platform/kernel/linux-rpi.git
3 years agoio_uring: fix flush cqring overflow list while TASK_INTERRUPTIBLE
Hao Xu [Wed, 27 Jan 2021 07:14:09 +0000 (15:14 +0800)]
io_uring: fix flush cqring overflow list while TASK_INTERRUPTIBLE

Abaci reported the follow warning:

[   27.073425] do not call blocking ops when !TASK_RUNNING; state=1 set at [] prepare_to_wait_exclusive+0x3a/0xc0
[   27.075805] WARNING: CPU: 0 PID: 951 at kernel/sched/core.c:7853 __might_sleep+0x80/0xa0
[   27.077604] Modules linked in:
[   27.078379] CPU: 0 PID: 951 Comm: a.out Not tainted 5.11.0-rc3+ #1
[   27.079637] Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011
[   27.080852] RIP: 0010:__might_sleep+0x80/0xa0
[   27.081835] Code: 65 48 8b 04 25 80 71 01 00 48 8b 90 c0 15 00 00 48 8b 70 18 48 c7 c7 08 39 95 82 c6 05 f9 5f de 08 01 48 89 d1 e8 00 c6 fa ff  0b eb bf 41 0f b6 f5 48 c7 c7 40 23 c9 82 e8 f3 48 ec 00 eb a7
[   27.084521] RSP: 0018:ffffc90000fe3ce8 EFLAGS: 00010286
[   27.085350] RAX: 0000000000000000 RBX: ffffffff82956083 RCX: 0000000000000000
[   27.086348] RDX: ffff8881057a0000 RSI: ffffffff8118cc9e RDI: ffff88813bc28570
[   27.087598] RBP: 00000000000003a7 R08: 0000000000000001 R09: 0000000000000001
[   27.088819] R10: ffffc90000fe3e00 R11: 00000000fffef9f0 R12: 0000000000000000
[   27.089819] R13: 0000000000000000 R14: ffff88810576eb80 R15: ffff88810576e800
[   27.091058] FS:  00007f7b144cf740(0000) GS:ffff88813bc00000(0000) knlGS:0000000000000000
[   27.092775] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   27.093796] CR2: 00000000022da7b8 CR3: 000000010b928002 CR4: 00000000003706f0
[   27.094778] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   27.095780] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[   27.097011] Call Trace:
[   27.097685]  __mutex_lock+0x5d/0xa30
[   27.098565]  ? prepare_to_wait_exclusive+0x71/0xc0
[   27.099412]  ? io_cqring_overflow_flush.part.101+0x6d/0x70
[   27.100441]  ? lockdep_hardirqs_on_prepare+0xe9/0x1c0
[   27.101537]  ? _raw_spin_unlock_irqrestore+0x2d/0x40
[   27.102656]  ? trace_hardirqs_on+0x46/0x110
[   27.103459]  ? io_cqring_overflow_flush.part.101+0x6d/0x70
[   27.104317]  io_cqring_overflow_flush.part.101+0x6d/0x70
[   27.105113]  io_cqring_wait+0x36e/0x4d0
[   27.105770]  ? find_held_lock+0x28/0xb0
[   27.106370]  ? io_uring_remove_task_files+0xa0/0xa0
[   27.107076]  __x64_sys_io_uring_enter+0x4fb/0x640
[   27.107801]  ? rcu_read_lock_sched_held+0x59/0xa0
[   27.108562]  ? lockdep_hardirqs_on_prepare+0xe9/0x1c0
[   27.109684]  ? syscall_enter_from_user_mode+0x26/0x70
[   27.110731]  do_syscall_64+0x2d/0x40
[   27.111296]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[   27.112056] RIP: 0033:0x7f7b13dc8239
[   27.112663] Code: 01 00 48 81 c4 80 00 00 00 e9 f1 fe ff ff 0f 1f 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05  3d 01 f0 ff ff 73 01 c3 48 8b 0d 27 ec 2c 00 f7 d8 64 89 01 48
[   27.115113] RSP: 002b:00007ffd6d7f5c88 EFLAGS: 00000286 ORIG_RAX: 00000000000001aa
[   27.116562] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f7b13dc8239
[   27.117961] RDX: 000000000000478e RSI: 0000000000000000 RDI: 0000000000000003
[   27.118925] RBP: 00007ffd6d7f5cb0 R08: 0000000020000040 R09: 0000000000000008
[   27.119773] R10: 0000000000000001 R11: 0000000000000286 R12: 0000000000400480
[   27.120614] R13: 00007ffd6d7f5d90 R14: 0000000000000000 R15: 0000000000000000
[   27.121490] irq event stamp: 5635
[   27.121946] hardirqs last  enabled at (5643): [] console_unlock+0x5c4/0x740
[   27.123476] hardirqs last disabled at (5652): [] console_unlock+0x4e7/0x740
[   27.125192] softirqs last  enabled at (5272): [] __do_softirq+0x3c5/0x5aa
[   27.126430] softirqs last disabled at (5267): [] asm_call_irq_on_stack+0xf/0x20
[   27.127634] ---[ end trace 289d7e28fa60f928 ]---

This is caused by calling io_cqring_overflow_flush() which may sleep
after calling prepare_to_wait_exclusive() which set task state to
TASK_INTERRUPTIBLE

Reported-by: Abaci <abaci@linux.alibaba.com>
Fixes: 6c503150ae33 ("io_uring: patch up IOPOLL overflow_flush sync")
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Hao Xu <haoxu@linux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agoRevert "block: simplify set_init_blocksize" to regain lost performance
Maxim Mikityanskiy [Tue, 26 Jan 2021 19:59:07 +0000 (21:59 +0200)]
Revert "block: simplify set_init_blocksize" to regain lost performance

The cited commit introduced a serious regression with SATA write speed,
as found by bisecting. This patch reverts this commit, which restores
write speed back to the values observed before this commit.

The performance tests were done on a Helios4 NAS (2nd batch) with 4 HDDs
(WD8003FFBX) using dd (bs=1M count=2000). "Direct" is a test with a
single HDD, the rest are different RAID levels built over the first
partitions of 4 HDDs. Test results are in MB/s, R is read, W is write.

                | Direct | RAID0 | RAID10 f2 | RAID10 n2 | RAID6
----------------+--------+-------+-----------+-----------+--------
9011495c9466    | R:256  | R:313 | R:276     | R:313     | R:323
(before faulty) | W:254  | W:253 | W:195     | W:204     | W:117
----------------+--------+-------+-----------+-----------+--------
5ff9f19231a0    | R:257  | R:398 | R:312     | R:344     | R:391
(faulty commit) | W:154  | W:122 | W:67.7    | W:66.6    | W:67.2
----------------+--------+-------+-----------+-----------+--------
5.10.10         | R:256  | R:401 | R:312     | R:356     | R:375
unpatched       | W:149  | W:123 | W:64      | W:64.1    | W:61.5
----------------+--------+-------+-----------+-----------+--------
5.10.10         | R:255  | R:396 | R:312     | R:340     | R:393
patched         | W:247  | W:274 | W:220     | W:225     | W:121

Applying this patch doesn't hurt read performance, while improves the
write speed by 1.5x - 3.5x (more impact on RAID tests). The write speed
is restored back to the state before the faulty commit, and even a bit
higher in RAID tests (which aren't HDD-bound on this device) - that is
likely related to other optimizations done between the faulty commit and
5.10.10 which also improved the read speed.

Signed-off-by: Maxim Mikityanskiy <maxtram95@gmail.com>
Fixes: 5ff9f19231a0 ("block: simplify set_init_blocksize")
Cc: Christoph Hellwig <hch@lst.de>
Cc: Jens Axboe <axboe@kernel.dk>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agomedia: venus: venc: set IDR period to FW only for H264 & HEVC
Dikshita Agarwal [Wed, 16 Dec 2020 07:06:50 +0000 (08:06 +0100)]
media: venus: venc: set IDR period to FW only for H264 & HEVC

HFI_PROPERTY_CONFIG_VENC_IDR_PERIOD is supported for only
H264 & HEVC codec. There is no need to set it for VP8 since
all key frames are treated as IDR frames for VP8.

Signed-off-by: Dikshita Agarwal <dikshita@codeaurora.org>
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: venus: venc: set inband mode property to FW.
Dikshita Agarwal [Fri, 8 Jan 2021 07:23:37 +0000 (08:23 +0100)]
media: venus: venc: set inband mode property to FW.

Set HFI_PROPERTY_CONFIG_VENC_SYNC_FRAME_SEQUENCE_HEADER to FW
to support inband sequence header mode.

