Make mach/irqs.h optional

From: Rob Herring <[email protected]>

Since asm/irq.h may not include mach/irqs.h, include mach/irqs.h directly.

Signed-off-by Rob Herring <[email protected]>

---
 drivers/rtc/rtc-sa1100.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

drivers/rtc/rtc-sa1100.c +1 -1

index 4595d3e..144fa06 100644
@@ -33,7 +33,7 @@
 #include <linux/io.h>
 
 #include <mach/hardware.h>
-#include <asm/irq.h>
+#include <mach/irqs.h>
 
 #define RTC_DEF_DIVIDER		(32768 - 1)
 #define RTC_DEF_TRIM		0

Signature · 1 line

1.7.5.4

Quoted 3 lines · Rob Herring

From: Rob Herring <[email protected]>

Since asm/irq.h may not include mach/irqs.h, include mach/irqs.h directly.

Rather than doing this, we should be eliminating the need for these old drivers to include mach/irqs.h at all.

Adding the resources to the platform device, and then using at least the IRQ resource(s) would be a definite step forward (and a relatively easy one to do.)

Quoted 22 lines

Signed-off-by Rob Herring <[email protected]>

---
 drivers/rtc/rtc-sa1100.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

drivers/rtc/rtc-sa1100.c +1 -1

index 4595d3e..144fa06 100644
@@ -33,7 +33,7 @@
 #include <linux/io.h>
 
 #include <mach/hardware.h>
-#include <asm/irq.h>
+#include <mach/irqs.h>
 
 #define RTC_DEF_DIVIDER		(32768 - 1)
 #define RTC_DEF_TRIM		0

Signature · 1 line

1.7.5.4

Quoted 8 lines · Russell King - ARM Linux

Quoted 3 lines · Rob Herring

From: Rob Herring <[email protected]>

Since asm/irq.h may not include mach/irqs.h, include mach/irqs.h directly.

Rather than doing this, we should be eliminating the need for these old drivers to include mach/irqs.h at all.

Quoted 4 lines

Adding the resources to the platform device, and then using at least the IRQ resource(s) would be a definite step forward (and a relatively easy one to do.)

Well, this has just happened in 3.3. So I can now drop this one I think.

Another problem that resources solve (and your original patch didn't make it any better nor worse) has to do with include file search paths. Once we start building more than one SOC in the kernel, we won't be able to support the inclusion of <mach/whatever.h> from driver code anymore since this could refer to more than one instances of arch/arm/mach-*/include/mach/whatever.h.

Why put so much work into a mostly dead platform? Well, it's one of the few decent platforms which I have, which actually has working DMA hardware. Why audio? It's one of those which provides a significant stress test on DMA stuff being correct. Why IrDA? It can reconfigure a DMA channel (eg, burst size) on the fly when it switches between SIR/FIR mode.

It may be 12 years old but it's still a very useful platform to develop non-hardware specific features on!

Quoted 14 lines · Rob Herring

Quoted 12 lines · Russell King - ARM Linux

Quoted 3 lines · Rob Herring

From: Rob Herring <[email protected]>

Since asm/irq.h may not include mach/irqs.h, include mach/irqs.h directly.

Rather than doing this, we should be eliminating the need for these old drivers to include mach/irqs.h at all.

Adding the resources to the platform device, and then using at least the IRQ resource(s) would be a definite step forward (and a relatively easy one to do.)

Quoted 1 line

Well, this has just happened in 3.3. So I can now drop this one I think.

Another problem that resources solve (and your original patch didn't make it any better nor worse) has to do with include file search paths. Once we start building more than one SOC in the kernel, we won't be able to support the inclusion of <mach/whatever.h> from driver code anymore since this could refer to more than one instances of arch/arm/mach-*/include/mach/whatever.h.

Quoted 6 lines · Nicolas Pitre

Another problem that resources solve (and your original patch didn't make it any better nor worse) has to do with include file search paths. Once we start building more than one SOC in the kernel, we won't be able to support the inclusion of <mach/whatever.h> from driver code anymore since this could refer to more than one instances of arch/arm/mach-*/include/mach/whatever.h.

I'm fully aware of this issue and was not trying to solve that here.

