Merge tag 'drm-next-2020-12-11' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie:
 "Not a huge amount of big things here, AMD has support for a few new HW
  variants (vangogh, green sardine, dimgrey cavefish), Intel has some
  more DG1 enablement. We have a few big reworks of the TTM layers and
  interfaces, GEM and atomic internal API reworks cross tree. fbdev is
  marked orphaned in here as well to reflect the current reality.

  core:
   - documentation updates
   - deprecate DRM_FORMAT_MOD_NONE
   - atomic crtc enable/disable rework
   - GEM convert drivers to gem object functions
   - remove SCATTER_LIST_MAX_SEGMENT

  sched:
   - avoid infinite waits

  ttm:
   - remove AGP support
   - don't modify caching for swapout
   - ttm pinning rework
   - major TTM reworks
   - new backend allocator
   - multihop support

  vram-helper:
   - top down BO placement fix
   - TTM changes
   - GEM object support

  displayport:
   - DP 2.0 DPCD prep work
   - DP MST extended DPCD caps

  fbdev:
   - mark as orphaned

  amdgpu:
   - Initial Vangogh support
   - Green Sardine support
   - Dimgrey Cavefish support
   - SG display support for renoir
   - SMU7 improvements
   - gfx9+ modiifier support
   - CI BACO fixes

  radeon:
   - expose voltage via hwmon on SUMO

  amdkfd:
   - fix unique id handling

  i915:
   - more DG1 enablement
   - bigjoiner support
   - integer scaling filter support
   - async flip support
   - ICL+ DSI command mode
   - Improve display shutdown
   - Display refactoring
   - eLLC machine fbdev loading fix
   - dma scatterlist fixes
   - TGL hang fixes
   - eLLC display buffer caching on SKL+
   - MOCS PTE seeting for gen9+

  msm:
   - Shutdown hook
   - GPU cooling device support
   - DSI 7nm and 10nm phy/pll updates
   - sm8150/sm2850 DPU support
   - GEM locking re-work
   - LLCC system cache support

  aspeed:
   - sysfs output config support

  ast:
   - LUT fix
   - new display mode

  gma500:
   - remove 2d framebuffer accel

  panfrost:
   - move gpu reset to a worker

  exynos:
   - new HDMI mode support

  mediatek:
   - MT8167 support
   - yaml bindings
   - MIPI DSI phy code moved

  etnaviv:
   - new perf counter
   - more lockdep annotation

  hibmc:
   - i2c DDC support

  ingenic:
   - pixel clock reset fix
   - reserved memory support
   - allow both DMA channels at once
   - different pixel format support
   - 30/24/8-bit palette modes

  tilcdc:
   - don't keep vblank irq enabled

  vc4:
   - new maintainer added
   - DSI registration fix

  virtio:
   - blob resource support
   - host visible and cross-device support
   - uuid api support"

* tag 'drm-next-2020-12-11' of git://anongit.freedesktop.org/drm/drm: (1754 commits)
  drm/amdgpu: Initialise drm_gem_object_funcs for imported BOs
  drm/amdgpu: fix size calculation with stolen vga memory
  drm/amdgpu: remove amdgpu_ttm_late_init  and amdgpu_bo_late_init
  drm/amdgpu: free the pre-OS console framebuffer after the first modeset
  drm/amdgpu: enable runtime pm using BACO on CI dGPUs
  drm/amdgpu/cik: enable BACO reset on Bonaire
  drm/amd/pm: update smu10.h WORKLOAD_PPLIB setting for raven
  drm/amd/pm: remove one unsupported smu function for vangogh
  drm/amd/display: setup system context for APUs
  drm/amd/display: add S/G support for Vangogh
  drm/amdkfd: Fix leak in dmabuf import
  drm/amdgpu: use AMDGPU_NUM_VMID when possible
  drm/amdgpu: fix sdma instance fw version and feature version init
  drm/amd/pm: update driver if version for dimgrey_cavefish
  drm/amd/display: 3.2.115
  drm/amd/display: [FW Promotion] Release 0.0.45
  drm/amd/display: Revert DCN2.1 dram_clock_change_latency update
  drm/amd/display: Enable gpu_vm_support for dcn3.01
  drm/amd/display: Fixed the audio noise during mode switching with HDCP mode on
  drm/amd/display: Add wm table for Renoir
  ...

