drm/amd/display: Update SubVP Visual Confirm

[Why]
The visual confirm bar is displayed on all displays, but is only
updated on a flip. If an update causes SubVP to be disabled on
all displays, but there are no updates on the other displays, it
will lead to stale information displayed on the non-active
displays. An example of this is playing a fullscreen video on a
rotated display.

[How]
Add a procedure to update visual confirm color for all pipes when
committing a plane.

Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: fix dpms_off issue when disabling bios mode

[Why]
disable_vbios_mode_if_required() will set dpms_off to false
during boot when pixel clk dismatches with driver requires.
This will cause extra backlight on and off if OS call 2
times setmode.

[How]
Set dpms_off to true to keep power_off and
let OS control BL by display's powerState.

Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Zhongwei Zhang <Zhongwei.Zhang@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix invalid DPIA AUX reply causing system hang

[Why]
Some DPIA AUX replies have incorrect data length from original request.
This could lead to overwriting of destination buffer if reply length is
larger, which could cause invalid access to stack since many destination
buffers are declared as local variables.

[How]
Check for invalid length from DPIA AUX replies and trigger a retry if
reply length is not the same as original request. A DRM_WARN() dmesg log
is also produced.

Reviewed-by: Roman Li <Roman.Li@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Stylon Wang <stylon.wang@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 6.0.x

drm/amdgpu: Add psp_13_0_10_ta firmware to modinfo

TA firmware loaded on psp v13_0_10, but it is missing in modinfo.

Signed-off-by: Candice Li <candice.li@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add HUBP surface flip interrupt handler

On IGT, there is a test named amd_hotplug, and when the subtest basic is
executed on DCN31, we get the following error:

[drm] *ERROR* [CRTC:71:crtc-0] flip_done timed out
[drm] *ERROR* flip_done timed out
[drm] *ERROR* [CRTC:71:crtc-0] commit wait timed out
[drm] *ERROR* flip_done timed out
[drm] *ERROR* [CONNECTOR:88:DP-1] commit wait timed out
[drm] *ERROR* flip_done timed out
[drm] *ERROR* [PLANE:59:plane-3] commit wait timed out

After enable the page flip log with the below command:

 echo -n 'format "[PFLIP]" +p' > /sys/kernel/debug/dynamic_debug/control

It is possible to see that the flip was submitted, but DC never replied
back, which generates time-out issues. This is an indication that the
HUBP surface flip is missing. This commit fixes this issue by adding
hubp1_set_flip_int to DCN31.

Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix access timeout to DPIA AUX at boot time

[Why]
Since introduction of patch "Query DPIA HPD status.", link detection at
boot could be accessing DPIA AUX, which will not succeed until
DMUB outbox messaging is enabled and results in below dmesg logs:

[  160.840227] [drm:amdgpu_dm_process_dmub_aux_transfer_sync [amdgpu]] *ERROR* wait_for_completion_timeout timeout!

[How]
Enable DMUB outbox messaging before link detection at boot time.

Reviewed-by: Wayne Lin <Wayne.Lin@amd.com>
Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Stylon Wang <stylon.wang@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 6.0.x

drm/amdgpu: clarify DC checks

There are several places where we don't want to check
if a particular asic could support DC, but rather, if
DC is enabled.  Set a flag if DC is enabled and check
for that rather than if a device supports DC or not.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: move non-DC vblank handling out of irq code

Move it into the DCE code for each generation. This avoids
confusion with the different display paths.

v2: no need for a hotplug worker for vkms

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: rework SR-IOV virtual display handling

virtual display is enabled unconditionally in SR-IOV, but
without specifying the virtual_display module, the number
of crtcs defaults to 0.  Set a single display by default
for SR-IOV if the virtual_display parameter is not set.
Only enable virtual display by default on SR-IOV on asics
which actually have display hardware.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use proper DC check in amdgpu_display_supported_domains()

amdgpu_device_asic_has_dc_support() just checks the asic itself.
amdgpu_device_has_dc_support() is a runtime check which not
only checks the asic, but also other things in the driver
like whether virtual display is enabled.  We want the latter
here.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove programming GCMC_VM_FB_LOCATION* on gfxhub_v3_0_3 in VF

[Why]
GCMC_VM related registers should be programmed by PSP on host side.
L1 and RLCG will block these regisers on VF.

