Unlike later versions, UVD 3 has firmware validation.
For this to work, the UVD should be powered up correctly.
When DPM is enabled and the display clock is off,
the SMU may choose a power state which doesn't power
the UVD, which can result in failure to initialize UVD.
v2:
Add code comments to explain about the UVD power state
and how UVD clock is turned on/off.
Fixes: b38f3e80ec ("drm amdgpu: SI UVD v3_1 (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>
The kfd CRIU checkpoint ioctl would return an error if trying
to checkpoint a process with no kfd buffer objects.
This is a normal case and should not be an error.
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: David Francis <David.Francis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add new GEM_OP_IOCTL option GET_MAPPING_INFO, which
returns a list of mappings associated with a given bo, along with
their positions and offsets.
Userspace for this and the previous change can be found at:
https://github.com/checkpoint-restore/criu/pull/2613
Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add new ioctl DRM_IOCTL_AMDGPU_GEM_LIST_HANDLES.
This ioctl returns a list of bos with their handles, sizes,
and flags and domains.
This ioctl is meant to be used during CRIU checkpoint and
provide information needed to reconstruct the bos
in CRIU restore.
Userspace for this and the next change can be found at
https://github.com/checkpoint-restore/criu/pull/2613
Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The GEM create flag AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE
specifies that gem memory contains sensitive information and
should be cleared to prevent snooping.
The COHERENT and UNCACHED gem create flags enable memory
features related to sharing memory across devices.
For CRIU we need to re-create KFD BOs through the
GEM_CREATE IOCTL, so allow those KFD specific flags here as well.
This will also aid us in the future and allows to move
the KFD components over using the render node for allocations.
Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
- Add madvise interface (Himal Prasad Ghimiray)
- Add DRM_IOCTL_XE_VM_QUERY_MEMORY_RANGE_ATTRS to query VMA count and
memory attributes (Himal Prasad Ghimiray)
- Handle Firmware reported Hardware Errors notifying userspace with
device wedged uevent (Riana Tauro)
Cross-subsystem Changes:
- Add a vendor-specific recovery method to drm device wedged uevent
(Riana Tauro)
Driver Changes:
- Use same directory structure in debugfs as in sysfs (Michal Wajdeczko)
- Cleanup and future-proof VRAM region initialization (Piotr Piórkowski)
- Add G-states and PCIe link states to debugfs (Soham Purkait)
- Cleanup eustall debug messages (Harish Chegondi)
- Add SR-IOV support to restore Compression Control Surface (CCS) to
Xe2 and later (Satyanarayana K V P)
- Enable SR-IOV PF mode by default on supported platforms without
needing CONFIG_DRM_XE_DEBUG and mark some platforms behind
force_probe as supported (Michal Wajdeczko)
- More targeted log messages (Michal Wajdeczko)
- Cleanup STEER_SEMAPHORE/MCFG_MCR_SELECTOR usage (Nitin Gote)
- Use common code to emit flush (Tvrtko Ursulin)
- Add/extend more HW workarounds and tunings for Xe2 and Xe3
(Sk Anirban, Tangudu Tilak Tirumalesh, Nitin Gote, Chaitanya Kumar Borah)
- Add a generic dependency scheduler to help with TLB invalidations
and future scenarios (Matthew Brost)
- Use DRM scheduler for delayed GT TLB invalidations (Matthew Brost)
- Error out on incorrect device use in configfs
(Michal Wajdeczko, Lucas De Marchi)
- Refactor configfs attributes (Michal Wajdeczko / Lucas De Marchi)
- Allow configuring future VF devices via configfs (Michal Wajdeczko)
- Implement some missing XeLP workarounds (Tvrtko Ursulin)
- Generalize WA BB setup/emission and add support for
mid context restore BB, aka indirect context (Tvrtko Ursulin)
- Prepare the driver to expose mmio regions to userspace
in future (Ilia Levi)
- Add more GuC load error status codes (John Harrison)
- Document DRM_XE_GEM_CREATE_FLAG_DEFER_BACKING (Priyanka Dandamudi)
- Disable CSC and RPM on VFs (Lukasz Laguna, Satyanarayana K V P)
- Fix oops in xe_gem_fault with PREEMPT_RT (Maarten Lankhorst)
- Skip LMTT update if no LMEM was provisioned (Michal Wajdeczko)