From: Rob Herring <[email protected]>

The include of mach/irqs.h isn't needed, so remove it. Compiled CMOS rtc driver.

From: Rob Herring <[email protected]>

In preparation to make mach/irqs.h optional, directly include mach/irq.h to get IRQ_BOARD_START.

Signed-off-by Rob Herring <[email protected]>

---
 arch/arm/include/asm/hardware/it8152.h |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

arch/arm/include/asm/hardware/it8152.h +0 -0

index 43cab49..73f84fa 100644

Quoted 16 lines · Rob Herring

From: Rob Herring <[email protected]>

In preparation to make mach/irqs.h optional, directly include mach/irq.h to get IRQ_BOARD_START.

Signed-off-by Rob Herring <[email protected]>

---
 arch/arm/include/asm/hardware/it8152.h |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

arch/arm/include/asm/hardware/it8152.h +0 -0

index 43cab49..73f84fa 100644
@@ -9,6 +9,9 @@
 

arch/arm/include/asm/hardware/it8152.h line 14

Quoted 5 lines

arch/arm/include/asm/hardware/it8152.h +3 -0

 #ifndef __ASM_HARDWARE_IT8152_H
 #define __ASM_HARDWARE_IT8152_H
+
+#include <mach/irqs.h>
+

This include is needed for IRQ_BOARD_START which could also be problematic outside a single mach dir. The IT8152 is only used on 1 family of pxa2xx boards. The part and board with it are EOL as well, so we're not going to see new users of this code. Perhaps the it8152 code should just be moved into mach-pxa?

From: Rob Herring <[email protected]>

SH intc has a compile time dependency on NR_IRQS. Make this dependency a local define so that shmobile (and ARM in general) can have run-time NR_IRQS setting. SH has NR_IRQS set to 512 and shmobile has NR_IRQS set to 1024, so we are using the maximum.

Signed-off-by Rob Herring <[email protected]>

---

?drivers/sh/intc/balancing.c | ? ?2 +-
?drivers/sh/intc/core.c ? ? ?| ? ?2 +-
?drivers/sh/intc/handle.c ? ?| ? ?2 +-
?drivers/sh/intc/internals.h | ? ?9 +++++++++
?drivers/sh/intc/virq.c ? ? ?| ? ?2 +-

?5 files changed, 13 insertions(+), 4 deletions(-)

drivers/sh/intc/balancing.c +0 -0

index cec7a96..bc78080 100644
@@ -9,7 +9,7 @@

?*/
?#include "internals.h"

drivers/sh/intc/balancing.c +1 -1

-static unsigned long dist_handle[NR_IRQS];
+static unsigned long dist_handle[INTC_NR_IRQS];

?void intc_balancing_enable(unsigned int irq)
?{

drivers/sh/intc/core.c +0 -0

index e53e449..2fde897 100644
@@ -42,7 +42,7 @@ unsigned int nr_intc_controllers;

?* - this needs to be at least 2 for 5-bit priorities on 7780 ?*/
?static unsigned int default_prio_level = 2; ? ?/* 2 - 16 */

drivers/sh/intc/core.c +1 -1

-static unsigned int intc_prio_level[NR_IRQS]; ?/* for now */
+static unsigned int intc_prio_level[INTC_NR_IRQS]; ? ? /* for now */

?unsigned int intc_get_dfl_prio_level(void)
?{

drivers/sh/intc/handle.c +0 -0

index 057ce56..f461d53 100644
@@ -13,7 +13,7 @@

?#include <linux/spinlock.h>
?#include "internals.h"

drivers/sh/intc/handle.c +1 -1

-static unsigned long ack_handle[NR_IRQS];
+static unsigned long ack_handle[INTC_NR_IRQS];

?static intc_enum __init intc_grp_id(struct intc_desc *desc, ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?intc_enum enum_id)

drivers/sh/intc/internals.h +0 -0

index b0e9155..469f092 100644
@@ -6,6 +6,15 @@

?#include <linux/radix-tree.h>
?#include <linux/device.h>

+#define INTC_NR_IRQS ? 1024

Hi,

2012/1/14 Rob Herring <[email protected]>:

Quoted 63 lines

From: Rob Herring <[email protected]>

SH intc has a compile time dependency on NR_IRQS. Make this dependency a local define so that shmobile (and ARM in general) can have run-time NR_IRQS setting. SH has NR_IRQS set to 512 and shmobile has NR_IRQS set to 1024, so we are using the maximum.

