Philip Yang [Thu, 4 Dec 2025 17:13:05 +0000 (12:13 -0500)]
drm/amdgpu: Fix gfx9 update PTE mtype flag
Fix copy&paste error, that should have been an assignment instead of an or,
otherwise MTYPE_UC 0x3 can not be updated to MTYPE_RW 0x1.
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>
Lijo Lazar [Tue, 2 Dec 2025 10:00:21 +0000 (15:30 +0530)]
drm/amd/pm: Use driver table for board temperature
GPU board and Baseboard temperatures come from system metrics table.
Driver keeps separate metrics table for both. Use the new driver table
structure to represent them.
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>
Tom St Denis [Tue, 2 Dec 2025 15:05:51 +0000 (10:05 -0500)]
drm/amd/amdgpu: Port over some missing registers and bits from GC 10.1 to 10.3 (v2)
v2: Added SPI bits to sh_mask header
Signed-off-by: Tom St Denis <tom.stdenis@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lijo Lazar [Tue, 2 Dec 2025 07:07:32 +0000 (12:37 +0530)]
drm/amd/pm: Add smu driver table structure
For interfaces like gpu metrics, driver returns a formatted structure
based on IP version. Add a separate data structure for such tables which
also tracks the cache intervals.
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>
Yang Wang [Thu, 11 Dec 2025 04:46:52 +0000 (12:46 +0800)]
drm/amd/pm: fix pp_dpm_pcie wrong state issue for smu v13.0.0
put wrong value into incorrect data into following function,
which caused it to fail to match the correct item on smu v13.0.0:
smu_cmn_print_pcie_levels()
Fixes: a95f01edd80b ("drm/amd/pm: Use common helper for smuv13.0.0 dpm") Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Yang Wang [Thu, 11 Dec 2025 04:49:35 +0000 (12:49 +0800)]
drm/amd/pm: fix pp_dpm_pcie wrong state issue for smu v13.0.7
put wrong value into incorrect data into following function,
which caused it to fail to match the correct item on smu v13.0.7:
smu_cmn_print_pcie_levels()
Fixes: b2debbbb60f1 ("drm/amd/pm: Use common helper for smuv13.0.7 dpm") Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Yang Wang [Wed, 10 Dec 2025 12:33:43 +0000 (20:33 +0800)]
drm/amd/pm: fix pp_dpm_pcie wrong state issue for smu v14.0.2
put wrong value into incorrect data into following function,
which caused it to fail to match the correct item on smu v14.0.2:
smu_cmn_print_pcie_levels()
Fixes: 03d11f8564ca ("drm/amd/pm: Use common helper for smuv14.0.2 dpm") Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Julia Lawall [Tue, 30 Dec 2025 16:17:17 +0000 (17:17 +0100)]
drm/amdkfd: update outdated comment
The function acquire_packet_buffer() was renamed
kq_acquire_packet_buffer() by commit a5a4d68c9326 ("drm/amdkfd:
Eliminate unnecessary kernel queue function pointers"). Update
the comment accordingly.
Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: Felix Kuehling <felix.kuehling@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Julia Lawall [Tue, 30 Dec 2025 17:53:53 +0000 (18:53 +0100)]
drm/amdgpu: update outdated comment
The function amdgpu_amdkfd_gpuvm_import_dmabuf() was split into
import_obj_create() and amdgpu_amdkfd_gpuvm_import_dmabuf_fd() in
commit 0188006d7c79 ("drm/amdkfd: Import DMABufs for interop
through DRM"). import_obj_create() now does the allocation for
the mem variable discussed in the comment.
Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr> Signed-off-by: Felix Kuehling <felix.kuehling@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Perry Yuan [Thu, 25 Dec 2025 08:43:49 +0000 (16:43 +0800)]
drm/amd/pm: Disable MMIO access during SMU Mode 1 reset
During Mode 1 reset, the ASIC undergoes a reset cycle and becomes
temporarily inaccessible via PCIe. Any attempt to access MMIO registers
during this window (e.g., from interrupt handlers or other driver threads)
can result in uncompleted PCIe transactions, leading to NMI panics or
system hangs.
To prevent this, set the `no_hw_access` flag to true immediately after
triggering the reset. This signals other driver components to skip
register accesses while the device is offline.
A memory barrier `smp_mb()` is added to ensure the flag update is
globally visible to all cores before the driver enters the sleep/wait
state.
Signed-off-by: Perry Yuan <perry.yuan@amd.com> Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: Refactor amdgpu_gem_va_ioctl for Handling Last Fence Update and Timeline Management v4
This commit simplifies the amdgpu_gem_va_ioctl function, key updates
include:
- Moved the logic for managing the last update fence directly into
amdgpu_gem_va_update_vm.
- Introduced checks for the timeline point to enable conditional
replacement or addition of fences.
v2: Addressed review comments from Christian.
v3: Updated comments (Christian).
v4: The previous version selected the fence too early and did not manage its
reference correctly, which could lead to stale or freed fences being used.
This resulted in refcount underflows and could crash when updating GPU
timelines.
The fence is now chosen only after the VA mapping work is completed, and its
reference is taken safely. After exporting it to the VM timeline syncobj, the
driver always drops its local fence reference, ensuring balanced refcounting
and avoiding use-after-free on dma_fence.
Link: https://patchwork.freedesktop.org/patch/654669/ Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Suggested-by: Christian König <christian.koenig@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Sat, 6 Dec 2025 02:31:06 +0000 (03:31 +0100)]
drm/amd/display: Don't repeat DAC load detection
The analog link detection code path had already performed the
DAC load detection by the time the EDID read is attempted.
So there is no need to repeat the DAC load detection,
we can know that no display is connected if no EDID is read.
Fixes: ac1bb4952267 ("drm/amd/display: Use DAC load detection on analog connectors (v2)") Suggested-by: Alex Hung <alex.hung@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Sat, 6 Dec 2025 02:31:04 +0000 (03:31 +0100)]
drm/amd/display: Add missing encoder setup to DACnEncoderControl
Apparently the DAC encoder needs to be set up before use.
The BIOS parser in DC did not support this so I assumed it was
not necessary, but the DAC doesn't work without it on some GPUs.
Fixes: 69b29b894660 ("drm/amd/display: Hook up DAC to bios_parser_encoder_control") Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Sat, 6 Dec 2025 02:31:03 +0000 (03:31 +0100)]
drm/amd/display: Correct color depth for SelectCRTC_Source
Pass the correct enum values as expected by the VBIOS.
Previously the actual bit depth integer value was passed,
which was a mistake.
Fixes: 7fb4f254c8eb ("drm/amd/display: Add SelectCRTC_Source to BIOS parser") Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Timur Kristóf [Sat, 6 Dec 2025 02:31:02 +0000 (03:31 +0100)]
drm/amd/display: Pass proper DAC encoder ID to VBIOS
Similarly to the analog_engine field, add a new analog_id field
which contains the encoder ID of the analog encoder that
corresponds to the link encoder.
Previously, the default encoder ID of the link encoder was used,
which meant that we passed the wrong ID in case of DVI-I.
Fixes: 5834c33fd3f6 ("drm/amd/display: Add concept of analog encoders (v2)") Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Wed, 20 Aug 2025 14:07:38 +0000 (10:07 -0400)]
drm/amdgpu/gfx9: Implement KGQ ring reset
GFX ring resets work differently on pre-GFX10 hardware since
there is no MQD managed by the scheduler.
For ring reset, you need issue the reset via CP_VMID_RESET
via KIQ or MMIO and submit the following to the gfx ring to
complete the reset:
1. EOP packet with EXEC bit set
2. WAIT_REG_MEM to wait for the fence
3. Clear CP_VMID_RESET to 0
4. EVENT_WRITE ENABLE_LEGACY_PIPELINE
5. EOP packet with EXEC bit set
6. WAIT_REG_MEM to wait for the fence
Once those commands have completed the reset should
be complete and the ring can accept new packets.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Tested-by: Jiqian Chen <Jiqian.Chen@amd.com> (v1) Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Replace WAIT_REG_MEM with EVENT_WRITE flushes for all
shader types and ACQUIRE_MEM. That should accomplish
the same thing and avoid having to wait on a fence
preventing any issues with pipeline syncs during
queue resets.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Fri, 12 Dec 2025 16:46:48 +0000 (11:46 -0500)]
drm/amdgpu: avoid a warning in timedout job handler
Only set an error on the fence if the fence is not
signalled. We can end up with a warning if the
per queue reset path signals the fence and sets an error
as part of the reset, but fails to recover.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Pratap Nirujogi [Wed, 10 Dec 2025 01:22:15 +0000 (20:22 -0500)]
drm/amd/amdgpu: Fix SMU warning during isp suspend-resume
ISP mfd child devices are using genpd and the system suspend-resume
operations between genpd and amdgpu parent device which uses only
runtime suspend-resume are not in sync.
Linux power manager during suspend-resume resuming the genpd devices
earlier than the amdgpu parent device. This is resulting in the below
warning as SMU is in suspended state when genpd attempts to resume ISP.
WARNING: CPU: 13 PID: 5435 at drivers/gpu/drm/amd/amdgpu/../pm/swsmu/amdgpu_smu.c:398 smu_dpm_set_power_gate+0x36f/0x380 [amdgpu]
To fix this warning isp suspend-resume is handled as part of amdgpu
parent device suspend-resume instead of genpd sequence. Each ISP MFD
child device is marked as dev_pm_syscore_device to skip genpd
suspend-resume and use pm_runtime_force api's to suspend-resume
the devices when callbacks from amdgpu are received.
Co-developed-by: Gjorgji Rosikopulos <grosikop@amd.com> Signed-off-by: Gjorgji Rosikopulos <grosikop@amd.com> Signed-off-by: Bin Du <bin.du@amd.com> Signed-off-by: Pratap Nirujogi <pratap.nirujogi@amd.com> Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Taimur Hassan [Sat, 13 Dec 2025 02:32:55 +0000 (21:32 -0500)]
drm/amd/display: Promote DC to 3.2.364
This version brings along the following updates:
- Add frame skip feature support flag.
- Add sink EDID data null check.
- Update function name to link_detect_connection_type_analog.
- Fix mismatched unlock for DMUB HW lock in HWSS fast path.
- Fix P010, NV12, YUY2 scale down by four times failure.
- Fix and reenable UPDATE_V3_FLOW_NEW_CONTEXT_MINIMAL.
- Consolidate dmub fb info to a single struct.
- Add new fields to fams2 config.
- Update timing source enums.
- Add signal type check for dcn401 get_phyd32clk_src.
- Fix dsc eDP issue.
- Remove unnecessary divider update flag.
- Update dc_connection_dac_load to dc_connection_analog_load.
- Check NULL before calling dac_load_detection.
- Replace log macro for analog display detection.
Reviewed-by: Rodrigo Siqueira <rodrigo.siqueira@amd.com> Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Acked-by: Wayne Lin <Wayne.Lin@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Hung [Mon, 8 Dec 2025 19:18:02 +0000 (12:18 -0700)]
drm/amd/display: Replace log macro for analog display detection
link detection should use LINK_INFO() macro.
Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Hung [Mon, 8 Dec 2025 19:11:43 +0000 (12:11 -0700)]
drm/amd/display: Check NULL before calling dac_load_detection
dac_load_detection can be NULL in some scenario, so checking it before
calling.
Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Hung [Mon, 8 Dec 2025 19:09:08 +0000 (12:09 -0700)]
drm/amd/display: Update dc_connection_dac_load to dc_connection_analog_load
Update to a more accurate name dc_connection_analog_load.
Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cruise Hung [Fri, 5 Dec 2025 13:22:44 +0000 (21:22 +0800)]
drm/amd/display: Remove unnecessary divider update flag
[Why]
When transitioning from 640x480 at RBRx1 to HBR3x1,
both output pixel mode and pixel rate divider should update.
The needs_divider_update flag was only for 8b10b and 128b132b transition.
[How]
Remove needs_divider_update flag.
Reviewed-by: Michael Strauss <michael.strauss@amd.com> Reviewed-by: Wenjing Liu <wenjing.liu@amd.com> Signed-off-by: Cruise Hung <Cruise.Hung@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/display: Fix and reenable UPDATE_V3_FLOW_NEW_CONTEXT_MINIMAL
[Why]
Reenable new split implementation, previously partially reverted due
to issues with ODM on high-bandwidth displays 4k144Hz, resulting
in a corrupted gray screen.
Minimal flows require two separate commits, with extra intermediate
commit to enable seamless transitions, each followed by a swap. Since
new design requires commit to be run in execute and swap in cleanup
stage, an attempt was made to reorder them from CSCS (Commit-Swap-Commit-Swap)
to CCSS (Commit-Commit-Swap-Swap). Not only is this not viable, but
was implemented incorrectly as CCS, one swap missing.
[How]
* Change UPDATE_V3_FLOW_NEW_CONTEXT_MINIMAL_NEW/CURRENT to execute
and cleanup one commit, then run UPDATE_V3_FLOW_NEW_CONTEXT_SEAMLESS,
which closely matches old implementation where minimal flows fall back
to seamless.
* Fix uninitialized variable error.
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Dominik Kaszewski <dominik.kaszewski@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Kaier Hsueg [Thu, 4 Dec 2025 16:33:59 +0000 (00:33 +0800)]
drm/amd/display: Fix P010, NV12, YUY2 scale down by four times failure
[WHY]
When performing 4:1 downscaling with subsampled formats,
the SPL remainder distribution logic (+1) overrides the
upper layer’s aligned width, resulting in odd segment
widths and causing hang.
The upper layer alignment ensures the width is sufficient
and even, so SPL should not modify it further.
[HOW]
In dc_spl.c within calculate_mpc_slice_in_timing_active,
add an extra condition: Skip the remainder distribution
(+1) when use_recout_width_aligned is true.This change
respects the upper layer’s alignment decision, prevents
odd widths, and is a minimal, safe fix.
Reviewed-by: Joshua Aberback <joshua.aberback@amd.com> Signed-off-by: Kaier Hsueh <Kaier.Hsueh@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd/display: Fix mismatched unlock for DMUB HW lock in HWSS fast path
[Why]
The evaluation for whether we need to use the DMUB HW lock isn't the
same as whether we need to unlock which results in a hang when the
fast path is used for ASIC without FAMS support.
[How]
Store a flag that indicates whether we should use the lock and use
that same flag to specify whether unlocking is needed.
Reviewed-by: Swapnil Patel <swapnil.patel@amd.com> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Hung [Tue, 2 Dec 2025 18:54:56 +0000 (11:54 -0700)]
drm/amd/display: Update function name to link_detect_connection_type_analog
[WHAT]
Update function "link_detect_analog" to a more accurate name
"link_detect_connection_type_analog".
Suggested-by: Wenjing Liu <wenjing.liu@amd.com> Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
ChunTao Tso [Mon, 8 Dec 2025 09:36:57 +0000 (17:36 +0800)]
drm/amd/display: Add frame skip feature support flag
[WHY]
The set_replay_frame_skip_number() function should not execute when
the link does not support the Frame Skipping feature.
[HOW]
Add a new field `frame_skip_supported` to struct replay_config to
indicate whether the link supports frame skipping. Check this flag
at the beginning of set_replay_frame_skip_number() and return early
if the feature is not supported.
Reviewed-by: Robin Chen <robin.chen@amd.com> Signed-off-by: ChunTao Tso <chuntao.tso@amd.com> Signed-off-by: Chenyu Chen <chen-yu.chen@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Yang Wang [Mon, 15 Dec 2025 09:51:11 +0000 (17:51 +0800)]
drm/amd/pm: force send pcie parmater on navi1x
v1:
the PMFW didn't initialize the PCIe DPM parameters
and requires the KMD to actively provide these parameters.
v2:
clean & remove unused code logic (lijo)
Fixes: 1a18607c07bb ("drm/amd/pm: override pcie dpm parameters only if it is necessary") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4671 Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add documentation for the uma/carveout_options and uma/carveout
attributes in sysfs
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Yo-Jung Leo Lin (AMD) <Leo.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: add UMA allocation interfaces to sysfs
Add a uma/ directory containing two sysfs files as interfaces to
inspect or change UMA carveout size. These files are:
- uma/carveout_options: a read-only file listing all the available
UMA allocation options and their index.
- uma/carveout: a file that is both readable and writable. On read,
it shows the index of the current setting. Writing a valid index
into this file allows users to change the UMA carveout size to that
option on the next boot.
Co-developed-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Yo-Jung Leo Lin (AMD) <Leo.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
On some platforms, UMA allocation size can be set using the ATCS
methods. Add helper functions to interact with this functionality.
Co-developed-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Yo-Jung Leo Lin (AMD) <Leo.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Currently, the available UMA allocation configs in the integrated system
information table have not been parsed. Add a helper function to retrieve
and store these configs.
Co-developed-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Yo-Jung Leo Lin (AMD) <Leo.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: parse UMA size-getting/setting bits in ATCS mask
The capabilities of getting and setting VRAM carveout size are exposed
in the ATCS mask. Parse and store these capabilities for future use.
Co-developed-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Yo-Jung Leo Lin (AMD) <Leo.Lin@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 13 Nov 2025 19:12:10 +0000 (14:12 -0500)]
drm/amdgpu: always backup and reemit fences
If when we backup the ring contents for reemit before a
ring reset, we skip jobs associated with the bad
context, however, we need to make sure the fences
are reemited as unprocessed submissions may depend on
them.
v2: clean up fence handling, make helpers static
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Alex Deucher [Thu, 13 Nov 2025 18:24:10 +0000 (13:24 -0500)]
drm/amdgpu: don't reemit ring contents more than once
If we cancel a bad job and reemit the ring contents, and
we get another timeout, cancel everything rather than reemitting.
The wptr markers are only relevant for the original emit. If
we reemit, the wptr markers are no longer correct.
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: Use explicit VCN instance 0 in SR-IOV init
vcn_v2_0_start_sriov() declares a local variable "i" initialized to zero
and uses it only as the instance index in SOC15_REG_OFFSET(UVD, i, ...).
The value is never changed and all other fields are taken from
adev->vcn.inst[0], so this path only ever programs VCN instance 0.
This triggered a Smatch:
warn: iterator 'i' not incremented
Replace the dummy iterator with an explicit instance index of 0 in
SOC15_REG_OFFSET() calls.
Fixes: dd26858a9cd8 ("drm/amdgpu: implement initialization part on VCN2.0 for SRIOV")
Reported by: Dan Carpenter <dan.carpenter@linaro.org> Cc: darlington Opara <darlington.opara@amd.com> Cc: Jinage Zhao <jiange.zhao@amd.com> Cc: Monk Liu <Monk.Liu@amd.com> Cc: Emily Deng <Emily.Deng@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Emily Deng <Emily.Deng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com> Reviewed-by: Lancelot Six <lancelot.six@amd.com> Cc: Joseph Greathouse <joseph.greathouse@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jay Cornwall [Thu, 23 Oct 2025 20:28:39 +0000 (15:28 -0500)]
drm/amdkfd: Fix VGPR bank state save in gfx12.1 trap handler
S_SETREG_IMM32_B32 does not apply a mask to the MODE bank bits.
SRC2 is consequently unconditonally cleared during context save.
Use S_SETREG_B32 instead to preserve SRC2.
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com> Reviewed-by: Lancelot Six <lancelot.six@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
David Yat Sin [Tue, 18 Mar 2025 19:49:55 +0000 (19:49 +0000)]
drm/amdkfd: Add metadata ring buffer for compute
Add support for separate ring-buffer for metadata packets when using
compute queues. Userspace application allocate the metadata ring-buffer
and the queue ring-buffer with a single allocation. The metadata
ring-buffer starts after the queue ring-buffer.
Signed-off-by: David Yat Sin <David.YatSin@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: Update TCP Control register on GFX 12.1
Update TCP CNTL register to disable some features not supported
on GFX 12.1.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Alex Sierra <alex.sierra@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: Add back CWSR trap handler for GFX 12.1
CWSR Trap handler for GFX 12.1 was missed when merging changes
from 6.14 NPI branch to 6.16 NPI branch. This change adds back
the CWSR trap handler for GFX 12.1.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Alex Sierra <alex.sierra@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
After the 6.16 merge, some changes not applicable to GFX 12.1 were
added in the gmc_v12_1_get_vm_pte function. Additionally, add the
case for MTYPE RW for GFX 12.1.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Alex Sierra <alex.sierra@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: Disable TCP Early Write Ack for GFX 12.1
Disable the TCP Early Write Ack feature on GFX 12.1.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Alex Sierra <alex.sierra@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Jonathan Kim [Tue, 9 Sep 2025 19:57:44 +0000 (15:57 -0400)]
drm/amdkfd: enable precise memory operations for gfx1250
Enable precise memory for GFX 1250.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
GFX 12 devices that support spatial partitioning should use the WREG32
per XCC macro when updating address watch settings, similar to GFX 9
devices that support spatial partitioning.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add CU masking implementation for GFX 12.1. Add a local
implementation for GFX 12.1 instead of using the generic
function defined in kfd_mqd_manager.c because of some
quirks in the way CU mask is handled on GFX 12.1.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Alex Sierra <alex.sierra@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Mukul Joshi [Thu, 14 Aug 2025 19:23:16 +0000 (15:23 -0400)]
drm/amdgpu: Always set PTE.B for device memory on GFX 12.1
On GFX 12.1, we need to set the atomics bit (PTE.B) always for
device memory.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Alex Sierra <alex.sierra@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Lang Yu [Tue, 19 Aug 2025 10:54:30 +0000 (18:54 +0800)]
drm/amdgpu/gfx12.1: Don't fetch default register values from hardware in mqd init
1. We can't assure the fetched values are always default register values.
Observing non-zero cp_hqd_pq_rptr in mes_v12_1_self_test->init_mqd()
where no GRBM_GFX_CNTL is specified.
2. See commit fc3c139cf043 ("drm/amdgpu/gfx12: don't read registers in mqd init").
Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amd: Pass `adev` to amdgpu_gfx_parse_disable_cu()
In order for messages to be attribute to the correct device
amdgpu_gfx_parse_disable_cu() needs to know what device is being
operated on. Pass the argument in.
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
At bootup on a Strix machine the following message comes up:
```
amdgpu: Topology: Add dGPU node [0x150e:0x1002]
```
This is an APU though. Clarify the messaging by only offer a
"CPU node" or "GPU node" message. Also set the message as
VID:DID instead which is how other messages work.
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdkfd: Fix signal_eviction_fence() bool return value
signal_eviction_fence() is declared to return bool, but returns -EINVAL
when no eviction fence is present. This makes the "no fence" or "the
NULL-fence" path evaluate to true and triggers a Smatch warning.
v2: Return true instead to explicitly indicate that there is no eviction
fence to signal and that eviction is already complete. This matches the
existing caller logic where a NULL fence means "nothing to do" and
allows restore handling to proceed normally. (Christian)
Fixes the below:
drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_process.c:2099 signal_eviction_fence()
warn: '(-22)' is not bool
Fixes: 37865e02e6cc ("drm/amdkfd: Fix eviction fence handling")
Reported by: Dan Carpenter <dan.carpenter@linaro.org> Cc: Philip Yang <Philip.Yang@amd.com> Cc: Gang BA <Gang.Ba@amd.com> Cc: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Yang Wang [Thu, 11 Dec 2025 02:47:18 +0000 (10:47 +0800)]
drm/amd/pm: fix wrong pcie parameter on navi1x
fix wrong pcie dpm parameter on navi1x
Fixes: 1a18607c07bb ("drm/amd/pm: override pcie dpm parameters only if it is necessary") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4671 Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Co-developed-by: Kenneth Feng <kenneth.feng@amd.com> Signed-off-by: Kenneth Feng <kenneth.feng@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
projects / thirdparty / kernel / linux.git / log
