Rocket Lake has a third DPLL (called 'DPLL4') that must be used to
enable a third display.  Unlike EHL's variant of DPLL4, the RKL variant
behaves the same as DPLL0/1.  And despite its name, the DPLL4 registers
are offset as if it were DPLL2.

v2:
 - Add new .update_ref_clks() hook.

v3:
 - Renumber TBT PLL to '3' and switch _MMIO_PLL3 to _MMIO_PLL (Lucas)

v4:
 - Don't drop _MMIO_PLL3; although it's now unused, we're going to need
   it very soon again for upcoming DG1 patches.  (Lucas)

v5:
 - Don't re-number TBT PLL and beyond, just use new RKL_DPLL_CFGCR
   macros to lookup the proper registers instead.  Although renumbering
   the PLLs might be something we want to consider down the road, it
   opens a big can of worms right now since a bunch of places in the
   code have an assumption that the PLL table has idx==id and no holes.
   Renumbering creates a hole for TGL, so we'd either need to allow
   holes in the table or break the idx==id invariant, both of which are
   somewhat invasive changes to the design.

Bspec: 49202
Bspec: 49443
Bspec: 50288
Bspec: 50289
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Signed-off-by: NMatt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200716220551.2730644-4-matthew.d.roper@intel.comReviewed-by: NLucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

RKL and TGL share some general gen12 workarounds, but each platform also
has its own platform-specific workarounds.

v2:
 - Add Wa_1604555607 for RKL.  This makes RKL's ctx WA list identical to
   TGL's, so we'll have both functions call the tgl_ function for now;
   this workaround isn't listed for DG1 so we don't want to add it to
   the general gen12_ function.

Cc: Matt Atwood <matthew.s.atwood@intel.com>
Signed-off-by: NMatt Roper <matthew.d.roper@intel.com>
Reviewed-by: NJosé Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200716220551.2730644-3-matthew.d.roper@intel.comSigned-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

RKL uses a slightly different bit layout for the DPCLKA_CFGCR0 register.

v2:
 - Fix inverted mask application when updating ICL_DPCLKA_CFGCR0
 - Checkpatch style fixes

Bspec: 50287
Cc: Aditya Swarup <aditya.swarup@intel.com>
Signed-off-by: NMatt Roper <matthew.d.roper@intel.com>
Reviewed-by: NJosé Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200716220551.2730644-2-matthew.d.roper@intel.comSigned-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

It's silly to have if(SKL) checks in gen9_init_clock_gating() when
we can just move those bits into skl_init_clock_gating().

I'm not entirely convinced we even need this w/a, or if we do
then maybe we want it for kbl/cfl as well. IIRC it was only
listed in the wadb, but that is now dead so can't double check
anymore. Bspec doesn't seem to have any purely skl specific
DOP clock gating workarounds listed.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200716190426.17047-1-ville.syrjala@linux.intel.comReviewed-by: NJosé Roberto de Souza <jose.souza@intel.com>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Hours Of Battery Life is a new GEN12+ power-saving feature that allows
supported motherboards to use a special voltage swing table for eDP
panels that uses less power.

So here if supported by HW, OEM will set it in VBT and i915 will try
to train link with HOBL vswing table if link training fails it fall
back to the original table.

intel_ddi_dp_preemph_max() was optimized to only check the HOBL flag
instead of do something like is done in intel_ddi_dp_voltage_max()
because it is only called after the first entry of the voltage swing
table was loaded so the HOBL flag is valid at that point.

v3:
- removed a few parameters of icl_ddi_combo_vswing_program() that
can be taken from encoder

v4:
- using the HOBL vswing table until training fails completely (Ville)

v5:
- not reducing lane or link rate when link training fails with HOBL
active
- duplicated the HOBL voltage swing entry to match DP spec requirement

v6:
- removed the optional VS 3 & pre-emp 0 from HOBL table
- changed from u8:1 to bool to store hobl_failed/active

BSpec: 49291
BSpec: 49399
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Animesh Manna <animesh.manna@intel.com>
Cc: Manasi Navare <manasi.d.navare@intel.com>
Signed-off-by: NJosé Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200715175637.33763-1-jose.souza@intel.comSigned-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

