Without MODULE_LICENSE set, I get the following with modprobe:
acpi_i2c: module license 'unspecified' taints kernel.
acpi_i2c: Unknown symbol i2c_new_device (err 0)
acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
Signed-off-by: Jerry Snitselaar <[email protected]>
---
drivers/acpi/acpi_i2c.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/drivers/acpi/acpi_i2c.c b/drivers/acpi/acpi_i2c.c
index 82045e3..2c5c65d 100644
--- a/drivers/acpi/acpi_i2c.c
+++ b/drivers/acpi/acpi_i2c.c
@@ -14,9 +14,14 @@
#include <linux/export.h>
#include <linux/i2c.h>
#include <linux/ioport.h>
+#include <linux/module.h>
ACPI_MODULE_NAME("i2c");
+MODULE_DESCRIPTION("ACPI I2C enumeration support");
+MODULE_AUTHOR("Mika Westerberg");
+MODULE_LICENSE("GPL");
+
static int acpi_i2c_add_resource(struct acpi_resource *ares, void *data)
{
struct i2c_board_info *info = data;
--
1.8.4.rc3.2.g2c2b664
On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> Without MODULE_LICENSE set, I get the following with modprobe:
>
> acpi_i2c: module license 'unspecified' taints kernel.
> acpi_i2c: Unknown symbol i2c_new_device (err 0)
> acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
>
> Signed-off-by: Jerry Snitselaar <[email protected]>
Looks good to me.
Acked-by: Mika Westerbeg <[email protected]>
On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > Without MODULE_LICENSE set, I get the following with modprobe:
> >
> > acpi_i2c: module license 'unspecified' taints kernel.
> > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> >
> > Signed-off-by: Jerry Snitselaar <[email protected]>
>
> Looks good to me.
>
> Acked-by: Mika Westerbeg <[email protected]>
Well, OK, but do we need to be able to build that as a module?
Maybe it should just be yes or no?
Rafael
On Tue Aug 20 13, Rafael J. Wysocki wrote:
> On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> > On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > > Without MODULE_LICENSE set, I get the following with modprobe:
> > >
> > > acpi_i2c: module license 'unspecified' taints kernel.
> > > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> > >
> > > Signed-off-by: Jerry Snitselaar <[email protected]>
> >
> > Looks good to me.
> >
> > Acked-by: Mika Westerbeg <[email protected]>
>
> Well, OK, but do we need to be able to build that as a module?
>
> Maybe it should just be yes or no?
>
> Rafael
>
Perhaps have depends on I2C=y and be a bool instead of tristate?
On Monday, August 19, 2013 04:35:29 PM Jerry Snitselaar wrote:
> On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> > > On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > > > Without MODULE_LICENSE set, I get the following with modprobe:
> > > >
> > > > acpi_i2c: module license 'unspecified' taints kernel.
> > > > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > > > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > > > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > > > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> > > >
> > > > Signed-off-by: Jerry Snitselaar <[email protected]>
> > >
> > > Looks good to me.
> > >
> > > Acked-by: Mika Westerbeg <[email protected]>
> >
> > Well, OK, but do we need to be able to build that as a module?
> >
> > Maybe it should just be yes or no?
> >
> > Rafael
> >
>
> Perhaps have depends on I2C=y and be a bool instead of tristate?
Yes, that's the idea.
On Tue Aug 20 13, Rafael J. Wysocki wrote:
> On Monday, August 19, 2013 04:35:29 PM Jerry Snitselaar wrote:
> > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> > > > On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > > > > Without MODULE_LICENSE set, I get the following with modprobe:
> > > > >
> > > > > acpi_i2c: module license 'unspecified' taints kernel.
> > > > > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > > > > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > > > > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > > > > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> > > > >
> > > > > Signed-off-by: Jerry Snitselaar <[email protected]>
> > > >
> > > > Looks good to me.
> > > >
> > > > Acked-by: Mika Westerbeg <[email protected]>
> > >
> > > Well, OK, but do we need to be able to build that as a module?
> > >
> > > Maybe it should just be yes or no?
> > >
> > > Rafael
> > >
> >
> > Perhaps have depends on I2C=y and be a bool instead of tristate?
>
> Yes, that's the idea.
>
Does this look okay Mika?
