- A Revert of a Kconfig patch that broke some builds (Jani)
- New fb_pin abstraction for xe and i915 fb transparent handling (Ville, Tvrtko)
- Skip inactive MST connectors on HDCP cases (Suraj)
- Reduce redundant intel_panel_fixed_mode (Ankit)
- Some general fixes (Imre, Chaitanya)
- Reorganize display documentation (Jani)
- Start switching to display specific reg types (Jani)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patch.msgid.link/agXbLMtMECnKy-YV@intel.com
strut intel_fb_pin_params will be an important part of the fb_pin
interface, so move the definition to the parent interface file.
Or maybe we should have a separate header for this kind of stuff
since the users of the parent interface will need the struct
definition but not the parent interface vfunc struct definitions?
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-3-ville.syrjala@linux.intel.com
Introduce the "fb_pin" parent interface, as the first trivial step
move the *_get_map() stuff there.
The whole "fb_pin" as an interface might not really make sense,
and perhaps this (and other stuff) should just be collected into
some kind of "bo" interface. But let's go with "fb_pin" for now
to match where things are implemented, and possibly restructure
it later.
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260508143426.26504-2-ville.syrjala@linux.intel.com
Add additional DPCDs required to be configured to support VRR with Panel
Replay. These DPCDs are specifically required for configuring Adaptive Sync
SDP and are introduced in DP v2.1.
v2:
- Correct the shift for the bits. (Ville)
- Add DP_PR_ prefix for the PR-related fields.
v3:
- Use macro values in their shifted form to match the convention. (Ville)
v4:
- Add macro for the mask. (Ville)
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260428074457.3566918-5-ankit.k.nautiyal@intel.com
drm-misc-next for v7.2-rc1:
UAPI Changes:
- Support medium/low power modes in amdxdna.
- Support limiting frequency in ivpu.
- Document license for drm core uAPI headers.
- Add the following DRM formats: P230, Y7, XYYY2101010, T430,
XVUY210101010.
Cross-subsystem Changes:
- Add and improve dt-bindings.
- Remove unused dma-fence-array's signal_on_any support.
Core Changes:
- Do not call drop_master on file close if not master.
- Convert drm-bridge and drm/atomic to use drm_printf_indent.
- Remove the extra call to drm_connector_attach_encoder after
drm_bridge_connector_init().
- Assorted docbook updates.
Driver Changes:
- Bugfixes in amdxdna, ivpu, mipi-dsi, imagination, nouveau, panthor,
bridge/analogix_dp, ipv3, lontium-lt8912b, verisilicon, tve200,
etnaviv, panel/focaltech-ota7290b, panel/jadard-jd9365da-h3,
bridge/ite-it6263, renesas, xlnx, bridge/cdns-dsi, gma500,
bridge/microchip-lvds, mgag200.
- Add support for MStar TSUMU88ADT3-LF-1 bridge.
- Add support for WaveShare 7, Novatek NT35532, Startek KD070HDFLD092,
ChipWealth CH13726A AMOLED, Team Source Display TST070WSNE-196C,
Displaytech DT050BTFT-PTS panels.
- Improve mipi-dsi shutdown and convert a panasonic panel to use the
mipi-dsi wrappers.
- Allowing dumping vbios over debugfs in GSP-RM mode.
- Update maintainers for ivpu, add reviewer for drm-bridge code
and update maintainers for LT8912B DRM HDMI bridge.
- Add test pattern support to bridge/ti-sn65dsi83.
- Convert vmwgfx to vblank timers.
- Add power management to sysfb drm drivers to allow suspend/resume.
- Support the aforementioned new drm formats in xlnx/qynqmp.
- Fix panel Kconfig dependencies.
- Add carveout support for debugging and bringup to amxdna.
- Add support for long command tx via videobuffer in bridge/tc358768.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patch.msgid.link/f73f342d-6efb-416d-81b0-1716bdd98d5f@linux.intel.com
The kernel-doc block above struct drm_gpusvm_pages duplicates the
descriptions of the bit-flags that live in struct drm_gpusvm_pages_flags
using dotted notation (@flags.migrate_devmem, @flags.unmapped, ...).
