]> git.karo-electronics.de Git - karo-tx-linux.git/log
karo-tx-linux.git
13 years agopm_runtime.h: explicitly requires notifier.h
Paul Gortmaker [Fri, 27 May 2011 11:08:41 +0000 (07:08 -0400)]
pm_runtime.h: explicitly requires notifier.h

This file was getting notifier.h via device.h --> module.h but
the module.h inclusion is going away, so add notifier.h directly.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agolinux/dmaengine.h needs bitmap.h
Stephen Rothwell [Fri, 29 Jul 2011 06:55:11 +0000 (16:55 +1000)]
linux/dmaengine.h needs bitmap.h

Presumably caused by some include file no longer being implicitly
included.

In file included from drivers/dma/iovlock.c:28:0:
include/linux/dmaengine.h: In function '__dma_cap_zero':
include/linux/dmaengine.h:736:2: error: implicit declaration of function 'bitmap_zero'
In file included from include/linux/nodemask.h:95:0,
                 from include/linux/mmzone.h:16,
                 from include/linux/gfp.h:4,
                 from include/linux/mm.h:8,
                 from include/linux/pagemap.h:7,
                 from drivers/dma/iovlock.c:29:
include/linux/bitmap.h: At top level:
include/linux/bitmap.h:159:20: warning: conflicting types for 'bitmap_zero'
include/linux/bitmap.h:159:20: error: static declaration of 'bitmap_zero' follows non-static declaration
include/linux/dmaengine.h:736:2: note: previous implicit declaration of 'bitmap_zero' was here

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agolinux/stop_machine.h: fix implicit use of smp.h for smp_processor_id
Paul Gortmaker [Sun, 17 Jul 2011 21:11:26 +0000 (17:11 -0400)]
linux/stop_machine.h: fix implicit use of smp.h for smp_processor_id

This will show up on MIPS when we fix all the implicit header presences
that are because of module.h being everywhere.

In file included from kernel/trace/ftrace.c:16:
include/linux/stop_machine.h: In function 'stop_one_cpu':
include/linux/stop_machine.h:50: error: implicit declaration of function 'smp_processor_id'
include/linux/stop_machine.h: In function 'stop_cpus':
include/linux/stop_machine.h:80: error: implicit declaration of function 'raw_smp_processor_id'

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agocrypto.h: remove unused crypto_tfm_alg_modname() inline
Paul Gortmaker [Thu, 26 May 2011 19:34:12 +0000 (15:34 -0400)]
crypto.h: remove unused crypto_tfm_alg_modname() inline

The <linux/crypto.h> (which is in turn in common headers
like tcp.h) wants to use module_name() in an inline fcn.
But having all of <linux/module.h> along for the ride is
overkill and slows down compiles by a measureable amount,
since it in turn includes lots of headers.

Since the inline is never used anywhere in the kernel[1],
we can just remove it, and then also remove the module.h
include as well.

In all the many crypto modules, there were some relying on
crypto.h including module.h -- for them we now explicitly
call out module.h for inclusion.

[1] git grep shows some staging drivers also define the same
static inline, but they also never ever use it.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodevice_cgroup.h: delete needless include <linux/module.h>
Paul Gortmaker [Thu, 26 May 2011 19:58:15 +0000 (15:58 -0400)]
device_cgroup.h: delete needless include <linux/module.h>

There is nothing modular in this file, and no reason to drag
in all the 357 headers that module.h brings with it, since
it just slows down compiles.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agonet: inet_timewait_sock doesnt need <linux/module.h>
Paul Gortmaker [Thu, 26 May 2011 17:36:45 +0000 (13:36 -0400)]
net: inet_timewait_sock doesnt need <linux/module.h>

There is nothing module specific in this header, and removing
it doesn't seem to uncover any implicit dependencies either.
Must be simply a vestige of an ancient legacy.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agosysdev.h: dont include <linux/module.h> for no reason
Paul Gortmaker [Thu, 26 May 2011 15:41:46 +0000 (11:41 -0400)]
sysdev.h: dont include <linux/module.h> for no reason

The <linux/module.h> pretty much brings in the kitchen sink along
with it, so it should be avoided wherever reasonably possible in
terms of being included from other commonly used <linux/something.h>
files, as it results in a measureable increase on compile times.

There doesn't appear to be any module specifics in this file.
The obvious people who were relying on the presence of
the vast amount of stuff module.h sucked in have been fixed.

If other files are implicitly relying on it, then lets see who
they are and fix them too.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/edac: Add module.h to mce_amd_inj.c
Paul Gortmaker [Sun, 3 Jul 2011 17:37:56 +0000 (13:37 -0400)]
drivers/edac: Add module.h to mce_amd_inj.c

This file really needs the full module.h header file present, but
was just getting it implicitly before.  Fix it up in advance so we
avoid build failures once the cleanup commit is present.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agolib: dma-debug needs export.h for EXPORT_SYMBOL
Paul Gortmaker [Fri, 1 Jul 2011 20:23:59 +0000 (16:23 -0400)]
lib: dma-debug needs export.h for EXPORT_SYMBOL

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/sfi: sfi_acpi.c needs sysfs.h
Randy Dunlap [Sat, 30 Jul 2011 01:09:12 +0000 (18:09 -0700)]
drivers/sfi: sfi_acpi.c needs sysfs.h

sfi_acpi.c needs to include linux/sysfs.h for data types.

