAT_VERSATILE_FPGA_IRQ_NR\n", __func__);
+		return;
 	}
 	f = &fpga_irq_devices[fpga_irq_id];
 	f->base = base;
@@ -147,29 +151,28 @@ fpga_irq_prep_struct(void __iomem *base, const char *name, u32 valid) {
 	f->chip.irq_mask = fpga_irq_mask;
 	f->chip.irq_unmask = fpga_irq_unmask;
 	f->valid = valid;
-	fpga_irq_id++;
-
-	return f;
-}
-
-void __init fpga_irq_init(void __iomem *base, const char *name, int irq_start,
-			  int parent_irq, u32 valid, struct device_node *node)
-{
-	struct fpga_irq_data *f;
-
-	f = fpga_irq_prep_struct(base, name, valid);
-	if (!f)
-		return;
 
 	if (parent_irq != -1) {
 		irq_set_handler_data(parent_irq, f);
 		irq_set_chained_handler(parent_irq, fpga_irq_handle);
 	}
 
-	f->domain = irq_domain_add_legacy(node, fls(valid), irq_start, 0,
+	/* This will also allocate irq descriptors */
+	f->domain = irq_domain_add_simple(node, fls(valid), irq_start,
 					  &fpga_irqdomain_ops, f);
+
+	/* This will allocate all valid descriptors in the linear case */
+	for (i = 0; i < fls(valid); i++)
+		if (valid & BIT(i)) {
+			if (!irq_start)
+				irq_create_mapping(f->domain, i);
+			f->used_irqs++;
+		}
+
 	pr_info("FPGA IRQ chip %d \"%s\" @ %p, %u irqs\n",
 		fpga_irq_id, name, base, f->used_irqs);
+
+	fpga_irq_id++;
 }
 
 #ifdef CONFIG_OF
@@ -193,18 +196,15 @@ int __init fpga_irq_of_init(struct device_node *node,
 	if (of_property_read_u32(node, "valid-mask", &valid_mask))
 		valid_mask = 0;
 
-	f = fpga_irq_prep_struct(base, node->name, valid_mask);
-	if (!f)
-		return -ENOMEM;
+	fpga_irq_init(base, node->name, 0, -1, valid_mask, node);
 
 	writel(clear_mask, base + IRQ_ENABLE_CLEAR);
 	writel(clear_mask, base + FIQ_ENABLE_CLEAR);
 
-	f->domain = irq_domain_add_linear(node, fls(valid_mask), &fpga_irqdomain_ops, f);
-	f->used_irqs = hweight32(valid_mask);
-
-	pr_info("FPGA IRQ chip %d \"%s\" @ %p, %u irqs\n",
-		fpga_irq_id, node->name, base, f->used_irqs);
 	return 0;
 }
+
+IRQCHIP_DECLARE(versatile_fpga_irq, "arm,versatile-fpga-irq",
+		fpga_irq_of_init);
+
 #endif
diff --git a/arch/arm/plat-versatile/include/plat/fpga-irq.h b/include/linux/irqchip/versatile-fpga.h
similarity index 73%
rename from arch/arm/plat-versatile/include/plat/fpga-irq.h
rename to include/linux/irqchip/versatile-fpga.h
index 1fac965..91bcfb6 100644
--- a/arch/arm/plat-versatile/include/plat/fpga-irq.h
+++ b/include/linux/irqchip/versatile-fpga.h
@@ -7,7 +7,5 @@ struct pt_regs;
 void fpga_handle_irq(struct pt_regs *regs);
 void fpga_irq_init(void __iomem *, const char *, int, int, u32,
 		struct device_node *node);
-int fpga_irq_of_init(struct device_node *node,
-		     struct device_node *parent);
 
 #endif
-- 
1.7.9.5

[PATCH 16/16] ARM: plat-versatile: move FPGA irq driver to drivers/irqchip

[Linus Walleij]

On Tue, Nov 20, 2012 at 11:01 PM, Thomas Petazzoni
<thomas.petazzoni@free-electrons.com> wrote:

> From: Linus Walleij <linus.walleij@linaro.org>
>
> This moves the Versatile FPGA interrupt controller driver, used in
> the Integrator/AP, Integrator/CP and some Versatile boards, out
> of arch/arm/plat-versatile and down to drivers/irqchip where we
> have consensus that such drivers belong. The header file is
> consequently moved to <linux/platform_data/irq-versatile-fpga.h>.
>
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>

It appears you have modified my commit so this is not the
whole story right?

This doesn't look like I wrote it:
+IRQCHIP_DECLARE(versatile_fpga_irq, "arm,versatile-fpga-irq",
+               fpga_irq_of_init);
+

Is it possible to do your changes on top of my commit
instead?

