Remove unneeded build directory traversals

When CONFIG_GPIOLIB is not set make will still descend into the gpio directory but nothing will be built. This produces unneeded build artifacts and messages in addition to slowing the build. Fix this here.

Signed-off-by Andrew F. Davis <[email protected]>

If the HSI core is built as a module hsi_boardinfo may still be built-in as its Kconfig type is bool, which can cause build issues. Fix this by building this code into the HSI core when enabled.

Reported-by kbuild test robot <[email protected]>

Signed-off-by Andrew F. Davis <[email protected]>

---
This build error seems to be due to Kconfig symbol CONFIG_HSI_BOARDINFO being a bool but depending on a tristate (CONFIG_HSI). This is normally okay when it is just a flag to enable a feature in source, but the helper code file hsi_boardinfo.c is built as a separate entity when enabled. This patch is probably how it was intended, and is more like how others do this kind of thing.

This patch should be applied before the parent patch:

When CONFIG_LGUEST is not set make will still descend into the lguest directory but nothing will be built. This produces unneeded build artifacts and messages in addition to slowing the build. Fix this here.

Signed-off-by Andrew F. Davis <[email protected]>

---
 drivers/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/Makefile +1 -1

index 19305e0..b178e2f 100644
@@ -122,7 +122,7 @@ obj-$(CONFIG_ACCESSIBILITY)	+= accessibility/
 obj-$(CONFIG_ISDN)		+= isdn/
 obj-$(CONFIG_EDAC)		+= edac/
 obj-$(CONFIG_EISA)		+= eisa/
-obj-y				+= lguest/
+obj-$(CONFIG_LGUEST)		+= lguest/
 obj-$(CONFIG_CPU_FREQ)		+= cpufreq/
 obj-$(CONFIG_CPU_IDLE)		+= cpuidle/
 obj-y				+= mmc/

Signature · 1 line

2.8.3

When CONFIG_NEW_LEDS is not set make will still descend into the leds directory but nothing will be built. This produces unneeded build artifacts and messages in addition to slowing the build. Fix this here.

Signed-off-by Andrew F. Davis <[email protected]>

---
  drivers/Makefile | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

drivers/Makefile +1 -1

index 567e32c..fa514d5 100644
@@ -127,7 +127,7 @@ obj-$(CONFIG_CPU_FREQ)		+= cpufreq/
  obj-$(CONFIG_CPU_IDLE)		+= cpuidle/
  obj-$(CONFIG_MMC)		+= mmc/
  obj-$(CONFIG_MEMSTICK)		+= memstick/
-obj-y				+= leds/
+obj-$(CONFIG_NEW_LEDS)		+= leds/
  obj-$(CONFIG_INFINIBAND)	+= infiniband/
  obj-$(CONFIG_SGI_SN)		+= sn/
  obj-y				+= firmware/

Hi Andrew,

Thanks for the patch.

Please address the issue [1] raised by test bot and resubmit.

Thanks,
Jacek Anaszewski

[1] https://lkml.org/lkml/2016/6/13/1091

Quoted 23 lines

When CONFIG_NEW_LEDS is not set make will still descend into the leds directory but nothing will be built. This produces unneeded build artifacts and messages in addition to slowing the build. Fix this here.

Signed-off-by Andrew F. Davis <[email protected]>

---
  drivers/Makefile | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

drivers/Makefile +1 -1

index 567e32c..fa514d5 100644
@@ -127,7 +127,7 @@ obj-$(CONFIG_CPU_FREQ)        += cpufreq/
  obj-$(CONFIG_CPU_IDLE)        += cpuidle/
  obj-$(CONFIG_MMC)        += mmc/
  obj-$(CONFIG_MEMSTICK)        += memstick/
-obj-y                += leds/
+obj-$(CONFIG_NEW_LEDS)        += leds/
  obj-$(CONFIG_INFINIBAND)    += infiniband/
  obj-$(CONFIG_SGI_SN)        += sn/
  obj-y                += firmware/

Quoted 11 lines · Jacek Anaszewski

Hi Andrew,

Thanks for the patch.

Please address the issue [1] raised by test bot and resubmit.

Thanks,
Jacek Anaszewski

[1] https://lkml.org/lkml/2016/6/13/1091

It looks like some systems use 'gpio_led_register_device' to make an in-memory copy of their LED device table so the original can be removed as .init.rodata. This doesn't necessarily depend on the LED subsystem but it kind of seems useless when the rest of the subsystem is disabled.

One solution could be to use a dummy 'gpio_led_register_device' when the subsystem is not enabled.

Another is just to remove the five or so uses
of 'gpio_led_register_device' and have those systems register LED device tables like other systems do.

If nether of these are acceptable then this patch can be dropped from this series for now.

Thanks,
Andrew

Quoted 26 lines · Andrew F. Davis

Quoted 23 lines

When CONFIG_NEW_LEDS is not set make will still descend into the leds directory but nothing will be built. This produces unneeded build artifacts and messages in addition to slowing the build. Fix this here.