[How]
Remove programming GCMC_VM_FB_LOCATION_BASE/TOP on gfxhub_v3_0_3 under SRIOV VF.

Signed-off-by: Yifan Zha <Yifan.Zha@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Stop clearing kiq position during unload

Do not clear kiq position in RLC_CP_SCHEDULER so that CP could perform
IDLE-SAVE after VF fini. CPG also needs to be active in save command.

v2: drop unused variable (Alex)

Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix memory leak in amdgpu_cs_pass1

When p->gang_size equals 0, amdgpu_cs_pass1() will return directly
without freeing chunk_array, which will cause a memory leak issue,
this patch fixes it.

Fixes: 4624459c84d7 ("drm/amdgpu: add gang submit frontend v6")
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: delete the duplicate .set_odm_bypass initialization in dcn314_tg_funcs

Fix below sparse warning:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn314/dcn314_optc.c:244:18: warning: Initializer entry defined twice
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn314/dcn314_optc.c:257:18:   also defined here

Fixes: 5ade1b951dec ("drm/amd/display: Add OTG/ODM functions")
Signed-off-by: Liu Jian <liujian56@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Replace one-element array with flex-array member

One-element arrays are deprecated, and we are replacing them with
flexible array members instead. So, replace one-element array with
flexible-array member in structs _ATOM_CONNECTOR_DEVICE_TAG_RECORD,
_ATOM_OBJECT_GPIO_CNTL_RECORD, _ATOM_BRACKET_LAYOUT_RECORD,
_ATOM_BRACKET_LAYOUT_RECORD, _ATOM_LEAKAGE_VOLTAGE_OBJECT_V3,
_ATOM_FUSION_SYSTEM_INFO_V3, _ATOM_I2C_DATA_RECORD,
_ATOM_I2C_DEVICE_SETUP_INFO, _ATOM_ASIC_MVDD_INFO and refactor the
rest of the code accordingly. While at it, removed a redundant casting.

Important to mention is that doing a build before/after this patch results
in no binary output differences.

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/79
Link: https://github.com/KSPP/linux/issues/238
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101836
Signed-off-by: Paulo Miguel Almeida <paulo.miguel.almeida.rodenas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Use drm_mode_init() for on-stack modes

Initialize on-stack modes with drm_mode_init() to guarantee
no stack garbage in the list head, or that we aren't copying
over another mode's list head.

Based on the following cocci script, with manual fixups:
@decl@
identifier M;
expression E;
@@
- struct drm_display_mode M = E;
+ struct drm_display_mode M;

@@
identifier decl.M;
expression decl.E;
statement S, S1;
@@
struct drm_display_mode M;
... when != S
+ drm_mode_init(&M, &E);
+
S1

@@
expression decl.E;
@@
- &*E
+ E

Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Leo Li <sunpeng.li@amd.com>
Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: amd-gfx@lists.freedesktop.org
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Ignore stop rlc on SRIOV environment.

For SRIOV, the guest driver should not do stop rlc. The host
handles programing RLC.

On SRIOV, the stop rlc will be hang (RLC related registers are
blocked by policy) when the RLCG interface is not enabled.

Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Gavin Wan <Gavin.Wan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add vram reservation based on vram_usagebyfirmware_v2_2

Move TMR region from top of FB to 2MB for FFBM, so we need to
reserve TMR region firstly to make sure TMR can be allocated at 2MB

Signed-off-by: Tong Liu01 <Tong.Liu01@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: remove the check of init status in psp_ras_initialize

The initialized status indicates RAS TA is loaded, but in some cases
(such as RAS fatal error) RAS TA could be destroyed although it's not
unloaded. Hence we load RAS TA unconditionally here.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Candice Li <candice.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: skip disabling all smu features on smu_v13_0_10 in suspend

skip disabling all smu features on smu_v13_0_10 in suspend

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: force read discovery file if set discovery=2

If discovery is set to 2 in module parameters explicitly, the
intention is to use the discovery file in FW rather than the one in
BIOS, usually because the latter is incorrect. This patch to force
read discovery file if set discovery=2.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Tim Huang <Tim.Huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm: Move radeon and amdgpu Kconfig options into their directories

Most Kconfig options to enable a driver are in the Kconfig file
inside the relevant directory, move these two to the same.