That dotted notation is intended for nested anonymous structs/unions that
the parser flattens into the parent's parameter list. Here, however,
flags is of a named external type, so the parser does not flatten its
members and the dotted entries do not match any member of
drm_gpusvm_pages. They also duplicate the canonical descriptions already
present in the kernel-doc of struct drm_gpusvm_pages_flags itself.
Drop the five @flags.* lines and replace them with a single @flags entry
that cross-references the type via kernel-doc's "&struct ..." syntax.
This eliminates the redundancy and removes warnings emitted by the new
parameterdescs check in scripts/kernel-doc:
Excess struct member 'flags.migrate_devmem' description in
'drm_gpusvm_pages'
Excess struct member 'flags.unmapped' description in 'drm_gpusvm_pages'
Excess struct member 'flags.partial_unmap' description in
'drm_gpusvm_pages'
Excess struct member 'flags.has_devmem_pages' description in
'drm_gpusvm_pages'
Excess struct member 'flags.has_dma_mapping' description in
'drm_gpusvm_pages'
No functional change.
Assisted-by: Claude:claude-opus-4.6
Cc: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patch.msgid.link/20260501175956.4054088-1-shuicheng.lin@intel.com
This reverts commit e3953ff665.
Seems to have been accidentally pushed without mandatory review.
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de> #irc
Acked-by: Maxime Ripard <mripard@kernel.org> #irc
drm-misc-next for v7.1-rc1:
UAPI Changes:
- Expose per-client BO memory usage via fdinfo in amdxdna. (Hou)
- Change the default priority of drm scheduler to fair. (Tvrtko)
Cross-subsystem Changes:
- Revert hugetlb support in udmabuf. (Gunthorpe)
- Fix error in udmabuf with CONFIG_DMA_API_DEBUG(/ _SG). (Gavrilov)
- Add Docbook for DRM_IOCTL_SYNCOBJ_EVENTFD, (Ser)
clarify drm_bridge_get/put. (Tvrtko)
- Change signature of drm_connector_attach_hdr_output_metadata_property. (Canal)
- Use IOVA allocations in gpusvm and pagemap APIs. (Brost)
- Fix tracepoints vs dma-fence lifetime. (Tvrtko)
- Convert st-dma*.c tests to use kunit. (Gunthorpe)
Core Changes:
- Deduplicate counter and timestamp retrieval in vblank code. (Ville)
- Parse AMD VSDB v3 in CTA extension blocks, and use it in amdgpu. (Chen)
- Prevent bridge and encoder chain changes at inopportune times. (Ceresoli)
- Map the run queue 1:1 to the drm scheduler. (Tvrtko)
Driver Changes:
- Assorted bugfixes and (documentation) updates to rockchip, bridge/synopsis,
panfrost, tidss, accel/qaic, tilcdc, vc4, ast, imagination, panthor,
renesas, accel/amdxdna, msxfb, bridge/imx8mp, nouveau.
bridge/analogix_dp, bridge/exynos_dp, omap.
- Add support for CSW PNB601LS1-2, LGD LP116WHA-SPB1, panels.
- Add support for a lot of waveshare panels (Baryshkov)
- Support for AIE4 devices in accel/wamdxdna. (Zhang)
- Enable support for GEM shrinking in panthor. (Goel/Brezillon)
- Runtime Power Management is added to v3d. (Canal)
- Allow panel probing and use the panel bridge helper in analogix_dp. (Ding)
- Support XRGB1555 and C8 in mgag and XRGB1555 in ast. (Zimmermann)
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patch.msgid.link/bf31b1a1-951b-4f60-b226-22e8c083697d@linux.intel.com
Signed-off-by: Dave Airlie <airlied@redhat.com>
UAPI Changes:
- Debugfs multi-LRC engine info (Xin Wang, 2 commits)
Expose multi-LRC engine classes in the debugfs engine info output
and improve the output readability.
- drm/drm_ras: Add clear-error-counter netlink command to drm_ras
Cross-subsystem Changes:
Core Changes:
- drm/gpusvm: Reject VMAs with VM_IO or VM_PFNMAP when creating SVM ranges
- drm/i915/display: switch to including common step file directly
Driver Changes:
- Engine initialization cleanup (Matt Roper, 10 commits)
Clean up engine initialization code inherited from i915 by moving
hardware programming (CCS enablement, HWSTAM, GFX_MODE, BLIT_CCTL,
STOP_RING) into the RTP infrastructure. This makes the programming
visible and verifiable via debugfs, and applies consistently across
both normal and execlist init paths. Also fixes the name/definition
of GFX_MODE, marks BCS engines as belonging to the GT forcewake
domain, and drops the now-unused xe_hw_engine_mmio_write32() helper.