Signed-off-by Andrew F. Davis <[email protected]>

---
   drivers/Makefile | 2 +-
   1 file changed, 1 insertion(+), 1 deletion(-)

drivers/Makefile +1 -1

index 567e32c..fa514d5 100644
@@ -127,7 +127,7 @@ obj-$(CONFIG_CPU_FREQ)        += cpufreq/
   obj-$(CONFIG_CPU_IDLE)        += cpuidle/
   obj-$(CONFIG_MMC)        += mmc/
   obj-$(CONFIG_MEMSTICK)        += memstick/
-obj-y                += leds/
+obj-$(CONFIG_NEW_LEDS)        += leds/
   obj-$(CONFIG_INFINIBAND)    += infiniband/
   obj-$(CONFIG_SGI_SN)        += sn/
   obj-y                += firmware/

Some systems use 'gpio_led_register_device' to make an in-memory copy of their LED device table so the original can be removed as .init.rodata. When the LED subsystem is not enabled source in the led directory is not built and so this function may be undefined. Fix this here.

Signed-off-by Andrew F. Davis <[email protected]>

---
  include/linux/leds.h | 8 ++++++++
  1 file changed, 8 insertions(+)

include/linux/leds.h +0 -0

index d2b1306..a4a3da6 100644

include/linux/leds.h +8 -0

  };
 
+#ifdef CONFIG_NEW_LEDS
  struct platform_device *gpio_led_register_device(
                 int id, const struct gpio_led_platform_data *pdata);
+#else
+static inline struct platform_device *gpio_led_register_device(
+               int id, const struct gpio_led_platform_data *pdata)
+{
+       return 0;
+}
+#endif
 
  enum cpu_led_event {
         CPU_LED_IDLE_START,     /* CPU enters idle */

Hi Andrew,

This patch doesn't apply, please rebase onto recent LED tree.

Quoted 15 lines · Andrew F. Davis

Some systems use 'gpio_led_register_device' to make an in-memory copy of their LED device table so the original can be removed as .init.rodata. When the LED subsystem is not enabled source in the led directory is not built and so this function may be undefined. Fix this here.

Signed-off-by Andrew F. Davis <[email protected]>

---
  include/linux/leds.h | 8 ++++++++
  1 file changed, 8 insertions(+)

include/linux/leds.h +0 -0

index d2b1306..a4a3da6 100644

include/linux/leds.h struct gpio_led_platform_data { · line 386

Quoted 1 line

include/linux/leds.h +0 -0

@@ -386,8 +386,16 @@ struct gpio_led_platform_data {
                                         unsigned long *delay_off);

Currently there is some stuff here, and in fact it has been for a long time.

Patch "[PATCH 12/12] leds: Only descend into leds directory when CONFIG_NEW_LEDS is set" also doesn't apply.
What repository are you using?

Quoted 13 lines

include/linux/leds.h +8 -0

  };
 
+#ifdef CONFIG_NEW_LEDS
  struct platform_device *gpio_led_register_device(
                 int id, const struct gpio_led_platform_data *pdata);
+#else
+static inline struct platform_device *gpio_led_register_device(
+               int id, const struct gpio_led_platform_data *pdata)
+{
+       return 0;
+}
+#endif
 

include/linux/leds.h struct gpio_led_platform_data { · line 401

Quoted 3 lines

include/linux/leds.h +0 -0

  enum cpu_led_event {
         CPU_LED_IDLE_START,     /* CPU enters idle */

Quoted 25 lines · Jacek Anaszewski

Hi Andrew,

This patch doesn't apply, please rebase onto recent LED tree.

Quoted 15 lines · Andrew F. Davis

Some systems use 'gpio_led_register_device' to make an in-memory copy of their LED device table so the original can be removed as .init.rodata. When the LED subsystem is not enabled source in the led directory is not built and so this function may be undefined. Fix this here.

Signed-off-by Andrew F. Davis <[email protected]>

---
   include/linux/leds.h | 8 ++++++++
   1 file changed, 8 insertions(+)

include/linux/leds.h +0 -0

index d2b1306..a4a3da6 100644

include/linux/leds.h struct gpio_led_platform_data { · line 386

Quoted 1 line

include/linux/leds.h +0 -0

@@ -386,8 +386,16 @@ struct gpio_led_platform_data {
                                          unsigned long *delay_off);

Currently there is some stuff here, and in fact it has been for a long time.

Quoted 4 lines

Patch "[PATCH 12/12] leds: Only descend into leds directory when CONFIG_NEW_LEDS is set" also doesn't apply.
What repository are you using?

v4.7-rc4, it may not apply due to the surrounding lines being changed in the other patches which may not be applied to your tree. It is a single line change per patch so hopefully the merge conflict resolutions will be trivial.

A better solution could have been getting an ack from each maintainer and having someone pull the whole series into one tree, but parts have already been picked so it may be a little late for that.

Quoted 18 lines

Quoted 16 lines

include/linux/leds.h +8 -0

   };
 
+#ifdef CONFIG_NEW_LEDS
   struct platform_device *gpio_led_register_device(
                  int id, const struct gpio_led_platform_data *pdata);
+#else
+static inline struct platform_device *gpio_led_register_device(
+               int id, const struct gpio_led_platform_data *pdata)
+{
+       return 0;
+}
+#endif
 
   enum cpu_led_event {
          CPU_LED_IDLE_START,     /* CPU enters idle */