Merge tag 'drm-for-v4.16' of git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
 "This seems to have been a comparatively quieter merge window, I assume
  due to holidays etc. The "biggest" change is AMD header cleanups, which
  merge/remove a bunch of them. The AMD gpu scheduler is now being made generic
  with the etnaviv driver wanting to reuse the code, hopefully other drivers
  can go in the same direction.

  Otherwise it's the usual lots of stuff in i915/amdgpu, not so much stuff
  elsewhere.

  Core:
   - Add .last_close and .output_poll_changed helpers to reduce driver footprints
   - Fix plane clipping
   - Improved debug printing support
   - Add panel orientation property
   - Update edid derived properties at edid setting
   - Reduction in fbdev driver footprint
   - Move amdgpu scheduler into core for other drivers to use.

  i915:
   - Selftest and IGT improvements
   - Fast boot prep work on IPS, pipe config
   - HW workarounds for Cannonlake, Geminilake
   - Cannonlake clock and HDMI2.0 fixes
   - GPU cache invalidation and context switch improvements
   - Display planes cleanup
   - New PMU interface for perf queries
   - New firmware support for KBL/SKL
   - Geminilake HW workaround for perforamce
   - Coffeelake stolen memory improvements
   - GPU reset robustness work
   - Cannonlake horizontal plane flipping
   - GVT work

  amdgpu/radeon:
   - RV and Vega header file cleanups (lots of lines gone!)
   - TTM operation context support
   - 48-bit GPUVM support for Vega/RV
   - ECC support for Vega
   - Resizeable BAR support
   - Multi-display sync support
   - Enable swapout for reserved BOs during allocation
   - S3 fixes on Raven
   - GPU reset cleanup and fixes
   - 2+1 level GPU page table

  amdkfd:
   - GFX7/8 SDMA user queues support
   - Hardware scheduling for multiple processes
   - dGPU prep work

  rcar:
   - Added R8A7743/5 support
   - System suspend/resume support

  sun4i:
   - Multi-plane support for YUV formats
   - A83T and LVDS support

  msm:
   - Devfreq support for GPU

  tegra:
   - Prep work for adding Tegra186 support
   - Tegra186 HDMI support
   - HDMI2.0 and zpos support by using generic helpers

  tilcdc:
   - Misc fixes

  omapdrm:
   - Support memory bandwidth limits
   - DSI command mode panel cleanups
   - DMM error handling

  exynos:
   - drop the old IPP subdriver.

  etnaviv:
   - Occlusion query fixes
   - Job handling fixes
   - Prep work for hooking in gpu scheduler

  armada:
   - Move closer to atomic modesetting
   - Allow disabling primary plane if overlay is full screen

  imx:
   - Format modifier support
   - Add tile prefetch to PRE
   - Runtime PM support for PRG

  ast:
   - fix LUT loading"

* tag 'drm-for-v4.16' of git://people.freedesktop.org/~airlied/linux: (1471 commits)
  drm/ast: Load lut in crtc_commit
  drm: Check for lessee in DROP_MASTER ioctl
  drm: fix gpu scheduler link order
  drm/amd/display: Demote error print to debug print when ATOM impl missing
  dma-buf: fix reservation_object_wait_timeout_rcu once more v2
  drm/amdgpu: Avoid leaking PM domain on driver unbind (v2)
  drm/amd/amdgpu: Add Polaris version check
  drm/amdgpu: Reenable manual GPU reset from sysfs
  drm/amdgpu: disable MMHUB power gating on raven
  drm/ttm: Don't unreserve swapped BOs that were previously reserved
  drm/ttm: Don't add swapped BOs to swap-LRU list
  drm/amdgpu: only check for ECC on Vega10
  drm/amd/powerplay: Fix smu_table_entry.handle type
  drm/ttm: add VADDR_FLAG_UPDATED_COUNT to correctly update dma_page global count
  drm: Fix PANEL_ORIENTATION_QUIRKS breaking the Kconfig DRM menuconfig
  drm/radeon: fill in rb backend map on evergreen/ni.
  drm/amdgpu/gfx9: fix ngg enablement to clear gds reserved memory (v2)
  drm/ttm: only free pages rather than update global memory count together
  drm/amdgpu: fix CPU based VM updates
  drm/amdgpu: fix typo in amdgpu_vce_validate_bo
  ...