Signed-off-by: Andrew Davis <afd@ti.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/radeon: Add HD-audio component notifier support (v6)

This patch adds the support for the notification of HD-audio hotplug
via the already existing drm_audio_component framework to radeon
driver.  This allows us more reliable hotplug notification and ELD
transfer without accessing HD-audio bus; it's more efficient, and more
importantly, it works without waking up the runtime PM.

The implementation is rather simplistic: radeon driver provides the
get_eld ops for HD-audio, and it notifies the audio hotplug via
pin_eld_notify callback upon each radeon_audio_enable() call.
The pin->id is referred as the port number passed to the notifier
callback, and the corresponding connector is looked through the
encoder list in the get_eld callback in turn.

The bind and unbind callbacks handle the device-link so that it
assures the PM call order.

Also, as a gratis bonus, this patch "fixes" the regression by the
recent change in HD-audio to be more strict for the HDMI/DP
connection, too.  Since the HD-audio HDMI/DP codec requires both the
connection bit and the valid ELD to be provided, it started failing on
some RADEON gfx boards where the ELD update performed instably.  As
this change switches the communication to a direct way between the
audio and the graphics drivers, now the system receives the proper
ELD, and the HDMI/DP hotplug starts working again.

[ v2: fix the logic in radeon_audio_component_get_eld to walk the
  connector list since that is where the EDID lives and we can
  derive the encoder from the connector because the encoder has
  not been assigned at this point (i.e., during monitor probe).

  v3: the component binding is moved outside radeon_audio_init() and
  _fini(), as those are called from suspend/resume, too.
  Drop modeset lock calls that caused Oops.
  Moved Kconfig change so that it can be applied on older kernels.

  v4: revive drm_modeset_lock*() again, add the missing
  device_link_remove() call at unbinding

  v5: squash in mutex fix

  v6: squash in fix for audio get_eld callback ]

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1569
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Set MTYPE in PTE based on BO flags

The same BO may need different MTYPEs and SNOOP flags in PTEs depending
on its current location relative to the mapping GPU. Setting MTYPEs from
clients ahead of time is not practical for coherent memory sharing.
Instead determine the correct MTYPE for the desired coherence model and
current BO location when updating the page tables.

To maintain backwards compatibility with MTYPE-selection in
AMDGPU_VA_OP_MAP, the coherence-model-based MTYPE selection is only
applied if it chooses an MTYPE other than MTYPE_NC (the default).

Add two AMDGPU_GEM_CREATE_... flags to indicate the coherence model. The
default if no flag is specified is non-coherent (i.e. coarse-grained
coherent at dispatch boundaries).

Update amdgpu_amdkfd_gpuvm.c to use this new method to choose the
correct MTYPE depending on the current memory location.

v2:
* check that bo is not NULL (e.g. PRT mappings)
* Fix missing ~ bitmask in gmc_v11_0.c
v3:
* squash in "drm/amdgpu: Inherit coherence flags on dmabuf import"

Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: only fill dirty rectangles when PSR is enabled

Currently, we are calling fill_dc_dirty_rects() even if PSR isn't
supported by the relevant link in amdgpu_dm_commit_planes(), this is
undesirable especially because when drm.debug is enabled we are printing
messages in fill_dc_dirty_rects() that are only useful for debugging PSR
(and confusing otherwise). So, we can instead limit the filling of dirty
rectangles to only when PSR is enabled.

Reviewed-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Have risk for memory exhaustion

In dcn*_clock_source_create when dcn*_clk_src_construct fails allocated
clk_src needs release. A local attack could use this to cause memory
exhaustion.

Signed-off-by: LongJun Tang <tanglongjun@kylinos.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Replace 1-element array with flexible-array member

One-element arrays are deprecated, and we are replacing them with
flexible array members instead. So, replace one-element array with
flexible-array member in structs _ATOM_GPIO_PIN_ASSIGNMENT,
_ATOM_DISPLAY_OBJECT_PATH, _ATOM_DISPLAY_OBJECT_PATH_TABLE,
_ATOM_OBJECT_TABLE and refactor the rest of the code accordingly.

