If NO_DMA=y:

    warning: (VIDEO_OMAP2_VOUT && VIDEO_VIU && VIDEO_TIMBERDALE) selects VIDEOBUF_DMA_CONTIG which has unmet direct dependencies (MEDIA_SUPPORT && HAS_DMA)

    drivers/built-in.o: In function `__videobuf_dc_free':
    videobuf-dma-contig.c:(.text+0x6f4d32): undefined reference to `dma_free_coherent'
    drivers/built-in.o: In function `__videobuf_dc_alloc':
    videobuf-dma-contig.c:(.text+0x6f4fe6): undefined reference to `dma_alloc_coherent'
    drivers/built-in.o: In function `__videobuf_mmap_mapper':
    videobuf-dma-contig.c:(.text+0x6f518e): undefined reference to `dma_free_coherent'

Commit 24482922 ("[media] timberdale: do not select TIMB_DMA")
dropped the dependency of VIDEO_TIMBERDALE on DMADEVICES, and thus the
implicit dependency on HAS_DMA.  VIDEO_TIMBERDALE selects
VIDEOBUF_DMA_CONTIG, which bypasses its dependency on HAS_DMA.  Make
VIDEO_TIMBERDALE depend on HAS_DMA to fix this.
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

If NO_DMA=y:

    warning: (VIDEO_AM437X_VPFE && VIDEO_DM365_VPFE && VIDEO_DT3155 && VIDEO_OMAP4) selects VIDEOBUF2_DMA_CONTIG which has unmet direct dependencies (MEDIA_SUPPORT && HAS_DMA)

    drivers/media/v4l2-core/videobuf2-dma-contig.c: In function ‘vb2_dc_mmap’:
    drivers/media/v4l2-core/videobuf2-dma-contig.c:207: error: implicit declaration of function ‘dma_mmap_coherent’
    drivers/media/v4l2-core/videobuf2-dma-contig.c: In function ‘vb2_dc_get_base_sgt’:
    drivers/media/v4l2-core/videobuf2-dma-contig.c:390: error: implicit declaration of function ‘dma_get_sgtable’

VIDEO_AM437X_VPFE selects VIDEOBUF2_DMA_CONTIG, which bypasses its
dependency on HAS_DMA.  Make VIDEO_AM437X_VPFE depend on HAS_DMA to fix
this.
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Acked-by: NLad, Prabhakar <prabhakar.csengg@gmail.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The if and the else code are identical - so the condition has no effect
on the effective code.
This patch removes the condition and the duplicated code.
Signed-off-by: NNicholas Mc Guire <hofrat@osadl.org>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

If the media controller (MC) is not enabled, it will compile
fine, but will fail at the linkedition:

 ERROR: "media_device_unregister" [drivers/media/usb/siano/smsusb.ko] undefined!
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

This is really a simple function, but using it avoids to have
if's inside the drivers.

Also, the kABI becomes a little more clearer.

This shouldn't generate any overhead, and the type check
will happen when compiling with MC DVB enabled.

So, let's do it.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

drivers/media/usb/dvb-usb-v2/dvb_usb_core.c: In function ‘dvb_usbv2_adapter_dvb_exit’:
drivers/media/usb/dvb-usb-v2/dvb_usb_core.c:531:25: warning: unused variable ‘d’ [-Wunused-variable]
  struct dvb_usb_device *d = adap_to_d(adap);
                         ^
drivers/media/usb/dvb-usb-v2/dvb_usb_core.c:403:13: warning: ‘dvb_usbv2_media_device_register’ defined but not used [-Wunused-function]
 static void dvb_usbv2_media_device_register(struct dvb_usb_adapter *adap)

drivers/media/usb/dvb-usb/dvb-usb-dvb.c:97:13: warning: ‘dvb_usb_media_device_register’ defined but not used [-Wunused-function]
 static void dvb_usb_media_device_register(struct dvb_usb_adapter *adap)
             ^
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Instead of using media_dev argument for dvb_create_media_graph(),
use the adapter.

That allows to create a stub for this function, if compiled
without DVB support, avoiding to add extra if's at the drivers.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Instead of assuming just one adapter, change the code to store
one media controller per adapter.

