- 25 Mar, 2023 12 commits
-
-
Rob Clark authored
Avoid allocation under idr_lock, to prevent deadlock against the job_free() path (which runs on same thread as job_run(), which makes it also part of the fence-signaling path. Signed-off-by:
Rob Clark <robdclark@chromium.org> Patchwork: https://patchwork.freedesktop.org/patch/527847/ Link: https://lore.kernel.org/r/20230320144356.803762-12-robdclark@gmail.com
-
Rob Clark authored
Needed to idr_preload() which returns with preemption disabled. Signed-off-by:
-
Rob Clark authored
Now that everything that controls which LRU an obj lives in *except* the backing pages is protected by the LRU lock, add a special path to unpin in the job_run() path, where we are assured that we already have backing pages and will not be racing against eviction (because the GEM object's dma_resv contains the fence that will be signaled when the submit/job completes). Signed-off-by:
-
Rob Clark authored
Since the LRU lock is already acquired when moving an obj between LRUs, we can use it to protect pin_count and madv, without any significant change in locking (ie. it just expands the scope of the lock by a hand- ful of instructions). This prepares the way to decrement the pin_count in the job_run() path without needing to hold the obj lock, to avoid a potential deadlock (or rather stall) caused by the fence-signaling path (job_run()) blocking on shrinker/reclaim. (Only a stall because the wait for fence signaling wait_for_idle() is not infinite.) Signed-off-by:
-
Rob Clark authored
Just code-motion. Signed-off-by:
-
Rob Clark authored
Export the locked version or lru's move_tail(). Signed-off-by:
-
Rob Clark authored
vmap'ing is just pinning in disguise. So treat it as such and simplify the LRU tracking. Signed-off-by:
-
Rob Clark authored
We need to use the inuse count to track that a BO is pinned until we have the hw_fence. But we want to remove the obj lock from the job_run() path as this could deadlock against reclaim/shrinker (because it is blocking the hw_fence from eventually being signaled). So split that tracking out into a per-vma lock with narrower scope. Signed-off-by:
-
Rob Clark authored
Stop open coding VMA construction, which will be needed in the next commit. And since the VMA already has a ptr to the adress space, stop passing that around everywhere. (Also, an aspace always has an mmu so we can drop a couple pointless NULL checks.) Signed-off-by:
-
Rob Clark authored
The flags are only accessed (1) when submit is constructed, before enqueuing to gpu sched (ie. when still visible to only the task calling the submit ioctl), (2) here, where we own a reference to the submit and are serialized on the gpu sched thread, and (3) after the submit is retired and last reference is dropped, which is serialized on the submit's reference count. Hence locking is unneeded here. Signed-off-by:
-
Rob Clark authored
Avoid allocating memory in job_run() by pre-allocating the hw_fence. Signed-off-by:
-
Arnd Bergmann authored
Selecting CONFIG_PM_GENERIC_DOMAINS causes a build failure when CONFIG_PM is not enabled: WARNING: unmet direct dependencies detected for PM_GENERIC_DOMAINS Depends on [n]: PM [=n] Selected by [m]: - DRM_MSM [=m] && HAS_IOMEM [=y] && DRM [=m] && (ARCH_QCOM [=y] || SOC_IMX5 || COMPILE_TEST [=y]) && COMMON_CLK [=y] && IOMMU_SUPPORT [=y] && (QCOM_OCMEM [=y] || QCOM_OCMEM [=y]=n) && (QCOM_LLCC [=n] || QCOM_LLCC [=n]=n) && (QCOM_COMMAND_DB [=y] || QCOM_COMMAND_DB [=y]=n) && DEVFREQ_GOV_SIMPLE_ONDEMAND [=y] drivers/base/power/domain.c:654:13: error: use of undeclared identifier 'pm_wq' queue_work(pm_wq, &genpd->power_off_work); ^ drivers/base/power/domain.c:853:26: error: no member named 'ignore_children' in 'struct dev_pm_info' if (!dev || dev->power.ignore_children) ~~~~~~~~~~ ^ Fixes: c11fa120 ("drm/msm/a6xx: Use genpd notifier to ensure cx-gdsc collapse") Signed-off-by:Arnd Bergmann <arnd@arndb.de> Reviewed-by:
Ulf Hansson <ulf.hansson@linaro.org> Patchwork: https://patchwork.freedesktop.org/patch/528597/ Link: https://lore.kernel.org/r/20230324095502.3289094-1-arnd@kernel.orgSigned-off-by:
-
- 21 Mar, 2023 1 commit
-
-
Rob Clark authored