1 files changed, 302 insertions, 35 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 55f4b8c3b933..ba1086784525 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -99,6 +99,23 @@
 #define mmGCR_GENERAL_CNTL_Sienna_Cichlid			0x1580
 #define mmGCR_GENERAL_CNTL_Sienna_Cichlid_BASE_IDX	0
 
+#define mmSPI_CONFIG_CNTL_1_Vangogh		 0x2441
+#define mmSPI_CONFIG_CNTL_1_Vangogh_BASE_IDX	 1
+#define mmVGT_TF_MEMORY_BASE_HI_Vangogh          0x2261
+#define mmVGT_TF_MEMORY_BASE_HI_Vangogh_BASE_IDX 1
+#define mmVGT_HS_OFFCHIP_PARAM_Vangogh           0x224f
+#define mmVGT_HS_OFFCHIP_PARAM_Vangogh_BASE_IDX  1
+#define mmVGT_TF_RING_SIZE_Vangogh               0x224e
+#define mmVGT_TF_RING_SIZE_Vangogh_BASE_IDX      1
+#define mmVGT_GSVS_RING_SIZE_Vangogh             0x2241
+#define mmVGT_GSVS_RING_SIZE_Vangogh_BASE_IDX    1
+#define mmVGT_TF_MEMORY_BASE_Vangogh             0x2250
+#define mmVGT_TF_MEMORY_BASE_Vangogh_BASE_IDX    1
+#define mmVGT_ESGS_RING_SIZE_Vangogh             0x2240
+#define mmVGT_ESGS_RING_SIZE_Vangogh_BASE_IDX    1
+#define mmSPI_CONFIG_CNTL_Vangogh                0x2440
+#define mmSPI_CONFIG_CNTL_Vangogh_BASE_IDX       1
+
 #define mmCP_HYP_PFP_UCODE_ADDR			0x5814
 #define mmCP_HYP_PFP_UCODE_ADDR_BASE_IDX	1
 #define mmCP_HYP_PFP_UCODE_DATA			0x5815
@@ -112,6 +129,13 @@
 #define mmCP_HYP_ME_UCODE_DATA			0x5817
 #define mmCP_HYP_ME_UCODE_DATA_BASE_IDX		1
 
+#define mmCPG_PSP_DEBUG				0x5c10
+#define mmCPG_PSP_DEBUG_BASE_IDX		1
+#define mmCPC_PSP_DEBUG				0x5c11
+#define mmCPC_PSP_DEBUG_BASE_IDX		1
+#define CPC_PSP_DEBUG__GPA_OVERRIDE_MASK	0x00000008L
+#define CPG_PSP_DEBUG__GPA_OVERRIDE_MASK	0x00000008L
+
 //CC_GC_SA_UNIT_DISABLE
 #define mmCC_GC_SA_UNIT_DISABLE                 0x0fe9
 #define mmCC_GC_SA_UNIT_DISABLE_BASE_IDX        0
@@ -131,6 +155,11 @@
 #define mmCGTT_SPI_CS_CLK_CTRL			0x507c
 #define mmCGTT_SPI_CS_CLK_CTRL_BASE_IDX         1
 
+#define mmGCUTCL2_CGTT_CLK_CTRL_Sienna_Cichlid		0x16f3
+#define mmGCUTCL2_CGTT_CLK_CTRL_Sienna_Cichlid_BASE_IDX	0
+#define mmGCVM_L2_CGTT_CLK_CTRL_Sienna_Cichlid          0x15db
+#define mmGCVM_L2_CGTT_CLK_CTRL_Sienna_Cichlid_BASE_IDX	0
+
 MODULE_FIRMWARE("amdgpu/navi10_ce.bin");
 MODULE_FIRMWARE("amdgpu/navi10_pfp.bin");
 MODULE_FIRMWARE("amdgpu/navi10_me.bin");
@@ -171,6 +200,20 @@ MODULE_FIRMWARE("amdgpu/navy_flounder_mec.bin");
 MODULE_FIRMWARE("amdgpu/navy_flounder_mec2.bin");
 MODULE_FIRMWARE("amdgpu/navy_flounder_rlc.bin");
 
+MODULE_FIRMWARE("amdgpu/vangogh_ce.bin");
+MODULE_FIRMWARE("amdgpu/vangogh_pfp.bin");
+MODULE_FIRMWARE("amdgpu/vangogh_me.bin");
+MODULE_FIRMWARE("amdgpu/vangogh_mec.bin");
+MODULE_FIRMWARE("amdgpu/vangogh_mec2.bin");
+MODULE_FIRMWARE("amdgpu/vangogh_rlc.bin");
+
+MODULE_FIRMWARE("amdgpu/dimgrey_cavefish_ce.bin");
+MODULE_FIRMWARE("amdgpu/dimgrey_cavefish_pfp.bin");
+MODULE_FIRMWARE("amdgpu/dimgrey_cavefish_me.bin");
+MODULE_FIRMWARE("amdgpu/dimgrey_cavefish_mec.bin");
+MODULE_FIRMWARE("amdgpu/dimgrey_cavefish_mec2.bin");
+MODULE_FIRMWARE("amdgpu/dimgrey_cavefish_rlc.bin");
+
 static const struct soc15_reg_golden golden_settings_gc_10_1[] =
 {
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0xffffffff, 0x00400014),
@@ -1366,23 +1409,14 @@ static void gfx_v10_rlcg_wreg(struct amdgpu_device *adev, u32 offset, u32 v)
 {
 	static void *scratch_reg0;
 	static void *scratch_reg1;
-	static void *scratch_reg2;
-	static void *scratch_reg3;
 	static void *spare_int;
-	static uint32_t grbm_cntl;
-	static uint32_t grbm_idx;
 	uint32_t i = 0;
 	uint32_t retries = 50000;
 
 	scratch_reg0 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0)*4;
 	scratch_reg1 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1)*4;
-	scratch_reg2 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG2)*4;
-	scratch_reg3 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3)*4;
 	spare_int = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT)*4;
 
