kernel/git/next/linux-next.git/drivers/gpu, branch pending-fixes

Merge branch 'for-linux-next-fixes' of https://gitlab.freedesktop.org/drm/misc/kernel.git

2026-09-15T12:53:28+00:00

Merge branch 'msm-fixes' of https://gitlab.freedesktop.org/drm/msm.git

2026-09-15T12:53:24+00:00

drm/vc4: Use managed KMS polling to fix UAF on unbind

2026-09-14T15:11:42+00:00

vc4_kms_load() calls drm_kms_helper_poll_init() but the driver provides no matching drm_kms_helper_poll_fini(). The output poll work stays scheduled after unbind and runs on the freed drm_device: # modprobe vc4; rmmod vc4; sleep 10 BUG: KASAN: slab-use-after-free in delayed_work_timer_fn BUG: KASAN: slab-use-after-free in drm_client_dev_hotplug [drm] Workqueue: events output_poll_execute [drm_kms_helper] Allocated by task 171: __devm_drm_dev_alloc Freed by task 262 (rmmod): drm_dev_put / component_del Use drmm_kms_helper_poll_init() so polling is finalized with the device, as other drivers do. Fixes: c8b75bca92cb ("drm/vc4: Add KMS support for Raspberry Pi.") Assisted-by: Claude:claude-fable-5 Signed-off-by: Karl Mehltretter Link: https://patch.msgid.link/20260822143110.68594-1-kmehltretter@gmail.com Reviewed-by: Maíra Canal Signed-off-by: Maíra Canal

Merge tag 'rust-fixes-7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux

2026-09-13T16:28:28+00:00

Pull Rust fixes from Miguel Ojeda: "Toolchain and infrastructure: - Work around a 'bindgen' 0.73.2 bug that emits an 'allow' attribute for 'unnecessary_transmutes', which is unknown in older compilers - Clean 'clippy::as_underscore' lints in generated code by the new 'bindgen' 0.73.0+ releases - Clean new 'clippy::needless_range_loop' lint for the upcoming Rust 1.100.0 (expected 2026-11-12) 'kernel' crate: - 'num' module: fix soundness issue in 'Bounded' by sealing the 'Integer' trait 'pin-init' crate: - Fix unreachable warning for the upcoming Rust 1.100.0 (expected 2026-11-12) due to 'Infallible' becoming an alias of '!' Samples: - Add missing newlines in 'pr_*!'s macro calls" * tag 'rust-fixes-7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux: rust: allow `unknown_lints` in generated bindings for Rust < 1.88 rust: allow `clippy::as_underscore` in the generated bindings rust: num: seal Integer drm/panic: clean new `clippy::needless_range_loop` lint for Rust 1.100.0 rust: samples: add missing newlines in rust_print_main rust: pin-init: use irrefutable pattern for `stack_pin_init`

drm/msm/a6xx+: Increase GMU FW init timeout

2026-09-12T17:01:54+00:00

We were using 10ms, kgsl uses 100ms. In practice it is usually takes less than 10ms, but very occasionally goes a bit above 10ms, leading to a "GMU firmware inialization timed out", from which point things go south. There are probably some things that could be done to speed up init, like increasing GMU freq. But to be safe, increase the timeout to match kgsl. Signed-off-by: Rob Clark Reviewed-by: Akhil P Oommen Patchwork: https://patchwork.freedesktop.org/patch/753014/ Message-ID: <20260912150915.28700-1-robin.clark@oss.qualcomm.com>

drm/gud: fix out-of-bounds write in gud_plane_atomic_check()

2026-09-12T12:56:44+00:00

The plane property loop uses req->properties[num_properties + i] as write index while simultaneously incrementing `num_properties` inside the loop. At iteration i, num_properties has also incremented by i, so the write is done at `initial_num_properties + 2*i`, skipping every other index and advancing by 2 per iteration. With just 2 connector and 32 plane properties the last write happens at index 64, one slot past the end of the 64-slot (indices 0–63) allocation. A USB device can trigger OOB by advertising the maximum number of properties. Fix by dropping the redundant `+ i`; num_properties is already the correct running index, as gud_connector_fill_properties() fills the preceding slots. Fixes: 40e1a70b4aed ("drm: Add GUD USB Display driver") Reported-by: Sashiko Link: https://sashiko.dev/#/patchset/20260821071812.16500-1-sajal2005gupta%40gmail.com?part=1 Signed-off-by: Sajal Gupta Cc: Acked-by: Ruben Wauters Signed-off-by: Ruben Wauters Link: https://patch.msgid.link/20260902123254.36987-1-sajal2005gupta@gmail.com

drm/gud: Ignore damage clips in full update mode

2026-09-12T12:52:10+00:00

When running in full update mode, previously small updates (such as moving the mouse across the screen) would cause many full frames to be generated. This would bog down the bus and lower the effective framerate significantly - I was seeing a drop from 60 FPS to 2 FPS. Set ignore_damage_clips in full update mode so the damage iterator yields a single full-plane rectangle instead of one per clip. Fixes: 73cfd166e045 ("drm/gud: Replace simple display pipe with DRM atomic helpers") Cc: # 6.18.x Signed-off-by: Sophie D Reviewed-by: Thomas Zimmermann Acked-by: Ruben Wauters Signed-off-by: Ruben Wauters Link: https://patch.msgid.link/20260910014910.8564-1-patches@scd31.com

drm/msm/a6xx: Use CX AO Counter register for timestamp on a750 GPUs

2026-09-11T18:36:41+00:00

The a750 uses the GMU CX AO Counters instead of the GMU_ALWAYS_ON_COUNTER register on A6xx and other A7xx GPUs, use it when running a A750 GPU. The GMU_ALWAYS_ON_COUNTER at offset 0x1f888 doesn't seem to exist on the SM8650 A750 GMU and returns 0, but the CX AO counter at offset 0x1f880 returns some proper timestamp data. Signed-off-by: Neil Armstrong Patchwork: https://patchwork.freedesktop.org/patch/752235/ Message-ID: <20260909-topic-sm8650-gmu-a750-timestamp-reg-v2-2-091d74958951@linaro.org> Signed-off-by: Rob Clark

drm/msm/a6xx: Add CX AO Counter registers used for a750 GPUs

2026-09-11T18:36:41+00:00

The a750 uses the CX AO Counters instead of the GMU_ALWAYS_ON_COUNTER register on A6xx and other A7xx GPUs. Signed-off-by: Neil Armstrong Patchwork: https://patchwork.freedesktop.org/patch/752234/ Message-ID: <20260909-topic-sm8650-gmu-a750-timestamp-reg-v2-1-091d74958951@linaro.org> Signed-off-by: Rob Clark

drm/msm: remove stale perf counter XML TODO

2026-09-11T18:33:27+00:00

The adreno_perfcntrs macro already takes the XML file stem as its second argument, allowing each perf counter JSON file to select the appropriate register XML. The a2xx and a5xx entries already pass their respective XML file stems. Remove the stale TODO comment. Signed-off-by: George Emmanuel Thomas Patchwork: https://patchwork.freedesktop.org/patch/746745/ Message-ID: <20260815164335.158958-1-georgeemmanuelthomas@gmail.com> Signed-off-by: Rob Clark