Reviewed-by: Fritz Koenig <frkoenig@chromium.org>
Signed-off-by: Dikshita Agarwal <dikshita@codeaurora.org>
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: venus: venc: fix handlig of S_SELECTION and G_SELECTION
Dikshita Agarwal [Thu, 5 Nov 2020 05:24:56 +0000 (06:24 +0100)]
media: venus: venc: fix handlig of S_SELECTION and G_SELECTION

- return correct width and height for G_SELECTION
- update capture port wxh with rectangle wxh.
- add support for HFI_PROPERTY_PARAM_UNCOMPRESSED_PLANE_ACTUAL_INFO
  to set stride info and chroma offset to FW.

Signed-off-by: Dikshita Agarwal <dikshita@codeaurora.org>
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: venus: pm_helpers: Control core power domain manually
Stanimir Varbanov [Sat, 9 Jan 2021 07:21:30 +0000 (08:21 +0100)]
media: venus: pm_helpers: Control core power domain manually

Presently we use device_link to control core power domain. But this
leads to issues because the genpd doesn't guarantee synchronous on/off
for supplier devices. Switch to manually control by pmruntime calls.

Tested-by: Fritz Koenig <frkoenig@chromium.org>
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: platform: Remove depends on interconnect
Georgi Djakov [Wed, 16 Sep 2020 11:09:47 +0000 (13:09 +0200)]
media: platform: Remove depends on interconnect

The dependency on interconnect in the Kconfig was introduced to avoid
the case of interconnect=m and driver=y, but the interconnect framework
has been converted from tristate to bool now. Remove the dependency as
the framework can't be a module anymore.

Signed-off-by: Georgi Djakov <georgi.djakov@linaro.org>
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: venus: core: add support to dump FW region
Dikshita Agarwal [Tue, 15 Dec 2020 18:08:36 +0000 (19:08 +0100)]
media: venus: core: add support to dump FW region

Add support to dump video FW region during FW crash
using devcoredump helpers.

Signed-off-by: Dikshita Agarwal <dikshita@codeaurora.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: venus: Mark bufreq_enc with static keyword
Zou Wei [Thu, 14 Jan 2021 09:05:21 +0000 (10:05 +0100)]
media: venus: Mark bufreq_enc with static keyword

Fix the following sparse warning:

hfi_plat_bufs_v6.c:1242:5: warning: symbol 'bufreq_enc' was not declared.
 Should it be static?

Signed-off-by: Zou Wei <zou_wei@huawei.com>
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: venus: Fix uninitialized variable count being checked for zero
Colin Ian King [Fri, 15 Jan 2021 10:11:26 +0000 (11:11 +0100)]
media: venus: Fix uninitialized variable count being checked for zero

In the case where plat->codecs is NULL the variable count is uninitialized
but is being checked to see if it is 0. Fix this by initializing
count to 0.

Addresses-Coverity: ("Uninitialized scalar variable")

Fixes: e29929266be1 ("media: venus: Get codecs and capabilities from hfi platform")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Stanimir Varbanov <stanimir.varbanov@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: add support for HEVC encoding
Michael Tretter [Thu, 3 Dec 2020 11:01:06 +0000 (12:01 +0100)]
media: allegro: add support for HEVC encoding

The Allegro Codec supports HEVC encoding. The messages to the MCU are
the same for H.264 and HEVC, but some options have to be changed. These
are actually only a few options.

The driver, however, must add the HEVC VPS/SPS/PPS NAL Units to the
coded stream and must properly provide the HEVC format and controls to
user space.

[hverkuil: fix warning for wrong enum type (h264 instead of hevc)]

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: activate v4l2-ctrls only for current codec
Michael Tretter [Thu, 3 Dec 2020 11:01:05 +0000 (12:01 +0100)]
media: allegro: activate v4l2-ctrls only for current codec

These controls are specific to H.264 and shall only be activated, if the
coded format is H.264.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: increase offset in CAPTURE buffer
Michael Tretter [Thu, 3 Dec 2020 11:01:04 +0000 (12:01 +0100)]
media: allegro: increase offset in CAPTURE buffer

The HEVC Non-VCL NAL units require more space than the H.264 Non-VCL NAL
units. Therefore, the driver needs to reserve more space in front of the
actual coded data that is written by the hardware codec.

Increase the offset that shall be used by the hardware codec from 64
bytes to 128 bytes which is as arbitrary as before.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: rename codec specific functions
Michael Tretter [Thu, 3 Dec 2020 11:01:03 +0000 (12:01 +0100)]
media: allegro: rename codec specific functions

The maximum bitrate and the size of the CPB are dependent on the level.
As the levels differ between the different codecs, the functions to get
the maximum bitrate and CPB must be codec specific, too.

Rename the functions to make it obvious that the function are only valid
for H.264.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: add helper to get entropy mode
Michael Tretter [Thu, 3 Dec 2020 11:01:02 +0000 (12:01 +0100)]
media: allegro: add helper to get entropy mode

The driver only supports CAVLC for H.264, but HEVC only uses CABAC. As
the driver has to explicitly tell the MCU to use CABAC for HEVC, add a
helper function to get the entropy mode.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: use accessor functions for QP values
Michael Tretter [Thu, 3 Dec 2020 11:01:01 +0000 (12:01 +0100)]
media: allegro: use accessor functions for QP values

V4L2 specifies different controls for the QP values of different codecs.

Simplify users that just care for the QP values by providing accessor
function that return the correct control based on the currently set
codec.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: remove profile and level from channel
Michael Tretter [Thu, 3 Dec 2020 11:01:00 +0000 (12:01 +0100)]
media: allegro: remove profile and level from channel

The profile and level are straight copies from the v4l2-ctrls. Avoid
duplication and directly read the value of the v4l2-ctrl.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: remove cpb_size and gop_size from channel
Michael Tretter [Thu, 3 Dec 2020 11:00:59 +0000 (12:00 +0100)]
media: allegro: remove cpb_size and gop_size from channel

The cpb_size and the gop_size are straight copies of the values in the
v4l2-ctrls. To avoid this duplication, directly get the values from the
v4l2-ctrls and remove the copies.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: implement scaling of cpb size in SPS
Michael Tretter [Thu, 3 Dec 2020 11:00:58 +0000 (12:00 +0100)]
media: allegro: implement scaling of cpb size in SPS

The cbp_size_value_minus1 in the SPS can scaled using cpb_size_scale to
reduce the number of bits necessary to encode the value. For simplicity,
the scaling was set to 1.

Restructure to the code to make it easier to drop the cbp_size from the
channel and as we are at it, also properly implement the scaling.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: initialize bitrate using v4l2_ctrl
Michael Tretter [Thu, 3 Dec 2020 11:00:57 +0000 (12:00 +0100)]
media: allegro: initialize bitrate using v4l2_ctrl

As the driver now uses the v4l2-ctrls to setup the channel, there is no
need to explicitly set the bitrate. The initial bitrate is now set via
the same path as if it is set from userspace using the v4l2-ctrl.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: use handler_setup to configure channel
Michael Tretter [Thu, 3 Dec 2020 11:00:56 +0000 (12:00 +0100)]
media: allegro: use handler_setup to configure channel

v4l2_ctrl_handler_setup() calls s_ctrl for all controls of the handler.
This ensures that the channel is initialized using the default values of
the v4l2-ctrls.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: fix log2_max_poc in firmware 2019.1
Michael Tretter [Thu, 3 Dec 2020 11:00:55 +0000 (12:00 +0100)]
media: allegro: fix log2_max_poc in firmware 2019.1

The log2_max_poc field is used to set log2_max_pic_order_cnt_lsb_minus4
for the coded stream. It has an allowed range of 0 to 12.

param contains the value without the minus4, but since firmware
version 2019.1, the value has to be log2_max_pic_order_cnt_lsb - 1,
presumably to fit the maximum value of 16 into a 4 bit field.

The driver does not support firmware version 2019.1. Thus, change the
behaviour starting from firmware version 2019.2.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: move encoding options to channel
Michael Tretter [Thu, 3 Dec 2020 11:00:54 +0000 (12:00 +0100)]
media: allegro: move encoding options to channel

There are several encoding options that are hard coded in the parameter
that is used to configure the hardware codec. However, some of the
options must be written to the SPS/PPS by the driver. Furthermore, some
of the options depend on the codec that is used (i.e. H.264 or HEVC).

Therefore, move options that depend on the codec to the channel and add
constants for options that are independent of the codec but must be
written to the SPS/PPS.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: adjust channel after format change
Michael Tretter [Thu, 3 Dec 2020 11:00:53 +0000 (12:00 +0100)]
media: allegro: adjust channel after format change

A format change (i.e. the frame size or the codec) has multiple effects
on a channel:

- The available controls

- The limits of the available controls

- The default encoding options

To avoid scattering the changes all over the driver, add a new function
that goes over the channel and does all required adjustments.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: implement S_FMT for CAPTURE
Michael Tretter [Fri, 15 Jan 2021 09:34:59 +0000 (10:34 +0100)]
media: allegro: implement S_FMT for CAPTURE

In order to support different codecs, the driver must support changing
the format on CAPTURE. Therefore, the driver needs to handle S_FMT on
CAPTURE.