- Add support to VF migration (Tomasz Lis)
- Use a helper for guc_waklv_enable functions (Jonathan Cavitt)
- Prepare GPU SVM for migration of THP (Francois Dugast)
- Program LMTT directory pointer on all GTs within a tile
(Piotr Piórkowski)
- Rename XE_WA to XE_GT_WA to better convey its scope vs the device WAs
(Matt Atwood)
- Allow to match devices on PCI devid/vendorid only (Lucas De Marchi)
- Improve PDE PAT index selection (Matthew Brost)
- Consolidate ASID allocation in xe_vm_create() vs
xe_vm_create_ioctl() (Piotr Piórkowski)
- Resize VF BARS to max possible size according to number of VFs
(Michał Winiarski)
- Untangle vm_bind_ioctl cleanup order (Christoph Manszewski)
- Start fixing usage of XE_PAGE_SIZE vs PAGE_SIZE to improve
compatibility with non-x86 arch (Simon Richter)
- Improve tile vs gt initialization order and accounting
(Gustavo Sousa)
- Extend WA kunit test to PTL
- Ensure data is initialized before transferring to pcode
(Stuart Summers)
- Add PSMI support for HW validation (Lucas De Marchi,
Vinay Belgaumkar, Badal Nilawar)
- Improve xe_dma_buf test (Thomas Hellström, Marcin Bernatowicz)
- Fix basename() usage in generator with !glibc (Carlos Llamas)
- Ensure GT is in C0 during resumes (Xin Wang)
- Add TLB invalidation abstraction (Matt Brost, Stuart Summers)
- Make MI_TLB_INVALIDATE conditional on migrate (Matthew Auld)
- Prepare xe_nvm to be initialized early for future use cases
(Riana Tauro)
-----BEGIN PGP SIGNATURE-----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=gqoN
-----END PGP SIGNATURE-----
Merge tag 'drm-xe-next-2025-08-29' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-next
UAPI Changes:
- Add madvise interface (Himal Prasad Ghimiray)
- Add DRM_IOCTL_XE_VM_QUERY_MEMORY_RANGE_ATTRS to query VMA count and
memory attributes (Himal Prasad Ghimiray)
- Handle Firmware reported Hardware Errors notifying userspace with
device wedged uevent (Riana Tauro)
Cross-subsystem Changes:
- Add a vendor-specific recovery method to drm device wedged uevent
(Riana Tauro)
Driver Changes:
- Use same directory structure in debugfs as in sysfs (Michal Wajdeczko)
- Cleanup and future-proof VRAM region initialization (Piotr Piórkowski)
- Add G-states and PCIe link states to debugfs (Soham Purkait)
- Cleanup eustall debug messages (Harish Chegondi)
- Add SR-IOV support to restore Compression Control Surface (CCS) to
Xe2 and later (Satyanarayana K V P)
- Enable SR-IOV PF mode by default on supported platforms without
needing CONFIG_DRM_XE_DEBUG and mark some platforms behind
force_probe as supported (Michal Wajdeczko)
- More targeted log messages (Michal Wajdeczko)
- Cleanup STEER_SEMAPHORE/MCFG_MCR_SELECTOR usage (Nitin Gote)
- Use common code to emit flush (Tvrtko Ursulin)
- Add/extend more HW workarounds and tunings for Xe2 and Xe3
(Sk Anirban, Tangudu Tilak Tirumalesh, Nitin Gote, Chaitanya Kumar Borah)
- Add a generic dependency scheduler to help with TLB invalidations
and future scenarios (Matthew Brost)
- Use DRM scheduler for delayed GT TLB invalidations (Matthew Brost)
- Error out on incorrect device use in configfs
(Michal Wajdeczko, Lucas De Marchi)
- Refactor configfs attributes (Michal Wajdeczko / Lucas De Marchi)
- Allow configuring future VF devices via configfs (Michal Wajdeczko)
- Implement some missing XeLP workarounds (Tvrtko Ursulin)
- Generalize WA BB setup/emission and add support for
mid context restore BB, aka indirect context (Tvrtko Ursulin)
- Prepare the driver to expose mmio regions to userspace
in future (Ilia Levi)
- Add more GuC load error status codes (John Harrison)
- Document DRM_XE_GEM_CREATE_FLAG_DEFER_BACKING (Priyanka Dandamudi)
- Disable CSC and RPM on VFs (Lukasz Laguna, Satyanarayana K V P)
- Fix oops in xe_gem_fault with PREEMPT_RT (Maarten Lankhorst)
- Skip LMTT update if no LMEM was provisioned (Michal Wajdeczko)
- Add support to VF migration (Tomasz Lis)
- Use a helper for guc_waklv_enable functions (Jonathan Cavitt)
- Prepare GPU SVM for migration of THP (Francois Dugast)
- Program LMTT directory pointer on all GTs within a tile
(Piotr Piórkowski)
- Rename XE_WA to XE_GT_WA to better convey its scope vs the device WAs
(Matt Atwood)
- Allow to match devices on PCI devid/vendorid only (Lucas De Marchi)
- Improve PDE PAT index selection (Matthew Brost)