Signed-off-by Rob Herring <[email protected]>

---
 drivers/sh/intc/balancing.c |    2 +-
 drivers/sh/intc/core.c      |    2 +-
 drivers/sh/intc/handle.c    |    2 +-
 drivers/sh/intc/internals.h |    9 +++++++++
 drivers/sh/intc/virq.c      |    2 +-
 5 files changed, 13 insertions(+), 4 deletions(-)

drivers/sh/intc/balancing.c +1 -1

index cec7a96..bc78080 100644
@@ -9,7 +9,7 @@
 */
 #include "internals.h"
 
-static unsigned long dist_handle[NR_IRQS];
+static unsigned long dist_handle[INTC_NR_IRQS];
 
 void intc_balancing_enable(unsigned int irq)
 {

drivers/sh/intc/core.c +1 -1

index e53e449..2fde897 100644
@@ -42,7 +42,7 @@ unsigned int nr_intc_controllers;
 * - this needs to be at least 2 for 5-bit priorities on 7780
 */
 static unsigned int default_prio_level = 2;    /* 2 - 16 */
-static unsigned int intc_prio_level[NR_IRQS];  /* for now */
+static unsigned int intc_prio_level[INTC_NR_IRQS];     /* for now */
 
 unsigned int intc_get_dfl_prio_level(void)
 {

drivers/sh/intc/handle.c +1 -1

index 057ce56..f461d53 100644
@@ -13,7 +13,7 @@
 #include <linux/spinlock.h>
 #include "internals.h"
 
-static unsigned long ack_handle[NR_IRQS];
+static unsigned long ack_handle[INTC_NR_IRQS];
 
 static intc_enum __init intc_grp_id(struct intc_desc *desc,
                                   intc_enum enum_id)

drivers/sh/intc/internals.h +0 -0

index b0e9155..469f092 100644
@@ -6,6 +6,15 @@
 #include <linux/radix-tree.h>
 #include <linux/device.h>
 

drivers/sh/intc/internals.h line 9

Quoted 1 line

drivers/sh/intc/internals.h +1 -0

+#define INTC_NR_IRQS   1024

On SH, INTC_NR_IRQS ( NR_IRQS ) is using to by arch/sh/kernel/machvec.c. And, this is defined by arch/sh/include/asm/irq.h.
You need to remove or rename from these.

drivers/sh/intc/internals.h line 17

Quoted 8 lines

drivers/sh/intc/internals.h +8 -0

+
+#ifndef evt2irq
+#define evt2irq(evt)            (((evt) >> 5) - 16)
+#endif
+#ifndef irq2evt
+#define irq2evt(irq)            (((irq) + 16) << 5)
+#endif
+

These are defined in arch/arm/mach-shmobile/include/mach/irqs.h and arch/sh/include/asm/irq.h.

I propose that linux/sh_intc.h defines these.

Best regards,
Nobuhiro

Quoted 66 lines

Hi,

2012/1/14 Rob Herring <[email protected]>:

Quoted 63 lines

From: Rob Herring <[email protected]>

SH intc has a compile time dependency on NR_IRQS. Make this dependency a local define so that shmobile (and ARM in general) can have run-time NR_IRQS setting. SH has NR_IRQS set to 512 and shmobile has NR_IRQS set to 1024, so we are using the maximum.

Signed-off-by Rob Herring <[email protected]>

---

?drivers/sh/intc/balancing.c | ? ?2 +-
?drivers/sh/intc/core.c ? ? ?| ? ?2 +-
?drivers/sh/intc/handle.c ? ?| ? ?2 +-
?drivers/sh/intc/internals.h | ? ?9 +++++++++
?drivers/sh/intc/virq.c ? ? ?| ? ?2 +-

?5 files changed, 13 insertions(+), 4 deletions(-)

drivers/sh/intc/balancing.c +0 -0

index cec7a96..bc78080 100644
@@ -9,7 +9,7 @@

?*/
?#include "internals.h"

drivers/sh/intc/balancing.c +1 -1

-static unsigned long dist_handle[NR_IRQS];
+static unsigned long dist_handle[INTC_NR_IRQS];

