[PATCH 0/5] PM / devfreq: remove modular references from non-modules

[PATCH 0/5] PM / devfreq: remove modular references from non-modules

[Paul Gortmaker]

For anyone new to the underlying goal of this cleanup, we are trying to
not use module support for code that can never be built as a module since:

 (1) it is easy to accidentally write unused module_exit and remove code
 (2) it can be misleading when reading the source, thinking it can be
     modular when the Makefile and/or Kconfig prohibit it
 (3) it requires the include of the module.h header file which in turn
     includes nearly everything else, thus adding to CPP overhead.
 (4) it gets copied/replicated into other code and spreads like weeds.

We have already merged over 100 of these for mainline to date, so there
is really nothing new to see here, in terms of the type of change.

That said, devfreq changes seen here cover the following categories:

  -just replacement of modular macros with their non-modular
   equivalents that CPP would have inserted anyway

  -the removal of including module.h ; replaced with init.h
   and/or export.h as required based on whether the file used it.

  -the removal of any/all unused/orphaned __exit functions that
   would never be called.

  -the removal of any ".remove" functions that were hooked into
   the driver struct.   This ".remove" function would of
   course not be called from the __exit function since that was
   never run.  However in theory, someone could have triggered it
   via sysfs unbind, even though there isn't a sensible use case
   for doing so.  So to cover that possibility, we've also disabled
   sysfs unbind in the driver.

There are no initcall level changes here; everything stays at the level
of initcall it was previously - either by not using modular versions to
begin with, or by using the builtin level equivalents.

Build tested for arm and arm64 allmodconfig (for which all the drivers
touched here get coverage) on the linux-next tree from today to ensure
no silly typos crept in.

If there is a desire for any of these to be modular, we can definitely
consider that, but by default the changes here keep the code consistent
with existing behaviour.  Thus I do not expand functionality into the
modular realm that I can't run time test, or even know if the modular
instance has a sensible modular use case.

Paul.
---

Cc: Chanwoo Choi <cw00.choi@samsung.com>
Cc: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Cc: Kukjin Kim <kgene@kernel.org>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: MyungJoo Ham <myungjoo.ham@samsung.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-samsung-soc@vger.kernel.org


Paul Gortmaker (5):
  PM / devfreq: make devfreq explicitly non-modular
  PM / devfreq: make devfreq-event explicitly non-modular
  PM / devfreq: make exynos-bus explicitly non-modular
  PM / devfreq: make event/exynos-nocp explicitly non-modular
  PM / devfreq: make event/exynos-ppmu explicitly non-modular

 drivers/devfreq/devfreq-event.c     | 12 +-----------
 drivers/devfreq/devfreq.c           | 13 +------------
 drivers/devfreq/event/exynos-nocp.c | 19 +++----------------
 drivers/devfreq/event/exynos-ppmu.c |  8 ++------
 drivers/devfreq/exynos-bus.c        |  9 ++-------
 5 files changed, 9 insertions(+), 52 deletions(-)

-- 
2.8.4

[PATCH 3/5] PM / devfreq: make exynos-bus explicitly non-modular

[Paul Gortmaker]

The Kconfig currently controlling compilation of this code is:

devfreq/Kconfig:config ARM_EXYNOS_BUS_DEVFREQ
devfreq/Kconfig:        bool "ARM EXYNOS Generic Memory Bus DEVFREQ Driver"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.

Also note that MODULE_DEVICE_TABLE is a no-op for non-modular code.

We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.

Cc: Chanwoo Choi <cw00.choi@samsung.com>
Cc: MyungJoo Ham <myungjoo.ham@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Kukjin Kim <kgene@kernel.org>
Cc: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-samsung-soc@vger.kernel.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
---
 drivers/devfreq/exynos-bus.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/devfreq/exynos-bus.c b/drivers/devfreq/exynos-bus.c
index 2363d0a189b7..2e018d80e0f7 100644
--- a/drivers/devfreq/exynos-bus.c
+++ b/drivers/devfreq/exynos-bus.c
@@ -17,7 +17,7 @@
 #include <linux/devfreq-event.h>
 #include <linux/device.h>
 #include <linux/export.h>
-#include <linux/module.h>
+#include <linux/init.h>
 #include <linux/of_device.h>
 #include <linux/pm_opp.h>
 #include <linux/platform_device.h>
@@ -553,7 +553,6 @@ static const struct of_device_id exynos_bus_of_match[] = {
 	{ .compatible = "samsung,exynos-bus", },
 	{ /* sentinel */ },
 };
-MODULE_DEVICE_TABLE(of, exynos_bus_of_match);
 