1 files changed, 139 insertions, 190 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 5aeb5f8816f3..c06479615e8a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -28,11 +28,10 @@
 #include "soc15.h"
 #include "soc15d.h"
 
-#include "vega10/soc15ip.h"
-#include "vega10/GC/gc_9_0_offset.h"
-#include "vega10/GC/gc_9_0_sh_mask.h"
-#include "vega10/vega10_enum.h"
-#include "vega10/HDP/hdp_4_0_offset.h"
+#include "gc/gc_9_0_offset.h"
+#include "gc/gc_9_0_sh_mask.h"
+#include "vega10_enum.h"
+#include "hdp/hdp_4_0_offset.h"
 
 #include "soc15_common.h"
 #include "clearstate_gfx9.h"
@@ -65,152 +64,84 @@ MODULE_FIRMWARE("amdgpu/raven_mec.bin");
 MODULE_FIRMWARE("amdgpu/raven_mec2.bin");
 MODULE_FIRMWARE("amdgpu/raven_rlc.bin");
 
-static const struct amdgpu_gds_reg_offset amdgpu_gds_reg_offset[] =
-{
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID0_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID0_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID0),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID0) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID1_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID1_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID1),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID1) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID2_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID2_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID2),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID2) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID3_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID3_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID3),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID3) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID4_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID4_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID4),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID4) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID5_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID5_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID5),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID5) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID6_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID6_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID6),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID6) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID7_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID7_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID7),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID7) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID8_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID8_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID8),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID8) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID9_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID9_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID9),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID9) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID10_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID10_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID10),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID10) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID11_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID11_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID11),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID11) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID12_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID12_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID12),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID12)},
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID13_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID13_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID13),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID13) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID14_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID14_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID14),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID14) },
-	{ SOC15_REG_OFFSET(GC, 0, mmGDS_VMID15_BASE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_VMID15_SIZE),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID15),
-	  SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID15) }
+static const struct soc15_reg_golden golden_settings_gc_9_0[] =
+{
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPC_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPG_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG2, 0xf00fffff, 0x00000420),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_GPU_ID, 0x0000000f, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmIA_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_EVENT_CNTL_3, 0x00000003, 0x82400024),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE, 0x3fffffff, 0x00000001),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_0, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_1, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_2, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_PREWALKER_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSH_MEM_CONFIG, 0x00001000, 0x00001000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_CONFIG_CNTL_1, 0x0000000f, 0x01000107),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQC_CONFIG, 0x03000000, 0x020a2000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfffffeef, 0x010b0000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_HI, 0xffffffff, 0x4a2c0e68),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_LO, 0xffffffff, 0xb5d3f197),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_CACHE_INVALIDATION, 0x3fff3af3, 0x19200000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000003ff),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmWD_UTCL1_CNTL, 0x08000000, 0x08000080)
 };
 