0x0155 is rather Ivy Bridge PCI-E Root Port.
0x0157 from the same commit ff049b6c ("drm/i915: bind driver to ValleyView chipsets")
is likely wrong too. Nowhere is it independetly confirmed or mentioned.
Signed-off-by: NAlexei Podtelezhnikov <apodtele@gmail.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200428034752.3975-1-apodtele@gmail.comSigned-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Rather than manually implement our own module reference counting for perf
pmu events, finally realise that there is a module parameter to struct
pmu for this very purpose.
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: stable@vger.kernel.org
Reviewed-by: NTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200716094643.31410-1-chris@chris-wilson.co.ukSigned-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

The value we program to DDI_BUF_CTL changes at the following places:
- At enabling/disabling the output to configure the port width etc, and
  to enable/disable the DDI BUF function.
- At the beginning/end of link re-training to disable/re-enable the DDI
  BUF function.
- On HSW/BDW/SKL to change the voltage swing/pre-emph levels.

Except of the above the value we program to the DDI_BUF_CTL register
(intel_dp->DP) doesn't change, so no need to reprogram the register when
changing the link training patterns (which is programmed via the
DP_TP_CTL register on DDI platforms).

v2:
- Fix the commit message wrt. voltage/pre-emph level values in
  intel_dp->DP. (Ville)
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200714153141.10280-2-imre.deak@intel.comSigned-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

According to BSpec this flag should not be changed while the DDI
function is enabled. On BDW+ the DP_TP_CTL register spec also states it
explicitly that the HW takes care of enabling/disabling the scrambling
for training patterns (and it must stay enabled for normal pixel
output). Assume that this HW automatic handling of scrambling is also
true for HSW.

BSpec: 8013, 7557, 50484
Signed-off-by: NImre Deak <imre.deak@intel.com>
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200714153141.10280-1-imre.deak@intel.comSigned-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Explicitly check for i830 when assigning the .get_cdclk() vfunc,
and then deal with the case of not having assigned the vfunc
separately. Less confusing, and gets rid of the checkpatch complaint
about using {} on one branch but not the others.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200714152626.380-3-ville.syrjala@linux.intel.comReviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Some spaces have snuck in where we want tabs. Fix it.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200714152626.380-2-ville.syrjala@linux.intel.comReviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

There's a pointless hole in struct intel_cdclk_vals, get rid of it.
Fortunately we already use named initializers so the order does not
matter.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200714152626.380-1-ville.syrjala@linux.intel.comReviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

Since g4x the CFB base only takes a 28bit offset into stolen.
Not sure if the CFB is allowed to start below that limit but
then extend beyond it. Let's assume not and just restrict the
allocation to the first 256MiB (in the unlikely case
we have more stolen than that).

v2: s/BIT/BIT_ULL/ (Chris)

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200714201945.18959-1-ville.syrjala@linux.intel.comReviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NRodrigo Vivi <rodrigo.vivi@intel.com>

This reverts commit 2ddef176.

As it turned out VMWGFX needs a much wider audit to fix this.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NDave Airlie <airlied@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200811092400.188124-1-christian.koenig@amd.com

The drm_mode_config_reset patches are very important fixing a recently
introduced kernel crash, the others fix various older issues which are
a bit less serious in practice.
Signed-off-by: NDave Airlie <airlied@redhat.com>
From: "Roland Scheidegger (VMware)" <rscheidegger.oss@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200812005941.19465-1-rscheidegger.oss@gmail.com

amd-drm-fixes-5.9-2020-08-07:

amdgpu:
- Re-add spelling typo fix
- Sienna Cichlid fixes
- Navy Flounder fixes
- DC fixes
- SMU i2c fix
- Power fixes
Signed-off-by: NDave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200807222843.3909-1-alexander.deucher@amd.com

 * backmerge from drm-fixes at v5.8-rc7
 * add orientation quirk for ASUS T103HAF
 * drm/omap: force runtime PM suspend on system suspend
 * drm/tidss: fix modeset init for DPI panels
 * re-added docs for drm_gem_flink_ioctl()
 * ttm: fix page-offset calculation within TTM
Signed-off-by: NDave Airlie <airlied@redhat.com>

From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20200804125510.GA29670@linux-uq9g

I need to backmerge 5.8 as I've got a bunch of fixes sitting
on an rc7 base that I want to land.
Signed-off-by: NDave Airlie <airlied@redhat.com>

There is a spelling mistake in a DRM_ERROR message. Fix it.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NRoland Scheidegger <sroland@vmware.com>

There is a spelling mistake in a DRM_ERROR message. Fix it.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NRoland Scheidegger <sroland@vmware.com>

Same problem as in stdu, same fix.

