Commit graph

126,670 commits

Author SHA1 Message Date
Maxime Ripard
98007a0d56
drm/bridge: Add encoder parameter to drm_bridge_funcs.attach
The drm_bridge structure contains an encoder pointer that is widely used
by bridge drivers. This pattern is largely documented as deprecated in
other KMS entities for atomic drivers.

However, one of the main use of that pointer is done in attach to just
call drm_bridge_attach on the next bridge to add it to the bridge list.
While this dereferences the bridge->encoder pointer, it's effectively
the same encoder the bridge was being attached to.

We can make it more explicit by adding the encoder the bridge is
attached to to the list of attach parameters. This also removes the need
to dereference bridge->encoder in most drivers.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Tested-by: Douglas Anderson <dianders@chromium.org>
Tested-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250313-bridge-connector-v6-1-511c54a604fb@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
2025-03-20 14:45:38 +01:00
Victor Skvortsov
9c05636ca7 drm/amdgpu: Skip pcie_replay_count sysfs creation for VF
VFs cannot read the NAK_COUNTER register. This information is only
available through PMFW metrics.

Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:56:27 -04:00
Candice Li
a5f7e90fe0 drm/amdgpu: Add active_umc_mask to ras init_flags
Add active_umc_mask to ras init_flags.

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>
2025-03-19 15:56:23 -04:00
Lijo Lazar
a7818b15cf Documentation/amdgpu: Add debug_mask documentation
Add description for debug_mask bit options.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:56:19 -04:00
Lijo Lazar
ab6893402a drm/amd/pm: Add debug bit for smu pool allocation
In certain cases, it's desirable to avoid PMFW log transactions to
system memory. Add a mask bit to decide whether to allocate smu pool in
device memory or system memory.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:56:13 -04:00
Alex Deucher
3b669df92c drm/amdgpu/vcn: adjust workload profile handling
No need to make the workload profile setup dependent
on the results of cancelling the delayed work thread.
We have all of the necessary checking in place for the
workload profile reference counting, so separate the
two.  As it is now, we can theoretically end up with
the call from begin_use happening while the worker
thread is executing which would result in the profile
not getting set for that submission.  It should not
affect the reference counting.

v2: bail early if the the profile is already active (Lijo)

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:56:09 -04:00
Alex Deucher
9e34d8d1a1 drm/amdgpu/gfx: adjust workload profile handling
No need to make the workload profile setup dependent
on the results of cancelling the delayed work thread.
We have all of the necessary checking in place for the
workload profile reference counting, so separate the
two.  As it is now, we can theoretically end up with
the call from begin_use happening while the worker
thread is executing which would result in the profile
not getting set for that submission.  It should not
affect the reference counting.

v2: bail early if the the profile is already active (Lijo)

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:56:04 -04:00
Candice Li
5762f9dcf7 drm/amdgpu: Add EEPROM I2C address support for smu v13_0_12
Add EEPROM I2C address support for smu v13_0_12.

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>
2025-03-19 15:56:00 -04:00
Alex Deucher
ca6575a32a drm/amdgpu/vcn: fix ref counting for ring based profile handling
We need to make sure the workload profile ref counts are
balanced.  This isn't currently the case because we can
increment the count on submissions, but the decrement may
be delayed as work comes in.  Track when we enable the
workload profile so the references are balanced.

v2: switch to a mutex and active flag
v3: fix mutex init

Fixes: 1443dd3c67 ("drm/amd/pm: fix and simplify workload handling")
Cc: Yang Wang <kevinyang.wang@amd.com>
Cc: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:54:36 -04:00
Alex Deucher
553673a3e1 drm/amdgpu/gfx: fix ref counting for ring based profile handling
We need to make sure the workload profile ref counts are
balanced.  This isn't currently the case because we can
increment the count on submissions, but the decrement may
be delayed as work comes in.  Track when we enable the
workload profile so the references are balanced.

v2: switch to a mutex and active flag
v3: fix mutex init

Fixes: 8fdb3958e3 ("drm/amdgpu/gfx: add ring helpers for setting workload profile")
Cc: Yang Wang <kevinyang.wang@amd.com>
Cc: Kenneth Feng <kenneth.feng@amd.com>
Tested-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:52:56 -04:00
Harish Kasiviswanathan
fed7efbb43 drm/amdkfd: Fix bug in config_dequeue_wait_counts
For certain ASICs where dequeue_wait_count don't need to be initialized,
pm_config_dequeue_wait_counts_v9 return without filling in the packet
information. However, the calling function interprets this as a success
and sends the uninitialized packet to firmware causing hang.