?void intc_balancing_enable(unsigned int irq)
?{

drivers/sh/intc/core.c +0 -0

index e53e449..2fde897 100644
@@ -42,7 +42,7 @@ unsigned int nr_intc_controllers;

?* - this needs to be at least 2 for 5-bit priorities on 7780 ?*/
?static unsigned int default_prio_level = 2; ? ?/* 2 - 16 */

drivers/sh/intc/core.c +1 -1

-static unsigned int intc_prio_level[NR_IRQS]; ?/* for now */
+static unsigned int intc_prio_level[INTC_NR_IRQS]; ? ? /* for now */

?unsigned int intc_get_dfl_prio_level(void)
?{

drivers/sh/intc/handle.c +0 -0

index 057ce56..f461d53 100644
@@ -13,7 +13,7 @@

?#include <linux/spinlock.h>
?#include "internals.h"

drivers/sh/intc/handle.c +1 -1

-static unsigned long ack_handle[NR_IRQS];
+static unsigned long ack_handle[INTC_NR_IRQS];

?static intc_enum __init intc_grp_id(struct intc_desc *desc, ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?intc_enum enum_id)

drivers/sh/intc/internals.h +0 -0

index b0e9155..469f092 100644
@@ -6,6 +6,15 @@

?#include <linux/radix-tree.h>
?#include <linux/device.h>

Quoted 6 lines

Quoted 1 line

+#define INTC_NR_IRQS ? 1024

On SH, INTC_NR_IRQS ( NR_IRQS ) ?is using to by arch/sh/kernel/machvec.c. And, this is defined by arch/sh/include/asm/irq.h.
You need to remove or rename from these.

machvec.c will still pickup NR_IRQS from arch/sh/include/asm/irq.h, so there is no change there. The value here is increased from 512 for SH to 1024, but that should not have any functional impact only array storage space. Where it is used is a bit of a hack as the comment indicates. The only other easy way to fix it I see is with an #ifdef CONFIG_SH or CONFIG_ARM here. I welcome patches if you've got better ideas.

2012/1/17 Rob Herring <[email protected]>:

Quoted 74 lines · Nobuhiro Iwamatsu

Quoted 66 lines

Hi,

2012/1/14 Rob Herring <[email protected]>:

Quoted 63 lines

From: Rob Herring <[email protected]>

SH intc has a compile time dependency on NR_IRQS. Make this dependency a local define so that shmobile (and ARM in general) can have run-time NR_IRQS setting. SH has NR_IRQS set to 512 and shmobile has NR_IRQS set to 1024, so we are using the maximum.

Signed-off-by Rob Herring <[email protected]>

---
 drivers/sh/intc/balancing.c |    2 +-
 drivers/sh/intc/core.c      |    2 +-
 drivers/sh/intc/handle.c    |    2 +-
 drivers/sh/intc/internals.h |    9 +++++++++
 drivers/sh/intc/virq.c      |    2 +-
 5 files changed, 13 insertions(+), 4 deletions(-)

drivers/sh/intc/balancing.c +1 -1

index cec7a96..bc78080 100644
@@ -9,7 +9,7 @@
 */
 #include "internals.h"
 
-static unsigned long dist_handle[NR_IRQS];
+static unsigned long dist_handle[INTC_NR_IRQS];
 
 void intc_balancing_enable(unsigned int irq)
 {

drivers/sh/intc/core.c +1 -1

index e53e449..2fde897 100644
@@ -42,7 +42,7 @@ unsigned int nr_intc_controllers;
 * - this needs to be at least 2 for 5-bit priorities on 7780
 */
 static unsigned int default_prio_level = 2;    /* 2 - 16 */
-static unsigned int intc_prio_level[NR_IRQS];  /* for now */
+static unsigned int intc_prio_level[INTC_NR_IRQS];     /* for now */
 
 unsigned int intc_get_dfl_prio_level(void)
 {

drivers/sh/intc/handle.c +1 -1

index 057ce56..f461d53 100644
@@ -13,7 +13,7 @@
 #include <linux/spinlock.h>
 #include "internals.h"
 