[PATCH] acpi_i2c: do not build as loadable module
Change from tristate to bool, and depend on I2C=y
Signed-off-by: Jerry Snitselaar <[email protected]>
---
drivers/acpi/Kconfig | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
index 100bd72..183a309 100644
--- a/drivers/acpi/Kconfig
+++ b/drivers/acpi/Kconfig
@@ -181,8 +181,9 @@ config ACPI_DOCK
drive bays such as the IBM Ultrabay and the Dell Module Bay.
config ACPI_I2C
- def_tristate I2C
- depends on I2C
+ bool "I2C"
+ depends on I2C=y
+ default n
help
ACPI I2C enumeration support.
--
1.8.4.rc3.2.g2c2b664
On Mon, Aug 19, 2013 at 08:34:03PM -0700, Jerry Snitselaar wrote:
> On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > On Monday, August 19, 2013 04:35:29 PM Jerry Snitselaar wrote:
> > > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > > On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> > > > > On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > > > > > Without MODULE_LICENSE set, I get the following with modprobe:
> > > > > >
> > > > > > acpi_i2c: module license 'unspecified' taints kernel.
> > > > > > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > > > > > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > > > > > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > > > > > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> > > > > >
> > > > > > Signed-off-by: Jerry Snitselaar <[email protected]>
> > > > >
> > > > > Looks good to me.
> > > > >
> > > > > Acked-by: Mika Westerbeg <[email protected]>
> > > >
> > > > Well, OK, but do we need to be able to build that as a module?
> > > >
> > > > Maybe it should just be yes or no?
> > > >
> > > > Rafael
> > > >
> > >
> > > Perhaps have depends on I2C=y and be a bool instead of tristate?
> >
> > Yes, that's the idea.
> >
> Does this look okay Mika?
>
> [PATCH] acpi_i2c: do not build as loadable module
>
> Change from tristate to bool, and depend on I2C=y
I'm not sure about this. Does the below mean that we can't build the ACPI
I2C enumeration at all if I2C core is compiled as module?
>
> Signed-off-by: Jerry Snitselaar <[email protected]>
> ---
> drivers/acpi/Kconfig | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
> index 100bd72..183a309 100644
> --- a/drivers/acpi/Kconfig
> +++ b/drivers/acpi/Kconfig
> @@ -181,8 +181,9 @@ config ACPI_DOCK
> drive bays such as the IBM Ultrabay and the Dell Module Bay.
>
> config ACPI_I2C
> - def_tristate I2C
> - depends on I2C
> + bool "I2C"
> + depends on I2C=y
> + default n
> help
> ACPI I2C enumeration support.
>
> --
> 1.8.4.rc3.2.g2c2b664
On Tue Aug 20 13, Mika Westerberg wrote:
> On Mon, Aug 19, 2013 at 08:34:03PM -0700, Jerry Snitselaar wrote:
> > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > On Monday, August 19, 2013 04:35:29 PM Jerry Snitselaar wrote:
> > > > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > > > On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> > > > > > On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > > > > > > Without MODULE_LICENSE set, I get the following with modprobe:
> > > > > > >
> > > > > > > acpi_i2c: module license 'unspecified' taints kernel.
> > > > > > > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > > > > > > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > > > > > > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > > > > > > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> > > > > > >
> > > > > > > Signed-off-by: Jerry Snitselaar <[email protected]>
> > > > > >
> > > > > > Looks good to me.
> > > > > >
> > > > > > Acked-by: Mika Westerbeg <[email protected]>
> > > > >
> > > > > Well, OK, but do we need to be able to build that as a module?
> > > > >
> > > > > Maybe it should just be yes or no?
> > > > >
> > > > > Rafael
> > > > >
> > > >
> > > > Perhaps have depends on I2C=y and be a bool instead of tristate?
> > >
> > > Yes, that's the idea.
> > >
> > Does this look okay Mika?
> >
> > [PATCH] acpi_i2c: do not build as loadable module
> >
> > Change from tristate to bool, and depend on I2C=y
>
> I'm not sure about this. Does the below mean that we can't build the ACPI
> I2C enumeration at all if I2C core is compiled as module?
Yes, that was what Rafael was suggesting. If the ability to compile as
a module if I2C is a module is needed, then we need the 1st patch.