 static struct platform_driver exynos_bus_platdrv = {
 	.probe		= exynos_bus_probe,
@@ -563,8 +562,4 @@ static struct platform_driver exynos_bus_platdrv = {
 		.of_match_table = of_match_ptr(exynos_bus_of_match),
 	},
 };
-module_platform_driver(exynos_bus_platdrv);
-
-MODULE_DESCRIPTION("Generic Exynos Bus frequency driver");
-MODULE_AUTHOR("Chanwoo Choi <cw00.choi@samsung.com>");
-MODULE_LICENSE("GPL v2");
+builtin_platform_driver(exynos_bus_platdrv);
-- 
2.8.4

[PATCH 4/5] PM / devfreq: make event/exynos-nocp explicitly non-modular

[Paul Gortmaker]

The Kconfig currently controlling compilation of this code is:

event/Kconfig:config DEVFREQ_EVENT_EXYNOS_NOCP
event/Kconfig:  bool "EXYNOS NoC (Network On Chip) Probe DEVFREQ event Driver"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

We explicitly disallow a driver unbind, since that doesn't have a
sensible use case anyway, and it allows us to drop the ".remove"
code for non-modular drivers.

Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.

We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.

Cc: Chanwoo Choi <cw00.choi@samsung.com>
Cc: MyungJoo Ham <myungjoo.ham@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Kukjin Kim <kgene@kernel.org>
Cc: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-samsung-soc@vger.kernel.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
---
 drivers/devfreq/event/exynos-nocp.c | 19 +++----------------
 1 file changed, 3 insertions(+), 16 deletions(-)

diff --git a/drivers/devfreq/event/exynos-nocp.c b/drivers/devfreq/event/exynos-nocp.c
index 6b6a5f310486..f954578abfdf 100644
--- a/drivers/devfreq/event/exynos-nocp.c
+++ b/drivers/devfreq/event/exynos-nocp.c
@@ -10,7 +10,7 @@
  */
 
 #include <linux/clk.h>
-#include <linux/module.h>
+#include <linux/init.h>
 #include <linux/devfreq-event.h>
 #include <linux/kernel.h>
 #include <linux/of_address.h>
@@ -280,25 +280,12 @@ static int exynos_nocp_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int exynos_nocp_remove(struct platform_device *pdev)
-{
-	struct exynos_nocp *nocp = platform_get_drvdata(pdev);
-
-	clk_disable_unprepare(nocp->clk);
-
-	return 0;
-}
-
 static struct platform_driver exynos_nocp_driver = {
 	.probe	= exynos_nocp_probe,
-	.remove	= exynos_nocp_remove,
 	.driver = {
 		.name	= "exynos-nocp",
+		.suppress_bind_attrs = true,
 		.of_match_table = exynos_nocp_id_match,
 	},
 };
-module_platform_driver(exynos_nocp_driver);
-
-MODULE_DESCRIPTION("Exynos NoC (Network on Chip) Probe driver");
-MODULE_AUTHOR("Chanwoo Choi <cw00.choi@samsung.com>");
-MODULE_LICENSE("GPL");
+builtin_platform_driver(exynos_nocp_driver);
-- 
2.8.4

[PATCH 5/5] PM / devfreq: make event/exynos-ppmu explicitly non-modular

[Paul Gortmaker]

The Kconfig currently controlling compilation of this code is:

config DEVFREQ_EVENT_EXYNOS_PPMU
   bool "EXYNOS PPMU (Platform Perf Monitoring Unit) DEVFREQ event Driver"

...meaning that it currently is not being built as a module by anyone.

Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.

Since module_platform_driver() uses the same init level priority as
builtin_platform_driver() the init ordering remains unchanged with
this commit.

We also delete the MODULE_LICENSE tag etc. since all that information
is already contained at the top of the file in the comments.