drivers/sfi/sfi_core.h:66: error: field 'attr' has incomplete type
drivers/sfi/sfi_acpi.c:179: warning: 'struct kobject' declared inside parameter list
drivers/sfi/sfi_acpi.c:179: warning: its scope is only this definition or declaration, which is probably not what you want
drivers/sfi/sfi_acpi.c:182: warning: type defaults to 'int' in declaration of '__mptr'
drivers/sfi/sfi_acpi.c:182: warning: initialization from incompatible pointer type

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/pcmcia: add module.h to db1xxx_ss.c
Paul Gortmaker [Thu, 28 Jul 2011 22:44:02 +0000 (18:44 -0400)]
drivers/pcmcia: add module.h to db1xxx_ss.c

As we will get this after the module.h usage cleanup.

drivers/pcmcia/db1xxx_ss.c:577: warning: type defaults to 'int' in declaration of 'MODULE_LICENSE'

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/dca: Add module.h to dca-core.c
Paul Gortmaker [Sun, 3 Jul 2011 17:37:11 +0000 (13:37 -0400)]
drivers/dca: Add module.h to dca-core.c

This file really needs the full module.h header file present, but
was just getting it implicitly before.  Call it out in advance so
that we don't get future build failures on this.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/hwmon: add module.h to ultra45_env.c
Paul Gortmaker [Mon, 18 Jul 2011 20:05:10 +0000 (16:05 -0400)]
drivers/hwmon: add module.h to ultra45_env.c

This file needs the full module.h header and up to now was just
implicitly capitalizing on it being present already.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agorapidio: Fix up those implicitly relying on getting stat.h
Paul Gortmaker [Fri, 15 Jul 2011 20:48:24 +0000 (16:48 -0400)]
rapidio: Fix up those implicitly relying on getting stat.h

Upon the trial removal of the implicit presence of module.h,
lots of files showed up that were getting the sub-includes
by default without calling them out.  Here is more of them.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/w1: Add export.h for EXPORT_SYMBOL to w1_family.c
Paul Gortmaker [Sun, 10 Jul 2011 17:21:52 +0000 (13:21 -0400)]
drivers/w1: Add export.h for EXPORT_SYMBOL to w1_family.c

Give this file export.h so that it can reliably get the EXPORT_SYMBOL
macros in the future, once module.h isn't implicitly everywhere.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/of: include export.h for EXPORT_SYMBOL_GPL in of_pci_irq.c
Stephen Rothwell [Fri, 29 Jul 2011 06:05:38 +0000 (16:05 +1000)]
drivers/of: include export.h for EXPORT_SYMBOL_GPL in of_pci_irq.c

This is an interaction with commit 98d9f30c820d ("pci/of: Match PCI
devices to OF nodes dynamically") which split off part of
drivers/of/of_pci.c, causing:

drivers/of/of_pci_irq.c:92:1: warning: data definition has no type or storage class
drivers/of/of_pci_irq.c:92:1: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL'
drivers/of/of_pci_irq.c:92:1: warning: parameter names (without types) in function declaration

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/sh: Add export.h for EXPORT_SYMBOL to intc/virq.c
Paul Gortmaker [Sun, 10 Jul 2011 16:57:12 +0000 (12:57 -0400)]
drivers/sh: Add export.h for EXPORT_SYMBOL to intc/virq.c

This is exporting symbols and will fail to build once we remove
the implicit presence of module.h

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/isdn: Add export.h for THIS_MODULE to capi/kcapi_proc.c
Paul Gortmaker [Sun, 10 Jul 2011 16:23:16 +0000 (12:23 -0400)]
drivers/isdn: Add export.h for THIS_MODULE to capi/kcapi_proc.c

In preparation of the module.h usage cleanup, call out the header
that we need to get THIS_MODULE into scope.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrm: Add export.h for EXPORT_SYMBOL to ttm/ttm_tt.c
Paul Gortmaker [Sun, 10 Jul 2011 16:20:27 +0000 (12:20 -0400)]
drm: Add export.h for EXPORT_SYMBOL to ttm/ttm_tt.c

This exports symbols, but wasn't calling out the module.h header
before.  Add in export.h so that it still builds once we halt the
spread of module.h to every file.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/firewire: Add export.h for EXPORT_SYMBOL to core-iso.c
Paul Gortmaker [Sun, 10 Jul 2011 16:16:32 +0000 (12:16 -0400)]
drivers/firewire: Add export.h for EXPORT_SYMBOL to core-iso.c

Ensure that the EXPORT_SYMBOL macros are present for when we clean up
the "module.h" is everywhere situation, to prevent build failures.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/dca: Add export.h for THIS_MODULE to dca-sysfs.c
Paul Gortmaker [Sun, 10 Jul 2011 16:15:51 +0000 (12:15 -0400)]
drivers/dca: Add export.h for THIS_MODULE to dca-sysfs.c

Ensure we have access to the THIS_MODLUE macro once we clean up
the implicit module.h usage.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/message: Add module.h to i2o/pci.c
Paul Gortmaker [Sun, 3 Jul 2011 19:13:20 +0000 (15:13 -0400)]
drivers/message: Add module.h to i2o/pci.c

This file really needs the full module.h header file present, but
was just getting it implicitly before.  Fix it up in advance so we
avoid build failures once the cleanup commit is present.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agouwb: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required
Paul Gortmaker [Sun, 10 Jul 2011 17:18:02 +0000 (13:18 -0400)]
uwb: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required

These macros are no longer in module.h and module.h is no longer
present everywhere.  Call out export.h for the real users who
are making use of these macros, or else we'll get things like:

 CC      drivers/uwb/umc-drv.o
drivers/uwb/umc-dev.c:42: warning: data definition has no type or storage class
drivers/uwb/umc-dev.c:42: warning: type defaults to ‘int’ in declaration of ‘EXPORT_SYMBOL_GPL’
drivers/uwb/umc-dev.c:42: warning: parameter names (without types) in function declaration

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agouwb: Add module.h to the real uwb modular users.
Paul Gortmaker [Sun, 17 Jul 2011 20:40:36 +0000 (16:40 -0400)]
uwb: Add module.h to the real uwb modular users.