Fix the above bug by not calling pm_config_dequeue_wait_counts_v9 for
ASICs that don't need the value to be initialized.

v2: Removed redudant code.
    Tidy up code based on review comments
v3: Don't call pm_config_dequeue_wait_counts_v9 for certain ASICs

Fixes: ed962f8d06 ("drm/amdkfd: Add pm_config_dequeue_wait_counts API")
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Reviewed-by: Jonathan Kim <jonathan.kim@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:51:44 -04:00
FengWei
5199e8ac07 drm/radeon/uvd: Replace nested max() with single max3()
Use max3() macro instead of nesting max() to simplify the return
statement.

Signed-off-by: FengWei <feng.wei8@zte.com.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:51:40 -04:00
Christian König
0d9a95099d drm/amdgpu: grab an additional reference on the gang fence v2
We keep the gang submission fence around in adev, make sure that it
stays alive.

v2: fix memory leak on retry

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-19 15:51:31 -04:00
Jason Gunthorpe
cb83f4b965 gpu: host1x: Do not assume that a NULL domain means no DMA IOMMU
Previously with tegra-smmu, even with CONFIG_IOMMU_DMA, the default domain
could have been left as NULL. The NULL domain is specially recognized by
host1x_iommu_attach() as meaning it is not the DMA domain and
should be replaced with the special shared domain.

This happened prior to the below commit because tegra-smmu was using the
NULL domain to mean IDENTITY.

Now that the domain is properly labled the test in DRM doesn't see NULL.
Check for IDENTITY as well to enable the special domains.

This is the same issue and basic fix as seen in
commit fae6e669cd ("drm/tegra: Do not assume that a NULL domain means no
DMA IOMMU").

Fixes: c8cc2655cc ("iommu/tegra-smmu: Implement an IDENTITY domain")
Reported-by: Diogo Ivo <diogo.ivo@tecnico.ulisboa.pt>
Closes: https://lore.kernel.org/all/c6a6f114-3acd-4d56-a13b-b88978e927dc@tecnico.ulisboa.pt/
Tested-by: Diogo Ivo <diogo.ivo@tecnico.ulisboa.pt>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/0-v1-10dcc8ce3869+3a7-host1x_identity_jgg@nvidia.com
2025-03-19 19:05:40 +01:00
Tomasz Rusinowicz
b96dabdba9 drm/xe: Fix exporting xe buffers multiple times
The `struct ttm_resource->placement` contains TTM_PL_FLAG_* flags, but
it was incorrectly tested for XE_PL_* flags.
This caused xe_dma_buf_pin() to always fail when invoked for
the second time. Fix this by checking the `mem_type` field instead.

Fixes: 7764222d54 ("drm/xe: Disallow pinning dma-bufs in VRAM")
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Nirmoy Das <nirmoy.das@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: intel-xe@lists.freedesktop.org
Cc: <stable@vger.kernel.org> # v6.8+
Signed-off-by: Tomasz Rusinowicz <tomasz.rusinowicz@intel.com>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250218100353.2137964-1-jacek.lawrynowicz@linux.intel.com
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
2025-03-19 12:09:38 +01:00
Thomas Hellström
52eb8cd788 Merge drm/drm-next into drm-xe-next
Backmerging to bring in the xe shrinker from drm-next.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
2025-03-19 11:08:52 +01:00
Ian Forbes
e95635d776 drm/vmwgfx: Switch to exclusively using GEM references
Currently we use a combination of TTM and GEM reference counting which is
cumbersome. TTM references are used for kernel internal BOs and operations
like validation. Simply switching the ttm_bo_(get|put) calls to their
GEM equivalents is insufficient as not all BOs are GEM BOs so we must set
the GEM vtable for all BOs even if they are not exposed to userspace.

Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Ian Forbes <ian.forbes@broadcom.com>
Reviewed-by: Zack Rusin <zack.rusin@broadcom.com>
Signed-off-by: Zack Rusin <zack.rusin@broadcom.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250131200321.193939-1-ian.forbes@broadcom.com
2025-03-18 23:59:51 -04:00
Ian Forbes
f42c09e614 drm/vmwgfx: Fix dumb buffer leak
Dumb buffers were not being freed because the GEM reference that was
acquired in gb_surface_define was not dropped like it is in the 2D case.
Dropping this ref uncovered a few additional issues with freeing the
resources associated with dirty tracking in vmw_bo_free/release.

Additionally the TTM object associated with the surface were also leaking
which meant that when the ttm_object_file was closed at process exit the
destructor unreferenced an already destroyed surface.

The solution is to remove the destructor from the vmw_user_surface
associated with the dumb_buffer and immediately unreferencing the TTM
object which his removes it from the ttm_object_file.

This also allows the early return in vmw_user_surface_base_release for the
dumb buffer case to be removed as it should no longer occur.

The chain of references now has the GEM handle(s) owning the dumb buffer.
The GEM handles have a singular GEM reference to the vmw_bo which is
dropped when all handles are closed. When the GEM reference count hits
zero the vmw_bo is freed which then unreferences the surface via
vmw_resource_release in vmw_bo_release.

Fixes: d6667f0ddf ("drm/vmwgfx: Fix handling of dumb buffers")
Signed-off-by: Ian Forbes <ian.forbes@broadcom.com>
Reviewed-by: Zack Rusin <zack.rusin@broadcom.com>
Signed-off-by: Zack Rusin <zack.rusin@broadcom.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250123204424.836896-1-ian.forbes@broadcom.com
2025-03-18 23:59:11 -04:00
Janusz Krzysztofik
2230387314 drm/i915: Fix harmful driver register/unregister asymmetry
Starting with commit ec3e00b4ee ("drm/i915: stop registering if
drm_dev_register() fails"), we return from i915_driver_register()
immediately if drm_dev_register() fails, skipping remaining registration
steps, and continue only with remaining probe steps.  However, the
_unregister() counterpart called at driver remove knows nothing about that
skip and executes reverts of all those steps.  As a consequence, a number
of kernel warnings that taint the kernel are triggered:

<3> [525.823143] i915 0000:00:02.0: [drm] *ERROR* Failed to register driver for userspace access!
...
<4> [525.831069] ------------[ cut here ]------------
<4> [525.831071] i915 0000:00:02.0: [drm] drm_WARN_ON(power_domains->init_wakeref)
<4> [525.831095] WARNING: CPU: 6 PID: 3440 at drivers/gpu/drm/i915/display/intel_display_power.c:2074 intel_power_domains_disable+0xc2/0xd0 [i915]
...
<4> [525.831328] CPU: 6 UID: 0 PID: 3440 Comm: i915_module_loa Tainted: G     U             6.14.0-rc1-CI_DRM_16076-g7a632b6798b6+ #1
...
<4> [525.831334] RIP: 0010:intel_power_domains_disable+0xc2/0xd0 [i915]
...
<4> [525.831483] Call Trace:
<4> [525.831484]  <TASK>
...
<4> [525.831943]  i915_driver_remove+0x4b/0x140 [i915]
<4> [525.832028]  i915_pci_remove+0x1e/0x40 [i915]
<4> [525.832099]  pci_device_remove+0x3e/0xb0
<4> [525.832103]  device_remove+0x40/0x80
<4> [525.832107]  device_release_driver_internal+0x215/0x280
...

Moreover, that unexpected PM reference is left untouched (not released)
but overwritten, then that triggers another kernel warning at driver
release phase:

<4> [526.685700] ------------[ cut here ]------------
<4> [526.685706] i915 0000:00:02.0: [drm] i915 raw-wakerefs=1 wakelocks=1 on cleanup
<4> [526.685734] WARNING: CPU: 1 PID: 3440 at drivers/gpu/drm/i915/intel_runtime_pm.c:443 intel_runtime_pm_driver_release+0x75/0x90 [i915]
...
<4> [526.686090] RIP: 0010:intel_runtime_pm_driver_release+0x75/0x90 [i915]
...
<4> [526.686294] Call Trace:
<4> [526.686296]  <TASK>
...
<4> [526.687025]  i915_driver_release+0x7e/0xb0 [i915]
<4> [526.687243]  drm_dev_put.part.0+0x47/0x90
<4> [526.687250]  devm_drm_dev_init_release+0x13/0x30
<4> [526.687255]  devm_action_release+0x12/0x30
<4> [526.687261]  release_nodes+0x3a/0x120
<4> [526.687268]  devres_release_all+0x97/0xe0
<4> [526.687277]  device_unbind_cleanup+0x12/0x80
<4> [526.687282]  device_release_driver_internal+0x23a/0x280
...

A call to intel_power_domains_disable() was already there.  It triggers
the drm_WARN_ON() when it finds a reference to a wakeref taken on device
probe and not released after device registration failure.  That wakeref is
then left held forever once its handle gets lost overwritten with another
wakeref, hence another WARN() is called from
intel_runtime_pm_driver_release().

The WARN() triggered by kernfs_remove_by_name_ns() from
i915_teardown_sysfs()->i915_gpu_error_sysfs_teardown(), formerly
i915_teardown_error_capture(), was also there when the return was added.

A call to intel_gt_sysfs_unregister() that triggers the WARN() from
kobject_put() was added to intel_gt_driver_unregister() with commit
69d6bf5c37 ("drm/i915/gt: Fix memory leaks in per-gt sysfs").

Fix the asymmetry by failing the driver probe on device registration
failure and going through rewind paths.

For that to work as expected, we apparently need to start the rewind path
of i915_driver_register() with drm_dev_unregister(), even if
drm_dev_register() returned an error.

v5: Drop unsigned keyword from ret variable declaration (Krzysztof),
  - keep the "Failed to register driver for userspace access" error
    message (Krzysztof),
  - split PXP cleanup addition to rewind path out to a separate patch.
v4: Switch to taking an error rewind path on device registration failure
    (Krzysztof, Lucas).
v3: Based on Andi's commitment on introducing a flag, try to address
    Jani's "must find another way" by finding a better place and name for
    the flag (in hope that's what Jani had on mind),
  - split into a series of patches and limit the scope of the first (this)
    one to a minimum of omitting conditionally only those unregister
    (sub)steps that trigger kernel warnings when not registered.
v2: Check in _unregister whether the drm_dev_register has succeeded and
    skip some of the _unregister() steps. (Andi)

Link: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/10047
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/10131
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/10887
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12817
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Chris Wilson <chris.p.wilson@linux.intel.com>
Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Krzysztof Niemiec <krzysztof.niemiec@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Reviewed-by: Krzysztof Niemiec <krzysztof.niemiec@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314205202.809563-8-janusz.krzysztofik@linux.intel.com
2025-03-19 02:44:42 +01:00
Janusz Krzysztofik
4cce01dc3e drm/i915: Fix PXP cleanup missing from probe error rewind
Commit f67986b011 ("drm/i915/pxp: Promote pxp subsystem to top-level
of i915") added PXP initialization to driver probe path, but didn't add a
respective PXP cleanup on probe error.  That lack of cleanup seems
harmless as long as PXP is still unused and idle when a probe failure
occurs and error rewind path is entered, but as soon as PXP starts
consuming device and driver resources keeping them busy, kernel warnings
may be triggered when cleaning up resources provided by memory regions,
GGTT, GEM and/or VMA cache from the probe error rewind and/or module
unload paths because of missing PXP cleanup.  That scenario was observed
on attempts to fail the probe and enter the rewind path on injection of
now ignored error in device registration path.

Fix it.

Cc: Alan Previn <alan.previn.teres.alexis@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Reviewed-by: Krzysztof Niemiec <krzysztof.niemiec@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314205202.809563-7-janusz.krzysztofik@linux.intel.com
2025-03-19 02:44:41 +01:00
Janusz Krzysztofik
2ae485e3d1 drm/i915: Downgrade device register error if injected
Commit 8f460e2c78 ("drm/i915: Demidlayer driver loading") which
introduced manual device registration also added a message that is
submitted on device registration failure as an error.  If that failure is
triggered by error injection test, that's an expected error, but CI still
reports it as a bug.  Fix it.

Suggested-by: Krzysztof Niemiec <krzysztof.niemiec@intel.com>
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/9820
Cc: Chris Wilson <chris.p.wilson@linux.intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Reviewed-by: Krzysztof Niemiec <krzysztof.niemiec@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314205202.809563-6-janusz.krzysztofik@linux.intel.com
2025-03-19 02:44:40 +01:00
Andi Shyti
5ba97b5925 drm/i915/gt: Use proper sleeping functions for timeouts shorter than 20ms
msleep is very imprecise for timeouts shorter than 20
milliseconds and most probably will sleep longer. Use
uslee_range() instead.

Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314021225.11813-5-andi.shyti@linux.intel.com
2025-03-19 02:41:17 +01:00
Andi Shyti
5bebf804ef drm/i915/gt: Remove trailing blank lines
Remove useless blank lines before and after the brackets.

Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314021225.11813-4-andi.shyti@linux.intel.com
2025-03-19 02:40:49 +01:00
Andi Shyti
b6aa4b8b3e drm/i915/gt: Fix SPDX license format
Header files need to declare the SPDX under /* ... */ style
comments at the beginning of the file.

Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250314021225.11813-3-andi.shyti@linux.intel.com
2025-03-19 02:40:19 +01:00
Tomasz Pakuła
d9d4cb224e drm/amdgpu/pm: Handle SCLK offset correctly in overdrive for smu 14.0.2
Currently, it seems like the code was carried over from RDNA3 because
it assumes two possible values to set. RDNA4, instead of having:
0: min SCLK
1: max SCLK
only has
0: SCLK offset

This change makes it so it only reports current offset value instead of
showing possible min/max values and their indices. Moreover, it now only
accepts the offset as a value, without the indice index.

Additionally, the lower bound was printed as %u by mistake.

Old:
OD_SCLK_OFFSET:
0: -500Mhz
1: 1000Mhz
OD_MCLK:
0: 97Mhz
1: 1259MHz
OD_VDDGFX_OFFSET:
0mV
OD_RANGE:
SCLK_OFFSET:    -500Mhz       1000Mhz
MCLK:      97Mhz       1500Mhz
VDDGFX_OFFSET:    -200mv          0mv

New:
OD_SCLK_OFFSET:
0Mhz
OD_MCLK:
0: 97Mhz
1: 1259MHz
OD_VDDGFX_OFFSET:
0mV
OD_RANGE:
SCLK_OFFSET:    -500Mhz       1000Mhz
MCLK:      97Mhz       1500Mhz
VDDGFX_OFFSET:    -200mv          0mv

Setting this offset:
Old: "s 1 <offset>"
New: "s <offset>"

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4036
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Tomasz Pakuła <tomasz.pakula.oficjalny@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 1cfeb60e6e)
Cc: stable@vger.kernel.org # 6.12.x
2025-03-18 16:34:06 -04:00
Lo-an Chen
d60073294c drm/amd/display: Fix incorrect fw_state address in dmub_srv
[WHY]
The fw_state in dmub_srv was assigned with wrong address.
The address was pointed to the firmware region.

[HOW]
Fix the firmware state by using DMUB_DEBUG_FW_STATE_OFFSET
in dmub_cmd.h.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Lo-an Chen <lo-an.chen@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit f57b38ac85)
2025-03-18 16:33:38 -04:00
Mario Limonciello
acbf16a6ae drm/amd/display: Use HW lock mgr for PSR1 when only one eDP
[WHY]
DMUB locking is important to make sure that registers aren't accessed
while in PSR.  Previously it was enabled but caused a deadlock in
situations with multiple eDP panels.

[HOW]
Detect if multiple eDP panels are in use to decide whether to use
lock. Refactor the function so that the first check is for PSR-SU
and then replay is in use to prevent having to look up number
of eDP panels for those configurations.

Fixes: f245b400a2 ("Revert "drm/amd/display: Use HW lock mgr for PSR1"")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3965
Reviewed-by: ChiaHsuan Chung <chiahsuan.chung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit ed569e1279)
Cc: stable@vger.kernel.org
2025-03-18 16:33:11 -04:00
Yilin Chen
35f0f9f421 drm/amd/display: Fix message for support_edp0_on_dp1
[WHY]
The info message was wrong when support_edp0_on_dp1 is enabled

[HOW]
Use correct info message for support_edp0_on_dp1

Fixes: f6d17270d1 ("drm/amd/display: add a quirk to enable eDP0 on DP1")
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Yilin Chen <Yilin.Chen@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 79538e6365)
Cc: stable@vger.kernel.org
2025-03-18 16:32:33 -04:00
Philip Yang
542c3bb836 drm/amdkfd: Fix user queue validation on Gfx7/8
To workaround queue full h/w issue on Gfx7/8, when application create
AQL queue, the ring buffer bo allocate size is queue_size/2 and
map queue_size ring buffer to GPU in 2 pieces using 2 attachments, each
attachment map size is queue_size/2, with same ring_bo backing memory.

For Gfx7/8, user queue buffer validation should use queue_size/2 to
verify ring_bo allocation and mapping size.

Fixes: 68e599db7a ("drm/amdkfd: Validate user queue buffers")
Suggested-by: Tomáš Trnka <trnka@scm.com>
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit e7a477735f)
Cc: stable@vger.kernel.org
2025-03-18 16:31:25 -04:00
David Belanger
35b6162bb7 drm/amdgpu: Restore uncached behaviour on GFX12
Always use MTYPE_UC if UNCACHED flag is specified.

This makes kernarg region uncached and it restores
usermode cache disable debug flag functionality.

Do not set MTYPE_UC for COHERENT flag, on GFX12 coherence is handled by
shader code.

Signed-off-by: David Belanger <david.belanger@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit eb6cdfb807)
Cc: stable@vger.kernel.org # 6.12.x
2025-03-18 16:29:52 -04:00
Wentao Liang
86730b5261 drm/amdgpu/gfx12: correct cleanup of 'me' field with gfx_v12_0_me_fini()
In gfx_v12_0_cp_gfx_load_me_microcode_rs64(), gfx_v12_0_pfp_fini() is
incorrectly used to free 'me' field of 'gfx', since gfx_v12_0_pfp_fini()
can only release 'pfp' field of 'gfx'. The release function of 'me' field
should be gfx_v12_0_me_fini().

Fixes: 52cb80c12e ("drm/amdgpu: Add gfx v12_0 ip block support (v6)")
Signed-off-by: Wentao Liang <vulab@iscas.ac.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit ebdc52607a)
Cc: stable@vger.kernel.org # 6.12.x
2025-03-18 16:29:16 -04:00
Jay Cornwall
424648c383 drm/amdkfd: Fix instruction hazard in gfx12 trap handler
VALU instructions with SGPR source need wait states to avoid hazard
with SALU using different SGPR.

v2: Eliminate some hazards to reduce code explosion

Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Reviewed-by: Lancelot Six <lancelot.six@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 7e0459d453)
Cc: stable@vger.kernel.org # 6.12.x
2025-03-18 16:28:34 -04:00
Alex Deucher
5ca0040ecf drm/amdgpu/pm: wire up hwmon fan speed for smu 14.0.2
Add callbacks for fan speed fetching.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4034
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 90df6db62f)
Cc: stable@vger.kernel.org # 6.12.x
2025-03-18 16:27:56 -04:00
Harish Kasiviswanathan
19b53f9685 drm/amd/pm: add unique_id for gfx12
Expose unique_id for gfx12

Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 16fbc18cb0)
Cc: stable@vger.kernel.org # 6.12.x
2025-03-18 16:26:53 -04:00
David Rosca
7fc0765208 drm/amdgpu: Remove JPEG from vega and carrizo video caps
JPEG is only supported for VCN1+.

Signed-off-by: David Rosca <david.rosca@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 0a6e7b06bd)
Cc: stable@vger.kernel.org
2025-03-18 16:26:12 -04:00
David Rosca
ec33964d9d drm/amdgpu: Fix JPEG video caps max size for navi1x and raven
8192x8192 is the maximum supported resolution.

Signed-off-by: David Rosca <david.rosca@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 6e0d2fde3a)
Cc: stable@vger.kernel.org
2025-03-18 16:25:46 -04:00
David Rosca
f0105e1731 drm/amdgpu: Fix MPEG2, MPEG4 and VC1 video caps max size
1920x1088 is the maximum supported resolution.

Signed-off-by: David Rosca <david.rosca@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 1a0807feb9)
Cc: stable@vger.kernel.org
2025-03-18 16:25:16 -04:00
Nikita Zhandarovich
dd8689b52a drm/radeon: fix uninitialized size issue in radeon_vce_cs_parse()
On the off chance that command stream passed from userspace via
ioctl() call to radeon_vce_cs_parse() is weirdly crafted and
first command to execute is to encode (case 0x03000001), the function
in question will attempt to call radeon_vce_cs_reloc() with size
argument that has not been properly initialized. Specifically, 'size'
will point to 'tmp' variable before the latter had a chance to be
assigned any value.

Play it safe and init 'tmp' with 0, thus ensuring that
radeon_vce_cs_reloc() will catch an early error in cases like these.

Found by Linux Verification Center (linuxtesting.org) with static
analysis tool SVACE.

Fixes: 2fc5703abd ("drm/radeon: check VCE relocation buffer range v3")
Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 2d52de55f9)
Cc: stable@vger.kernel.org
2025-03-18 16:24:44 -04:00
Lijo Lazar
6c11d4a87d drm/amdgpu: Use wafl version for xgmi
XGMI and WAFL share the same versions. Use WAFL version if XGMI version
is not present in discovery.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:47 -04:00
Jesse.zhang@amd.com
cc63bcfd14 drm/amdgpu: Fix SDMA engine reset logic
The scheduler should restart only if the reset operation
succeeds  This ensures that new tasks are only submitted
to the queues after a successful reset.

Fixes: 4c02f73016 ("drm/amdgpu: Introduce conditional user queue suspension for SDMA resets")
Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse.Zhang <Jesse.zhang@amd.com>
Reviewed-by: Tim Huang <tim.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:47 -04:00
Tomasz Pakuła
1cfeb60e6e drm/amdgpu/pm: Handle SCLK offset correctly in overdrive for smu 14.0.2
Currently, it seems like the code was carried over from RDNA3 because
it assumes two possible values to set. RDNA4, instead of having:
0: min SCLK
1: max SCLK
only has
0: SCLK offset

This change makes it so it only reports current offset value instead of
showing possible min/max values and their indices. Moreover, it now only
accepts the offset as a value, without the indice index.

Additionally, the lower bound was printed as %u by mistake.

Old:
OD_SCLK_OFFSET:
0: -500Mhz
1: 1000Mhz
OD_MCLK:
0: 97Mhz
1: 1259MHz
OD_VDDGFX_OFFSET:
0mV
OD_RANGE:
SCLK_OFFSET:    -500Mhz       1000Mhz
MCLK:      97Mhz       1500Mhz
VDDGFX_OFFSET:    -200mv          0mv

New:
OD_SCLK_OFFSET:
0Mhz
OD_MCLK:
0: 97Mhz
1: 1259MHz
OD_VDDGFX_OFFSET:
0mV
OD_RANGE:
SCLK_OFFSET:    -500Mhz       1000Mhz
MCLK:      97Mhz       1500Mhz
VDDGFX_OFFSET:    -200mv          0mv

Setting this offset:
Old: "s 1 <offset>"
New: "s <offset>"

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4036
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Tomasz Pakuła <tomasz.pakula.oficjalny@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:47 -04:00
Flora Cui
b5aaa82e2b drm/amdgpu: release xcp_mgr on exit
Free on driver cleanup.

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Flora Cui <flora.cui@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:47 -04:00
Taimur Hassan
bed6bc66e8 drm/amd/display: 3.2.325
This version brings along following fixes:
- Use DPM table clk setting for dml2 soc dscclk
- Update static soc table
- Fix incorrect fw_state address in dmub_srv
- Use HW lock mgr for PSR1 when only one eDP
- Revert "Support for reg inbox0 for host->DMUB CMDs"
- Change notification of link BW allocation
- Fix message for support_edp0_on_dp1
- Guard against setting dispclk low for dcn31x
- Prevent VStartup Overflow
- Check pipe->stream before passing it to a function

Acked-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:47 -04:00
Charlene Liu
15b959534a drm/amd/display: Use DPM table clk setting for dml2 soc dscclk
[WHY]
Not like dppclk/dispclk, dml2 will calculate the minimum required clocks.
For dscclk, it is used for pure comparision.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:47 -04:00
Charlene Liu
20c13ca5ba drm/amd/display: Update static soc table
[WHY]
Update the static soc table dcn3_5_soc.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:47 -04:00
Lo-an Chen
f57b38ac85 drm/amd/display: Fix incorrect fw_state address in dmub_srv
[WHY]
The fw_state in dmub_srv was assigned with wrong address.
The address was pointed to the firmware region.

[HOW]
Fix the firmware state by using DMUB_DEBUG_FW_STATE_OFFSET
in dmub_cmd.h.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Lo-an Chen <lo-an.chen@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:47 -04:00
Mario Limonciello
ed569e1279 drm/amd/display: Use HW lock mgr for PSR1 when only one eDP
[WHY]
DMUB locking is important to make sure that registers aren't accessed
while in PSR.  Previously it was enabled but caused a deadlock in
situations with multiple eDP panels.

[HOW]
Detect if multiple eDP panels are in use to decide whether to use
lock. Refactor the function so that the first check is for PSR-SU
and then replay is in use to prevent having to look up number
of eDP panels for those configurations.

Fixes: f245b400a2 ("Revert "drm/amd/display: Use HW lock mgr for PSR1"")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3965
Reviewed-by: ChiaHsuan Chung <chiahsuan.chung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:46 -04:00
Dillon Varone
b3d58262dc drm/amd/display: Revert "Support for reg inbox0 for host->DMUB CMDs"
This reverts commit 15d1c2e6bf.

Reason: Cursor movement causes system to hang.

Reviewed-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:46 -04:00
Cruise Hung
52af17eabb drm/amd/display: Change notification of link BW allocation
[WHY & HOW]
The response of DP BW allocation is handled in Outbox ISR.
When it failed to request the DP BW allocation, it sent another
DPCD request in Outbox ISR immediately. The DP AUX reply also
uses the Outbox ISR. So, no AUX reply happened in this case.
Change to use HPD IRQ for the notification.

Reviewed-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
Signed-off-by: Cruise Hung <Cruise.Hung@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:46 -04:00
Yilin Chen
79538e6365 drm/amd/display: Fix message for support_edp0_on_dp1
[WHY]
The info message was wrong when support_edp0_on_dp1 is enabled

[HOW]
Use correct info message for support_edp0_on_dp1

Fixes: f6d17270d1 ("drm/amd/display: add a quirk to enable eDP0 on DP1")
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Yilin Chen <Yilin.Chen@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-03-18 14:03:46 -04:00