In addition to generic VFIO PCI functionality, the driver implements
VFIO migration uAPI, allowing userspace to enable migration for Intel
Graphics SR-IOV Virtual Functions.
The driver binds to VF device and uses API exposed by Xe driver to
transfer the VF migration data under the control of PF device.
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Alex Williamson <alex@shazbot.org>
Link: https://patch.msgid.link/20251127093934.1462188-5-michal.winiarski@intel.com
Link: https://lore.kernel.org/all/20251128125322.34edbeaf.alex@shazbot.org/
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 2e38c50ae4)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Device specific VFIO driver variant for Xe will implement VF migration.
Export everything that's needed for migration ops.
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251127093934.1462188-4-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 17f22465c5)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
In certain scenarios (such as VF migration), VF driver needs to interact
with PF driver.
Add a helper to allow VF driver access to PF xe_device.
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251127093934.1462188-3-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 8b3cce3ad9)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
All of the necessary building blocks are now in place to support SR-IOV
VF migration.
Flip the enable/disable logic to match VF code and disable the feature
only for platforms that don't meet the necessary prerequisites.
To allow more testing and experiments, on DEBUG builds any missing
prerequisites will be ignored.
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251127093934.1462188-2-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 01c724aa7b)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Add a scope-based helpers for runtime PM that may be used to simplify
cleanup logic and potentially avoid goto-based cleanup.
For example, using
guard(xe_pm_runtime)(xe);
will get runtime PM and cause a corresponding put to occur automatically
when the current scope is exited. 'xe_pm_runtime_noresume' can be used
as a guard replacement for the corresponding 'noresume' variant.
There's also an xe_pm_runtime_ioctl conditional guard that can be used
as a replacement for xe_runtime_ioctl():
ACQUIRE(xe_pm_runtime_ioctl, pm)(xe);
if ((ret = ACQUIRE_ERR(xe_pm_runtime_ioctl, &pm)) < 0)
/* failed */
In a few rare cases (such as gt_reset_worker()) we need to ensure that
runtime PM is dropped when the function is exited by any means
(including error paths), but the function does not need to acquire
runtime PM because that has already been done earlier by a different
function. For these special cases, an 'xe_pm_runtime_release_only'
guard can be used to handle the release without doing an acquisition.
These guards will be used in future patches to eliminate some of our
goto-based cleanup.
v2:
- Specify success condition for xe_pm runtime_ioctl as _RET >= 0 so
that positive values will be properly identified as success and
trigger destructor cleanup properly.
v3:
- Add comments to the kerneldoc for the existing 'get' functions
indicating that scope-based handling should be preferred where
possible. (Gustavo)
Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://patch.msgid.link/20251118164338.3572146-31-matthew.d.roper@intel.com
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
(cherry picked from commit 59e7528dbf)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
- Restrict the pointer size of flush pages to prevent a regression.
-----BEGIN PGP SIGNATURE-----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=9d7X
-----END PGP SIGNATURE-----
Merge tag 'drm-misc-next-fixes-2025-11-26' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next
drm-misc-next-fixes for v6.19:
- Restrict the pointer size of flush pages to prevent a regression.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patch.msgid.link/0090a4fc-9cc4-4c03-bfe5-d1b1f0cc7e05@linux.intel.com
Ensure that the DMA address of the framebuffer flush page is not larger
than its hardware register.
On GPUs older than Hopper, the register for the address can hold up to a
40-bit address (right-shifted by 8 so that it fits in the 32-bit
register), and on Hopper and later it can be 52 bits (64-bit register
where bits 52-63 must be zero).
Recently it was discovered that under certain conditions, the flush page
could be allocated outside this range. Although this bug was fixed, we
can ensure that any future changes to this code don't accidentally
generate an invalid page address.
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Link: https://patch.msgid.link/20251113230323.1271726-2-ttabi@nvidia.com
The flush page DMA address is stored in a special register that is not
associated with the GPU's standard DMA range. For example, on Turing,
the GPU's MMU can handle 47-bit addresses, but the flush page address
register is limited to 40 bits.
At the point during device initialization when the flush page is
allocated, the DMA mask is still at its default of 32 bits. So even
though it's unlikely that the flush page could exist above a 40-bit
address, the dma_map_page() call could fail, e.g. if IOMMU is disabled
and the address is above 32 bits. The simplest way to achieve all
constraints is to allocate the page in the DMA32 zone. Since the flush
page is literally just a page, this is an acceptable limitation. The
alternative is to temporarily set the DMA mask to 40 (or 52 for Hopper
and later) bits, but that could have unforseen side effects.