Important to mention is that doing a build before/after this patch results
in no functional binary output differences.

This helps with the ongoing efforts to tighten the FORTIFY_SOURCE
routines on memcpy() and help us make progress towards globally
enabling -fstrict-flex-arrays=3 [1].

Link: https://github.com/KSPP/linux/issues/79
Link: https://github.com/KSPP/linux/issues/238
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101836
Signed-off-by: Paulo Miguel Almeida <paulo.miguel.almeida.rodenas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/radeon: Add the missed acpi_put_table() to fix memory leak

When the radeon driver reads the bios information from ACPI
table in radeon_acpi_vfct_bios(), it misses to call acpi_put_table()
to release the ACPI memory after the init, so add acpi_put_table()
properly to fix the memory leak.

v2: fix text formatting (Alex)

Fixes: 268ba0a99f89 ("drm/radeon: implement ACPI VFCT vbios fetch (v3)")
Signed-off-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: complete gfxoff allow signal during suspend without delay

change guarantees that gfxoff is allowed before moving further in
s2idle sequence to add more reliablity about gfxoff in amdgpu IP's
suspend flow

Signed-off-by: Harsh Jain <harsh.jain@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Decouple RAS EEPROM addresses from chips

Abstract RAS I2C EEPROM addresses from chip names, and set their macro
definition names to the address they set, not the chip they attach
to. Since most chips either use I2C EEPROM address 0 or 40000h for the RAS
table start offset, this leaves us with only two macro definitions as
opposed to five, and removes the redundancy of four.

Cc: Candice Li <candice.li@amd.com>
Cc: Tao Zhou <tao.zhou1@amd.com>
Cc: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Kent Russell <kent.russell@amd.com>
Reviewed-by: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove redundant I2C EEPROM address

Remove redundant EEPROM_I2C_MADDR_54H address, since we already have it
represented (ARCTURUS), and since we don't include the I2C device type
identifier in EEPROM memory addresses, i.e. that high up in the device
abstraction--we only use EEPROM memory addresses, as memory is continuously
represented by EEPROM device(s) on the I2C bus.

Add a comment describing what these memory addresses are, how they come
about and how they're usually extracted from the device address byte.

Cc: Candice Li <candice.li@amd.com>
Cc: Tao Zhou <tao.zhou1@amd.com>
Cc: Alex Deucher <Alexander.Deucher@amd.com>
Fixes: c9bdc6c3cf39df ("drm/amdgpu: Add EEPROM I2C address support for ip discovery")
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Alex Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Make kfd_fill_cache_non_crat_info() as static

kfd_fill_cache_non_crat_info() is only used in kfd_topology.c,
so make it as static.

Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: enable mode1 reset on smu_v13_0_10

enable mode1 reset and prioritize debug port on smu_v13_0_10
as a more reliable message processing

v2 - move mode1 reset callback to smu_v13_0_0_ppt.c

Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Fix the memory overrun

Fix the memory overrun issue caused by wrong array size.

Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reported-by: coverity-bot <keescook+coverity-bot@chromium.org>
Addresses-Coverity-ID: 1527133 ("Memory - corruptions")
Fixes: c0cc999f3c32e6 ("drm/amdkfd: Fix the warning of array-index-out-of-bounds")
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: disable BACO on special BEIGE_GOBY card

Still avoid intermittent failure.

Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Drop eviction lock when allocating PT BO

Re-take the eviction lock immediately again after the allocation is
completed, to fix circular locking warning with drm_buddy allocator.

Move amdgpu_vm_eviction_lock/unlock/trylock to amdgpu_vm.h as they are
called from multiple files.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Unlock bo_list_mutex after error handling

Get below kernel WARNING backtrace when pressing ctrl-C to kill kfdtest
application.

If amdgpu_cs_parser_bos returns error after taking bo_list_mutex, as
caller amdgpu_cs_ioctl will not unlock bo_list_mutex, this generates the
kernel WARNING.