Jerry
>
> >
> > Signed-off-by: Jerry Snitselaar <[email protected]>
> > ---
> > drivers/acpi/Kconfig | 5 +++--
> > 1 file changed, 3 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
> > index 100bd72..183a309 100644
> > --- a/drivers/acpi/Kconfig
> > +++ b/drivers/acpi/Kconfig
> > @@ -181,8 +181,9 @@ config ACPI_DOCK
> > drive bays such as the IBM Ultrabay and the Dell Module Bay.
> >
> > config ACPI_I2C
> > - def_tristate I2C
> > - depends on I2C
> > + bool "I2C"
> > + depends on I2C=y
> > + default n
> > help
> > ACPI I2C enumeration support.
> >
> > --
> > 1.8.4.rc3.2.g2c2b664
On Tue, Aug 20, 2013 at 01:00:08AM -0700, Jerry Snitselaar wrote:
> On Tue Aug 20 13, Mika Westerberg wrote:
> > On Mon, Aug 19, 2013 at 08:34:03PM -0700, Jerry Snitselaar wrote:
> > > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > > On Monday, August 19, 2013 04:35:29 PM Jerry Snitselaar wrote:
> > > > > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > > > > On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> > > > > > > On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > > > > > > > Without MODULE_LICENSE set, I get the following with modprobe:
> > > > > > > >
> > > > > > > > acpi_i2c: module license 'unspecified' taints kernel.
> > > > > > > > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > > > > > > > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > > > > > > > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > > > > > > > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> > > > > > > >
> > > > > > > > Signed-off-by: Jerry Snitselaar <[email protected]>
> > > > > > >
> > > > > > > Looks good to me.
> > > > > > >
> > > > > > > Acked-by: Mika Westerbeg <[email protected]>
> > > > > >
> > > > > > Well, OK, but do we need to be able to build that as a module?
> > > > > >
> > > > > > Maybe it should just be yes or no?
> > > > > >
> > > > > > Rafael
> > > > > >
> > > > >
> > > > > Perhaps have depends on I2C=y and be a bool instead of tristate?
> > > >
> > > > Yes, that's the idea.
> > > >
> > > Does this look okay Mika?
> > >
> > > [PATCH] acpi_i2c: do not build as loadable module
> > >
> > > Change from tristate to bool, and depend on I2C=y
> >
> > I'm not sure about this. Does the below mean that we can't build the ACPI
> > I2C enumeration at all if I2C core is compiled as module?
>
> Yes, that was what Rafael was suggesting. If the ability to compile as
> a module if I2C is a module is needed, then we need the 1st patch.
In that case I would prefer the first patch. Otherwise we lose the ability
to enumerate I2C devices from ACPI namespace on some distros (at least
Debian builds I2C core as a module).
Rafael?
On Tue, Aug 20, 2013 at 11:14:42AM +0300, Mika Westerberg wrote:
> On Tue, Aug 20, 2013 at 01:00:08AM -0700, Jerry Snitselaar wrote:
> > On Tue Aug 20 13, Mika Westerberg wrote:
> > > On Mon, Aug 19, 2013 at 08:34:03PM -0700, Jerry Snitselaar wrote:
> > > > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > > > On Monday, August 19, 2013 04:35:29 PM Jerry Snitselaar wrote:
> > > > > > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > > > > > On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> > > > > > > > On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > > > > > > > > Without MODULE_LICENSE set, I get the following with modprobe:
> > > > > > > > >
> > > > > > > > > acpi_i2c: module license 'unspecified' taints kernel.
> > > > > > > > > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > > > > > > > > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > > > > > > > > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > > > > > > > > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> > > > > > > > >
> > > > > > > > > Signed-off-by: Jerry Snitselaar <[email protected]>
> > > > > > > >
> > > > > > > > Looks good to me.
> > > > > > > >
> > > > > > > > Acked-by: Mika Westerbeg <[email protected]>
> > > > > > >
> > > > > > > Well, OK, but do we need to be able to build that as a module?
> > > > > > >
> > > > > > > Maybe it should just be yes or no?
> > > > > > >
> > > > > > > Rafael
> > > > > > >
> > > > > >
> > > > > > Perhaps have depends on I2C=y and be a bool instead of tristate?
> > > > >
> > > > > Yes, that's the idea.