-	grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL;
-	grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX;
-
 	if (amdgpu_sriov_runtime(adev)) {
 		pr_err("shouldn't call rlcg write register during runtime\n");
 		return;
@@ -3108,6 +3142,8 @@ static const struct soc15_reg_golden golden_settings_gc_10_3[] =
 	SOC15_REG_GOLDEN_VALUE(GC, 0 ,mmGCEA_SDP_TAG_RESERVE0, 0xffffffff, 0x10100100),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCEA_SDP_TAG_RESERVE1, 0xffffffff, 0x17000088),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL_Sienna_Cichlid, 0x1ff1ffff, 0x00000500),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCUTCL2_CGTT_CLK_CTRL_Sienna_Cichlid, 0xff000000, 0xff008080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCVM_L2_CGTT_CLK_CTRL_Sienna_Cichlid, 0xff000000, 0xff008080),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGE_PC_CNTL, 0x003fffff, 0x00280400),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2A_ADDR_MATCH_MASK, 0xffffffff, 0xffffffcf),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xffffffcf),
@@ -3146,6 +3182,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_3_sienna_cichlid[] =
 
 static const struct soc15_reg_golden golden_settings_gc_10_3_2[] =
 {
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_PS_CLK_CTRL, 0xff7f0fff, 0x78000100),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA0_CLK_CTRL, 0xff7f0fff, 0x30000100),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA1_CLK_CTRL, 0xff7f0fff, 0x7e000100),
@@ -3154,6 +3191,8 @@ static const struct soc15_reg_golden golden_settings_gc_10_3_2[] =
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0xffffffff, 0x00800000),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_EXCEPTION_CONTROL, 0x7fff0f1f, 0x00b80000),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL_Sienna_Cichlid, 0x1ff1ffff, 0x00000500),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCUTCL2_CGTT_CLK_CTRL_Sienna_Cichlid, 0xffffffff, 0xff008080),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCVM_L2_CGTT_CLK_CTRL_Sienna_Cichlid, 0xffff8fff, 0xff008080),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGE_PC_CNTL, 0x003fffff, 0x00280400),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2A_ADDR_MATCH_MASK, 0xffffffff, 0xffffffcf),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xffffffcf),
@@ -3183,7 +3222,79 @@ static const struct soc15_reg_golden golden_settings_gc_10_3_2[] =
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER9_SELECT, 0xf0f001ff, 0x00000000),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000),
 	SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffbfffff, 0x00a00000),
-	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000003ff)
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000003ff),
+
+	/* This is not in GDB yet. Don't remove it. It fixes a GPU hang on Navy Flounder. */
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG,  0x00000020, 0x00000020),
+};
+
+static const struct soc15_reg_golden golden_settings_gc_10_3_vangogh[] =
+{
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA0_CLK_CTRL, 0xff7f0fff, 0x30000100),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA1_CLK_CTRL, 0xff7f0fff, 0x7e000100),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x000000e4),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG3, 0xffffffff, 0x00000200),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0xffffffff, 0x00800000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_EXCEPTION_CONTROL, 0x7fff0f1f, 0x00b80000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGB_ADDR_CONFIG, 0x0c1807ff, 0x00000142),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL, 0x1ff1ffff, 0x00000500),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL1_PIPE_STEER, 0x000000ff, 0x000000e4),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_0, 0x77777777, 0x32103210),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2_PIPE_STEER_1, 0x77777777, 0x32103210),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2A_ADDR_MATCH_MASK, 0xffffffff, 0xfffffff3),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xfffffff3),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CM_CTRL1, 0xff8fff0f, 0x580f1008),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL3, 0xf7ffffff, 0x00f80988),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_CL_ENHANCE, 0xf17fffff, 0x01200007),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_TIMEOUT_COUNTER, 0xffffffff, 0x00000800),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_2, 0xffffffbf, 0x00000020),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_CONFIG_CNTL_1_Vangogh, 0xffffffff, 0x00070103),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQG_CONFIG, 0x000017ff, 0x00001000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00400000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmVGT_GS_MAX_WAVE_ID, 0x00000fff, 0x000000ff),
+
+	/* This is not in GDB yet. Don't remove it. It fixes a GPU hang on VanGogh. */
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG,  0x00000020, 0x00000020),
+};
+
+static const struct soc15_reg_golden golden_settings_gc_10_3_4[] =
+{
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0x78000000, 0x78000100),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA0_CLK_CTRL, 0x30000000, 0x30000100),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA1_CLK_CTRL, 0x7e000000, 0x7e000100),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_GCR_CNTL, 0x0007ffff, 0x0000c000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG3, 0x00000280, 0x00000280),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0x07800000, 0x00800000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL, 0x00001d00, 0x00000500),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGE_PC_CNTL, 0x003c0000, 0x00280400),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2A_ADDR_MATCH_MASK, 0xffffffff, 0xffffffcf),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_ADDR_MATCH_MASK, 0xffffffff, 0xffffffcf),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CM_CTRL1, 0x40000000, 0x580f1008),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL3, 0x00040000, 0x00f80988),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_CL_ENHANCE, 0x01000000, 0x01200007),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_TIMEOUT_COUNTER, 0xffffffff, 0x00000800),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_2, 0x00000800, 0x00000820),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_CONFIG, 0x0000001f, 0x00180070),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER0_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER1_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER10_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER11_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER12_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER13_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER14_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER15_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER2_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER3_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER4_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER5_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER6_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER7_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER8_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER9_SELECT, 0xf0f001ff, 0x00000000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0x01030000, 0x01030000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x03a00000, 0x00a00000),
+	SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG,  0x00000020, 0x00000020)
 };
 