Add unlock bo_list_mutex after amdgpu_cs_parser_bos error handling to
cleanup bo_list userptr bo.

 WARNING: kfdtest/2930 still has locks held!
 1 lock held by kfdtest/2930:
  (&list->bo_list_mutex){+.+.}-{3:3}, at: amdgpu_cs_ioctl+0xce5/0x1f10 [amdgpu]
  stack backtrace:
   dump_stack_lvl+0x44/0x57
   get_signal+0x79f/0xd00
   arch_do_signal_or_restart+0x36/0x7b0
   exit_to_user_mode_prepare+0xfd/0x1b0
   syscall_exit_to_user_mode+0x19/0x40
   do_syscall_64+0x40/0x80

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: 3.2.211

DC version 3.2.211 brings along the following fixes:

- Wait for VBLANK during pipe programming
- Adding HDMI SCDC DEVICE_ID define
- Cursor update refactor: PSR-SU support condition
- Update 709 gamma to 2.222 as stated in the standerd
- Consider dp cable id only when data is non zero
- Waiting for 1 frame to fix the flash issue on PSR1
- Update SR watermarks for DCN314
- Allow tuning DCN314 bounding box
- Zeromem mypipe heap struct before using it
- Use min transition for SubVP into MPO
- Disable phantom OTG after enable for plane disable
- Disable DRR actions during state commit
- Fix fallback issues for DP LL 1.4a tests
- Fix FCLK deviation and tool compile issues
- Fix reg timeout in enc314_enable_fifo
- Fix gpio port mapping issue
- Only update link settings after successful MST link train
- Enforce minimum prefetch time for low memclk on DCN32
- Set correct EOTF and Gamut flag in VRR info
- Add margin for max vblank time for SubVP + DRR
- Populate DP2.0 output type for DML pipe

Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Reviewed-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Aric Cyr <Aric.Cyr@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Revert "drm/amdgpu: Revert "drm/amdgpu: getting fan speed pwm for vega10 properly""

This reverts commit 4545ae2ed3f2f7c3f615a53399c9c8460ee5bca7.

The origin patch "drm/amdgpu: getting fan speed pwm for vega10 properly" works fine.
Test failure is caused by test case self.

Signed-off-by: Asher Song <Asher.Song@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix the kerneldoc description

amdgpu_ttm_tt_set_userptr() is also called by the KFD as part of
initializing the user pages for userptr BOs and also while initializing
the GPUVM for a KFD process so update the function description.

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Populate DP2.0 output type for DML pipe

[Why]
DCN3.2 DML logic uses a new output type for DP2.0,
which will enable validation to pass for higher BW
timings that require DP2.0 link rates.

[How]
Populate the DML pipe with DP2.0 output type if
the signal type of the pipe_ctx is 128b/132b.

Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Acked-by: Jasdeep Dhillon <jasdeep.dhillon@amd.com>
Signed-off-by: George Shen <george.shen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add margin for max vblank time for SubVP + DRR

[Description]
- Incorporate FW delays as port of max VTOTAL calculated for
  SubVP + DRR cases (since it is part of the microschedule).
- Also add margin for the max VTOTAL possible for SubVP + DRR cases.
- Due to rounding errors in FW (integer arithmetic), the microschedule
  calculation can get pushed to the next frame (incorrectly) in cases
  where we use the max VTOTAL possible to complete the MCLK switch.
- When the rounding error occurs, we are only off by 1-2 lines,
  use 40us margin which is working consistently.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Set correct EOTF and Gamut flag in VRR info

[Why] FreeSync always use G2.2 EOTF and Native gamut
[How] Set EOTF and Gamut flags accordingly

Reviewed-by: Krunoslav Kovac <Krunoslav.Kovac@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Mike Hsieh <Mike.Hsieh@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Enforce minimum prefetch time for low memclk on DCN32

[WHY?]
Data return times when using lowest memclk can be <= 60us, which can cause
underflow on high bandwidth displays with a workload.

[HOW?]
Enforce a minimum prefetch time during validation for low memclk modes.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Only update link settings after successful MST link train

[WHY]
Currently driver reduces verified link caps on DPIA devices if a link is
trained at a link rate below the max rate verified during link detection.
This blocks high bandwidth modes after setting a low bandwidth mode.