- Consolidate ASID allocation in xe_vm_create() vs
xe_vm_create_ioctl() (Piotr Piórkowski)
- Resize VF BARS to max possible size according to number of VFs
(Michał Winiarski)
- Untangle vm_bind_ioctl cleanup order (Christoph Manszewski)
- Start fixing usage of XE_PAGE_SIZE vs PAGE_SIZE to improve
compatibility with non-x86 arch (Simon Richter)
- Improve tile vs gt initialization order and accounting
(Gustavo Sousa)
- Extend WA kunit test to PTL
- Ensure data is initialized before transferring to pcode
(Stuart Summers)
- Add PSMI support for HW validation (Lucas De Marchi,
Vinay Belgaumkar, Badal Nilawar)
- Improve xe_dma_buf test (Thomas Hellström, Marcin Bernatowicz)
- Fix basename() usage in generator with !glibc (Carlos Llamas)
- Ensure GT is in C0 during resumes (Xin Wang)
- Add TLB invalidation abstraction (Matt Brost, Stuart Summers)
- Make MI_TLB_INVALIDATE conditional on migrate (Matthew Auld)
- Prepare xe_nvm to be initialized early for future use cases
(Riana Tauro)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/nuejxdhnalyok7tzwkrj67dwjgdafwp4mhdejpyyqnrh4f2epq@nlldovuflnbx
Update the legacy (non-DC) display code to respect the maximum
pixel clock for HDMI and DVI-D. Reject modes that would require
a higher pixel clock than can be supported.
Also update the maximum supported HDMI clock value depending on
the ASIC type.
For reference, see the DC code:
check max_hdmi_pixel_clock in dce*_resource.c
v2:
Fix maximum clocks for DVI-D and DVI/HDMI adapters.
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>
[why]
Read link setting inside mode validation is not always the final downlink setting.
It is found true in Synaptics branch device.
At bootup, the preferred mode being set right after 1080p is set. It occurred
before graphic load. That modeset switch in a short period of time makes
the branch device switch back and forth from lower and higher link rate,
observed at Synaptics branch device.
DP2 RTK hub on the other hand, sticks to highest available downlink rate after bootup.
Existing check of runtime downlink setting in mode validation shows asynchronous at
branch device link switch, i.e., downlink switch to higher link rate not yet complete
when the mode validation tries to probe the downlink setting. That makes mode validation
checking downlink setting making wrong decision by pruning modes that should pass the
validation after the downlink setting switch is complete.
[how]
If Synaptics is found at the last branch, skip checking downlink setting
at mode validation.
Reviewed-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why&How]
ON DCN314, clearing DPP SW structure without power gating it can cause a
double cursor in full screen with non-native scaling.
A W/A that clears CURSOR0_CONTROL cursor_enable flag if
dcn10_plane_atomic_power_down is called and DPP power gating is disabled.
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4168
Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why & How]
Previously, when calculating dto phase, we would incorrectly fail when phase
<=0 without additionally checking for the integer value. This meant that
calculations would incorrectly fail when the desired pixel clock was an exact
multiple of the reference clock.
Reviewed-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: Clay King <clayking@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Bounding box values can be stored in multiple locations. (e.g. PMFW, VBIOS, DMUB).
The source and interpretation of these values can vary with DCN revision
so there should be a component that can gather these values and translate
them accordingly
[How]
Have component start with the statically defined values as a base.
Then update them as needed with DCN-specific logic
Guard this component with FPU flags since values need to be in float point.
Reviewed-by: Jun Lei <jun.lei@amd.com>
Signed-off-by: Austin Zheng <Austin.Zheng@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Custom brightness curve works by walking through all data points one
by one. When the brightness value is at either extreme this is a lot
of data points to walk. This is especially noticeable when moving a
brightness slider around how it can lag.
[How]
Bisect the data points to find the closest for interpolation.
