- 08 Jun, 2018 2 commits
-
-
Dave Airlie authored
Merge tag 'drm-misc-next-fixes-2018-05-31' of git://anongit.freedesktop.org/drm/drm-misc into drm-next drm-misc-next-fixes for v4.18: Driver changes: - Plug small memory leak in vc4. (anholt) - Depend on MMU in v3d. (arnd) Signed-off-by:
Dave Airlie <airlied@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/12faab25-e809-e73c-409c-5e9c08aa351c@linux.intel.com
-
git://people.freedesktop.org/~robclark/linuxDave Airlie authored
A bunch of msm fixes and some atomic work. Signed-off-by:
-
- 04 Jun, 2018 6 commits
-
-
Sean Paul authored
This patch avoids dereferencing msm_host->dev when it is NULL. If we find ourselves tearing down dsi before calling (mdp4|mdp5|dpu)_kms_init(), we'll end up in a state where the dev pointer is NULL and trying to extract priv from it will fail. This was introduced in a seemingly innocuous commit to ensure the arguments to msm_gem_put_iova() are correct (even though that function has been a stub for ~5 years). Correctness FTW! \o/ Fixes: b01884a286b0 drm/msm: use correct aspace pointer in msm_gem_put_iova() Cc: Daniel Mack <daniel@zonque.org> Cc: Rob Clark <robdclark@gmail.com> Signed-off-by:
Sean Paul <seanpaul@chromium.org> Signed-off-by:
Rob Clark <robdclark@gmail.com>
-
Sean Paul authored
Now that all of the msm-specific goo is tucked safely away we can switch over to using the atomic helper commit directly. \o/ Changes in v2: - None Changes in v3: - Rebased on Archit's private_obj set Changes in v4: - None Cc: Abhinav Kumar <abhinavk@codeaurora.org> Signed-off-by:
-
Sean Paul authored
Moving further towards switching fully to the the atomic helpers, this patch removes the hand-rolled worker nonblock commit code and uses the atomic helpers commit_work model. Changes in v2: - Remove commit_destroy() - Shuffle order of commit_tail calls to further serialize commits - Use stall in swap_state to avoid abandoned events on disable Changes in v3: - Rebased on Archit's private_obj set Changes in v4: - None Signed-off-by:
-
Sean Paul authored
Ensure that any queued events are issued when disabling the crtc. This avoids timeouts when we come back and wait for dependencies (like the previous frame's flip_done). Changes in v2: - None Changes in v3: - Rebased on Archit's private_obj set Changes in v4: - None Reviewed-by:
Archit Taneja <architt@codeaurora.org> Signed-off-by:
-
Sean Paul authored
In preparation for moving to atomic helpers, move the implicit sync fence handling out of atomic commit and into the plane->prepare_fb() hook. While we're at it, de-duplicate the mdp*_prepare_fb functions. Changes in v4: - Added Reported-by:
-
Sean Paul authored
Factor out the commit_tail() portions of complete_commit() into a separate function to facilitate moving to the atomic helpers in future patches. Changes in v2: - None Changes in v3: - Rebased on Archit's private_obj set Changes in v4: - None Cc: Jeykumar Sankaran <jsanka@codeaurora.org> Reviewed-by:
-
- 03 Jun, 2018 13 commits
-
-
Archit Taneja authored
With the addition of "private_objs" in drm_atomic_state, we no longer need to subclass drm_atomic_state to store state of share resources that don't perfectly fit within planes/crtc/connector state information. We can now save this state within drm_atomic_state itself using the private objects. Remove the infrastructure that allowed subclassing of drm_atomic_state in the driver. Changes in v3: - Added to the msm atomic helper patch set Changes in v4: - None Signed-off-by:
-
Archit Taneja authored