[HOW]
Only update link rate after a successful link train if link is MST.

Reviewed-by: Mustapha Ghaddar <Mustapha.Ghaddar@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Michael Strauss <michael.strauss@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix gpio port mapping issue

[Why]
1. Port of gpio has different mapping.

[How]
1. Add a dummy entry in mapping table.
2. Fix incorrect mask bit field access.

Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Steve Su <steve.su@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix reg timeout in enc314_enable_fifo

[Why]
The link enablement sequence can end up resetting the encoder while
the PHY symclk isn't yet on.

This means that waiting for symclk on will timeout, along with the reset
bit never asserting high.

This causes unnecessary delay when enabling the link and produces a
warning affecting multiple IGT tests.

[How]
Don't wait for the symclk to be on here because firmware already does.

Don't wait for reset if we know the symclk isn't on.

Split the reset into a helper function that checks the bit and decides
whether or not a delay is sufficient.

Reviewed-by: Roman Li <Roman.Li@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix FCLK deviation and tool compile issues

[Why]
Recent backports from open source do not have header inclusion pattern
that is consistent with inclusion style in the rest of the file. This
breaks the internal tool builds as well. A recent commit erronously
modified the original DML formula for calculating
ActiveClockChangeLatencyHidingY. This resulted in a FCLK deviation
from the golden values.

[How]
Change the way in which display_mode_vba.h is included so that it is
consistent with the inclusion style in rest of the file which also fixes
the tool build. Restore the DML formula to its original state to fix the
FCLK deviation.

Reviewed-by: Aurabindo Pillai <Aurabindo.Pillai@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Chaitanya Dhere <chaitanya.dhere@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix fallback issues for DP LL 1.4a tests

[WHY]
Unlike DP or USBC, the USB4 link does not get its own encoder and
has to share therefore verify_caps is skipped.

[HOW]
Fix the fallback logic for automated tests and take that
into consideration for LT and LS.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Mustapha Ghaddar <mustapha.ghaddar@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Disable DRR actions during state commit

[WHY]
Committing a state while performing DRR actions can cause underflow.

[HOW]
Disabled features performing DRR actions during state commit.
Need to follow-up on why DRR actions affect state commit.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Wesley Chalmers <Wesley.Chalmers@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Disable phantom OTG after enable for plane disable

[Description]
- Need to disable phantom OTG after it's enabled
  in order to restore it to it's original state.
- If it's enabled and then an MCLK switch comes in
  we may not prefetch the correct data since the phantom
  OTG could already be in the middle of the frame.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Use min transition for SubVP into MPO

[Description]
- For SubVP transitioning into MPO, we want to
  use a minimal transition to prevent transient
  underflow
- Transitioning a phantom pipe directly into a
  "real" pipe can result in underflow due to the
  HUBP still having it's "phantom" programming
  when HUBP is unblanked (have to wait for next
  VUPDATE of the new OTG)
- Also ensure subvp pipe lock is acquired early
  enough for programming in dc_commit_state_no_check
- When disabling phantom planes, enable phantom OTG
  first so the disable gets the double buffer update

Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Zeromem mypipe heap struct before using it

[Why&How]
Bug was caused when moving variable from stack to heap because it was reusable
and garbage was left over, so we need to zero mem.

Reviewed-by: Martin Leung <Martin.Leung@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Martin Leung <Martin.Leung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Allow tuning DCN314 bounding box

[Why]
We're missing the helpers from dcn20 that would allow
overriding these with DC debug options.

[How]
Use dcn20_patch_bounding_box to support overriding all the
relevant values.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Update SR watermarks for DCN314

[Why & How]
New values requested by hardware after fine-tuning.
Update for all memory types.

Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Waiting for 1 frame to fix the flash issue on PSR1

[Why]
Needs more frames waiting before the PSR_Exit sending for the specific
TCON.

[How]
Add relock_delay_frame_cnt to control how many frames waiting are needed
before the PSR_Exit sending. The default value is 0. The Driver side can
set this variable for specific TCONs.