Lots of files were getting an implicit include of module.h
for years now.  But that will be fixed shortly.  So get the
real users to call out the include explicitly.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/uwb: fix implicit use of stat.h
Paul Gortmaker [Sun, 17 Jul 2011 20:50:39 +0000 (16:50 -0400)]
drivers/uwb: fix implicit use of stat.h

The module.h (via device.h) was pulling handfuls of implicitly present
header files -- including basic stuff like stat.h.  Fix up these stat.h
users in advance so they don't break on the cleanup.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/gpio: Fix drivers who are implicit users of module.h
Paul Gortmaker [Sun, 3 Jul 2011 17:38:09 +0000 (13:38 -0400)]
drivers/gpio: Fix drivers who are implicit users of module.h

A pending cleanup will mean that module.h won't be implicitly
everywhere anymore.  Make sure the modular drivers in gpio
are actually calling out for <module.h> explicitly in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/of: Add module.h to of_spi.c
Paul Gortmaker [Sun, 17 Jul 2011 19:58:06 +0000 (15:58 -0400)]
drivers/of: Add module.h to of_spi.c

This file really needs the full module.h header file present, but
was just getting it implicitly before.  Fix it up in advance so we
avoid build failures once the cleanup commit is present.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/spi: Add export.h for THIS_MODULE/EXPORT_SYMBOL to spi.c
Paul Gortmaker [Sun, 10 Jul 2011 16:57:55 +0000 (12:57 -0400)]
drivers/spi: Add export.h for THIS_MODULE/EXPORT_SYMBOL to spi.c

This uses both EXPORT_SYMBOL and THIS_MODULE, both which come from
the export.h file now

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agospi: spi-au1550.c was implicitly using module.h
Paul Gortmaker [Thu, 28 Jul 2011 23:33:33 +0000 (19:33 -0400)]
spi: spi-au1550.c was implicitly using module.h

It needs the full module.h header, else you get:

drivers/spi/spi-au1550.c:1030: warning: type defaults to 'int' in declaration of 'MODULE_DESCRIPTION'

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agospi: Add module.h to implicit users in drivers/spi
Paul Gortmaker [Sun, 3 Jul 2011 19:44:29 +0000 (15:44 -0400)]
spi: Add module.h to implicit users in drivers/spi

We are clipping down the presence of module.h, since it was
everywhere.  If you really need it, you better call it out,
as per this changeset.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoregulator/driver.h: Fix implicit use of notifier.h
Paul Gortmaker [Sun, 17 Jul 2011 20:24:35 +0000 (16:24 -0400)]
regulator/driver.h: Fix implicit use of notifier.h

This was implicitly appearing by way of module.h -- but when
we fix that, we'll get this:

In file included from drivers/regulator/dummy.c:21:
include/linux/regulator/driver.h:197: error: field 'notifier' has incomplete type
make[3]: *** [drivers/regulator/dummy.o] Error 1

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoregulator: Add module.h to drivers/regulator users as required
Paul Gortmaker [Sun, 17 Jul 2011 20:28:23 +0000 (16:28 -0400)]
regulator: Add module.h to drivers/regulator users as required

Another group of drivers that are taking advantage of the implicit
presence of module.h -- and will break when we pull the carpet out
from under them during a cleanup.  Fix 'em now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/regulator: Add export.h for THIS_MODULE to dummy.c
Paul Gortmaker [Sun, 17 Jul 2011 19:48:22 +0000 (15:48 -0400)]
drivers/regulator: Add export.h for THIS_MODULE to dummy.c

So that THIS_MODULE is still in scope once we clean up the presence
of module.h being everywhere.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers: pci-sysfs.c needs export.h for THIS_MODULE
Paul Gortmaker [Fri, 27 May 2011 13:20:21 +0000 (09:20 -0400)]
drivers: pci-sysfs.c needs export.h for THIS_MODULE

Otherwise, once you break the chain of implicit includes from
device.h --> module.h then you will get this:

pci-sysfs.c: In function ‘remove_store’:
pci-sysfs.c:350: error: ‘THIS_MODULE’ undeclared (first use in this function)
pci-sysfs.c:350: error: (Each undeclared identifier is reported only once
pci-sysfs.c:350: error: for each function it appears in.)
make[3]: *** [drivers/pci/pci-sysfs.o] Error 1

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/pci: Add module.h to hotplug/pciehp_acpi.c
Paul Gortmaker [Sun, 3 Jul 2011 19:27:05 +0000 (15:27 -0400)]
drivers/pci: Add module.h to hotplug/pciehp_acpi.c

This file really needs the full module.h header file present, but
was just getting it implicitly before.  Fix it up in advance so we
avoid build failures once the cleanup commit is present.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agovirtio: Add module.h to drivers/virtio users.
Paul Gortmaker [Sun, 3 Jul 2011 20:20:30 +0000 (16:20 -0400)]
virtio: Add module.h to drivers/virtio users.

