Skip to contentSkip to conversation

of/pci: add pci_pio_to_address dummy for !CONFIG_OF

PATCH In review Arnd Bergmann 4 participants 10 messages Sep 30, 2014 – Oct 1, 2014

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 of/pci: add pci_pio_to_address dummy for !CONFIG_OF

The pci_register_io_range() and pci_pio_to_address() were recently introduced to generalize the handling of memory mapped PCI I/O space, but they are only valid when CONFIG_OF is set, leading to a possible build error:

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_setup_window':
drivers/pci/host/pcie-rcar.c:340:3: error: implicit declaration of function 'pci_pio_to_address' [-Werror=implicit-function-declaration]
res_start = pci_pio_to_address(res->start);
^
drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_probe':
drivers/pci/host/pcie-rcar.c:945:3: error: implicit declaration of function 'of_pci_range_to_resource' [-Werror=implicit-function-declaration]
err = of_pci_range_to_resource(&range, pdev->dev.of_node,
^
cc1: some warnings being treated as errors

This provides inline dummy implementations for the case that CONFIG_OF is disabled, to allow better build testing.

Fixes 279c5dd046 ("of/pci: Add pci_register_io_range() and pci_pio_to_address()")
include/linux/of_address.h +11 -0
index 7ebb877b07c2..851097aab115 100644
@@ -71,6 +71,11 @@ static inline const __be32 *of_get_address(struct device_node *dev, int index,
 	return NULL;
 }
 
+static inline phys_addr_t pci_pio_to_address(unsigned long pio)
+{
+	return 0;
+}
+
 static inline int of_pci_range_parser_init(struct of_pci_range_parser *parser,
 			struct device_node *node)
 {
@@ -144,6 +149,12 @@ static inline const __be32 *of_get_pci_address(struct device_node *dev,
 {
 	return NULL;
 }
+static inline int of_pci_range_to_resource(struct of_pci_range *range,
+					   struct device_node *np,
+					   struct resource *res)
+{
+	return -ENOSYS;
+}
 #endif /* CONFIG_OF_ADDRESS && CONFIG_PCI */
 
 #endif /* __OF_ADDRESS_H */
Acked-by on v1 1/1
#1 lore raw
Quoted 15 lines · Arnd Bergmann

The pci_register_io_range() and pci_pio_to_address() were recently introduced to generalize the handling of memory mapped PCI I/O space, but they are only valid when CONFIG_OF is set, leading to a possible build error:

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_setup_window':
drivers/pci/host/pcie-rcar.c:340:3: error: implicit declaration of function 'pci_pio_to_address' [-Werror=implicit-function-declaration]

   res_start = pci_pio_to_address(res->start);
   ^

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_probe':
drivers/pci/host/pcie-rcar.c:945:3: error: implicit declaration of function 'of_pci_range_to_resource' [-Werror=implicit-function-declaration]
err = of_pci_range_to_resource(&range, pdev->dev.of_node,
^
cc1: some warnings being treated as errors

Quoted 5 lines

This provides inline dummy implementations for the case that CONFIG_OF is disabled, to allow better build testing.

Fixes 279c5dd046 ("of/pci: Add pci_register_io_range() and pci_pio_to_address()")

Is that triggered by shmobile_defconfig?

Best regards,
Liviu

Quoted 33 lines
include/linux/of_address.h +11 -0
index 7ebb877b07c2..851097aab115 100644
@@ -71,6 +71,11 @@ static inline const __be32 *of_get_address(struct device_node *dev, int index,
 	return NULL;
 }
 
+static inline phys_addr_t pci_pio_to_address(unsigned long pio)
+{
+	return 0;
+}
+
 static inline int of_pci_range_parser_init(struct of_pci_range_parser *parser,
 			struct device_node *node)
 {
@@ -144,6 +149,12 @@ static inline const __be32 *of_get_pci_address(struct device_node *dev,
 {
 	return NULL;
 }
+static inline int of_pci_range_to_resource(struct of_pci_range *range,
+					   struct device_node *np,
+					   struct resource *res)
+{
+	return -ENOSYS;
+}
 #endif /* CONFIG_OF_ADDRESS && CONFIG_PCI */
 