In situations where the flush page is allocated above 32 bits and IOMMU
is disabled, you will get an error like this:
nouveau 0000:65:00.0: DMA addr 0x0000000107c56000+4096 overflow (mask ffffffff, bus limit 0).
Fixes: 5728d06419 ("drm/nouveau/fb: handle sysmem flush page from common code")
Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Link: https://patch.msgid.link/20251113230323.1271726-1-ttabi@nvidia.com
In xe_oa_add_config_ioctl(), we accessed oa_config->id after dropping
metrics_lock. Since this lock protects the lifetime of oa_config, an
attacker could guess the id and call xe_oa_remove_config_ioctl() with
perfect timing, freeing oa_config before we dereference it, leading to
a potential use-after-free.
Fix this by caching the id in a local variable while holding the lock.
v2: (Matt A)
- Dropped mutex_unlock(&oa->metrics_lock) ordering change from
xe_oa_remove_config_ioctl()
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6614
Fixes: cdf02fe1a9 ("drm/xe/oa/uapi: Add/remove OA config perf ops")
Cc: <stable@vger.kernel.org> # v6.11+
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Sanjay Yadav <sanjay.kumar.yadav@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patch.msgid.link/20251118114859.3379952-2-sanjay.kumar.yadav@intel.com
(cherry picked from commit 28aeaed130)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
The code incorrectly assumes that the VRAM save/restore fence is valid.
Fix it by checking for error.
Fixes: 49cf1b9b60 ("drm/xe/pf: Handle VRAM migration data as part of PF control")
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251114122339.1791026-1-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 78ff838a8a)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
The reference is only dropped on error. Fix it by adding the missing
xe_bo_put().
Fixes: 49cf1b9b60 ("drm/xe/pf: Handle VRAM migration data as part of PF control")
Reported-by: Adam Miszczak <adam.miszczak@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patch.msgid.link/20251114100713.1776073-1-michal.winiarski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit dab751b424)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
The kernel-doc for xe_sriov_pf_migration_save_consume() contained
multiple "Return:" sections, causing a warning.
Fix it by removing the extra line.
Fixes: 67df4a5cbc ("drm/xe/pf: Add data structures and handlers for migration rings")
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://patch.msgid.link/20251114134030.1795947-1-michal.winiarski@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit 562b0f254d)
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Changes for v6.19:
GPU:
- Gen8 support: A840 (Kaanapali) and X2-85 (Glymur)
- A612 support
- A few NULL check fixes
MDSS:
- Added support for Glymur and QCS8300 platforms
DPU:
- Enabled Quad-Pipe support, unlocking higher resolutions support
- Added support for Glymur platform
- Documented DPU on QCS8300 platform as supported
- Misc small fixes
DisplayPort:
- Added support for Glymur platform
- Added support lame remapping inside DP block
- Documented DisplayPort controller on QCS8300 and SM6150/QCS615 as
supported
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rob Clark <rob.clark@oss.qualcomm.com>
Link: https://patch.msgid.link/CACSVV00sFi65XrZipHCU3C0bYji7vgu7OgWvLeOQ1Cg475_pUA@mail.gmail.com
-----BEGIN PGP SIGNATURE-----
iQFSBAABCgA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAmkaT2IeHHRvcnZhbGRz
QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiGnakIAIq6O35+abAlDGe7
n8IwX/4hwDJ9Ol2vddg6jTdio3szSlvcXRAyijkG35etLap+i+Q78Q9H1fRcQ8CF
5t9kfs5oCVWEDlsHMI+Q38vYB3tWikaIUIuao6pH5U50fsA0KM/287TLYdzvmBsK
gy0BHoZhl++xdMsD2HNwt08RrK/IyY+Zp9i7pKhcsxJiJCgy8nq5ZxhmRPbqTKNc
IqIBPzG+UYcOd88Iz377Ax+TnzWaTpwyvfaiU8DVFMN0FeENdk1FY8ClYfpYxCDH
Pm5NzL/f54VtdZ3aL1fKmYwWN7WFJT4FH3l6DOvtCc+Gx3KW1a3Y508CbqYBXvgY
QV4oH4Y=
=iSoY
-----END PGP SIGNATURE-----
Merge tag 'v6.18-rc6' into drm-next
Linux 6.18-rc6
Backmerge in order to merge msm next
Signed-off-by: Dave Airlie <airlied@redhat.com>
In older versions of Rust, the compiler doesn't know about the newer
`as_flattened*` methods, thus `rustdoc` complains about the intra-doc
links, e.g.