Up to now, the module.h header was as hard to keep out as
sunlight.  But we are cleaning that up.  Fix the virtio users
who simply expect module.h to be there in every C file.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/lguest: add export.h to lguest_device.c
Paul Gortmaker [Thu, 21 Jul 2011 17:03:20 +0000 (13:03 -0400)]
drivers/lguest: add export.h to lguest_device.c

We need this in advance of the module.h cleanup, or we'll get:

  CC      drivers/lguest/lguest_device.o
drivers/lguest/lguest_device.c: In function ‘lguest_devices_init’:
drivers/lguest/lguest_device.c:490: error: ‘THIS_MODULE’ undeclared (first use in this function)

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agokvm: iommu.c file requires the full module.h present.
Paul Gortmaker [Thu, 28 Jul 2011 01:25:05 +0000 (21:25 -0400)]
kvm: iommu.c file requires the full module.h present.

This file has things like module_param_named() and MODULE_PARM_DESC()
so it needs the full module.h header present.  Without it, you'll get:

  CC      arch/x86/kvm/../../../virt/kvm/iommu.o
virt/kvm/iommu.c:37: error: expected ‘)’ before ‘bool’
virt/kvm/iommu.c:39: error: expected ‘)’ before string constant
make[3]: *** [arch/x86/kvm/../../../virt/kvm/iommu.o] Error 1
make[2]: *** [arch/x86/kvm] Error 2

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agokvm: fix implicit use of stat.h header file
Paul Gortmaker [Thu, 28 Jul 2011 01:17:59 +0000 (21:17 -0400)]
kvm: fix implicit use of stat.h header file

This was coming in via an implicit module.h (and its sub-includes)
before, but we'll be cleaning that up shortly.  Call out the stat.h
include requirement in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoxen: fix implicit stat.h/module.h include usage in xen-selfballoon.c
Paul Gortmaker [Thu, 28 Jul 2011 14:08:24 +0000 (10:08 -0400)]
xen: fix implicit stat.h/module.h include usage in xen-selfballoon.c

The module.h cleanup has revealed that this file had the following
implicit dependencies as follows:

  CC      drivers/xen/xen-selfballoon.o
drivers/xen/xen-selfballoon.c:272: error: ‘S_IRUGO’ undeclared here (not in a function)
drivers/xen/xen-selfballoon.c:272: error: ‘S_IWUSR’ undeclared here (not in a function)
drivers/xen/xen-selfballoon.c:485: warning: type defaults to ‘int’ in declaration of ‘MODULE_LICENSE’
drivers/xen/xen-selfballoon.c:485: warning: function declaration isn’t a prototype
make[3]: *** [drivers/xen/xen-selfballoon.o] Error 1

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoxen: Add export.h for THIS_MODULE/EXPORT_SYMBOL to various xen users.
Paul Gortmaker [Sun, 10 Jul 2011 17:22:07 +0000 (13:22 -0400)]
xen: Add export.h for THIS_MODULE/EXPORT_SYMBOL to various xen users.

Things like THIS_MODULE and EXPORT_SYMBOL were simply everywhere
because module.h was also everywhere.  But we are fixing the latter.
So we need to call out the real users in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoxen: Add module.h to modular drivers/xen users.
Paul Gortmaker [Sun, 3 Jul 2011 20:20:57 +0000 (16:20 -0400)]
xen: Add module.h to modular drivers/xen users.

Previously these drivers just got module.h implicitly, but we
are cleaning that up and it will be no longer.  Call out the
real users of it.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoscsi: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required
Paul Gortmaker [Sun, 10 Jul 2011 16:55:15 +0000 (12:55 -0400)]
scsi: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required

For the basic SCSI infrastructure files that are exporting symbols
but not modules themselves, add in the basic export.h header file
to allow the exports.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoscsi: Add module.h to drivers/scsi users who really need it.
Paul Gortmaker [Sun, 17 Jul 2011 19:23:40 +0000 (15:23 -0400)]
scsi: Add module.h to drivers/scsi users who really need it.

The module.h was really just about everywhere.  But no more.
With a cleanup, you'll really need to grab it if you want it.
Fix up the users who really need it in SCSI.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agostaging: fix implicit use of slab.h in dt3155v4l/dt3155v4l.c
Paul Gortmaker [Mon, 18 Jul 2011 18:27:17 +0000 (14:27 -0400)]
staging: fix implicit use of slab.h in dt3155v4l/dt3155v4l.c

Add in <linux/slab.h> to fix:

drivers/staging/dt3155v4l/dt3155v4l.c: In function 'dt3155_open':
drivers/staging/dt3155v4l/dt3155v4l.c:435: error: implicit declaration of function 'kzalloc'

during builds on ARMi architecture.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agostaging: Add export.h for THIS_MODULE/EXPORT_SYMBOL to drivers/staging users.
Paul Gortmaker [Sun, 10 Jul 2011 17:09:12 +0000 (13:09 -0400)]
staging: Add export.h for THIS_MODULE/EXPORT_SYMBOL to drivers/staging users.

Lots of drivers have in the past expected the presence of basic things
like THIS_MODULE and EXPORT_SYMBOL.  With the header cleanup, they wont
have these.  Call out the include explicitly.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agostaging: Add module.h to drivers/staging users
Paul Gortmaker [Sun, 3 Jul 2011 19:49:50 +0000 (15:49 -0400)]
staging: Add module.h to drivers/staging users

Lots of people expect module.h to just "be there" without
any #include effort.  But we are crushing that.  So fix those
people in staging with broken expectations.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agommc: Fix implicit use of stat.h header in associated files
Paul Gortmaker [Fri, 15 Jul 2011 16:01:27 +0000 (12:01 -0400)]
mmc: Fix implicit use of stat.h header in associated files

Once the implicit use of module.h is prevented, these files will
fail to find the stat.h header content.