As the driver will have a different number of formats for OUTPUT and
CAPTURE, split the check for the format index in ENUM_FMT into CAPTURE
and OUTPUT.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: add HEVC NAL unit generator
Michael Tretter [Thu, 3 Dec 2020 11:00:51 +0000 (12:00 +0100)]
media: allegro: add HEVC NAL unit generator

When encoding a video as HEVC, the allegro driver needs to generate the
Non-VCL NAL units for HEVC. Do the same as for H.264 and add a module
that takes C structs for the VPS/SPS/PPS and encodes the fields as RBPS
as specified by "ITU-T Rec. H.265 (02/2018) high efficiency video
coding".

[hverkuil: add missing static for nal_hevc_write/read_start_code_prefix]
[hverkuil: fix typo: pps -> vps in nal_hevc_write_vps]

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: add helper to report unsupported fields
Michael Tretter [Thu, 3 Dec 2020 11:00:50 +0000 (12:00 +0100)]
media: allegro: add helper to report unsupported fields

Allow generators to explicitly signal an error if the C structs contain
unsupported or invalid fields.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: allegro: extract RBSP handler from H.264 NAL generator
Michael Tretter [Thu, 3 Dec 2020 11:00:49 +0000 (12:00 +0100)]
media: allegro: extract RBSP handler from H.264 NAL generator

The RBSP structure is the same for HEVC and H.264. In order to be able
to reuse the RBSP handler for generating HEVC NAL units, extract the
functions from the H.264 generator.

Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: dt-bindings: media: ov5647: Fix filename
Jacopo Mondi [Fri, 15 Jan 2021 10:16:52 +0000 (11:16 +0100)]
media: dt-bindings: media: ov5647: Fix filename

Commit 1b5071af8240 ("media: dt-bindings: media: i2c: Rename
ov5647.yaml") renamed the bindings file but did not update the
Id: field there.

Fix it by using the new filename.

Fixes: 1b5071af8240 ("media: dt-bindings: media: i2c: Rename ov5647.yaml")
Signed-off-by: Jacopo Mondi <jacopo@jmondi.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: v4l2-async: Safely unregister an non-registered async subdev
Laurent Pinchart [Thu, 7 Jan 2021 22:54:58 +0000 (23:54 +0100)]
media: v4l2-async: Safely unregister an non-registered async subdev

Make the V4L2 async framework a bit more robust by allowing to
unregister a non-registered async subdev. Otherwise the
v4l2_async_cleanup() will attempt to delete the async subdev from the
subdev_list with the corresponding list_head not initialized.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: i2c/Kconfig: Select FWNODE for OV772x sensor
Lad Prabhakar [Wed, 20 Jan 2021 09:01:48 +0000 (10:01 +0100)]
media: i2c/Kconfig: Select FWNODE for OV772x sensor

Fix OV772x build breakage by selecting V4L2_FWNODE config:

ia64-linux-ld: drivers/media/i2c/ov772x.o: in function `ov772x_probe':
ov772x.c:(.text+0x1ee2): undefined reference to `v4l2_fwnode_endpoint_alloc_parse'
ia64-linux-ld: ov772x.c:(.text+0x1f12): undefined reference to `v4l2_fwnode_endpoint_free'
ia64-linux-ld: ov772x.c:(.text+0x2212): undefined reference to `v4l2_fwnode_endpoint_alloc_parse'

Fixes: 8a10b4e3601e ("media: i2c: ov772x: Parse endpoint properties")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Acked-by: Jacopo Mondi <jacopo+renesas@jmondi.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: v4l2-async: Add waiting subdevices debugfs
Ezequiel Garcia [Fri, 8 Jan 2021 17:17:28 +0000 (18:17 +0100)]
media: v4l2-async: Add waiting subdevices debugfs

There is currently little to no information available about the reasons
why a v4l2-async device hasn't probed completely.

Inspired by the "devices_deferred" debugfs file, add a file to list
information about the subdevices that are on waiting lists, for each
notifier.

This is useful to debug v4l2-async subdevices and notifiers, for instance
when doing device bring-up.

For instance, a typical output would be:

$ cat /sys/kernel/debug/video4linux/pending_async_subdevices
ipu1_csi1:
 [fwnode] dev=20e0000.iomuxc-gpr:ipu1_csi1_mux, node=/soc/bus@2000000/iomuxc-gpr@20e0000/ipu1_csi1_mux
ipu1_csi0:
 [fwnode] dev=20e0000.iomuxc-gpr:ipu1_csi0_mux, node=/soc/bus@2000000/iomuxc-gpr@20e0000/ipu1_csi0_mux
imx6-mipi-csi2:
 [fwnode] dev=1-003c, node=/soc/bus@2100000/i2c@21a4000/camera@3c
imx-media:

Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agox86/xen: avoid warning in Xen pv guest with CONFIG_AMD_MEM_ENCRYPT enabled
Juergen Gross [Mon, 25 Jan 2021 13:42:07 +0000 (14:42 +0100)]
x86/xen: avoid warning in Xen pv guest with CONFIG_AMD_MEM_ENCRYPT enabled

When booting a kernel which has been built with CONFIG_AMD_MEM_ENCRYPT
enabled as a Xen pv guest a warning is issued for each processor:

[    5.964347] ------------[ cut here ]------------
[    5.968314] WARNING: CPU: 0 PID: 1 at /home/gross/linux/head/arch/x86/xen/enlighten_pv.c:660 get_trap_addr+0x59/0x90
[    5.972321] Modules linked in:
[    5.976313] CPU: 0 PID: 1 Comm: swapper/0 Tainted: G        W         5.11.0-rc5-default #75
[    5.980313] Hardware name: Dell Inc. OptiPlex 9020/0PC5F7, BIOS A05 12/05/2013
[    5.984313] RIP: e030:get_trap_addr+0x59/0x90
[    5.988313] Code: 42 10 83 f0 01 85 f6 74 04 84 c0 75 1d b8 01 00 00 00 c3 48 3d 00 80 83 82 72 08 48 3d 20 81 83 82 72 0c b8 01 00 00 00 eb db <0f> 0b 31 c0 c3 48 2d 00 80 83 82 48 ba 72 1c c7 71 1c c7 71 1c 48
[    5.992313] RSP: e02b:ffffc90040033d38 EFLAGS: 00010202
[    5.996313] RAX: 0000000000000001 RBX: ffffffff82a141d0 RCX: ffffffff8222ec38
[    6.000312] RDX: ffffffff8222ec38 RSI: 0000000000000005 RDI: ffffc90040033d40
[    6.004313] RBP: ffff8881003984a0 R08: 0000000000000007 R09: ffff888100398000
[    6.008312] R10: 0000000000000007 R11: ffffc90040246000 R12: ffff8884082182a8
[    6.012313] R13: 0000000000000100 R14: 000000000000001d R15: ffff8881003982d0
[    6.016316] FS:  0000000000000000(0000) GS:ffff888408200000(0000) knlGS:0000000000000000
[    6.020313] CS:  e030 DS: 0000 ES: 0000 CR0: 0000000080050033
[    6.024313] CR2: ffffc900020ef000 CR3: 000000000220a000 CR4: 0000000000050660
[    6.028314] Call Trace:
[    6.032313]  cvt_gate_to_trap.part.7+0x3f/0x90
[    6.036313]  ? asm_exc_double_fault+0x30/0x30
[    6.040313]  xen_convert_trap_info+0x87/0xd0
[    6.044313]  xen_pv_cpu_up+0x17a/0x450
[    6.048313]  bringup_cpu+0x2b/0xc0
[    6.052313]  ? cpus_read_trylock+0x50/0x50
[    6.056313]  cpuhp_invoke_callback+0x80/0x4c0
[    6.060313]  _cpu_up+0xa7/0x140
[    6.064313]  cpu_up+0x98/0xd0
[    6.068313]  bringup_nonboot_cpus+0x4f/0x60
[    6.072313]  smp_init+0x26/0x79
[    6.076313]  kernel_init_freeable+0x103/0x258
[    6.080313]  ? rest_init+0xd0/0xd0
[    6.084313]  kernel_init+0xa/0x110
[    6.088313]  ret_from_fork+0x1f/0x30
[    6.092313] ---[ end trace be9ecf17dceeb4f3 ]---

Reason is that there is no Xen pv trap entry for X86_TRAP_VC.

Fix that by adding a generic trap handler for unknown traps and wire all
unknown bare metal handlers to this generic handler, which will just
crash the system in case such a trap will ever happen.

Fixes: 0786138c78e793 ("x86/sev-es: Add a Runtime #VC Exception Handler")
Cc: <stable@vger.kernel.org> # v5.10
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
3 years agomedia: v4l2-async: Remove V4L2_ASYNC_MATCH_CUSTOM
Ezequiel Garcia [Fri, 8 Jan 2021 17:17:27 +0000 (18:17 +0100)]
media: v4l2-async: Remove V4L2_ASYNC_MATCH_CUSTOM

Custom/driver-specific v4l2-async match support was introduced
in 2013, as V4L2_ASYNC_BUS_CUSTOM.