Reviewed-by: Robin Chen <robin.chen@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Ryan Lin <tsung-hua.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Consider dp cable id only when data is non zero

Cable ID is a DP2 feature to identify max certified link rate that
a cable can carry. The cable identification method requires both
cable and display hardware support. Since the specs comes late, it is
anticipated that the first round of DP2 cables and displays may not
be fully compatible to reliably return cable ID data. Therefore the
decision of our cable id policy is that if the cable can return non
zero cable id data, we will take cable's link rate capability into
account. However if we get zero data, the cable link rate capability
is considered inconclusive. In this case, we will not take cable's
capability into account to avoid of over limiting hardware capability
from users. The max overall link rate capability is still determined
after actual dp pre-training. Cable id is considered as an auxiliary
method of determining max link bandwidth capability.

Reviewed-by: George Shen <George.Shen@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Wenjing Liu <wenjing.liu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Update 709 gamma to 2.222 as stated in the standerd

[WHY]
    Previously driver use gamma 2.2 for 709 color space,
    but the standard is to use gamma of 2.222

[HOW]
    Change it gamma to 2.222

Reviewed-by: Krunoslav Kovac <Krunoslav.Kovac@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Nawwar Ali <nawwar.ali@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Cursor update refactor: PSR-SU support condition

[Why]
PSR-SU requires extra conditions while cursor update.

Reviewed-by: Robin Chen <robin.chen@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Max Tseng <Max.Tseng@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Adding HDMI SCDC DEVICE_ID define

[Why && How]
We will need to differentiate vendor behavior in the future.

Reviewed-by: Chris Park <Chris.Park@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Leo Ma <hanghong.ma@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Wait for VBLANK during pipe programming

[Description]
- Wait for vblank during front end programming
  for global sync to ensure all double buffer
  updates take.
- This prevents underflow in some cases.

Reviewed-by: Martin Leung <Martin.Leung@amd.com>
Acked-by: Alan Liu <HaoPing.Liu@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: workaround for TLB seq race

It can happen that we query the sequence value before the callback
had a chance to run.

Workaround that by grabbing the fence lock and releasing it again.
Should be replaced by hw handling soon.

Signed-off-by: Christian König <christian.koenig@amd.com>
CC: stable@vger.kernel.org # 5.19+
Fixes: 5255e146c99a6 ("drm/amdgpu: rework TLB flushing")
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2113
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Philip Yang <Philip.Yang@amd.com>
Tested-by: Stefan Springer <stefanspr94@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Fix error handling in criu_checkpoint

Checkpoint BOs last. That way we don't need to close dmabuf FDs if
something else fails later. This avoids problematic access to user mode
memory in the error handling code path.

criu_checkpoint_bos has its own error handling and cleanup that does not
depend on access to user memory.

In the private data, keep BOs before the remaining objects. This is
necessary to restore things in the correct order as restoring events
depends on the events-page BO being restored first.

Fixes: be072b06c739 ("drm/amdkfd: CRIU export BOs as prime dmabuf objects")
Reported-by: Jann Horn <jannh@google.com>
CC: Rajneesh Bhardwaj <Rajneesh.Bhardwaj@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-and-tested-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Fix error handling in kfd_criu_restore_events

mutex_unlock before the exit label because all the error code paths that
jump there didn't take that lock. This fixes unbalanced locking errors
in case of restore errors.

Fixes: 40e8a766a761 ("drm/amdkfd: CRIU checkpoint and restore events")
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: update SMU IP v13.0.4 msg interface header

Some of the unused messages that were used earlier in development have
been freed up as spare messages, no intended functional changes.

Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Tim Huang <tim.huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Merge branch '00.06-gr-ampere' of https://gitlab.freedesktop.org/skeggsb/nouveau into drm-next

This is the pull request for a whole bunch of fixes and prep-work that
was done to support Ampere acceleration prior to GSP-RM being
available.  It uses the ACR firmware released by NVIDIA in
linux-firmware, as we do on earlier GPUs.  The work to support running
on top of GSP-RM also heavily depends on various pieces of this
series.

