Before, it was called from individual encoder (de)init procedures, which
lead to spare threads running (which were actually lost, leaked).
The current fix uses trivial approach, and the downside is that the
processing thread is working always, even when there's no consumer.
Signed-off-by: NAndrey Utkin <andrey.krieger.utkin@gmail.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

Fixes this warning:

[  956.730136] ------------[ cut here ]------------
[  956.730143] WARNING: CPU: 1 PID: 10134 at lib/dma-debug.c:963 dma_debug_device_change+0x191/0x1f0()
[  956.730146] pci 0000:07:05.0: DMA-API: device driver has pending DMA allocations while released from device [count=8]
One of leaked entries details: [device address=0x00000000d3d57000] [size=512 bytes] [mapped with DMA_BIDIRECTIONAL] [mapped as coherent]
[  956.730147] Modules linked in: solo6x10(-) videobuf2_dma_contig videobuf2_dma_sg videobuf2_memops videobuf2_core ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_nat_ipv4 nf_nat videobuf_vmalloc videobuf_core v4l2_common videodev rt2800usb rt2800lib rt2x00usb rt2x00lib mac80211 cfg80211 crc_ccitt usbkbd hid_a4tech hid_generic usbhid snd_hda_codec_hdmi snd_hda_codec_via snd_hda_codec_generic snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm x86_pkg_temp_thermal snd_timer snd soundcore
[  956.730172] CPU: 1 PID: 10134 Comm: rmmod Not tainted 3.18.0-rc1-next-20141023-zver-dirty #24
[  956.730173] Hardware name: System manufacturer System Product Name/P8H77-V, BIOS 0501 02/28/2012
[  956.730175]  0000000000000009 ffff8801df9e3c58 ffffffff817ffe6b 0000000000000001
[  956.730177]  ffff8801df9e3ca8 ffff8801df9e3c98 ffffffff81091ec7 0000000000000046
[  956.730180]  ffff880215457e90 0000000000000008 ffffffff81cbb10f ffff880215570098
[  956.730183] Call Trace:
[  956.730188]  [<ffffffff817ffe6b>] dump_stack+0x4f/0x7c
[  956.730192]  [<ffffffff81091ec7>] warn_slowpath_common+0x87/0xb0
[  956.730194]  [<ffffffff81091f91>] warn_slowpath_fmt+0x41/0x50
[  956.730197]  [<ffffffff81412558>] ? dma_debug_device_change+0xb8/0x1f0
[  956.730199]  [<ffffffff81412631>] dma_debug_device_change+0x191/0x1f0
[  956.730203]  [<ffffffff810b14ad>] notifier_call_chain+0x4d/0x70
[  956.730205]  [<ffffffff810b15f9>] __blocking_notifier_call_chain+0x59/0x80
[  956.730207]  [<ffffffff810b1631>] blocking_notifier_call_chain+0x11/0x20
[  956.730211]  [<ffffffff815873af>] __device_release_driver+0xcf/0xf0
[  956.730213]  [<ffffffff81587ee8>] driver_detach+0xc8/0xd0
[  956.730215]  [<ffffffff81587147>] bus_remove_driver+0x57/0xd0
[  956.730218]  [<ffffffff815887e9>] driver_unregister+0x29/0x60
[  956.730221]  [<ffffffff81420131>] pci_unregister_driver+0x21/0x90
[  956.730225]  [<ffffffffa03219d7>] solo_pci_driver_exit+0x10/0x12 [solo6x10]
[  956.730228]  [<ffffffff81112ee0>] SyS_delete_module+0x170/0x1f0
[  956.730232]  [<ffffffff813eb76e>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[  956.730234]  [<ffffffff8180abd2>] system_call_fastpath+0x12/0x17
[  956.730235] ---[ end trace e730af02713a6c53 ]---
[  956.730237] Mapped at:
[  956.730238]  [<ffffffff8141186c>] debug_dma_alloc_coherent+0x3c/0xb0
[  956.730240]  [<ffffffffa03203f6>] solo_enc_v4l2_init+0x706/0xba0 [solo6x10]
[  956.730243]  [<ffffffffa03165b3>] solo_pci_probe+0x503/0x700 [solo6x10]
[  956.730245]  [<ffffffff81420459>] local_pci_probe+0x49/0xa0
[  956.730248]  [<ffffffff814207a1>] pci_device_probe+0xd1/0x120
Signed-off-by: NAndrey Utkin <andrey.krieger.utkin@gmail.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