> > > > >
> > > > Does this look okay Mika?
> > > >
> > > > [PATCH] acpi_i2c: do not build as loadable module
> > > >
> > > > Change from tristate to bool, and depend on I2C=y
> > >
> > > I'm not sure about this. Does the below mean that we can't build the ACPI
> > > I2C enumeration at all if I2C core is compiled as module?
> >
> > Yes, that was what Rafael was suggesting. If the ability to compile as
> > a module if I2C is a module is needed, then we need the 1st patch.
>
> In that case I would prefer the first patch. Otherwise we lose the ability
> to enumerate I2C devices from ACPI namespace on some distros (at least
> Debian builds I2C core as a module).
>
> Rafael?
Actually there's a patch that moves DT I2C helpers to the I2C core here:
https://lkml.org/lkml/2013/8/19/349
we should probably do the same for the ACPI case. Doing that solves this
problem as well.
On Tuesday, August 20, 2013 11:18:52 AM Mika Westerberg wrote:
> On Tue, Aug 20, 2013 at 11:14:42AM +0300, Mika Westerberg wrote:
> > On Tue, Aug 20, 2013 at 01:00:08AM -0700, Jerry Snitselaar wrote:
> > > On Tue Aug 20 13, Mika Westerberg wrote:
> > > > On Mon, Aug 19, 2013 at 08:34:03PM -0700, Jerry Snitselaar wrote:
> > > > > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > > > > On Monday, August 19, 2013 04:35:29 PM Jerry Snitselaar wrote:
> > > > > > > On Tue Aug 20 13, Rafael J. Wysocki wrote:
> > > > > > > > On Monday, August 19, 2013 09:16:14 AM Mika Westerberg wrote:
> > > > > > > > > On Fri, Aug 16, 2013 at 06:26:35PM -0700, Jerry Snitselaar wrote:
> > > > > > > > > > Without MODULE_LICENSE set, I get the following with modprobe:
> > > > > > > > > >
> > > > > > > > > > acpi_i2c: module license 'unspecified' taints kernel.
> > > > > > > > > > acpi_i2c: Unknown symbol i2c_new_device (err 0)
> > > > > > > > > > acpi_i2c: Unknown symbol acpi_dev_get_resources (err 0)
> > > > > > > > > > acpi_i2c: Unknown symbol acpi_dev_resource_interrupt (err 0)
> > > > > > > > > > acpi_i2c: Unknown symbol acpi_dev_free_resource_list (err 0)
> > > > > > > > > >
> > > > > > > > > > Signed-off-by: Jerry Snitselaar <[email protected]>
> > > > > > > > >
> > > > > > > > > Looks good to me.
> > > > > > > > >
> > > > > > > > > Acked-by: Mika Westerbeg <[email protected]>
> > > > > > > >
> > > > > > > > Well, OK, but do we need to be able to build that as a module?
> > > > > > > >
> > > > > > > > Maybe it should just be yes or no?
> > > > > > > >
> > > > > > > > Rafael
> > > > > > > >
> > > > > > >
> > > > > > > Perhaps have depends on I2C=y and be a bool instead of tristate?
> > > > > >
> > > > > > Yes, that's the idea.
> > > > > >
> > > > > Does this look okay Mika?
> > > > >
> > > > > [PATCH] acpi_i2c: do not build as loadable module
> > > > >
> > > > > Change from tristate to bool, and depend on I2C=y
> > > >
> > > > I'm not sure about this. Does the below mean that we can't build the ACPI
> > > > I2C enumeration at all if I2C core is compiled as module?
> > >
> > > Yes, that was what Rafael was suggesting. If the ability to compile as
> > > a module if I2C is a module is needed, then we need the 1st patch.
> >
> > In that case I would prefer the first patch. Otherwise we lose the ability
> > to enumerate I2C devices from ACPI namespace on some distros (at least
> > Debian builds I2C core as a module).
> >
> > Rafael?
>
> Actually there's a patch that moves DT I2C helpers to the I2C core here:
>
> https://lkml.org/lkml/2013/8/19/349
>
> we should probably do the same for the ACPI case. Doing that solves this
> problem as well.
Yes, and I'd prefer it to be done this way. Having ACPI support as a separate
module doesn't really buy us anything.
Thanks,
Rafael