-static unsigned long ack_handle[NR_IRQS];
+static unsigned long ack_handle[INTC_NR_IRQS];
 
 static intc_enum __init intc_grp_id(struct intc_desc *desc,
                                   intc_enum enum_id)

drivers/sh/intc/internals.h +0 -0

index b0e9155..469f092 100644
@@ -6,6 +6,15 @@
 #include <linux/radix-tree.h>
 #include <linux/device.h>
 

Quoted 6 lines

drivers/sh/intc/internals.h line 9

Quoted 1 line

drivers/sh/intc/internals.h +1 -0

+#define INTC_NR_IRQS   1024

On SH, INTC_NR_IRQS ( NR_IRQS ) is using to by arch/sh/kernel/machvec.c. And, this is defined by arch/sh/include/asm/irq.h.
You need to remove or rename from these.

Quoted 7 lines

machvec.c will still pickup NR_IRQS from arch/sh/include/asm/irq.h, so there is no change there. The value here is increased from 512 for SH to 1024, but that should not have any functional impact only array storage space. Where it is used is a bit of a hack as the comment indicates. The only other easy way to fix it I see is with an #ifdef CONFIG_SH or CONFIG_ARM here. I welcome patches if you've got better ideas.

I also cared about the problem from which array increases in SH. Since drivers/sh/intc/internals.h is referred to only from intc function, it needs to define INTC_NR_IRQS as other places.

How is it that defines by Kconfig?
I created a patch on your patch. Could you give comment?

Best regards,
Nobuhiro

------

arch/sh/include/asm/irq.h +0 -1

index 45d08b6..9278bb0 100644
@@ -9,7 +9,6 @@
  * advised to cap this at the hard limit that they're interested in
  * through the machvec.
  */
-#define NR_IRQS			512
 #define NR_IRQS_LEGACY		8	/* Legacy external IRQ0-7 */
 
 /*

arch/sh/kernel/machvec.c +1 -1

index 3d722e4..e6b12b4 100644
@@ -123,5 +123,5 @@ void __init sh_mv_setup(void)
 	mv_set(mem_init);
 
 	if (!sh_mv.mv_nr_irqs)
-		sh_mv.mv_nr_irqs = NR_IRQS;
+		sh_mv.mv_nr_irqs = CONFIG_SH_NR_IRQS;
 }

drivers/sh/intc/Kconfig +6 -0

index c88cbcc..38d51f1 100644
@@ -33,3 +33,9 @@ config INTC_MAPPING_DEBUG
 	  between system IRQs and the per-controller id tables.
 
 	  If in doubt, say N.
+
+config SH_NR_IRQS
+	int
+	depends on ARCH_SHMOBILE || SUPERH
+	default 1024 if ARCH_SHMOBILE
+	default 512 if SUPERH

drivers/sh/intc/internals.h +0 -0

index 469f092..076c286 100644

Quoted 142 lines

2012/1/17 Rob Herring <[email protected]>:

Quoted 74 lines · Nobuhiro Iwamatsu

Quoted 66 lines

Hi,

2012/1/14 Rob Herring <[email protected]>:

Quoted 63 lines

From: Rob Herring <[email protected]>

SH intc has a compile time dependency on NR_IRQS. Make this dependency a local define so that shmobile (and ARM in general) can have run-time NR_IRQS setting. SH has NR_IRQS set to 512 and shmobile has NR_IRQS set to 1024, so we are using the maximum.

Signed-off-by Rob Herring <[email protected]>

---

?drivers/sh/intc/balancing.c | ? ?2 +-
?drivers/sh/intc/core.c ? ? ?| ? ?2 +-
?drivers/sh/intc/handle.c ? ?| ? ?2 +-
?drivers/sh/intc/internals.h | ? ?9 +++++++++
?drivers/sh/intc/virq.c ? ? ?| ? ?2 +-

?5 files changed, 13 insertions(+), 4 deletions(-)

drivers/sh/intc/balancing.c +0 -0

index cec7a96..bc78080 100644
@@ -9,7 +9,7 @@

?*/
?#include "internals.h"

drivers/sh/intc/balancing.c +1 -1

-static unsigned long dist_handle[NR_IRQS];
+static unsigned long dist_handle[INTC_NR_IRQS];