Atleast add your Signed-off-by please...

Yours,
Linus Walleij

[PATCH 16/16] ARM: plat-versatile: move FPGA irq driver to drivers/irqchip

[Thomas Petazzoni]

Linus,

On Wed, 21 Nov 2012 13:22:27 +0100, Linus Walleij wrote:

> > This moves the Versatile FPGA interrupt controller driver, used in
> > the Integrator/AP, Integrator/CP and some Versatile boards, out
> > of arch/arm/plat-versatile and down to drivers/irqchip where we
> > have consensus that such drivers belong. The header file is
> > consequently moved to <linux/platform_data/irq-versatile-fpga.h>.
> >
> > Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> 
> It appears you have modified my commit so this is not the
> whole story right?
> 
> This doesn't look like I wrote it:
> +IRQCHIP_DECLARE(versatile_fpga_irq, "arm,versatile-fpga-irq",
> +               fpga_irq_of_init);
> +

Right, sorry. As I said earlier, this v4 is not intended for merging,
but only to show the current state, and define together what remains to
be fixed/changed.

> Is it possible to do your changes on top of my commit
> instead?

Not if we want to keep bisectability.

> Atleast add your Signed-off-by please...

Will do, thanks for the notice.

Best regards,

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[PATCH v4] Introduce irqchip infrastructure

[Rob Herring]

On 11/20/2012 04:00 PM, Thomas Petazzoni wrote:
> Hello,
> 
> Here is a set of patches that introduces a small irqchip
> infrastructure that allows the registration of irqchip drivers without
> having each of those drivers to expose a public API and header in
> include/linux/irqchip/, and moves a number of existing irqchip
> controller to this infrastructure.
> 
> This new version (detailed changelog below) provides a number of
> improvements: usage of a special link section to avoid the central
> registration in irqchip.{c,h} of all irqchip drivers, and addition of
> the GIC, VIC and Versatile FPGA IRQ drivers move, thanks to Rob
> Herring and Linus Walleij.
> 
> Remaining issues to solve:
> 
>  * Russell would like arch/arm/include/asm/hardware/vic.h and
>    arch/arm/include/asm/hardware/gic.h to move elsewhere, since the
>    corresponding code is no longer in arch/arm/. I can move them in
>    <linux/irqchip>, but that will cause a fairly large change as there
>    are a big number of users of those headers in arch/arm.

We need to do this. KVM will need the header as well, so we can't move
the register definitions out either.

>  * The arch/arm/include/asm/hardware/vic.h has a few offset macros
>    that should move into the .c file of the irqchip driver, but some
>    of those offsets are bizarrely used in some other pieces of code in
>    arch/arm/.

Not really a good solution that I can see there. We could punt on doing
the VIC for now.

> 
>  * How to get rid entirely of the headers in <linux/irqchip/>. The
>    remaining problematic functions are <foo>_handle_irq() and
>    <foo>_irq_init() that are used by non-DT platforms. And the special
>    case of gic_cascade_irq().

I have patches for removing vic/gic_handle_irq. It is going to take a
while for the others, so I don't think we should wait for that. KVM
needs the GIC header anyway.

>  * Move all the users of gic_of_init() to the irqchip mechanism
>    (Exynos, i.MX 6, MSM, OMAP, SH Mobile, socfpga, spear13xx, tegra,
>    ux500, vexpress) so that gic_of_init() no longer has to be
>    exported.

I have a patch doing this. I will try to get this sent out today. I've
split this into clean-up and then the move, so the clean-up could go in
for 3.8. It is only dependent on patch 2.

> Of course, I don't expect any of this to go in 3.8, it is 3.9 material
> if we manage to reach a consensus on the remaining issues to solve
> (and the other issues that will certainly show up or be raised by
> other people).

I still would like to try to get this in for 3.8. If not the move, some
of the clean-up.

