- 17 Sep, 2018 33 commits
-
-
Philipp Zabel authored
Signed-off-by:
Philipp Zabel <p.zabel@pengutronix.de> Signed-off-by:
Marco Felsch <m.felsch@pengutronix.de> Signed-off-by:
Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
-
Philipp Zabel authored
Issue a V4L2_EVENT_SOURCE_CHANGE notification when the TVP5150 locks onto a signal and when it loses the lock. Signed-off-by:
-
Philipp Zabel authored
To avoid short frames on stream start, keep output pins at high impedance while we are not properly locked onto the input signal. Signed-off-by:
-
Philipp Zabel authored
This patch adds an optional interrupt handler to handle the sync lock interrupt and sync lock status. Signed-off-by:
Lucas Stach <l.stach@pengutronix.de> [m.felsch@pengutronix.de: move added .g_std callback to separate patch] Signed-off-by:
-
Philipp Zabel authored
To allow optional interrupt support, we want to configure the pin settings dynamically. Move those register accesses out of the static initialization tables. Signed-off-by:
-
Philipp Zabel authored
To trigger standard autodetection only the reset part of the routine is necessary during probe(). Split this out to make it callable on its own. [m.felsch@pengutronix.de: adapt commit message] [m.felsch@pengutronix.de: add tvp5150_enable() to tvp5150_s_stream()] [m.chehab@samsung.com: fix a compilation breakage] Signed-off-by:
-
Philipp Zabel authored
Make sure to not overwrite decoder->norm when setting the standard in hardware, but only when instructed by V4L2 API calls. Signed-off-by:
-
Philipp Zabel authored
If cropping isn't set explicitly by userspace, reset it to the maximum possible rectangle in subdevice open if a standard change is detected. Signed-off-by:
-
Marco Felsch authored
The patch adds three macros to bundle the mbus_framefmt default values and a helper function to set the the default crop and mbus_framefmt values. Signed-off-by:
-
Marco Felsch authored
Add two helper functions to check if two rectangles have the same position (top/left) and if two rectangles equals (same size and same position). Signed-off-by:
Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by:
-
Marco Felsch authored
Since commit 9a4c7e68f7e0 ("media: tvp5150: convert register access to regmap")' the driver supports regmap. Now we can drop the handmade bit update sequence and move to the regmap provided helpers. Signed-off-by: -
Philipp Zabel authored
Regmap provides built in debugging, caching and provides dedicated accessors for bit manipulations in registers, which make the following changes a lot simpler. Signed-off-by:
-
Marco Felsch authored
The function only consists of a single switch case block without a default case. Unsupported control requests are indicated by the -EINVAL return code trough the last return statement at the end of the function. So exiting just the switch case block returns the -EINVAL error code but the hue control is supported and a zero should be returned instead. Replace the break by a 'return 0' to fix this behaviour. Fixes: d183e4ef ("[media] v4l: tvp5150: Add missing break in set control handler") Signed-off-by:
-
Marco Felsch authored
The driver ignored the width alignment which exists due to the UYVY colorspace format. Fix the width alignment and make use of the the provided v4l2 helper function to set the width, height and all alignments in one. Fixes: 963ddc63 ("[media] media: tvp5150: Add cropping support") Signed-off-by:
-
Mauro Carvalho Chehab authored
Now that all drivers are using pad signal types, we can get rid of the global static definition, as routes are stablished using the pad signal type. The tuner and IF-PLL pads are now used only by the tuner core, so move the definitions to be there. Acked-by:
Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by:
-
Mauro Carvalho Chehab authored
As we don't need anymore to share pad numbers with similar drivers, use its own pad definition instead of a global model. Acked-by:
-
Mauro Carvalho Chehab authored
As we don't need anymore to share pad numbers with similar drivers, use its own pad definition instead of a global model. Acked-by:
-
Mauro Carvalho Chehab authored
As we don't need anymore to share pad numbers with similar drivers, use its own pad definition instead of a global model. Acked-by:
-
Mauro Carvalho Chehab authored
As we don't need anymore to share pad numbers with similar drivers, use its own pad definition instead of a global model. Acked-by:
-
Mauro Carvalho Chehab authored
As we don't need anymore to share pad numbers with similar drivers, use its own pad definition instead of a global model. Acked-by:
-
Mauro Carvalho Chehab authored