?void intc_balancing_enable(unsigned int irq)
?{

drivers/sh/intc/core.c +0 -0

index e53e449..2fde897 100644
@@ -42,7 +42,7 @@ unsigned int nr_intc_controllers;

?* - this needs to be at least 2 for 5-bit priorities on 7780 ?*/
?static unsigned int default_prio_level = 2; ? ?/* 2 - 16 */

drivers/sh/intc/core.c +1 -1

-static unsigned int intc_prio_level[NR_IRQS]; ?/* for now */
+static unsigned int intc_prio_level[INTC_NR_IRQS]; ? ? /* for now */

?unsigned int intc_get_dfl_prio_level(void)
?{

drivers/sh/intc/handle.c +0 -0

index 057ce56..f461d53 100644
@@ -13,7 +13,7 @@

?#include <linux/spinlock.h>
?#include "internals.h"

drivers/sh/intc/handle.c +1 -1

-static unsigned long ack_handle[NR_IRQS];
+static unsigned long ack_handle[INTC_NR_IRQS];

?static intc_enum __init intc_grp_id(struct intc_desc *desc, ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?intc_enum enum_id)

drivers/sh/intc/internals.h +0 -0

index b0e9155..469f092 100644
@@ -6,6 +6,15 @@

?#include <linux/radix-tree.h>
?#include <linux/device.h>

Quoted 6 lines

Quoted 1 line

+#define INTC_NR_IRQS ? 1024

On SH, INTC_NR_IRQS ( NR_IRQS ) ?is using to by arch/sh/kernel/machvec.c. And, this is defined by arch/sh/include/asm/irq.h.
You need to remove or rename from these.

Quoted 7 lines

machvec.c will still pickup NR_IRQS from arch/sh/include/asm/irq.h, so there is no change there. The value here is increased from 512 for SH to 1024, but that should not have any functional impact only array storage space. Where it is used is a bit of a hack as the comment indicates. The only other easy way to fix it I see is with an #ifdef CONFIG_SH or CONFIG_ARM here. I welcome patches if you've got better ideas.

I also cared about the problem from which array increases in SH. Since drivers/sh/intc/internals.h is referred to only from intc function, it needs to define INTC_NR_IRQS as other places.

How is it that defines by Kconfig?
I created a patch on your patch. Could you give comment?

Best regards,
? Nobuhiro

------

arch/sh/include/asm/irq.h +0 -0

index 45d08b6..9278bb0 100644
@@ -9,7 +9,6 @@

? * advised to cap this at the hard limit that they're interested in ? * through the machvec.
? */
-#define NR_IRQS ? ? ? ? ? ? ? ? ? ? ?512
?#define NR_IRQS_LEGACY ? ? ? ? ? ? ? 8 ? ? ? /* Legacy external IRQ0-7 */

?/*

arch/sh/kernel/machvec.c +1 -1

index 3d722e4..e6b12b4 100644
@@ -123,5 +123,5 @@ void __init sh_mv_setup(void)
 ? ? ? mv_set(mem_init);
 
 ? ? ? if (!sh_mv.mv_nr_irqs)
- ? ? ? ? ? ? sh_mv.mv_nr_irqs = NR_IRQS;
+ ? ? ? ? ? ? sh_mv.mv_nr_irqs = CONFIG_SH_NR_IRQS;
 ?}

drivers/sh/intc/Kconfig +0 -0

index c88cbcc..38d51f1 100644
@@ -33,3 +33,9 @@ config INTC_MAPPING_DEBUG

? ? ? ? between system IRQs and the per-controller id tables.