 #endif /* __OF_ADDRESS_H */
Signature · 8 lines
====================
| I would like to |
| fix the world,  |
| but they're not |
| giving me the   |
 \ source code!  /
  ---------------
    ?\_(?)_/?
#2 lore raw
Quoted 6 lines · On Tuesday 30 September 2014 15:45:06 Liviu Dudau wrote:

Is that triggered by shmobile_defconfig?

No, I only found it using randconfig testing, shmobile enables CONFIG_OF by default.

Arnd

#3 lore raw
Quoted 10 lines · Arnd Bergmann
Quoted 6 lines · On Tuesday 30 September 2014 15:45:06 Liviu Dudau wrote:

Is that triggered by shmobile_defconfig?

No, I only found it using randconfig testing, shmobile enables CONFIG_OF by default.

Hmm, I was looking at who defines CONFIG_PCI_RCAR_GEN2_PCIE and it's only present in shmobile_defconfig, but that one does not select any OF options (other than CONFIG_ARM_APPENDED_DTB). So there is no way of triggering this other than by randconfig?

Best regards,
Liviu

Quoted 4 lines

Arnd

Signature · 8 lines
====================
| I would like to |
| fix the world,  |
| but they're not |
| giving me the   |
 \ source code!  /
  ---------------
    ?\_(?)_/?
#4 lore raw
Quoted 9 lines · On Tuesday 30 September 2014 16:42:02 Liviu Dudau wrote:
Quoted 8 lines · Arnd Bergmann
Quoted 6 lines · On Tuesday 30 September 2014 15:45:06 Liviu Dudau wrote:

Is that triggered by shmobile_defconfig?

Quoted 7 lines
Quoted 2 lines

No, I only found it using randconfig testing, shmobile enables CONFIG_OF by default.

Hmm, I was looking at who defines CONFIG_PCI_RCAR_GEN2_PCIE and it's only present in shmobile_defconfig, but that one does not select any OF options (other than CONFIG_ARM_APPENDED_DTB). So there is no way of triggering this other than by randconfig?

It should be easy enough to build a kernel without CONFIG_OF that enables this driver manually on another platform.

Arnd

maintainer
#5 lore raw
In reply to Arnd Bergmann
Quoted 15 lines · Arnd Bergmann

The pci_register_io_range() and pci_pio_to_address() were recently introduced to generalize the handling of memory mapped PCI I/O space, but they are only valid when CONFIG_OF is set, leading to a possible build error:

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_setup_window':
drivers/pci/host/pcie-rcar.c:340:3: error: implicit declaration of function 'pci_pio_to_address' [-Werror=implicit-function-declaration]

   res_start = pci_pio_to_address(res->start);
   ^

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_probe':
drivers/pci/host/pcie-rcar.c:945:3: error: implicit declaration of function 'of_pci_range_to_resource' [-Werror=implicit-function-declaration]
err = of_pci_range_to_resource(&range, pdev->dev.of_node,
^
cc1: some warnings being treated as errors

Quoted 5 lines

This provides inline dummy implementations for the case that CONFIG_OF is disabled, to allow better build testing.

Fixes 279c5dd046 ("of/pci: Add pci_register_io_range() and pci_pio_to_address()")

I'm rebuilding the pci/host-generic branch to pick up the other v13 fixes, so I folded this fix directly into the "of/pci: Add pci_register_io_range() and pci_pio_to_address()" patch. Thanks for finding this; the config dependencies are a bit of a mess.