 #define DEFAULT_SH_MEM_CONFIG \
@@ -3198,7 +3309,7 @@ static void gfx_v10_0_set_irq_funcs(struct amdgpu_device *adev);
 static void gfx_v10_0_set_gds_init(struct amdgpu_device *adev);
 static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev);
 static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev,
-                                 struct amdgpu_cu_info *cu_info);
+				 struct amdgpu_cu_info *cu_info);
 static uint64_t gfx_v10_0_get_gpu_clock_counter(struct amdgpu_device *adev);
 static void gfx_v10_0_select_se_sh(struct amdgpu_device *adev, u32 se_num,
 				   u32 sh_num, u32 instance);
@@ -3395,7 +3506,16 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev)
 						golden_settings_gc_10_3_2,
 						(const u32)ARRAY_SIZE(golden_settings_gc_10_3_2));
 		break;
-
+	case CHIP_VANGOGH:
+		soc15_program_register_sequence(adev,
+						golden_settings_gc_10_3_vangogh,
+						(const u32)ARRAY_SIZE(golden_settings_gc_10_3_vangogh));
+		break;
+	case CHIP_DIMGREY_CAVEFISH:
+		soc15_program_register_sequence(adev,
+                                                golden_settings_gc_10_3_4,
+                                                (const u32)ARRAY_SIZE(golden_settings_gc_10_3_4));
+		break;
 	default:
 		break;
 	}
@@ -3579,6 +3699,8 @@ static void gfx_v10_0_check_fw_write_wait(struct amdgpu_device *adev)
 		break;
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		adev->gfx.cp_fw_write_wait = true;
 		break;
 	default:
@@ -3646,7 +3768,7 @@ static void gfx_v10_0_check_gfxoff_flag(struct amdgpu_device *adev)
 		if (!gfx_v10_0_navi10_gfxoff_should_enable(adev))
 			adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
 		break;
-	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
 		adev->pm.pp_feature &= ~PP_GFXOFF_MASK;
 		break;
 	default:
@@ -3691,6 +3813,12 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev)
 	case CHIP_NAVY_FLOUNDER:
 		chip_name = "navy_flounder";
 		break;
+	case CHIP_VANGOGH:
+		chip_name = "vangogh";
+		break;
+	case CHIP_DIMGREY_CAVEFISH:
+		chip_name = "dimgrey_cavefish";
+		break;
 	default:
 		BUG();
 	}
@@ -4206,11 +4334,26 @@ static void gfx_v10_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd,
 }
 
 static void gfx_v10_0_select_me_pipe_q(struct amdgpu_device *adev,
-									  u32 me, u32 pipe, u32 q, u32 vm)
- {
-       nv_grbm_select(adev, me, pipe, q, vm);
- }
+				       u32 me, u32 pipe, u32 q, u32 vm)
+{
+	nv_grbm_select(adev, me, pipe, q, vm);
+}
+
+static void gfx_v10_0_update_perfmon_mgcg(struct amdgpu_device *adev,
+					  bool enable)
+{
+	uint32_t data, def;
+
+	data = def = RREG32_SOC15(GC, 0, mmRLC_PERFMON_CLK_CNTL);
+
+	if (enable)
+		data |= RLC_PERFMON_CLK_CNTL__PERFMON_CLOCK_STATE_MASK;
+	else
+		data &= ~RLC_PERFMON_CLK_CNTL__PERFMON_CLOCK_STATE_MASK;
 