Fix up the implicit usage expectations in advance of the cleanup.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agommc: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required
Paul Gortmaker [Sun, 10 Jul 2011 16:42:00 +0000 (12:42 -0400)]
mmc: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required

These two basic defines were everywhere, simply because module.h
was also everywhere.   But we are cleaning up the latter.  So make
the exporters actually call out their need for the include.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agommc: Add module.h to drivers/mmc users assuming implicit presence.
Paul Gortmaker [Sun, 3 Jul 2011 19:15:51 +0000 (15:15 -0400)]
mmc: Add module.h to drivers/mmc users assuming implicit presence.

We are cleaning up the implicit presence of module.h; these guys are
some of the people who just assume it will be there.  Call it out
explitly for those that really need it.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/misc: Add export.h for EXPORT_SYMBOL to atmel_tclib.c
Paul Gortmaker [Sun, 10 Jul 2011 16:41:41 +0000 (12:41 -0400)]
drivers/misc: Add export.h for EXPORT_SYMBOL to atmel_tclib.c

This exports symbols, and so will need this to avoid build failures
pending a future module.h presence cleanup.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/misc: Add module.h to files who are really modular.
Paul Gortmaker [Sun, 3 Jul 2011 19:14:56 +0000 (15:14 -0400)]
drivers/misc: Add module.h to files who are really modular.

These files really need the full module.h header file present, but
were just getting it implicitly before.  Fix it up in advance so we
avoid build failures once the cleanup commit is present.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agomfd: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required
Paul Gortmaker [Sun, 10 Jul 2011 16:41:10 +0000 (12:41 -0400)]
mfd: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required

These macros are in <linux/export.h> and will no longer be
implicitly everywhere anymore.  Fix it up in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agomfd: Add module.h to the implicit drivers/mfd users
Paul Gortmaker [Sun, 3 Jul 2011 19:13:27 +0000 (15:13 -0400)]
mfd: Add module.h to the implicit drivers/mfd users

With the pending module.h cleanup, these files will fail to compile,
unless they explicitly call out the include of this file.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/leds: Add export.h for THIS_MODULE to leds-cobalt-raq.c
Paul Gortmaker [Sun, 10 Jul 2011 16:24:44 +0000 (12:24 -0400)]
drivers/leds: Add export.h for THIS_MODULE to leds-cobalt-raq.c

In preparation of the module.h usage cleanup, call out the export.h
to avoid build failures when that happens.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/leds: Add module.h to files using it implicitly
Paul Gortmaker [Sun, 3 Jul 2011 17:56:03 +0000 (13:56 -0400)]
drivers/leds: Add module.h to files using it implicitly

A pending cleanup will mean that module.h won't be implicitly
everywhere anymore.  Make sure the modular drivers in the leds
dir are actually calling out for <module.h> explicitly in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/firmware: Add export.h for EXPORT_SYMBOL to sigma.c
Paul Gortmaker [Sun, 10 Jul 2011 16:16:38 +0000 (12:16 -0400)]
drivers/firmware: Add export.h for EXPORT_SYMBOL to sigma.c

It exports symbols, so it will need this once we remove the implicit
everywhere presence of module.h

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/firmware: Add module.h to google/gsmi.c
Paul Gortmaker [Sun, 3 Jul 2011 17:38:03 +0000 (13:38 -0400)]
drivers/firmware: Add module.h to google/gsmi.c

This file really needs the full module.h header file present, but
was just getting it implicitly before.  Fix it up in advance so we
avoid build failures once the cleanup commit is present.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agobcma: fix implicit use of export.h contents
Paul Gortmaker [Thu, 28 Jul 2011 01:21:04 +0000 (21:21 -0400)]
bcma: fix implicit use of export.h contents

Fix in advance, or we will get this:

drivers/bcma/core.c:20: warning: data definition has no type or storage class
drivers/bcma/core.c:20: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL'
drivers/bcma/core.c:20: warning: parameter names (without types) in function declaration

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agobcma: add module.h to the modular portions of this driver
Paul Gortmaker [Fri, 1 Jul 2011 20:06:37 +0000 (16:06 -0400)]
bcma: add module.h to the modular portions of this driver

This will ensure that it continues to build once we remove
the implicit module.h presence from everywhere later on.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agouio: Add module.h to implicit drivers/uio users
Paul Gortmaker [Sun, 17 Jul 2011 20:38:40 +0000 (16:38 -0400)]
uio: Add module.h to implicit drivers/uio users

We are cleaning up the omnipresent module.h stuff, so people
who really use it need to call it out explicitly.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agossb: driver_gige.c needs export.h
Paul Gortmaker [Thu, 28 Jul 2011 23:10:21 +0000 (19:10 -0400)]
ssb: driver_gige.c needs export.h

Or else we'll get this:

drivers/ssb/driver_gige.c:249: warning: ty
pe defaults to 'int' in declaration of 'EXPORT_SYMBOL'

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agossb: fixup another group of export.h implicit users
Paul Gortmaker [Thu, 28 Jul 2011 02:07:02 +0000 (22:07 -0400)]
ssb: fixup another group of export.h implicit users