Fixes: 51f644b4 ("drm/atomic-helper: reset vblank on crtc reset")
Acked-by: NCharmaine Lee <charmainel@vmware.com>
Reviewed-by: NZack Rusin <zackr@vmware.com>
Signed-off-by: NRoland Scheidegger <sroland@vmware.com>

Same problem as in stdu, same fix.

Fixes: 51f644b4 ("drm/atomic-helper: reset vblank on crtc reset")
Acked-by: NCharmaine Lee <charmainel@vmware.com>
Reviewed-by: NZack Rusin <zackr@vmware.com>
Signed-off-by: NRoland Scheidegger <sroland@vmware.com>

When converting to atomic the state reset was done by directly calling
the functions, and before the modeset object was fully initialized.
This means the various ->dev pointers weren't set up.

After

commit 51f644b4
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Fri Jun 12 18:00:49 2020 +0200

    drm/atomic-helper: reset vblank on crtc reset

this started to oops because now we're trying to derefence
drm_crtc->dev. Fix this up by entirely switching over to
drm_mode_config_reset, called once everything is set up.

Fixes: 51f644b4 ("drm/atomic-helper: reset vblank on crtc reset")
Reported-by: NTetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Tested-by: NRoland Scheidegger <sroland@vmware.com>
Signed-off-by: NRoland Scheidegger <sroland@vmware.com>

These if statements are supposed to be true if we ended the
list_for_each_entry() loops without hitting a break statement but they
don't work.

In the first loop, we increment "i" after the "if (i == unit)" condition
so we don't necessarily know that "i" is not equal to unit at the end of
the loop.

In the second loop we exit when mode is not pointing to a valid
drm_display_mode struct so it doesn't make sense to check "mode->type".

Fixes: a278724a ("drm/vmwgfx: Implement fbdev on kms v2")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: NRoland Scheidegger <sroland@vmware.com>
Signed-off-by: NRoland Scheidegger <sroland@vmware.com>

The "entry" pointer is an offset from the list head and it doesn't
point to a valid vmw_legacy_display_unit struct.  Presumably the
intent was to point to the last entry.

Also the "i++" wasn't used so I have removed that as well.

Fixes: d7e1958d ("drm/vmwgfx: Support older hardware.")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: NRoland Scheidegger <sroland@vmware.com>
Signed-off-by: NRoland Scheidegger <sroland@vmware.com>

Make use of the struct_size() helper instead of an open-coded version
in order to avoid any potential type mistakes.

This code was detected with the help of Coccinelle and, audited and
fixed manually.
Signed-off-by: NGustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: NRoland Scheidegger <sroland@vmware.com>

Merge tag 'drm-misc-next-fixes-2020-08-05' of git://anongit.freedesktop.org/drm/drm-misc into drm-next

drm-misc-next-fixes for v5.9-rc1:
- Fix drm_dp_mst_port refcount leaks in drm_dp_mst_allocate_vcpi
- Fix a fbcon OOB read in fbdev, found by syzbot.
- Mark vga_tryget static as it's not used elsewhere.
- Small fixes to xlnx.
- Remove null check for kfree in drm_dev_release.
- Fix DRM_FORMAT_MOD_AMLOGIC_FBC definition.
- Fix mode initialization in omap_connector_mode_valid().
Signed-off-by: NDave Airlie <airlied@redhat.com>

From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/b2043dad-f118-bd19-54a6-f23bf6264007@linux.intel.com

Reproducing bug report here:
After hibernating and resuming, DPM is not enabled. This remains the case
even if you test hibernate using the steps here:
https://www.kernel.org/doc/html/latest/power/basic-pm-debugging.html

I debugged the problem, and figured out that in the file hardwaremanager.c,
in the function, phm_enable_dynamic_state_management(), the check
'if (!hwmgr->pp_one_vf && smum_is_dpm_running(hwmgr) && !amdgpu_passthrough(adev) && adev->in_suspend)'
returns true for the hibernate case, and false for the suspend case.