-static const u32 golden_settings_gc_9_0[] =
-{
-	SOC15_REG_OFFSET(GC, 0, mmCPC_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmCPF_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmCPG_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmDB_DEBUG2), 0xf00fffff, 0x00000420,
-	SOC15_REG_OFFSET(GC, 0, mmGB_GPU_ID), 0x0000000f, 0x00000000,
-	SOC15_REG_OFFSET(GC, 0, mmIA_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_BINNER_EVENT_CNTL_3), 0x00000003, 0x82400024,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_ENHANCE), 0x3fffffff, 0x00000001,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_LINE_STIPPLE_STATE), 0x0000ff0f, 0x00000000,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UTCL1_CNTL_0), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UTCL1_CNTL_1), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UTCL1_CNTL_2), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_PREWALKER_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmSH_MEM_CONFIG), 0x00001000, 0x00001000,
-	SOC15_REG_OFFSET(GC, 0, mmSPI_CONFIG_CNTL_1), 0x0000000f, 0x01000107,
-	SOC15_REG_OFFSET(GC, 0, mmSQC_CONFIG), 0x03000000, 0x020a2000,
-	SOC15_REG_OFFSET(GC, 0, mmTA_CNTL_AUX), 0xfffffeef, 0x010b0000,
-	SOC15_REG_OFFSET(GC, 0, mmTCP_CHAN_STEER_HI), 0xffffffff, 0x4a2c0e68,
-	SOC15_REG_OFFSET(GC, 0, mmTCP_CHAN_STEER_LO), 0xffffffff, 0xb5d3f197,
-	SOC15_REG_OFFSET(GC, 0, mmVGT_CACHE_INVALIDATION), 0x3fff3af3, 0x19200000,
-	SOC15_REG_OFFSET(GC, 0, mmVGT_GS_MAX_WAVE_ID), 0x00000fff, 0x000003ff,
-	SOC15_REG_OFFSET(GC, 0, mmWD_UTCL1_CNTL), 0x08000000, 0x08000080
-};
-
-static const u32 golden_settings_gc_9_0_vg10[] =
+static const struct soc15_reg_golden golden_settings_gc_9_0_vg10[] =
 {
-	SOC15_REG_OFFSET(GC, 0, mmCB_HW_CONTROL), 0x0000f000, 0x00012107,
-	SOC15_REG_OFFSET(GC, 0, mmCB_HW_CONTROL_3), 0x30000000, 0x10000000,
-	SOC15_REG_OFFSET(GC, 0, mmGB_ADDR_CONFIG), 0xffff77ff, 0x2a114042,
-	SOC15_REG_OFFSET(GC, 0, mmGB_ADDR_CONFIG_READ), 0xffff77ff, 0x2a114042,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_ENHANCE_1), 0x00008000, 0x00048000,
-	SOC15_REG_OFFSET(GC, 0, mmRMI_UTCL1_CNTL2), 0x00030000, 0x00020000,
-	SOC15_REG_OFFSET(GC, 0, mmTD_CNTL), 0x00001800, 0x00000800
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL, 0x0000f000, 0x00012107),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_3, 0x30000000, 0x10000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0xffff77ff, 0x2a114042),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG_READ, 0xffff77ff, 0x2a114042),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_1, 0x00008000, 0x00048000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRMI_UTCL1_CNTL2, 0x00030000, 0x00020000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTD_CNTL, 0x00001800, 0x00000800)
 };
 
-static const u32 golden_settings_gc_9_1[] =
-{
-	SOC15_REG_OFFSET(GC, 0, mmCB_HW_CONTROL), 0xfffdf3cf, 0x00014104,
-	SOC15_REG_OFFSET(GC, 0, mmCPC_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmCPF_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmCPG_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmDB_DEBUG2), 0xf00fffff, 0x00000420,
-	SOC15_REG_OFFSET(GC, 0, mmGB_GPU_ID), 0x0000000f, 0x00000000,
-	SOC15_REG_OFFSET(GC, 0, mmIA_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_BINNER_EVENT_CNTL_3), 0x00000003, 0x82400024,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_ENHANCE), 0x3fffffff, 0x00000001,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_LINE_STIPPLE_STATE), 0x0000ff0f, 0x00000000,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UTCL1_CNTL_0), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UTCL1_CNTL_1), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UTCL1_CNTL_2), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_PREWALKER_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_UTCL1_CNTL), 0x08000000, 0x08000080,
-	SOC15_REG_OFFSET(GC, 0, mmTA_CNTL_AUX), 0xfffffeef, 0x010b0000,
-	SOC15_REG_OFFSET(GC, 0, mmTCP_CHAN_STEER_HI), 0xffffffff, 0x00000000,
-	SOC15_REG_OFFSET(GC, 0, mmTCP_CHAN_STEER_LO), 0xffffffff, 0x00003120,
-	SOC15_REG_OFFSET(GC, 0, mmVGT_CACHE_INVALIDATION), 0x3fff3af3, 0x19200000,
-	SOC15_REG_OFFSET(GC, 0, mmVGT_GS_MAX_WAVE_ID), 0x00000fff, 0x000000ff,
-	SOC15_REG_OFFSET(GC, 0, mmWD_UTCL1_CNTL), 0x08000000, 0x08000080
+static const struct soc15_reg_golden golden_settings_gc_9_1[] =
+{
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL, 0xfffdf3cf, 0x00014104),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPC_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPG_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG2, 0xf00fffff, 0x00000420),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_GPU_ID, 0x0000000f, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmIA_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_EVENT_CNTL_3, 0x00000003, 0x82400024),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE, 0x3fffffff, 0x00000001),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_0, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_1, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_GPM_UTCL1_CNTL_2, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_PREWALKER_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_UTCL1_CNTL, 0x08000000, 0x08000080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfffffeef, 0x010b0000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_HI, 0xffffffff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTCP_CHAN_STEER_LO, 0xffffffff, 0x00003120),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_CACHE_INVALIDATION, 0x3fff3af3, 0x19200000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000000ff),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmWD_UTCL1_CNTL, 0x08000000, 0x08000080)
 };
 
-static const u32 golden_settings_gc_9_1_rv1[] =
+static const struct soc15_reg_golden golden_settings_gc_9_1_rv1[] =
 {
-	SOC15_REG_OFFSET(GC, 0, mmCB_HW_CONTROL_3), 0x30000000, 0x10000000,
-	SOC15_REG_OFFSET(GC, 0, mmGB_ADDR_CONFIG), 0xffff77ff, 0x24000042,
-	SOC15_REG_OFFSET(GC, 0, mmGB_ADDR_CONFIG_READ), 0xffff77ff, 0x24000042,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_ENHANCE_1), 0xffffffff, 0x04048000,
-	SOC15_REG_OFFSET(GC, 0, mmPA_SC_MODE_CNTL_1), 0x06000000, 0x06000000,
-	SOC15_REG_OFFSET(GC, 0, mmRMI_UTCL1_CNTL2), 0x00030000, 0x00020000,
-	SOC15_REG_OFFSET(GC, 0, mmTD_CNTL), 0x01bd9f33, 0x00000800
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_3, 0x30000000, 0x10000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0xffff77ff, 0x24000042),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG_READ, 0xffff77ff, 0x24000042),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_1, 0xffffffff, 0x04048000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_MODE_CNTL_1, 0x06000000, 0x06000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmRMI_UTCL1_CNTL2, 0x00030000, 0x00020000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTD_CNTL, 0x01bd9f33, 0x00000800)
 };
 
-static const u32 golden_settings_gc_9_x_common[] =
+static const struct soc15_reg_golden golden_settings_gc_9_x_common[] =
 {
-	SOC15_REG_OFFSET(GC, 0, mmGRBM_CAM_INDEX), 0xffffffff, 0x00000000,
-	SOC15_REG_OFFSET(GC, 0, mmGRBM_CAM_DATA), 0xffffffff, 0x2544c382
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_CAM_INDEX, 0xffffffff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_CAM_DATA, 0xffffffff, 0x2544c382)
 };
 
 #define VEGA10_GB_ADDR_CONFIG_GOLDEN 0x2a114042
@@ -230,26 +161,26 @@ static void gfx_v9_0_init_golden_registers(struct amdgpu_device *adev)
 {
 	switch (adev->asic_type) {
 	case CHIP_VEGA10:
-		amdgpu_program_register_sequence(adev,
+		soc15_program_register_sequence(adev,
 						 golden_settings_gc_9_0,
-						 (const u32)ARRAY_SIZE(golden_settings_gc_9_0));
-		amdgpu_program_register_sequence(adev,
+						 ARRAY_SIZE(golden_settings_gc_9_0));
+		soc15_program_register_sequence(adev,
 						 golden_settings_gc_9_0_vg10,
-						 (const u32)ARRAY_SIZE(golden_settings_gc_9_0_vg10));
+						 ARRAY_SIZE(golden_settings_gc_9_0_vg10));
 		break;
 	case CHIP_RAVEN:
-		amdgpu_program_register_sequence(adev,
+		soc15_program_register_sequence(adev,
 						 golden_settings_gc_9_1,
-						 (const u32)ARRAY_SIZE(golden_settings_gc_9_1));
-		amdgpu_program_register_sequence(adev,
+						 ARRAY_SIZE(golden_settings_gc_9_1));
+		soc15_program_register_sequence(adev,
 						 golden_settings_gc_9_1_rv1,
-						 (const u32)ARRAY_SIZE(golden_settings_gc_9_1_rv1));
+						 ARRAY_SIZE(golden_settings_gc_9_1_rv1));
 		break;
 	default:
 		break;
 	}
 
-	amdgpu_program_register_sequence(adev, golden_settings_gc_9_x_common,
+	soc15_program_register_sequence(adev, golden_settings_gc_9_x_common,
 					(const u32)ARRAY_SIZE(golden_settings_gc_9_x_common));
 }
 
@@ -327,7 +258,7 @@ static int gfx_v9_0_ring_test_ring(struct amdgpu_ring *ring)
 		DRM_UDELAY(1);
 	}
 	if (i < adev->usec_timeout) {
-		DRM_INFO("ring test on %d succeeded in %d usecs\n",
+		DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
 			 ring->idx, i);
 	} else {
 		DRM_ERROR("amdgpu: ring %d test failed (scratch(0x%04X)=0x%08X)\n",
@@ -379,7 +310,7 @@ static int gfx_v9_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
         }
         tmp = RREG32(scratch);
         if (tmp == 0xDEADBEEF) {
-                DRM_INFO("ib test on ring %d succeeded\n", ring->idx);
+                DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
                 r = 0;
         } else {
                 DRM_ERROR("amdgpu: ib test failed (scratch(0x%04X)=0x%08X)\n",
@@ -1137,8 +1068,8 @@ static int gfx_v9_0_ngg_init(struct amdgpu_device *adev)
 	adev->gfx.ngg.gds_reserve_size = ALIGN(5 * 4, 0x40);
 	adev->gds.mem.total_size -= adev->gfx.ngg.gds_reserve_size;
 	adev->gds.mem.gfx_partition_size -= adev->gfx.ngg.gds_reserve_size;
-	adev->gfx.ngg.gds_reserve_addr = amdgpu_gds_reg_offset[0].mem_base;
-	adev->gfx.ngg.gds_reserve_addr += adev->gds.mem.gfx_partition_size;
+	adev->gfx.ngg.gds_reserve_addr = RREG32_SOC15(GC, 0, mmGDS_VMID0_BASE);
+	adev->gfx.ngg.gds_reserve_addr += RREG32_SOC15(GC, 0, mmGDS_VMID0_SIZE);
 
 	/* Primitive Buffer */
 	r = gfx_v9_0_ngg_create_buf(adev, &adev->gfx.ngg.buf[NGG_PRIM],
@@ -1243,23 +1174,24 @@ static int gfx_v9_0_ngg_en(struct amdgpu_device *adev)
 	}
 
 	gfx_v9_0_write_data_to_reg(ring, 0, false,
-				   amdgpu_gds_reg_offset[0].mem_size,
+				   SOC15_REG_OFFSET(GC, 0, mmGDS_VMID0_SIZE),
 			           (adev->gds.mem.total_size +
 				    adev->gfx.ngg.gds_reserve_size) >>
 				   AMDGPU_GDS_SHIFT);
 
 	amdgpu_ring_write(ring, PACKET3(PACKET3_DMA_DATA, 5));
 	amdgpu_ring_write(ring, (PACKET3_DMA_DATA_CP_SYNC |
+				PACKET3_DMA_DATA_DST_SEL(1) |
 				PACKET3_DMA_DATA_SRC_SEL(2)));
 	amdgpu_ring_write(ring, 0);
 	amdgpu_ring_write(ring, 0);
 	amdgpu_ring_write(ring, adev->gfx.ngg.gds_reserve_addr);
 	amdgpu_ring_write(ring, 0);
-	amdgpu_ring_write(ring, adev->gfx.ngg.gds_reserve_size);
-
+	amdgpu_ring_write(ring, PACKET3_DMA_DATA_CMD_RAW_WAIT |
+				adev->gfx.ngg.gds_reserve_size);
 
 	gfx_v9_0_write_data_to_reg(ring, 0, false,
-				   amdgpu_gds_reg_offset[0].mem_size, 0);
+				   SOC15_REG_OFFSET(GC, 0, mmGDS_VMID0_SIZE), 0);
 
 	amdgpu_ring_commit(ring);
 
@@ -1464,7 +1396,6 @@ static int gfx_v9_0_sw_fini(void *handle)
 	amdgpu_gfx_compute_mqd_sw_fini(adev);
 	amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq.ring, &adev->gfx.kiq.irq);
 	amdgpu_gfx_kiq_fini(adev);
-	amdgpu_bo_free_kernel(&adev->virt.csa_obj, &adev->virt.csa_vmid0_addr, NULL);
 
 	gfx_v9_0_mec_fini(adev);
 	gfx_v9_0_ngg_fini(adev);
@@ -1596,14 +1527,21 @@ static void gfx_v9_0_gpu_init(struct amdgpu_device *adev)
 	/* XXX SH_MEM regs */
 	/* where to put LDS, scratch, GPUVM in FSA64 space */
 	mutex_lock(&adev->srbm_mutex);
-	for (i = 0; i < 16; i++) {
+	for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB].num_ids; i++) {
 		soc15_grbm_select(adev, 0, 0, 0, i);
 		/* CP and shaders */
-		tmp = 0;
-		tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, ALIGNMENT_MODE,
-				    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
-		WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, tmp);
-		WREG32_SOC15(GC, 0, mmSH_MEM_BASES, 0);
+		if (i == 0) {
+			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
+					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
+			WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, tmp);
+			WREG32_SOC15(GC, 0, mmSH_MEM_BASES, 0);
+		} else {
+			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE,
+					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
+			WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, tmp);
+			tmp = adev->mc.shared_aperture_start >> 48;
+			WREG32_SOC15(GC, 0, mmSH_MEM_BASES, tmp);
+		}
 	}
 	soc15_grbm_select(adev, 0, 0, 0, 0);
 
@@ -1645,6 +1583,14 @@ static void gfx_v9_0_wait_for_rlc_serdes(struct amdgpu_device *adev)
 					break;
 				udelay(1);
 			}
+			if (k == adev->usec_timeout) {
+				gfx_v9_0_select_se_sh(adev, 0xffffffff,
+						      0xffffffff, 0xffffffff);
+				mutex_unlock(&adev->grbm_idx_mutex);
+				DRM_INFO("Timeout wait for RLC serdes %u,%u\n",
+					 i, j);
+				return;
+			}
 		}
 	}
 	gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
@@ -2749,7 +2695,7 @@ static int gfx_v9_0_kiq_init_queue(struct amdgpu_ring *ring)
 
 	gfx_v9_0_kiq_setting(ring);
 
-	if (adev->in_sriov_reset) { /* for GPU_RESET case */
+	if (adev->in_gpu_reset) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation));
@@ -2787,7 +2733,7 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring)
 	struct v9_mqd *mqd = ring->mqd_ptr;
 	int mqd_idx = ring - &adev->gfx.compute_ring[0];
 
-	if (!adev->in_sriov_reset && !adev->gfx.in_suspend) {
+	if (!adev->in_gpu_reset && !adev->gfx.in_suspend) {
 		memset((void *)mqd, 0, sizeof(struct v9_mqd_allocation));
 		((struct v9_mqd_allocation *)mqd)->dynamic_cu_mask = 0xFFFFFFFF;
 		((struct v9_mqd_allocation *)mqd)->dynamic_rb_mask = 0xFFFFFFFF;
@@ -2799,7 +2745,7 @@ static int gfx_v9_0_kcq_init_queue(struct amdgpu_ring *ring)
 