Or we will get things like this when we remove the implicit path:

drivers/ssb/embedded.c:32: warning: data definition has no type or storage class
drivers/ssb/embedded.c:32: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL'
drivers/ssb/embedded.c:32: warning: parameter names (without types) in function declaration

drivers/ssb/driver_chipcommon.c:432: warning: data definition has no type or storage class
drivers/ssb/driver_chipcommon.c:432: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL'
drivers/ssb/driver_chipcommon.c:432: warning: parameter names (without types) in function declaration

drivers/ssb/driver_chipcommon_pmu.c:607: warning: data definition has no type or storage class
drivers/ssb/driver_chipcommon_pmu.c:607: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL'
drivers/ssb/driver_chipcommon_pmu.c:607: warning: parameter names (without types) in function declaration

drivers/ssb/pcihost_wrapper.c: In function ‘ssb_pcihost_register’:
drivers/ssb/pcihost_wrapper.c:120: error: ‘THIS_MODULE’ undeclared (first use in this function)
drivers/ssb/pcihost_wrapper.c:120: error: (Each undeclared identifier is reported only once
drivers/ssb/pcihost_wrapper.c:120: error: for each function it appears in.)

drivers/ssb/driver_pcicore.c:721: warning: data definition has no type or storage class
drivers/ssb/driver_pcicore.c:721: warning: type defaults to ‘int’ in declaration of ‘EXPORT_SYMBOL’
drivers/ssb/driver_pcicore.c:721: warning: parameter names (without types) in function declaration

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agossb: Add module.h to the real modules in drivers/ssb
Paul Gortmaker [Sun, 17 Jul 2011 20:03:40 +0000 (16:03 -0400)]
ssb: Add module.h to the real modules in drivers/ssb

The header cleanup means that module.h is no longer simply
everywhere.  So real modules need to actively include it.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/media: Add module.h to all files using it implicitly
Paul Gortmaker [Sun, 3 Jul 2011 18:03:12 +0000 (14:03 -0400)]
drivers/media: Add module.h to all files using it implicitly

A pending cleanup will mean that module.h won't be implicitly
everywhere anymore.  Make sure the modular drivers in clocksource
are actually calling out for <module.h> explicitly in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agomedia: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required
Paul Gortmaker [Fri, 15 Jul 2011 16:02:56 +0000 (12:02 -0400)]
media: Add export.h for EXPORT_SYMBOL/THIS_MODULE as required

These two macros were in module.h but now module.h is no longer
sprayed across every source file imaginable, so the users need
to expicitly call out their use of them.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/char in ramoops.c fix implicit use of err.h
Paul Gortmaker [Thu, 28 Jul 2011 01:31:17 +0000 (21:31 -0400)]
drivers/char in ramoops.c fix implicit use of err.h

drivers/char/ramoops.c: In function 'ramoops_init':
drivers/char/ramoops.c:221: error: implicit declaration of function 'IS_ERR'
drivers/char/ramoops.c:222: error: implicit declaration of function 'PTR_ERR'
make[3]: *** [drivers/char/ramoops.o] Error 1

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/char: Add export.h for THIS_MODULE to ttyprintk.c
Paul Gortmaker [Sun, 10 Jul 2011 16:14:53 +0000 (12:14 -0400)]
drivers/char: Add export.h for THIS_MODULE to ttyprintk.c

It will need it called out explicitly in the near future due
to a module.h usage cleanup.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/char: Add module.h to those who were using it implicitly
Paul Gortmaker [Sun, 3 Jul 2011 17:35:48 +0000 (13:35 -0400)]
drivers/char: Add module.h to those who were using it implicitly

A pending cleanup will mean that module.h won't be implicitly
everywhere anymore.  Make sure the modular drivers in char are
actually calling out for <module.h> explicitly in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/clocksource: Add module.h to those who were using it implicitly
Paul Gortmaker [Sun, 3 Jul 2011 17:36:22 +0000 (13:36 -0400)]
drivers/clocksource: Add module.h to those who were using it implicitly

A pending cleanup will mean that module.h won't be implicitly
everywhere anymore.  Make sure the modular drivers in clocksource
are actually calling out for <module.h> explicitly in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/power: Add module.h to the actual modules in drivers/power
Paul Gortmaker [Sun, 3 Jul 2011 19:28:29 +0000 (15:28 -0400)]
drivers/power: Add module.h to the actual modules in drivers/power

The module.h header is no longer going to be implicitly present
everywhere.  So real modular users need to call out its use
explicitly in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agomemstick: Add module.h to the modular users in drivers/memstick
Paul Gortmaker [Sun, 3 Jul 2011 19:12:37 +0000 (15:12 -0400)]
memstick: Add module.h to the modular users in drivers/memstick

This is another group of drivers that simply assumed that module.h was
everywhere.  But it won't be once we clean up its presence from device.h
Call out the real users of it in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers: base.h implicitly depends on <linux/notifier.h>
Paul Gortmaker [Thu, 26 May 2011 22:08:35 +0000 (18:08 -0400)]
drivers: base.h implicitly depends on <linux/notifier.h>

This file is currently relying on <linux/module.h> sneaking it in
through the twisted include paths from device.h.  Once that
is cleaned up, this will happen:

In file included from drivers/base/init.c:12:
drivers/base/base.h:34: error: field ‘bus_notifier’ has incomplete type
make[3]: *** [drivers/base/init.o] Error 1

Fix it up in advance, so the cleanup can continue.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers: base/dma-coherent.c is a module and needs module.h
Paul Gortmaker [Fri, 1 Jul 2011 20:07:32 +0000 (16:07 -0400)]
drivers: base/dma-coherent.c is a module and needs module.h

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agodrivers/base: Fix various files needing export.h for EXPORT_SYMBOL
Paul Gortmaker [Fri, 27 May 2011 11:12:15 +0000 (07:12 -0400)]
drivers/base: Fix various files needing export.h for EXPORT_SYMBOL