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
-Pipe splitting allows for clocks to be reduced, but when using TMDS 420,
reduced clocks lead to missed clocks cycles on clock resyncing
[How]
-Impose a minimum clock when using TMDS 420
Reviewed-by: Chris Park <chris.park@amd.com>
Signed-off-by: Relja Vojvodic <rvojvodi@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We already disable the audio pins in hw_fini so
there is no need to do it again in sw_fini.
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4481
Cc: oushixiong <oushixiong1025@163.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Although compositors will add their own modes, Xorg won't use it's own
modes and will only stick to modes advertised by the driver. This mean a
user that used to pick 1024x768 could no longer access it unless the
panel's native resolution was 1024x768.
[How]
Revert commit 6d396e7ac1 ("drm/amd/display: Disable common modes for
LVDS") and commit 7948afb46a ("drm/amd/display: Disable common modes
for eDP").
The panel will still use scaling for any non-native modes due to
commit 978fa2f6d0 ("drm/amd/display: Use scaling for non-native
resolutions on eDP")
Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4538
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250828140856.2887993-1-superm1@kernel.org
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use __free(kfree) for memory alloc cleanups in SMUv13.0.6
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>
The patch uses power state of VCN instances for requesting video
profile.
In idle worker of a vcn instance, when there is no outstanding
submisssion or fence, the instance is put to power gated state. When
all instances are powered off that means video profile is no longer
required. A request is made to turn off video profile.
A job submission starts with begin_use of ring, and at that time
vcn instance state is changed to power on. Subsequently a check is
made for active video profile, and if not active, a request is made.
Fixes: 3b669df92c ("drm/amdgpu/vcn: adjust workload profile handling")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Sriov guest side doesn't init ras feature hence the poison irq shouldn't
be put during hw fini
Signed-off-by: Mangesh Gadre <Mangesh.Gadre@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Sriov guest side doesn't init ras feature hence the poison irq shouldn't
be put during hw fini
Signed-off-by: Mangesh Gadre <Mangesh.Gadre@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
v1:
1. Unified amdgpu ip block name print with format
"{ip_type}_v{major}_{minor}_{rev}"
2. Avoid IP block name conflicts for SMU/PSP ip block
v2:
Update IP block print format to keep legacy IP block name (Alex)
"{ip_type}_v{major}_{minor}_{rev} ({funcs->name})"
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Using the previous firmware could lead to problems with
PROTECTED_FENCE_SIGNAL commands, specifically causing register
conflicts between MCU_DBG0 and MCU_DBG1.
The updated firmware versions ensure proper alignment
and unification of the SDMA_SUBOP_PROTECTED_FENCE_SIGNAL value with SDMA 7.x,
resolving these hardware coordination issues
Fixes: e8cca30d8b ("drm/amdgpu/sdma6: add ucode version checks for userq support")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Notify pmfw when bad page threshold is exceeded, no matter the module
parameter 'bad_page_threshold' is set or not.
Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
For multiple VCN instances case we get multiple lines of the same
message like below:
amdgpu 0000:43:00.0: amdgpu: Found VCN firmware Version ENC: 1.24 DEC: 9 VEP: 0 Revision: 11
amdgpu 0000:43:00.0: amdgpu: Found VCN firmware Version ENC: 1.24 DEC: 9 VEP: 0 Revision: 11
By adding instance number to the log message for multiple VCN instances,
each line will clearly indicate which VCN instance it refers to.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use max() to reduce the code and improve readability.
No functional changes.
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fix a kernel-doc warning by documenting the num_rmcm_3dluts member of struct mpc_color_caps.
v2: improve comment (Melissa)
Signed-off-by: Kavithesh A.S <kavitheshnitt@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If an object is backed up to shmem it is incorrectly identified
as not having valid data by the move code. This means moving
to VRAM skips the -EMULTIHOP step and the bo is cleared. This
causes all sorts of weird behaviour on DGFX if an already evicted
object is targeted by the shrinker.
Fix this by using ttm_tt_is_swapped() to identify backed-up
objects.
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5996
Fixes: 00c8efc318 ("drm/xe: Add a shrinker for xe bos")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: <stable@vger.kernel.org> # v6.15+
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250828134837.5709-1-thomas.hellstrom@linux.intel.com
Correct kernel-doc formatting issues in the UAPI definitions for
madvise and VMA query interfaces to resolve docutils warnings during
documentation build.