- PF fair scheduling auto-provisioning (Michal Wajdeczko, 13 commits)
Fix several corner cases in SR-IOV PF scheduling policy provisioning,
then auto-provision PF and all VFs with 16ms execution-quantum and
preemption-timeout defaults. This prevents a VF from monopolizing
the GPU by submitting workloads without gaps, without requiring the
user to configure sysfs manually.
- System Controller support for CRI/Xe3p (Anoop Vijay + Raag Jadav, 10 commits)
Add xe_sysctrl infrastructure for communicating with the System
Controller firmware entity on CRI/Xe3p discrete GPU platforms via a
mailbox interface. Includes type definitions, register definitions,
mailbox communication, initialization, power management, interrupt
handling, and event dispatch. Builds on this to add initial RAS
correctable error handling, using sysctrl interrupts to receive
threshold-crossed events.
- PXP state machine fixes (Daniele Ceraolo Spurio, 4 commits)
Fix PXP state machine handling: reject PXP sessions on PTL platforms
with older GSC firmware that does not support it, fix the restart
flag not being cleared after jumping back in pxp_start, remove
incorrect handling of an impossible state during suspend, and clean
up termination status on failure.
- Reset/wedge/unload corner case fixes (Zhanjun Dong + Matthew Brost, 5 commits)
Fix memory leaks and fence signal failures that occurred during GPU
reset, device wedging, and driver unload by forcefully tearing down
remaining exec queues in GuC submit fini, always killing queues in
pause/abort, and triggering queue cleanup when not in wedged mode 2.
Also ensures GuC CT state transitions via STOP before DISABLED.
- Wedge path memory allocation fixes (Matthew Brost, 3 commits)
Avoid GFP_KERNEL allocations in xe_device_declare_wedged(), which
runs in the DMA-fence signaling path. Also drops the
guc_submit_wedged_fini devm registration from xe_guc_submit_wedge()
to clean up the wedge shutdown sequence.
- PAT type cleanup and invalid index hardening (Xin Wang, 3 commits)
Standardize pat_index fields to u16 across the driver, default
XE_CACHE_NONE_COMPRESSION to XE_PAT_INVALID_IDX (matching
WB_COMPRESSION), and introduce xe_cache_pat_idx() — a macro helper
that validates cache_mode bounds and asserts on invalid PAT indices
before returning the index.
- Reject unsafe PAT indices for CPU cached memory (Jia Yao, 2 commits)
Reject incoherent (coh_none) PAT indices for CPU cached memory in
both the madvise ioctl and vm_bind with CPU_ADDR_MIRROR flag, closing
a security gap where the GPU could bypass CPU caches and observe
stale or sensitive data.
- OA improvements for CRI device memory (Ashutosh Dixit, 3 commits)
Move OA buffer access to the xe_map layer to support both system and
device memory (required for CRI), switch OA buffer mmap to use
drm_gem_mmap_obj, and implement workaround Wa_14026633728.
- xe_drm.h documentation fixes (Shuicheng Lin, 6 commits)
Fix multiple documentation issues in the xe_drm.h UAPI header:
typos, spelling errors, grammar, wrong names and references,
kernel-doc cross-reference syntax, and broken code examples.
- kernel-doc syntax fixes in xe headers (Shuicheng Lin, 4 commits)
Fix kernel-doc syntax issues across xe header files: missing '@'
prefixes on member tags, stale/mismatched member tags, comment
syntax errors, and type/parameter name mismatches in references.
- Buffer object and DMA-buf resource leak fixes (Shuicheng Lin, 4 commits)
Fix resource leaks on error paths: DMA-buf attachment leak in
xe_gem_prime_import(), BO leak in xe_dma_buf_init_obj() on
allocation failure, and BO leaks in xe_bo_init_locked() on GGTT
flag validation and unaligned size validation failures.
- Include guard cleanup (Shuicheng Lin, 5 commits)
Fix and standardize include guards across xe header files: normalize
double-underscore guards to single, add missing leading/trailing
underscores, add missing _H suffixes, and add guards to previously
unprotected headers.
