drm/i915/perf: Drop lockdep assert for i915_oa_init_reg_state()

This is called from execlist context init which we need to be unlocked.
Commit f89823c21224 ("drm/i915/perf: Implement
I915_PERF_ADD/REMOVE_CONFIG interface") added a lockdep assert to this
path for unclear reasons, remove it again!

Fixes: f89823c21224 ("drm/i915/perf: Implement I915_PERF_ADD/REMOVE_CONFIG interface")
Signed-off-by: Chris Wilson <[email protected]>
Cc: Lionel Landwerlin <[email protected]>
Cc: Matthew Auld <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Matthew Auld <[email protected]>

drm/i915/perf: Initialise dynamic sysfs group before creation

Another case where we need to call sysfs_attr_init() to setup the
internal lockdep class prior to use:

[    9.325229] BUG: key ffff880168bc7bb0 not in .data!
[    9.325240] DEBUG_LOCKS_WARN_ON(1)
[    9.325250] ------------[ cut here ]------------
[    9.325280] WARNING: CPU: 1 PID: 275 at kernel/locking/lockdep.c:3156 lockdep_init_map+0x1b2/0x1c0
[    9.325301] Modules linked in: intel_powerclamp(+) coretemp crct10dif_pclmul crc32_pclmul ghash_clmulni_intel i915(+) snd_hda_intel snd_hda_codec snd_hwdep r8169 mii snd_hda_core snd_pcm prime_numbers i2c_hid pinctrl_geminilake pinctrl_intel
[    9.325375] CPU: 1 PID: 275 Comm: modprobe Not tainted 4.13.0-rc4-CI-Trybot_1040+ #1
[    9.325395] Hardware name: Intel Corp. Geminilake/GLK RVP2 LP4SD (07), BIOS GELKRVPA.X64.0045.B51.1704281422 04/28/2017
[    9.325422] task: ffff8801721a4ec0 task.stack: ffffc900001dc000
[    9.325440] RIP: 0010:lockdep_init_map+0x1b2/0x1c0
[    9.325456] RSP: 0018:ffffc900001dfa10 EFLAGS: 00010282
[    9.325473] RAX: 0000000000000016 RBX: ffff880168d54b80 RCX: 0000000000000000
[    9.325488] RDX: 0000000080000001 RSI: 0000000000000001 RDI: ffffffff810f0800
[    9.325505] RBP: ffffc900001dfa30 R08: 0000000000000001 R09: 0000000000000000
[    9.325521] R10: 0000000000000000 R11: 0000000000000000 R12: ffff880168bc7bb0
[    9.325537] R13: 0000000000000000 R14: ffff880168bc7b98 R15: ffffffff81a263a0
[    9.325554] FS:  00007fb60c3fd700(0000) GS:ffff88017fc80000(0000) knlGS:0000000000000000
[    9.325574] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    9.325588] CR2: 0000006582777d80 CR3: 000000016d818000 CR4: 00000000003406e0
[    9.325604] Call Trace:
[    9.325618]  __kernfs_create_file+0x76/0xe0
[    9.325632]  sysfs_add_file_mode_ns+0x8a/0x1a0
[    9.325646]  internal_create_group+0xea/0x2c0
[    9.325660]  sysfs_create_group+0x13/0x20
[    9.325737]  i915_perf_register+0xde/0x220 [i915]
[    9.325800]  i915_driver_load+0xa77/0x16c0 [i915]
[    9.325863]  i915_pci_probe+0x37/0x90 [i915]
[    9.325880]  pci_device_probe+0xa8/0x130
[    9.325894]  driver_probe_device+0x29c/0x450
[    9.325908]  __driver_attach+0xe3/0xf0
[    9.325922]  ? driver_probe_device+0x450/0x450
[    9.325935]  bus_for_each_dev+0x62/0xa0
[    9.325948]  driver_attach+0x1e/0x20
[    9.325960]  bus_add_driver+0x173/0x270
[    9.325974]  driver_register+0x60/0xe0
[    9.325986]  __pci_register_driver+0x60/0x70
[    9.326044]  i915_init+0x6f/0x78 [i915]
[    9.326066]  ? 0xffffffffa024e000
[    9.326079]  do_one_initcall+0x43/0x170
[    9.326094]  ? rcu_read_lock_sched_held+0x7a/0x90
[    9.326109]  ? kmem_cache_alloc_trace+0x261/0x2d0
[    9.326124]  do_init_module+0x5f/0x206
[    9.326137]  load_module+0x2561/0x2da0
[    9.326150]  ? show_coresize+0x30/0x30
[    9.326165]  ? kernel_read_file+0x105/0x190
[    9.326180]  SyS_finit_module+0xc1/0x100
[    9.326192]  ? SyS_finit_module+0xc1/0x100
[    9.326210]  entry_SYSCALL_64_fastpath+0x1c/0xb1
[    9.326223] RIP: 0033:0x7fb60bf359f9
[    9.326234] RSP: 002b:00007fff92b47c48 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
[    9.326255] RAX: ffffffffffffffda RBX: ffffffff814898a3 RCX: 00007fb60bf359f9
[    9.326271] RDX: 0000000000000000 RSI: 00000028a9ceef8b RDI: 0000000000000000
[    9.326287] RBP: ffffc900001dff88 R08: 0000000000000000 R09: 0000000000000000
[    9.326303] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000040000
[    9.326319] R13: 00000028aaef2a70 R14: 0000000000000000 R15: 00000028aaeee5d0
[    9.326339]  ? __this_cpu_preempt_check+0x13/0x20
[    9.326353] Code: f1 39 00 85 c0 0f 84 38 ff ff ff 83 3d 9f 44 ce 01 00 0f 85 2b ff ff ff 48 c7 c6 b2 a2 c7 81 48 c7 c7 53 40 c5 81 e8 3f 82 01 00 <0f> ff e9 11 ff ff ff 0f 1f 80 00 00 00 00 55 31 c9 31 d2 31 f6

Fixes: 701f8231a2fe ("drm/i915/perf: prune OA configs")
Signed-off-by: Chris Wilson <[email protected]>
Cc: Lionel Landwerlin <[email protected]>
Cc: Matthew Auld <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Lionel Landwerlin <[email protected]>

drm/i915: add register macro definition style guide

This is not to try to force a new style; this is my interpretation of
what the most common existing style is.

With hopes I don't need to answer so many questions about style going
forward.

Start a new style section in the i915 document to bolt the register
style guide into.

v2: vertical alignment, incorporate to kernel-doc, and more

Cc: Daniel Vetter <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Reviewed-by: Daniel Vetter <[email protected]>
Reviewed-by: Dhinakaran Pandiyan <[email protected]>
Signed-off-by: Jani Nikula <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/9de4a5b1bea4e76461c70a1dd66751581de0124f.1502368010.git.jani.nikula@intel.com

drm/i915: enum i915_power_well_id is not proper kernel-doc

Revert to a normal comment, as the enum isn't properly documented
anyway.

Signed-off-by: Jani Nikula <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/f2c44a0aa00ea7d9b71e7a3183a7507f98811146.1502368010.git.jani.nikula@intel.com

Documentation/i915: remove sphinx conversion artefact

Remove old warning about docproc directive that's not supported in the
Sphinx toolchain.