This works fine for dvb-usb, as, on all drivers here, it is not
possible to write a media graph that would mix resources between
the two different adapters.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Create a struct media_device and add it to the dvb adapter.

Please notice that the tuner is not mapped yet by the dvb core.

[mchehab@osg.samsung.com: use config option MEDIA_CONTROLLER_DVB]
Signed-off-by: NRafael Lourenço de Lima Chehab <chehabrafael@gmail.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Instead of assuming just one adapter, change the code to store
one media controller per adapter.

This works fine for dvb-usb, as, on all drivers here, it is not
possible to write a media graph that would mix resources between
the two different adapters.
Acked-by: NAntti Palosaari <crope@iki.fi>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Create a struct media_device and add it to the dvb adapter.

Please notice that the tuner is not mapped yet by the dvb core.

[mchehab@osg.samsung.com: use config option MEDIA_CONTROLLER_DVB due to a rebase]
Signed-off-by: NRafael Lourenço de Lima Chehab <chehabrafael@gmail.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

We need to initialize the media controller earlier, as the core
will call the smsdvb hotplug during register time. Ok, this is
an async operation, so, when the module is not loaded, the media
controller works.

However, if the module is already loaded, nothing will be
registered at the media controller, as it will load too late.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Right now, this is a debug message, misplaced. Promote it
to an info message, as it helps to discover if something
bad happened during device init.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

All siano modules have a sms_dbg parameter. Now that we're using
the standard pr_debug() macro, we can get rid of it.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

On most cases, sms_info() should actually be pr_debug(), but,
on other places, it should be pr_info().
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

There's no reason to use a macro here. Just replace everything,
and let those debug messages to be activated via dynamic printk.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Despite its name, those functions are acutally debug
prints for the IR part of the driver. So, properly
map them using pr_debug()
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Originally, sms_err() would be also displaying the line where
the error occurs, but the messages are clear enough. Also,
the function is always printed. So, no need for it.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

There's no reason for a sms' own sms_warn macro. Just replace
it by the standard pr_warn().
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Instead of defining its own set of printk functions, let's
use the common Kernel debug logic provided by pr_foo functions.

As a first step, let's just define the existing macros as the
Kernel ones.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Adding support for the media controller for a pure DVB device
is simple: just create a struct media_device and add it to the
dvb adapter. After creating all DVB devices, we need to call
the DVB core, for it to create the media graph.

More work is needed for pure DVB tuners, but this is hidden
at the Siano driver, just like several others non-hybrid
devices. So, this is streight forward.
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Use an enum to better document the media pads.

No functional changes.
Suggested-by: NPrabhakar Lad <prabhakar.csengg@gmail.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

There's no need to free state, as it was allocated via devm_kzalloc().

Also, let's return the error code, instead of something else.
Reported-by: NPrabhakar Lad <prabhakar.csengg@gmail.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

buf_prepare callback is called for every queued buffer. This is
an overkill. Call it at buf_setup, as this should be enough.
Acked-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

There are two problems at the comment:
- it is badly idented;
- its comment doesn't mean anything.

Fix it.
Requested-by: NHans Verkuil <hans.verkuil@cisco.com>
Acked-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Analog video inputs are the tuner, plus composite, svideo, etc,
 e. g. the input pat should actually be like:

                ___________
TUNER --------> |         |
                |         |
SVIDEO .......> | cx25840 |
                |         |
COMPOSITE1 ...> |_________|

(in the above, dashes represent the enabled link, and periods
represent the disabled ones)

In other words, if we want to properly represent the pipeline,
it should be possible to see via the media controller if the tuner
is being used as an image source, or if the source is something else.