? ? ? ? If in doubt, say N.

drivers/sh/intc/Kconfig +6 -0

+
+config SH_NR_IRQS
+ ? ? int
+ ? ? depends on ARCH_SHMOBILE || SUPERH
+ ? ? default 1024 if ARCH_SHMOBILE
+ ? ? default 512 if SUPERH

drivers/sh/intc/internals.h +0 -0

index 469f092..076c286 100644
@@ -6,7 +6,7 @@

?#include <linux/radix-tree.h>
?#include <linux/device.h>

drivers/sh/intc/internals.h +1 -1

-#define INTC_NR_IRQS 1024
+#define INTC_NR_IRQS CONFIG_SH_NR_IRQS

Quoted 3 lines

?#ifndef evt2irq
?#define evt2irq(evt) ? ? ? ? ? ?(((evt) >> 5) - 16)

If we went this route I wonder if it would be better for this to be more generic and have CONFIG_NR_IRQS like powerpc. However, I don't see having CONFIG_NR_IRQS as being that useful in the SPARSE_IRQ case. Plus it would be a much more invasive. I think I'll just add this to linux/sh_intc.h:

#ifdef CONFIG_SUPERH
#define INTC_NR_IRQS ? 512
#else
#define INTC_NR_IRQS ? 1024
#endif

From: Rob Herring <[email protected]>

Remove NR_IRQS and add a per machine .nr_irqs setting.

Signed-off-by Rob Herring <[email protected]>

---
 arch/arm/mach-mmp/avengers_lite.c     |    1 +
 arch/arm/mach-mmp/include/mach/irqs.h |    2 --
 arch/arm/mach-mmp/irq-mmp2.c          |    1 +
 arch/arm/mach-mmp/jasper.c            |    1 +
 arch/arm/mach-mmp/tavorevb.c          |    1 +
 5 files changed, 4 insertions(+), 2 deletions(-)

arch/arm/mach-mmp/avengers_lite.c +1 -0

index b148a9d..0fd60fb 100644
@@ -43,6 +43,7 @@ static void __init avengers_lite_init(void)
 
 MACHINE_START(AVENGERS_LITE, "PXA168 Avengers lite Development Platform")
 	.map_io		= mmp_map_io,
+	.nr_irqs	= IRQ_BOARD_START,
 	.init_irq       = pxa168_init_irq,
 	.timer          = &pxa168_timer,
 	.init_machine   = avengers_lite_init,

arch/arm/mach-mmp/include/mach/irqs.h +0 -2

index 34635a0..25e6d60 100644
@@ -224,6 +224,4 @@
 
 #define IRQ_BOARD_START			(IRQ_GPIO_START + MMP_NR_BUILTIN_GPIO)
 
-#define NR_IRQS				(IRQ_BOARD_START)
-
 #endif /* __ASM_MACH_IRQS_H */

arch/arm/mach-mmp/irq-mmp2.c +1 -0

index d21c544..7895d27 100644
@@ -15,6 +15,7 @@
 #include <linux/irq.h>
 #include <linux/io.h>
 
+#include <mach/irqs.h>
 #include <mach/regs-icu.h>
 #include <mach/mmp2.h>
 

arch/arm/mach-mmp/jasper.c +1 -0

index 96cf5c8..8f80c31 100644
@@ -19,6 +19,7 @@
 #include <linux/mfd/max8925.h>
 #include <linux/interrupt.h>
 
+#include <mach/irqs.h>
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
 #include <mach/addr-map.h>

arch/arm/mach-mmp/tavorevb.c +1 -0

index 8e3b5af0..d3ba655 100644
@@ -102,6 +102,7 @@ static void __init tavorevb_init(void)
 
 MACHINE_START(TAVOREVB, "PXA910 Evaluation Board (aka TavorEVB)")
 	.map_io		= mmp_map_io,
+	.nr_irqs	= IRQ_BOARD_START,
 	.init_irq       = pxa910_init_irq,
 	.timer          = &pxa910_timer,
 	.init_machine   = tavorevb_init,

Signature · 1 line

1.7.5.4

From: Rob Herring <[email protected]>

This series removes the include of mach/irqs.h from asm/irq.h when SPARSE_IRQ is selected. Platforms which select SPARSE_IRQ have been updated.

This breaks platforms where SPARSE_IRQ is user selectable. This is on purpose so that SPARSE_IRQ gets properly supported. We could push HAVE_SPARSE_IRQ option down to the platforms that select it if that is preferred. But as recent experience has shown, perhaps breaking platforms is the only way to get some platforms updated.

From 2d07b017bb06be86bd01686f98c699f1eff0ac10 Mon Sep 17 00:00:00 2001