Signed-off-by: Jani Nikula <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/7fc8a110b78a9dc9a585dce643b68b4200b7e793.1502368010.git.jani.nikula@intel.com

drm/i915: Add format modifiers for Intel

This was based on a patch originally by Kristian. It has been modified
pretty heavily to use the new callbacks from the previous patch.

v2:
  - Add LINEAR and Yf modifiers to list (Ville)
  - Combine i8xx and i965 into one list of formats (Ville)
  - Allow 1010102 formats for Y/Yf tiled (Ville)

v3:
  - Handle cursor formats (Ville)
  - Put handling for LINEAR in the mod_support functions (Ville)

v4:
  - List each modifier explicitly in supported modifiers (Ville)
  - Handle the CURSOR plane (Ville)

v5:
  - Split out cursor and sprite handling (Ville)

v6:
  - Actually use the sprite funcs (Emil)
  - Use unreachable (Emil)

v7:
  - Only allow Intel modifiers and LINEAR (Ben)

v8
  - Fix spite assert introduced in v6 (Daniel)

v9
  - Change vendor check logic to avoid magic 56 (Emil)
  - Reorder skl_mod_support (Ville)
  - make intel_plane_funcs static, could be done as of v5 (Ville)
  - rename local variable intel_format_modifiers to modifiers (Ville)
    - actually use sprite modifiers
  - split out modifier/formats by platform (Ville)

v10:
  - Undo vendor check from v9

v11:
  - Squash CCS advertisement into this patch (daniels)
  - Don't advertise CCS on higher sprite planes (daniels)

v12:
  - Don't advertise Y-tiled or CCS on any sprite planes, since we don't
    allocate enough DDB space for it to work. (daniels)

Cc: Ville Syrjälä <[email protected]>
Reviewed-by: Emil Velikov <[email protected]> (v8)
Signed-off-by: Ben Widawsky <[email protected]>
Signed-off-by: Daniel Stone <[email protected]>

drm/i915: Add render decompression support

SKL+ display engine can scan out certain kinds of compressed surfaces
produced by the render engine. This involved telling the display engine
the location of the color control surfae (CCS) which describes
which parts of the main surface are compressed and which are not. The
location of CCS is provided by userspace as just another plane with its
own offset.

Add the required stuff to validate the user provided AUX plane metadata
and convert the user provided linear offset into something the hardware
can consume.

Due to hardware limitations we require that the main surface and
the AUX surface (CCS) be part of the same bo. The hardware also
makes life hard by not allowing you to provide separate x/y offsets
for the main and AUX surfaces (excpet with NV12), so finding suitable
offsets for both requires a bit of work. Assuming we still want keep
playing tricks with the offsets. I've just gone with a dumb "search
backward for suitable offsets" approach, which is far from optimal,
but it works.

Also not all planes will be capable of scanning out compressed surfaces,
and eg. 90/270 degree rotation is not supported in combination with
decompression either.

This patch may contain work from at least the following people:
* Vandana Kannan <[email protected]>
* Daniel Vetter <[email protected]>
* Ben Widawsky <[email protected]>

v2: Deal with display workarounds 0390, 0531, 1125 (Paulo)
v3: Pretend CCS tiles are regular 128 byte wide Y tiles (Jason)
    Put the AUX register defines to the correct place
    Fix up the slightly bogus rotation check
v4: Use I915_WRITE_FW() due to plane update locking changes
    s/return -EINVAL/goto err/ in intel_framebuffer_init()
    Eliminate a bunch hardcoded numbers in CCS code

v5: (By Ben)
conflict resolution +
-               res_blocks += fixed_16_16_to_u32_round_up(y_tile_minimum);
+               res_blocks += fixed16_to_u32_round_up(y_tile_minimum);

v6: (daniels) Fix botched commit message.

Cc: Paulo Zanoni <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Ben Widawsky <[email protected]>
Cc: Jason Ekstrand <[email protected]>
Signed-off-by: Ville Syrjä <[email protected]>
Reviewed-by: Ben Widawsky <[email protected]> (v1)
Reviewed-by: Daniel Stone <[email protected]>
Signed-off-by: Ben Widawsky <[email protected]>
Signed-off-by: Daniel Stone <[email protected]>
Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915: Implement .get_format_info() hook for CCS

SKL+ display engine can scan out certain kinds of compressed surfaces
produced by the render engine. This involved telling the display engine
the location of the color control surfae (CCS) which describes which
parts of the main surface are compressed and which are not. The location
of CCS is provided by userspace as just another plane with its own offset.

By providing our own format information for the CCS formats, we should
be able to make framebuffer_check() do the right thing for the CCS
surface as well.

Note that we'll return the same format info for both Y and Yf tiled
format as that's what happens with the non-CCS Y vs. Yf as well. If
desired, we could potentially return a unique pointer for each
pixel_format+tiling+ccs combination, in which case we immediately be
able to tell if any of that stuff changed by just comparing the
pointers. But that does sound a bit wasteful space wise.

v2: Drop the 'dev' argument from the hook
v3: Include the description of the CCS surface layout
v4: Pretend CCS tiles are regular 128 byte wide Y tiles (Jason)
v5: Re-drop 'dev', fix commit message, add missing drm_fourcc.h
    description of CCS layout. (daniels)

Cc: Daniel Vetter <[email protected]>
Cc: Ben Widawsky <[email protected]>
Cc: Jason Ekstrand <[email protected]>
Acked-by: Jason Ekstrand <[email protected]>
Reviewed-by: Ben Widawsky <[email protected]> (v3)
Reviewed-by: Daniel Stone <[email protected]>
Signed-off-by: Ville Syrjä <[email protected]>
Signed-off-by: Ben Widawsky <[email protected]>
Signed-off-by: Daniel Stone <[email protected]>

Merge airlied/drm-next into drm-intel-next-queued

Ben Widawsky/Daniel Stone need the extended modifier support from
drm-misc to be able to merge CCS support for i915.ko

Signed-off-by: Daniel Vetter <[email protected]>

drm/i915: Supply the engine-id for our mock_engine()

In the original selftest, we didn't care what the engine->id was, just
that it could uniquely identify it. Later though, we started tracking
the mock engines in the fixed size arrays around the drm_i915_private and
so we now require their indices to be correct. This becomes an issue when
using the standalone harness which runs all available tests at module load,
and so we quickly assign an out-of-bounds index to an engine as we
reallocate the mock GEM device between tests. It doesn't show up in
igt/drv_selftest as that runs each subtest individually.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102045
Signed-off-by: Chris Wilson <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Mika Kuoppala <[email protected]>

Merge tag 'drm-misc-next-2017-08-08' of git://anongit.freedesktop.org/git/drm-misc into drm-next

UAPI Changes:
- vc4: Add ioctl to allow attaching a label to a bo (Eric)
- Add new format/modifier blob plane property (Ben)
- armada: Use __u32/__u64 instead of uint32_t/uint64_t (Mikko)
- [kinda uapi] fb_helper: Expose display_info size via fb_info (David)

Core Changes:
- Default gem_dumb_[map_offset|destroy] as mmap/destroy implementations (Noralf)
- Simplify atomic properties by removing the helpers and handling in core (Daniel)

Driver Changes:
- stm: Add STM32 DSI controller driver (Phillipe)
- vc4: Add HDMI CEC support (Hans)
- rockchip: Refactor register init & soc version handling (Mark)
- misc: Remove .load_lut, .gamma_set, .gamma_get dead code (Peter)
- dw-hdmi: Add HDMI CEC support (Russell)

Cc: Philippe CORNU <[email protected]>
Cc: Hans Verkuil <[email protected]>
Cc: Eric Anholt <[email protected]>
Cc: Noralf Trønnes <[email protected]>
Cc: Ben Widawsky <[email protected]>
Cc: Mark yao <[email protected]>
Cc: Peter Rosin <[email protected]>
Cc: Russell King <[email protected]>
Cc: Mikko Rapeli <[email protected]>
Cc: David Lechner <[email protected]>
Cc: Daniel Vetter <[email protected]>
* tag 'drm-misc-next-2017-08-08' of git://anongit.freedesktop.org/git/drm-misc: (107 commits)
  drm: Nuke drm_atomic_legacy_backoff
  drm: Nuke drm_atomic_helper_connector_dpms
  drm: Nuke drm_atomic_helper_connector_set_property
  drm: Nuke drm_atomic_helper_plane_set_property
  drm: Nuke drm_atomic_helper_crtc_set_property
  drm: Handle properties in the core for atomic drivers
  drm: Don't update property values for atomic drivers
  drm/omap: Rework the rotation-on-crtc hack
  drm/radeon: Use the drm_driver.dumb_destroy default
  drm/i915: Use the drm_driver.dumb_destroy default
  drm/sti: Use .dumb_map_offset and .dumb_destroy defaults
  drm: bridge: synopsys/dw-hdmi: Provide default configuration function for HDMI 2.0 PHY
  drm/fb-helper: pass physical dimensions to fbdev
  uapi drm/armada_drm.h: use __u32 and __u64 instead of uint32_t and uint64_t
  drm/bridge: dw-hdmi: remove CEC engine register definitions
  drm/bridge: dw-hdmi: add cec driver
  drm/bridge: dw-hdmi: add missing cec_notifier_put
  drm: remove unused and redundant callbacks
  staging: vboxvideo: remove dead gamma lut code
  drm: dw-hdmi-i2s: add missing company name on Copyright
  ...

drm/i915/psr: Preserve SRD_CTL bit 29 on PSR init

Bit 29 of SRD_CTL needs to have its value preserved according to the
B-Spec, so right before we write out the register we go ahead and read
the register and preserve the value of that bit before we write out
the configured register value.

v2: Spaces => tabs, minor name change, and commit message wording (Rodrigo)

Cc: Rodrigo Vivi <[email protected]>
Cc: Chris Wilson <[email protected]>
Cc: Jani Nikula <[email protected]>
Signed-off-by: Jim Bride <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Signed-off-by: Rodrigo Vivi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915/cnl: Removing missing DDI_E bits from CNL.