This type of match never had any user, so it's fair
to conclude it's not required and that safe for removal.
If the support is ever needed, it can always be restored.

Suggested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Rename debug functions
Joe Perches [Wed, 23 Dec 2020 13:35:27 +0000 (14:35 +0100)]
media: uvcvideo: Rename debug functions

trace isn't a good name as it's not a trace mechanism, it is a
typical debug mechanism.

Rename uvc_trace/uvc_trace_cont macros to uvc_dbg/uvc_dbg_cont.
Rename uvc_trace_param to uvc_dbg_param
Rename UVC_TRACE_<FOO> defines to UVC_DBG_<FOO>
Use ## concatenation in uvc_dbg macros to avoid overly long and
repetitive UVC_DBG uses

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: use dev_printk() for uvc_trace()
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:26 +0000 (14:35 +0100)]
media: uvcvideo: use dev_printk() for uvc_trace()

Instead of calling prink() inside uvc_trace, use dev_printk(), which adds
context to the output.

Now that we are at it, regroup the strings so the messages can be easily
"grepable".

Suggested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: New macro uvc_trace_cont
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:25 +0000 (14:35 +0100)]
media: uvcvideo: New macro uvc_trace_cont

Remove all the duplicated code around printk(KERN_CONT, with a new macro.

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Use dev_ printk aliases
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:24 +0000 (14:35 +0100)]
media: uvcvideo: Use dev_ printk aliases

Replace all the uses of printk() and uvc_printk() with its
equivalent dev_ alias macros.

Modify uvc_warn_once() macro to use dev_info instead printk().

They are more standard across the kernel tree and provide
more context about the error.

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Add Privacy control based on EXT_GPIO
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:23 +0000 (14:35 +0100)]
media: uvcvideo: Add Privacy control based on EXT_GPIO

Add a new control and mapping for Privacy controls connected to
UVC_GUID_EXT_GPIO_CONTROLLERs.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Implement UVC_EXT_GPIO_UNIT
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:22 +0000 (14:35 +0100)]
media: uvcvideo: Implement UVC_EXT_GPIO_UNIT

Some devices can implement a physical switch to disable the input of the
camera on demand. Think of it like an elegant privacy sticker.

The system can read the status of the privacy switch via a GPIO.

It is important to know the status of the switch, e.g. to notify the
user when the camera will produce black frames and a videochat
application is used.

In some systems, the GPIO is connected to the main SoC instead of the
camera controller, with the connection reported by the system firmware
(ACPI or DT). In that case, the UVC device isn't aware of the GPIO. We
need to implement a virtual entity to handle the GPIO fully on the
driver side.

For example, for ACPI-based systems, the GPIO is reported in the USB
device object:

  Scope (\_SB.PCI0.XHCI.RHUB.HS07)
  {

  /.../

    Name (_CRS, ResourceTemplate ()  // _CRS: Current Resource Settings
    {
        GpioIo (Exclusive, PullDefault, 0x0000, 0x0000, IoRestrictionOutputOnly,
            "\\_SB.PCI0.GPIO", 0x00, ResourceConsumer, ,
            )
            {   // Pin list
                0x0064
            }
    })
    Name (_DSD, Package (0x02)  // _DSD: Device-Specific Data
    {
        ToUUID ("daffd814-6eba-4d8c-8a91-bc9bbf4aa301") /* Device Properties for _DSD */,
        Package (0x01)
        {
            Package (0x02)
            {
                "privacy-gpio",
                Package (0x04)
                {
                    \_SB.PCI0.XHCI.RHUB.HS07,
                    Zero,
                    Zero,
                    One
                }
            }
        }
    })
  }

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Allow entity-defined get_info and get_cur
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:21 +0000 (14:35 +0100)]
media: uvcvideo: Allow entity-defined get_info and get_cur

Allows controls to get their properties and current value
from an entity-defined function instead of via a query to the USB
device.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Provide sync and async uvc_ctrl_status_event
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:20 +0000 (14:35 +0100)]
media: uvcvideo: Provide sync and async uvc_ctrl_status_event

Split the functionality of void uvc_ctrl_status_event_work in two, so it
can be called by functions outside interrupt context and not part of an
URB.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Allow entities with no pads
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:19 +0000 (14:35 +0100)]
media: uvcvideo: Allow entities with no pads

Avoid an underflow while calculating the number of inputs for entities
with zero pads.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Allow extra entities
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:18 +0000 (14:35 +0100)]
media: uvcvideo: Allow extra entities

Increase the size of the id, to avoid collisions with entities
implemented by the driver that are not part of the UVC device.

Entities exposed by the UVC device use IDs 0-255, extra entities
implemented by the driver (such as the GPIO entity) use IDs 256 and
up.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Move guid to entity
Ricardo Ribalda [Wed, 23 Dec 2020 13:35:17 +0000 (14:35 +0100)]
media: uvcvideo: Move guid to entity

Instead of having multiple copies of the entity guid on the code, move
it to the entity structure.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Accept invalid bFormatIndex and bFrameIndex values
Laurent Pinchart [Sun, 20 Dec 2020 14:11:13 +0000 (15:11 +0100)]
media: uvcvideo: Accept invalid bFormatIndex and bFrameIndex values

The Renkforce RF AC4K 300 Action Cam 4K reports invalid bFormatIndex and
bFrameIndex values when negotiating the video probe and commit controls.
The UVC descriptors report a single supported format and frame size,
with bFormatIndex and bFrameIndex both equal to 2, but the video probe
and commit controls report bFormatIndex and bFrameIndex set to 1.

The device otherwise operates correctly, but the driver rejects the
values and fails the format try operation. Fix it by ignoring the
invalid indices, and assuming that the format and frame requested by the
driver are accepted by the device.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=210767
Fixes: 8a652a17e3c0 ("media: uvcvideo: Ensure all probed info is returned to v4l2")
Reported-by: Till Dörges <doerges@pre-sense.de>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: uvcvideo: Force UVC version to 1.0a for 1bcf:0b40
Laurent Pinchart [Fri, 20 Nov 2020 10:57:48 +0000 (11:57 +0100)]
media: uvcvideo: Force UVC version to 1.0a for 1bcf:0b40

The Shenzhen Aoni Electronic Co.,Ltd 2K FHD camera reports a UVC 1.10
version, but implements UVC 1.0a as shown by the UVC probe control being
26 bytes long. Force the UVC version for that device.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reported-by: Doncho Minkov <donchominkov@gmail.com>
Tested-by: Doncho Minkov <donchominkov@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: zr364xx: fix memory leaks in probe()
Dan Carpenter [Thu, 21 Jan 2021 06:44:00 +0000 (07:44 +0100)]
media: zr364xx: fix memory leaks in probe()

Syzbot discovered that the probe error handling doesn't clean up the
resources allocated in zr364xx_board_init().  There are several
related bugs in this code so I have re-written the error handling.

1)  Introduce a new function zr364xx_board_uninit() which cleans up
    the resources in zr364xx_board_init().
2)  In zr364xx_board_init() if the call to zr364xx_start_readpipe()
    fails then release the "cam->buffer.frame[i].lpvbits" memory
    before returning.  This way every function either allocates
    everything successfully or it cleans up after itself.
3)  Re-write the probe function so that each failure path goto frees
    the most recent allocation.  That way we don't free anything
    before it has been allocated and we can also verify that
    everything is freed.
4)  Originally, in the probe function the "cam->v4l2_dev.release"
    pointer was set to "zr364xx_release" near the start but I moved
    that assignment to the end, after everything had succeeded.  The
    release function was never actually called during the probe cleanup
    process, but with this change I wanted to make it clear that we
    don't want to call zr364xx_release() until everything is
    allocated successfully.

Next I re-wrote the zr364xx_release() function.  Ideally this would
have been a simple matter of copy and pasting the cleanup code from
probe and adding an additional call to video_unregister_device().  But
there are a couple quirks to note.

1)  The probe function does not call videobuf_mmap_free() and I don't
    know where the videobuf_mmap is allocated.  I left the code as-is to
    avoid introducing a bug in code I don't understand.
2)  The zr364xx_board_uninit() has a call to zr364xx_stop_readpipe()
    which is a change from the original behavior with regards to
    unloading the driver.  Calling zr364xx_stop_readpipe() on a stopped
    pipe is not a problem so this is safe and is potentially a bugfix.