error: unresolved link to `slice::as_flattened`
--> rust/kernel/slice.rs:19:23
|
19 | /// [`as_flattened`]: slice::as_flattened
| ^^^^^^^^^^^^^^^^^^^ the primitive type `slice` has no associated item named `as_flattened`
|
= note: `-D rustdoc::broken-intra-doc-links` implied by `-D warnings`
= help: to override `-D warnings` add `#[allow(rustdoc::broken_intra_doc_links)]`
Thus fix them by using an URL instead.
Fixes: 88622323dd ("rust: enable slice_flatten feature and provide it through an extension trait")
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://patch.msgid.link/20251119185125.1411151-1-ojeda@kernel.org
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
Rust
- Make slice::as_flattened usable on all supported versions of rustc.
- Add FromBytes::from_bytes_prefix() method.
Core Changes:
- Update Tyr in MAINTAINERS file.
- Remove redundant device ptr from Rust GEM object.
- Change how AlwaysRefCounted is implemented for GEM objects.
- Add deferred vm_bo cleanup to GPUVM and use it in Panthor.
Driver Changes:
Nova Core
- Introduction of bitfield! macro, with support for different storage sizes
and custom visibility.
- Introduction of safe converters between integer types for which the
conversion is lossless.
- GSP initialized up to fully booted state on Ampere.
- Use more future-proof register for GPU identification.
- Various simplifications and optimizations.
Nova
- Select NOVA_CORE.
- Depend on CONFIG_64BIT.
-----BEGIN PGP SIGNATURE-----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=yKU+
-----END PGP SIGNATURE-----
Merge tag 'drm-rust-next-2025-11-18' of https://gitlab.freedesktop.org/drm/rust/kernel into drm-next
Cross-subsystem Changes:
Rust
- Make slice::as_flattened usable on all supported versions of rustc.
- Add FromBytes::from_bytes_prefix() method.
Core Changes:
- Update Tyr in MAINTAINERS file.
- Remove redundant device ptr from Rust GEM object.
- Change how AlwaysRefCounted is implemented for GEM objects.
- Add deferred vm_bo cleanup to GPUVM and use it in Panthor.
Driver Changes:
Nova Core
- Introduction of bitfield! macro, with support for different storage sizes
and custom visibility.
- Introduction of safe converters between integer types for which the
conversion is lossless.
- GSP initialized up to fully booted state on Ampere.
- Use more future-proof register for GPU identification.
- Various simplifications and optimizations.
Nova
- Select NOVA_CORE.
- Depend on CONFIG_64BIT.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alice Ryhl <aliceryhl@google.com>
Link: https://patch.msgid.link/aRxtJC0D1pQUepF4@google.com
The function mtk_dp_dt_parse() calls of_graph_get_endpoint_by_regs()
to get the endpoint device node, but fails to call of_node_put() to release
the reference when the function returns. This results in a device node
reference leak.
Fix this by adding the missing of_node_put() call before returning from
the function.
Found via static analysis and code review.
Fixes: f70ac097a2 ("drm/mediatek: Add MT8195 Embedded DisplayPort driver")
Cc: stable@vger.kernel.org
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Markus Schneider-Pargmann <msp@baylibre.com>
Reviewed-by: CK Hu <ck.hu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20251029072307.10955-1-linmq006@gmail.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Document Adreno X2-85 GMU found in Glymur chipsets in the
dt-binding specification. It is very similar to Adreno 840
GMU with the additional requirement of RSCC HUB clock.
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689032/
Message-ID: <20251118-kaana-gpu-support-v4-21-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Document Adreno 840 GMU in the dt-binding specification.
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689031/
Message-ID: <20251118-kaana-gpu-support-v4-20-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Update the devicetree bindings to document the GPU SMMUs present in
Kaanapali and Glymur chipsets.
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689028/
Message-ID: <20251118-kaana-gpu-support-v4-19-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Adreno X2-85 GPU is found in the next generation of Qualcomm's compute
series chipset called Snapdragon X2 Elite (a.k.a Glymur). It is based
on the new A8x slice architecture and features up to 4 slices. Due to
the wider 12 channel DDR support, there is higher DDR bandwidth available
than previous generation to improve performance.
Add a new entry in the catalog along with the necessary register
configurations to enable support for it.