+	if (data != def)
+		WREG32_SOC15(GC, 0, mmRLC_PERFMON_CLK_CNTL, data);
+}
 
 static const struct amdgpu_gfx_funcs gfx_v10_0_gfx_funcs = {
 	.get_gpu_clock_counter = &gfx_v10_0_get_gpu_clock_counter,
@@ -4220,6 +4363,7 @@ static const struct amdgpu_gfx_funcs gfx_v10_0_gfx_funcs = {
 	.read_wave_vgprs = &gfx_v10_0_read_wave_vgprs,
 	.select_me_pipe_q = &gfx_v10_0_select_me_pipe_q,
 	.init_spm_golden = &gfx_v10_0_init_spm_golden_registers,
+	.update_perfmon_mgcg = &gfx_v10_0_update_perfmon_mgcg,
 };
 
 static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev)
@@ -4241,6 +4385,8 @@ static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev)
 		break;
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		adev->gfx.config.max_hw_contexts = 8;
 		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
 		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
@@ -4334,7 +4480,8 @@ static int gfx_v10_0_compute_ring_init(struct amdgpu_device *adev, int ring_id,
 	irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP
 		+ ((ring->me - 1) * adev->gfx.mec.num_pipe_per_mec)
 		+ ring->pipe;
-	hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue) ?
+	hw_prio = amdgpu_gfx_is_high_priority_compute_queue(adev, ring->pipe,
+							    ring->queue) ?
 			AMDGPU_GFX_PIPE_PRIO_HIGH : AMDGPU_GFX_PIPE_PRIO_NORMAL;
 	/* type-2 packets are deprecated on MEC, use type-3 instead */
 	r = amdgpu_ring_init(adev, ring, 1024,
@@ -4364,6 +4511,8 @@ static int gfx_v10_0_sw_init(void *handle)
 		break;
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		adev->gfx.me.num_me = 1;
 		adev->gfx.me.num_pipe_per_me = 1;
 		adev->gfx.me.num_queue_per_pipe = 1;
@@ -4623,8 +4772,7 @@ static u32 gfx_v10_0_init_pa_sc_tile_steering_override(struct amdgpu_device *ade
 
 	/* for ASICs that integrates GFX v10.3
 	 * pa_sc_tile_steering_override should be set to 0 */
-	if (adev->asic_type == CHIP_SIENNA_CICHLID ||
-	    adev->asic_type == CHIP_NAVY_FLOUNDER)
+	if (adev->asic_type >= CHIP_SIENNA_CICHLID)
 		return 0;
 
 	/* init num_sc */
@@ -4696,7 +4844,7 @@ static void gfx_v10_0_init_gds_vmid(struct amdgpu_device *adev)
 	 * the driver can enable them for graphics. VMID0 should maintain
 	 * access so that HWS firmware can save/restore entries.
 	 */
-	for (vmid = 1; vmid < 16; vmid++) {
+	for (vmid = 1; vmid < AMDGPU_NUM_VMID; vmid++) {
 		WREG32_SOC15_OFFSET(GC, 0, mmGDS_VMID0_BASE, 2 * vmid, 0);
 		WREG32_SOC15_OFFSET(GC, 0, mmGDS_VMID0_SIZE, 2 * vmid, 0);
 		WREG32_SOC15_OFFSET(GC, 0, mmGDS_GWS_VMID0, vmid, 0);
@@ -4866,7 +5014,7 @@ static int gfx_v10_0_init_csb(struct amdgpu_device *adev)
 	return 0;
 }
 
-void gfx_v10_0_rlc_stop(struct amdgpu_device *adev)
+static void gfx_v10_0_rlc_stop(struct amdgpu_device *adev)
 {
 	u32 tmp = RREG32_SOC15(GC, 0, mmRLC_CNTL);
 
@@ -5847,20 +5995,24 @@ static void gfx_v10_0_cp_gfx_set_doorbell(struct amdgpu_device *adev,
 {
 	u32 tmp;
 