- VF CCS memory pool (Satyanarayana K V P, 2 commits)
Switch VF CCS read/write operations from the DRM sub-allocator to
DRM mm, fixing allocation failures in fence-disabled mode where
the sub-allocator's hole cursor assumption breaks. Also introduces
a general BO-backed memory pool with shadow support using drm_mm.
- i915/xe step definition unification (Jani Nikula, 3 commits)
Complete the migration to the shared intel_step header: switch xe
from its own xe_step enum to the shared intel_step naming and
definitions, and update i915 display code to include the common
step header directly.
- Xe3p GT tuning fixes (3 commits)
Three Xe3p GT tuning corrections: fix the register offset for
GAMSTLB_CTRL, stop applying the CCCHKNREG1 tuning from Xe3p onward
(no longer needed), and mark ROW_CHICKEN5 as a masked register.
- Forcewake cleanup in GT and GuC PC (Raag Jadav, 3 commits)
Drop a redundant forcewake reference in xe_gt, reorder forcewake
usage in xe_guc_pc_fini_hw() to avoid a redundant hold, and convert
xe_guc_pc_stop() to void since it can no longer fail.
- SVM garbage collector fix on close (Matthew Brost, 2 commits)
Disable the SVM garbage collector work item when an SVM is closed
to prevent use-after-free when the GC fires after the SVM is freed.
- Admin-only PF mode (Satyanarayana K V P, 2 commits)
Derive an "admin-only PF" mode flag from xe_device state instead of
using a local flag, and restrict device query responses when running
in admin-only PF mode to avoid exposing internal state.
- Enable hwmon energy attributes and accepted power limit for CRI (2 commits)
Enable energy consumption hwmon attributes for the CRI platform and
add support for reading the accepted (sustained) power limit via
hwmon.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Hellstrom <thomas.hellstrom@linux.intel.com>
Link: https://patch.msgid.link/afNSeZJHC3X2m3-N@fedora
The KMS framework uses two slightly different definitions for the state
concept. For a given object (plane, CRTC, encoder, etc., so
drm_$OBJECT_state), the state is the entire state of that object.
However, at the device level, drm_atomic_state refers to a state update
for a limited number of objects.
Thus, drm_atomic_state isn't the entire device state, but only the full
state of some objects in that device. This has been an endless source of
confusion and thus bugs.
We can rename the drm_atomic_state structure to drm_atomic_commit to
make it less confusing.
This patch was created using:
rg -l drm_atomic_state | \
xargs sed -i 's/drm_atomic_state/drm_atomic_commit/g; s/drm_atomic_commit_helper/drm_atomic_state_helper/g'
mv drivers/gpu/drm/tests/drm_atomic_state_test.c drivers/gpu/drm/tests/drm_atomic_commit_test.c
Acked-by: Simona Vetter <simona.vetter@ffwll.ch>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Tested-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patch.msgid.link/20260427-drm-drm-atomic-update-v4-1-c0e713bfdf25@kernel.org
The merge-commit 02e778f123 ("Merge tag 'amd-drm-next-7.1-2026-03-12' of
https://gitlab.freedesktop.org/agd5f/linux into drm-next") removes the stub
for drm_fb_helper_gem_is_fb(), so the buld gets broken if DRM_FBDEV_EMULATION
is not set.
‘drm_fb_helper_gem_is_fb’; did you mean ‘drm_fb_helper_from_client’? [-Wimplicit-function-declaration]
1777 | if (!drm_fb_helper_gem_is_fb(dev->fb_helper, fb->obj[0])) {
| ^~~~~~~~~~~~~~~~~~~~~~~
| drm_fb_helper_from_client
Restore it.
Fixes: 02e778f123 ("Merge tag 'amd-drm-next-7.1-2026-03-12' of https://gitlab.freedesktop.org/agd5f/linux into drm-next")
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Yury Norov <ynorov@nvidia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 7b81bc38e9)
Video Format Data Blocks (VFDBs) contain the necessary information that
needs to be fed to the Optimized Video Timings (OVT) Algorithm.
Also, we require OVT support to cover modes that aren't supported by
earlier standards (e.g. CVT). So, parse all of the relevant VFDB data
and feed it to the OVT Algorithm, to extract all of the missing OVT
modes.