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(adev->gfx.mec.mqd_backup[mqd_idx], mqd, sizeof(struct v9_mqd_allocation));
-	} else if (adev->in_sriov_reset) { /* for GPU_RESET case */
+	} else if (adev->in_gpu_reset) { /* for GPU_RESET case */
 		/* reset MQD to a clean status */
 		if (adev->gfx.mec.mqd_backup[mqd_idx])
 			memcpy(mqd, adev->gfx.mec.mqd_backup[mqd_idx], sizeof(struct v9_mqd_allocation));
@@ -3139,6 +3085,8 @@ static void gfx_v9_0_ring_emit_gds_switch(struct amdgpu_ring *ring,
 					  uint32_t gws_base, uint32_t gws_size,
 					  uint32_t oa_base, uint32_t oa_size)
 {
+	struct amdgpu_device *adev = ring->adev;
+
 	gds_base = gds_base >> AMDGPU_GDS_SHIFT;
 	gds_size = gds_size >> AMDGPU_GDS_SHIFT;
 
@@ -3150,22 +3098,22 @@ static void gfx_v9_0_ring_emit_gds_switch(struct amdgpu_ring *ring,
 
 	/* GDS Base */
 	gfx_v9_0_write_data_to_reg(ring, 0, false,
-				   amdgpu_gds_reg_offset[vmid].mem_base,
+				   SOC15_REG_OFFSET(GC, 0, mmGDS_VMID0_BASE) + 2 * vmid,
 				   gds_base);
 
 	/* GDS Size */
 	gfx_v9_0_write_data_to_reg(ring, 0, false,
-				   amdgpu_gds_reg_offset[vmid].mem_size,
+				   SOC15_REG_OFFSET(GC, 0, mmGDS_VMID0_SIZE) + 2 * vmid,
 				   gds_size);
 
 	/* GWS */
 	gfx_v9_0_write_data_to_reg(ring, 0, false,
-				   amdgpu_gds_reg_offset[vmid].gws,
+				   SOC15_REG_OFFSET(GC, 0, mmGDS_GWS_VMID0) + vmid,
 				   gws_size << GDS_GWS_VMID0__SIZE__SHIFT | gws_base);
 
 	/* OA */
 	gfx_v9_0_write_data_to_reg(ring, 0, false,
-				   amdgpu_gds_reg_offset[vmid].oa,
+				   SOC15_REG_OFFSET(GC, 0, mmGDS_OA_VMID0) + vmid,
 				   (1 << (oa_size + oa_base)) - (1 << oa_base));
 }
 
@@ -3610,13 +3558,9 @@ static void gfx_v9_0_ring_set_wptr_gfx(struct amdgpu_ring *ring)
 
 static void gfx_v9_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 {
+	struct amdgpu_device *adev = ring->adev;
 	u32 ref_and_mask, reg_mem_engine;
-	const struct nbio_hdp_flush_reg *nbio_hf_reg;
-
-	if (ring->adev->flags & AMD_IS_APU)
-		nbio_hf_reg = &nbio_v7_0_hdp_flush_reg;
-	else
-		nbio_hf_reg = &nbio_v6_1_hdp_flush_reg;
+	const struct nbio_hdp_flush_reg *nbio_hf_reg = adev->nbio_funcs->hdp_flush_reg;
 
 	if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) {
 		switch (ring->me) {
@@ -3636,20 +3580,22 @@ static void gfx_v9_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 	}
 
 	gfx_v9_0_wait_reg_mem(ring, reg_mem_engine, 0, 1,
-			      nbio_hf_reg->hdp_flush_req_offset,
-			      nbio_hf_reg->hdp_flush_done_offset,
+			      adev->nbio_funcs->get_hdp_flush_req_offset(adev),
+			      adev->nbio_funcs->get_hdp_flush_done_offset(adev),
 			      ref_and_mask, ref_and_mask, 0x20);
 }
 