I didn't map those other inputs here yet, due to a few things:
- The extra inputs would require subdevs that won't be controlled
- I was in doubt about the best way for doing that
- That would likely require some extra setup for cx25840 caller
  drivers, in order to represent what of the possible internal
  inputs are actually used on each specific board

Actually, at least for now, I was unable to see much benefit
on adding such map now, so let's just document it, as this could
be added later on, as needed.
Acked-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Prefix all DVB media controller entities with "dvb-" and use dash
instead of underline at the names.
Requested-by: NHans Verkuil <hans.verkuil@cisco.com>
Acked-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

drivers/media/v4l2-core/tuner-core.c:440:7: error: 'struct v4l2_subdev' has no member named 'entity'
     t->sd.entity.name = t->name;
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Acked-by: NLad, Prabhakar <prabhakar.csengg@gmail.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

drivers/media/usb/cx231xx/cx231xx-cards.c: In function ‘cx231xx_usb_probe’:
drivers/media/usb/cx231xx/cx231xx-cards.c:1589:15: error: ‘struct v4l2_device’ has no member named ‘mdev’
  dev->v4l2_dev.mdev = dev->media_dev;
               ^
drivers/media/usb/cx231xx/cx231xx-cards.c:1589:26: error: ‘struct cx231xx’ has no member named ‘media_dev’
  dev->v4l2_dev.mdev = dev->media_dev;
                          ^
scripts/Makefile.build:257: recipe for target 'drivers/media/usb/cx231xx/cx231xx-cards.o' failed
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

if CONFIG_MEDIA_CONTROLLER_DVB is not selected, it is now
producing this warning:

drivers/media/dvb-core/dvb_frontend.c: In function ‘dvb_frontend_thread’:
drivers/media/dvb-core/dvb_frontend.c:695:6: warning: unused variable ‘ret’ [-Wunused-variable]
  int ret;
      ^
Reported-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The alsa struct in struct media_entity_desc is now marked as deprecated.
However, the alsa struct should remain as it is since it cannot be replaced
by a simple major/minor device node description. The alsa struct was designed
to be used as an alsa card description so V4L2 drivers could use this to expose
the alsa card that they create to carry the captured audio. Such a card is not
just a PCM device, but also needs to contain the alsa subdevice information,
and it may map to multiple devices, e.g. a PCM and a mixer device, such as the
au0828 usb stick creates.

This is exactly as intended and this cannot and should not be replaced by a
simple major/minor.

However, whether this information is in the right form for an ALSA device such
that it can handle udev renaming rules as well is another matter. So mark this
alsa struct as TODO and document the problems involved.

Updated the documentation as well to reflect this and to add the 'major'
and 'minor' field documentation.

Updated the documentation to clearly state that struct dev is to be used for
(sub-)devices that create a single device node. Other devices need their own
structure here.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

The logic there tries to start the media entity even if it
doesn't exist, causing this bug:

	[  314.356162] BUG: unable to handle kernel NULL pointer dereference at 0000000000000010
	[  314.356202] IP: [<ffffffffa02ef74c>] media_entity_pipeline_start+0x1c/0x390 [media]
Reported-by: NGert-Jan van der Stroom <gjstroom@gmail.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Linux 34.0-rc1

* tag 'v4.0-rc1': (8947 commits)
  Linux 4.0-rc1
  autofs4 copy_dev_ioctl(): keep the value of ->size we'd used for allocation
  procfs: fix race between symlink removals and traversals
  debugfs: leave freeing a symlink body until inode eviction
  Documentation/filesystems/Locking: ->get_sb() is long gone
  trylock_super(): replacement for grab_super_passive()
  fanotify: Fix up scripted S_ISDIR/S_ISREG/S_ISLNK conversions
  Cachefiles: Fix up scripted S_ISDIR/S_ISREG/S_ISLNK conversions
  VFS: (Scripted) Convert S_ISLNK/DIR/REG(dentry->d_inode) to d_is_*(dentry)
  SELinux: Use d_is_positive() rather than testing dentry->d_inode
  Smack: Use d_is_positive() rather than testing dentry->d_inode
  TOMOYO: Use d_is_dir() rather than d_inode and S_ISDIR()
  Apparmor: Use d_is_positive/negative() rather than testing dentry->d_inode
  Apparmor: mediated_filesystem() should use dentry->d_sb not inode->i_sb
  VFS: Split DCACHE_FILE_TYPE into regular and special types
  VFS: Add a fallthrough flag for marking virtual dentries
  VFS: Add a whiteout dentry type
  VFS: Introduce inode-getting helpers for layered/unioned fs environments
  kernel: make READ_ONCE() valid on const arguments
  blk-throttle: check stats_cpu before reading it from sysfs
  ...