Suggested-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Hamza Mahfooz <someguy@effective-light.com>
Link: https://patch.msgid.link/20260225175709.408010-1-someguy@effective-light.com
Pull char / misc / IIO / and others driver updates from Greg KH:
"Here is the char/misc/iio and other smaller driver subsystem updates
for 7.1-rc1. Lots of stuff in here, all tiny, but relevant for the
different drivers they touch. Major points in here is:
- the usual large set of new IIO drivers and updates for that
subsystem (the large majority of this diffstat)
- lots of comedi driver updates and bugfixes
- coresight driver updates
- interconnect driver updates and additions
- mei driver updates
- binder (both rust and C versions) updates and fixes
- lots of other smaller driver subsystem updates and additions
All of these have been in linux-next for a while with no reported
issues"
* tag 'char-misc-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (405 commits)
coresight: tpdm: fix invalid MMIO access issue
mei: me: add nova lake point H DID
mei: lb: add late binding version 2
mei: bus: add mei_cldev_uuid
w1: ds2490: drop redundant device reference
bus: mhi: host: pci_generic: Add Telit FE912C04 modem support
mei: csc: wake device while reading firmware status
mei: csc: support controller with separate PCI device
mei: convert PCI error to common errno
mei: trace: print return value of pci_cfg_read
mei: me: move trace into firmware status read
mei: fix idle print specifiers
mei: me: use PCI_DEVICE_DATA macro
sonypi: Convert ACPI driver to a platform one
misc: apds990x: fix all kernel-doc warnings
most: usb: Use kzalloc_objs for endpoint address array
hpet: Convert ACPI driver to a platform one
misc: vmw_vmci: Fix spelling mistakes in comments
parport: Remove completed item from to-do list
char: remove unnecessary module_init/exit functions
...
Pull more drm fixes from Dave Airlie:
"These are the regular fixes that have built up over last couple of
weeks, all pretty minor and spread all over.
atomic:
- raise the vblank timeout to avoid it on virtual drivers
- fix colorop duplication
bridge:
- stm_lvds: state check fix
- dw-mipi-dsi: bridge reference leak fix
panel:
- visionx-rm69299: init fix
dma-fence:
- fix sparse warning
dma-buf:
- UAF fix
panthor:
- mapping fix
arcgpu:
- device_node reference leak fix
nouveau:
- memory leak in error path fix
- overflow in reloc path for old hw fix
hv:
- Kconfig fix
v3d:
- infinite loop fix"
* tag 'drm-fixes-2026-04-24' of https://gitlab.freedesktop.org/drm/kernel:
drm/nouveau: fix u32 overflow in pushbuf reloc bounds check
MAINTAINERS: split hisilicon maintenance and add Yongbang Shi for hibmc-drm matainers
drm/v3d: Reject empty multisync extension to prevent infinite loop
drm/panel: visionox-rm69299: Make use of prepare_prev_first
drm/drm_atomic: duplicate colorop states if plane color pipeline in use
drm/nouveau: fix nvkm_device leak on aperture removal failure
hv: Select CONFIG_SYSFB only for CONFIG_HYPERV_VMBUS
dma-fence: Silence sparse warning in dma_fence_describe
drm/bridge: dw-mipi-dsi: Fix bridge leak when host attach fails
drm/arcpgu: fix device node leak
drm/panthor: Fix outdated function documentation
drm/panthor: Extend VM locked region for remap case to be a superset
dma-buf: fix UAF in dma_buf_put() tracepoint
drm/bridge: stm_lvds: Do not fail atomic_check on disabled connector
drm/atomic: Increase timeout in drm_atomic_helper_wait_for_vblanks()
This week in drm-misc-fixes, we have:
- A patch to raise the vblank timeout to avoid it on virtual drivers
- a state check fix for stm_lvds
- a use-after-free fix for dma-buf
- a mapping fix for panthor
- a device_node reference leak fix for arcgpu
- a bridge reference leak fix for dw-mipi-dsi
- a sparse warning fix for dma-fence
- a kconfig fix for hv
- a memory leak fix for nouveau
- a fix to duplicate colorop when duplicating states
- a panel initialisation order fix for visionox-rm69299
- a fix to prevent an infinite loop for v3d
- an overflow fix for nouveau
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maxime Ripard <mripard@redhat.com>
Link: https://patch.msgid.link/20260423-realistic-eager-reindeer-4dacf7@houat
If the DSI panel supports versatile lanes configuration, its driver
might require determining the number of DSI data lanes, which is usually
specified on the DSI host side of the OF graph. Add new helper as a
pair to drm_of_get_data_lanes_count_ep() that lets callers determine
number of data-lanes on the remote side of the OF graph.