DDI_E is not supported on CNL-U and CNL-Y

When adding the initial support we noticed DDI_E wasn't supported
and removed it on v4 and v5 of that patch.
However for some reason I missed or put back these 2 chunks.

Time to clean it up to avoid later confusion.

Fixes: 8bcd3dd41766 ("drm/i915/cnl: Add power wells for CNL")
Cc: Clint Taylor <[email protected]>
Cc: Ville Syrjälä <[email protected]>
Cc: Imre Deak <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Jani Nikula <[email protected]>
Signed-off-by: Rodrigo Vivi <[email protected]>
Reviewed-by: Imre Deak <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915: Perform an invalidate prior to executing golden renderstate

As we may have just bound the renderstate into the GGTT for execution, we
need to ensure that the GTT TLB are also flushed.

On snb-gt2, this would cause a random GPU hang at the start of a new
context (e.g. boot) and on snb-gt1, it was causing the renderstate batch
to take ~10s. It was the GPU hang that revealed the truth, as the CS
gleefully executed beyond the end of the golden renderstate batch, a good
indicator for a GTT TLB miss.

Fixes: 20fe17aa52dc ("drm/i915: Remove redundant TLB invalidate on switching contexts")
Signed-off-by: Chris Wilson <[email protected]>
Cc: Mika Kuoppala <[email protected]>
Cc: [email protected]
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Mika Kuoppala <[email protected]>
Cc: <[email protected]> # v4.12-rc1+

drm: Nuke drm_atomic_legacy_backoff

Finally all users are gone!

Reviewed-by: Maarten Lankhorst <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: Nuke drm_atomic_helper_connector_dpms

It's dead code, the core handles all this directly now.

The only special case is nouveau and tda988x which used one function
for both legacy modeset code and -nv50 atomic world instead of 2
vtables. But amounts to exactly the same.

v2: Rebase over the panel/brideg refactorings in stm/ltdc.

Signed-off-by: Daniel Vetter <[email protected]>
Cc: Archit Taneja <[email protected]>
Cc: Andrzej Hajda <[email protected]>
Cc: Laurent Pinchart <[email protected]>
Cc: Peter Senna Tschudin <[email protected]>
Cc: Martin Donnelly <[email protected]>
Cc: Martyn Welch <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Sean Paul <[email protected]>
Cc: David Airlie <[email protected]>
Cc: Inki Dae <[email protected]>
Cc: Joonyoung Shim <[email protected]>
Cc: Seung-Woo Kim <[email protected]>
Cc: Kyungmin Park <[email protected]>
Cc: Kukjin Kim <[email protected]>
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Stefan Agner <[email protected]>
Cc: Alison Wang <[email protected]>
Cc: Russell King <[email protected]>
Cc: Philipp Zabel <[email protected]>
Cc: CK Hu <[email protected]>
Cc: Matthias Brugger <[email protected]>
Cc: Neil Armstrong <[email protected]>
Cc: Carlo Caione <[email protected]>
Cc: Kevin Hilman <[email protected]>
Cc: Marek Vasut <[email protected]>
Cc: Ben Skeggs <[email protected]>
Cc: Tomi Valkeinen <[email protected]>
Cc: Eric Anholt <[email protected]>
Cc: Mark Yao <[email protected]>
Cc: Heiko Stuebner <[email protected]>
Cc: Benjamin Gaignard <[email protected]>
Cc: Vincent Abriou <[email protected]>
Cc: Yannick Fertre <[email protected]>
Cc: Philippe Cornu <[email protected]>
Cc: Maxime Ripard <[email protected]>
Cc: Chen-Yu Tsai <[email protected]>
Cc: Thierry Reding <[email protected]>
Cc: Jonathan Hunter <[email protected]>
Cc: Jyri Sarha <[email protected]>
Cc: Gerd Hoffmann <[email protected]>
Cc: Shawn Guo <[email protected]>
Cc: John Stultz <[email protected]>
Cc: Lars-Peter Clausen <[email protected]>
Cc: Sergei Shtylyov <[email protected]>
Cc: Jeffy Chen <[email protected]>
Cc: Tomeu Vizoso <[email protected]>
Cc: Yakir Yang <[email protected]>
Cc: Marek Szyprowski <[email protected]>
Cc: Jose Abreu <[email protected]>
Cc: Romain Perier <[email protected]>
Cc: Kieran Bingham <[email protected]>
Cc: Xinliang Liu <[email protected]>
Cc: Alexey Brodkin <[email protected]>
Cc: Alex Deucher <[email protected]>
Cc: Rongrong Zou <[email protected]>
Cc: Rob Clark <[email protected]>
Cc: Hai Li <[email protected]>
Cc: "Noralf Trønnes" <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: zain wang <[email protected]>
Cc: Baoyou Xie <[email protected]>
Cc: Boris Brezillon <[email protected]>
Reviewed-by: Maarten Lankhorst <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Acked-by: Neil Armstrong <[email protected]>
Reviewed-by: Neil Armstrong <[email protected]>
Acked-by: Philipp Zabel <[email protected]>
Acked-by: Archit Taneja <[email protected]>
Tested-by: Philippe Cornu <[email protected]> (on stm)
Reviewed-by: Laurent Pinchart <[email protected]>
Acked-by: Shawn Guo <[email protected]>
Acked-by: Shawn Guo <[email protected]>
Acked-by: Noralf Trønnes <[email protected]>
Acked-by: Vincent Abriou <[email protected]>

drm: Nuke drm_atomic_helper_connector_set_property

It's dead code, the core handles all this directly now. This also
allows us to unexport drm_atomic_helper_connector_set_property.

The only special case is nouveau which used one function for both
pre-nv50 legacy modeset code and post-nv50 atomic world instead of 2
vtables. But amounts to exactly the same.

What is rather strange here is how few drivers set this up, I suspect
the earlier patch to handle properties in the core did end up fixing a
pile of possible issues.

Signed-off-by: Daniel Vetter <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Sean Paul <[email protected]>
Cc: David Airlie <[email protected]>
Cc: Ben Skeggs <[email protected]>
Cc: Benjamin Gaignard <[email protected]>
Cc: Vincent Abriou <[email protected]>
Cc: Eric Anholt <[email protected]>
Cc: [email protected]
Cc: [email protected]
Reviewed-by: Maarten Lankhorst <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Acked-by: Vincent Abriou <[email protected]>

drm: Nuke drm_atomic_helper_plane_set_property

It's dead code, the core handles all this directly now. This also
allows us to unexport drm_atomic_plane_set_property.

Signed-off-by: Daniel Vetter <[email protected]>
Cc: Liviu Dudau <[email protected]>
Cc: Brian Starkey <[email protected]>
Cc: Mali DP Maintainers <[email protected]>
Cc: Boris Brezillon <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Sean Paul <[email protected]>
Cc: David Airlie <[email protected]>
Cc: Inki Dae <[email protected]>
Cc: Joonyoung Shim <[email protected]>
Cc: Seung-Woo Kim <[email protected]>
Cc: Kyungmin Park <[email protected]>
Cc: Kukjin Kim <[email protected]>
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Ben Skeggs <[email protected]>
Cc: Tomi Valkeinen <[email protected]>
Cc: Laurent Pinchart <[email protected]>
Cc: Benjamin Gaignard <[email protected]>
Cc: Vincent Abriou <[email protected]>
Cc: Yannick Fertre <[email protected]>
Cc: Philippe Cornu <[email protected]>
Cc: Jyri Sarha <[email protected]>
Cc: "Ville Syrjälä" <[email protected]>
Cc: Rongrong Zou <[email protected]>
Cc: Shawn Guo <[email protected]>
Cc: Alexey Brodkin <[email protected]>
Cc: Eric Engestrom <[email protected]>
Cc: Chris Wilson <[email protected]>
Cc: Rob Clark <[email protected]>
Cc: Archit Taneja <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: Thomas Hellstrom <[email protected]>
Cc: Maxime Ripard <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Archit Taneja <[email protected]>
Acked-by: Philippe Cornu <[email protected]>
Tested-by: Philippe Cornu <[email protected]>
Acked-by: Liviu Dudau <[email protected]>
Acked-by: Vincent Abriou <[email protected]>
Reviewed-by: Maarten Lankhorst <[email protected]>
Reviewed-by: Laurent Pinchart <[email protected]>

drm: Nuke drm_atomic_helper_crtc_set_property

It's dead code because this is now handled in the core.