It's been a bit overdue. Regen headers to pull in a2xx perfcntr updates, etc. Signed-off-by:
-
- 20 Mar, 2023 14 commits
-
-
Rob Clark authored
The next generated header update will drop the _LO/_HI suffix, now that the userspace tooling properly understands 64b vs 32b regs (and the _LO/ _HI workarounds are getting cleaned up). So convert to using the 64b reg helpers in prep. Signed-off-by:
Konrad Dybcio <konrad.dybcio@linaro.org> Patchwork: https://patchwork.freedesktop.org/patch/527923/ Link: https://lore.kernel.org/r/20230320185416.938842-1-robdclark@gmail.com
-
Adam Skladowski authored
Downstream driver appears to not support preemption on A510 target, trying to use one make device slow and fill log with rings related errors. Set num_rings to 1 to disable preemption. Suggested-by:
Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Fixes: e20c9284 ("drm/msm/adreno: Add support for Adreno 510 GPU") Signed-off-by:
Adam Skladowski <a39.skl@gmail.com> Reviewed-by:
-
Abhinav Kumar authored
Update the URI for MSM DRM bugs for users to be able to file bugs at a centralized location. Signed-off-by:
Abhinav Kumar <quic_abhinavk@quicinc.com> Acked-by:
Rob Clark <robdclark@gmail.com> Patchwork: https://patchwork.freedesktop.org/patch/525026/ Link: https://lore.kernel.org/r/1677972416-7353-1-git-send-email-quic_abhinavk@quicinc.comSigned-off-by:
-
Johan Hovold authored
Clean up the component ops initialisers which were indented one level too far. Signed-off-by:
Johan Hovold <johan+linaro@kernel.org> Patchwork: https://patchwork.freedesktop.org/patch/524973/ Link: https://lore.kernel.org/r/20230303164807.13124-5-johan+linaro@kernel.orgSigned-off-by:
-
Johan Hovold authored
The runtime PM status can only be updated while runtime PM is disabled. Drop the bogus pm_runtime_set_active() call that was made after enabling runtime PM and which (incidentally but correctly) left the runtime PM status set to 'suspended'. Fixes: 2c087a33 ("drm/msm/adreno: Load the firmware before bringing up the hardware") Signed-off-by:
-
Johan Hovold authored
A recent commit moved enabling of runtime PM to GPU load time (first open()) but failed to update the error paths so that runtime PM is disabled if initialisation of the GPU fails. This would trigger a warning about the unbalanced disable count on the next open() attempt. Note that pm_runtime_put_noidle() is sufficient to balance the usage count when pm_runtime_put_sync() fails (and is chosen over pm_runtime_resume_and_get() for consistency reasons). Fixes: 4b18299b ("drm/msm/adreno: Defer enabling runpm until hw_init()") Cc: stable@vger.kernel.org # 6.0 Signed-off-by:
-
Konrad Dybcio authored
Add the dev_pm_opp_of_find_icc_paths() call to let the OPP framework handle bus voting as part of power level setting. Reviewed-by:
-
Konrad Dybcio authored
Add support for gpu_busy on a4xx, which is required for devfreq support. Reviewed-by:
-
Konrad Dybcio authored
Add support for gpu_busy on a3xx, which is required for devfreq support. Tested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> #ifc6410 Reviewed-by:
-
Konrad Dybcio authored
Some older GPUs (namely a2xx with no opp tables at all and a320 with downstream-remnants gpu pwrlevels) used not to have OPP tables. They both however had just one frequency defined, making it extremely easy to construct such an OPP table from within the driver if need be. Do so and switch all clk_set_rate calls on core_clk to their OPP counterparts. Reviewed-by:
-
Akhil P Oommen authored
As per the recommended recovery sequence of adreno gpu, cx gdsc should collapse at hardware before it is turned back ON. This helps to clear out the stale states in hardware before it is reinitialized. Use the genpd notifier along with the newly introduced dev_pm_genpd_synced_poweroff() api to ensure that cx gdsc has collapsed before we turn it back ON. Signed-off-by:
Akhil P Oommen <quic_akhilpo@quicinc.com> Reviewed-by:
-
Akhil P Oommen authored
Remove the unused 'reset' interface which was supposed to help to ensure that cx gdsc has collapsed during gpu recovery. This is was not enabled so far due to missing gpucc driver support. Similar functionality using genpd framework will be implemented in the upcoming patch. This effectively reverts commit 1f6cca40 ("drm/msm/a6xx: Ensure CX collapse during gpu recovery"). Signed-off-by:
Philipp Zabel <p.zabel@pengutronix.de> Patchwork: https://patchwork.freedesktop.org/patch/516470/ Link: https://lore.kernel.org/r/20230102161757.v5.4.I96e0bf9eaf96dd866111c1eec8a4c9b70fd7cbcb@changeidSigned-off-by:
-
Akhil P Oommen authored
When a device has multiple power domains, dev->power_domain is left empty during probe. That didn't cause any issue so far because we are freeloading on smmu driver's vote on cx gdsc. Instead of that, create a device_link between cx genpd device and gmu device to keep a vote from gpu driver. Before this patch: localhost ~ # cat /sys/kernel/debug/pm_genpd/pm_genpd_summary gx_gdsc on 0 /devices/genpd:1:3d6a000.gmu active 0 cx_gdsc on 0 /devices/platform/soc@0/3da0000.iommu active 0 After this patch: localhost ~ # cat /sys/kernel/debug/pm_genpd/pm_genpd_summary gx_gdsc on 0 /devices/genpd:1:3d6a000.gmu active 0 cx_gdsc on 0 /devices/platform/soc@0/3da0000.iommu active 0 /devices/genpd:0:3d6a000.gmu active 0 Signed-off-by: -
Rob Clark authored
Merge drm-next into msm-next to pick up external clk and PM dependencies for improved a6xx GPU reset sequence. Signed-off-by:
-
- 15 Mar, 2023 1 commit
-
-
git://anongit.freedesktop.org/drm/drm-intelDave Airlie authored
Cross-subsystem Changes: - MEI patches to fix suspend/resume issues with the i915's PXP. (Alexander) Driver Changes: - Registers helpers and clean-ups. (Lucas) - PXP fixes and clean-ups. (Alan, Alexander) - CDCLK related fixes and w/a (Chaitanya, Stanislav) - Move display code to use RMW whenever possible (Andrzej) - PSR fixes (Jouni, Ville) - Implement async_flip mode per plane tracking (Andrzej) - Remove pre-production Workarounds (Matt) - HDMI related fixes (Ankit) - LVDS cleanup (Ville) - Watermark fixes and cleanups (Ville, Jani, Stanilav) - DMC code related fixes, cleanups and improvements (Jani) - Implement fb_dirty for PSR,FBC,DRRS fixes (Jouni) - Initial DSB improvements targeting LUTs loading (Ville) - HWMON related fixes (Ashutosh) - PCI ID updates (Jonathan, Matt Roper) - Fix leak in scatterlist (Matt Atwood) - Fix eDP+DSI dual panel systems (Ville) - Cast iomem to avoid sparese warnings (Jani) - Set default backlight controller index (Jani) - More MTL enabling (RK) - Conversion of display dev_priv towards i915 (Nirmoy) - Improvements in log/debug messages (Ville) - Increase slice_height for DP VDSC (Suraj) - VBT ports improvements (Ville) - Fix platforms without Display (Imre) - Other generic display code clean-ups (Ville, Jani, Rodrigo) - Add RPL-U sub platform (Chaitanya) - Add inverted backlight quirk for HP 14-r206nv (Mavroudis) - Transcoder timing improvements (Ville) - Track audio state per-transcoder (Ville) - Error/underrun interrupt fixes (Ville) - Update combo PHY init sequence (Matt Roper) - Get HDR DPCD refresh timeout (Ville) - Vblank improvements (Ville) - DSS fixes and cleanups (Jani) - PM code cleanup (Jani) - Split display parts related to RPS (Jani) Signed-off-by:
Dave Airlie <airlied@redhat.com> From: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/ZAez4aekcob8fTeh@intel.com
-
- 14 Mar, 2023 1 commit
-
-
git://anongit.freedesktop.org/drm/drm-miscDave Airlie authored
drm-misc-next for v6.4-rc1: Note: Only changes since pull request from 2023-02-23 are included here. UAPI Changes: - Convert rockchip bindings to YAML. - Constify kobj_type structure in dma-buf. - FBDEV cmdline parser fixes, and other small fbdev fixes for mode parsing. Cross-subsystem Changes: - Add Neil Armstrong as linaro maintainer. - Actually signal the private stub dma-fence. Core Changes: - Add function for adding syncobj dep to sched_job and use it in panfrost, v3d. - Improve DisplayID 2.0 topology parsing and EDID parsing in general. - Add a gem eviction function and callback for generic GEM shrinker purposes. - Prepare to convert shmem helper to use the GEM reservation lock instead of own locking. (Actual commit itself got reverted for now) - Move the suballocator from radeon and amdgpu drivers to core in preparation for