.. after extensive statistical analysis of my G+ polling, I've come to
the inescapable conclusion that internet polls are bad.

Big surprise.

But "Hurr durr I'ma sheep" trounced "I like online polls" by a 62-to-38%
margin, in a poll that people weren't even supposed to participate in.
Who can argue with solid numbers like that? 5,796 votes from people who
can't even follow the most basic directions?

In contrast, "v4.0" beat out "v3.20" by a slimmer margin of 56-to-44%,
but with a total of 29,110 votes right now.

Now, arguably, that vote spread is only about 3,200 votes, which is less
than the almost six thousand votes that the "please ignore" poll got, so
it could be considered noise.

But hey, I asked, so I'll honor the votes.

Pull ext4 fixes from Ted Ts'o:
 "Ext4 bug fixes.

  We also reserved code points for encryption and read-only images (for
  which the implementation is mostly just the reserved code point for a
  read-only feature :-)"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: fix indirect punch hole corruption
  ext4: ignore journal checksum on remount; don't fail
  ext4: remove duplicate remount check for JOURNAL_CHECKSUM change
  ext4: fix mmap data corruption in nodelalloc mode when blocksize < pagesize
  ext4: support read-only images
  ext4: change to use setup_timer() instead of init_timer()
  ext4: reserve codepoints used by the ext4 encryption feature
  jbd2: complain about descriptor block checksum errors

Pull more vfs updates from Al Viro:
 "Assorted stuff from this cycle.  The big ones here are multilayer
  overlayfs from Miklos and beginning of sorting ->d_inode accesses out
  from David"

* 'for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (51 commits)
  autofs4 copy_dev_ioctl(): keep the value of ->size we'd used for allocation
  procfs: fix race between symlink removals and traversals
  debugfs: leave freeing a symlink body until inode eviction
  Documentation/filesystems/Locking: ->get_sb() is long gone
  trylock_super(): replacement for grab_super_passive()
  fanotify: Fix up scripted S_ISDIR/S_ISREG/S_ISLNK conversions
  Cachefiles: Fix up scripted S_ISDIR/S_ISREG/S_ISLNK conversions
  VFS: (Scripted) Convert S_ISLNK/DIR/REG(dentry->d_inode) to d_is_*(dentry)
  SELinux: Use d_is_positive() rather than testing dentry->d_inode
  Smack: Use d_is_positive() rather than testing dentry->d_inode
  TOMOYO: Use d_is_dir() rather than d_inode and S_ISDIR()
  Apparmor: Use d_is_positive/negative() rather than testing dentry->d_inode
  Apparmor: mediated_filesystem() should use dentry->d_sb not inode->i_sb
  VFS: Split DCACHE_FILE_TYPE into regular and special types
  VFS: Add a fallthrough flag for marking virtual dentries
  VFS: Add a whiteout dentry type
  VFS: Introduce inode-getting helpers for layered/unioned fs environments
  Infiniband: Fix potential NULL d_inode dereference
  posix_acl: fix reference leaks in posix_acl_create
  autofs4: Wrong format for printing dentry
  ...

Pull ARM fix from Russell King:
 "Just one fix this time around.  __iommu_alloc_buffer() can cause a
  BUG() if dma_alloc_coherent() is called with either __GFP_DMA32 or
  __GFP_HIGHMEM set.  The patch from Alexandre addresses this"

* 'fixes' of git://ftp.arm.linux.org.uk/~rmk/linux-arm:
  ARM: 8305/1: DMA: Fix kzalloc flags in __iommu_alloc_buffer()

X-Coverup: just ask spender
Cc: stable@vger.kernel.org
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

use_pde()/unuse_pde() in ->follow_link()/->put_link() resp.

Cc: stable@vger.kernel.org
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

As it is, we have debugfs_remove() racing with symlink traversals.
Supply ->evict_inode() and do freeing there - inode will remain
pinned until we are done with the symlink body.

And rip the idiocy with checking if dentry is positive right after
we'd verified debugfs_positive(), which is a stronger check...

Cc: stable@vger.kernel.org
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>