- 27 Nov, 2023 2 commits
-
-
Luben Tuikov authored
Fix compilation issues with DRM scheduler priority rename MIN to LOW. Signed-off-by:
Luben Tuikov <ltuikov89@gmail.com> Reported-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202311252109.WgbJsSkG-lkp@intel.com/ Cc: Danilo Krummrich <dakr@redhat.com> Cc: Frank Binns <frank.binns@imgtec.com> Cc: Donald Robson <donald.robson@imgtec.com> Cc: Matt Coster <matt.coster@imgtec.com> Cc: Direct Rendering Infrastructure - Development <dri-devel@lists.freedesktop.org> Fixes: fe375c74 ("drm/sched: Rename priority MIN to LOW") Fixes: 38f922a5 ("drm/sched: Reverse run-queue priority enumeration") Fixes: 5f03a507 ("drm/nouveau: implement 1:1 scheduler - entity relationship") Link: https://patchwork.freedesktop.org/patch/msgid/20231125192246.87268-2-ltuikov89@gmail.comReviewed-by:
Christian König <christian.koenig@amd.com> Link: https://lore.kernel.org/r/7429262c-6dea-4dcc-bf7e-54d2277dabf1@amd.com
-
Rajneesh Bhardwaj authored
Try to allocate system memory on the NUMA node the device is closest to and try to run delayed_delete workers on a CPU of this node as well. To optimize the memory clearing operation when a TTM BO gets freed by the delayed_delete worker, scheduling it closer to a NUMA node where the memory was initially allocated helps avoid the cases where the worker gets randomly scheduled on the CPU cores that are across interconnect boundaries such as xGMI, PCIe etc. This change helps USWC GTT allocations on NUMA systems (dGPU) and AMD APU platforms such as GFXIP9.4.3. Signed-off-by:
Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Acked-by:
Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by:
-
- 25 Nov, 2023 3 commits
-
-
Luben Tuikov authored
Reverse run-queue priority enumeration such that the higest priority is now 0, and for each consecutive integer the prioirty diminishes. Run-queues correspond to priorities. To an external observer a scheduler created with a single run-queue, and another created with DRM_SCHED_PRIORITY_COUNT number of run-queues, should always schedule sched->sched_rq[0] with the same "priority", as that index run-queue exists in both schedulers, i.e. a scheduler with one run-queue or many. This patch makes it so. In other words, the "priority" of sched->sched_rq[n], n >= 0, is the same for any scheduler created with any allowable number of run-queues (priorities), 0 to DRM_SCHED_PRIORITY_COUNT. Cc: Rob Clark <robdclark@gmail.com> Cc: Abhinav Kumar <quic_abhinavk@quicinc.com> Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Cc: Danilo Krummrich <dakr@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by:
-
Luben Tuikov authored
Rename DRM_SCHED_PRIORITY_MIN to DRM_SCHED_PRIORITY_LOW. This mirrors DRM_SCHED_PRIORITY_HIGH, for a list of DRM scheduler priorities in ascending order, DRM_SCHED_PRIORITY_LOW, DRM_SCHED_PRIORITY_NORMAL, DRM_SCHED_PRIORITY_HIGH, DRM_SCHED_PRIORITY_KERNEL. Cc: Rob Clark <robdclark@gmail.com> Cc: Abhinav Kumar <quic_abhinavk@quicinc.com> Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Cc: Danilo Krummrich <dakr@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by:
-
Luben Tuikov authored
If we're given a malformed entity in drm_sched_entity_init()--shouldn't happen, but we verify--with out-of-bounds priority value, we set it to an allowed value. Fix the expression which sets this limit. Signed-off-by:
-
- 24 Nov, 2023 17 commits
-
-
Danilo Krummrich authored
Make use of the scheduler's credit limit and scheduler job's credit count to account for the actual size of a job, such that we fill up the ring efficiently. Signed-off-by:
Danilo Krummrich <dakr@redhat.com> Reviewed-by:
Dave Airlie <airlied@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231114002728.3491-2-dakr@redhat.com
-
Danilo Krummrich authored
Recent patches to the DRM scheduler [1][2] allow for a variable number of run-queues and add support for (shared) workqueues rather than dedicated kthreads per scheduler. This allows us to create a 1:1 relationship between a GPU scheduler and a scheduler entity, in order to properly support firmware schedulers being able to handle an arbitrary amount of dynamically allocated command ring buffers. This perfectly matches Nouveau's needs, hence make use of it. Topology wise we create one scheduler instance per client (handling VM_BIND jobs) and one scheduler instance per channel (handling EXEC jobs). All channel scheduler instances share a workqueue, but every client scheduler instance has a dedicated workqueue. The latter is required to ensure that for VM_BIND job's free_job() work and run_job() work can always run concurrently and hence, free_job() work can never stall run_job() work. For EXEC jobs we don't have this requirement, since EXEC job's free_job() does not require to take any locks which indirectly or directly are held for allocations elsewhere. [1] https://lore.kernel.org/all/8f53f7ef-7621-4f0b-bdef-d8d20bc497ff@redhat.com/T/ [2] https://lore.kernel.org/all/20231031032439.1558703-1-matthew.brost@intel.com/T/Signed-off-by:
-
Danilo Krummrich authored
GPUVM provides common infrastructure to track external and evicted GEM objects as well as locking and validation helpers. Especially external and evicted object tracking is a huge improvement compared to the current brute force approach of iterating all mappings in order to lock and validate the GPUVM's GEM objects. Hence, make us of it. Signed-off-by:
-
Javier Martinez Canillas authored
Currently, only damage tracking for frame damage is supported. If a driver needs to do buffer damage (e.g: the framebuffer attached to plane's state has changed since the last page-flip), the damage helpers just fallback to a full plane update. Add en entry in the TODO about implementing buffer age or any other damage accumulation algorithm for buffer damage handling. Suggested-by:
Simon Ser <contact@emersion.fr> Signed-off-by:
Javier Martinez Canillas <javierm@redhat.com> Reviewed-by:
Thomas Zimmermann <tzimmermann@suse.de> Reviewed-by:
Zack Rusin <zackr@vmware.com> Acked-by:
Sima Vetter <daniel.vetter@ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20231123221315.3579454-6-javierm@redhat.com
-
Javier Martinez Canillas authored
The "Damage Tracking Properties" section in the documentation doesn't have info about the two type of damage handling: frame damage vs buffer damage. Add it to the section and mention that helpers only support frame damage, and how drivers handling buffer damage can indicate that the damage clips should be ignored. Also add references to further documentation about the two damage types. Suggested-by:
-
Javier Martinez Canillas authored
The driver does per-buffer uploads and needs to force a full plane update if the plane's attached framebuffer has change since the last page-flip. Suggested-by:
-
Javier Martinez Canillas authored
The driver does per-buffer uploads and needs to force a full plane update if the plane's attached framebuffer has change since the last page-flip. Fixes: 01f05940 ("drm/virtio: Enable fb damage clips property for the primary plane") Cc: <stable@vger.kernel.org> # v6.4+ Reported-by:
nerdopolis <bluescreen_avenger@verizon.net> Closes: https://bugzilla.kernel.org/show_bug.cgi?id=218115Suggested-by:
-
Javier Martinez Canillas authored
It allows drivers to set a struct drm_plane_state .ignore_damage_clips in their plane's .atomic_check callback, as an indication to damage helpers such as drm_atomic_helper_damage_iter_init() that the damage clips should be ignored. To be used by drivers that do per-buffer (e.g: virtio-gpu) uploads (rather than per-plane uploads), since these type of drivers need to handle buffer damages instead of frame damages. That way, these drivers could force a full plane update if the framebuffer attached to a plane's state has changed since the last update (page-flip). Fixes: 01f05940 ("drm/virtio: Enable fb damage clips property for the primary plane") Cc: <stable@vger.kernel.org> # v6.4+ Reported-by:
-
Michael Banack authored
To clarify the intent and reasoning behind the hotspot properties introduce userspace documentation that goes over cursor handling in para-virtualized environments. The documentation is generic enough to not special case for any specific hypervisor and should apply equally to all. Signed-off-by:
Pekka Paalanen <pekka.paalanen@collabora.com> Reviewed-by:
Michael Banack <banackm@vmware.com> Signed-off-by:
-
Zack Rusin authored
Virtualized drivers place additional restrictions on the cursor plane which breaks the contract of universal planes. To allow atomic modesettings with virtualized drivers the clients need to advertise that they're capable of dealing with those extra restrictions. To do that introduce DRM_CLIENT_CAP_CURSOR_PLANE_HOTSPOT which lets DRM know that the client is aware of and capable of dealing with the extra restrictions on the virtual cursor plane. Setting this option to true makes DRM expose the cursor plane on virtualized drivers. The userspace is expected to set the hotspots and handle mouse events on that plane. Signed-off-by:
-
Zack Rusin authored
Atomic modesetting supports mouse cursor offsets via the hotspot properties that are created on cursor planes. All drivers which support hotspots are atomic and the legacy code has been implemented in terms of the atomic properties as well. Due to the above the lagacy cursor hotspot code is no longer used or needed and can be removed. Signed-off-by:
-
Zack Rusin authored
Atomic modesetting got support for mouse hotspots via the hotspot properties. Port the legacy kms hotspot handling to the new properties on cursor planes. Signed-off-by:
Gerd Hoffmann <kraxel@redhat.com> Cc: David Airlie <airlied@linux.ie> Cc: Gurchetan Singh <gurchetansingh@chromium.org> Cc: Chia-I Wu <olvaffe@gmail.com> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: virtualization@lists.linux-foundation.org Reviewed-by:
-
Zack Rusin authored
Atomic modesetting got support for mouse hotspots via the hotspot properties. Port the legacy kms hotspot handling to the new properties on cursor planes. Signed-off-by:
-
Zack Rusin authored
Atomic modesetting got support for mouse hotspots via the hotspot properties. Port the legacy kms hotspot handling to the new properties on cursor planes. Signed-off-by:
-
Zack Rusin authored
Atomic modesetting got support for mouse hotspots via the hotspot properties. Port the legacy kms hotspot handling to the new properties on cursor planes. Signed-off-by:
Martin Krastev <krastevm@vmware.com> Signed-off-by:
-
Zack Rusin authored
Atomic modesetting code lacked support for specifying mouse cursor hotspots. The legacy kms DRM_IOCTL_MODE_CURSOR2 had support for setting the hotspot but the functionality was not implemented in the new atomic paths. Due to the lack of hotspots in the atomic paths userspace compositors completely disable atomic modesetting for drivers that require it (i.e. all paravirtualized drivers). This change adds hotspot properties to the atomic codepaths throughtout the DRM core and will allow enabling atomic modesetting for virtualized drivers in the userspace. Signed-off-by:
-
Zack Rusin authored
Cursor planes on virtualized drivers have special meaning and require that the clients handle them in specific ways, e.g. the cursor plane should react to the mouse movement the way a mouse cursor would be expected to and the client is required to set hotspot properties on it in order for the mouse events to be routed correctly. This breaks the contract as specified by the "universal planes". Fix it by disabling the cursor planes on virtualized drivers while adding a foundation on top of which it's possible to special case mouse cursor planes for clients that want it. Disabling the cursor planes makes some kms compositors which were broken, e.g. Weston, fallback to software cursor which works fine or at least better than currently while having no effect on others, e.g. gnome-shell or kwin, which put virtualized drivers on a deny-list when running in atomic context to make them fallback to legacy kms and avoid this issue. Signed-off-by:
-
- 23 Nov, 2023 18 commits
-
-
Simon Ser authored
This new kernel capability indicates whether async page-flips are supported via the atomic uAPI. DRM clients can use it to check for support before feeding DRM_MODE_PAGE_FLIP_ASYNC to the kernel. Make it clear that DRM_CAP_ASYNC_PAGE_FLIP is for legacy uAPI only. Signed-off-by:
André Almeida <andrealmeid@igalia.com> Reviewed-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
-
Simon Ser authored
If the driver supports it, allow user-space to supply the DRM_MODE_PAGE_FLIP_ASYNC flag to request an async page-flip. Set drm_crtc_state.async_flip accordingly. Document that drivers will reject atomic commits if an async flip isn't possible. This allows user-space to fall back to something else. For instance, Xorg falls back to a blit. Another option is to wait as close to the next vblank as possible before performing the page-flip to reduce latency. Signed-off-by:
-
André Almeida authored
Given that prop changes may lead to modesetting, which would defeat the fast path of the async flip, refuse any atomic prop change for async flips in atomic API. The only exception is the framebuffer ID to flip to. Currently the only plane type supported is the primary one. Signed-off-by:
-
Sarah Walker authored
Add documentation for the UAPI. Changes since v5: - Remove obsolete VM documentation Co-developed-by:
Matt Coster <matt.coster@imgtec.com> Signed-off-by:
Donald Robson <donald.robson@imgtec.com> Signed-off-by:
Sarah Walker <sarah.walker@imgtec.com> Reviewed-by:
Maxime Ripard <mripard@kernel.org> Link: https://lore.kernel.org/r/76a7b18cfbe93066efcee3311ae795176ce7c65d.1700668843.git.donald.robson@imgtec.comSigned-off-by:
-
Sarah Walker authored
Firmware trace is exposed at /sys/debug/dri/<dev_nr>/pvr_fw/trace_0. Trace is enabled via the group mask at /sys/debug/dri/<dev_nr>/pvr_params/fw_trace_mask. Changes since v8: - Corrected license identifiers Changes since v3: - Use drm_dev_{enter,exit} Co-developed-by: -
Sarah Walker authored
Changes since v8: - Corrected license identifiers Changes since v5: - Split up header commit due to size Signed-off-by:
-
Sarah Walker authored
Implement job submission ioctl. Job scheduling is implemented using drm_sched. Jobs are submitted in a stream format. This is intended to allow the UAPI data format to be independent of the actual FWIF structures in use, which vary depending on the GPU in use. The stream formats are documented at: https://gitlab.freedesktop.org/mesa/mesa/-/blob/f8d2b42ae65c2f16f36a43e0ae39d288431e4263/src/imagination/csbgen/rogue_kmd_stream.xml Changes since v8: - Updated for upstreamed DRM scheduler changes - Removed workaround code for the pending_list previously being updated after run_job() returned - Fixed null deref in pvr_queue_cleanup_fw_context() for bad stream ptr given to create_context ioctl - Corrected license identifiers Changes since v7: - Updated for v8 "DRM scheduler changes for XE" patchset Changes since v6: - Fix fence handling in pvr_sync_signal_array_add() - Add handling for SUBMIT_JOB_FRAG_CMD_DISABLE_PIXELMERGE flag - Fix missing dma_resv locking in job submit path Changes since v5: - Fix leak in job creation error path Changes since v4: - Use a regular workqueue for job scheduling Changes since v3: - Support partial render jobs - Add job timeout handler - Split sync handling out of job code - Use drm_dev_{enter,exit} Changes since v2: - Use drm_sched for job scheduling Co-developed-by:
Boris Brezillon <boris.brezillon@collabora.com> Signed-off-by:
-
Sarah Walker authored
Implement ioctls for the creation and destruction of contexts. Contexts are used for job submission and each is associated with a particular job type. Changes since v8: - Fixed one error path in pvr_stream_process_1() - Corrected license identifiers Changes since v5: - Fix context release in final error path in pvr_context_create() Changes since v3: - Use drm_dev_{enter,exit} Co-developed-by: -
Sarah Walker authored
Implement ioctls to create and destroy free lists and HWRT datasets. Free lists are used for GPU-side memory allocation during geometry processing. HWRT datasets are the FW-side structures representing render targets. Changes since v8: - Corrected license identifiers Changes since v6: - Fix out-of-bounds shift in get_cr_multisamplectl_val() Changes since v4: - Remove use of drm_gem_shmem_get_pages() Changes since v3: - Support free list grow requests from FW - Use drm_dev_{enter,exit} Co-developed-by: -
Sarah Walker authored
Add support for the MIPS firmware processor, used in the Series AXE GPU. The MIPS firmware processor uses a separate MMU to the rest of the GPU, so this patch adds support for that as well. Changes since v8: - Corrected license identifiers Changes since v6: - Fix integer overflow in VM map error path Changes since v5: - Use alloc_page() when allocating MIPS pagetable Changes since v3: - Get regs resource (removed from GPU resources commit) Signed-off-by:
-
Sarah Walker authored
The infrastructure includes parsing of the firmware image, initialising FW-side structures, handling the kernel and firmware command ringbuffers and starting & stopping the firmware processor. This patch also adds the necessary support code for the META firmware processor. Changes since v8: - Fix documentation for pvr_fwccb_process() - Corrected license identifiers Changes since v6: - Add a minimum retry count to pvr_kccb_reserve_slot_sync() Changes since v5: - Add workaround for BRN 71242 - Attempt to recover GPU on MMU flush command failure Changes since v4: - Remove use of drm_gem_shmem_get_pages() - Remove interrupt resource name Changes since v3: - Hard reset FW processor on watchdog timeout - Switch to threaded IRQ - Rework FW object creation/initialisation to aid hard reset - Added MODULE_FIRMWARE() - Use drm_dev_{enter,exit} Signed-off-by: -
Sarah Walker authored
Add power management to the driver, using runtime pm. The power off sequence depends on firmware commands which are not implemented in this patch. Changes since v8: - Corrected license identifiers Changes since v5: - Use RUNTIME_PM_OPS() to declare PM callbacks - Add Kconfig dependency on CONFIG_PM Changes since v4: - Suspend runtime PM before unplugging device on rmmod Changes since v3: - Don't power device when calling pvr_device_gpu_fini() - Documentation for pvr_dev->lost has been improved - pvr_power_init() renamed to pvr_watchdog_init() - Use drm_dev_{enter,exit} Changes since v2: - Use runtime PM - Implement watchdog Signed-off-by: -
Donald Robson authored
Add a GEM implementation based on drm_gem_shmem, and support code for the PowerVR GPU MMU. The GPU VA manager is used for address space management. Changes since v8: - Updated for changes to drm_gpuvm - Switched to dma_resv locking for vm ops - Removed linked lists for collecting BOs in vm_context and for freeing after ops. This is now done internally in drm_gpuvm - Corrected license identifiers Changes since v7: - kernel-doc fixes - Remove prefixes from DRM_PVR_BO_* flags - CREATE_BO ioctl now returns an error if provided size isn't page aligned - Optimised MMU flushes Changes since v6: - Don't initialise kernel_vm_ctx when using MIPS firmware processor - Rename drm_gpuva_manager uses to drm_gpuvm - Sync GEM object to device on creation Changes since v5: - Use WRITE_ONCE() when writing to page tables - Add memory barriers to page table insertion - Fixed double backing page alloc on page table objects - Fix BO mask checks in DRM_IOCTL_PVR_CREATE_BO handler - Document use of pvr_page_table_*_idx when preallocing page table objs - Remove pvr_vm_gpuva_mapping_init() - Remove NULL check for unmap op in remap function - Protect gem object with mutex during drm_gpuva_link/unlink - Defer free or release of page table pages until after TLB flush - Use drm_gpuva_op_remap_get_unmap_range() helper Changes since v4: - Correct sync function in vmap/vunmap function documentation - Update for upstream GPU VA manager - Fix missing frees when unmapping drm_gpuva objects - Always zero GEM BOs on creation Changes since v3: - Split MMU and VM code - Register page table allocations with kmemleak - Use drm_dev_{enter,exit} Changes since v2: - Use GPU VA manager - Use drm_gem_shmem Co-developed-by: -
Sarah Walker authored
Read the GPU ID register at probe time and select the correct features/quirks/enhancements. Use the GPU ID to form the firmware file name and load the firmware. The features/quirks/enhancements arrays are currently hardcoded in the driver for the supported GPUs. We are looking at moving this information to the firmware image. Changes since v8: - Corrected license identifiers Changes since v7: - Fix kerneldoc for pvr_device_info_set_enhancements() Changes since v5: - Add BRN 71242 to device info Changes since v4: - Retrieve device information from firmware header - Pull forward firmware header parsing from FW infrastructure patch - Use devm_add_action_or_reset to release firmware Changes since v3: - Use drm_dev_{enter,exit} Co-developed-by:Frank Binns <frank.binns@imgtec.com> Signed-off-by:
-
Sarah Walker authored
Changes since v8: - Corrected license identifiers Changes since v7: - Add padding to struct rogue_fwif_ccb_ctl to place read and write offsets in different cache lines Changes since v5: - Split up header commit due to size - Add BRN 71242 to device info Changes since v4: - Add FW header device info Signed-off-by:
-
Sarah Walker authored
Changes since v8: - Corrected license identifiers Changes since v5: - Split up header commit due to size Signed-off-by:
-
Sarah Walker authored
Changes since v8: - Corrected license identifiers Changes since v5: - Split up header commit due to size Signed-off-by:
-
Sarah Walker authored
Acquire clock and register resources, and enable/map as appropriate. Changes since v8: - Corrected license identifiers Changes since v3: - Remove regulator resource (not used on supported platform) - Use devm helpers - Use devm_clk_get_optional() for optional clocks - Don't prepare clocks on resource acquisition - Drop pvr_device_clk_core_get_freq() helper - Drop pvr_device_reg_fini() - Drop NULLing of clocks in pvr_device_clk_init() - Use dev_err_probe() on clock acquisition failure - Remove PVR_CR_READ/WRITE helper macros - Improve documentation for GPU clocks - Remove regs resource (not used in this commit) Co-developed-by:
-