This replaces the usage of the subclassed atomic state (mdp5_state) with a private_obj state embedded within drm_atomic_state. The latter method is the preferred approach, since it's simpler to implement and less prone to errors. The new API replaces the older and equivalent mdp5_state usage in the following pattern: - References to "mdp5_kms->state" (i.e, the old/existing state) is replaced with mdp5_get_existing_global_state(). In the atomic_check path, this should be called with the glob_state_lock drm_modeset_lock alredy taken. - References to "mdp5_get_state()" are replaced with mdp5_get_global_state(). This acquires glob_state_lock and uses drm_atomic_get_private_obj_state() to create a new duplicated state. Changes in v3: - Acquire glob_state_lock in mdp5_smp.c - Added to the msm atomic helper patch set Changes in v4: - None Signed-off-by:
-
Archit Taneja authored
Global shared resources (hwpipes, hwmixers and SMP) for MDP5 are implemented as a part of atomic state by subclassing drm_atomic_state. The preferred approach is to use the drm_private_obj infrastructure available in the atomic core. mdp5_global_state is introduced as a drm atomic private object. The two funcs mdp5_get_global_state() and mdp5_get_existing_global_state() are the two variants that will be used to access mdp5_global_state. This will replace the existing mdp5_state struct (which subclasses drm_atomic_state) and the funcs around it. These will be removed later once we mdp5_global_state is put to use everywhere. Changes in v3: - Added glob_state_lock instead of pushing it into the core - Added to the msm atomic helper patch set Changes in v4: - None Signed-off-by:
-
Daniel Mack authored
Even though msm_gem_put_iova() is currently a NOP function, the caller should pass in the address space pointer it used to obtain the object. Other call sites were changed in 8bdcd949 ("drm/msm: pass address-space to _get_iova() and friends"), but this one seems to have been forgotten. Signed-off-by:
Daniel Mack <daniel@zonque.org> Cc: Rob Clark <robdclark@gmail.com> Signed-off-by:
-
Daniel Mack authored
This regression stems from 0e08270a ("drm/msm: Separate locking of buffer resources from struct_mutex"). Signed-off-by:
-
Emil Velikov authored
Currently the error pointer returned by msm_alloc_stolen_fb gets passed to drm_framebuffer_remove. The latter handles only NULL pointers, thus a nasty crash will occur. Drop the unnecessary fail label and the associated checks - both err and fb will be set at this stage. Cc: Rob Clark <robdclark@gmail.com> Cc: linux-arm-msm@vger.kernel.org Cc: dri-devel@lists.freedesktop.org Cc: freedreno@lists.freedesktop.org Signed-off-by:
Emil Velikov <emil.velikov@collabora.com> Signed-off-by:
-
Stefan Agner authored
The function dsi_get_cmd_fmt returns enum dsi_cmd_dst_format, use the correct enum value also for MIPI_DSI_FMT_RGB666/_PACKED. This has been discovered using clang: drivers/gpu/drm/msm/dsi/dsi_host.c:743:35: warning: implicit conversion from enumeration type 'enum dsi_vid_dst_format' to different enumeration type 'enum dsi_cmd_dst_format' [-Wenum-conversion] case MIPI_DSI_FMT_RGB666: return VID_DST_FORMAT_RGB666; ~~~~~~ ^~~~~~~~~~~~~~~~~~~~~ Signed-off-by:Stefan Agner <stefan@agner.ch> Reviewed-by:
-
Ben Hutchings authored
Commit 62e3a3e3 changed get_pages() to initialise msm_gem_object::pages before trying to initialise msm_gem_object::sgt, so that put_pages() would properly clean up pages in the failure case. However, this means that put_pages() now needs to check that msm_gem_object::sgt is not null before trying to clean it up, and this check was only applied to part of the cleanup code. Move it all into the conditional block. (Strictly speaking we don't need to make the kfree() conditional, but since we can't avoid checking for null ourselves we may as well do so.) Fixes: 62e3a3e3 ("drm/msm: fix leak in failed get_pages") Signed-off-by:
Ben Hutchings <ben.hutchings@codethink.co.uk> Reviewed-by:
Jordan Crouse <jcrouse@codeaurora.org> Signed-off-by:
-
Jeykumar Sankaran authored
This change plumbs the new fb modifier through the various mdp/disp get_format hooks. Signed-off-by:
Jeykumar Sankaran <jsanka@codeaurora.org> [seanpaul pimped out commit message a bit] Signed-off-by:
-
Sean Paul authored
Don't leave the event != NULL once it's consumed, this is used a signal to the atomic helpers that the event will be handled by the driver. Changes in v2: - None Changes in v3: - Rebased on Archit's private_obj set Changes in v4: - None Cc: Jeykumar Sankaran <jsanka@codeaurora.org> Reviewed-by:
-
Abhinav Kumar authored
Currently the DSI PHY timings are hard-coded for a specific panel for the 10nm PHY. Replace this with the auto PHY timing calculator which can calculate the PHY timings for any panel. Changes in v4: - None Changes in v3: - None Changes in v2: - None Reviewed-by:
Abhinav Kumar <abhinavk@codeaurora.org> Signed-off-by:
-
Abhinav Kumar authored
Make sure the video mode engine is on before waiting for the video done interrupt. Changes in v4: - Move setting enabled to false earlier Changes in v3: - Move the return value check to another patch Changes in v2: - Replace pr_err with dev_err - Changed error message Signed-off-by:
-
Abhinav Kumar authored
Check for the return value of wait for video done waits and print appropriate error message. Signed-off-by:
-
- 30 May, 2018 4 commits
-
-
Arnd Bergmann authored
Without CONFIG_MMU, we get a link error: drivers/gpu/drm/v3d/v3d_bo.o: In function `v3d_gem_fault': v3d_bo.c:(.text+0x3ca): undefined reference to `vm_insert_mixed' The other drivers with this problem already depend on CONFIG_MMU, so let's do the same thing here. Fixes: 57692c94 ("drm/v3d: Introduce a new DRM driver for Broadcom V3D V3.x+") Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
Eric Anholt <eric@anholt.net> Link: https://patchwork.freedesktop.org/patch/msgid/20180525155030.3667352-5-arnd@arndb.de
-
Eric Anholt authored
Signed-off-by:
Boris Brezillon <boris.brezillon@bootlin.com>
-
git://people.freedesktop.org/~gabbayo/linuxDave Airlie authored
- Build amdkfd's related files inside amdgpu only if amdkfd is built - Fix compile warning - Print info message in case ASIC is not supported by amdkfd Signed-off-by:
-
Dave Airlie authored
Merge tag 'exynos-drm-next-for-v4.18-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos into drm-next Add more HW overlays support - It enables hardware overlay number 4 and 5. For this, this patch series adds required clocks. Several fixups - Fix default value of zpos according to real hardware overlay number. - Fix error value of exynos_Drm_crtc_get_by_type function correctly. - Fix static checker warning of scaler_task_done function. - Fix signedness bug in fimc_setup_clocks function. One cleanup - Disable framedone interrupt of DSI device which is not required. Signed-off-by:
-
- 25 May, 2018 4 commits
-
-
Arnd Bergmann authored
Casting a pointer to a 64-bit type causes a warning on 32-bit targets: drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c:473:24: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast] lower_32_bits((uint64_t)wptr)); ^ drivers/gpu/drm/amd/amdgpu/amdgpu.h:1701:53: note: in definition of macro 'WREG32' #define WREG32(reg, v) amdgpu_mm_wreg(adev, (reg), (v), 0) ^ drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c:473:10: note: in expansion of macro 'lower_32_bits' lower_32_bits((uint64_t)wptr)); ^~~~~~~~~~~~~ The correct method is to cast to 'uintptr_t'. Fixes: d5a114a6 ("drm/amdgpu: Add GFXv9 kfd2kgd interface functions") Signed-off-by:Oded Gabbay <oded.gabbay@gmail.com> Signed-off-by:
-
Dan Carpenter authored
"id" needs to be signed for the error handling to work. Fixes: 7a2d5c77 ("drm/exynos: fimc: Convert driver to IPP v2 core API") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Inki Dae <inki.dae@samsung.com>
-
Inki Dae authored
drivers/gpu/drm/exynos/exynos_drm_scaler.c:402 scaler_task_done() warn: signedness bug returning '(-22)' Reported-by:
-
git://people.freedesktop.org/~agd5f/linuxDave Airlie authored
Last feature request for 4.18. Mostly vega20 support. - Vega20 support - clock and powergating for VCN - misc bug fixes Signed-off-by:
-
- 24 May, 2018 11 commits
-
-
Tom Stellard authored
This is an important message, so it should be visible to users without having to enable extra debugging. Signed-off-by:
Tom Stellard <tstellar@redhat.com> Reviewed-by:
-
David Francis authored
In fixed31_32.h, in dc_fixpt_shl,'/' was used for division of one long long int by another long long int. As there is no inbuilt long long int division function in c, gcc inserted its own. However, gcc does not link the library that contains this function. To avoid this, use bitwise operators instead of / Reviewed-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
David Francis <David.Francis@amd.com> Reviewed-by:
Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Signed-off-by:
-
Shaoyun Liu authored
Update to the latest version from the vbios team. Signed-off-by:
Shaoyun Liu <Shaoyun.Liu@amd.com> Acked-by:
-
Alex Deucher authored
is_dpm_running callback was assigned to the same value twice. Drop the duplicate. Reviewed-by:
Michel Dänzer <michel.daenzer@amd.com> Signed-off-by:
-
Emily Deng authored
To free the fence from the amdgpu_fence_slab, need twice call_rcu, to avoid the amdgpu_fence_slab_fini call kmem_cache_destroy(amdgpu_fence_slab) before kmem_cache_free(amdgpu_fence_slab, fence), add rcu_barrier after drm_sched_entity_fini. The kmem_cache_free(amdgpu_fence_slab, fence)'s call trace as below: 1.drm_sched_entity_fini -> drm_sched_entity_cleanup -> dma_fence_put(entity->last_scheduled) -> drm_sched_fence_release_finished -> drm_sched_fence_release_scheduled -> call_rcu(&fence->finished.rcu, drm_sched_fence_free) 2.drm_sched_fence_free -> dma_fence_put(fence->parent) -> amdgpu_fence_release -> call_rcu(&f->rcu, amdgpu_fence_free) -> kmem_cache_free(amdgpu_fence_slab, fence); v2:put the barrier before the kmem_cache_destroy v3:put the dma_fence_put(fence->parent) before call_rcu in drm_sched_fence_release_scheduled Signed-off-by:
Emily Deng <Emily.Deng@amd.com> Reviewed-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
-
Christian König authored
Move all BOs belonging to a VM on the LRU with every submission. Signed-off-by:
Chunming Zhou <david1.zhou@amd.com> Signed-off-by:
-
Christian König authored
Instead of sometimes checking if the vm_status is empty use the moved flag and also reset it when the BO leaves the state machine. Signed-off-by:
Junwei Zhang <Jerry.Zhang@amd.com> Signed-off-by:
-
Christian König authored
In theory it is possible that PDs/PTs can move without eviction. Signed-off-by:
-
Christian König authored
Splice the moved list to a local one to avoid taking the lock over and over again. Signed-off-by:
-
Christian König authored
Use list_for_each_entry_safe here. v2: Drop the optimization, it doesn't work as expected. Signed-off-by:
-
Christian König authored
Only the moved state needs a separate spin lock protection. All other states are protected by reserving the VM anyway. v2: fix some more incorrect cases Signed-off-by:
-