Quoted 30 lines
include/linux/of_address.h +11 -0
index 7ebb877b07c2..851097aab115 100644
@@ -71,6 +71,11 @@ static inline const __be32 *of_get_address(struct device_node *dev, int index,
 	return NULL;
 }
 
+static inline phys_addr_t pci_pio_to_address(unsigned long pio)
+{
+	return 0;
+}
+
 static inline int of_pci_range_parser_init(struct of_pci_range_parser *parser,
 			struct device_node *node)
 {
@@ -144,6 +149,12 @@ static inline const __be32 *of_get_pci_address(struct device_node *dev,
 {
 	return NULL;
 }
+static inline int of_pci_range_to_resource(struct of_pci_range *range,
+					   struct device_node *np,
+					   struct resource *res)
+{
+	return -ENOSYS;
+}
 #endif /* CONFIG_OF_ADDRESS && CONFIG_PCI */
 
 #endif /* __OF_ADDRESS_H */
#6 lore raw
Quoted 19 lines · Bjorn Helgaas
Quoted 18 lines · Arnd Bergmann

The pci_register_io_range() and pci_pio_to_address() were recently introduced to generalize the handling of memory mapped PCI I/O space, but they are only valid when CONFIG_OF is set, leading to a possible build error:

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_setup_window':
drivers/pci/host/pcie-rcar.c:340:3: error: implicit declaration of function 'pci_pio_to_address' [-Werror=implicit-function-declaration]

   res_start = pci_pio_to_address(res->start);
   ^

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_probe':
drivers/pci/host/pcie-rcar.c:945:3: error: implicit declaration of function 'of_pci_range_to_resource' [-Werror=implicit-function-declaration]
err = of_pci_range_to_resource(&range, pdev->dev.of_node,
^
cc1: some warnings being treated as errors

This provides inline dummy implementations for the case that CONFIG_OF is disabled, to allow better build testing.

Quoted 7 lines
Quoted 2 lines
Fixes 279c5dd046 ("of/pci: Add pci_register_io_range() and pci_pio_to_address()")

I'm rebuilding the pci/host-generic branch to pick up the other v13 fixes, so I folded this fix directly into the "of/pci: Add pci_register_io_range() and pci_pio_to_address()" patch. Thanks for finding this; the config dependencies are a bit of a mess.

Bjorn,

I suggest splitting this patch into 2 parts, move the one that adds pci_pio_to_address() in patch 2 and the other in patch 5.

Best regards,
Liviu

Quoted 36 lines
Quoted 21 lines
include/linux/of_address.h +5 -0
index 7ebb877b07c2..851097aab115 100644
@@ -71,6 +71,11 @@ static inline const __be32 *of_get_address(struct device_node *dev, int index,
 	return NULL;
 }
 
+static inline phys_addr_t pci_pio_to_address(unsigned long pio)
+{
+	return 0;
+}
+
 static inline int of_pci_range_parser_init(struct of_pci_range_parser *parser,
 			struct device_node *node)
 {
@@ -144,6 +149,12 @@ static inline const __be32 *of_get_pci_address(struct device_node *dev,
 {
 	return NULL;

include/linux/of_address.h static inline const __be32 *of_get_pci_address(struct device_node *dev, · line 160

Quoted 10 lines
include/linux/of_address.h +6 -0
 }
+static inline int of_pci_range_to_resource(struct of_pci_range *range,
+					   struct device_node *np,
+					   struct resource *res)
+{
+	return -ENOSYS;
+}
 #endif /* CONFIG_OF_ADDRESS && CONFIG_PCI */
 
 #endif /* __OF_ADDRESS_H */

--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo at vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 10 lines
-------------------
   .oooO
   (   )
    \ (  Oooo.
     \_) (   )
          ) /
         (_/

 One small step
   for me ...
maintainer
#7 lore raw
Quoted 28 lines · Liviu Dudau
Quoted 19 lines · Bjorn Helgaas
Quoted 18 lines · Arnd Bergmann

The pci_register_io_range() and pci_pio_to_address() were recently introduced to generalize the handling of memory mapped PCI I/O space, but they are only valid when CONFIG_OF is set, leading to a possible build error:

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_setup_window':
drivers/pci/host/pcie-rcar.c:340:3: error: implicit declaration of function 'pci_pio_to_address' [-Werror=implicit-function-declaration]

   res_start = pci_pio_to_address(res->start);
   ^

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_probe':
drivers/pci/host/pcie-rcar.c:945:3: error: implicit declaration of function 'of_pci_range_to_resource' [-Werror=implicit-function-declaration]
err = of_pci_range_to_resource(&range, pdev->dev.of_node,
^
cc1: some warnings being treated as errors

This provides inline dummy implementations for the case that CONFIG_OF is disabled, to allow better build testing.

Quoted 7 lines
Quoted 2 lines
Fixes 279c5dd046 ("of/pci: Add pci_register_io_range() and pci_pio_to_address()")

I'm rebuilding the pci/host-generic branch to pick up the other v13 fixes, so I folded this fix directly into the "of/pci: Add pci_register_io_range() and pci_pio_to_address()" patch. Thanks for finding this; the config dependencies are a bit of a mess.

Quoted 4 lines

Bjorn,

I suggest splitting this patch into 2 parts, move the one that adds pci_pio_to_address() in patch 2 and the other in patch 5.

Yeah, I screwed that up. I had just applied Arnd's patch completely to 279c5dd046, which doesn't quite make sense.

I tried again like this:

- In "of/pci: Add pci_register_io_range() and pci_pio_to_address()",
I added the dummy pci_pio_to_address(), so there's an extern and a definition in address.c when CONFIG_OF_ADDRESS=y, and an inline dummy otherwise.

- I dropped "of/pci: Define of_pci_range_to_resource() only when
CONFIG_PCI=y".

- In "of/pci: Move of_pci_range_to_resource() to of/address.c", I
added the dummy of_pci_range_to_resource(), so there's an extern an a definition in address.c when CONFIG_OF_ADDRESS=y and CONFIG_PCI=y, and an inline dummy otherwise.

I'm not really set up to build test this, and my head is swimming a bit, so I hope this is it. I pushed the pci/host-generic branch again, so you can look at it yourself.

Bjorn

#8 lore raw
Quoted 47 lines · Bjorn Helgaas
Quoted 28 lines · Liviu Dudau
Quoted 19 lines · Bjorn Helgaas
Quoted 18 lines · Arnd Bergmann

The pci_register_io_range() and pci_pio_to_address() were recently introduced to generalize the handling of memory mapped PCI I/O space, but they are only valid when CONFIG_OF is set, leading to a possible build error:

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_setup_window':
drivers/pci/host/pcie-rcar.c:340:3: error: implicit declaration of function 'pci_pio_to_address' [-Werror=implicit-function-declaration]

   res_start = pci_pio_to_address(res->start);
   ^

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_probe':
drivers/pci/host/pcie-rcar.c:945:3: error: implicit declaration of function 'of_pci_range_to_resource' [-Werror=implicit-function-declaration]
err = of_pci_range_to_resource(&range, pdev->dev.of_node,
^
cc1: some warnings being treated as errors

This provides inline dummy implementations for the case that CONFIG_OF is disabled, to allow better build testing.

Quoted 7 lines
Quoted 2 lines
Fixes 279c5dd046 ("of/pci: Add pci_register_io_range() and pci_pio_to_address()")

I'm rebuilding the pci/host-generic branch to pick up the other v13 fixes, so I folded this fix directly into the "of/pci: Add pci_register_io_range() and pci_pio_to_address()" patch. Thanks for finding this; the config dependencies are a bit of a mess.

Quoted 4 lines

Bjorn,

I suggest splitting this patch into 2 parts, move the one that adds pci_pio_to_address() in patch 2 and the other in patch 5.

Yeah, I screwed that up. I had just applied Arnd's patch completely to 279c5dd046, which doesn't quite make sense.

