Each V3D queue works independently and all the dependencies between the
jobs are handled through the DRM scheduler. Therefore, there is no need
to use one single lock for all queues. Using it, creates unnecessary
contention between different queues that can operate independently.
Replace the global spinlock with per-queue locks to improve parallelism
and reduce contention between different V3D queues (BIN, RENDER, TFU,
CSD). This allows independent queues to operate concurrently while
maintaining proper synchronization within each queue.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-3-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Instead of storing the queue's active job in four different variables,
store the active job inside the queue's state. This way, it's possible
to access all active jobs using an index based in `enum v3d_queue`.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-2-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Instead of storing a pointer to the DRM file data, store a pointer
directly to the private V3D file struct. No functional change, this
commit only avoids multiple levels of pointer indirection and makes
the code more straightforward.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-1-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
ADV75xx hardware supports sending SPD InfoFrame over the HDMI link. Also
it provides support for two generic (Spare) InfoFrames. Use those
capabilities to be able to send SPD and HDMI Vendor-Specific Infoframes.
Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-adv7511-infoframes-v1-1-f89b9690f89c@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
1. Add error handling for old state CRTC in atomic_disable
2. Fix DSI host and panel bridge pre-enable order
3. Fix device/node reference count leaks in mtk_drm_get_all_drm_priv
4. mtk_hdmi: Fix inverted parameters in some regmap_update_bits calls
-----BEGIN PGP SIGNATURE-----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=j/Jq
-----END PGP SIGNATURE-----
Merge tag 'mediatek-drm-fixes-20250829' of https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux into drm-fixes
Mediatek DRM Fixes - 20250829
1. Add error handling for old state CRTC in atomic_disable
2. Fix DSI host and panel bridge pre-enable order
3. Fix device/node reference count leaks in mtk_drm_get_all_drm_priv
4. mtk_hdmi: Fix inverted parameters in some regmap_update_bits calls
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Link: https://lore.kernel.org/r/20250828234116.4960-1-chunkuang.hu@kernel.org
In mtk_hdmi driver, a recent change replaced custom register access
function calls by regmap ones, but two replacements by regmap_update_bits
were done incorrectly, because original offset and mask parameters were
inverted, so fix them.
Fixes: d6e25b3590 ("drm/mediatek: hdmi: Use regmap instead of iomem for main registers")
Signed-off-by: Louis-Alexis Eyraud <louisalexis.eyraud@collabora.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20250818-mt8173-fix-hdmi-issue-v1-1-55aff9b0295d@collabora.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series.
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Brian Masney <bmasney@redhat.com>
Patchwork: https://patchwork.freedesktop.org/patch/667982/
Link: https://lore.kernel.org/r/20250811-drm-clk-round-rate-v2-4-4a91ccf239cf@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series.
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Brian Masney <bmasney@redhat.com>
Patchwork: https://patchwork.freedesktop.org/patch/667980/
Link: https://lore.kernel.org/r/20250811-drm-clk-round-rate-v2-3-4a91ccf239cf@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.
Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667878/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-7-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.
Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667876/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-6-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.
Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667874/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-5-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.
Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667872/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-4-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.
Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667870/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-3-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.
Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667869/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-2-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.
Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667866/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-1-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
irq_domain_create_simple() takes fwnode as the first argument. It can be
extracted from the struct device using dev_fwnode() helper instead of
using of_node with of_fwnode_handle().
So use the dev_fwnode() helper.
Signed-off-by: Jiri Slaby (SUSE) <jirislaby@kernel.org>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>
Cc: Dmitry Baryshkov <lumag@kernel.org>
Cc: Sean Paul <sean@poorly.run>
Cc: Marijn Suijten <marijn.suijten@somainline.org>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/665227/
Link: https://lore.kernel.org/r/20250723055512.1800438-1-jirislaby@kernel.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Use the newly added of_reserved_mem_region_to_resource() function to
handle "memory-region" properties.