Xe. - Assorted small fixes and documentation. - Fixes to HPD polling. - Assorted small fixes in simpledrm, bridge, accel, shmem-helper, and the selftest of format-helper. - Remove dummy resource when ttm bo is created, and during pipelined gutting. Fix all drivers to accept a NULL ttm_bo->resource. - Handle pinned BO moving prevention in ttm core. - Set drm panel-bridge orientation before connector is registered. - Remove dumb_destroy callback. - Add documentation to GEM_CLOSE, PRIME_HANDLE_TO_FD, PRIME_FD_TO_HANDLE, GETFB2 ioctl's. - Add atomic enable_plane callback, use it in ast, mgag200, tidss. Driver Changes: - Use drm_gem_objects_lookup in vc4. - Assorted small fixes to virtio, ast, bridge/tc358762, meson, nouveau. - Allow virtio KMS to be disabled and compiled out. - Add Radxa 8/10HD, Samsung AMS495QA01 panels. - Fix ivpu compiler errors. - Assorted fixes to drm/panel, malidp, rockchip, ivpu, amdgpu, vgem, nouveau, vc4. - Assorted cleanups, simplifications and fixes to vmwgfx. Signed-off-by:
-
- 13 Mar, 2023 11 commits
-
-
Vinod Polimera authored
For the PSR to kick in, self_refresh_aware has to be set. Initialize it based on the PSR support for the eDP interface. Signed-off-by:
Vinod Polimera <quic_vpolimer@quicinc.com> Reviewed-by:
-
Vinod Polimera authored
Populate the enocder software structure to reflect the updated crtc appropriately during crtc enable/disable for a new commit while taking care of the self refresh transitions when crtc disable is triggered from the drm self refresh library. Signed-off-by:
-
Vinod Polimera authored
Enable PSR on eDP interface using drm self-refresh librabry. This patch uses a trigger from self-refresh library to enter/exit into PSR, when there are no updates from framework. Signed-off-by:
Kalyan Thota <quic_kalyant@quicinc.com> Signed-off-by:
-
Vinod Polimera authored
Use atomic variants for encoder callback functions such that certain states like self-refresh can be accessed as part of enable/disable sequence. Signed-off-by:
-
Vinod Polimera authored
The eDP and DP interfaces shared the bridge operations and the eDP specific changes were implemented under is_edp check. To add psr support for eDP, we started using a new set of eDP bridge ops. We are moving the eDP specific code in the dp_bridge_mode_valid function to a new eDP function, edp_bridge_mode_valid under the eDP bridge ops. Signed-off-by:
Sankeerth Billakanti <quic_sbillaka@quicinc.com> Signed-off-by:
-
Vinod Polimera authored
Add support for basic panel self refresh (PSR) feature for eDP. Add a new interface to set PSR state in the sink from DPU. Program the eDP controller to issue PSR enter and exit SDP to the sink. Signed-off-by:
-
Vinod Polimera authored
Use atomic variants for DP bridge callback functions so that the atomic state can be accessed in the interface drivers. The atomic state will help the driver find out if the display is in self refresh state. Signed-off-by:
Douglas Anderson <dianders@chromium.org> Patchwork: https://patchwork.freedesktop.org/patch/524731/ Link: https://lore.kernel.org/r/1677774797-31063-9-git-send-email-quic_vpolimer@quicinc.comSigned-off-by:
-
Vinod Polimera authored
Reset the datapath after disabling the timing gen, such that it can start on a clean slate when the intf is enabled back. This was a recommended sequence from the DPU HW programming guide. Signed-off-by:
-
Vinod Polimera authored
There can be a race between timing gen disable and vblank irq. The wait post timing gen disable may return early but intf disable sequence might not be completed. Ensure that, intf status is disabled before we retire the function. Signed-off-by:
-
Vinod Polimera authored
Recommended way of reading the interface timing gen status is via status register. Timing gen status register will give a reliable status of the interface especially during ON/OFF transitions. This support was added from DPU version 5.0.0. Signed-off-by:
-
Vinod Polimera authored
According to KMS documentation, The driver must not release any shared resources if active is set to false but enable still true. Fixes: ccc862b9 ("drm/msm/dpu: Fix reservation failures in modeset") Signed-off-by:
-