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://patch.msgid.link/20260413-waveshare-dsi-touch-v3-6-3aeb53022c32@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Since the new FAIR policy is in general better than FIFO and almost as
good as round-robin in interactive use cases, plus the latter has not been
the default policy in a long time, we can afford to remove both and leave
just FAIR.
By doing so we can simplify the scheduler code by making the scheduler to
run queue relationship always 1:1 and remove some code.
Also, now that the FIFO policy is gone the tree of entities is not a FIFO
tree any more so rename it to just the tree.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Philipp Stanner <phasta@kernel.org>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://patch.msgid.link/20260417103744.76020-14-tvrtko.ursulin@igalia.com
GPUs do not always implement preemption and DRM scheduler definitely
does not support it at the front end scheduling level. This means
execution quanta can be quite long and is controlled by userspace,
consequence of which is picking the "wrong" entity to run can have a
larger negative effect than it would have with a virtual runtime based CPU
scheduler.
Another important consideration is that rendering clients often have
shallow submission queues, meaning they will be entering and exiting the
scheduler's runnable queue often.
Relevant scenario here is what happens when an entity re-joins the
runnable queue with other entities already present. One cornerstone of the
virtual runtime algorithm is to let it re-join at the head and rely on the
virtual runtime accounting and timeslicing to sort it out.
However, as explained above, this may not work perfectly in the GPU world.
Entity could always get to overtake the existing entities, or not,
depending on the submission order and rbtree equal key insertion
behaviour.
Allow interactive jobs to overtake entities already queued up for the
limited case when interactive entity is re-joining the queue after
being idle.
This gives more opportunity for the compositors to have their rendering
executed before the GPU hogs even if they have been configured with the
same scheduling priority.
To classify a client as interactive we look at its average job duration
versus the average for the whole scheduler. We can track this easily by
plugging into the existing job runtime tracking and applying the
exponential moving average window on the past submissions. Then, all other
things being equal, we let the more interactive jobs go first.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Philipp Stanner <phasta@kernel.org>
Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://patch.msgid.link/20260417103744.76020-12-tvrtko.ursulin@igalia.com
The FAIR scheduling policy is built upon the same concepts as the well
known CFS CPU scheduler - entity run queue is sorted by the virtual GPU
time consumed by entities in a way that the entity with least vruntime
runs first.
It is able to avoid total priority starvation, which is one of the
problems with FIFO, and it also does not need for per priority run queues.
As it scales the actual GPU runtime by an exponential factor as the
priority decreases, the virtual runtime for low priority entities grows
faster than for normal priority, pushing them further down the runqueue
order for the same real GPU time spent.
Apart from this fundamental fairness, fair policy is especially strong in
oversubscription workloads where it is able to give more GPU time to short
and bursty workloads when they are running in parallel with GPU heavy
clients submitting deep job queues.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Philipp Stanner <phasta@kernel.org>
Cc: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://patch.msgid.link/20260417103744.76020-11-tvrtko.ursulin@igalia.com
To implement fair scheduling we need a view into the GPU time consumed by
entities. Problem we have is that jobs and entities objects have decoupled
lifetimes, where at the point we have a view into accurate GPU time, we
cannot link back to the entity any longer.
Solve this by adding a light weight entity stats object which is reference
counted by both entity and the job and hence can safely be used from
either side.
With that, the only other thing we need is to add a helper for adding the
job's GPU time into the respective entity stats object, and call it once
the accurate GPU time has been calculated.
The most convenient place to do that is the free job worker for several
reasons. Doing the accounting from the job completion callback would mean
a few locks would need to become irq safe and we would also need to worry
about out of order completions (via dma_fence_is_signaled calls which we
cannot control). In-order completions are critical for GPU time accuracy
which is currently adjusted per fence in the free worker and requires
looking at the next job in the scheduler pending list. We would also need
to add a new lock to protect the scheduler average stats update.