Reported-by: syzbot+b4d54814b339b5c6bbd4@syzkaller.appspotmail.com
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: imx6-mipi-csi2: Call remote subdev get_mbus_config to get active lanes
Ezequiel Garcia [Sat, 16 Jan 2021 16:40:38 +0000 (17:40 +0100)]
media: imx6-mipi-csi2: Call remote subdev get_mbus_config to get active lanes

Currently, the CSI2 subdevice is using the data-lanes from the
neareast endpoint to config the CSI2 lanes.

While this may work, the proper way to configure the hardware is
to obtain the remote subdevice in v4l2_async_notifier_operations.bound(),
and then call get_mbus_config using the remote subdevice to get
the active lanes.

Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: smipcie: switch from 'pci_' to 'dma_' API
Christophe JAILLET [Fri, 15 Jan 2021 21:34:03 +0000 (22:34 +0100)]
media: smipcie: switch from 'pci_' to 'dma_' API

The wrappers in include/linux/pci-dma-compat.h should go away.

The patch has been generated with the coccinelle script below and has been
hand modified to replace GFP_ with a correct flag.
It has been compile tested.

When memory is allocated in 'smi_port_init()' GFP_KERNEL can be used
because this function is called only from the probe function and no lock
is taken in between.

The call chain is:
  smi_probe()
    --> smi_port_attach()
      --> smi_port_init()

@@
@@
-    PCI_DMA_BIDIRECTIONAL
+    DMA_BIDIRECTIONAL

@@
@@
-    PCI_DMA_TODEVICE
+    DMA_TO_DEVICE

@@
@@
-    PCI_DMA_FROMDEVICE
+    DMA_FROM_DEVICE

@@
@@
-    PCI_DMA_NONE
+    DMA_NONE

@@
expression e1, e2, e3;
@@
-    pci_alloc_consistent(e1, e2, e3)
+    dma_alloc_coherent(&e1->dev, e2, e3, GFP_)

@@
expression e1, e2, e3;
@@
-    pci_zalloc_consistent(e1, e2, e3)
+    dma_alloc_coherent(&e1->dev, e2, e3, GFP_)

@@
expression e1, e2, e3, e4;
@@
-    pci_free_consistent(e1, e2, e3, e4)
+    dma_free_coherent(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_map_single(e1, e2, e3, e4)
+    dma_map_single(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_unmap_single(e1, e2, e3, e4)
+    dma_unmap_single(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4, e5;
@@
-    pci_map_page(e1, e2, e3, e4, e5)
+    dma_map_page(&e1->dev, e2, e3, e4, e5)

@@
expression e1, e2, e3, e4;
@@
-    pci_unmap_page(e1, e2, e3, e4)
+    dma_unmap_page(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_map_sg(e1, e2, e3, e4)
+    dma_map_sg(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_unmap_sg(e1, e2, e3, e4)
+    dma_unmap_sg(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_dma_sync_single_for_cpu(e1, e2, e3, e4)
+    dma_sync_single_for_cpu(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_dma_sync_single_for_device(e1, e2, e3, e4)
+    dma_sync_single_for_device(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_dma_sync_sg_for_cpu(e1, e2, e3, e4)
+    dma_sync_sg_for_cpu(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_dma_sync_sg_for_device(e1, e2, e3, e4)
+    dma_sync_sg_for_device(&e1->dev, e2, e3, e4)

@@
expression e1, e2;
@@
-    pci_dma_mapping_error(e1, e2)
+    dma_mapping_error(&e1->dev, e2)

@@
expression e1, e2;
@@
-    pci_set_dma_mask(e1, e2)
+    dma_set_mask(&e1->dev, e2)

@@
expression e1, e2;
@@
-    pci_set_consistent_dma_mask(e1, e2)
+    dma_set_coherent_mask(&e1->dev, e2)

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: hdmi: cec: replace broken link to HDMI specs
ivan tkachenko [Wed, 23 Dec 2020 21:43:54 +0000 (22:43 +0100)]
media: hdmi: cec: replace broken link to HDMI specs

Current link died, according to Wayback Machine, back in 2017. And the
website is completely down since 2019. Moreover, there was a custom
cover on that PDF, i.e. it was modified.

According to HDMI licence agreement (LA), HDMI specification and technical
information are supposed to be hosted on www.hdmi.org exclusively, and not
redistributed by third-parties.

Sure, there are still many more or less reliable "mirrors" out there with
a direct download straight from a search engine's page. However, for
example, from FPGA4fun[1] website it was removed "per HDMI LA request".

Unfortunately, the official download page is protected by email CAPTCHA,
but that seems to be the only legit way to obtain a copy.

[1] https://www.fpga4fun.com/HDMI.html

Signed-off-by: ivan tkachenko <me@ratijas.tk>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: Fix RTL2832 not depending on REGMAP_I2C
Maxim Plotnikov [Tue, 1 Dec 2020 00:14:27 +0000 (01:14 +0100)]
media: Fix RTL2832 not depending on REGMAP_I2C

Prevents this compile time error:

ERROR: modpost: "__regmap_init_i2c" [drivers/media/dvb-frontends/rtl2832.ko] undefined!

Signed-off-by: Maxim Plotnikov <wgh@torlan.ru>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: rcar-vin: Route events to correct video device
Niklas Söderlund [Fri, 15 Jan 2021 00:21:46 +0000 (01:21 +0100)]
media: rcar-vin: Route events to correct video device

The event route for VIN running with a media controller (Gen3) is
incorrect as all events are only routed to the video device that are
used to register the async notifier.

Remedy this be examining which subdevice generated the event and route
it to all VIN(s) that are connected to that subdevice.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Jacopo Mondi <jacopo+renesas@jmondi.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: rcar-vin: Do not try to stop stream if not running
Niklas Söderlund [Fri, 15 Jan 2021 00:21:45 +0000 (01:21 +0100)]
media: rcar-vin: Do not try to stop stream if not running

Do not attempt to stop the streaming if the stream is not running.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Jacopo Mondi <jacopo+renesas@jmondi.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: videobuf2-v4l2: remove redundant error test
Helen Koike [Thu, 14 Jan 2021 18:01:49 +0000 (19:01 +0100)]
media: videobuf2-v4l2: remove redundant error test

request_fd is validated under media_request_get_by_fd() just below this
check. Thus remove it.

Suggested-by: Tomasz Figa <tfiga@chromium.org>
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: v4l2-ioctl: print capabilities in v4l_print_create_buffers()
Helen Koike [Thu, 14 Jan 2021 18:01:47 +0000 (19:01 +0100)]
media: v4l2-ioctl: print capabilities in v4l_print_create_buffers()

Print capabilities field from struct v4l2_create_buffers for better
debugging.

Signed-off-by: Helen Koike <helen.koike@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: pxa_camera: declare variable when DEBUG is defined
Tom Rix [Mon, 18 Jan 2021 13:45:13 +0000 (14:45 +0100)]
media: pxa_camera: declare variable when DEBUG is defined

When DEBUG is defined this error occurs

drivers/media/platform/pxa_camera.c:1410:7: error:
  ‘i’ undeclared (first use in this function)
  for (i = 0; i < vb->num_planes; i++)
       ^
The variable 'i' is missing, so declare it.

Fixes: 6f28435d1c15 ("[media] media: platform: pxa_camera: trivial move of functions")
Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: mtk-vcodec: fix argument used when DEBUG is defined
Tom Rix [Sun, 17 Jan 2021 22:21:38 +0000 (23:21 +0100)]
media: mtk-vcodec: fix argument used when DEBUG is defined

When DEBUG is defined this error occurs

drivers/media/platform/mtk-vcodec/mtk_vcodec_enc_drv.c:306:41:
  error: ‘i’ undeclared (first use in this function)
  mtk_v4l2_debug(2, "reg[%d] base=0x%p", i, dev->reg_base[VENC_SYS]);

Reviewing the old line

mtk_v4l2_debug(2, "reg[%d] base=0x%p", i, dev->reg_base[i]);

All the i's need to be changed to VENC_SYS.
Fix a similar error for VENC_LT_SYS.

Fixes: 0dc4b3286125 ("media: mtk-vcodec: venc: support SCP firmware")
Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: cx25821: Fix a bug when reallocating some dma memory
Christophe JAILLET [Sat, 16 Jan 2021 21:21:46 +0000 (22:21 +0100)]
media: cx25821: Fix a bug when reallocating some dma memory

This function looks like a realloc.

However, if 'risc->cpu != NULL', the memory will be freed, but never
reallocated with the bigger 'size'.
Explicitly set 'risc->cpu' to NULL, so that the reallocation is
correctly performed a few lines below.

[hverkuil: NULL != risc->cpu -> risc->cpu]

Fixes: 5ede94c70553 ("[media] cx25821: remove bogus btcx_risc dependency)
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: saa7164: switch from 'pci_' to 'dma_' API
Christophe JAILLET [Fri, 15 Jan 2021 21:24:15 +0000 (22:24 +0100)]
media: saa7164: switch from 'pci_' to 'dma_' API

The wrappers in include/linux/pci-dma-compat.h should go away.