This fixes warning from drivers/media/v4l2-core/videobuf2-core.c,
WARN_ON(atomic_read(&q->owned_by_drv_count)).
Signed-off-by: NAndrey Utkin <andrey.krieger.utkin@gmail.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>

drivers/media/pci/solo6x10/solo6x10-disp.c:221 solo_set_motion_block() error: potential null dereference
'buf'.  (kzalloc returns null)

Also propagate this error up the chain.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Now that the custom motion detection API has been replaced with a
standard API there is no reason anymore to keep it in staging.

So (finally!) move it to drivers/media/pci.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Added a blank line after variable declarations where checkpatch
requested that, and removed the 'write to the FSF' paragraph,
again as requested.

This is in preparation of the move out of staging into drivers/media.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Previously, it was needed to reopen device to update GOP size and
quantization parameter. Now we update device registers with new values
immediately.
Signed-off-by: NAndrey Utkin <andrey.utkin@corp.bluecherry.net>
Signed-off-by: NIsmael Luceno <ismael.luceno@corp.bluecherry.net>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

solo6*10 boards have configurable quantization parameter which takes
values from 0 to 31, inclusively.

This change enables setting it with ioctl VIDIOC_S_CTRL with id
V4L2_CID_MPEG_VIDEO_H264_MIN_QP.
Signed-off-by: NAndrey Utkin <andrey.utkin@corp.bluecherry.net>
Signed-off-by: NIsmael Luceno <ismael.luceno@corp.bluecherry.net>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

The dma_map_sg() function can fail, so check for the return value.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Use the new motion detection event.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Replace the custom ioctls to set motion detection thresholds by standard
matrix controls.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Rather than having two unions for all types just keep 'val' and
'cur.val' and use the p_cur and p_new unions to access all others.

The only reason for keeping 'val' and 'cur.val' is that it is used
all over, so converting this as well would be a huge job.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Since all the drivers that use `struct v4l2_fh' use the core
priority checking, the setting of the flag in the drivers can
be removed.
Signed-off-by: NRamakrishnan Muthukrishnan <ramakrmu@cisco.com>
Reviewed-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

The vb2 core ignores any return code from the stop_streaming op.
And there really isn't anything it can do anyway in case of an error.
So change the return type to void and update any drivers that implement it.

The int return gave drivers the idea that this operation could actually
fail, but that's really not the case.

The pwc amd sdr-msi3101 drivers both had this construction:

        if (mutex_lock_interruptible(&s->v4l2_lock))
                return -ERESTARTSYS;

This has been updated to just call mutex_lock(). The stop_streaming op
expects this to really stop streaming and I very much doubt this will
work reliably if stop_streaming just returns without really stopping the
DMA.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Acked-by: NPawel Osciak <pawel@osciak.com>
Acked-by: NSakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

There are some expressions that compute the roundup of a number, but don't use
the existing macro defined in /include/kernel.h. This patch uses the following
Coccinelle semantic patch:
@ haskernel @
@@

@ depends on haskernel @
expression E1, E2;
@@

- (E1 + (E2 - 1)) & ~(E2 - 1)
+ ALIGN(E1, E2)
Signed-off-by: NAndreea-Cristina Bernat <bernat.ada@gmail.com>
Acked-by: NBob Copeland <me@bobcopeland.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

Fix checkpatch.pl issues with line over 80 characters in solo6x10-v4l2-enc.c
Signed-off-by: NAybuke Ozdemir <aybuke.147@gmail.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>

The timestamp_type field used to contain only the timestamp type. Soon it
will be used for timestamp source flags as well. Rename the field
accordingly.

[m.chehab@samsung.com: do the change also to drivers/staging/media and at s2255]
Signed-off-by: NSakari Ailus <sakari.ailus@iki.fi>
Acked-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

The video_type was never set correctly for PAL: it's not a bool, instead
it is a register value.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Tested-by: NHans Verkuil <hans.verkuil@cisco.com>
Reported-by: tomdev@freenet.de
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Mask v4l2_buf.flags to indicate only I/P/B frames.
Signed-off-by: NKrzysztof Ha?asa <khalasa@piap.pl>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

On certain platforms a sequence of dma_map_sg() and dma_unmap_sg()
discards data previously stored in the buffers. Build video headers
only after the DMA is completed.
Signed-off-by: NKrzysztof Ha?asa <khalasa@piap.pl>
[hans.verkuil@cisco.com: fix merge problems due to the recent solo sg_table changes]
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Signed-off-by: NKrzysztof Ha?asa <khalasa@piap.pl>
[hans.verkuil@cisco.com: fix merge problems due to the recent solo sg_table changes]
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Replace the private struct vb2_dma_sg_desc with the struct sg_table so
we can benefit from all the helping functions in lib/scatterlist.c for
things like allocating the sg or compacting the descriptor.
marvel-ccic and solo6x10 drivers, that use this API have been updated.
Acked-by: NMarek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: NAndre Heider <a.heider@gmail.com>
Signed-off-by: NRicardo Ribalda Delgado <ricardo.ribalda@gmail.com>
[s.nawrocki@samsung.com: minor corrections of the changelog]
Signed-off-by: NSylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: NMauro Carvalho Chehab <m.chehab@samsung.com>

Instead of falling back to 1/25 (PAL) or 1/30 (NTSC).
Signed-off-by: NIsmael Luceno <ismael.luceno@corp.bluecherry.net>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

The 6010 produces MPEG-4 part 2, while 6110 produces H.264.
Signed-off-by: NIsmael Luceno <ismael.luceno@corp.bluecherry.net>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NIsmael Luceno <ismael.luceno@corp.bluecherry.net>
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

ioctl's number is unsigned int. Fix it at vidioc_default.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Prefix the headers with 'solo6x10-'.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

An earlier patch temporarily disabled regional motion detection. This patch
adds it back: the 'Motion Detection Enable' control is now a 'Motion Detection Mode'.
And to set/get the regional thresholds two new ioctls were added to get/set those
thresholds.
The BUF_FLAG constants were also updated to prevent clashing with existing buffer
flags.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

FMT_MPEG is for multiplexed streams, not elementary streams. The same is
true for the V4L2_CID_MPEG_VIDEO_ENCODING control.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

As a consequence the ioctl op has been replaced by unlocked_ioctl.
Since we are now using the core lock the locking scheme has been
simplified as well.
The main reason for converting this driver to vb2 was that the locking
scheme in v4l2.c was hopeless. It was easier to just convert the driver
then to try and salvage a threading and videobuf nightmare.
The videobuf2 framework is far, far superior compared to the old videobuf.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

As a consequence the ioctl op has been replaced by unlocked_ioctl.
Since we are now using the core lock the locking scheme has been
simplified as well.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

All fields in solo_enc_fh do not belong there since they refer to global
properties. After moving all these fields to solo_enc_dev the solo_dev_fh
struct can be removed completely.
Note that this also kills the 'listener' feature of this driver. This
feature (where multiple filehandles can read the video) is illegal in the
V4L2 API. Do this in userspace: it's much more efficient to copy memory
than it is to DMA to every listener.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

- try_fmt should never return -EBUSY.
- invalid pix->field values were not mapped to a valid value.
- the priv field of struct v4l2_pix_format wasn't zeroed.
- the try_fmt error code was not checked in set_fmt.
- enum_framesizes/intervals is valid for both MJPEG and MPEG pixel formats.
- enum_frameintervals didn't check width and height and reported the
  wrong range.
- s_parm didn't set readbuffers.
- don't fail on invalid colorspace, just replace with the valid colorspace.
- bytesperline should be 0 for compressed formats.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Note that the MOTION_THRESHOLD functionality has been temporarily reduced:
only the global threshold can be set, not the per-block. This will be
addressed in a later patch: controls are not the proper way to do this.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Add support for struct v4l2_device.
Signed-off-by: NHans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>