In contrast to those complications, having the accounting done from the
free worker is serialized by definition and all the above complications
are avoided. Downside is there is potential for a time lag between job
completions and GPU time being accounted against the entity. Since that is
partly alleviated by batch processing the completed job queue, and the
scheduling algorithm does not attempt to be completely fair, which would
even be rather impossible to achieve in the GPU world with the current
DRM scheduler design and hardware with no or poor preemption support,
this downside is not considered critical. Plus, in practice the scheduler
is also affected by worker scheduling delays from other angles too. Not
least being able to promptly feed the GPU with new work.
We therefore choose the simple option and can later consider improving
upon it if the need arises.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Philipp Stanner <phasta@kernel.org>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://patch.msgid.link/20260417103744.76020-9-tvrtko.ursulin@igalia.com
Round-robin being the non-default policy and unclear how much it is used,
we can notice that it can be implemented using the FIFO data structures if
we only invent a fake submit timestamp which is monotonically increasing
inside drm_sched_rq instances.
So instead of remembering which was the last entity the scheduler worker
picked we can simply bump the picked one to the bottom of the tree, which
ensures round-robin behaviour between all active queued jobs.
If the picked job was the last from a given entity, we remember the
assigned fake timestamp and use it to re-insert the job once it re-joins
the queue. This ensures the job neither overtakes all already queued jobs,
neither it goes last. Instead it keeps the position after the currently
queued jobs and before the ones which haven't yet been queued at the point
the entity left the queue.
Advantage is that we can consolidate to a single code path and remove a
bunch of code. Downside is round-robin mode now needs to lock on the job
pop path but that should not have a measurable performance impact.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Philipp Stanner <phasta@kernel.org>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Tested-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://patch.msgid.link/20260417103744.76020-7-tvrtko.ursulin@igalia.com
Some drivers do not care on teardown whether the last jobs pending in an
entity are actually executed before teardown completed. For such
scenarios, drm_sched_entity_flush() is not the ideal function since it's
intended to wait for jobs to complete.
Make drm_sched_entity_kill() public for that use-case and update the
documentation.
Suggested-by: Danilo Krummrich <dakr@kernel.org>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://patch.msgid.link/20260415144956.272506-2-phasta@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
For suspend/resume to work correctly, do for colorop state the same we
do for plane/crtc/connector states: duplicate the state of colorops in a
color pipeline if it's in use by a given plane when suspending and
restore cached colorop states when resuming. While at it, prevent
unused-variable warning when using for_each_new_colorop_in_stage here.
Fixes: 2afc3184f3 ("drm/plane: Add COLOR PIPELINE property")
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Signed-off-by: Melissa Wen <mwen@igalia.com>
Link: https://patch.msgid.link/20260318163629.300627-1-mwen@igalia.com
Signed-off-by: Melissa Wen <melissa.srw@gmail.com>
drm_for_each_bridge_in_chain_scoped() and
drm_for_each_bridge_in_chain_from() currently get/put the bridge at each
iteration. But they don't protect the encoder chain, so it could change
(bridges added/removed) while some code is iterating over the list
itself. Such code can then derail on incorrect pointers.
To make iterations safe, augment these for_each macros to lock the encoder
chain mutex at the beginning and unlock it at the end of the loop (be it at
the end of the list, or earlier due to a 'break' or 'return' statement).
This change requires more operations when starting and ending the loop. To
avoid making the macros even more complex, move these operations to helper
functions. Also remname some of the existing helper functions for
consistency.
Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
Link: https://patch.msgid.link/20260324-drm-bridge-alloc-encoder-chain-mutex-v5-4-8bf786c5c7e6@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
The initial plane vblank wait operates on display registers,
so it really belongs in the display code proper. Move it there.
We can use intel_parent_irq_enabled() to determine if we can
rely on interrupts or not.
On average we should end up waiting half a frame here, so the
polling interval can be fairly long. 1 ms (which actually
makes poll_timeout_us() use ~250-1000 usec) seems good enough
to me.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260410150449.9699-6-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
The "plane_config" vs. "plane_configs" naming difference is very
subtle, making it far too easy to use the wrong thing by accident.
Introduce a separate type for the array, making it impossible to
pass in the wrong thing. And while at it name the variable
"all_plane_configs" to help the poor reader make sense of things.
The .config_fini() prototype also mistakenly used the plural
form despite only taking in a singular plane_config. So fix that
one up as well.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patch.msgid.link/20260410150449.9699-2-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