I tried again like this:

- In "of/pci: Add pci_register_io_range() and pci_pio_to_address()",
I added the dummy pci_pio_to_address(), so there's an extern and a definition in address.c when CONFIG_OF_ADDRESS=y, and an inline dummy otherwise.

- I dropped "of/pci: Define of_pci_range_to_resource() only when
CONFIG_PCI=y".

Quoted 8 lines

- In "of/pci: Move of_pci_range_to_resource() to of/address.c", I
added the dummy of_pci_range_to_resource(), so there's an extern an a definition in address.c when CONFIG_OF_ADDRESS=y and CONFIG_PCI=y, and an inline dummy otherwise.

I'm not really set up to build test this, and my head is swimming a bit, so I hope this is it. I pushed the pci/host-generic branch again, so you can look at it yourself.

Hi Bjorn,

It looks good to me. Functionality-wise I can only test it tomorrow UK time, but from the diffs I've done I thinks we're good.

Will wait for more reports from the build bot.

Take care,
Liviu

Quoted 3 lines

Bjorn

#9 lore raw
In reply to Arnd Bergmann
Quoted 41 lines · Arnd Bergmann

The pci_register_io_range() and pci_pio_to_address() were recently introduced to generalize the handling of memory mapped PCI I/O space, but they are only valid when CONFIG_OF is set, leading to a possible build error:

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_setup_window':
drivers/pci/host/pcie-rcar.c:340:3: error: implicit declaration of function 'pci_pio_to_address' [-Werror=implicit-function-declaration]

   res_start = pci_pio_to_address(res->start);
   ^

drivers/pci/host/pcie-rcar.c: In function 'rcar_pcie_probe':
drivers/pci/host/pcie-rcar.c:945:3: error: implicit declaration of function 'of_pci_range_to_resource' [-Werror=implicit-function-declaration]
err = of_pci_range_to_resource(&range, pdev->dev.of_node,
^
cc1: some warnings being treated as errors

This provides inline dummy implementations for the case that CONFIG_OF is disabled, to allow better build testing.

Fixes 279c5dd046 ("of/pci: Add pci_register_io_range() and pci_pio_to_address()")
include/linux/of_address.h +5 -0
index 7ebb877b07c2..851097aab115 100644
@@ -71,6 +71,11 @@ static inline const __be32 *of_get_address(struct device_node *dev, int index,
 	return NULL;
 }
 
+static inline phys_addr_t pci_pio_to_address(unsigned long pio)
+{
+	return 0;
+}
+
 static inline int of_pci_range_parser_init(struct of_pci_range_parser *parser,
 			struct device_node *node)
 {
@@ -144,6 +149,12 @@ static inline const __be32 *of_get_pci_address(struct device_node *dev,
 {
 	return NULL;

include/linux/of_address.h static inline const __be32 *of_get_pci_address(struct device_node *dev, · line 160

Quoted 10 lines
include/linux/of_address.h +6 -0
 }
+static inline int of_pci_range_to_resource(struct of_pci_range *range,
+					   struct device_node *np,
+					   struct resource *res)
+{
+	return -ENOSYS;
+}
 #endif /* CONFIG_OF_ADDRESS && CONFIG_PCI */
 
 #endif /* __OF_ADDRESS_H */

Arnd,

I have a more general question that is related only vaguely to this patch but it was prompted by it:
given that this pcie-rcar.c driver is so dependent on the OF functionality, why the fix(es) (in general) tend to prefer creating these dummy functions for !CONFIG_OF rather than making CONFIG_PCI_RCAR_GEN2_PCIE dependent on CONFIG_OF? We are basically compiling a driver that we can guarantee it will fail when used without OF support and contribute to the overall size of the kernel. Given the presentation you have done at Linaro Connect last month on this topic, I started to wonder if there is a deeper API style agreement that is not apparent to me?

Thanks for your time,
Liviu

Signature · 8 lines
====================
| I would like to |
| fix the world,  |
| but they're not |
| giving me the   |
 \ source code!  /
  ---------------
    ?\_(?)_/?