-	tmp = RREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_CONTROL);
-	if (ring->use_doorbell) {
-		tmp = REG_SET_FIELD(tmp, CP_RB_DOORBELL_CONTROL,
-				    DOORBELL_OFFSET, ring->doorbell_index);
-		tmp = REG_SET_FIELD(tmp, CP_RB_DOORBELL_CONTROL,
-				    DOORBELL_EN, 1);
-	} else {
-		tmp = REG_SET_FIELD(tmp, CP_RB_DOORBELL_CONTROL,
-				    DOORBELL_EN, 0);
+	if (!amdgpu_async_gfx_ring) {
+		tmp = RREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_CONTROL);
+		if (ring->use_doorbell) {
+			tmp = REG_SET_FIELD(tmp, CP_RB_DOORBELL_CONTROL,
+						DOORBELL_OFFSET, ring->doorbell_index);
+			tmp = REG_SET_FIELD(tmp, CP_RB_DOORBELL_CONTROL,
+						DOORBELL_EN, 1);
+		} else {
+			tmp = REG_SET_FIELD(tmp, CP_RB_DOORBELL_CONTROL,
+						DOORBELL_EN, 0);
+		}
+		WREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_CONTROL, tmp);
 	}
-	WREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_CONTROL, tmp);
 	switch (adev->asic_type) {
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		tmp = REG_SET_FIELD(0, CP_RB_DOORBELL_RANGE_LOWER,
 				    DOORBELL_RANGE_LOWER_Sienna_Cichlid, ring->doorbell_index);
 		WREG32_SOC15(GC, 0, mmCP_RB_DOORBELL_RANGE_LOWER, tmp);
@@ -5994,6 +6146,8 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
 		switch (adev->asic_type) {
 		case CHIP_SIENNA_CICHLID:
 		case CHIP_NAVY_FLOUNDER:
+		case CHIP_VANGOGH:
+		case CHIP_DIMGREY_CAVEFISH:
 			WREG32_SOC15(GC, 0, mmCP_MEC_CNTL_Sienna_Cichlid, 0);
 			break;
 		default:
@@ -6004,6 +6158,8 @@ static void gfx_v10_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
 		switch (adev->asic_type) {
 		case CHIP_SIENNA_CICHLID:
 		case CHIP_NAVY_FLOUNDER:
+		case CHIP_VANGOGH:
+		case CHIP_DIMGREY_CAVEFISH:
 			WREG32_SOC15(GC, 0, mmCP_MEC_CNTL_Sienna_Cichlid,
 				     (CP_MEC_CNTL__MEC_ME1_HALT_MASK |
 				      CP_MEC_CNTL__MEC_ME2_HALT_MASK));
@@ -6098,6 +6254,8 @@ static void gfx_v10_0_kiq_setting(struct amdgpu_ring *ring)
 	switch (adev->asic_type) {
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		tmp = RREG32_SOC15(GC, 0, mmRLC_CP_SCHEDULERS_Sienna_Cichlid);
 		tmp &= 0xffffff00;
 		tmp |= (ring->me << 5) | (ring->pipe << 3) | (ring->queue);
@@ -6193,6 +6351,11 @@ static int gfx_v10_0_gfx_mqd_init(struct amdgpu_ring *ring)
 				    DOORBELL_EN, 0);
 	mqd->cp_rb_doorbell_control = tmp;
 
+	/*if there are 2 gfx rings, set the lower doorbell range of the first ring,
+	 *otherwise the range of the second ring will override the first ring */
+	if (ring->doorbell_index == adev->doorbell_index.gfx_ring0 << 1)
+		gfx_v10_0_cp_gfx_set_doorbell(adev, ring);
+
 	/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
 	ring->wptr = 0;
 	mqd->cp_gfx_hqd_rptr = RREG32_SOC15(GC, 0, mmCP_GFX_HQD_RPTR);
@@ -6360,7 +6523,8 @@ static void gfx_v10_0_compute_mqd_set_priority(struct amdgpu_ring *ring, struct
 	struct amdgpu_device *adev = ring->adev;
 
 	if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) {
-		if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->queue)) {
+		if (amdgpu_gfx_is_high_priority_compute_queue(adev, ring->pipe,
+							      ring->queue)) {
 			mqd->cp_hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH;
 			mqd->cp_hqd_queue_priority =
 				AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM;
@@ -6806,6 +6970,7 @@ static bool gfx_v10_0_check_grbm_cam_remapping(struct amdgpu_device *adev)
 	switch (adev->asic_type) {
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_DIMGREY_CAVEFISH:
 		data = RREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_Sienna_Cichlid);
 		WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_Sienna_Cichlid, 0);
 		WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_UMD, pattern);
@@ -6818,6 +6983,8 @@ static bool gfx_v10_0_check_grbm_cam_remapping(struct amdgpu_device *adev)
 			return false;
 		}
 		break;
+	case CHIP_VANGOGH:
+		return true;
 	default:
 		data = RREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE);
 		WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE, 0);
@@ -6845,6 +7012,8 @@ static void gfx_v10_0_setup_grbm_cam_remapping(struct amdgpu_device *adev)
 	switch (adev->asic_type) {
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		/* mmVGT_TF_RING_SIZE_UMD -> mmVGT_TF_RING_SIZE */
 		data = (SOC15_REG_OFFSET(GC, 0, mmVGT_TF_RING_SIZE_UMD) <<
 			GRBM_CAM_DATA__CAM_ADDR__SHIFT) |
@@ -6960,6 +7129,18 @@ static void gfx_v10_0_setup_grbm_cam_remapping(struct amdgpu_device *adev)
 	WREG32_SOC15(GC, 0, mmGRBM_CAM_DATA, data);
 }
 