Cc: MyungJoo Ham <myungjoo.ham@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Kukjin Kim <kgene@kernel.org>
Cc: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Cc: Chanwoo Choi <cw00.choi@samsung.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-samsung-soc@vger.kernel.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
---
 drivers/devfreq/event/exynos-ppmu.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/devfreq/event/exynos-ppmu.c b/drivers/devfreq/event/exynos-ppmu.c
index f312485f1451..38a997807043 100644
--- a/drivers/devfreq/event/exynos-ppmu.c
+++ b/drivers/devfreq/event/exynos-ppmu.c
@@ -14,7 +14,7 @@
 #include <linux/clk.h>
 #include <linux/io.h>
 #include <linux/kernel.h>
-#include <linux/module.h>
+#include <linux/init.h>
 #include <linux/mutex.h>
 #include <linux/of_address.h>
 #include <linux/platform_device.h>
@@ -524,8 +524,4 @@ static struct platform_driver exynos_ppmu_driver = {
 		.of_match_table = exynos_ppmu_id_match,
 	},
 };
-module_platform_driver(exynos_ppmu_driver);
-
-MODULE_DESCRIPTION("Exynos PPMU(Platform Performance Monitoring Unit) driver");
-MODULE_AUTHOR("Chanwoo Choi <cw00.choi@samsung.com>");
-MODULE_LICENSE("GPL");
+builtin_platform_driver(exynos_ppmu_driver);
-- 
2.8.4

Re: [PATCH 0/5] PM / devfreq: remove modular references from non-modules

[Chanwoo Choi]

Hi Paul,

It is good for devfreq/devfreq-event framework(patch1/patch2).

But, I think that the device driver (patch3/4/5) should change
the type from 'bool' to 'tristate' instead of using the .

Thanks,
Chanwoo Choi

On 2016년 06월 21일 14:14, Paul Gortmaker wrote:
> For anyone new to the underlying goal of this cleanup, we are trying to
> not use module support for code that can never be built as a module since:
> 
>  (1) it is easy to accidentally write unused module_exit and remove code
>  (2) it can be misleading when reading the source, thinking it can be
>      modular when the Makefile and/or Kconfig prohibit it
>  (3) it requires the include of the module.h header file which in turn
>      includes nearly everything else, thus adding to CPP overhead.
>  (4) it gets copied/replicated into other code and spreads like weeds.
> 
> We have already merged over 100 of these for mainline to date, so there
> is really nothing new to see here, in terms of the type of change.
> 
> That said, devfreq changes seen here cover the following categories:
> 
>   -just replacement of modular macros with their non-modular
>    equivalents that CPP would have inserted anyway
> 
>   -the removal of including module.h ; replaced with init.h
>    and/or export.h as required based on whether the file used it.
> 
>   -the removal of any/all unused/orphaned __exit functions that
>    would never be called.
> 
>   -the removal of any ".remove" functions that were hooked into
>    the driver struct.   This ".remove" function would of
>    course not be called from the __exit function since that was
>    never run.  However in theory, someone could have triggered it
>    via sysfs unbind, even though there isn't a sensible use case
>    for doing so.  So to cover that possibility, we've also disabled
>    sysfs unbind in the driver.
> 
> There are no initcall level changes here; everything stays at the level
> of initcall it was previously - either by not using modular versions to
> begin with, or by using the builtin level equivalents.
> 
> Build tested for arm and arm64 allmodconfig (for which all the drivers
> touched here get coverage) on the linux-next tree from today to ensure
> no silly typos crept in.
> 
> If there is a desire for any of these to be modular, we can definitely
> consider that, but by default the changes here keep the code consistent
> with existing behaviour.  Thus I do not expand functionality into the
> modular realm that I can't run time test, or even know if the modular
> instance has a sensible modular use case.
> 
> Paul.
> ---
> 
> Cc: Chanwoo Choi <cw00.choi@samsung.com>
> Cc: Krzysztof Kozlowski <k.kozlowski@samsung.com>
> Cc: Kukjin Kim <kgene@kernel.org>
> Cc: Kyungmin Park <kyungmin.park@samsung.com>
> Cc: MyungJoo Ham <myungjoo.ham@samsung.com>
> Cc: linux-pm@vger.kernel.org
> Cc: linux-samsung-soc@vger.kernel.org
> 
> 
> Paul Gortmaker (5):
>   PM / devfreq: make devfreq explicitly non-modular
>   PM / devfreq: make devfreq-event explicitly non-modular
>   PM / devfreq: make exynos-bus explicitly non-modular
>   PM / devfreq: make event/exynos-nocp explicitly non-modular
>   PM / devfreq: make event/exynos-ppmu explicitly non-modular
> 
>  drivers/devfreq/devfreq-event.c     | 12 +-----------
>  drivers/devfreq/devfreq.c           | 13 +------------
>  drivers/devfreq/event/exynos-nocp.c | 19 +++----------------
>  drivers/devfreq/event/exynos-ppmu.c |  8 ++------
>  drivers/devfreq/exynos-bus.c        |  9 ++-------
>  5 files changed, 9 insertions(+), 52 deletions(-)
>