The original code did not set 'zap_available' to false if
of_address_to_resource() failed which seems like an oversight.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/662275/
Link: https://lore.kernel.org/r/20250703183442.2073717-1-robh@kernel.org
[DB: dropped part related to VRAM, no longer applicable]
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
less restrictive with the formats it accepts. A fix for amdgpu to pin
vmapped dma-buf, and a revert for tegra for a regression in the dma-buf
/ GEM code.
-----BEGIN PGP SIGNATURE-----
iJUEABMJAB0WIQTkHFbLp4ejekA/qfgnX84Zoj2+dgUCaK/35wAKCRAnX84Zoj2+
duAqAX9IIovrilRa/6169LIVEFbI3NfhWJorjDp5GB6jEilKxu6LBWFNYmyiLLwI
0+c4WiEBfAk/jHkpO4fdtAbIczic5d1WdSlO3VipH6w8WMDca1A3DcDM4BXO2x5Z
74f5CctKSA==
=PWA/
-----END PGP SIGNATURE-----
Merge tag 'drm-misc-fixes-2025-08-28' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes
Several nouveau fixes to remove unused code, fix an error path and be
less restrictive with the formats it accepts. A fix for amdgpu to pin
vmapped dma-buf, and a revert for tegra for a regression in the dma-buf
/ GEM code.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maxime Ripard <mripard@redhat.com>
Link: https://lore.kernel.org/r/20250828-hypersonic-colorful-squirrel-64f04b@houat
Handle LVDS pixel clock.
The LTDC operates with multiple clock domains for register access,
requiring all clocks to be provided during read/write operations. This
imposes a dependency between the LVDS and LTDC to access correctly all
LTDC registers. And because both IPs' pixel rates must be synchronized,
the LTDC has to handle the LVDS clock.
Signed-off-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Philippe Cornu <philippe.cornu@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-8-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
STM32MP25 SoC features a new version of the LTDC IP. Add its compatible
to the list of device to probe and implement its quirks.
This hardware supports a pad frequency of 150MHz and a peripheral bus
clock.
Signed-off-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Philippe Cornu <philippe.cornu@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-7-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.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>
Add driver extension for Synopsys DesignWare DPTX IP used
on Rockchip RK3588 SoC.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Acked-by: Dmitry Baryshkov <lumag@kernel.org>
Tested-by: Nicolas Frattaroli <nicolas.frattaroli@collabora.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250822063959.692098-4-andyshrk@163.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
The DW DP TX Controller is compliant with the DisplayPort Specification
Version 1.4 with the following features:
* DisplayPort 1.4a
* Main Link: 1/2/4 lanes
* Main Link Support 1.62Gbps, 2.7Gbps, 5.4Gbps and 8.1Gbps
* AUX channel 1Mbps
* Single Stream Transport(SST)
* Multistream Transport (MST)
* Type-C support (alternate mode)
* HDCP 2.2, HDCP 1.3
* Supports up to 8/10 bits per color component
* Supports RBG, YCbCr4:4:4, YCbCr4:2:2, YCbCr4:2:0
* Pixel clock up to 594MHz
* I2S, SPDIF audio interface
Add library with common helpers to make it can be shared with
other SoC.
Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Tested-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Link: https://lore.kernel.org/r/20250822063959.692098-3-andyshrk@163.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.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>
The only callers to mmu_hw_do_operation_locked() pass an 'op' of either
AS_COMAND_FLUSH_MEM or AS_COMMAND_FLUSH_PT. This means the code paths
after that are dead. Removing those paths means the
mmu_hw_do_flush_on_gpu_ctrl() function might has well be inlined.
Simplify everything by having a switch statement for the type of 'op'
(warning if we get an unexpected value) and removing the dead cases.
Suggested-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Steven Price <steven.price@arm.com>
Reviewed-by: Karunika Choo <karunika.choo@arm.com>
Link: https://lore.kernel.org/r/20250815134226.57703-1-steven.price@arm.com
Since all users of gem.gpuva.lock_dep_map now rely on the mutex directly
in gpuva, we may remove it. Whether the mutex is used is now tracked by
a flag in gpuvm rather than by whether lock_dep_map is null.
Note that a GEM object may not be pushed to multiple gpuvms that
disagree on the value of this new flag. But that's okay because a single
driver should use the same locking scheme everywhere, and a GEM object
is driver specific (when a GEM is exported with prime, a new GEM object
instance is created from the backing dma-buf).
The flag is present even with CONFIG_LOCKDEP=n because the intent is
that the flag will also cause vm_bo cleanup to become deferred. However,
that will happen in a follow-up patch.
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250827-gpuva-mutex-in-gem-v3-3-bd89f5a82c0d@google.com
[ Use lockdep_is_held() instead of lock_is_held(). - Danilo ]
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Now that drm_gem_object has a dedicated mutex for the gpuva list that is
intended to be used in cases that must be fence signalling safe, use it
in Panthor.
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250827-gpuva-mutex-in-gem-v3-2-bd89f5a82c0d@google.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
There are two main ways that GPUVM might be used:
* staged mode, where VM_BIND ioctls update the GPUVM immediately so that
the GPUVM reflects the state of the VM *including* staged changes that
are not yet applied to the GPU's virtual address space.
* immediate mode, where the GPUVM state is updated during run_job(),
i.e., in the DMA fence signalling critical path, to ensure that the
GPUVM and the GPU's virtual address space has the same state at all
times.
Currently, only Panthor uses GPUVM in immediate mode, but the Rust
drivers Tyr and Nova will also use GPUVM in immediate mode, so it is
worth to support both staged and immediate mode well in GPUVM. To use
immediate mode, the GEMs gpuva list must be modified during the fence
signalling path, which means that it must be protected by a lock that is
fence signalling safe.
For this reason, a mutex is added to struct drm_gem_object that is
intended to achieve this purpose. Adding it directly in the GEM object
both makes it easier to use GPUVM in immediate mode, but also makes it
possible to take the gpuva lock from core drm code.
As a follow-up, another change that should probably be made to support
immediate mode is a mechanism to postpone cleanup of vm_bo objects, as
dropping a vm_bo object in the fence signalling path is problematic for
two reasons:
* When using DRM_GPUVM_RESV_PROTECTED, you cannot remove the vm_bo from
the extobj/evicted lists during the fence signalling path.
* Dropping a vm_bo could lead to the GEM object getting destroyed.
The requirement that GEM object cleanup is fence signalling safe is
dubious and likely to be violated in practice.
Panthor already has its own custom implementation of postponing vm_bo
cleanup.
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250827-gpuva-mutex-in-gem-v3-1-bd89f5a82c0d@google.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Remove unnecessary semicolons reported by Coccinelle/coccicheck and the
semantic patch at scripts/coccinelle/misc/semicolon.cocci.
Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250828085708.3383760-1-nichen@iscas.ac.cn
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
In drm_sched_fini() all entities are marked as stopped - without taking
the appropriate lock, because that would deadlock. That means that
drm_sched_fini() and drm_sched_entity_push_job() can race against each
other.
This should most likely be fixed by establishing the rule that all
entities associated with a scheduler must be torn down first. Then,
however, the locking should be removed from drm_sched_fini() alltogether
with an appropriate comment.
Reported-by: James Flowers <bold.zone2373@fastmail.com>
Link: https://lore.kernel.org/dri-devel/20250720235748.2798-1-bold.zone2373@fastmail.com/
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250813085654.102504-2-phasta@kernel.org
The header file <linux/atomic.h> is already included on line 8. Remove the
redundant include.
Fixes: 5a99350794 ("drm/sched: Add scheduler unit testing infrastructure and some basic tests")
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250819142630.368796-1-liaoyuanhong@vivo.com
Indirect buffers are an AMD term describing essentialy a job submitted to
the scheduler, just a lower level one. Since scheduler was promoted to be
generic long ago, lets replace those references with jobs.
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>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250814133627.2550-1-tvrtko.ursulin@igalia.com
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>