 static void gfx_v9_0_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
 {
+	struct amdgpu_device *adev = ring->adev;
+
 	gfx_v9_0_write_data_to_reg(ring, 0, true,
 				   SOC15_REG_OFFSET(HDP, 0, mmHDP_READ_CACHE_INVALIDATE), 1);
 }
 
 static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
                                       struct amdgpu_ib *ib,
-                                      unsigned vm_id, bool ctx_switch)
+                                      unsigned vmid, bool ctx_switch)
 {
 	u32 header, control = 0;
 
@@ -3658,7 +3604,7 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 	else
 		header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw | (vm_id << 24);
+	control |= ib->length_dw | (vmid << 24);
 
 	if (amdgpu_sriov_vf(ring->adev) && (ib->flags & AMDGPU_IB_FLAG_PREEMPT)) {
 		control |= INDIRECT_BUFFER_PRE_ENB(1);
@@ -3680,9 +3626,9 @@ BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
 
 static void gfx_v9_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
                                           struct amdgpu_ib *ib,
-                                          unsigned vm_id, bool ctx_switch)
+                                          unsigned vmid, bool ctx_switch)
 {
-        u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vm_id << 24);
+        u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
 
         amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
 	BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
@@ -3738,22 +3684,23 @@ static void gfx_v9_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
 }
 
 static void gfx_v9_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
-					unsigned vm_id, uint64_t pd_addr)
+					unsigned vmid, uint64_t pd_addr)
 {
 	struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->funcs->vmhub];
 	int usepfp = (ring->funcs->type == AMDGPU_RING_TYPE_GFX);
-	uint32_t req = ring->adev->gart.gart_funcs->get_invalidate_req(vm_id);
+	uint32_t req = ring->adev->gart.gart_funcs->get_invalidate_req(vmid);
+	uint64_t flags = AMDGPU_PTE_VALID;
 	unsigned eng = ring->vm_inv_eng;
 
-	pd_addr = amdgpu_gart_get_vm_pde(ring->adev, pd_addr);
-	pd_addr |= AMDGPU_PTE_VALID;
+	amdgpu_gart_get_vm_pde(ring->adev, -1, &pd_addr, &flags);
+	pd_addr |= flags;
 
 	gfx_v9_0_write_data_to_reg(ring, usepfp, true,
-				   hub->ctx0_ptb_addr_lo32 + (2 * vm_id),
+				   hub->ctx0_ptb_addr_lo32 + (2 * vmid),
 				   lower_32_bits(pd_addr));
 
 	gfx_v9_0_write_data_to_reg(ring, usepfp, true,
-				   hub->ctx0_ptb_addr_hi32 + (2 * vm_id),
+				   hub->ctx0_ptb_addr_hi32 + (2 * vmid),
 				   upper_32_bits(pd_addr));
 
 	gfx_v9_0_write_data_to_reg(ring, usepfp, true,
@@ -3761,7 +3708,7 @@ static void gfx_v9_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
 
 	/* wait for the invalidate to complete */
 	gfx_v9_0_wait_reg_mem(ring, 0, 0, 0, hub->vm_inv_eng0_ack +
-			      eng, 0, 1 << vm_id, 1 << vm_id, 0x20);
+			      eng, 0, 1 << vmid, 1 << vmid, 0x20);
 
 	/* compute doesn't have PFP */
 	if (usepfp) {
@@ -3804,6 +3751,8 @@ static void gfx_v9_0_ring_set_wptr_compute(struct amdgpu_ring *ring)
 static void gfx_v9_0_ring_emit_fence_kiq(struct amdgpu_ring *ring, u64 addr,
 					 u64 seq, unsigned int flags)
 {
+	struct amdgpu_device *adev = ring->adev;
+
 	/* we only allocate 32bit for each seq wb address */
 	BUG_ON(flags & AMDGPU_FENCE_FLAG_64BIT);