The patch has been generated with the coccinelle script below and has been
hand modified to replace GFP_ with a correct flag.
It has been compile tested.

When memory is allocated in 'saa7164_buffer_alloc()' GFP_KERNEL can be used
because this function is already using this flag just a few lines above.

@@
@@
-    PCI_DMA_BIDIRECTIONAL
+    DMA_BIDIRECTIONAL

@@
@@
-    PCI_DMA_TODEVICE
+    DMA_TO_DEVICE

@@
@@
-    PCI_DMA_FROMDEVICE
+    DMA_FROM_DEVICE

@@
@@
-    PCI_DMA_NONE
+    DMA_NONE

@@
expression e1, e2, e3;
@@
-    pci_alloc_consistent(e1, e2, e3)
+    dma_alloc_coherent(&e1->dev, e2, e3, GFP_)

@@
expression e1, e2, e3;
@@
-    pci_zalloc_consistent(e1, e2, e3)
+    dma_alloc_coherent(&e1->dev, e2, e3, GFP_)

@@
expression e1, e2, e3, e4;
@@
-    pci_free_consistent(e1, e2, e3, e4)
+    dma_free_coherent(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_map_single(e1, e2, e3, e4)
+    dma_map_single(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_unmap_single(e1, e2, e3, e4)
+    dma_unmap_single(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4, e5;
@@
-    pci_map_page(e1, e2, e3, e4, e5)
+    dma_map_page(&e1->dev, e2, e3, e4, e5)

@@
expression e1, e2, e3, e4;
@@
-    pci_unmap_page(e1, e2, e3, e4)
+    dma_unmap_page(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_map_sg(e1, e2, e3, e4)
+    dma_map_sg(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_unmap_sg(e1, e2, e3, e4)
+    dma_unmap_sg(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_dma_sync_single_for_cpu(e1, e2, e3, e4)
+    dma_sync_single_for_cpu(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_dma_sync_single_for_device(e1, e2, e3, e4)
+    dma_sync_single_for_device(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_dma_sync_sg_for_cpu(e1, e2, e3, e4)
+    dma_sync_sg_for_cpu(&e1->dev, e2, e3, e4)

@@
expression e1, e2, e3, e4;
@@
-    pci_dma_sync_sg_for_device(e1, e2, e3, e4)
+    dma_sync_sg_for_device(&e1->dev, e2, e3, e4)

@@
expression e1, e2;
@@
-    pci_dma_mapping_error(e1, e2)
+    dma_mapping_error(&e1->dev, e2)

@@
expression e1, e2;
@@
-    pci_set_dma_mask(e1, e2)
+    dma_set_mask(&e1->dev, e2)

@@
expression e1, e2;
@@
-    pci_set_consistent_dma_mask(e1, e2)
+    dma_set_coherent_mask(&e1->dev, e2)

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: mtk-vcodec: Do not zero reserved fields
Ricardo Ribalda [Tue, 12 Jan 2021 12:20:53 +0000 (13:20 +0100)]
media: mtk-vcodec: Do not zero reserved fields

Core code already clears reserved fields of struct
v4l2_pix_format_mplane, check commit 4e1e0eb0e074 ("media: v4l2-ioctl: Zero
v4l2_plane_pix_format reserved fields").

[hverkuil: drop unused variable i]

Cc: Matthias Brugger <matthias.bgg@gmail.com>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: sun4i-csi: Do not zero reserved fields
Ricardo Ribalda [Tue, 12 Jan 2021 12:20:52 +0000 (13:20 +0100)]
media: sun4i-csi: Do not zero reserved fields

Core code already clears reserved fields of struct
v4l2_pix_format_mplane, check commit 4e1e0eb0e074 ("media: v4l2-ioctl: Zero
v4l2_plane_pix_format reserved fields").

Cc: Maxime Ripard <mripard@kernel.org>
Cc: Chen-Yu Tsai <wens@csie.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agoACPI/IORT: Do not blindly trust DMA masks from firmware
Moritz Fischer [Fri, 22 Jan 2021 01:24:19 +0000 (17:24 -0800)]
ACPI/IORT: Do not blindly trust DMA masks from firmware

Address issue observed on real world system with suboptimal IORT table
where DMA masks of PCI devices would get set to 0 as result.

iort_dma_setup() would query the root complex'/named component IORT
entry for a DMA mask, and use that over the one the device has been
configured with earlier.

Ideally we want to use the minimum mask of what the IORT contains for
the root complex and what the device was configured with.

Fixes: 5ac65e8c8941 ("ACPI/IORT: Support address size limit for root complexes")
Signed-off-by: Moritz Fischer <mdf@kernel.org>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Acked-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Link: https://lore.kernel.org/r/20210122012419.95010-1-mdf@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
3 years agomedia: vicodec: Do not zero reserved fields
Ricardo Ribalda [Mon, 11 Jan 2021 14:54:44 +0000 (15:54 +0100)]
media: vicodec: Do not zero reserved fields

Core code already clears reserved fields of struct
v4l2_pix_format_mplane, check commit 4e1e0eb0e074 ("media: v4l2-ioctl: Zero
v4l2_plane_pix_format reserved fields").

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: ti-vpe: Do not zero reserved fields
Ricardo Ribalda [Mon, 11 Jan 2021 14:54:43 +0000 (15:54 +0100)]
media: ti-vpe: Do not zero reserved fields

Core code already clears reserved fields of struct
v4l2_pix_format_mplane, check commit 4e1e0eb0e074 ("media: v4l2-ioctl: Zero
v4l2_plane_pix_format reserved fields").

Cc: Benoit Parrot <bparrot@ti.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: jpu: Do not zero reserved fields
Ricardo Ribalda [Mon, 11 Jan 2021 14:54:41 +0000 (15:54 +0100)]
media: jpu: Do not zero reserved fields

Core code already clears reserved fields of struct
v4l2_pix_format_mplane, check commit 4e1e0eb0e074 ("media: v4l2-ioctl: Zero
v4l2_plane_pix_format reserved fields").

Cc: Mikhail Ulyanov <mikhail.ulyanov@cogentembedded.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: fdp1: Do not zero reserved fields
Ricardo Ribalda [Mon, 11 Jan 2021 14:54:40 +0000 (15:54 +0100)]
media: fdp1: Do not zero reserved fields

Core code already clears reserved fields of struct
v4l2_pix_format_mplane, check commit 4e1e0eb0e074 ("media: v4l2-ioctl: Zero
v4l2_plane_pix_format reserved fields").

Cc: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: mtk-mdp: Do not zero reserved fields
Ricardo Ribalda [Mon, 11 Jan 2021 14:54:38 +0000 (15:54 +0100)]
media: mtk-mdp: Do not zero reserved fields

Core code already clears reserved fields of struct
v4l2_pix_format_mplane, check commit 4e1e0eb0e074 ("media: v4l2-ioctl: Zero
v4l2_plane_pix_format reserved fields").

Cc: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: rc: select CONFIG_BITREVERSE where needed
Arnd Bergmann [Thu, 3 Dec 2020 23:17:03 +0000 (00:17 +0100)]
media: rc: select CONFIG_BITREVERSE where needed

A number of remote control drivers require the bitreverse
helper, and run into a link error when it is disabled:

arm-linux-gnueabi-ld: drivers/media/rc/img-ir/img-ir-nec.o: in function `img_ir_nec_scancode':
img-ir-nec.c:(.text+0x10c): undefined reference to `byte_rev_table'
arm-linux-gnueabi-ld: drivers/media/rc/img-ir/img-ir-nec.o: in function `img_ir_nec_filter':
img-ir-nec.c:(.text+0x2dc): undefined reference to `byte_rev_table'
arm-linux-gnueabi-ld: drivers/media/usb/cx231xx/cx231xx-input.o: in function `get_key_isdbt':
cx231xx-input.c:(.text+0x38c): undefined reference to `byte_rev_table'
arm-linux-gnueabi-ld: drivers/media/usb/em28xx/em28xx-input.o: in function `em28xx_get_key_em_haup':
em28xx-input.c:(.text+0x1704): undefined reference to `byte_rev_table'

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Sean Young <sean@mess.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: mtk-vcodec: Fix order of log arguments
Evan Benn [Thu, 26 Nov 2020 03:08:42 +0000 (04:08 +0100)]
media: mtk-vcodec: Fix order of log arguments

Fix order of log arguments.

[hverkuil: fix checkpatch parenthesis alignment warnings]

Signed-off-by: Evan Benn <evanbenn@chromium.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agos390: uv: Fix sysfs max number of VCPUs reporting
Janosch Frank [Wed, 13 Jan 2021 16:56:26 +0000 (11:56 -0500)]
s390: uv: Fix sysfs max number of VCPUs reporting

The number reported by the query is N-1 and I think people reading the
sysfs file would expect N instead. For users creating VMs there's no
actual difference because KVM's limit is currently below the UV's
limit.

Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Fixes: a0f60f8431999 ("s390/protvirt: Add sysfs firmware interface for Ultravisor information")
Cc: stable@vger.kernel.org
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
3 years agos390/vfio-ap: No need to disable IRQ after queue reset
Tony Krowiak [Wed, 23 Dec 2020 01:15:53 +0000 (20:15 -0500)]
s390/vfio-ap: No need to disable IRQ after queue reset

The queues assigned to a matrix mediated device are currently reset when:

* The VFIO_DEVICE_RESET ioctl is invoked
* The mdev fd is closed by userspace (QEMU)
* The mdev is removed from sysfs.

Immediately after the reset of a queue, a call is made to disable
interrupts for the queue. This is entirely unnecessary because the reset of
a queue disables interrupts, so this will be removed.

Furthermore, vfio_ap_irq_disable() does an unconditional PQAP/AQIC which
can result in a specification exception (when the corresponding facility
is not available), so this is actually a bugfix.

Signed-off-by: Tony Krowiak <akrowiak@linux.ibm.com>
[pasic@linux.ibm.com: minor rework before merging]
Signed-off-by: Halil Pasic <pasic@linux.ibm.com>
Fixes: ec89b55e3bce ("s390: ap: implement PAPQ AQIC interception in kernel")
Cc: <stable@vger.kernel.org>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
3 years agos390/vfio-ap: clean up vfio_ap resources when KVM pointer invalidated
Tony Krowiak [Wed, 23 Dec 2020 01:20:13 +0000 (20:20 -0500)]
s390/vfio-ap: clean up vfio_ap resources when KVM pointer invalidated

The vfio_ap device driver registers a group notifier with VFIO when the
file descriptor for a VFIO mediated device for a KVM guest is opened to
receive notification that the KVM pointer is set (VFIO_GROUP_NOTIFY_SET_KVM
event). When the KVM pointer is set, the vfio_ap driver takes the
following actions:
1. Stashes the KVM pointer in the vfio_ap_mdev struct that holds the state
   of the mediated device.
2. Calls the kvm_get_kvm() function to increment its reference counter.
3. Sets the function pointer to the function that handles interception of
   the instruction that enables/disables interrupt processing.
4. Sets the masks in the KVM guest's CRYCB to pass AP resources through to
   the guest.

In order to avoid memory leaks, when the notifier is called to receive
notification that the KVM pointer has been set to NULL, the vfio_ap device
driver should reverse the actions taken when the KVM pointer was set.

Fixes: 258287c994de ("s390: vfio-ap: implement mediated device open callback")
Signed-off-by: Tony Krowiak <akrowiak@linux.ibm.com>
Reviewed-by: Halil Pasic <pasic@linux.ibm.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20201223012013.5418-1-akrowiak@linux.ibm.com
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
3 years agomedia: qm1d1c0042: fix error return code in qm1d1c0042_init()
Luo Meng [Wed, 25 Nov 2020 01:34:37 +0000 (02:34 +0100)]
media: qm1d1c0042: fix error return code in qm1d1c0042_init()

Fix to return a negative error code from the error handling case
instead of 0 in function qm1d1c0042_init(), as done elsewhere
in this function.

Fixes: ab4d14528fdf ("[media] em28xx: add support for PLEX PX-BCUD (ISDB-S)")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Luo Meng <luomeng12@huawei.com>
Acked-by: Akihiro Tsukada <tskd08@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: saa7134: Fix fall-through warnings for Clang
Gustavo A. R. Silva [Fri, 20 Nov 2020 18:37:08 +0000 (19:37 +0100)]
media: saa7134: Fix fall-through warnings for Clang

In preparation to enable -Wimplicit-fallthrough for Clang, fix a warning
by explicitly adding a break statement instead of letting the code fall
through to the next case.

Link: https://github.com/KSPP/linux/issues/115
[hverkuil: fix checkpatch TAB indendation issues]

Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: rcar_jpu: Fix fall-through warnings for Clang
Gustavo A. R. Silva [Fri, 20 Nov 2020 18:37:02 +0000 (19:37 +0100)]
media: rcar_jpu: Fix fall-through warnings for Clang

In preparation to enable -Wimplicit-fallthrough for Clang, fix a warning
by explicitly adding a break statement instead of letting the code fall
through to the next case.

Link: https://github.com/KSPP/linux/issues/115
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: dvb_frontend: Fix fall-through warnings for Clang
Gustavo A. R. Silva [Fri, 20 Nov 2020 18:36:57 +0000 (19:36 +0100)]
media: dvb_frontend: Fix fall-through warnings for Clang

In preparation to enable -Wimplicit-fallthrough for Clang, fix a warning
by explicitly adding a break statement instead of letting the code fall
through to the next case.

Link: https://github.com/KSPP/linux/issues/115
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: atomisp: Fix fall-through warnings for Clang
Gustavo A. R. Silva [Fri, 20 Nov 2020 18:36:50 +0000 (19:36 +0100)]
media: atomisp: Fix fall-through warnings for Clang

In preparation to enable -Wimplicit-fallthrough for Clang, fix a warning
by explicitly adding a break statement instead of letting the code fall
through to the next case.

Link: https://github.com/KSPP/linux/issues/115
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: usb: dvb-usb-v2: Fix fall-through warnings for Clang
Gustavo A. R. Silva [Fri, 20 Nov 2020 18:26:16 +0000 (19:26 +0100)]
media: usb: dvb-usb-v2: Fix fall-through warnings for Clang

In preparation to enable -Wimplicit-fallthrough for Clang, fix multiple
warnings by explicitly adding a couple of break statements instead of
just letting the code fall through to the next case.

Link: https://github.com/KSPP/linux/issues/115
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: dvb-frontends: Fix fall-through warnings for Clang
Gustavo A. R. Silva [Fri, 20 Nov 2020 18:26:09 +0000 (19:26 +0100)]
media: dvb-frontends: Fix fall-through warnings for Clang

In preparation to enable -Wimplicit-fallthrough for Clang, fix multiple
warnings by explicitly adding multiple break and a return statements
instead of just letting the code fall through to the next case.

Link: https://github.com/KSPP/linux/issues/115
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: atomisp: Fix a buffer overflow in debug code
Dan Carpenter [Fri, 20 Nov 2020 16:27:18 +0000 (17:27 +0100)]
media: atomisp: Fix a buffer overflow in debug code

The "pad" variable is a user controlled string and we haven't properly
clamped it at this point so the debug code could print from beyond the
of the array.

Fixes: a49d25364dfb ("staging/atomisp: Add support for the Intel IPU v2")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: cedrus: Remove checking for required controls
Jernej Skrabec [Wed, 23 Dec 2020 11:06:58 +0000 (12:06 +0100)]
media: cedrus: Remove checking for required controls

According to v4l2 request api specifications, it's allowed to skip
control if its content isn't changed for performance reasons. Cedrus
driver predates that, so it has implemented mechanism to check if all
required controls are included in one request.

Conform to specifications with removing that mechanism.

Note that this mechanism with static required flag isn't very good
anyway because need for control is usually signaled in other controls.

Fixes: 50e761516f2b ("media: platform: Add Cedrus VPU decoder driver")
Signed-off-by: Jernej Skrabec <jernej.skrabec@siol.net>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: vidtv: print message when driver is removed
Daniel W. S. Almeida [Wed, 6 Jan 2021 03:04:49 +0000 (04:04 +0100)]
media: vidtv: print message when driver is removed

Print a message when the driver is removed so that we get some
visual confirmation when unbinding vidtv.

Signed-off-by: Daniel W. S. Almeida <dwlsalmeida@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: vidtv: use a simpler name in platform_{device|driver}
Daniel W. S. Almeida [Wed, 6 Jan 2021 03:04:48 +0000 (04:04 +0100)]
media: vidtv: use a simpler name in platform_{device|driver}

Change from "vidtv_bridge" to simply "vidtv" so that vidtv looks
more similar to the other media virtual drivers in /sys/bus/platform.

Signed-off-by: Daniel W. S. Almeida <dwlsalmeida@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: vidtv: reinstate sysfs bind attrs
Daniel W. S. Almeida [Wed, 6 Jan 2021 03:04:47 +0000 (04:04 +0100)]
media: vidtv: reinstate sysfs bind attrs

Reinstate sysfs bind attrs so that vidtv can be bound and unbound
via sysfs. This is useful for automated regression testing in
userspace.

Signed-off-by: Daniel W. S. Almeida <dwlsalmeida@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: vidtv: Add media controller support
Daniel W. S. Almeida [Wed, 6 Jan 2021 03:04:46 +0000 (04:04 +0100)]
media: vidtv: Add media controller support

Add media controller support when CONFIG_MEDIA_CONTROLLER_DVB is set
so that, in the future, a test sequence in v4l-utils can be written
without having to know which /dev/fooX device should be used.