This means that for the hibernate case, the AMDGPU driver doesn't enable DPM
(even though it should) and simply returns from that function.
In the suspend case, it goes ahead and enables DPM, even though it doesn't need to.

I debugged further, and found out that in the case of suspend, for the
CIK/Hawaii GPUs, smum_is_dpm_running(hwmgr) returns false, while in the case of
hibernate, smum_is_dpm_running(hwmgr) returns true.

For CIK, the ci_is_dpm_running() function calls the ci_is_smc_ram_running() function,
which is ultimately used to determine if DPM is currently enabled or not,
and this seems to provide the wrong answer.

I've changed the ci_is_dpm_running() function to instead use the same method that
some other AMD GPU chips do (e.g Fiji), which seems to read the voltage controller.
I've tested on my R9 390 and it seems to work correctly for both suspend and
hibernate use cases, and has been stable so far.

Bug: https://bugzilla.kernel.org/show_bug.cgi?id=208839Signed-off-by: NSandeep Raghuraman <sandy.8925@gmail.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

As VCN related dpm table setup needs VCN be in PG ungate state. Same logics
applies to JPEG.

V2: fix paste typo
V3: code cosmetic
Signed-off-by: NEvan Quan <evan.quan@amd.com>
Tested-by: NMatt Coffin <mcoffin13@gmail.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Add lock protections and avoid unnecessary actions
if the PG state is already the same as required.
Signed-off-by: NEvan Quan <evan.quan@amd.com>
Tested-by: NMatt Coffin <mcoffin13@gmail.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Swith default gpu reset method for sienna_cichlid to MODE1 reset.
Signed-off-by: NLikun Gao <Likun.Gao@amd.com>
Reviewed-by: NHawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

The i2c init/fini functions just register the i2c adapter.
There is no need to call them during hw init/fini.  They only
need to be called once per driver init/fini.  The previous
behavior broke runtime pm because we unregistered the i2c
adapter during suspend.
Tested-by: NTom St Denis <tom.stdenis@amd.com>
Reviewed-by: NKevin Wang <kevin1.wang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[why]
Display goes blank after driver installation.
Aux tuning parameters must be used for 2.x only.
Wrong dc_golden_table offset was used.

[How]
Implement a new enc3_hw_init function without VBIOS constants usage to
be called for 3.x
Calculate dc_golden_table offset using sum of
base dce_info offset and golden table offset
Signed-off-by: NIgor Kravchenko <Igor.Kravchenko@amd.com>
Reviewed-by: NAric Cyr <Aric.Cyr@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
During SetPathMode and UpdatePlanes, the plane state can be null. We default
to linear swizzle mode when plane state is null. This resulted in bandwidth
validation failing when trying to set 8K60 mode (which previously passed validation
during rebuild timing list).

[How]
Change the default swizzle mode from linear to 4kb_s and update pitch accordingly.
Signed-off-by: NGeorge Shen <george.shen@amd.com>
Reviewed-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[How]
Use dc_is_hdmi_signal to determine signal type.
Signed-off-by: NJinZe.Xu <JinZe.Xu@amd.com>
Reviewed-by: NCharlene Liu <Charlene.Liu@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
Sink OUI supported cap is not set so driver skips programming it.

[How]
Revert the change the skips OUI programming if the cap is not set
Signed-off-by: NAric Cyr <aric.cyr@amd.com>
Reviewed-by: NAnthony Koo <Anthony.Koo@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why&How]
use correct logger context
Signed-off-by: NHarry Wentland <harry.wentland@amd.com>
Reviewed-by: NRoman Li <Roman.Li@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Signed-off-by: NDmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Reviewed-by: NEric Bernstein <Eric.Bernstein@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

[Why]
For ver.4.4 and higher VBIOS contains default setting table.

{How]
Read Golden Settings Table from VBIOS, apply Aux tuning parameters.
Signed-off-by: NIgor Kravchenko <Igor.Kravchenko@amd.com>
Reviewed-by: NAric Cyr <Aric.Cyr@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Signed-off-by: NEric Bernstein <eric.bernstein@amd.com>
Reviewed-by: NChris Park <Chris.Park@amd.com>
Acked-by: NAurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>