Rob

> 
> This series has been built and boot tested for BCM2835 and Armada
> XP. It has been built tested on Versatile with OF and without OF,
> spear3xx, spear6xx and a multi_v6_v7 configuration with all possible
> architectures enabled (therefore including picoxcell).
> 
> Changes since v3:
> 
>  * Switch to a link-trick based registration of the irqchip drivers,
>    using the IRQCHIP_DECLARE() macro.
> 
>  * Integrate Rob Herring's patches to move the GIC and VIC irq
>    controller drivers in drivers/irqchip/.
> 
>  * Migrate picoxcell, spear3xx and spear6xx to using the irqchip
>    platform, so that no direct user of vic_of_init() remains.
> 
>  * Don't export vic_of_init() anymore: all users go through
>    irqchip_init() now.
> 
>  * Integrate Linus Walleij patch that moves the Versatile FPGA irq
>    controller driver to drivers/irqchip.
> 
>  * A set_handle_irq() ARM function was added to properly set the
>    handle_arch_irq pointer to the right handler of the parent IRQ
>    controller.
> 
> Changes since v2:
> 
>  * Fixed the entry in the MAINTAINERS file. Noticed by Rob Herring.
> 
>  * Simplified the Kconfig logic by making the IRQCHIP option enabled
>    by default as soon as OF_IRQ is enabled. The individual ARCH_<foo>
>    Kconfig options no longer have to select IRQCHIP. Also, the option
>    was renamed from USE_IRQCHIP to just IRQCHIP. Suggested by Rob
>    Herring.
> 
>  * Added Reviewed-by tags given by Rob Herring.
> 
> Changes since v1:
> 
>  * Add a new patch mentionning the drivers/irqchip in the list of
>    directories part of the IRQ subsystem maintained by Thomas Gleixner
>    in the MAINTAINERS file. Requested by Arnd Bergmann.
> 
>  * Reduce the amount of code movement in the irq-bcm2835.c and
>    irq-armada-370-xp.c files by using one forward declaration for the
>    IRQ handling entry point. Requested by Stephen Warren.
> 
>  * Rename the armctrl_of_init() function to bcm2835_irqchip_init() as
>    requested by Stephen Warren.
> 
>  * Added the formal Acked-by and Reviewed-by received from Stephen
>    Warren.
> 
> Thanks,
> 
> Thomas
> 

[PATCH v4] Introduce irqchip infrastructure

[Thomas Petazzoni]

Rob,

On Tue, 20 Nov 2012 16:38:20 -0600, Rob Herring wrote:

> > Remaining issues to solve:
> > 
> >  * Russell would like arch/arm/include/asm/hardware/vic.h and
> >    arch/arm/include/asm/hardware/gic.h to move elsewhere, since the
> >    corresponding code is no longer in arch/arm/. I can move them in
> >    <linux/irqchip>, but that will cause a fairly large change as there
> >    are a big number of users of those headers in arch/arm.
> 
> We need to do this. KVM will need the header as well, so we can't move
> the register definitions out either.

Which register definitions are you talking about? The one from vic.h ?

> >  * The arch/arm/include/asm/hardware/vic.h has a few offset macros
> >    that should move into the .c file of the irqchip driver, but some
> >    of those offsets are bizarrely used in some other pieces of code in
> >    arch/arm/.
> 
> Not really a good solution that I can see there. We could punt on doing
> the VIC for now.

Well, most of the VIC macro users are strange. Most of them probably
need their own macro definition instead of shamelessly borrowing the
VIC definitions.

> >  * How to get rid entirely of the headers in <linux/irqchip/>. The
> >    remaining problematic functions are <foo>_handle_irq() and
> >    <foo>_irq_init() that are used by non-DT platforms. And the special
> >    case of gic_cascade_irq().
> 
> I have patches for removing vic/gic_handle_irq. It is going to take a
> while for the others, so I don't think we should wait for that. KVM
> needs the GIC header anyway.

What is your approach to remove vic/gic_handle_irq? I'll see in your
patches I guess.

That said, I think that having <linux/irqchip/gic.h> and
<linux/irqchip/vic.h> for now is not a big problem. What we want to
avoid is to have gazillions of headers here, but as new platforms are
DT-capable, those platforms shouldn't need to add a header in
<linux/irqchip/>, so I think we could live with
<linux/irqchip/{gic,vic}.h> for a while, and progressively clean
up/improve what needs to be done. Doing the perfect migration as an
unique step is going to be difficult.

> >  * Move all the users of gic_of_init() to the irqchip mechanism
> >    (Exynos, i.MX 6, MSM, OMAP, SH Mobile, socfpga, spear13xx, tegra,
> >    ux500, vexpress) so that gic_of_init() no longer has to be
> >    exported.
> 
> I have a patch doing this. I will try to get this sent out today. I've
> split this into clean-up and then the move, so the clean-up could go in
> for 3.8. It is only dependent on patch 2.

Ok.

> > Of course, I don't expect any of this to go in 3.8, it is 3.9 material
> > if we manage to reach a consensus on the remaining issues to solve
> > (and the other issues that will certainly show up or be raised by
> > other people).
> 
> I still would like to try to get this in for 3.8. If not the move, some
> of the clean-up.