Signed-off-by: Daniel Vetter <[email protected]>
Cc: Boris Brezillon <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Jani Nikula <[email protected]>
Cc: Sean Paul <[email protected]>
Cc: David Airlie <[email protected]>
Cc: Ben Skeggs <[email protected]>
Cc: Tomi Valkeinen <[email protected]>
Cc: Laurent Pinchart <[email protected]>
Cc: Alexey Brodkin <[email protected]>
Cc: Shawn Guo <[email protected]>
Cc: Eric Engestrom <[email protected]>
Cc: Chris Wilson <[email protected]>
Cc: "Ville Syrjälä" <[email protected]>
Cc: Rob Clark <[email protected]>
Cc: Philippe Cornu <[email protected]>
Cc: Masahiro Yamada <[email protected]>
Cc: Sushmita Susheelendra <[email protected]>
Cc: Archit Taneja <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: Philipp Zabel <[email protected]>
Cc: Maxime Ripard <[email protected]>
Cc: Thomas Hellstrom <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Archit Taneja <[email protected]>
Acked-by: Philippe Cornu <[email protected]>
Tested-by: Philippe Cornu <[email protected]>
Reviewed-by: Maarten Lankhorst <[email protected]>
Acked-by: Thomas Hellstrom <[email protected]>

drm: Handle properties in the core for atomic drivers

The reason behind the original indirection through the helper
functions was to allow existing drivers to overwrite how they handle
properties. For example when a vendor-specific userspace had
expectations that didn't match atomic. That seemed likely, since
atomic is standardizing a _lot_ more of the behaviour of a kms driver.

But 20 drivers later there's no such need at all. Worse, this forces
all drivers to hook up the default behaviour, breaking userspace if
they forget to do that. And it forces us to export a bunch of core
function just for those helpers.

And finally, these helpers are the last places using
drm_atomic_legacy_backoff() and the implicit acquire_ctx.

This patch here just implements the new behaviour and updates the
docs. Follow-up patches will garbage-collect all the dead code.

v2: Fixup docs even better!

v3: Make it actually work ...

v4: Drop the uses_atomic_modeset() checks from the previous patch
again, since they're now moved up in the callchain.

Cc: Maarten Lankhorst <[email protected]>
Reviewed-by: Archit Taneja <[email protected]> (v3)
Reviewed-by: Maarten Lankhorst <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: Don't update property values for atomic drivers

Atomic drivers only use the property value store for immutable (i.e.
can't be set by userspace, but the kernel can still adjust it)
properties. The only tricky part is the removal of the update in
drm_atomic_helper_update_legacy_modeset_state().

This was added in

commit 8c10342cb48f3140d9abeadcfd2fa6625d447282 (tag: topic/drm-misc-2015-07-28)
Author: Maarten Lankhorst <[email protected]>
Date:   Mon Jul 27 13:24:29 2015 +0200

    drm/atomic: Update legacy DPMS state during modesets, v3.

by copying it from the i915 code, where it was originally added in

commit 68d3472047a572936551f8ff0b6f4016c5a1fdef
Author: Daniel Vetter <[email protected]>
Date:   Thu Sep 6 22:08:35 2012 +0200

    drm/i915: update dpms property in set_mode

for the legacy modeset code. The reason we needed this hack was that
i915 didn't yet set DRIVER_ATOMIC, and we checked for that instead of
the newer-ish drm_drv_uses_atomic_modeset(), which avoids such
troubles. With the correct feature checks this isn't needed anymore at
all.

Also make sure that drivers don't accidentally get this wrong by
making the exported version of drm_object_property_get_value() only
work for legacy drivers. Only gma500 uses it anyway.

v2: Fixup the uses_atomic_modeset() checks (Maarten)

Cc: Maarten Lankhorst <[email protected]>
Reviewed-by: Maarten Lankhorst <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/omap: Rework the rotation-on-crtc hack

I want/need to rework the core property handling, and this hack is
getting in the way. But since it's a non-standard propety only used by
legacy userspace we know that this will only every be called from
ioctl code. And never on some other free-standing state struct, where
this old hack wouldn't work either.

v2: don't forget zorder and get_property!

v3: Shadow the legacy state to avoid locking issues in get_property
(Maarten).

v4: Review from Laurent
- Move struct omap_crtc_state into omap_crtc.c
- Clean up comments.
- Don't forget to copy the shadowed state over on duplicate.

v5: Don't forget to update the reset handler (Maarten).
v6: Update omap_crtc_state shadow values in omap_crtc_atomic_check (Maarten).
v7:
- Fix get_property to return 0 and set value in *val (Maarten).
- Update comment in set_property for changes in v6 (Maarten).

Reviewed-by: Laurent Pinchart <[email protected]> (v4)
Cc: Maarten Lankhorst <[email protected]>
Cc: Tomi Valkeinen <[email protected]
Cc: Laurent Pinchart <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]> (v4)
Signed-off-by: Maarten Lankhorst <[email protected]>
Reviewed-by: Tomi Valkeinen <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/radeon: Use the drm_driver.dumb_destroy default

drm_gem_dumb_destroy() is the drm_driver.dumb_destroy default,
so no need to set it.

Cc: Alex Deucher <[email protected]>
Cc: Christian König <[email protected]>
Signed-off-by: Noralf Trønnes <[email protected]>
Acked-by: Alex Deucher <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915: Use the drm_driver.dumb_destroy default

drm_gem_dumb_destroy() is the drm_driver.dumb_destroy default,
so no need to set it.

Cc: Daniel Vetter <[email protected]>
Cc: Jani Nikula <[email protected]>
Signed-off-by: Noralf Trønnes <[email protected]>
Reviewed-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/sti: Use .dumb_map_offset and .dumb_destroy defaults

This driver can use the drm_driver.dumb_destroy and
drm_driver.dumb_map_offset defaults, so no need to set them.

Cc: Benjamin Gaignard <[email protected]>
Cc: Vincent Abriou <[email protected]>
Signed-off-by: Noralf Trønnes <[email protected]>
Acked-by: Vincent Abriou <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: bridge: synopsys/dw-hdmi: Provide default configuration function for HDMI 2.0 PHY

Currently HDMI 2.0 PHYs do not have a default configuration function.

As *some* of the HDMI 2.0 PHYs have the same register layout as the 3D
PHYs we can provide the same default configuration function for both
and still let user overwrite this with custom configuration function
if needed.

If, for some reason, the PHY is custom or has a register different
register layout then custom configuration function *must* be provided
in order for the system to work correctly. As we prefer the pdata
provided configuration function over the internal one this change
will not make any impact in custom platforms.

Signed-off-by: Jose Abreu <[email protected]>
Tested-by: Mark Yao <[email protected]>
Cc: Kieran Bingham <[email protected]>
Cc: Laurent Pinchart <[email protected]>
Cc: Archit Taneja <[email protected]>
Cc: Andrzej Hajda <[email protected]>
Cc: Mark Yao <[email protected]>
Cc: Carlos Palminha <[email protected]>
Cc: Heiko Stübner <[email protected]>
Reviewed-by: Laurent Pinchart <[email protected]>
Signed-off-by: Archit Taneja <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/185ccf7d4473fa557044732402ca20b3d4007952.1498209896.git.joabreu@synopsys.com

drm/i915: remove unused function declaration

This function is not part of the driver anymore.

Signed-off-by: Lionel Landwerlin <[email protected]>
Fixes: 90f4fcd56bda ("drm/i915: Remove forced stop ring on suspend/unload")
Reviewed-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/fb-helper: pass physical dimensions to fbdev

The fbdev subsystem has a place for physical dimensions (width and height
in mm) that is readable by userspace. Since DRM also knows these
dimensions, pass this information to the fbdev device.