+static void gfx_v10_0_disable_gpa_mode(struct amdgpu_device *adev)
+{
+	uint32_t data;
+	data = RREG32_SOC15(GC, 0, mmCPC_PSP_DEBUG);
+	data |= CPC_PSP_DEBUG__GPA_OVERRIDE_MASK;
+	WREG32_SOC15(GC, 0, mmCPC_PSP_DEBUG, data);
+
+	data = RREG32_SOC15(GC, 0, mmCPG_PSP_DEBUG);
+	data |= CPG_PSP_DEBUG__GPA_OVERRIDE_MASK;
+	WREG32_SOC15(GC, 0, mmCPG_PSP_DEBUG, data);
+}
+
 static int gfx_v10_0_hw_init(void *handle)
 {
 	int r;
@@ -6974,7 +7155,7 @@ static int gfx_v10_0_hw_init(void *handle)
 		 * loaded firstly, so in direct type, it has to load smc ucode
 		 * here before rlc.
 		 */
-		if (adev->smu.ppt_funcs != NULL) {
+		if (adev->smu.ppt_funcs != NULL && !(adev->flags & AMD_IS_APU)) {
 			r = smu_load_microcode(&adev->smu);
 			if (r)
 				return r;
@@ -6985,6 +7166,7 @@ static int gfx_v10_0_hw_init(void *handle)
 				return r;
 			}
 		}
+		gfx_v10_0_disable_gpa_mode(adev);
 	}
 
 	/* if GRBM CAM not remapped, set up the remapping */
@@ -7142,6 +7324,8 @@ static int gfx_v10_0_soft_reset(void *handle)
 	switch (adev->asic_type) {
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		if (REG_GET_FIELD(tmp, GRBM_STATUS2, RLC_BUSY_Sienna_Cichlid))
 			grbm_soft_reset = REG_SET_FIELD(grbm_soft_reset,
 							GRBM_SOFT_RESET,
@@ -7241,13 +7425,16 @@ static int gfx_v10_0_early_init(void *handle)
 		break;
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS_Sienna_Cichlid;
 		break;
 	default:
 		break;
 	}
 
-	adev->gfx.num_compute_rings = amdgpu_num_kcq;
+	adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
+					  AMDGPU_MAX_COMPUTE_RINGS);
 
 	gfx_v10_0_set_kiq_pm4_funcs(adev);
 	gfx_v10_0_set_ring_funcs(adev);
@@ -7294,6 +7481,8 @@ static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev)
 	switch (adev->asic_type) {
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE_Sienna_Cichlid, data);
 
 		/* wait for RLC_SAFE_MODE */
@@ -7326,6 +7515,8 @@ static void gfx_v10_0_unset_safe_mode(struct amdgpu_device *adev)
 	switch (adev->asic_type) {
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE_Sienna_Cichlid, data);
 		break;
 	default:
@@ -7350,6 +7541,7 @@ static void gfx_v10_0_update_medium_grain_clock_gating(struct amdgpu_device *ade
 		/* 1 - RLC_CGTT_MGCG_OVERRIDE */
 		def = data = RREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE);
 		data &= ~(RLC_CGTT_MGCG_OVERRIDE__GRBM_CGTT_SCLK_OVERRIDE_MASK |
+			  RLC_CGTT_MGCG_OVERRIDE__RLC_CGTT_SCLK_OVERRIDE_MASK |
 			  RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK |
 			  RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGLS_OVERRIDE_MASK |
 			  RLC_CGTT_MGCG_OVERRIDE__ENABLE_CGTS_LEGACY_MASK);
@@ -7486,12 +7678,50 @@ static void gfx_v10_0_update_coarse_grain_clock_gating(struct amdgpu_device *ade
 	}
 }
 
+static void gfx_v10_0_update_fine_grain_clock_gating(struct amdgpu_device *adev,
+						      bool enable)
+{
+	uint32_t def, data;
+
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_FGCG)) {
+		def = data = RREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE);
+		/* unset FGCG override */
+		data &= ~RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK;
+		/* update FGCG override bits */
+		if (def != data)
+			WREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE, data);
+
+		def = data = RREG32_SOC15(GC, 0, mmRLC_CLK_CNTL);
+		/* unset RLC SRAM CLK GATER override */
+		data &= ~RLC_CLK_CNTL__RLC_SRAM_CLK_GATER_OVERRIDE_MASK;
+		/* update RLC SRAM CLK GATER override bits */
+		if (def != data)
+			WREG32_SOC15(GC, 0, mmRLC_CLK_CNTL, data);
+	} else {
+		def = data = RREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE);
+		/* reset FGCG bits */
+		data |= RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK;
+		/* disable FGCG*/
+		if (def != data)
+			WREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE, data);
+
+		def = data = RREG32_SOC15(GC, 0, mmRLC_CLK_CNTL);
+		/* reset RLC SRAM CLK GATER bits */
+		data |= RLC_CLK_CNTL__RLC_SRAM_CLK_GATER_OVERRIDE_MASK;
+		/* disable RLC SRAM CLK*/
+		if (def != data)
+			WREG32_SOC15(GC, 0, mmRLC_CLK_CNTL, data);
+	}
+}
+
 static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 					    bool enable)
 {
 	amdgpu_gfx_rlc_enter_safe_mode(adev);
 