So the cleanup series for 3.8 would contain:

 [PATCH 02/16], needed for the clean up you will send later today
 [PATCH 06/16]
 [PATCH 07/16]

I think all the other ones are directly related to the introduction of
the irqchip infrastructure, so they most likely cannot be part of 3.8.

So, I can prepare a pull request with 2, 6, 7 and your upcoming clean
up patch, if you want.

Thomas
-- 
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com

[PATCH v4] Introduce irqchip infrastructure

[Rob Herring]

On 11/20/2012 05:12 PM, Thomas Petazzoni wrote:
> Rob,
> 
> On Tue, 20 Nov 2012 16:38:20 -0600, Rob Herring wrote:
> 
>>> Remaining issues to solve:
>>>
>>>  * Russell would like arch/arm/include/asm/hardware/vic.h and
>>>    arch/arm/include/asm/hardware/gic.h to move elsewhere, since the
>>>    corresponding code is no longer in arch/arm/. I can move them in
>>>    <linux/irqchip>, but that will cause a fairly large change as there
>>>    are a big number of users of those headers in arch/arm.
>>
>> We need to do this. KVM will need the header as well, so we can't move
>> the register definitions out either.
> 
> Which register definitions are you talking about? The one from vic.h ?
> 
>>>  * The arch/arm/include/asm/hardware/vic.h has a few offset macros
>>>    that should move into the .c file of the irqchip driver, but some
>>>    of those offsets are bizarrely used in some other pieces of code in
>>>    arch/arm/.
>>
>> Not really a good solution that I can see there. We could punt on doing
>> the VIC for now.
> 
> Well, most of the VIC macro users are strange. Most of them probably
> need their own macro definition instead of shamelessly borrowing the
> VIC definitions.
> 
>>>  * How to get rid entirely of the headers in <linux/irqchip/>. The
>>>    remaining problematic functions are <foo>_handle_irq() and
>>>    <foo>_irq_init() that are used by non-DT platforms. And the special
>>>    case of gic_cascade_irq().
>>
>> I have patches for removing vic/gic_handle_irq. It is going to take a
>> while for the others, so I don't think we should wait for that. KVM
>> needs the GIC header anyway.
> 
> What is your approach to remove vic/gic_handle_irq? I'll see in your
> patches I guess.
> 
> That said, I think that having <linux/irqchip/gic.h> and
> <linux/irqchip/vic.h> for now is not a big problem. What we want to
> avoid is to have gazillions of headers here, but as new platforms are
> DT-capable, those platforms shouldn't need to add a header in
> <linux/irqchip/>, so I think we could live with
> <linux/irqchip/{gic,vic}.h> for a while, and progressively clean
> up/improve what needs to be done. Doing the perfect migration as an
> unique step is going to be difficult.
> 
>>>  * Move all the users of gic_of_init() to the irqchip mechanism
>>>    (Exynos, i.MX 6, MSM, OMAP, SH Mobile, socfpga, spear13xx, tegra,
>>>    ux500, vexpress) so that gic_of_init() no longer has to be
>>>    exported.
>>
>> I have a patch doing this. I will try to get this sent out today. I've
>> split this into clean-up and then the move, so the clean-up could go in
>> for 3.8. It is only dependent on patch 2.

I've pushed out patches that moves gic.h and converts all GIC DT
platforms over to irqchip_init:

git://sources.calxeda.com/kernel/linux.git gic-irqchip

I haven't gotten to rebasing the VIC ones yet.

One issue I found is it would be easier to move things if
IRQCHIP_DECLARE was accessible in arch/arm. Then I can convert all the
init over to irqchip_init first and then move gic.c and gic.h together
in one patch.

Rob

[PATCH v4] Introduce irqchip infrastructure

[Grant Likely]

On Tue, 20 Nov 2012 23:00:51 +0100, Thomas Petazzoni <thomas.petazzoni@free-electrons.com> wrote:
> Hello,
> 
> Here is a set of patches that introduces a small irqchip
> infrastructure that allows the registration of irqchip drivers without
> having each of those drivers to expose a public API and header in
> include/linux/irqchip/, and moves a number of existing irqchip
> controller to this infrastructure.
> 
> This new version (detailed changelog below) provides a number of
> improvements: usage of a special link section to avoid the central
> registration in irqchip.{c,h} of all irqchip drivers, and addition of
> the GIC, VIC and Versatile FPGA IRQ drivers move, thanks to Rob
> Herring and Linus Walleij.

Hey Thomas,

This looks pretty darn sensible. Good job.

I've not reviewed deeply, but you can add my:

Reviewed-by: Grant Likely <grant.likely@secretlab.ca>

g.