As we don't need anymore to share pad numbers with similar drivers, use its own pad definition instead of a global model. Acked-by:
-
Mauro Carvalho Chehab authored
As we don't need anymore to share pad numbers with similar drivers, use its own pad definition instead of a global model. Acked-by:
-
Mauro Carvalho Chehab authored
When creating the audio link, use pad signals, instead of hardcoding using the pad index number. Acked-by:
-
Mauro Carvalho Chehab authored
On tuner pads, multiple signals are present. Be sure to get the right PAD by using them. Acked-by:
-
Mauro Carvalho Chehab authored
It is not trivial to debug troubles at media graph build. So, add print messages to help debug what's happening, in the case of an error occurs. Signed-off-by: -
Mauro Carvalho Chehab authored
Instead of relying on a static map for pids, use the new sig_type "taint" type to setup the pipelines with the same tipe between different entities. Acked-by:
-
Mauro Carvalho Chehab authored
Consumer devices are provided with a wide different range of types supported by the same driver, allowing different configutations. In order to make easier to setup media controller links, "taint" pads with the signal type it carries. While here, get rid of DEMOD_PAD_VBI_OUT, as the signal it carries is actually the same as the normal video output. The difference happens at the video/VBI interface: - for VBI, only the hidden lines are streamed; - for video, the stream is usually cropped to hide the vbi lines. Acked-by:
-
Mauro Carvalho Chehab authored
The signal there is the same as the video output (well, except for sliced VBI, but let's simplify the model and ignore it, at least for now - as it is routed together with raw VBI). Acked-by:
-
Mauro Carvalho Chehab authored
The v4l2-compliance tool complains if a video doesn't start with a zero sequence number. While this shouldn't cause any real problem for apps, let's make it happier, in order to better check the v4l2-compliance differences before and after patchsets. This is actually an old issue. It is there since at least its videobuf2 conversion, e. g. changeset 3829fadc461 ("[media] em28xx: convert to videobuf2"), if VB1 wouldn't suffer from the same issue. Cc: stable@vger.kernel.org Fixes: d3829fad ("[media] em28xx: convert to videobuf2") Signed-off-by: -
Mauro Carvalho Chehab authored
Instead of using a register value, use an AMUX name, as otherwise VIDIOC_G_AUDIO would fail. Cc: stable@vger.kernel.org Fixes: 766ed64d ("V4L/DVB (11827): Add support for Terratec Grabster AV350") Signed-off-by:
-
Mauro Carvalho Chehab authored
Follow the V4L2 spec, as warned by v4l2-compliance: warn: v4l2-test-formats.cpp(732): TRY_FMT cannot handle an invalid pixelformat. warn: v4l2-test-formats.cpp(733): This may or may not be a problem. For more information see: warn: v4l2-test-formats.cpp(734): http://www.mail-archive.com/linux-media@vger.kernel.org/msg56550.html Cc: stable@vger.kernel.org Fixes: bddcf633 ("V4L/DVB (9927): em28xx: use a more standard way to specify video formats") Signed-off-by:
-
Mauro Carvalho Chehab authored
The a->index is not the name of the internal amux entry, but, instead a value from zero to the maximum number of audio inputs. As the actual available inputs depend on each board, build it dynamically. This is broken for a really long time. On a quick check, since at least commit 195a4ef6 ("V4L/DVB (6585): Convert em28xx to video_ioctl2") this was not implemented right. Fixes: 195a4ef6 ("V4L/DVB (6585): Convert em28xx to video_ioctl2") Cc: stable@vger.kernel.org Signed-off-by:
-
Mauro Carvalho Chehab authored
The parameters of v4l2_ctrl_new_std_menu_items() are tricky: instead of the number of possible values, it requires the number of the maximum value. In other words, the ARRAY_SIZE() value should be decremented, otherwise it will go past the array bounds, as warned by KASAN: [ 279.839688] BUG: KASAN: global-out-of-bounds in v4l2_querymenu+0x10d/0x180 [videodev] [ 279.839709] Read of size 8 at addr ffffffffc10a4cb0 by task v4l2-compliance/16676 [ 279.839736] CPU: 1 PID: 16676 Comm: v4l2-compliance Not tainted 4.18.0-rc2+ #120 [ 279.839741] Hardware name: /NUC5i7RYB, BIOS RYBDWi35.86A.0364.2017.0511.0949 05/11/2017 [ 279.839743] Call Trace: [ 279.839758] dump_stack+0x71/0xab [ 279.839807] ? v4l2_querymenu+0x10d/0x180 [videodev] [ 279.839817] print_address_description+0x1c9/0x270 [ 279.839863] ? v4l2_querymenu+0x10d/0x180 [videodev] [ 279.839871] kasan_report+0x237/0x360 [ 279.839918] v4l2_querymenu+0x10d/0x180 [videodev] [ 279.839964] __video_do_ioctl+0x2c8/0x590 [videodev] [ 279.840011] ? copy_overflow+0x20/0x20 [videodev] [ 279.840020] ? avc_ss_reset+0xa0/0xa0 [ 279.840028] ? check_stack_object+0x21/0x60 [ 279.840036] ? __check_object_size+0xe7/0x240 [ 279.840080] video_usercopy+0xed/0x730 [videodev] [ 279.840123] ? copy_overflow+0x20/0x20 [videodev] [ 279.840167] ? v4l_enumstd+0x40/0x40 [videodev] [ 279.840177] ? __handle_mm_fault+0x9f9/0x1ba0 [ 279.840186] ? __pmd_alloc+0x2c0/0x2c0 [ 279.840193] ? __vfs_write+0xb6/0x350 [ 279.840200] ? kernel_read+0xa0/0xa0 [ 279.840244] ? video_usercopy+0x730/0x730 [videodev] [ 279.840284] v4l2_ioctl+0xa1/0xb0 [videodev] [ 279.840295] do_vfs_ioctl+0x117/0x8a0 [ 279.840303] ? selinux_file_ioctl+0x211/0x2f0 [ 279.840313] ? ioctl_preallocate+0x120/0x120 [ 279.840319] ? selinux_capable+0x20/0x20 [ 279.840332] ksys_ioctl+0x70/0x80 [ 279.840342] __x64_sys_ioctl+0x3d/0x50 [ 279.840351] do_syscall_64+0x6d/0x1c0 [ 279.840361] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 279.840367] RIP: 0033:0x7fdfb46275d7 [ 279.840369] Code: b3 66 90 48 8b 05 b1 48 2d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 81 48 2d 00 f7 d8 64 89 01 48 [ 279.840474] RSP: 002b:00007ffee1179038 EFLAGS: 00000202 ORIG_RAX: 0000000000000010 [ 279.840483] RAX: ffffffffffffffda RBX: 00007ffee1179180 RCX: 00007fdfb46275d7 [ 279.840488] RDX: 00007ffee11790c0 RSI: 00000000c02c5625 RDI: 0000000000000003 [ 279.840493] RBP: 0000000000000002 R08: 0000000000000020 R09: 00000000009f0902 [ 279.840497] R10: 0000000000000000 R11: 0000000000000202 R12: 00007ffee117a5a0 [ 279.840501] R13: 00007ffee11790c0 R14: 0000000000000002 R15: 0000000000000000 [ 279.840515] The buggy address belongs to the variable: [ 279.840535] tvp5150_test_patterns+0x10/0xffffffffffffe360 [tvp5150] Fixes: c43875f6 ("[media] tvp5150: replace MEDIA_ENT_F_CONN_TEST by a control") Cc: stable@vger.kernel.org Signed-off-by:
-
- 12 Sep, 2018 7 commits
-
-
Gustavo A. R. Silva authored
Replace "falltrough" with a proper "fall through" annotation. This fix is part of the ongoing efforts to enabling -Wimplicit-fallthrough Signed-off-by:
Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by:
-
Rob Herring authored
In preparation to remove the node name pointer from struct device_node, convert printf users to use the %pOFn format specifier. Cc: "Lad, Prabhakar" <prabhakar.csengg@gmail.com> Cc: Benoit Parrot <bparrot@ti.com> Cc: Philipp Zabel <p.zabel@pengutronix.de> Cc: Hyun Kwon <hyun.kwon@xilinx.com> Cc: Michal Simek <michal.simek@xilinx.com> Signed-off-by:
Rob Herring <robh@kernel.org> Reviewed-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by:
-
Arnd Bergmann authored
The VIDEO_STILLPICTURE is only implemented by one driver, while VIDEO_GET_EVENT has two users in tree. In both cases, it is fairly easy to handle the compat ioctls in the native handler rather than relying on translation in fs/compat_ioctls. In effect, this means that now the drivers implement both structure layouts in both native and compat mode, but I don't see anything wrong with that. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Arnd Bergmann authored
Most DVB audio and video ioctl commands are completely compatible, and are implemented by just two drivers: ttpci and ivtv. In both cases, we can use the same ioctl handler for both native and compat ioctl handling, and remove the entries from the global lookup table. In case of ttpci, this directly hooks into the file_operations structure, and for ivtv, we have to set the compat_ioctl32 method in v4l2_file_operations. For all I can tell, setting it to video_ioctl2 will still do the right thing for all commands. Note that for the VIDEO_STILLPICTURE and VIDEO_GET_EVENT commands, a translation handler in fs/compat_ioctl.c is still used. This works because the command numbers are different on 32-bit systems. Signed-off-by:
-
Arnd Bergmann authored
All the CEC ioctls are compatible, and they are only implemented in one driver, so we can simply let this driver handle them natively. Signed-off-by:
-
Arnd Bergmann authored
All dmx ioctls are compatible, and they are only implemented in one file, so we can replace the list of commands in fs/compat_ioctl.c with a single line in dmxdev. Signed-off-by:
-
Arnd Bergmann authored
The VIDEO_GET_EVENT and VIDEO_STILLPICTURE was added back in 2005 but it never worked because the command number is wrong. Using the right command number means we have a better chance of them actually doing the right thing, though clearly nobody has ever tried it successfully. I noticed these while auditing the remaining users of compat_time_t for y2038 bugs. This one is fine in that regard, it just never did anything. Fixes: 6e87abd0 ("[DVB]: Add compat ioctl handling.") Signed-off-by:
-