Most of these files were implicitly getting EXPORT_SYMBOL via
device.h which was including module.h, but that path will be broken
soon.  Similar for wakeup.c; it needs EXPORT_SYMBOL and THIS_MODULE.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoide/ata: Add module.h to the implicit modular users
Paul Gortmaker [Sun, 3 Jul 2011 17:41:29 +0000 (13:41 -0400)]
ide/ata: Add module.h to the implicit modular users

A pending cleanup will mean that module.h won't be implicitly
everywhere anymore.  Make sure the modular drivers in the ide dir
are actually calling out for <module.h> explicitly in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoide/ata: Add export.h for EXPORT_SYMBOL/THIS_MODULE where needed
Paul Gortmaker [Sun, 17 Jul 2011 19:33:58 +0000 (15:33 -0400)]
ide/ata: Add export.h for EXPORT_SYMBOL/THIS_MODULE where needed

They were getting this implicitly by an include of module.h
from device.h -- but we are going to clean that up and break
that include chain, so include export.h explicitly now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoinfiniband: add in export.h for files using EXPORT_SYMBOL/THIS_MODULE
Paul Gortmaker [Fri, 27 May 2011 19:29:33 +0000 (15:29 -0400)]
infiniband: add in export.h for files using EXPORT_SYMBOL/THIS_MODULE

These were getting it implicitly via device.h --> module.h but
we are going to stop that when we clean up the headers.

Fix these in advance so the tree remains biscect-clean.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoinfiniband: Fix up module files that need to include module.h
Paul Gortmaker [Fri, 27 May 2011 19:35:46 +0000 (15:35 -0400)]
infiniband: Fix up module files that need to include module.h

They had been getting it implicitly via device.h but we can't
rely on that for the future, due to a pending cleanup so fix
it now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoinfiniband: Fix up users implicitly relying on getting stat.h
Paul Gortmaker [Fri, 27 May 2011 17:27:45 +0000 (13:27 -0400)]
infiniband: Fix up users implicitly relying on getting stat.h

They get it via module.h (via device.h) but we want to clean that up.
When we do, we'll get things like:

  CC [M]  drivers/infiniband/core/sysfs.o
  sysfs.c:361: error: 'S_IRUGO' undeclared here (not in a function)
  sysfs.c:654: error: 'S_IWUSR' undeclared here (not in a function)

so add in the stat header it is using explicitly in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agovideo: Add module.h to drivers/video files who really use it.
Paul Gortmaker [Sun, 3 Jul 2011 20:17:28 +0000 (16:17 -0400)]
video: Add module.h to drivers/video files who really use it.

They were getting this implicitly by an include of module.h
from device.h -- but we are going to clean that up and break
that include chain, so include module.h explicitly now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agovideo: Add export.h for THIS_MODULE/EXPORT_SYMBOL to drivers/video
Paul Gortmaker [Sun, 10 Jul 2011 17:20:26 +0000 (13:20 -0400)]
video: Add export.h for THIS_MODULE/EXPORT_SYMBOL to drivers/video

With module.h being implicitly everywhere via device.h, the absence
of explicitly including something for EXPORT_SYMBOL went unnoticed.
Since we are heading to fix things up and clean module.h from the
device.h file, we need to explicitly include these files now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoselinux: netlink.c uses THIS_MODULE; give it assoc. header
Paul Gortmaker [Fri, 27 May 2011 00:52:10 +0000 (20:52 -0400)]
selinux: netlink.c uses THIS_MODULE; give it assoc. header

The pervasive, but implicit presence of <linux/module.h> meant
that things like this file would happily compile as-is.  But
with the desire to phase out the module.h being included everywhere,
point this file at export.h which will give it THIS_MODULE.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agocpufreq: Fix build of s3c64xx cpufreq driver for header change
Mark Brown [Fri, 29 Jul 2011 15:19:26 +0000 (16:19 +0100)]
cpufreq: Fix build of s3c64xx cpufreq driver for header change

The header change has removed an implicit include of module.h, breaking
the build due to the use of THIS_MODULE. Fix that.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agocpufreq: cpufreq_stats.c is a module, and should include module.h
Paul Gortmaker [Fri, 27 May 2011 17:23:32 +0000 (13:23 -0400)]
cpufreq: cpufreq_stats.c is a module, and should include module.h

So that we can clean up the header files and not be relying
on implicit includes from device.h ---> module.h

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoi2c: i2c-pxa-pci.c needs module.h
Randy Dunlap [Sat, 30 Jul 2011 04:14:30 +0000 (21:14 -0700)]
i2c: i2c-pxa-pci.c needs module.h

i2c-pxa-pci.c uses interfaces from linux/module.h, so it should
include that file.  This fixes build errors.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoi2c: add export.h to i2c-boardinfo.c for EXPORT_SYMBOL
Paul Gortmaker [Fri, 27 May 2011 14:46:24 +0000 (10:46 -0400)]
i2c: add export.h to i2c-boardinfo.c for EXPORT_SYMBOL

With module.h being implicitly everywhere via device.h, the absence
of explicitly including something for EXPORT_SYMBOL went unnoticed.
Since we are heading to fix things up and clean module.h from the
device.h file, we need to explicitly include these files now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agortc: rtc-tx4939.c is modular and so needs module.h
Paul Gortmaker [Fri, 29 Jul 2011 00:23:50 +0000 (20:23 -0400)]
rtc: rtc-tx4939.c is modular and so needs module.h