[mchehab: avoided usage of C99 comments]
Signed-off-by: Daniel W. S. Almeida <dwlsalmeida@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: vidtv: Use kmemdup instead of kzalloc and memcpy
Tian Tao [Tue, 29 Dec 2020 08:55:30 +0000 (09:55 +0100)]
media: vidtv: Use kmemdup instead of kzalloc and memcpy

Fixes coccicheck warning:
drivers/media/test-drivers/vidtv/vidtv_psi.c:509:10-17: WARNING
opportunity for kmemdup

Signed-off-by: Tian Tao <tiantao6@hisilicon.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: vidtv: remove unused field from 'struct vidtv_mpeg_ts'
Daniel W. S. Almeida [Thu, 24 Dec 2020 15:04:01 +0000 (16:04 +0100)]
media: vidtv: remove unused field from 'struct vidtv_mpeg_ts'

Commit 3d1387b3b8f6 ("media: vidtv: fix some warnings") replaced
the unused flexible array member at the end of struct vidtv_mpeg_ts
with a pointer. This broke the 188-byte alignment since the struct
no longer was 4 bytes in size.

Fix this by removing this field entirely.

Fixes: 3d1387b3b8f6 ("media: vidtv: fix some warnings")
Signed-off-by: Daniel W. S. Almeida <dwlsalmeida@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: vidtv: psi: fix missing crc for PMT
Daniel W. S. Almeida [Thu, 24 Dec 2020 15:04:00 +0000 (16:04 +0100)]
media: vidtv: psi: fix missing crc for PMT

The PMT write function was refactored and this broke the CRC computation.

Fix it.

Fixes: db9569f67e2e ("media: vidtv: cleanup PMT write table function")
Signed-off-by: Daniel W. S. Almeida <dwlsalmeida@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agocan: dev: prevent potential information leak in can_fill_info()
Dan Carpenter [Thu, 21 Jan 2021 06:08:05 +0000 (09:08 +0300)]
can: dev: prevent potential information leak in can_fill_info()

The "bec" struct isn't necessarily always initialized. For example, the
mcp251xfd_get_berr_counter() function doesn't initialize anything if the
interface is down.

Fixes: 52c793f24054 ("can: netlink support for bus-error reporting and counters")
Link: https://lore.kernel.org/r/YAkaRdRJncsJO8Ve@mwanda
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
3 years agortc: mc146818: Detect and handle broken RTCs
Thomas Gleixner [Tue, 26 Jan 2021 17:02:11 +0000 (18:02 +0100)]
rtc: mc146818: Detect and handle broken RTCs

The recent fix for handling the UIP bit unearthed another issue in the RTC
code. If the RTC is advertised but the readout is straight 0xFF because
it's not available, the old code just proceeded with crappy values, but the
new code hangs because it waits for the UIP bit to become low.

Add a sanity check in the RTC CMOS probe function which reads the RTC_VALID
register (Register D) which should have bit 0-6 cleared. If that's not the
case then fail to register the CMOS.

Add the same check to mc146818_get_time(), warn once when the condition
is true and invalidate the rtc_time data.

Reported-by: Mickaël Salaün <mic@digikod.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Mickaël Salaün <mic@linux.microsoft.com>
Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/87tur3fx7w.fsf@nanos.tec.linutronix.de
3 years agoxen: Fix XenStore initialisation for XS_LOCAL
David Woodhouse [Tue, 26 Jan 2021 17:01:49 +0000 (17:01 +0000)]
xen: Fix XenStore initialisation for XS_LOCAL

In commit 3499ba8198ca ("xen: Fix event channel callback via INTX/GSI")
I reworked the triggering of xenbus_probe().

I tried to simplify things by taking out the workqueue based startup
triggered from wake_waiting(); the somewhat poorly named xenbus IRQ
handler.

I missed the fact that in the XS_LOCAL case (Dom0 starting its own
xenstored or xenstore-stubdom, which happens after the kernel is booted
completely), that IRQ-based trigger is still actually needed.

So... put it back, except more cleanly. By just spawning a xenbus_probe
thread which waits on xb_waitq and runs the probe the first time it
gets woken, just as the workqueue-based hack did.

This is actually a nicer approach for *all* the back ends with different
interrupt methods, and we can switch them all over to that without the
complex conditions for when to trigger it. But not in -rc6. This is
the minimal fix for the regression, although it's a step in the right
direction instead of doing a partial revert and actually putting the
workqueue back. It's also simpler than the workqueue.

Fixes: 3499ba8198ca ("xen: Fix event channel callback via INTX/GSI")
Reported-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Juergen Gross <jgross@suse.com>
Link: https://lore.kernel.org/r/4c9af052a6e0f6485d1de43f2c38b1461996db99.camel@infradead.org
Signed-off-by: Juergen Gross <jgross@suse.com>
3 years agomedia: i2c/ov02a10.c: add cast to fix type mismatch
Hans Verkuil [Wed, 20 Jan 2021 09:27:08 +0000 (10:27 +0100)]
media: i2c/ov02a10.c: add cast to fix type mismatch

By adding this cast '(__force __le16)' this sparse warning
is fixed:

drivers/media/i2c/ov02a10.c:391:19: warning: cast to restricted __le16

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: atomisp/pci: add missing include
Hans Verkuil [Wed, 20 Jan 2021 09:11:33 +0000 (10:11 +0100)]
media: atomisp/pci: add missing include

Fix two smatch warnings:

drivers/staging/media/atomisp//pci/ia_css_firmware.h:52:29: warning: 'struct device' declared inside parameter list will not be visible outside of this definition or declaration
drivers/staging/media/atomisp//pci/ia_css_control.h:49:24: warning: 'struct device' declared inside parameter list will not be visible outside of this definition or declaration

Add '#include <linux/device.h>' to ia_css_firmware.h so struct device is defined.

ia_css_control.h includes ia_css_firmware.h, so it is sufficient to
just modify ia_css_firmware.h.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: dvb-frontends/af9033.c: fix missing error codes
Hans Verkuil [Wed, 20 Jan 2021 09:09:05 +0000 (10:09 +0100)]
media: dvb-frontends/af9033.c: fix missing error codes

Fixes two smatch warnings:

drivers/media/dvb-frontends/af9033.c:128 af9033_init() warn: missing error code 'ret'
drivers/media/dvb-frontends/af9033.c:855 af9033_read_snr() warn: missing error code 'ret'

Set error codes correctly.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: dvb-frontends/rtl2832.c: fix missing error code
Hans Verkuil [Wed, 20 Jan 2021 09:03:21 +0000 (10:03 +0100)]
media: dvb-frontends/rtl2832.c: fix missing error code

Fixes this smatch warning:

drivers/media/dvb-frontends/rtl2832.c:702 rtl2832_read_status() warn: missing error code 'ret'

'ret' is indeed not set, so set it to -EINVAL so a proper error
code is returned.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: usb/dvb-usb-v2/rtl28xxu.c: clean up code to fix smatch warning
Hans Verkuil [Wed, 20 Jan 2021 08:58:38 +0000 (09:58 +0100)]
media: usb/dvb-usb-v2/rtl28xxu.c: clean up code to fix smatch warning

Fixes this smatch warning:

drivers/media/usb/dvb-usb-v2/rtl28xxu.c:1040 rtl2832u_frontend_attach() warn: missing error code 'ret'

It is actually a bogus warning since in this particular case ret isn't
meant to be set. But by reworking the code a bit the code is actually
a lot more understandable and it fixes this warning.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: pci/ivtv: release memory regions on error
Hans Verkuil [Wed, 20 Jan 2021 08:52:53 +0000 (09:52 +0100)]
media: pci/ivtv: release memory regions on error

Fix this smatch warning:

drivers/media/pci/ivtv/ivtv-driver.c:900 ivtv_setup_pci() warn: 'itv->base_addr' not released on lines: 876.

One error path didn't release the memory regions.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: sti/hva: add missing clk_disable_unprepare()
Hans Verkuil [Wed, 20 Jan 2021 08:47:20 +0000 (09:47 +0100)]
media: sti/hva: add missing clk_disable_unprepare()

Fixes this smatch warning:

drivers/media/platform/sti/hva/hva-hw.c:453 hva_hw_runtime_resume() warn: 'hva->clk' not released on lines: 450.

Call clk_disable_unprepare() when clk_set_rate() fails.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Cc: Jean-Christophe Trotin <jean-christophe.trotin@st.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
3 years agomedia: sti/c8sectpfe: set correct return code
Hans Verkuil [Wed, 20 Jan 2021 08:45:23 +0000 (09:45 +0100)]
media: sti/c8sectpfe: set correct return code

Fixes this smatch warning:

drivers/media/platform/sti/c8sectpfe/c8sectpfe-core.c:829 c8sectpfe_probe() warn: missing error code 'ret'

Set ret to -EINVAL if the reset gpio was not found.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>