This has to be done in drm_setup_crtcs_fb() instead of drm_setup_crtcs()
because fb_helper->fbdev may be NULL in drm_setup_crtcs().

Signed-off-by: David Lechner <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

uapi drm/armada_drm.h: use __u32 and __u64 instead of uint32_t and uint64_t

These are defined in linux/types.h or drm/drm.h. Fixes
user space compilation errors like:

drm/armada_drm.h:26:2: error: unknown type name ‘uint32_t’
  uint32_t handle;
  ^~~~~~~~

Signed-off-by: Mikko Rapeli <[email protected]>
Cc: Emil Velikov <[email protected]>
Cc: Gabriel Laskar <[email protected]>
Cc: Russell King <[email protected]>
Cc: Rob Clark <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/bridge: dw-hdmi: remove CEC engine register definitions

We don't need the CEC engine register definitions, so let's remove them.

Signed-off-by: Russell King <[email protected]>
Acked-by: Hans Verkuil <[email protected]>
Tested-by: Hans Verkuil <[email protected]>
Tested-by: Laurent Pinchart <[email protected]>
Signed-off-by: Archit Taneja <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/bridge: dw-hdmi: add cec driver

Add a CEC driver for the dw-hdmi hardware.

Reviewed-by: Neil Armstrong <[email protected]>
Signed-off-by: Russell King <[email protected]>
[hans.verkuil: unsigned -> unsigned int]
[hans.verkuil: cec_transmit_done -> cec_transmit_attempt_done]
[hans.verkuil: add missing CEC_CAP_PASSTHROUGH]
Acked-by: Hans Verkuil <[email protected]>
Tested-by: Hans Verkuil <[email protected]>
Tested-by: Laurent Pinchart <[email protected]>
Signed-off-by: Archit Taneja <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/bridge: dw-hdmi: add missing cec_notifier_put

The __dw_hdmi_remove() function was missing a call to cec_notifier_put
to balance the cec_notifier_get in the probe function.

Signed-off-by: Hans Verkuil <[email protected]>
Signed-off-by: Archit Taneja <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: remove unused and redundant callbacks

Drivers no longer have any need for these callbacks, and there are no
users. Zap. Zap-zap-zzzap-p-pp-p.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

staging: vboxvideo: remove dead gamma lut code

The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code that was not doing anything
sensible anyway.

Signed-off-by: Peter Rosin <[email protected]>
Acked-by: Hans de Goede <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: dw-hdmi-i2s: add missing company name on Copyright

This driver's Copyright is under Renesas Solutions Corp.
This patch updates the year, because this driver was moved
into synopsys folder in 2017.

Signed-off-by: Kuninori Morimoto <[email protected]>
Signed-off-by: Archit Taneja <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/87tw1khvnq.wl%[email protected]

drm/i915/selftests: Retarget igt_render_engine_reset_fallback()

The purpose of the test was to check per-engine resets would fallback to
the global reset when required, but first we actually need a test for a
basic i915_handle_error()!

Cc: Mika Kuoppala <[email protected]>
Cc: Michel Thierry <[email protected]>
Signed-off-by: Chris Wilson <[email protected]>
Reviewed-by: Michel Thierry <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/bridge: dw-hdmi: add better clock disable control

The video setup path aways sets the clock disable register to a specific
value, which has the effect of disabling the CEC engine.  When we add the
CEC driver, this becomes a problem.

Fix this by only setting/clearing the bits that the video path needs to.

Reviewed-by: Jose Abreu <[email protected]>
Signed-off-by: Russell King <[email protected]>
Tested-by: Laurent Pinchart <[email protected]>
Acked-by: Hans Verkuil <[email protected]>
Signed-off-by: Archit Taneja <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/bridge: dw-hdmi: add cec notifier support

Add CEC notifier support to the HDMI bridge driver, so that the CEC
part of the IP can receive its physical address.

Signed-off-by: Russell King <[email protected]>
Tested-by: Neil Armstrong <[email protected]>
Reviewed-by: Neil Armstrong <[email protected]>
Tested-by: Laurent Pinchart <[email protected]>
Acked-by: Hans Verkuil <[email protected]>
Signed-off-by: Archit Taneja <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/tinydrm: remove call to mipi_dbi_init() from mipi_dbi_spi_init()

This removes the call to mipi_dbi_init() from mipi_dbi_spi_init() so that
drivers can have a driver-specific implementation if needed.

Suggested-by: Noralf Trønnes <[email protected]>
Signed-off-by: David Lechner <[email protected]>
Reviewed-by: Noralf Trønnes <[email protected]>
Signed-off-by: Noralf Trønnes <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/fsl-dcu: Use .dumb_map_offset and .dumb_destroy defaults

This driver can use the drm_driver.dumb_destroy and
drm_driver.dumb_map_offset defaults, so no need to set them.

Cc: Stefan Agner <[email protected]>
Cc: Alison Wang <[email protected]>
Signed-off-by: Noralf Trønnes <[email protected]>
Acked-by: Stefan Agner <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915: fix backlight invert for non-zero minimum brightness

When we started following the backlight minimum brightness in
6dda730e55f4 ("drm/i915: respect the VBT minimum backlight brightness")
we overlooked the brightness invert quirk. Even if we invert the
brightness, we need to take the min limit into account. We probably
missed this because the invert has only been required on gen4 for proper
operation.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101127
Fixes: 6dda730e55f4 ("drm/i915: respect the VBT minimum backlight brightness")
Cc: Daniel Vetter <[email protected]>
Reviewed-by: Daniel Vetter <[email protected]>
Signed-off-by: Jani Nikula <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915/shrinker: Wrap need_resched() inside preempt-disable

In order for us to successfully detect the end of a timeslice,
preemption must be disabled. Otherwise, inside the loop we may be
preempted many times without our noticing, and each time our timeslice
will be reset, invalidating need_resched()

Reported-by: Joonas Lahtinen <[email protected]>
Reported-by: Tomi Sarvela <[email protected]>
Fixes: 290271de34f6 ("drm/i915: Spin for struct_mutex inside shrinker")
Signed-off-by: Chris Wilson <[email protected]>
Cc: Mika Kuoppala <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Cc: <[email protected]> # v4.13-rc1+
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Tested-by: Joonas Lahtinen <[email protected]>
Reviewed-by: Joonas Lahtinen <[email protected]>

drm/i915/perf: Initialise the dynamic sysfs attr

Use sysfs_attr_init() to dynamically initialise the
oa_config->sysfs_metric_id.attr as it has the important side-effect of
setting the lockdep key.

[    4.971513] [drm] Initialized i915 1.6.0 20170731 for 0000:00:02.0 on minor 0
[    4.973489] BUG: key ffff88026f6e7bb0 not in .data!
[    4.973506] DEBUG_LOCKS_WARN_ON(1)
[    4.973518] ------------[ cut here ]------------
[    4.973547] WARNING: CPU: 1 PID: 258 at kernel/locking/lockdep.c:3156 lockdep_init_map+0x1b2/0x1c0
[    4.973567] Modules linked in: i915(+) x86_pkg_temp_thermal intel_powerclamp coretemp crct10dif_pclmul crc32_pclmul ghash_clmulni_intel snd_hda_intel snd_hda_codec snd_hwdep snd_hda_core snd_pcm r8169 mei_me mii mei lpc_ich prime_numbers i2c_hid pinctrl_broxton pinctrl_intel
[    4.973645] CPU: 1 PID: 258 Comm: systemd-udevd Not tainted 4.13.0-rc3-CI-CI_DRM_2915+ #1
[    4.973664] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./J4205-ITX, BIOS P1.10 09/29/2016
[    4.973686] task: ffff8802704c2740 task.stack: ffffc90000224000
[    4.973700] RIP: 0010:lockdep_init_map+0x1b2/0x1c0
[    4.973712] RSP: 0018:ffffc90000227a10 EFLAGS: 00010282
[    4.973726] RAX: 0000000000000016 RBX: ffff880262aac010 RCX: 0000000000000000
[    4.973741] RDX: 0000000080000001 RSI: 0000000000000001 RDI: ffffffff810ed1ab
[    4.973757] RBP: ffffc90000227a30 R08: 0000000000000001 R09: 0000000000000000
[    4.973774] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88026f6e7bb0
[    4.973789] R13: 0000000000000000 R14: ffff88026f6e7b98 R15: ffffffff81a24da0
[    4.973805] FS:  00007f588d7f58c0(0000) GS:ffff88027fc80000(0000) knlGS:0000000000000000
[    4.973823] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    4.973837] CR2: 00000082482e32a0 CR3: 0000000270531000 CR4: 00000000003406e0
[    4.973852] Call Trace:
[    4.973864]  __kernfs_create_file+0x71/0xe0
[    4.973876]  sysfs_add_file_mode_ns+0x85/0x1a0
[    4.973890]  internal_create_group+0xe5/0x2b0
[    4.973903]  sysfs_create_group+0xe/0x10
[    4.973985]  i915_perf_register+0xd9/0x220 [i915]
[    4.974044]  i915_driver_load+0xa72/0x16b0 [i915]
[    4.974124]  i915_pci_probe+0x32/0x90 [i915]

Annoyingly detected by CI, but not reported due to it occurring during boot
and disabling lockdep for later runs.

Fixes: f89823c21224 ("drm/i915/perf: Implement I915_PERF_ADD/REMOVE_CONFIG interface")
Signed-off-by: Chris Wilson <[email protected]>
Cc: Matthew Auld <[email protected]>
Cc: Lionel Landwerlin <[email protected]>
Cc: Andrzej Datczuk <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Matthew Auld <[email protected]>
Reviewed-by: Lionel Landwerlin <[email protected]>

drm: stm: remove dead code and pointless local lut storage

The redundant fb helper .load_lut is no longer used, and can not
work right without also providing the fb helpers .gamma_set and
.gamma_get thus rendering the code in this driver suspect.

Just remove the dead code.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: radeon: remove dead code and pointless local lut storage

The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Acked-by: Alex Deucher <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: nouveau: remove dead code and pointless local lut storage

The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: mgag200: remove dead code and pointless local lut storage

The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: i915: remove dead code and pointless local lut storage

The driver stores lut values from the fbdev interface, and is able
to give them back, but does not appear to do anything with these
lut values. The generic fb helpers have replaced this function,
and may even have made the driver work for the C8 mode from the
fbdev interface. But that is untested.

Since the fb helpers .gamma_set and .gamma_get are obsolete,
remove the dead code.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: gma500: remove dead code and pointless local lut storage

The redundant fb helpers .gamma_set and .gamma_get are no longer
used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: cirrus: remove dead code and pointless local lut storage

The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: ast: remove dead code and pointless local lut storage

The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: armada: remove dead empty functions

The redundant fb helpers .gamma_set and .gamma_get are no longer used.
Remove the dead code.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm: amd: remove dead code and pointless local lut storage

The redundant fb helpers .load_lut, .gamma_set and .gamma_get are
no longer used. Remove the dead code and hook up the crtc .gamma_set
to use the crtc gamma_store directly instead of duplicating that
info locally.

Acked-by: Daniel Vetter <[email protected]>
Signed-off-by: Peter Rosin <[email protected]>
Acked-by: Alex Deucher <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/rockchip: fix race with kms hotplug and fbdev

According to the kerneldoc[0], should do fbdev setup before calling
drm_kms_helper_poll_init(), otherwise, Kms hotplug event may race
into fbdev helper initial, and fb_helper->dev may be NULL pointer,
that would cause the bug:
[    0.735411] [00000200] *pgd=00000000f6ffe003, *pud=00000000f6ffe003, *pmd=0000000000000000
[    0.736156] Internal error: Oops: 96000005 [#1] PREEMPT SMP
[    0.736648] Modules linked in:
[    0.736930] CPU: 2 PID: 20 Comm: kworker/2:0 Not tainted 4.4.41 #20
[    0.737480] Hardware name: Rockchip RK3399 Board rev2 (BOX) (DT)
[    0.738020] Workqueue: events cdn_dp_pd_event_work
[    0.738447] task: ffffffc0f21f3100 ti: ffffffc0f2218000 task.ti: ffffffc0f2218000
[    0.739109] PC is at mutex_lock+0x14/0x44
[    0.739469] LR is at drm_fb_helper_hotplug_event+0x30/0x114
[    0.756253] [<ffffff8008a344f4>] mutex_lock+0x14/0x44
[    0.756260] [<ffffff8008445708>] drm_fb_helper_hotplug_event+0x30/0x114
[    0.756271] [<ffffff8008473c84>] rockchip_drm_output_poll_changed+0x18/0x20
[    0.756280] [<ffffff8008439fcc>] drm_kms_helper_hotplug_event+0x28/0x34
[    0.756286] [<ffffff800846c444>] cdn_dp_pd_event_work+0x394/0x3c4
[    0.756295] [<ffffff80080b2b38>] process_one_work+0x218/0x3e0
[    0.756302] [<ffffff80080b3538>] worker_thread+0x2e8/0x404
[    0.756308] [<ffffff80080b7e70>] kthread+0xe8/0xf0
[    0.756316] [<ffffff8008082690>] ret_from_fork+0x10/0x40

[0]: https://01.org/linuxgraphics/gfx-docs/drm/gpu/drm-kms-helpers.html

Signed-off-by: Mark Yao <[email protected]>
Reviewed-by: Sandy huang <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/rockchip: vop: report error when check resource error

The user would be confused while facing a error commit without
any error report.

Signed-off-by: Mark Yao <[email protected]>
Reviewed-by: Sandy huang <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/rockchip: vop: round_up pitches to word align

VOP pitch register is word align, need align to word.

VOP_WIN0_VIR:
  bit[31:16] win0_vir_stride_uv
    Number of words of Win0 uv Virtual width
  bit[15:0] win0_vir_width
    Number of words of Win0 yrgb Virtual width
    ARGB888 : win0_vir_width
    RGB888 : (win0_vir_width*3/4) + (win0_vir_width%3)
    RGB565 : ceil(win0_vir_width/2)
    YUV : ceil(win0_vir_width/4)

Signed-off-by: Mark Yao <[email protected]>
Reviewed-by: Sandy huang <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/rockchip: vop: fix NV12 video display error

fixup the scale calculation formula on the case
src_height == (dst_height/2).

Signed-off-by: Mark Yao <[email protected]>
Reviewed-by: Sandy huang <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/rockchip: vop: fix iommu page fault when resume

Iommu would get page fault with following path:
   vop_disable:
      1, disable all windows and set vop config done
      2, vop enter to standy, all windows not works, but their registers
         are not clean, when you read window's enable bit, may found the
         window is enable.

   vop_enable:
      1, memcpy(vop->regsbak, vop->regs, len)
         save current vop registers to vop->regsbak, then you can found
         window is enable on regsbak.
      2, VOP_WIN_SET(vop, win, gate, 1);
         force enable window gate, but gate and enable are on same
         hardware register, then window enable bit rewrite to vop hardware.
      3, vop power on, and vop might try to scan destroyed buffer,
         then iommu get page fault.

Move windows disable after vop regsbak restore, then vop regsbak mechanism
would keep tracing the modify, everything would be safe.

Signed-off-by: Mark Yao <[email protected]>
Reviewed-by: Sandy huang <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/rockchip: vop: no need wait vblank on crtc enable

Since atomic framework, crtc enable and disable are in pairs,
no need to wait vblank.

Signed-off-by: Mark Yao <[email protected]>
Reviewed-by: Sandy huang <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

agp: nvidia: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

agp: amd64: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

agp: sis: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

agp: efficeon: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

agp: ati: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

agp: ali: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

agp: intel: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

agp: amd-k7: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

agp: uninorth: constify pci_device_id.

pci_device_id are not supposed to change at runtime. All functions
working with pci_device_id provided by <linux/pci.h> work with
const pci_device_id. So mark the non-const structs as const.

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>

Merge branch 'topic-arcpgu-updates' of https://github.com/foss-for-synopsys-dwc-arc-processors/linux into drm-next

arcgpu minor updates.

* 'topic-arcpgu-updates' of https://github.com/foss-for-synopsys-dwc-arc-processors/linux:
  drm: arcpgu: Allow some clock deviation in crtc->mode_valid() callback
  drm: arcpgu: Fix module unload
  drm: arcpgu: Fix mmap() callback
  arcpgu: Simplify driver name
  drm/arcpgu: Opt in debugfs

Merge tag 'drm-next-du-20170803' of git://linuxtv.org/pinchartl/media into drm-next

rcar-du updates, contains vsp1 updates as well.

* tag 'drm-next-du-20170803' of git://linuxtv.org/pinchartl/media: (24 commits)
  drm: rcar-du: Use new iterator macros
  drm: rcar-du: Repair vblank for DRM page flips using the VSP
  drm: rcar-du: Fix race condition when disabling planes at CRTC stop
  drm: rcar-du: Wait for flip completion instead of vblank in commit tail
  drm: rcar-du: Use the VBK interrupt for vblank events
  drm: rcar-du: Add HDMI outputs to R8A7796 device description
  drm: rcar-du: Remove an unneeded NULL check
  drm: rcar-du: Setup planes before enabling CRTC to avoid flicker
  drm: rcar-du: Configure DPAD0 routing through last group on Gen3
  drm: rcar-du: Restrict DPLL duty cycle workaround to H3 ES1.x
  drm: rcar-du: Support multiple sources from the same VSP
  drm: rcar-du: Fix comments to comply with the kernel coding style
  drm: rcar-du: Use of_graph_get_remote_endpoint()
  v4l: vsp1: Add support for header display lists in continuous mode
  v4l: vsp1: Add support for multiple DRM pipelines
  v4l: vsp1: Add support for multiple LIF instances
  v4l: vsp1: Add support for new VSP2-BS, VSP2-DL and VSP2-D instances
  v4l: vsp1: Add support for the BRS entity
  v4l: vsp1: Add pipe index argument to the VSP-DU API
  v4l: vsp1: Don't create links for DRM pipeline
  ...

drm/i915: enable WaDisableDopClkGating for skl

This WA is required when decoupled frequencies for slice and unslice
are enabled. This disables DOP clock gating for skl.

v2: enable the WA for all gen9 platforms (not just for SKL GT4 where
    the hang issue is originally reported) to avoid rare hangs (David)
v3: as per WaDatabase, enable it only for SKL (Rodrigo)

Cc: David Weinehall <[email protected]>
Reviewed-by: David Weinehall <[email protected]>
Signed-off-by: Praveen Paneri <[email protected]>
Signed-off-by: Rodrigo Vivi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915: Fix PCH names for KBP and CNP.

No functional change.

KBP was based on SPT and spec wasn't clear about the full name.
There was the initial point of the "Point" confusion.

Later the split with Coffee Lake and Cannon Lake both using CNP
and also some uncertainty from the specs we had at that time
made us to propagated the mistake along.

So, let's fix this now and avoid propagating these wrong
"points".

Cc: Anusha Srivatsa <[email protected]>
Cc: Paulo Zanoni <[email protected]>
Signed-off-by: Rodrigo Vivi <[email protected]>
Reviewed-by: Anusha Srivatsa <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/fb-helper: add new drm_setup_crtcs_fb() function

This adds a new drm_setup_crtcs_fb() function to handle the parts of
drm_setup_crtcs() that touch fb_helper->fb and fb_helper->fbdev. When
drm_setup_crtcs() is called during initialization, these fields are NULL
because they have not been allocated yet.

There is currently a hack at the end of drm_fb_helper_single_fb_probe()
that sets fb_helper->fb, so it is moved to the new drm_setup_crtcs_fb()
function.

This is also done in preparation for addition setup that requires access
to fb_helper->fbdev.

Signed-off-by: David Lechner <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915/perf: Implement I915_PERF_ADD/REMOVE_CONFIG interface

The motivation behind this new interface is expose at runtime the
creation of new OA configs which can be used as part of the i915 perf
open interface. This will enable the kernel to learn new configs which
may be experimental, or otherwise not part of the core set currently
available through the i915 perf interface.

v2: Drop DRM_ERROR for userspace errors (Matthew)
    Add padding to userspace structure (Matthew)
    s/guid/uuid/ (Matthew)

v3: Use u32 instead of int to iterate through registers (Matthew)

v4: Lock access to dynamic config list (Lionel)

v5: by Matthew:
    Fix uninitialized error values
    Fix incorrect unwiding when opening perf stream
    Use kmalloc_array() to store register
    Use uuid_is_valid() to valid config uuids
    Declare ioctls as write only
    Check padding members are set to 0
    by Lionel:
    Return ENOENT rather than EINVAL when trying to remove non
    existing config

v6: by Chris:
    Use ref counts for OA configs
    Store UUID in drm_i915_perf_oa_config rather then using pointer
    Shuffle fields of drm_i915_perf_oa_config to avoid padding

v7: by Chris
    Rename uapi pointers fields to end with '_ptr'

v8: by Andrzej, Marek, Sebastian
    Update register whitelisting
    by Lionel
    Add more register names for documentation
    Allow configuration programming in non-paranoid mode
    Add support for value filter for a couple of registers already
    programmed in other part of the kernel

v9: Documentation fix (Lionel)
    Allow writing WAIT_FOR_RC6_EXIT only on Gen8+ (Andrzej)

v10: Perform read access_ok() on register pointers (Lionel)

Signed-off-by: Matthew Auld <[email protected]>
Signed-off-by: Lionel Landwerlin <[email protected]>
Signed-off-by: Andrzej Datczuk <[email protected]>
Reviewed-by: Andrzej Datczuk <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915: reorder NOA register definition to follow addresses

It makes things easier to read when implementing whitelisting in the
following patches.

Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Matthew Auld <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915/perf: disable NOA logic when not used

We already do it on Haswell and the documentation says it saves power.

Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Matthew Auld <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915/perf: leave GDT_CHICKEN_BITS programming in configs

There will be a need for userspaces configurations to set this
register. We can apply the same model inside the kernel for test
configs.

Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Matthew Auld <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915/perf: prune OA configs

In the following commit we'll introduce loadable userspace
configs. This change reworks how configurations are handled in the
perf driver and retains only the test configurations in kernel space.

We now store the test config in dev_priv and resolve the id only once
when opening the perf stream. The OA config is then handled through a
pointer to the structure holding the configuration details.

v2: Rework how test configs are handled (Lionel)

v3: Use u32 to hold number of register (Matthew)

v4: Removed unused dev_priv->perf.oa.current_config variable (Matthew)

v5: Lock device when accessing exclusive_stream (Lionel)

v6: Ensure OACTXCONTROL is always reprogrammed (Lionel)

v7: Switch a couple of index variable from int to u32 (Matthew)

Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Matthew Auld <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/i915/perf: fix flex eu registers programming

We were reserving fewer dwords in the ring than necessary. Indeed
we're always writing all registers once, so discard the actual number
of registers given by the user and just program the whitelisted ones
once.

Fixes: 19f81df2859e ("drm/i915/perf: Add OA unit support for Gen 8+")
Reported-by: Matthew Auld <[email protected]>
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Matthew Auld <[email protected]>
Cc: <[email protected]> # v4.12+
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/atmel-hlcdc: switch to drm_*{get,put} helpers

drm_*_reference() and drm_*_unreference() functions are just
compatibility alias for drm_*_get() and drm_*_put() and should not be
used by new code. So convert all users of compatibility functions to use
the new APIs.

Signed-off-by: Cihangir Akturk <[email protected]>
Acked-by: Boris Brezillon <[email protected]>
Signed-off-by: Boris Brezillon <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/atmel-hlcdc : constify drm_plane_helper_funcs and drm_plane_funcs.

drm_plane_helper_funcs and drm_plane_funcsare not supposed to change
at runtime. All functions working with drm_plane_helper_funcs and
drm_plane_funcs work with const. So mark the non-const structs as const.

File size before:
   text    data     bss     dec     hex filename
   6072     596       0    6668    1a0c atmel_hlcdc_plane.o

File size After adding 'const':
   text    data     bss     dec     hex filename
   6218     436       0    6654    19fe atmel_hlcdc_plane.o

Signed-off-by: Arvind Yadav <[email protected]>
Signed-off-by: Boris Brezillon <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/646415a3b2e62182f85254115e8491e5caf4b2c7.1499098826.git.arvind.yadav.cs@gmail.com

drm: rcar-du: Use new iterator macros

for_each_obj_in_state is about to be removed, so use the correct new
iterator macros.

Also look at new_plane_state instead of plane->state when looking up
the hw planes in use. They should be the same except when reallocating,
(in which case this code is skipped) and we should really stop looking
at obj->state whenever possible.