In addition to the new HW support, general stability of the driver
should be improved, especially around recovering HW from bugs that can
be generated by userspace driver components.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Ben Skeggs <bskeggs@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/CABDvA==s+nZD0n7CuRWLPE=Pj+02CN13r+ZQJxoHQ_EmR+o=XQ@mail.gmail.com

drm/nouveau/gr/ga102: initial support

v2:
- whitespace

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Gourav Samaiya <gsamaiya@nvidia.com>

drm/nouveau/ltc/ga102: initial support

v2. fixup for ga103 early merge

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/acr/ga102: initial support

v2. fixup for ga103 early merge

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Gourav Samaiya <gsamaiya@nvidia.com>

drm/nouveau/fb/ga102: load and boot VPR scrubber FW

v2. fixup for ga103 early merge

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Gourav Samaiya <gsamaiya@nvidia.com>

drm/nouveau/gr/tu102: remove gv100_grctx_unkn88c

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/tu102: add gv100_gr_init_4188a4

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/tu102-: fix support for sw_bundle64_init

We weren't sending the high bits, though they're zero currently anyway.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/tu102-: use sw_veid_bundle_init from firmware

NVIDIA provided this on Turing, but we kept using the hardcoded version
from Volta (where they didn't).

Switch to the firmware version prior to Ampere.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gv100-: drop a write from init_shader_exceptions()

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gv100-: move init_419bd8() after sw_ctx load

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gv100-: add NV_PGRAPH_PRI_PD_AB_DIST_CONFIG_1 to patch list

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gv100-: fix number of tile map registers

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gv100-: port smid mapping code from nvgpu

Essentially ripped verbatim from NVGPU, comments and all, and adapted to
nvkm's structs and style.

- maybe fixes an nvgpu bug though, a small tweak was needed to match RM

v2:
- remove unnecessary WARN_ON

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gp100-: modify init_fecs_exceptions

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gk20a,gm20b,gp10b: split out netlist parsing from fw loading

We'll want to reuse the former for loading from proper netlist images.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gp100-: fix number of zcull tile regs

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf117-: make ppc_nr[gpc] accurate

We're going to be pulling in a chunk of code from NVGPU to fixup our
SMID mappings on Volta and above, which depends on ppc_nr[gpc]
reflecting the actual number of PPCs present, not the maximum number.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: switch to newer style interrupt handler

Ampere.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: move some init to init_exception2()

Ampere.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: move some init to init_rop_exceptions()

Ampere.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: move reset during golden ctx init to fecs_reset()

Ampere.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: wfi after register-bashing golden init

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: gpfifo_ctl zero before init

Match RM.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: wait for FE_PWR_MODE_AUTO

This doesn't fix any known issue, but RM started doing it at some point,
so presumably it's needed for something.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: call FECS HALT_PIPE method before RC reset

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: call FECS WFI_GOLDEN_SAVE method

This won't work on Ampere, and, it's questionable whether we should have
been using our FW's method of storing the golden context image with NV's
firmware to begin with.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: stop using NV_PGRAPH_FECS_CTXSW_MAILBOX_CLEAR

This doesn't work on Ampere for some reason, switch to directly modifying
NV_PGRAPH_FECS_CTXSW_MAILBOX instead.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: make global attrib_cb actually global

This was thought to be per-channel initially - it's not.  The backing
pages for the VMM mappings are shared for all channels.

- switches to more straight-forward patch interfaces
- prepares for sub-context support
- this is saving a *sizeable* amount of vram

v2:
- whitespace

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: move misc context patching out of attrib_cb funcs

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: make global bundle_cb actually global

This was thought to be per-channel initially - it's not.  The backing
pages for the VMM mappings are shared for all channels.

- switches to more straight-forward patch interfaces
- prepares for sub-context support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: make global pagepool actually global

This was thought to be per-channel initially - it's not.  The backing
pages for the VMM mappings are shared for all channels.

- switches to more straight-forward patch interfaces
- prepares for sub-context support

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: generate golden context during first object alloc

Needed for GV100 (and only GV100 for some reason) for WFI_GOLDEN_SAVE.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/gr/gf100-: move some code around to make next commits nicer

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>

drm/nouveau/fifo: expose function to read engine ctxsw status

Needed to support Ampere differences in gr/gf100-:

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>