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689026/
Message-ID: <20251118-kaana-gpu-support-v4-18-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
GMU lies on the CX domain and accesses CX GBIF. So do CX GBIF
configurations before GMU wakes up. This was not a problem so far, but
A840 GPU is very sensitive to this requirement. Also, move these
registers to the catalog.
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689024/
Message-ID: <20251118-kaana-gpu-support-v4-17-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Adreno 840 present in Kaanapali SoC is the second generation GPU in
A8x family. It comes in 2 variants with either 2 or 3 Slices. This is
in addition to the SKUs supported based on the GPU FMAX.
Add the necessary register configurations to the catalog and enable
support for it.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689022/
Message-ID: <20251118-kaana-gpu-support-v4-16-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
AQE (Applicaton Qrisc Engine) is a dedicated core inside CP which aides
in Raytracing related workloads. Add support for loading the AQE firmware
and initialize the necessary registers.
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689020/
Message-ID: <20251118-kaana-gpu-support-v4-15-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
A8x is the next generation of Adreno GPUs, featuring a significant
hardware design change. A major update to the design is the introduction
of Slice architecture. Slices are sort of mini-GPUs within the GPU which
are more independent in processing Graphics and compute workloads. Also,
in addition to the BV and BR pipe we saw in A7x, CP has more concurrency
with additional pipes.
From a software interface perspective, these changes have a significant
impact on the KMD side. First, the GPU register space has been extensively
reorganized. Second, to avoid a register space explosion caused by the
new slice architecture and additional pipes, many registers are now
virtualized, instead of duplicated as in A7x. KMD must configure an
aperture register with the appropriate slice and pipe ID before accessing
these virtualized registers.
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689019/
Message-ID: <20251118-kaana-gpu-support-v4-14-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
A8x GMU firmwares expect a separate vote table which describes the
relationship between the Gx rail and MxA rail (and possibly Cx rail).
Create this new vote table and implement the new HFI message which
allows passing vote tables to send this data to GMU.
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689016/
Message-ID: <20251118-kaana-gpu-support-v4-13-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Current logic assumes that the voltage corners in both MxG and MxA are
always same. This is not true for recent targets. So, rework the rpmh init
sequence to probe and calculate the votes with the respective rails, ie,
GX rails should use MxG as secondary rail and Cx rail should use MxA as
the secondary rail.
Fixes: d6225e0cd0 ("drm/msm/adreno: Add support for X185 GPU")
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689014/
Message-ID: <20251118-kaana-gpu-support-v4-12-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
A8x GMU configurations are very similar to A7x. Unfortunately, there are
minor shuffling in the register offsets in the GMU CX register region.
So, update the driver to use the correct register offsets on A8x hw.
Some A8x GPUs have more than 16 powerlevels on GX domain and 4 on CX
domain. To accommodate this, increase the arrays' sizes which hold gx and
cx power levels.
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689013/
Message-ID: <20251118-kaana-gpu-support-v4-11-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
GMU registers are always at a fixed offset from the GPU base address,
a consistency maintained at least within a given architecture generation.
In A8x family, the base address of the GMU has changed, but the offsets
of the gmu registers remain largely the same. To enable reuse of the gmu
code for A8x chipsets, update the gmu register offsets to be relative
to the GPU's base address instead of GMU's.
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689010/
Message-ID: <20251118-kaana-gpu-support-v4-10-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Sync the latest register definitions from Mesa which includes the
updates for A8x family.
Co-developed-by: Rob Clark <robin.clark@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689009/
Message-ID: <20251118-kaana-gpu-support-v4-9-86eeb8e93fb6@oss.qualcomm.com>
Move MMU fault handler for each generation to adreno function list. This
will help to use common code for mmu pagefault handler registration between
a6x/a7x and a8x layer.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689007/
Message-ID: <20251118-kaana-gpu-support-v4-8-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Move the gbif halt fn to adreno_gpu_func so that we can call different
implementation from common code. This will come handy when we implement
A8x layer.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689005/
Message-ID: <20251118-kaana-gpu-support-v4-7-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
In A6x family (which is a pretty big one), there are separate
adreno_func definitions for each sub-generations. To streamline the
identification of the correct struct for a gpu, move it to the
catalogue and move the gpu_init routine to struct adreno_gpu_funcs.
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689003/
Message-ID: <20251118-kaana-gpu-support-v4-6-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Newer gen's introduce pipe enums which do not exist on older gens, but
the numeric values do not conflict. IOW, they are backward compatible.
