drm_atomic_helper_swap_state could previously never fail, in order to still continue it would set a limit of 10s on waiting for previous updates to complete, then it moved forward. This can be very bad when ignoring previous updates, because the hw state and sw state may get out of sync when for example a modeset is ignored.
By converting the swap_state to interruptible and handling the error in each driver, we fix this issue because if a update takes forever it can be aborted through signals. Changes since first version: - Split out driver conversions. - Fix nouveau error handling first. Maarten Lankhorst (12): drm/nouveau: Fix error handling in nv50_disp_atomic_commit drm/atomic: Change drm_atomic_helper_swap_state to return an error. drm/nouveau: Handle drm_atomic_helper_swap_state failure drm/atmel-hlcdc: Handle drm_atomic_helper_swap_state failure drm/i915: Handle drm_atomic_helper_swap_state failure drm/mediatek: Handle drm_atomic_helper_swap_state failure drm/msm: Handle drm_atomic_helper_swap_state failure drm/tegra: Handle drm_atomic_helper_swap_state failure drm/tilcdc: Handle drm_atomic_helper_swap_state failure drm/vc4: Handle drm_atomic_helper_swap_state failure drm/atomic: Add __must_check to drm_atomic_helper_swap_state. drm/atomic: Allow drm_atomic_helper_swap_state to fail drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c | 13 ++++++----- drivers/gpu/drm/drm_atomic_helper.c | 35 +++++++++++++++++----------- drivers/gpu/drm/i915/intel_display.c | 10 +++++++- drivers/gpu/drm/mediatek/mtk_drm_drv.c | 7 +++++- drivers/gpu/drm/msm/msm_atomic.c | 16 ++++++------- drivers/gpu/drm/nouveau/nv50_display.c | 12 +++++++--- drivers/gpu/drm/tegra/drm.c | 7 +++++- drivers/gpu/drm/tilcdc/tilcdc_drv.c | 6 ++++- drivers/gpu/drm/vc4/vc4_kms.c | 2 +- include/drm/drm_atomic_helper.h | 4 ++-- 10 files changed, 75 insertions(+), 37 deletions(-) -- 2.11.0 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