Signed-off-by: Maarten Lankhorst <[email protected]>
Reviewed-by: Laurent Pinchart <[email protected]>
Signed-off-by: Laurent Pinchart <[email protected]>

drm: rcar-du: Repair vblank for DRM page flips using the VSP

The driver recently switched from handling page flip completion in the
DU vertical blanking handler to the VSP frame end handler to fix a race
condition. This unfortunately resulted in incorrect timestamps in the
vertical blanking events sent to userspace as vertical blanking is now
handled after sending the event.

To fix this we must reverse the order of the two operations. The easiest
way is to handle vertical blanking in the VSP frame end handler before
sending the event. The VSP frame end interrupt occurs approximately 50µs
earlier than the DU frame end interrupt, but this should not cause any
undue harm.

As we need to handle vertical blanking even when page flip completion is
delayed, the VSP driver now needs to call the frame end completion
callback unconditionally, with a new argument to report whether page
flip has completed.

With this new scheme the DU vertical blanking interrupt isn't needed
anymore, so we can stop enabling it.

Fixes: d503a43ac06a ("drm: rcar-du: Register a completion callback with VSP1")
Signed-off-by: Kieran Bingham <[email protected]>
Signed-off-by: Laurent Pinchart <[email protected]>
Acked-by: Mauro Carvalho Chehab <[email protected]>

drm: rcar-du: Fix race condition when disabling planes at CRTC stop

When stopping the CRTC the driver must disable all planes and wait for
the change to take effect at the next vblank. Merely calling
drm_crtc_wait_one_vblank() is not enough, as the function doesn't
include any mechanism to handle the race with vblank interrupts.

Replace the drm_crtc_wait_one_vblank() call with a manual mechanism that
handles the vblank interrupt race.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>

drm: rcar-du: Wait for flip completion instead of vblank in commit tail

Page flips can take more than one vertical blanking to complete if
arming the page flips races with the vertical blanking interrupt.
Waiting for one vblank to complete the atomic commit in the commit tail
handler is thus incorrect, and can lead to framebuffers being released
while still being scanned out.

Fix this by waiting for flip completion instead, using the
drm_atomic_helper_wait_for_flip_done() helper.

Fixes: 0d230422d256 ("drm: rcar-du: Register a completion callback with VSP1")
Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>

drm: rcar-du: Use the VBK interrupt for vblank events

When implementing support for interlaced modes, the driver switched from
reporting vblank events on the vertical blanking (VBK) interrupt to the
frame end interrupt (FRM). This incorrectly divided the reported refresh
rate by two. Fix it by moving back to the VBK interrupt.

Fixes: 906eff7fcada ("drm: rcar-du: Implement support for interlaced modes")
Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>

drm: rcar-du: Add HDMI outputs to R8A7796 device description

Update the device description with the HDMI output.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>

drm: rcar-du: Remove an unneeded NULL check

"params" can't be NULL here.  The next lines assume that we either
hit the break statement of "params->mpixelclock == ~0UL".  The
inconsistent NULL checking makes static checkers complain.  I've just
removed the test.

Signed-off-by: Dan Carpenter <[email protected]>
Reviewed-by: Laurent Pinchart <[email protected]>
Signed-off-by: Laurent Pinchart <[email protected]>

drm: rcar-du: Setup planes before enabling CRTC to avoid flicker

Commit 52055bafa1ff ("drm: rcar-du: Move plane commit code from CRTC
start to CRTC resume") changed the order of the plane commit and CRTC
enable operations to accommodate the runtime PM requirements. However,
this introduced corruption in the first displayed frame, as the CRTC is
now enabled without any plane configured. On Gen2 hardware the first
frame will be black and likely unnoticed, but on Gen3 hardware we end up
starting the display before the VSP compositor, which is more
noticeable.

To fix this, revert the order of the commit operations back, and handle
runtime PM requirements in the CRTC .atomic_begin() and .atomic_enable()
helper operation handlers.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>

drm: rcar-du: Configure DPAD0 routing through last group on Gen3

On Gen3 SoCs DPAD0 routing is configured through the last CRTC group,
unlike on Gen2 where it is configured through the first CRTC group. Fix
the driver accordingly.

Fixes: 2427b3037710 ("drm: rcar-du: Add R8A7795 device support")
Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>

drm: rcar-du: Restrict DPLL duty cycle workaround to H3 ES1.x

The H3 ES1.x exhibits dot clock duty cycle stability issues. We can work
around them by configuring the DPLL to twice the desired frequency,
coupled with a /2 post-divider. This isn't needed on other SoCs and
breaks HDMI output on M3-W for a currently unknown reason, so restrict
the workaround to H3 ES1.x.

From an implementation point of view, move work around handling outside
of the rcar_du_dpll_divider() function by requesting a x2 DPLL output
frequency explicitly. The existing post-divider calculation mechanism
will then take care of dividing the clock by two automatically.

While at it, print a more useful debugging message to ease debugging
clock rate issues.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>

drm: rcar-du: Support multiple sources from the same VSP

On R-Car H3 ES2.0, DU channels 0 and 3 are served by two separate
pipelines from the same VSP. Support this in the DU driver.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>

drm: rcar-du: Fix comments to comply with the kernel coding style

To avoid mixing comment styles when new comments complying with the
kernel coding style are introduced, fix all multiline comments in one
go.

Signed-off-by: Laurent Pinchart <[email protected]>
Acked-by: Kieran Bingham <[email protected]>

drm: rcar-du: Use of_graph_get_remote_endpoint()

Now, we can use of_graph_get_remote_endpoint(). Let's use it.

Signed-off-by: Kuninori Morimoto <[email protected]>
Signed-off-by: Laurent Pinchart <[email protected]>

v4l: vsp1: Add support for header display lists in continuous mode

The VSP supports both header and headerless display lists. The latter is
easier to use when the VSP feeds data directly to the DU in continuous
mode, and the driver thus uses headerless display lists for DU operation
and header display lists otherwise.

Headerless display lists are only available on WPF.0. This has never
been an issue so far, as only WPF.0 is connected to the DU. However, on
H3 ES2.0, the VSP-DL instance has both WPF.0 and WPF.1 connected to the
DU. We thus can't use headerless display lists unconditionally for DU
operation.

Implement support for continuous mode with header display lists, and use
it for DU operation on WPF outputs that don't support headerless mode.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>
Acked-by: Mauro Carvalho Chehab <[email protected]>

v4l: vsp1: Add support for multiple DRM pipelines

The R-Car H3 ES2.0 VSP-DL instance has two LIF entities and can drive
two display pipelines at the same time. Refactor the VSP DRM code to
support that by introducing a vsp_drm_pipeline object that models one
display pipeline.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>
Acked-by: Mauro Carvalho Chehab <[email protected]>

v4l: vsp1: Add support for multiple LIF instances

The VSP2-DL instance (present in the H3 ES2.0 and M3-N SoCs) has two LIF
instances. Adapt the driver infrastructure to support multiple LIFs.
Support for multiple display pipelines will be added separately.

The change to the entity routing table removes the ability to connect
the LIF output to the HGO or HGT histogram generators. This feature is
only available on Gen2 hardware, isn't supported by the rest of the
driver, and has no known use case, so this isn't an issue.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>
Acked-by: Mauro Carvalho Chehab <[email protected]>

v4l: vsp1: Add support for new VSP2-BS, VSP2-DL and VSP2-D instances

New Gen3 SoCs come with two new VSP2 variants names VSP2-BS and VSP2-DL,
as well as a new VSP2-D variant on V3M and V3H SoCs. Add new entries for
them in the VSP device info table.

Signed-off-by: Laurent Pinchart <[email protected]>
Reviewed-by: Kieran Bingham <[email protected]>
Acked-by: Mauro Carvalho Chehab <[email protected]>

drm/i915: add const to bin_attribute

Add const to bin_attribute structures as they are only passed to the
functions sysfs_{remove/create}_bin_file or
device_{remove/create}_bin_file. The corresponding arguments are of
type const, so declare the structures to be const.

Signed-off-by: Bhumika Goyal <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/fb: Fix pointer dereference before null check.

fb_crtc is used before a null check, so move the use after the null check.

This was just identified by inspection. I haven't actually observed a crash
here, so it is possible that the null check could be unnecessary.

Signed-off-by: David Lechner <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]