So move its definition to adreno_common.xml.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689001/
Message-ID: <20251118-kaana-gpu-support-v4-5-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Crashdec doesn't require SCRATCH registers anymore for a6xx and newer
architectures. So skip dumping them during recovery.
Suggested-by: Rob Clark <rob.clark@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689000/
Message-ID: <20251118-kaana-gpu-support-v4-4-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Correct the register offset and enable this workaround for all A7x
and newer GPUs to match the recommendation. Also, downstream does this
w/a after moving the fence to allow mode. So do the same.
Fixes: dbfbb376b5 ("drm/msm/a6xx: Add A621 support")
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/688997/
Message-ID: <20251118-kaana-gpu-support-v4-3-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
As per the recommendation, A7x and newer GPUs should flush the LRZ cache
before switching the pagetable. Update a6xx_set_pagetable() to do this.
While we are at it, sync both BV and BR before issuing a
CP_RESET_CONTEXT_STATE command, to match the downstream sequence.
Fixes: af66706acc ("drm/msm/a6xx: Add skeleton A7xx support")
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/688995/
Message-ID: <20251118-kaana-gpu-support-v4-2-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
REG_A6XX_GMU_AO_AHB_FENCE_CTRL register falls under GMU's register
range. So, use gmu_write() routines to write to this register.
Fixes: 1707add815 ("drm/msm/a6xx: Add a6xx gpu state")
Cc: stable@vger.kernel.org
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/688993/
Message-ID: <20251118-kaana-gpu-support-v4-1-86eeb8e93fb6@oss.qualcomm.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Based on mesa commit 3f70b0578402 ("freedreno/registers: Fix encoding
fields in 64b registers"), but with some fixes to not skip emitting
interrupt enum values.
v2: Don't append "ull" to 32b reg MASK defines, to avoid printf format
conversion warnings all over the place
Co-developed-by: Connor Abbott <cwabbott0@gmail.com>
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/689141/
Message-ID: <20251118152952.226510-1-robin.clark@oss.qualcomm.com>
if matrixbit is 11,
The range of color matrix is from 0 to (BIT(12) - 1).
Values from 0 to (BIT(11) - 1) represent positive numbers,
values from BIT(11) to (BIT(12) - 1) represent negative numbers.
For example, -1 need converted to 8191.
so convert S31.32 to HW Q2.11 format by drm_color_ctm_s31_32_to_qm_n,
and set int_bits to 2.
Fixes: 738ed4156f ("drm/mediatek: Add matrix_bits private data for ccorr")
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Jay Liu <jay.liu@mediatek.com>
Link: https://patchwork.kernel.org/project/dri-devel/patch/20250921055416.25588-2-jay.liu@mediatek.com/
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
amdgpu:
- RAS updates
- GC12 DCC P2P fix
- Documentation fixes
- Power limit code cleanup
- Userq updates
- VRR fix
- SMART OLED support
- DSC refactor for DCN 3.5
- Replay updates
- DC clockgating updates
- HDCP refactor
- ISP fix
- SMU 13.0.12 updates
- JPEG 5.0.1 fix
- VCE1 support
- Enable DC by default on SI
- Refactor CIK and SI enablement
- Enable amdgpu by default for CI dGPUs
- XGMI fixes
- SR-IOV fixes
- Memory allocation critical path fixes
- Enable amdgpu by default on SI dGPUs
amdkfd:
- Relax checks on save area overallocations
- Fix GPU mappings after prefetch
radeon:
- Refactor CIK and SI enablement
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQQgO5Idg2tXNTSZAr293/aFa7yZ2AUCaReBigAKCRC93/aFa7yZ
2JqCAQDZYnl6lkfzXcm3z5qqI3DmEyREO75hn0IEffyehAhUYwD9FpC48tTCyzeY
1MQZ+ORH12VXuiT4Y+wBIiFetQV96QA=
=uThs
-----END PGP SIGNATURE-----
Merge tag 'amd-drm-next-6.19-2025-11-14' of https://gitlab.freedesktop.org/agd5f/linux into drm-next
amd-drm-next-6.19-2025-11-14:
amdgpu:
- RAS updates
- GC12 DCC P2P fix
- Documentation fixes
- Power limit code cleanup
- Userq updates
- VRR fix
- SMART OLED support
- DSC refactor for DCN 3.5
- Replay updates
- DC clockgating updates
- HDCP refactor
- ISP fix
- SMU 13.0.12 updates
- JPEG 5.0.1 fix
- VCE1 support
- Enable DC by default on SI
- Refactor CIK and SI enablement
- Enable amdgpu by default for CI dGPUs
- XGMI fixes
- SR-IOV fixes
- Memory allocation critical path fixes
- Enable amdgpu by default on SI dGPUs
amdkfd:
- Relax checks on save area overallocations
- Fix GPU mappings after prefetch
radeon:
- Refactor CIK and SI enablement
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patch.msgid.link/20251114192553.442621-1-alexander.deucher@amd.com
Driver Changes:
Fixes/improvements/new stuff:
- Avoid lock inversion when pinning to GGTT on CHV/BXT+VTD (Janusz Krzysztofik)
- Use standard API for seqcount read in TLB invalidation [gt] (Andi Shyti)
Miscellaneous:
- Wait longer for threads in migrate selftest on CHV/BXT+VTD (Janusz Krzysztofik)
- Wait for page_sizes_gtt in gtt selftest on CHV/BXT+VTD (Janusz Krzysztofik)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Tvrtko Ursulin <tursulin@igalia.com>
Link: https://patch.msgid.link/aRdXOAKlTVX_b0en@linux
drm/i915 feature pull #2 for v6.19:
Features and functionality:
- Add initial display support for Xe3p_LPD, display version 35 (Sai Teja, Matt
R, Gustavo, Matt A, Ankit, Juha-pekka, Luca, Ravi Kumar)
- Compute LT PHY HDMI params when port clock not in predefined tables (Suraj)
Refactoring and cleanups:
- Refactor intel_frontbuffer split between i915, xe, and display (Ville)
- Clean up intel_de_wait_custom() usage (Ville)
- Unify display register polling interfaces (Ville)
- Finish removal of the expensive format info lookups (Ville)
- Cursor code cleanups (Ville)
- Convert intel_rom interfaces to struct drm_device (Jani)
Fixes:
- Fix uninitialized variable in DSI exec packet (Jonathan)
- Fix PIPEDMC logging (Alok Tiwari)
- Fix PSR pipe to vblank conversion (Jani)
- Fix intel_frontbuffer lifetime handling (Ville)
- Disable Panel Replay on DP MST for the time being (Imre)
Merges:
- Backmerge drm-next to get the drm_print.h changes (Jani)
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Jani Nikula <jani.nikula@intel.com>
Link: https://patch.msgid.link/b131309bb7310ab749f1770aa6e36fa8d6a82fa5@intel.com
If there is a flood of faults then the MMU can become saturated while it
waits for the kernel to process the first fault and resume it, so that
the GMU becomes blocked. This is mainly a problem when the kernel reads
the state of the GPU for a devcoredump, because this takes a while. If
we timeout waiting for the GMU, check if this has happened and retry
after we're finished.
Signed-off-by: Connor Abbott <cwabbott0@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/664685/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
If the rootfs have a legacy A200 firmware, currently the driver will
complain each time the hw is reinited (which can happen a lot). E.g.
with GL testsuite the hw is reinited after each test, spamming the
console.
Make sure that the message is printed only once: when we detect the
firmware that doesn't support protection.
Fixes: 302295070d ("drm/msm/a2xx: support loading legacy (iMX) firmware")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/688098/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
The crashstate_get_bos() function allocates memory for `state->bos`
using kcalloc(), but the vmbind path does not check for allocation
failure before dereferencing it in the following drm_gpuvm_for_each_va()
loop. This could lead to a NULL pointer dereference if memory allocation
fails.
Fix this by wrapping the drm_gpuvm_for_each_va() loop with a NULL check
on state->bos, similar to the safety check in the non-vmbind path.
Fixes: af9aa6f316 ("drm/msm: Crashdump support for sparse")
Signed-off-by: Huiwen He <hehuiwen@kylinos.cn>
Patchwork: https://patchwork.freedesktop.org/patch/687556/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
crashstate_get_vm_logs() did not check the return value of
kmalloc_array(). In low-memory situations, kmalloc_array() may return
NULL, leading to a NULL pointer dereference when the function later
accesses state->vm_logs.
Fix this by checking the return value of kmalloc_array() and setting
state->nr_vm_logs to 0 if allocation fails.
Fixes: 9edc52967c ("drm/msm: Add VM logging for VM_BIND updates")
Signed-off-by: Huiwen He <hehuiwen@kylinos.cn>
Patchwork: https://patchwork.freedesktop.org/patch/687555/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