Fixes: 418807860e ("drm/xe/uapi: Add UAPI for querying VMA count and memory attributes")
Fixes: 231bb0ee7a ("drm/xe/uapi: Add madvise interface")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250828071516.3838110-1-himal.prasad.ghimiray@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
To allow initialization of nvm during early probe for future usecases,
use root tile instead of root gt to access mmios, as gt is not
yet initialized at early probe.
v2: fix commit message (Lucas)
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250825103537.2551837-1-riana.tauro@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Small-BAR systems (e.g., SR-IOV VFs in VMs) expose only a subset of
VRAM via PCI/BAR. Exporting a BO outside that window fails, and the
selftests also do CPU fill/verify.
Set XE_BO_FLAG_NEEDS_CPU_ACCESS for cross-device variants to force
CPU-mappable placement and keep tests reliable. Large-BAR/P2P setups
are unaffected.
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250814145950.430231-1-marcin.bernatowicz@linux.intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
When clearing VRAM we should be able to skip invalidating the TLBs if we
are only using the identity map to access VRAM (which is the common
case), since no modifications are made to PTEs on the fly. Also since we
use huge 1G entries within the identity map, there should be a pretty
decent chance that the next packet(s) (if also clears) can avoid a tree
walk if we don't shoot down the TLBs, like if we have to process a long
stream of clears.
For normal moves/copies, we usually always end up with the src or dst
being system memory, meaning we can't only rely on the identity map and
will also need to emit PTEs and so will always require a TLB flush.
v2:
- Update commit to explain the situation for normal copies (Matt B)
- Rebase on latest changes
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250808110452.467513-2-matthew.auld@intel.com
Bitwise AND was intended instead of OR. With the current code the
condition is always true.
Fixes: cbdd16b818 ("HID: i2c-hid: Make elan touch controllers power on after panel is enabled")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Pin-yen Lin <treapking@chromium.org>
Acked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/aK8Au3CgZSTvfEJ6@stanley.mountain
The frontend exposes an API to the driver to send invalidations, handles
sequence number assignment, synchronization (fences), and provides a
timeout mechanism. The backend issues the actual invalidation to the
hardware (or firmware).
The new layering easily allows issuing TLB invalidations to different
hardware or firmware interfaces.
Normalize some naming while here too.
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-10-stuart.summers@intel.com
Break out the GuC specific code into helpers as part of the process to
decouple frontback TLB invalidation code from the backend.
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-9-stuart.summers@intel.com
It is a bit backwards to add a TLB invalidation fence to the pending
list after issuing the invalidation. Perform this step before issuing
the TLB invalidation in a helper function.
v2: Make sure the seqno_lock mutex covers the send as well (Matt)
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-8-stuart.summers@intel.com
Decouple TLB invalidations from the GT by updating the TLB invalidation
layer to accept a `struct xe_tlb_inval` instead of a `struct xe_gt`.
Also, rename *gt_tlb* to *tlb*. The internals of the TLB invalidation
code still operate on a GT, but this is now hidden from the rest of the
driver.
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-7-stuart.summers@intel.com
Extract TLB invalidation state into a structure to decouple TLB
invalidations from the GT, allowing the structure to be embedded
anywhere in the driver.
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-5-stuart.summers@intel.com
tlb_invalidation is a bit verbose leading to ugly wraps in the code,
shorten to tlb_inval.
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-4-stuart.summers@intel.com
Add a new _fini() routine on the GT TLB invalidation
side to handle this worker cleanup on driver teardown.
v2: Move the TLB teardown to the gt fini() routine called during
gt_init rather than in gt_alloc. This way the GT structure stays
alive for while we reset the TLB state.
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-3-stuart.summers@intel.com
Currently the CT lock is used to cover TLB invalidation
sequence number updates. In an effort to separate the GuC
back end tracking of communication with the firmware from
the front end TLB sequence number tracking, add a new lock
here to specifically track those sequence number updates
coming in from the user.
Apart from the CT lock, we also have a pending lock to
cover both pending fences and sequence numbers received
from the back end. Those cover interrupt cases and so
it makes not to overload those with sequence numbers
coming in from new transactions. In that way, we'll employ
a mutex here.
v2: Actually add the correct lock rather than just dropping
it... (Matt)
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-2-stuart.summers@intel.com
Although it's possible to change the attributes in runtime, they have no
effect after the driver is already bound to the device. Check for that
and return -EBUSY in that case.
This should help users understand what's going on when the behavior is
not changing even if the value from the configfs is "right", but it got
to that state too late.
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250826153210.3068808-2-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
If the doorbell is invalid, be sure to set the r to an error
state so the function returns an error.
Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
