PF: i915_gem_get_tiling_ioctl returns 0 with tiling_mode=2, swizzle_mode=0, phys_swizzle_mode=0
VF: i915_gem_get_tiling_ioctl returns 0 with tiling_mode=0, swizzle_mode=0, phys_swizzle_mode=0
The tiling_mode is lost on the VF. By disabling these ioctls, users should fallback to default values.
This fixes the blocky artifact when sharing surfaces between VAAPI and OpenCL.
I’m not sure if this is the correct way to do it, but it works.
There may be unknown side effects, so further testing is still needed...
Signed-off-by: bbaa <bbaa@bbaa.fun>
2679bb2f692e3d9b6c20863ba0fd8a793ea24f16 [PATCH 01/12] drm/i915/guc: Check if expecting reply before decrementing outstanding_submission_g2h
ba282cfba33c8cee46382a6045190af3bd687e29 [PATCH 02/12] drm/i915/psr: Fix using wrong mask in REG_FIELD_PREP
798fd944fbcffdfffeef0b7a5cfdfab1c9dbc56b [PATCH 03/12] drm/i915/guc: Handle race condition where wakeref count drops below 0
fcc95344aa1c419d54f46c19011219363f05c593 [PATCH 04/12] drm/i915/pmu: Fix build error with GCOV and AutoFDO enabled
5b930e72aaeaf16dae45eef1aa1ae4b8114d3460 [PATCH 05/12] drm/i915/gem: Allow EXEC_CAPTURE on recoverable contexts on DG1
7952c4bb509259a4c26e35cdc7e5bd7cb4da2882 [PATCH 06/12] Revert "drm/i915/gem: Allow EXEC_CAPTURE on recoverable contexts on DG1"
18ec560e3e9b476e63c671514dcb8b9163a578a2 [PATCH 07/12] drm/i915: fix build error some more
cc0a3fd781bf83d631d988568871808d06046e75 [PATCH 08/12] drm/i915/dsi: Fix off by one in BXT_MIPI_TRANS_VTOTAL
9d4064787d8d11f69b01fd9d08bb7fd248041a56 [PATCH 09/12] drm/i915/selftests: Change mock_request() to return
5a7ae7bebdc4c2ecd48a2c061319956f65c09473 [PATCH 10/12] drm/i915/gt: Fix timeline left held on VMA alloc error
6a17e0d27fbe0dd6f3061aa627e61b4ac7afbb95 [PATCH 11/12] drm/i915/gsc: mei interrupt top half should be in irq disabled context
710deaff6aebd5c23e307f03e62bb02e23989ab7 [PATCH 12/12] drm/i915/dp_mst: Work around Thunderbolt sink disconnect after SINK_COUNT_ESI read