Or else you'll get this:

drivers/rtc/rtc-tx4939.c:317: warning: type defaults to 'int' in declaration of 'MODULE_LICENSE'

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agortc: Add module.h to implicit users in drivers/rtc
Paul Gortmaker [Fri, 27 May 2011 13:57:25 +0000 (09:57 -0400)]
rtc: Add module.h to implicit users in drivers/rtc

The module.h was implicitly everywhere, but when we clean
that up, the implicit users will compile fail; fix them up
in advance.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agotty: Add export.h for EXPORT_SYMBOL/THIS_MODULE to exporters
Paul Gortmaker [Fri, 27 May 2011 14:46:24 +0000 (10:46 -0400)]
tty: Add export.h for EXPORT_SYMBOL/THIS_MODULE to exporters

With module.h being implicitly everywhere via device.h, the absence
of explicitly including something for EXPORT_SYMBOL went unnoticed.
Since we are heading to fix things up and clean module.h from the
device.h file, we need to explicitly include these files now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agotty: Add module.h to drivers/tty users who just expect it there.
Paul Gortmaker [Fri, 27 May 2011 20:14:23 +0000 (16:14 -0400)]
tty: Add module.h to drivers/tty users who just expect it there.

We are cleaning up the issue that means module.h is omnipresent.
These tty users are the people who implictly are relying on that.
Fix up the real users to call out the include that they really need.

In the case of jsm_driver.c file, it had moduleparam.h but that
isn't enough and it needs the full module.h

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agousb: fix implicit usage of gfp.h in host/xhci-hub.c
Paul Gortmaker [Mon, 18 Jul 2011 18:42:00 +0000 (14:42 -0400)]
usb: fix implicit usage of gfp.h in host/xhci-hub.c

To fix this build error on ARM:

drivers/usb/host/xhci-hub.c: In function 'xhci_stop_device':
drivers/usb/host/xhci-hub.c:261: error: 'GFP_NOIO' undeclared (first use in this function)
make[4]: *** [drivers/usb/host/xhci-hub.o] Error 1

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agousb: Add export.h for EXPORT_SYMBOL/THIS_MODULE where needed
Paul Gortmaker [Fri, 27 May 2011 13:56:31 +0000 (09:56 -0400)]
usb: Add export.h for EXPORT_SYMBOL/THIS_MODULE where needed

With module.h being implicitly everywhere via device.h, the absence
of explicitly including something for EXPORT_SYMBOL went unnoticed.
Since we are heading to fix things up and clean module.h from the
device.h file, we need to explicitly include these files now.

Use the lightweight version of the header that has just THIS_MODULE
and EXPORT_SYMBOL variants.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agousb: Add module.h to drivers/usb consumers who really use it.
Paul Gortmaker [Sun, 3 Jul 2011 20:09:31 +0000 (16:09 -0400)]
usb: Add module.h to drivers/usb consumers who really use it.

The situation up to this point meant that module.h was pretty
much everywhere, regardless of whether you asked for it or not.
We are fixing that, so give the USB folks who want it an actual
include of it.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoscsi: Fix up files implicitly depending on module.h inclusion
Paul Gortmaker [Fri, 27 May 2011 13:47:43 +0000 (09:47 -0400)]
scsi: Fix up files implicitly depending on module.h inclusion

The module.h header was implicitly present everywhere, so files
with no explicit include of the module infrastructure would build
anyway.  We are now fixing the include bird's nest, and so we need
to call out the module.h file that we need explicitly.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agoscsi: Fix files needing export.h since they use EXPORT_SYMBOL
Paul Gortmaker [Fri, 27 May 2011 13:37:25 +0000 (09:37 -0400)]
scsi: Fix files needing export.h since they use EXPORT_SYMBOL

They were implicitly getting it from device.h --> module.h but
we want to clean that up and untangle that mess.  So add the
minimal header for the exports.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agopci: add module.h to files implicitly relying on its presence.
Paul Gortmaker [Fri, 27 May 2011 13:42:30 +0000 (09:42 -0400)]
pci: add module.h to files implicitly relying on its presence.

These were getting module.h implicitly from device.h but we want
to clean that up, so we fix it here to avoid things like:

pci/slot.c: In function ‘pci_hp_create_module_link’:
pci/slot.c:383: error: ‘module_kset’ undeclared (first use in this function)

Similarly, rpadlpar_core.c is modular, so add module.h to its includes.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agopci: Fix files needing export.h since they use EXPORT_SYMBOL
Paul Gortmaker [Fri, 27 May 2011 13:37:25 +0000 (09:37 -0400)]
pci: Fix files needing export.h since they use EXPORT_SYMBOL

They were implicitly getting it from device.h --> module.h but
we want to clean that up and untangle that mess.  So add the
minimal header for the exports.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agosound: Add export.h for THIS_MODULE/EXPORT_SYMBOL where needed
Paul Gortmaker [Sun, 17 Jul 2011 19:08:17 +0000 (15:08 -0400)]
sound: Add export.h for THIS_MODULE/EXPORT_SYMBOL where needed

These aren't modules, but they do make use of these macros, so
they will need export.h to get that definition.  Previously,
they got it via the implicit module.h inclusion.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
13 years agosound: Add module.h to the previously silent sound users
Paul Gortmaker [Fri, 15 Jul 2011 16:38:28 +0000 (12:38 -0400)]
sound: Add module.h to the previously silent sound users

Lots of sound drivers were getting module.h via the implicit presence
of it in <linux/device.h> but we are going to clean that up.  So
fix up those users now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>