 	if (enable) {
+		/* enable FGCG firstly*/
+		gfx_v10_0_update_fine_grain_clock_gating(adev, enable);
 		/* CGCG/CGLS should be enabled after MGCG/MGLS
 		 * ===  MGCG + MGLS ===
 		 */
@@ -7509,6 +7739,8 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 		gfx_v10_0_update_3d_clock_gating(adev, enable);
 		/* ===  MGCG + MGLS === */
 		gfx_v10_0_update_medium_grain_clock_gating(adev, enable);
+		/* disable fgcg at last*/
+		gfx_v10_0_update_fine_grain_clock_gating(adev, enable);
 	}
 
 	if (adev->cg_flags &
@@ -7570,6 +7802,27 @@ static bool gfx_v10_0_is_rlcg_access_range(struct amdgpu_device *adev, u32 offse
 	return gfx_v10_0_check_rlcg_range(adev, offset, NULL, 0);
 }
 
+static void gfx_v10_cntl_power_gating(struct amdgpu_device *adev, bool enable)
+{
+	u32 data = RREG32_SOC15(GC, 0, mmRLC_PG_CNTL);
+
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG))
+		data |= RLC_PG_CNTL__GFX_POWER_GATING_ENABLE_MASK;
+	else
+		data &= ~RLC_PG_CNTL__GFX_POWER_GATING_ENABLE_MASK;
+
+	WREG32_SOC15(GC, 0, mmRLC_PG_CNTL, data);
+}
+
+static void gfx_v10_cntl_pg(struct amdgpu_device *adev, bool enable)
+{
+	amdgpu_gfx_rlc_enter_safe_mode(adev);
+
+	gfx_v10_cntl_power_gating(adev, enable);
+
+	amdgpu_gfx_rlc_exit_safe_mode(adev);
+}
+
 static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs = {
 	.is_rlc_enabled = gfx_v10_0_is_rlc_enabled,
 	.set_safe_mode = gfx_v10_0_set_safe_mode,
@@ -7615,8 +7868,12 @@ static int gfx_v10_0_set_powergating_state(void *handle,
 	case CHIP_NAVI12:
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_DIMGREY_CAVEFISH:
 		amdgpu_gfx_off_ctrl(adev, enable);
 		break;
+	case CHIP_VANGOGH:
+		gfx_v10_cntl_pg(adev, enable);
+		break;
 	default:
 		break;
 	}
@@ -7637,6 +7894,8 @@ static int gfx_v10_0_set_clockgating_state(void *handle,
 	case CHIP_NAVI12:
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		gfx_v10_0_update_gfx_clock_gating(adev,
 						 state == AMD_CG_STATE_GATE);
 		break;
@@ -7651,6 +7910,11 @@ static void gfx_v10_0_get_clockgating_state(void *handle, u32 *flags)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int data;
 
+	/* AMD_CG_SUPPORT_GFX_FGCG */
+	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE));
+	if (!(data & RLC_CGTT_MGCG_OVERRIDE__GFXIP_FGCG_OVERRIDE_MASK))
+		*flags |= AMD_CG_SUPPORT_GFX_FGCG;
+
 	/* AMD_CG_SUPPORT_GFX_MGCG */
 	data = RREG32_KIQ(SOC15_REG_OFFSET(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE));
 	if (!(data & RLC_CGTT_MGCG_OVERRIDE__GFXIP_MGCG_OVERRIDE_MASK))
@@ -8400,6 +8664,7 @@ static int gfx_v10_0_set_priv_inst_fault_state(struct amdgpu_device *adev,
 		WREG32_FIELD15(GC, 0, CP_INT_CNTL_RING0,
 			       PRIV_INSTR_INT_ENABLE,
 			       state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
+		break;
 	default:
 		break;
 	}
@@ -8739,6 +9004,8 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev)
 	case CHIP_NAVI14:
 	case CHIP_SIENNA_CICHLID:
 	case CHIP_NAVY_FLOUNDER:
+	case CHIP_VANGOGH:
+	case CHIP_DIMGREY_CAVEFISH:
 		adev->gfx.rlc.funcs = &gfx_v10_0_rlc_funcs;
 		break;
 	case CHIP_NAVI12: