Merge tag 'amd-drm-next-6.7-2023-11-10' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.7-2023-11-10: amdgpu: - SR-IOV fixes - DMCUB fixes - DCN3.5 fixes - DP2 fixes - SubVP fixes - SMU14 fixes - SDMA4.x fixes - Suspend/resume fixes - AGP regression fix - UAF fixes for some error cases - SMU 13.0.6 fixes - Documentation fixes - RAS fixes - Hotplug fixes - Scheduling entity ordering fix - GPUVM fixes Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> From: Alex Deucher <alexander.deucher@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231110190703.4741-1-alexander.deucher@amd.com
author: Daniel Vetter <daniel.vetter@ffwll.ch> 2023-11-10 20:51:37 +0100
committer: Daniel Vetter <daniel.vetter@ffwll.ch> 2023-11-10 20:51:38 +0100
commit: 03df0fc007ca4713fa1e716329af556f981807e4 (patch)
tree: afb3271fe3d62fb9516ab9d468f66e42affc6858 /drivers
parent: aec3e2e23b08f188c22f36c4108467f80e980b15 (diff)
parent: 8473bfdcb5b1a32fd05629c4535ccacd73bc5567 (diff)
85 files changed, 1688 insertions, 773 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 8b4ca2576a3d..afec09930efa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1159,11 +1159,18 @@ uint32_t amdgpu_device_rreg(struct amdgpu_device *adev,
 			    uint32_t reg, uint32_t acc_flags);
 u32 amdgpu_device_indirect_rreg_ext(struct amdgpu_device *adev,
 				    u64 reg_addr);
+uint32_t amdgpu_device_xcc_rreg(struct amdgpu_device *adev,
+				uint32_t reg, uint32_t acc_flags,
+				uint32_t xcc_id);
 void amdgpu_device_wreg(struct amdgpu_device *adev,
 			uint32_t reg, uint32_t v,
 			uint32_t acc_flags);
 void amdgpu_device_indirect_wreg_ext(struct amdgpu_device *adev,
 				     u64 reg_addr, u32 reg_data);
+void amdgpu_device_xcc_wreg(struct amdgpu_device *adev,
+			    uint32_t reg, uint32_t v,
+			    uint32_t acc_flags,
+			    uint32_t xcc_id);
 void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev,
 			     uint32_t reg, uint32_t v, uint32_t xcc_id);
 void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value);
@@ -1204,8 +1211,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 #define RREG32_NO_KIQ(reg) amdgpu_device_rreg(adev, (reg), AMDGPU_REGS_NO_KIQ)
 #define WREG32_NO_KIQ(reg, v) amdgpu_device_wreg(adev, (reg), (v), AMDGPU_REGS_NO_KIQ)
 
-#define RREG32_KIQ(reg) amdgpu_kiq_rreg(adev, (reg))
-#define WREG32_KIQ(reg, v) amdgpu_kiq_wreg(adev, (reg), (v))
+#define RREG32_KIQ(reg) amdgpu_kiq_rreg(adev, (reg), 0)
+#define WREG32_KIQ(reg, v) amdgpu_kiq_wreg(adev, (reg), (v), 0)
 
 #define RREG8(reg) amdgpu_mm_rreg8(adev, (reg))
 #define WREG8(reg, v) amdgpu_mm_wreg8(adev, (reg), (v))
@@ -1215,6 +1222,8 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 #define WREG32(reg, v) amdgpu_device_wreg(adev, (reg), (v), 0)
 #define REG_SET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
 #define REG_GET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
+#define RREG32_XCC(reg, inst) amdgpu_device_xcc_rreg(adev, (reg), 0, inst)
+#define WREG32_XCC(reg, v, inst) amdgpu_device_xcc_wreg(adev, (reg), (v), 0, inst)
 #define RREG32_PCIE(reg) adev->pcie_rreg(adev, (reg))
 #define WREG32_PCIE(reg, v) adev->pcie_wreg(adev, (reg), (v))
 #define RREG32_PCIE_PORT(reg) adev->pciep_rreg(adev, (reg))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index d62e49758635..e550067e5c5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -1497,6 +1497,9 @@ bool amdgpu_acpi_is_s0ix_active(struct amdgpu_device *adev)
 	if (adev->asic_type < CHIP_RAVEN)
 		return false;
 
+	if (!(adev->pm.pp_feature & PP_GFXOFF_MASK))
+		return false;
+
 	/*
 	 * If ACPI_FADT_LOW_POWER_S0 is not set in the FADT, it is generally
 	 * risky to do any special firmware-related preparations for entering
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
index 490c8f5ddb60..f6598b9e4faa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c
@@ -300,14 +300,13 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 	hqd_end = SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_AQL_DISPATCH_ID_HI);
 
 	for (reg = hqd_base; reg <= hqd_end; reg++)
-		WREG32_RLC(reg, mqd_hqd[reg - hqd_base]);
+		WREG32_XCC(reg, mqd_hqd[reg - hqd_base], inst);
 
 
 	/* Activate doorbell logic before triggering WPTR poll. */
 	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
 			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_DOORBELL_CONTROL),
-				data);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), regCP_HQD_PQ_DOORBELL_CONTROL, data);
 
 	if (wptr) {
 		/* Don't read wptr with get_user because the user
@@ -336,27 +335,24 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd,
 		guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1);
 		guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32;
 
-		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_LO),
-		       lower_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_HI),
-		       upper_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_POLL_ADDR),
-		       lower_32_bits((uintptr_t)wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst),
-			regCP_HQD_PQ_WPTR_POLL_ADDR_HI),
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_LO,
+			lower_32_bits(guessed_wptr));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_HI,
+			upper_32_bits(guessed_wptr));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_POLL_ADDR,
+			lower_32_bits((uintptr_t)wptr));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), regCP_HQD_PQ_WPTR_POLL_ADDR_HI,
 			upper_32_bits((uintptr_t)wptr));
-		WREG32(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_PQ_WPTR_POLL_CNTL1),
-		       (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id,
-			       queue_id));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), regCP_PQ_WPTR_POLL_CNTL1,
+			(uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id, queue_id));
 	}
 
 	/* Start the EOP fetcher */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_EOP_RPTR),
-	       REG_SET_FIELD(m->cp_hqd_eop_rptr,
-			     CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), regCP_HQD_EOP_RPTR,
+	       REG_SET_FIELD(m->cp_hqd_eop_rptr, CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
 
 	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), regCP_HQD_ACTIVE), data);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), regCP_HQD_ACTIVE, data);
 
 	kgd_gfx_v9_release_queue(adev, inst);
 
@@ -494,15 +490,15 @@ static uint32_t kgd_gfx_v9_4_3_set_address_watch(
 			VALID,
 			1);
 
-	WREG32_RLC((SOC15_REG_OFFSET(GC, GET_INST(GC, inst),
+	WREG32_XCC((SOC15_REG_OFFSET(GC, GET_INST(GC, inst),
 			regTCP_WATCH0_ADDR_H) +
 			(watch_id * TCP_WATCH_STRIDE)),
-			watch_address_high);
+			watch_address_high, inst);
 
-	WREG32_RLC((SOC15_REG_OFFSET(GC, GET_INST(GC, inst),
+	WREG32_XCC((SOC15_REG_OFFSET(GC, GET_INST(GC, inst),
 			regTCP_WATCH0_ADDR_L) +
 			(watch_id * TCP_WATCH_STRIDE)),
-			watch_address_low);
+			watch_address_low, inst);
 
 	return watch_address_cntl;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 51011e8ee90d..00fbc0f44c92 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -91,8 +91,8 @@ void kgd_gfx_v9_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmi
 {
 	kgd_gfx_v9_lock_srbm(adev, 0, 0, 0, vmid, inst);
 
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmSH_MEM_CONFIG), sh_mem_config);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmSH_MEM_BASES), sh_mem_bases);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmSH_MEM_CONFIG, sh_mem_config);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmSH_MEM_BASES, sh_mem_bases);
 	/* APE1 no longer exists on GFX9 */
 
 	kgd_gfx_v9_unlock_srbm(adev, inst);
@@ -239,14 +239,13 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 
 	for (reg = hqd_base;
 	     reg <= SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI); reg++)
-		WREG32_RLC(reg, mqd_hqd[reg - hqd_base]);
+		WREG32_XCC(reg, mqd_hqd[reg - hqd_base], inst);
 
 
 	/* Activate doorbell logic before triggering WPTR poll. */
 	data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control,
 			     CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_DOORBELL_CONTROL),
-					data);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_HQD_PQ_DOORBELL_CONTROL, data);
 
 	if (wptr) {
 		/* Don't read wptr with get_user because the user
@@ -275,25 +274,24 @@ int kgd_gfx_v9_hqd_load(struct amdgpu_device *adev, void *mqd,
 		guessed_wptr += m->cp_hqd_pq_wptr_lo & ~(queue_size - 1);
 		guessed_wptr += (uint64_t)m->cp_hqd_pq_wptr_hi << 32;
 
-		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_LO),
-		       lower_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI),
-		       upper_32_bits(guessed_wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_POLL_ADDR),
-		       lower_32_bits((uintptr_t)wptr));
-		WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_POLL_ADDR_HI),
-		       upper_32_bits((uintptr_t)wptr));
-		WREG32_SOC15(GC, GET_INST(GC, inst), mmCP_PQ_WPTR_POLL_CNTL1,
-		       (uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id, queue_id));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_LO,
+			lower_32_bits(guessed_wptr));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_HI,
+			upper_32_bits(guessed_wptr));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_POLL_ADDR,
+			lower_32_bits((uintptr_t)wptr));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_HQD_PQ_WPTR_POLL_ADDR_HI,
+			upper_32_bits((uintptr_t)wptr));
+		WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_PQ_WPTR_POLL_CNTL1,
+			(uint32_t)kgd_gfx_v9_get_queue_mask(adev, pipe_id, queue_id));
 	}
 
 	/* Start the EOP fetcher */
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_EOP_RPTR),
-	       REG_SET_FIELD(m->cp_hqd_eop_rptr,
-			     CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_HQD_EOP_RPTR,
+	       REG_SET_FIELD(m->cp_hqd_eop_rptr, CP_HQD_EOP_RPTR, INIT_FETCHER, 1));
 
 	data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1);
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE), data);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_HQD_ACTIVE, data);
 
 	kgd_gfx_v9_release_queue(adev, inst);
 
@@ -556,7 +554,7 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd,
 		break;
 	}
 
-	WREG32_RLC(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), mmCP_HQD_DEQUEUE_REQUEST), type);
+	WREG32_SOC15_RLC(GC, GET_INST(GC, inst), mmCP_HQD_DEQUEUE_REQUEST, type);
 
 	end_jiffies = (utimeout * HZ / 1000) + jiffies;
 	while (true) {
@@ -908,8 +906,8 @@ void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev,
 					uint32_t inst)
 
 {
-	*wait_times = RREG32(SOC15_REG_OFFSET(GC, GET_INST(GC, inst),
-			mmCP_IQ_WAIT_TIME2));
+	*wait_times = RREG32_SOC15_RLC(GC, GET_INST(GC, inst),
+			mmCP_IQ_WAIT_TIME2);
 }
 
 void kgd_gfx_v9_set_vm_context_page_table_base(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index 781e5c5ce04d..702f6610d024 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -172,6 +172,7 @@ int amdgpu_bo_list_get(struct amdgpu_fpriv *fpriv, int id,
 	}
 
 	rcu_read_unlock();
+	*result = NULL;
 	return -ENOENT;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 86f88512eb7e..df3ecfa9e13f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -1415,7 +1415,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		if (r == -ENOMEM)
 			DRM_ERROR("Not enough memory for command submission!\n");
 		else if (r != -ERESTARTSYS && r != -EAGAIN)
-			DRM_ERROR("Failed to process the buffer list %d!\n", r);
+			DRM_DEBUG("Failed to process the buffer list %d!\n", r);
 		goto error_fini;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index fd8cd8e2d3f2..7eeaf0aa7f81 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -73,6 +73,7 @@
 #include "amdgpu_pmu.h"
 #include "amdgpu_fru_eeprom.h"
 #include "amdgpu_reset.h"
+#include "amdgpu_virt.h"
 
 #include <linux/suspend.h>
 #include <drm/task_barrier.h>
@@ -472,7 +473,7 @@ uint32_t amdgpu_device_rreg(struct amdgpu_device *adev,
 		if (!(acc_flags & AMDGPU_REGS_NO_KIQ) &&
 		    amdgpu_sriov_runtime(adev) &&
 		    down_read_trylock(&adev->reset_domain->sem)) {
-			ret = amdgpu_kiq_rreg(adev, reg);
+			ret = amdgpu_kiq_rreg(adev, reg, 0);
 			up_read(&adev->reset_domain->sem);
 		} else {
 			ret = readl(((void __iomem *)adev->rmmio) + (reg * 4));
@@ -509,6 +510,49 @@ uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset)
 	BUG();
 }
 
+
+/**
+ * amdgpu_device_xcc_rreg - read a memory mapped IO or indirect register with specific XCC
+ *
+ * @adev: amdgpu_device pointer
+ * @reg: dword aligned register offset
+ * @acc_flags: access flags which require special behavior
+ * @xcc_id: xcc accelerated compute core id
+ *
+ * Returns the 32 bit value from the offset specified.
+ */
+uint32_t amdgpu_device_xcc_rreg(struct amdgpu_device *adev,
+				uint32_t reg, uint32_t acc_flags,
+				uint32_t xcc_id)
+{
+	uint32_t ret, rlcg_flag;
+
+	if (amdgpu_device_skip_hw_access(adev))
+		return 0;
+
+	if ((reg * 4) < adev->rmmio_size) {
+		if (amdgpu_sriov_vf(adev) &&
+		    !amdgpu_sriov_runtime(adev) &&
+		    adev->gfx.rlc.rlcg_reg_access_supported &&
+		    amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags,
+							 GC_HWIP, false,
+							 &rlcg_flag)) {
+			ret = amdgpu_virt_rlcg_reg_rw(adev, reg, 0, rlcg_flag, xcc_id);
+		} else if (!(acc_flags & AMDGPU_REGS_NO_KIQ) &&
+		    amdgpu_sriov_runtime(adev) &&
+		    down_read_trylock(&adev->reset_domain->sem)) {
+			ret = amdgpu_kiq_rreg(adev, reg, xcc_id);
+			up_read(&adev->reset_domain->sem);
+		} else {
+			ret = readl(((void __iomem *)adev->rmmio) + (reg * 4));
+		}
+	} else {
+		ret = adev->pcie_rreg(adev, reg * 4);
+	}
+
+	return ret;
+}
+
 /*
  * MMIO register write with bytes helper functions
  * @offset:bytes offset from MMIO start
@@ -556,7 +600,7 @@ void amdgpu_device_wreg(struct amdgpu_device *adev,
 		if (!(acc_flags & AMDGPU_REGS_NO_KIQ) &&
 		    amdgpu_sriov_runtime(adev) &&
 		    down_read_trylock(&adev->reset_domain->sem)) {
-			amdgpu_kiq_wreg(adev, reg, v);
+			amdgpu_kiq_wreg(adev, reg, v, 0);
 			up_read(&adev->reset_domain->sem);
 		} else {
 			writel(v, ((void __iomem *)adev->rmmio) + (reg * 4));
@@ -598,6 +642,47 @@ void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev,
 }
 
 /**
+ * amdgpu_device_xcc_wreg - write to a memory mapped IO or indirect register with specific XCC
+ *
+ * @adev: amdgpu_device pointer
+ * @reg: dword aligned register offset
+ * @v: 32 bit value to write to the register
+ * @acc_flags: access flags which require special behavior
+ * @xcc_id: xcc accelerated compute core id
+ *
+ * Writes the value specified to the offset specified.
+ */
+void amdgpu_device_xcc_wreg(struct amdgpu_device *adev,
+			uint32_t reg, uint32_t v,
+			uint32_t acc_flags, uint32_t xcc_id)
+{
+	uint32_t rlcg_flag;
+
+	if (amdgpu_device_skip_hw_access(adev))
+		return;
+
+	if ((reg * 4) < adev->rmmio_size) {
+		if (amdgpu_sriov_vf(adev) &&
+		    !amdgpu_sriov_runtime(adev) &&
+		    adev->gfx.rlc.rlcg_reg_access_supported &&
+		    amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags,
+							 GC_HWIP, true,
+							 &rlcg_flag)) {
+			amdgpu_virt_rlcg_reg_rw(adev, reg, v, rlcg_flag, xcc_id);
+		} else if (!(acc_flags & AMDGPU_REGS_NO_KIQ) &&
+		    amdgpu_sriov_runtime(adev) &&
+		    down_read_trylock(&adev->reset_domain->sem)) {
+			amdgpu_kiq_wreg(adev, reg, v, xcc_id);
+			up_read(&adev->reset_domain->sem);
+		} else {
+			writel(v, ((void __iomem *)adev->rmmio) + (reg * 4));
+		}
+	} else {
+		adev->pcie_wreg(adev, reg * 4, v);
+	}
+}
+
+/**
  * amdgpu_device_indirect_rreg - read an indirect register
  *
  * @adev: amdgpu_device pointer
@@ -2499,6 +2584,18 @@ static int amdgpu_device_init_schedulers(struct amdgpu_device *adev)
 				  ring->name);
 			return r;
 		}
+		r = amdgpu_uvd_entity_init(adev, ring);
+		if (r) {
+			DRM_ERROR("Failed to create UVD scheduling entity on ring %s.\n",
+				  ring->name);
+			return r;
+		}
+		r = amdgpu_vce_entity_init(adev, ring);
+		if (r) {
+			DRM_ERROR("Failed to create VCE scheduling entity on ring %s.\n",
+				  ring->name);
+			return r;
+		}
 	}
 
 	amdgpu_xcp_update_partition_sched_list(adev);
@@ -4486,19 +4583,18 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 	}
 	amdgpu_fence_driver_hw_init(adev);
 
-	r = amdgpu_device_ip_late_init(adev);
-	if (r)
-		goto exit;
-
-	queue_delayed_work(system_wq, &adev->delayed_init_work,
-			   msecs_to_jiffies(AMDGPU_RESUME_MS));
-
 	if (!adev->in_s0ix) {
 		r = amdgpu_amdkfd_resume(adev, adev->in_runpm);
 		if (r)
 			goto exit;
 	}
 
+	r = amdgpu_device_ip_late_init(adev);
+	if (r)
+		goto exit;
+
+	queue_delayed_work(system_wq, &adev->delayed_init_work,
+			   msecs_to_jiffies(AMDGPU_RESUME_MS));
 exit:
 	if (amdgpu_sriov_vf(adev)) {
 		amdgpu_virt_init_data_exchange(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index a2a29dcb2422..b9674c57c436 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -931,12 +931,12 @@ void amdgpu_gfx_ras_error_func(struct amdgpu_device *adev,
 		func(adev, ras_error_status, i);
 }
 
-uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg)
+uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg, uint32_t xcc_id)
 {
 	signed long r, cnt = 0;
 	unsigned long flags;
 	uint32_t seq, reg_val_offs = 0, value = 0;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 	struct amdgpu_ring *ring = &kiq->ring;
 
 	if (amdgpu_device_skip_hw_access(adev))
@@ -999,12 +999,12 @@ failed_kiq_read:
 	return ~0;
 }
 
-void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v)
+void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t xcc_id)
 {
 	signed long r, cnt = 0;
 	unsigned long flags;
 	uint32_t seq;
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id];
 	struct amdgpu_ring *ring = &kiq->ring;
 
 	BUG_ON(!ring->funcs->emit_wreg);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 7088c5015675..f23bafec71c5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -521,8 +521,8 @@ int amdgpu_gfx_process_ras_data_cb(struct amdgpu_device *adev,
 int amdgpu_gfx_cp_ecc_error_irq(struct amdgpu_device *adev,
 				  struct amdgpu_irq_src *source,
 				  struct amdgpu_iv_entry *entry);
-uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg);
-void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v);
+uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg, uint32_t xcc_id);
+void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v, uint32_t xcc_id);
 int amdgpu_gfx_get_num_kcq(struct amdgpu_device *adev);
 void amdgpu_gfx_cp_init_microcode(struct amdgpu_device *adev, uint32_t ucode_id);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 2dce338b0f1e..5f71414190e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -826,7 +826,10 @@ void amdgpu_gmc_noretry_set(struct amdgpu_device *adev)
 				gc_ver == IP_VERSION(9, 4, 3) ||
 				gc_ver >= IP_VERSION(10, 3, 0));
 
-	gmc->noretry = (amdgpu_noretry == -1) ? noretry_default : amdgpu_noretry;
+	if (!amdgpu_sriov_xnack_support(adev))
+		gmc->noretry = 1;
+	else
+		gmc->noretry = (amdgpu_noretry == -1) ? noretry_default : amdgpu_noretry;
 }
 
 void amdgpu_gmc_set_vm_fault_masks(struct amdgpu_device *adev, int hub_type,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
index 5a828c175e3a..cf33eb219e25 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
@@ -143,6 +143,46 @@ int amdgpu_mca_mpio_ras_sw_init(struct amdgpu_device *adev)
 	return 0;
 }
 
+void amdgpu_mca_bank_set_init(struct mca_bank_set *mca_set)
+{
+	if (!mca_set)
+		return;
+
+	memset(mca_set, 0, sizeof(*mca_set));
+	INIT_LIST_HEAD(&mca_set->list);
+}
+
+int amdgpu_mca_bank_set_add_entry(struct mca_bank_set *mca_set, struct mca_bank_entry *entry)
+{
+	struct mca_bank_node *node;
+
+	if (!entry)
+		return -EINVAL;
+
+	node = kvzalloc(sizeof(*node), GFP_KERNEL);
+	if (!node)
+		return -ENOMEM;
+
+	memcpy(&node->entry, entry, sizeof(*entry));
+
+	INIT_LIST_HEAD(&node->node);
+	list_add_tail(&node->node, &mca_set->list);
+
+	mca_set->nr_entries++;
+
+	return 0;
+}
+
+void amdgpu_mca_bank_set_release(struct mca_bank_set *mca_set)
+{
+	struct mca_bank_node *node, *tmp;
+
+	list_for_each_entry_safe(node, tmp, &mca_set->list, node) {
+		list_del(&node->node);
+		kvfree(node);
+	}
+}
+
 void amdgpu_mca_smu_init_funcs(struct amdgpu_device *adev, const struct amdgpu_mca_smu_funcs *mca_funcs)
 {
 	struct amdgpu_mca *mca = &adev->mca;
@@ -160,6 +200,65 @@ int amdgpu_mca_smu_set_debug_mode(struct amdgpu_device *adev, bool enable)
 	return -EOPNOTSUPP;
 }
 
+static void amdgpu_mca_smu_mca_bank_dump(struct amdgpu_device *adev, int idx, struct mca_bank_entry *entry)
+{
+	dev_info(adev->dev, "[Hardware error] Accelerator Check Architecture events logged\n");
+	dev_info(adev->dev, "[Hardware error] aca entry[%02d].STATUS=0x%016llx\n",
+		 idx, entry->regs[MCA_REG_IDX_STATUS]);
+	dev_info(adev->dev, "[Hardware error] aca entry[%02d].ADDR=0x%016llx\n",
+		 idx, entry->regs[MCA_REG_IDX_ADDR]);
+	dev_info(adev->dev, "[Hardware error] aca entry[%02d].MISC0=0x%016llx\n",
+		 idx, entry->regs[MCA_REG_IDX_MISC0]);
+	dev_info(adev->dev, "[Hardware error] aca entry[%02d].IPID=0x%016llx\n",
+		 idx, entry->regs[MCA_REG_IDX_IPID]);
+	dev_info(adev->dev, "[Hardware error] aca entry[%02d].SYND=0x%016llx\n",
+		 idx, entry->regs[MCA_REG_IDX_SYND]);
+}
+
+int amdgpu_mca_smu_log_ras_error(struct amdgpu_device *adev, enum amdgpu_ras_block blk, enum amdgpu_mca_error_type type, struct ras_err_data *err_data)
+{
+	struct amdgpu_smuio_mcm_config_info mcm_info;
+	struct mca_bank_set mca_set;
+	struct mca_bank_node *node;
+	struct mca_bank_entry *entry;
+	uint32_t count;
+	int ret, i = 0;
+
+	amdgpu_mca_bank_set_init(&mca_set);
+
+	ret = amdgpu_mca_smu_get_mca_set(adev, blk, type, &mca_set);
+	if (ret)
+		goto out_mca_release;
+
+	list_for_each_entry(node, &mca_set.list, node) {
+		entry = &node->entry;
+
+		amdgpu_mca_smu_mca_bank_dump(adev, i++, entry);
+
+		count = 0;
+		ret = amdgpu_mca_smu_parse_mca_error_count(adev, blk, type, entry, &count);
+		if (ret)
+			goto out_mca_release;
+
+		if (!count)
+			continue;
+
+		mcm_info.socket_id = entry->info.socket_id;
+		mcm_info.die_id = entry->info.aid;
+
+		if (type == AMDGPU_MCA_ERROR_TYPE_UE)
+			amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, (uint64_t)count);
+		else
+			amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, (uint64_t)count);
+	}
+
+out_mca_release:
+	amdgpu_mca_bank_set_release(&mca_set);
+
+	return ret;
+}
+
+
 int amdgpu_mca_smu_get_valid_mca_count(struct amdgpu_device *adev, enum amdgpu_mca_error_type type, uint32_t *count)
 {
 	const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs;
@@ -173,17 +272,77 @@ int amdgpu_mca_smu_get_valid_mca_count(struct amdgpu_device *adev, enum amdgpu_m
 	return -EOPNOTSUPP;
 }
 
-int amdgpu_mca_smu_get_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
-				   enum amdgpu_mca_error_type type, uint32_t *count)
+int amdgpu_mca_smu_get_mca_set_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
+					    enum amdgpu_mca_error_type type, uint32_t *total)
 {
 	const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs;
-	if (!count)
+	struct mca_bank_set mca_set;
+	struct mca_bank_node *node;
+	struct mca_bank_entry *entry;
+	uint32_t count;
+	int ret;
+
+	if (!total)
 		return -EINVAL;
 
-	if (mca_funcs && mca_funcs->mca_get_error_count)
-		return mca_funcs->mca_get_error_count(adev, blk, type, count);
+	if (!mca_funcs)
+		return -EOPNOTSUPP;
 
-	return -EOPNOTSUPP;
+	if (!mca_funcs->mca_get_ras_mca_set || !mca_funcs->mca_get_valid_mca_count)
+		return -EOPNOTSUPP;
+
+	amdgpu_mca_bank_set_init(&mca_set);
+
+	ret = mca_funcs->mca_get_ras_mca_set(adev, blk, type, &mca_set);
+	if (ret)
+		goto err_mca_set_release;
+
+	*total = 0;
+	list_for_each_entry(node, &mca_set.list, node) {
+		entry = &node->entry;
+
+		count = 0;
+		ret = mca_funcs->mca_parse_mca_error_count(adev, blk, type, entry, &count);
+		if (ret)
+			goto err_mca_set_release;
+
+		*total += count;
+	}
+
+err_mca_set_release:
+	amdgpu_mca_bank_set_release(&mca_set);
+
+	return ret;
+}
+
+int amdgpu_mca_smu_parse_mca_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
+					 enum amdgpu_mca_error_type type, struct mca_bank_entry *entry, uint32_t *count)
+{
+	const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs;
+	if (!count || !entry)
+		return -EINVAL;
+
+	if (!mca_funcs || !mca_funcs->mca_parse_mca_error_count)
+		return -EOPNOTSUPP;
+
+
+	return mca_funcs->mca_parse_mca_error_count(adev, blk, type, entry, count);
+}
+
+int amdgpu_mca_smu_get_mca_set(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
+			       enum amdgpu_mca_error_type type, struct mca_bank_set *mca_set)
+{
+	const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs;
+
+	if (!mca_set)
+		return -EINVAL;
+
+	if (!mca_funcs || !mca_funcs->mca_get_ras_mca_set)
+		return -EOPNOTSUPP;
+
+	WARN_ON(!list_empty(&mca_set->list));
+
+	return mca_funcs->mca_get_ras_mca_set(adev, blk, type, mca_set);
 }
 
 int amdgpu_mca_smu_get_mca_entry(struct amdgpu_device *adev, enum amdgpu_mca_error_type type,
@@ -230,14 +389,21 @@ static int amdgpu_mca_smu_debug_mode_set(void *data, u64 val)
 static void mca_dump_entry(struct seq_file *m, struct mca_bank_entry *entry)
 {
 	int i, idx = entry->idx;
+	int reg_idx_array[] = {
+		MCA_REG_IDX_STATUS,
+		MCA_REG_IDX_ADDR,
+		MCA_REG_IDX_MISC0,
+		MCA_REG_IDX_IPID,
+		MCA_REG_IDX_SYND,
+	};
 
 	seq_printf(m, "mca entry[%d].type: %s\n", idx, entry->type == AMDGPU_MCA_ERROR_TYPE_UE ? "UE" : "CE");
 	seq_printf(m, "mca entry[%d].ip: %d\n", idx, entry->ip);
 	seq_printf(m, "mca entry[%d].info: socketid:%d aid:%d hwid:0x%03x mcatype:0x%04x\n",
 		   idx, entry->info.socket_id, entry->info.aid, entry->info.hwid, entry->info.mcatype);
 
-	for (i = 0; i < ARRAY_SIZE(entry->regs); i++)
-		seq_printf(m, "mca entry[%d].regs[%d]: 0x%016llx\n", idx, i, entry->regs[i]);
+	for (i = 0; i < ARRAY_SIZE(reg_idx_array); i++)
+		seq_printf(m, "mca entry[%d].regs[%d]: 0x%016llx\n", idx, reg_idx_array[i], entry->regs[reg_idx_array[i]]);
 }
 
 static int mca_dump_show(struct seq_file *m, enum amdgpu_mca_error_type type)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h
index 28ad463cf5c9..2b488fcf2f95 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h
@@ -25,6 +25,27 @@
 
 #define MCA_MAX_REGS_COUNT	(16)
 
+#define MCA_REG_FIELD(x, h, l)			(((x) & GENMASK_ULL(h, l)) >> l)
+#define MCA_REG__STATUS__VAL(x)			MCA_REG_FIELD(x, 63, 63)
+#define MCA_REG__STATUS__OVERFLOW(x)		MCA_REG_FIELD(x, 62, 62)
+#define MCA_REG__STATUS__UC(x)			MCA_REG_FIELD(x, 61, 61)
+#define MCA_REG__STATUS__EN(x)			MCA_REG_FIELD(x, 60, 60)
+#define MCA_REG__STATUS__MISCV(x)		MCA_REG_FIELD(x, 59, 59)
+#define MCA_REG__STATUS__ADDRV(x)		MCA_REG_FIELD(x, 58, 58)
+#define MCA_REG__STATUS__PCC(x)			MCA_REG_FIELD(x, 57, 57)
+#define MCA_REG__STATUS__ERRCOREIDVAL(x)	MCA_REG_FIELD(x, 56, 56)
+#define MCA_REG__STATUS__TCC(x)			MCA_REG_FIELD(x, 55, 55)
+#define MCA_REG__STATUS__SYNDV(x)		MCA_REG_FIELD(x, 53, 53)
+#define MCA_REG__STATUS__CECC(x)		MCA_REG_FIELD(x, 46, 46)
+#define MCA_REG__STATUS__UECC(x)		MCA_REG_FIELD(x, 45, 45)
+#define MCA_REG__STATUS__DEFERRED(x)		MCA_REG_FIELD(x, 44, 44)
+#define MCA_REG__STATUS__POISON(x)		MCA_REG_FIELD(x, 43, 43)
+#define MCA_REG__STATUS__SCRUB(x)		MCA_REG_FIELD(x, 40, 40)
+#define MCA_REG__STATUS__ERRCOREID(x)		MCA_REG_FIELD(x, 37, 32)
+#define MCA_REG__STATUS__ADDRLSB(x)		MCA_REG_FIELD(x, 29, 24)
+#define MCA_REG__STATUS__ERRORCODEEXT(x)	MCA_REG_FIELD(x, 21, 16)
+#define MCA_REG__STATUS__ERRORCODE(x)		MCA_REG_FIELD(x, 15, 0)
+
 enum amdgpu_mca_ip {
 	AMDGPU_MCA_IP_UNKNOW = -1,
 	AMDGPU_MCA_IP_PSP = 0,
@@ -33,6 +54,7 @@ enum amdgpu_mca_ip {
 	AMDGPU_MCA_IP_SMU,
 	AMDGPU_MCA_IP_MP5,
 	AMDGPU_MCA_IP_UMC,
+	AMDGPU_MCA_IP_PCS_XGMI,
 	AMDGPU_MCA_IP_COUNT,
 };
 
@@ -57,6 +79,15 @@ struct amdgpu_mca {
 	const struct amdgpu_mca_smu_funcs *mca_funcs;
 };
 
+enum mca_reg_idx {
+	MCA_REG_IDX_STATUS		= 1,
+	MCA_REG_IDX_ADDR		= 2,
+	MCA_REG_IDX_MISC0		= 3,
+	MCA_REG_IDX_IPID		= 5,
+	MCA_REG_IDX_SYND		= 6,
+	MCA_REG_IDX_COUNT		= 16,
+};
+
 struct mca_bank_info {
 	int socket_id;
 	int aid;
@@ -72,18 +103,28 @@ struct mca_bank_entry {
 	uint64_t regs[MCA_MAX_REGS_COUNT];
 };
 
+struct mca_bank_node {
+	struct mca_bank_entry entry;
+	struct list_head node;
+};
+
+struct mca_bank_set {
+	int nr_entries;
+	struct list_head list;
+};
+
 struct amdgpu_mca_smu_funcs {
 	int max_ue_count;
 	int max_ce_count;
 	int (*mca_set_debug_mode)(struct amdgpu_device *adev, bool enable);
-	int (*mca_get_error_count)(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
-				   enum amdgpu_mca_error_type type, uint32_t *count);
+	int (*mca_get_ras_mca_set)(struct amdgpu_device *adev, enum amdgpu_ras_block blk, enum amdgpu_mca_error_type type,
+				   struct mca_bank_set *mca_set);
+	int (*mca_parse_mca_error_count)(struct amdgpu_device *adev, enum amdgpu_ras_block blk, enum amdgpu_mca_error_type type,
+					 struct mca_bank_entry *entry, uint32_t *count);
 	int (*mca_get_valid_mca_count)(struct amdgpu_device *adev, enum amdgpu_mca_error_type type,
 				       uint32_t *count);
 	int (*mca_get_mca_entry)(struct amdgpu_device *adev, enum amdgpu_mca_error_type type,
 				 int idx, struct mca_bank_entry *entry);
-	int (*mca_get_ras_mca_idx_array)(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
-					 enum amdgpu_mca_error_type type, int *idx_array, int *idx_array_size);
 };
 
 void amdgpu_mca_query_correctable_error_count(struct amdgpu_device *adev,
@@ -107,11 +148,22 @@ int amdgpu_mca_mpio_ras_sw_init(struct amdgpu_device *adev);
 void amdgpu_mca_smu_init_funcs(struct amdgpu_device *adev, const struct amdgpu_mca_smu_funcs *mca_funcs);
 int amdgpu_mca_smu_set_debug_mode(struct amdgpu_device *adev, bool enable);
 int amdgpu_mca_smu_get_valid_mca_count(struct amdgpu_device *adev, enum amdgpu_mca_error_type type, uint32_t *count);
+int amdgpu_mca_smu_get_mca_set_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
+					   enum amdgpu_mca_error_type type, uint32_t *total);
 int amdgpu_mca_smu_get_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
 				   enum amdgpu_mca_error_type type, uint32_t *count);
+int amdgpu_mca_smu_parse_mca_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
+					 enum amdgpu_mca_error_type type, struct mca_bank_entry *entry, uint32_t *count);
+int amdgpu_mca_smu_get_mca_set(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
+			       enum amdgpu_mca_error_type type, struct mca_bank_set *mca_set);
 int amdgpu_mca_smu_get_mca_entry(struct amdgpu_device *adev, enum amdgpu_mca_error_type type,
 				 int idx, struct mca_bank_entry *entry);
 
 void amdgpu_mca_smu_debugfs_init(struct amdgpu_device *adev, struct dentry *root);
 
+void amdgpu_mca_bank_set_init(struct mca_bank_set *mca_set);
+int amdgpu_mca_bank_set_add_entry(struct mca_bank_set *mca_set, struct mca_bank_entry *entry);
+void amdgpu_mca_bank_set_release(struct mca_bank_set *mca_set);
+int amdgpu_mca_smu_log_ras_error(struct amdgpu_device *adev, enum amdgpu_ras_block blk, enum amdgpu_mca_error_type type, struct ras_err_data *err_data);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 0dcb6c36b02c..cef920a93924 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -1062,9 +1062,6 @@ static const char * const amdgpu_vram_names[] = {
  */
 int amdgpu_bo_init(struct amdgpu_device *adev)
 {
-	/* set the default AGP aperture state */
-	amdgpu_gmc_set_agp_default(adev, &adev->gmc);
-
 	/* On A+A platform, VRAM can be mapped as WB */
 	if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) {
 		/* reserve PAT memory space to WC for VRAM */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index b7fe5951b166..84e5987b14e0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1165,13 +1165,53 @@ static void amdgpu_rasmgr_error_data_statistic_update(struct ras_manager *obj, s
 	}
 }
 
-/* query/inject/cure begin */
-int amdgpu_ras_query_error_status(struct amdgpu_device *adev,
-				  struct ras_query_if *info)
+static int amdgpu_ras_query_error_status_helper(struct amdgpu_device *adev,
+						struct ras_query_if *info,
+						struct ras_err_data *err_data,
+						unsigned int error_query_mode)
 {
+	enum amdgpu_ras_block blk = info ? info->head.block : AMDGPU_RAS_BLOCK_COUNT;
 	struct amdgpu_ras_block_object *block_obj = NULL;
+
+	if (error_query_mode == AMDGPU_RAS_INVALID_ERROR_QUERY)
+		return -EINVAL;
+
+	if (error_query_mode == AMDGPU_RAS_DIRECT_ERROR_QUERY) {
+		if (info->head.block == AMDGPU_RAS_BLOCK__UMC) {
+			amdgpu_ras_get_ecc_info(adev, err_data);
+		} else {
+			block_obj = amdgpu_ras_get_ras_block(adev, info->head.block, 0);
+			if (!block_obj || !block_obj->hw_ops) {
+				dev_dbg_once(adev->dev, "%s doesn't config RAS function\n",
+					     get_ras_block_str(&info->head));
+				return -EINVAL;
+			}
+
+			if (block_obj->hw_ops->query_ras_error_count)
+				block_obj->hw_ops->query_ras_error_count(adev, &err_data);
+
+			if ((info->head.block == AMDGPU_RAS_BLOCK__SDMA) ||
+			    (info->head.block == AMDGPU_RAS_BLOCK__GFX) ||
+			    (info->head.block == AMDGPU_RAS_BLOCK__MMHUB)) {
+				if (block_obj->hw_ops->query_ras_error_status)
+					block_obj->hw_ops->query_ras_error_status(adev);
+			}
+		}
+	} else {
+		/* FIXME: add code to check return value later */
+		amdgpu_mca_smu_log_ras_error(adev, blk, AMDGPU_MCA_ERROR_TYPE_UE, err_data);
+		amdgpu_mca_smu_log_ras_error(adev, blk, AMDGPU_MCA_ERROR_TYPE_CE, err_data);
+	}
+
+	return 0;
+}
+
+/* query/inject/cure begin */
+int amdgpu_ras_query_error_status(struct amdgpu_device *adev, struct ras_query_if *info)
+{
 	struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head);
 	struct ras_err_data err_data;
+	unsigned int error_query_mode;
 	int ret;
 
 	if (!obj)
@@ -1181,27 +1221,14 @@ int amdgpu_ras_query_error_status(struct amdgpu_device *adev,
 	if (ret)
 		return ret;
 
-	if (info->head.block == AMDGPU_RAS_BLOCK__UMC) {
-		amdgpu_ras_get_ecc_info(adev, &err_data);
-	} else {
-		block_obj = amdgpu_ras_get_ras_block(adev, info->head.block, 0);
-		if (!block_obj || !block_obj->hw_ops)   {
-			dev_dbg_once(adev->dev, "%s doesn't config RAS function\n",
-				     get_ras_block_str(&info->head));
-			ret = -EINVAL;
-			goto out_fini_err_data;
-		}
-
-		if (block_obj->hw_ops->query_ras_error_count)
-			block_obj->hw_ops->query_ras_error_count(adev, &err_data);
+	if (!amdgpu_ras_get_error_query_mode(adev, &error_query_mode))
+		return -EINVAL;
 
-		if ((info->head.block == AMDGPU_RAS_BLOCK__SDMA) ||
-		    (info->head.block == AMDGPU_RAS_BLOCK__GFX) ||
-		    (info->head.block == AMDGPU_RAS_BLOCK__MMHUB)) {
-				if (block_obj->hw_ops->query_ras_error_status)
-					block_obj->hw_ops->query_ras_error_status(adev);
-			}
-	}
+	ret = amdgpu_ras_query_error_status_helper(adev, info,
+						   &err_data,
+						   error_query_mode);
+	if (ret)
+		goto out_fini_err_data;
 
 	amdgpu_rasmgr_error_data_statistic_update(obj, &err_data);
 
@@ -1537,7 +1564,8 @@ static void amdgpu_ras_sysfs_remove_bad_page_node(struct amdgpu_device *adev)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 
-	sysfs_remove_file_from_group(&adev->dev->kobj,
+	if (adev->dev->kobj.sd)
+		sysfs_remove_file_from_group(&adev->dev->kobj,
 				&con->badpages_attr.attr,
 				RAS_FS_NAME);
 }
@@ -1556,7 +1584,8 @@ static int amdgpu_ras_sysfs_remove_dev_attr_node(struct amdgpu_device *adev)
 		.attrs = attrs,
 	};
 
-	sysfs_remove_group(&adev->dev->kobj, &group);
+	if (adev->dev->kobj.sd)
+		sysfs_remove_group(&adev->dev->kobj, &group);
 
 	return 0;
 }
@@ -1603,7 +1632,8 @@ int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev,
 	if (!obj || !obj->attr_inuse)
 		return -EINVAL;
 
-	sysfs_remove_file_from_group(&adev->dev->kobj,
+	if (adev->dev->kobj.sd)
+		sysfs_remove_file_from_group(&adev->dev->kobj,
 				&obj->sysfs_attr.attr,
 				RAS_FS_NAME);
 	obj->attr_inuse = 0;
@@ -3397,6 +3427,26 @@ bool amdgpu_ras_get_mca_debug_mode(struct amdgpu_device *adev)
 		return true;
 }
 
+bool amdgpu_ras_get_error_query_mode(struct amdgpu_device *adev,
+				     unsigned int *error_query_mode)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs;
+
+	if (!con) {
+		*error_query_mode = AMDGPU_RAS_INVALID_ERROR_QUERY;
+		return false;
+	}
+
+	if (mca_funcs && mca_funcs->mca_set_debug_mode)
+		*error_query_mode =
+			(con->is_mca_debug_mode) ? AMDGPU_RAS_DIRECT_ERROR_QUERY : AMDGPU_RAS_FIRMWARE_ERROR_QUERY;
+	else
+		*error_query_mode = AMDGPU_RAS_DIRECT_ERROR_QUERY;
+
+	return true;
+}
+
 /* Register each ip ras block into amdgpu ras */
 int amdgpu_ras_register_ras_block(struct amdgpu_device *adev,
 		struct amdgpu_ras_block_object *ras_block_obj)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 665414c22ca9..19161916ac46 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -320,6 +320,12 @@ enum amdgpu_ras_ret {
 	AMDGPU_RAS_PT,
 };
 
+enum amdgpu_ras_error_query_mode {
+	AMDGPU_RAS_INVALID_ERROR_QUERY		= 0,
+	AMDGPU_RAS_DIRECT_ERROR_QUERY		= 1,
+	AMDGPU_RAS_FIRMWARE_ERROR_QUERY		= 2,
+};
+
 /* ras error status reisger fields */
 #define ERR_STATUS_LO__ERR_STATUS_VALID_FLAG__SHIFT	0x0
 #define ERR_STATUS_LO__ERR_STATUS_VALID_FLAG_MASK	0x00000001L
@@ -769,6 +775,8 @@ int amdgpu_ras_set_context(struct amdgpu_device *adev, struct amdgpu_ras *ras_co
 
 void amdgpu_ras_set_mca_debug_mode(struct amdgpu_device *adev, bool enable);
 bool amdgpu_ras_get_mca_debug_mode(struct amdgpu_device *adev);
+bool amdgpu_ras_get_error_query_mode(struct amdgpu_device *adev,
+				     unsigned int *mode);
 
 int amdgpu_ras_register_ras_block(struct amdgpu_device *adev,
 				struct amdgpu_ras_block_object *ras_block_obj);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index 815b7c34ed33..65949cc7abb9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -399,20 +399,20 @@ int amdgpu_uvd_sw_fini(struct amdgpu_device *adev)
  *
  * @adev: amdgpu_device pointer
  *
+ * Initialize the entity used for handle management in the kernel driver.
  */
-int amdgpu_uvd_entity_init(struct amdgpu_device *adev)
+int amdgpu_uvd_entity_init(struct amdgpu_device *adev, struct amdgpu_ring *ring)
 {
-	struct amdgpu_ring *ring;
-	struct drm_gpu_scheduler *sched;
-	int r;
+	if (ring == &adev->uvd.inst[0].ring) {
+		struct drm_gpu_scheduler *sched = &ring->sched;
+		int r;
 
-	ring = &adev->uvd.inst[0].ring;
-	sched = &ring->sched;
-	r = drm_sched_entity_init(&adev->uvd.entity, DRM_SCHED_PRIORITY_NORMAL,
-				  &sched, 1, NULL);
-	if (r) {
-		DRM_ERROR("Failed setting up UVD kernel entity.\n");
-		return r;
+		r = drm_sched_entity_init(&adev->uvd.entity, DRM_SCHED_PRIORITY_NORMAL,
+					  &sched, 1, NULL);
+		if (r) {
+			DRM_ERROR("Failed setting up UVD kernel entity.\n");
+			return r;
+		}
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h
index a9f342537c68..9dfad2f48ef4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h
@@ -73,7 +73,7 @@ struct amdgpu_uvd {
 
 int amdgpu_uvd_sw_init(struct amdgpu_device *adev);
 int amdgpu_uvd_sw_fini(struct amdgpu_device *adev);
-int amdgpu_uvd_entity_init(struct amdgpu_device *adev);
+int amdgpu_uvd_entity_init(struct amdgpu_device *adev, struct amdgpu_ring *ring);
 int amdgpu_uvd_prepare_suspend(struct amdgpu_device *adev);
 int amdgpu_uvd_suspend(struct amdgpu_device *adev);
 int amdgpu_uvd_resume(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index 1904edf68407..0954447f689d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -231,20 +231,20 @@ int amdgpu_vce_sw_fini(struct amdgpu_device *adev)
  *
  * @adev: amdgpu_device pointer
  *
+ * Initialize the entity used for handle management in the kernel driver.
  */
-int amdgpu_vce_entity_init(struct amdgpu_device *adev)
+int amdgpu_vce_entity_init(struct amdgpu_device *adev, struct amdgpu_ring *ring)
 {
-	struct amdgpu_ring *ring;
-	struct drm_gpu_scheduler *sched;
-	int r;
-
-	ring = &adev->vce.ring[0];
-	sched = &ring->sched;
-	r = drm_sched_entity_init(&adev->vce.entity, DRM_SCHED_PRIORITY_NORMAL,
-				  &sched, 1, NULL);
-	if (r != 0) {
-		DRM_ERROR("Failed setting up VCE run queue.\n");
-		return r;
+	if (ring == &adev->vce.ring[0]) {
+		struct drm_gpu_scheduler *sched = &ring->sched;
+		int r;
+
+		r = drm_sched_entity_init(&adev->vce.entity, DRM_SCHED_PRIORITY_NORMAL,
+					  &sched, 1, NULL);
+		if (r != 0) {
+			DRM_ERROR("Failed setting up VCE run queue.\n");
+			return r;
+		}
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
index ea680fc9a6c3..6e53f872d084 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
@@ -55,7 +55,7 @@ struct amdgpu_vce {
 
 int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size);
 int amdgpu_vce_sw_fini(struct amdgpu_device *adev);
-int amdgpu_vce_entity_init(struct amdgpu_device *adev);
+int amdgpu_vce_entity_init(struct amdgpu_device *adev, struct amdgpu_ring *ring);
 int amdgpu_vce_suspend(struct amdgpu_device *adev);
 int amdgpu_vce_resume(struct amdgpu_device *adev);
 void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index a0aa624f5a92..3a632c3b1a2c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -73,9 +73,10 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev)
 
 void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev,
 					uint32_t reg0, uint32_t reg1,
-					uint32_t ref, uint32_t mask)
+					uint32_t ref, uint32_t mask,
+					uint32_t xcc_inst)
 {
-	struct amdgpu_kiq *kiq = &adev->gfx.kiq[0];
+	struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_inst];
 	struct amdgpu_ring *ring = &kiq->ring;
 	signed long r, cnt = 0;
 	unsigned long flags;
@@ -942,7 +943,7 @@ void amdgpu_virt_update_sriov_video_codec(struct amdgpu_device *adev,
 	}
 }
 
-static bool amdgpu_virt_get_rlcg_reg_access_flag(struct amdgpu_device *adev,
+bool amdgpu_virt_get_rlcg_reg_access_flag(struct amdgpu_device *adev,
 						 u32 acc_flags, u32 hwip,
 						 bool write, u32 *rlcg_flag)
 {
@@ -975,7 +976,7 @@ static bool amdgpu_virt_get_rlcg_reg_access_flag(struct amdgpu_device *adev,
 	return ret;
 }
 
-static u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 flag, u32 xcc_id)
+u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 flag, u32 xcc_id)
 {
 	struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl;
 	uint32_t timeout = 50000;
@@ -1093,3 +1094,13 @@ u32 amdgpu_sriov_rreg(struct amdgpu_device *adev,
 	else
 		return RREG32(offset);
 }
+
+bool amdgpu_sriov_xnack_support(struct amdgpu_device *adev)
+{
+	bool xnack_mode = true;
+
+	if (amdgpu_sriov_vf(adev) && adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2))
+		xnack_mode = false;
+
+	return xnack_mode;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index 858ef21ae515..d4207e44141f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -334,7 +334,8 @@ bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev);
 void amdgpu_virt_init_setting(struct amdgpu_device *adev);
 void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev,
 					uint32_t reg0, uint32_t rreg1,
-					uint32_t ref, uint32_t mask);
+					uint32_t ref, uint32_t mask,
+					uint32_t xcc_inst);
 int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init);
 int amdgpu_virt_release_full_gpu(struct amdgpu_device *adev, bool init);
 int amdgpu_virt_reset_gpu(struct amdgpu_device *adev);
@@ -365,4 +366,9 @@ u32 amdgpu_sriov_rreg(struct amdgpu_device *adev,
 bool amdgpu_virt_fw_load_skip_check(struct amdgpu_device *adev,
 			uint32_t ucode_id);
 void amdgpu_virt_post_reset(struct amdgpu_device *adev);
+bool amdgpu_sriov_xnack_support(struct amdgpu_device *adev);
+bool amdgpu_virt_get_rlcg_reg_access_flag(struct amdgpu_device *adev,
+					  u32 acc_flags, u32 hwip,
+					  bool write, u32 *rlcg_flag);
+u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 flag, u32 xcc_id);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 904252456d25..d1b8afd105c9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1098,8 +1098,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 				bo = gem_to_amdgpu_bo(gobj);
 		}
 		mem = bo->tbo.resource;
-		if (mem->mem_type == TTM_PL_TT ||
-		    mem->mem_type == AMDGPU_PL_PREEMPT)
+		if (mem && (mem->mem_type == TTM_PL_TT ||
+			    mem->mem_type == AMDGPU_PL_PREEMPT))
 			pages_addr = bo->tbo.ttm->dma_address;
 	}
 
@@ -2139,7 +2139,8 @@ long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout)
  * Returns:
  * 0 for success, error for failure.
  */
-int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, int32_t xcp_id)
+int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+		   int32_t xcp_id)
 {
 	struct amdgpu_bo *root_bo;
 	struct amdgpu_bo_vm *root;
@@ -2158,6 +2159,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, int32_t xcp
 	INIT_LIST_HEAD(&vm->done);
 	INIT_LIST_HEAD(&vm->pt_freed);
 	INIT_WORK(&vm->pt_free_work, amdgpu_vm_pt_free_work);
+	INIT_KFIFO(vm->faults);
 
 	r = amdgpu_vm_init_entities(adev, vm);
 	if (r)
@@ -2192,34 +2194,33 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, int32_t xcp
 				false, &root, xcp_id);
 	if (r)
 		goto error_free_delayed;
-	root_bo = &root->bo;
+
+	root_bo = amdgpu_bo_ref(&root->bo);
 	r = amdgpu_bo_reserve(root_bo, true);
-	if (r)
-		goto error_free_root;
+	if (r) {
+		amdgpu_bo_unref(&root->shadow);
+		amdgpu_bo_unref(&root_bo);
+		goto error_free_delayed;
+	}
 
+	amdgpu_vm_bo_base_init(&vm->root, vm, root_bo);
 	r = dma_resv_reserve_fences(root_bo->tbo.base.resv, 1);
 	if (r)
-		goto error_unreserve;
-
-	amdgpu_vm_bo_base_init(&vm->root, vm, root_bo);
+		goto error_free_root;
 
 	r = amdgpu_vm_pt_clear(adev, vm, root, false);
 	if (r)
-		goto error_unreserve;
+		goto error_free_root;
 
 	amdgpu_bo_unreserve(vm->root.bo);
-
-	INIT_KFIFO(vm->faults);
+	amdgpu_bo_unref(&root_bo);
 
 	return 0;
 
-error_unreserve:
-	amdgpu_bo_unreserve(vm->root.bo);
-
 error_free_root:
-	amdgpu_bo_unref(&root->shadow);
+	amdgpu_vm_pt_free_root(adev, vm);
+	amdgpu_bo_unreserve(vm->root.bo);
 	amdgpu_bo_unref(&root_bo);
-	vm->root.bo = NULL;
 
 error_free_delayed:
 	dma_fence_put(vm->last_tlb_flush);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 9d5d742ee9d3..bd20cb3b9819 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -103,6 +103,53 @@ static const int walf_pcs_err_noncorrectable_mask_reg_aldebaran[] = {
 	smnPCS_GOPX1_PCS_ERROR_NONCORRECTABLE_MASK + 0x100000
 };
 
+static const int xgmi3x16_pcs_err_status_reg_v6_4[] = {
+	smnPCS_XGMI3X16_PCS_ERROR_STATUS,
+	smnPCS_XGMI3X16_PCS_ERROR_STATUS + 0x100000
+};
+
+static const int xgmi3x16_pcs_err_noncorrectable_mask_reg_v6_4[] = {
+	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK,
+	smnPCS_XGMI3X16_PCS_ERROR_NONCORRECTABLE_MASK + 0x100000
+};
+
+static const u64 xgmi_v6_4_0_mca_base_array[] = {
+	0x11a09200,
+	0x11b09200,
+};
+
+static const char *xgmi_v6_4_0_ras_error_code_ext[32] = {
+	[0x00] = "XGMI PCS DataLossErr",
+	[0x01] = "XGMI PCS TrainingErr",
+	[0x02] = "XGMI PCS FlowCtrlAckErr",
+	[0x03] = "XGMI PCS RxFifoUnderflowErr",
+	[0x04] = "XGMI PCS RxFifoOverflowErr",
+	[0x05] = "XGMI PCS CRCErr",
+	[0x06] = "XGMI PCS BERExceededErr",
+	[0x07] = "XGMI PCS TxMetaDataErr",
+	[0x08] = "XGMI PCS ReplayBufParityErr",
+	[0x09] = "XGMI PCS DataParityErr",
+	[0x0a] = "XGMI PCS ReplayFifoOverflowErr",
+	[0x0b] = "XGMI PCS ReplayFifoUnderflowErr",
+	[0x0c] = "XGMI PCS ElasticFifoOverflowErr",
+	[0x0d] = "XGMI PCS DeskewErr",
+	[0x0e] = "XGMI PCS FlowCtrlCRCErr",
+	[0x0f] = "XGMI PCS DataStartupLimitErr",
+	[0x10] = "XGMI PCS FCInitTimeoutErr",
+	[0x11] = "XGMI PCS RecoveryTimeoutErr",
+	[0x12] = "XGMI PCS ReadySerialTimeoutErr",
+	[0x13] = "XGMI PCS ReadySerialAttemptErr",
+	[0x14] = "XGMI PCS RecoveryAttemptErr",
+	[0x15] = "XGMI PCS RecoveryRelockAttemptErr",
+	[0x16] = "XGMI PCS ReplayAttemptErr",
+	[0x17] = "XGMI PCS SyncHdrErr",
+	[0x18] = "XGMI PCS TxReplayTimeoutErr",
+	[0x19] = "XGMI PCS RxReplayTimeoutErr",
+	[0x1a] = "XGMI PCS LinkSubTxTimeoutErr",
+	[0x1b] = "XGMI PCS LinkSubRxTimeoutErr",
+	[0x1c] = "XGMI PCS RxCMDPktErr",
+};
+
 static const struct amdgpu_pcs_ras_field xgmi_pcs_ras_fields[] = {
 	{"XGMI PCS DataLossErr",
 	 SOC15_REG_FIELD(XGMI0_PCS_GOPX16_PCS_ERROR_STATUS, DataLossErr)},
@@ -926,7 +973,7 @@ static void pcs_clear_status(struct amdgpu_device *adev, uint32_t pcs_status_reg
 	WREG32_PCIE(pcs_status_reg, 0);
 }
 
-static void amdgpu_xgmi_reset_ras_error_count(struct amdgpu_device *adev)
+static void amdgpu_xgmi_legacy_reset_ras_error_count(struct amdgpu_device *adev)
 {
 	uint32_t i;
 
@@ -952,6 +999,49 @@ static void amdgpu_xgmi_reset_ras_error_count(struct amdgpu_device *adev)
 	default:
 		break;
 	}
+
+	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
+	case IP_VERSION(6, 4, 0):
+		for (i = 0; i < ARRAY_SIZE(xgmi3x16_pcs_err_status_reg_v6_4); i++)
+			pcs_clear_status(adev,
+					xgmi3x16_pcs_err_status_reg_v6_4[i]);
+		break;
+	default:
+		break;
+	}
+}
+
+static void __xgmi_v6_4_0_reset_error_count(struct amdgpu_device *adev, int xgmi_inst, u64 mca_base)
+{
+	WREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS, 0ULL);
+}
+
+static void xgmi_v6_4_0_reset_error_count(struct amdgpu_device *adev, int xgmi_inst)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(xgmi_v6_4_0_mca_base_array); i++)
+		__xgmi_v6_4_0_reset_error_count(adev, xgmi_inst, xgmi_v6_4_0_mca_base_array[i]);
+}
+
+static void xgmi_v6_4_0_reset_ras_error_count(struct amdgpu_device *adev)
+{
+	int i;
+
+	for_each_inst(i, adev->aid_mask)
+		xgmi_v6_4_0_reset_error_count(adev, i);
+}
+
+static void amdgpu_xgmi_reset_ras_error_count(struct amdgpu_device *adev)
+{
+	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
+	case IP_VERSION(6, 4, 0):
+		xgmi_v6_4_0_reset_ras_error_count(adev);
+		break;
+	default:
+		amdgpu_xgmi_legacy_reset_ras_error_count(adev);
+		break;
+	}
 }
 
 static int amdgpu_xgmi_query_pcs_error_status(struct amdgpu_device *adev,
@@ -969,7 +1059,9 @@ static int amdgpu_xgmi_query_pcs_error_status(struct amdgpu_device *adev,
 
 	if (is_xgmi_pcs) {
 		if (amdgpu_ip_version(adev, XGMI_HWIP, 0) ==
-		    IP_VERSION(6, 1, 0)) {
+		    IP_VERSION(6, 1, 0) ||
+		    amdgpu_ip_version(adev, XGMI_HWIP, 0) ==
+		    IP_VERSION(6, 4, 0)) {
 			pcs_ras_fields = &xgmi3x16_pcs_ras_fields[0];
 			field_array_size = ARRAY_SIZE(xgmi3x16_pcs_ras_fields);
 		} else {
@@ -1003,11 +1095,11 @@ static int amdgpu_xgmi_query_pcs_error_status(struct amdgpu_device *adev,
 	return 0;
 }
 
-static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev,
-					     void *ras_error_status)
+static void amdgpu_xgmi_legacy_query_ras_error_count(struct amdgpu_device *adev,
+						     void *ras_error_status)
 {
 	struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
-	int i;
+	int i, supported = 1;
 	uint32_t data, mask_data = 0;
 	uint32_t ue_cnt = 0, ce_cnt = 0;
 
@@ -1071,7 +1163,25 @@ static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev,
 		}
 		break;
 	default:
-		dev_warn(adev->dev, "XGMI RAS error query not supported");
+		supported = 0;
+		break;
+	}
+
+	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
+	case IP_VERSION(6, 4, 0):
+		/* check xgmi3x16 pcs error */
+		for (i = 0; i < ARRAY_SIZE(xgmi3x16_pcs_err_status_reg_v6_4); i++) {
+			data = RREG32_PCIE(xgmi3x16_pcs_err_status_reg_v6_4[i]);
+			mask_data =
+				RREG32_PCIE(xgmi3x16_pcs_err_noncorrectable_mask_reg_v6_4[i]);
+			if (data)
+				amdgpu_xgmi_query_pcs_error_status(adev, data,
+						mask_data, &ue_cnt, &ce_cnt, true, true);
+		}
+		break;
+	default:
+		if (!supported)
+			dev_warn(adev->dev, "XGMI RAS error query not supported");
 		break;
 	}
 
@@ -1081,32 +1191,116 @@ static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev,
 	err_data->ce_count += ce_cnt;
 }
 
+static enum amdgpu_mca_error_type xgmi_v6_4_0_pcs_mca_get_error_type(struct amdgpu_device *adev, u64 status)
+{
+	const char *error_str;
+	int ext_error_code;
+
+	ext_error_code = MCA_REG__STATUS__ERRORCODEEXT(status);
+
+	error_str = ext_error_code < ARRAY_SIZE(xgmi_v6_4_0_ras_error_code_ext) ?
+		xgmi_v6_4_0_ras_error_code_ext[ext_error_code] : NULL;
+	if (error_str)
+		dev_info(adev->dev, "%s detected\n", error_str);
+
+	switch (ext_error_code) {
+	case 0:
+		return AMDGPU_MCA_ERROR_TYPE_UE;
+	case 6:
+		return AMDGPU_MCA_ERROR_TYPE_CE;
+	default:
+		return -EINVAL;
+	}
+
+	return -EINVAL;
+}
+
+static void __xgmi_v6_4_0_query_error_count(struct amdgpu_device *adev, struct amdgpu_smuio_mcm_config_info *mcm_info,
+					    u64 mca_base, struct ras_err_data *err_data)
+{
+	int xgmi_inst = mcm_info->die_id;
+	u64 status = 0;
+
+	status = RREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS);
+	if (!MCA_REG__STATUS__VAL(status))
+		return;
+
+	switch (xgmi_v6_4_0_pcs_mca_get_error_type(adev, status)) {
+	case AMDGPU_MCA_ERROR_TYPE_UE:
+		amdgpu_ras_error_statistic_ue_count(err_data, mcm_info, 1ULL);
+		break;
+	case AMDGPU_MCA_ERROR_TYPE_CE:
+		amdgpu_ras_error_statistic_ce_count(err_data, mcm_info, 1ULL);
+		break;
+	default:
+		break;
+	}
+
+	WREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS, 0ULL);
+}
+
+static void xgmi_v6_4_0_query_error_count(struct amdgpu_device *adev, int xgmi_inst, struct ras_err_data *err_data)
+{
+	struct amdgpu_smuio_mcm_config_info mcm_info = {
+		.socket_id = adev->smuio.funcs->get_socket_id(adev),
+		.die_id = xgmi_inst,
+	};
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(xgmi_v6_4_0_mca_base_array); i++)
+		__xgmi_v6_4_0_query_error_count(adev, &mcm_info, xgmi_v6_4_0_mca_base_array[i], err_data);
+}
+
+static void xgmi_v6_4_0_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status)
+{
+	struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
+	int i;
+
+	for_each_inst(i, adev->aid_mask)
+		xgmi_v6_4_0_query_error_count(adev, i, err_data);
+}
+
+static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev,
+					      void *ras_error_status)
+{
+	switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) {
+	case IP_VERSION(6, 4, 0):
+		xgmi_v6_4_0_query_ras_error_count(adev, ras_error_status);
+		break;
+	default:
+		amdgpu_xgmi_legacy_query_ras_error_count(adev, ras_error_status);
+		break;
+	}
+}
+
 /* Trigger XGMI/WAFL error */
 static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev,
 			void *inject_if, uint32_t instance_mask)
 {
-	int ret = 0;
+	int ret1, ret2;
 	struct ta_ras_trigger_error_input *block_info =
 				(struct ta_ras_trigger_error_input *)inject_if;
 
 	if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW))
 		dev_warn(adev->dev, "Failed to disallow df cstate");
 
-	if (amdgpu_dpm_set_xgmi_plpd_mode(adev, XGMI_PLPD_DISALLOW))
+	ret1 = amdgpu_dpm_set_xgmi_plpd_mode(adev, XGMI_PLPD_DISALLOW);
+	if (ret1 && ret1 != -EOPNOTSUPP)
 		dev_warn(adev->dev, "Failed to disallow XGMI power down");
 
-	ret = psp_ras_trigger_error(&adev->psp, block_info, instance_mask);
+	ret2 = psp_ras_trigger_error(&adev->psp, block_info, instance_mask);
 
 	if (amdgpu_ras_intr_triggered())
-		return ret;
+		return ret2;
 
-	if (amdgpu_dpm_set_xgmi_plpd_mode(adev, XGMI_PLPD_DEFAULT))
+	ret1 = amdgpu_dpm_set_xgmi_plpd_mode(adev, XGMI_PLPD_DEFAULT);
+	if (ret1 && ret1 != -EOPNOTSUPP)
 		dev_warn(adev->dev, "Failed to allow XGMI power down");
 
 	if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW))
 		dev_warn(adev->dev, "Failed to allow df cstate");
 
-	return ret;
+	return ret2;
 }
 
 struct amdgpu_ras_block_hw_ops  xgmi_ras_hw_ops = {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 41bbabd9ad4d..40d06d32bb74 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -1102,6 +1102,7 @@ static void gfx_v9_4_3_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev)
 		reg_access_ctrl->grbm_idx = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regGRBM_GFX_INDEX);
 		reg_access_ctrl->spare_int = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_SPARE_INT);
 	}
+	adev->gfx.rlc.rlcg_reg_access_supported = true;
 }
 
 static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev)
@@ -2738,16 +2739,16 @@ static void gfx_v9_4_3_xcc_set_compute_eop_interrupt_state(
 
 	switch (state) {
 	case AMDGPU_IRQ_STATE_DISABLE:
-		mec_int_cntl = RREG32(mec_int_cntl_reg);
+		mec_int_cntl = RREG32_XCC(mec_int_cntl_reg, xcc_id);
 		mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL,
 					     TIME_STAMP_INT_ENABLE, 0);
-		WREG32(mec_int_cntl_reg, mec_int_cntl);
+		WREG32_XCC(mec_int_cntl_reg, mec_int_cntl, xcc_id);
 		break;
 	case AMDGPU_IRQ_STATE_ENABLE:
-		mec_int_cntl = RREG32(mec_int_cntl_reg);
+		mec_int_cntl = RREG32_XCC(mec_int_cntl_reg, xcc_id);
 		mec_int_cntl = REG_SET_FIELD(mec_int_cntl, CP_ME1_PIPE0_INT_CNTL,
 					     TIME_STAMP_INT_ENABLE, 1);
-		WREG32(mec_int_cntl_reg, mec_int_cntl);
+		WREG32_XCC(mec_int_cntl_reg, mec_int_cntl, xcc_id);
 		break;
 	default:
 		break;
@@ -3799,6 +3800,27 @@ static void gfx_v9_4_3_inst_query_ras_err_count(struct amdgpu_device *adev,
 		}
 	}
 
+	/* handle extra register entries of UE */
+	for (; i < ARRAY_SIZE(gfx_v9_4_3_ue_reg_list); i++) {
+		for (j = 0; j < gfx_v9_4_3_ue_reg_list[i].se_num; j++) {
+			for (k = 0; k < gfx_v9_4_3_ue_reg_list[i].reg_entry.reg_inst; k++) {
+				/* no need to select if instance number is 1 */
+				if (gfx_v9_4_3_ue_reg_list[i].se_num > 1 ||
+					gfx_v9_4_3_ue_reg_list[i].reg_entry.reg_inst > 1)
+					gfx_v9_4_3_xcc_select_se_sh(adev, j, 0, k, xcc_id);
+
+				amdgpu_ras_inst_query_ras_error_count(adev,
+					&(gfx_v9_4_3_ue_reg_list[i].reg_entry),
+					1,
+					gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ue_reg_list[i].mem_id_type].mem_id_ent,
+					gfx_v9_4_3_ras_mem_list_array[gfx_v9_4_3_ue_reg_list[i].mem_id_type].size,
+					GET_INST(GC, xcc_id),
+					AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
+					&ue_count);
+			}
+		}
+	}
+
 	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
 			xcc_id);
 	mutex_unlock(&adev->grbm_idx_mutex);
@@ -3838,6 +3860,23 @@ static void gfx_v9_4_3_inst_reset_ras_err_count(struct amdgpu_device *adev,
 		}
 	}
 
+	/* handle extra register entries of UE */
+	for (; i < ARRAY_SIZE(gfx_v9_4_3_ue_reg_list); i++) {
+		for (j = 0; j < gfx_v9_4_3_ue_reg_list[i].se_num; j++) {
+			for (k = 0; k < gfx_v9_4_3_ue_reg_list[i].reg_entry.reg_inst; k++) {
+				/* no need to select if instance number is 1 */
+				if (gfx_v9_4_3_ue_reg_list[i].se_num > 1 ||
+					gfx_v9_4_3_ue_reg_list[i].reg_entry.reg_inst > 1)
+					gfx_v9_4_3_xcc_select_se_sh(adev, j, 0, k, xcc_id);
+
+				amdgpu_ras_inst_reset_ras_error_count(adev,
+					&(gfx_v9_4_3_ue_reg_list[i].reg_entry),
+					1,
+					GET_INST(GC, xcc_id));
+			}
+		}
+	}
+
 	gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff,
 			xcc_id);
 	mutex_unlock(&adev->grbm_idx_mutex);
@@ -4300,7 +4339,7 @@ const struct amdgpu_ip_block_version gfx_v9_4_3_ip_block = {
 	.type = AMD_IP_BLOCK_TYPE_GFX,
 	.major = 9,
 	.minor = 4,
-	.rev = 0,
+	.rev = 3,
 	.funcs = &gfx_v9_4_3_ip_funcs,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index d8a4fddab9c1..0ec7b061d7c2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -268,7 +268,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	if (adev->gfx.kiq[0].ring.sched.ready && !adev->enable_mes &&
 	    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) {
 		amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, inv_req,
-				1 << vmid);
+				1 << vmid, GET_INST(GC, 0));
 		return;
 	}
 
@@ -672,6 +672,7 @@ static void gmc_v10_0_vram_gtt_location(struct amdgpu_device *adev,
 	/* add the xgmi offset of the physical node */
 	base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
 
+	amdgpu_gmc_set_agp_default(adev, mc);
 	amdgpu_gmc_vram_location(adev, &adev->gmc, base);
 	amdgpu_gmc_gart_location(adev, mc, AMDGPU_GART_PLACEMENT_BEST_FIT);
 	if (!amdgpu_sriov_vf(adev))
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index 4713a62ad586..6dce9b29f675 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -229,7 +229,7 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	if ((adev->gfx.kiq[0].ring.sched.ready || adev->mes.ring.sched.ready) &&
 	    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) {
 		amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, inv_req,
-				1 << vmid);
+				1 << vmid, GET_INST(GC, 0));
 		return;
 	}
 
@@ -637,6 +637,7 @@ static void gmc_v11_0_vram_gtt_location(struct amdgpu_device *adev,
 
 	base = adev->mmhub.funcs->get_fb_location(adev);
 
+	amdgpu_gmc_set_agp_default(adev, mc);
 	amdgpu_gmc_vram_location(adev, &adev->gmc, base);
 	amdgpu_gmc_gart_location(adev, mc, AMDGPU_GART_PLACEMENT_HIGH);
 	if (!amdgpu_sriov_vf(adev) ||
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
index 7f66954fd302..42e103d7077d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
@@ -211,6 +211,7 @@ static void gmc_v6_0_vram_gtt_location(struct amdgpu_device *adev,
 
 	base <<= 24;
 
+	amdgpu_gmc_set_agp_default(adev, mc);
 	amdgpu_gmc_vram_location(adev, mc, base);
 	amdgpu_gmc_gart_location(adev, mc, AMDGPU_GART_PLACEMENT_BEST_FIT);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 61ca1a82b651..efc16e580f1e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -239,6 +239,7 @@ static void gmc_v7_0_vram_gtt_location(struct amdgpu_device *adev,
 
 	base <<= 24;
 
+	amdgpu_gmc_set_agp_default(adev, mc);
 	amdgpu_gmc_vram_location(adev, mc, base);
 	amdgpu_gmc_gart_location(adev, mc, AMDGPU_GART_PLACEMENT_BEST_FIT);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index fa59749c2aef..ff4ae73d27ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -413,6 +413,7 @@ static void gmc_v8_0_vram_gtt_location(struct amdgpu_device *adev,
 		base = RREG32(mmMC_VM_FB_LOCATION) & 0xFFFF;
 	base <<= 24;
 
+	amdgpu_gmc_set_agp_default(adev, mc);
 	amdgpu_gmc_vram_location(adev, mc, base);
 	amdgpu_gmc_gart_location(adev, mc, AMDGPU_GART_PLACEMENT_BEST_FIT);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index b66c5f7e1c56..bde25eb4ed8e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -817,7 +817,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 					uint32_t vmhub, uint32_t flush_type)
 {
 	bool use_semaphore = gmc_v9_0_use_invalidate_semaphore(adev, vmhub);
-	u32 j, inv_req, tmp, sem, req, ack;
+	u32 j, inv_req, tmp, sem, req, ack, inst;
 	const unsigned int eng = 17;
 	struct amdgpu_vmhub *hub;
 
@@ -832,13 +832,17 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	/* This is necessary for a HW workaround under SRIOV as well
 	 * as GFXOFF under bare metal
 	 */
-	if (adev->gfx.kiq[0].ring.sched.ready &&
+	if (vmhub >= AMDGPU_MMHUB0(0))
+		inst = GET_INST(GC, 0);
+	else
+		inst = vmhub;
+	if (adev->gfx.kiq[inst].ring.sched.ready &&
 	    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) {
 		uint32_t req = hub->vm_inv_eng0_req + hub->eng_distance * eng;
 		uint32_t ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng;
 
 		amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, inv_req,
-						   1 << vmid);
+						   1 << vmid, inst);
 		return;
 	}
 
@@ -856,9 +860,9 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 		for (j = 0; j < adev->usec_timeout; j++) {
 			/* a read return value of 1 means semaphore acquire */
 			if (vmhub >= AMDGPU_MMHUB0(0))
-				tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, sem);
+				tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, sem, inst);
 			else
-				tmp = RREG32_SOC15_IP_NO_KIQ(GC, sem);
+				tmp = RREG32_SOC15_IP_NO_KIQ(GC, sem, inst);
 			if (tmp & 0x1)
 				break;
 			udelay(1);
@@ -869,9 +873,9 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	}
 
 	if (vmhub >= AMDGPU_MMHUB0(0))
-		WREG32_SOC15_IP_NO_KIQ(MMHUB, req, inv_req);
+		WREG32_SOC15_IP_NO_KIQ(MMHUB, req, inv_req, inst);
 	else
-		WREG32_SOC15_IP_NO_KIQ(GC, req, inv_req);
+		WREG32_SOC15_IP_NO_KIQ(GC, req, inv_req, inst);
 
 	/*
 	 * Issue a dummy read to wait for the ACK register to
@@ -884,9 +888,9 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 
 	for (j = 0; j < adev->usec_timeout; j++) {
 		if (vmhub >= AMDGPU_MMHUB0(0))
-			tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, ack);
+			tmp = RREG32_SOC15_IP_NO_KIQ(MMHUB, ack, inst);
 		else
-			tmp = RREG32_SOC15_IP_NO_KIQ(GC, ack);
+			tmp = RREG32_SOC15_IP_NO_KIQ(GC, ack, inst);
 		if (tmp & (1 << vmid))
 			break;
 		udelay(1);
@@ -899,9 +903,9 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 		 * write with 0 means semaphore release
 		 */
 		if (vmhub >= AMDGPU_MMHUB0(0))
-			WREG32_SOC15_IP_NO_KIQ(MMHUB, sem, 0);
+			WREG32_SOC15_IP_NO_KIQ(MMHUB, sem, 0, inst);
 		else
-			WREG32_SOC15_IP_NO_KIQ(GC, sem, 0);
+			WREG32_SOC15_IP_NO_KIQ(GC, sem, 0, inst);
 	}
 
 	spin_unlock(&adev->gmc.invalidate_lock);
@@ -1176,7 +1180,10 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 		if (uncached) {
 			mtype = MTYPE_UC;
 		} else if (ext_coherent) {
-			mtype = is_local ? MTYPE_CC : MTYPE_UC;
+			if (adev->rev_id)
+				mtype = is_local ? MTYPE_CC : MTYPE_UC;
+			else
+				mtype = MTYPE_UC;
 		} else if (adev->flags & AMD_IS_APU) {
 			mtype = is_local ? mtype_local : MTYPE_NC;
 		} else {
@@ -1297,7 +1304,7 @@ static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev,
 
 			*flags = (*flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) |
 				 AMDGPU_PTE_MTYPE_VG10(mtype_local);
-		} else {
+		} else if (adev->rev_id) {
 			/* MTYPE_UC case */
 			*flags = (*flags & ~AMDGPU_PTE_MTYPE_VG10_MASK) |
 				 AMDGPU_PTE_MTYPE_VG10(MTYPE_CC);
@@ -1614,6 +1621,8 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev,
 {
 	u64 base = adev->mmhub.funcs->get_fb_location(adev);
 
+	amdgpu_gmc_set_agp_default(adev, mc);
+
 	/* add the xgmi offset of the physical node */
 	base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
 	if (adev->gmc.xgmi.connected_to_cpu) {
diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
index 3f3a6445c006..49e934975719 100644
--- a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c
@@ -145,6 +145,10 @@ static void hdp_v4_0_init_registers(struct amdgpu_device *adev)
 		break;
 	}
 
+	/* Do not program registers if VF */
+	if (amdgpu_sriov_vf(adev))
+		return;
+
 	WREG32_FIELD15(HDP, 0, HDP_MISC_CNTL, FLUSH_INVALIDATE_CACHE, 1);
 
 	if (amdgpu_ip_version(adev, HDP_HWIP, 0) == IP_VERSION(4, 4, 0))
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index 9a8ec4d7e333..82b6b62c170b 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -654,9 +654,11 @@ static void jpeg_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring)
  */
 static void jpeg_v4_0_3_dec_ring_insert_start(struct amdgpu_ring *ring)
 {
-	amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
-		0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, 0x62a04); /* PCTL0_MMHUB_DEEPSLEEP_IB */
+	if (!amdgpu_sriov_vf(ring->adev)) {
+		amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
+			0, 0, PACKETJ_TYPE0));
+		amdgpu_ring_write(ring, 0x62a04); /* PCTL0_MMHUB_DEEPSLEEP_IB */
+	}
 
 	amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
 		0, 0, PACKETJ_TYPE0));
@@ -672,9 +674,11 @@ static void jpeg_v4_0_3_dec_ring_insert_start(struct amdgpu_ring *ring)
  */
 static void jpeg_v4_0_3_dec_ring_insert_end(struct amdgpu_ring *ring)
 {
-	amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
-		0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, 0x62a04);
+	if (!amdgpu_sriov_vf(ring->adev)) {
+		amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET,
+			0, 0, PACKETJ_TYPE0));
+		amdgpu_ring_write(ring, 0x62a04);
+	}
 
 	amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR,
 		0, 0, PACKETJ_TYPE0));
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
index c46bc6aa4f48..0f24af6f2810 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c
@@ -427,6 +427,7 @@ static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev,
 				      uint32_t inst_mask)
 {
 	struct amdgpu_ring *sdma[AMDGPU_MAX_SDMA_INSTANCES];
+	u32 doorbell_offset, doorbell;
 	u32 rb_cntl, ib_cntl;
 	int i, unset = 0;
 
@@ -444,6 +445,18 @@ static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev,
 		ib_cntl = RREG32_SDMA(i, regSDMA_GFX_IB_CNTL);
 		ib_cntl = REG_SET_FIELD(ib_cntl, SDMA_GFX_IB_CNTL, IB_ENABLE, 0);
 		WREG32_SDMA(i, regSDMA_GFX_IB_CNTL, ib_cntl);
+
+		if (sdma[i]->use_doorbell) {
+			doorbell = RREG32_SDMA(i, regSDMA_GFX_DOORBELL);
+			doorbell_offset = RREG32_SDMA(i, regSDMA_GFX_DOORBELL_OFFSET);
+
+			doorbell = REG_SET_FIELD(doorbell, SDMA_GFX_DOORBELL, ENABLE, 0);
+			doorbell_offset = REG_SET_FIELD(doorbell_offset,
+					SDMA_GFX_DOORBELL_OFFSET,
+					OFFSET, 0);
+			WREG32_SDMA(i, regSDMA_GFX_DOORBELL, doorbell);
+			WREG32_SDMA(i, regSDMA_GFX_DOORBELL_OFFSET, doorbell_offset);
+		}
 	}
 }
 
@@ -631,12 +644,6 @@ static void sdma_v4_4_2_gfx_resume(struct amdgpu_device *adev, unsigned int i)
 	rb_cntl = sdma_v4_4_2_rb_cntl(ring, rb_cntl);
 	WREG32_SDMA(i, regSDMA_GFX_RB_CNTL, rb_cntl);
 
-	/* Initialize the ring buffer's read and write pointers */
-	WREG32_SDMA(i, regSDMA_GFX_RB_RPTR, 0);
-	WREG32_SDMA(i, regSDMA_GFX_RB_RPTR_HI, 0);
-	WREG32_SDMA(i, regSDMA_GFX_RB_WPTR, 0);
-	WREG32_SDMA(i, regSDMA_GFX_RB_WPTR_HI, 0);
-
 	/* set the wb address whether it's enabled or not */
 	WREG32_SDMA(i, regSDMA_GFX_RB_RPTR_ADDR_HI,
 	       upper_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFF);
@@ -654,6 +661,12 @@ static void sdma_v4_4_2_gfx_resume(struct amdgpu_device *adev, unsigned int i)
 	/* before programing wptr to a less value, need set minor_ptr_update first */
 	WREG32_SDMA(i, regSDMA_GFX_MINOR_PTR_UPDATE, 1);
 
+	/* Initialize the ring buffer's read and write pointers */
+	WREG32_SDMA(i, regSDMA_GFX_RB_RPTR, 0);
+	WREG32_SDMA(i, regSDMA_GFX_RB_RPTR_HI, 0);
+	WREG32_SDMA(i, regSDMA_GFX_RB_WPTR, 0);
+	WREG32_SDMA(i, regSDMA_GFX_RB_WPTR_HI, 0);
+
 	doorbell = RREG32_SDMA(i, regSDMA_GFX_DOORBELL);
 	doorbell_offset = RREG32_SDMA(i, regSDMA_GFX_DOORBELL_OFFSET);
 
@@ -2048,7 +2061,7 @@ const struct amdgpu_ip_block_version sdma_v4_4_2_ip_block = {
 	.type = AMD_IP_BLOCK_TYPE_SDMA,
 	.major = 4,
 	.minor = 4,
-	.rev = 0,
+	.rev = 2,
 	.funcs = &sdma_v4_4_2_ip_funcs,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
index da683afa0222..242b24f73c17 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h
+++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h
@@ -69,7 +69,7 @@
 
 #define RREG32_SOC15_IP(ip, reg) __RREG32_SOC15_RLC__(reg, 0, ip##_HWIP, 0)
 
-#define RREG32_SOC15_IP_NO_KIQ(ip, reg) __RREG32_SOC15_RLC__(reg, AMDGPU_REGS_NO_KIQ, ip##_HWIP, 0)
+#define RREG32_SOC15_IP_NO_KIQ(ip, reg, inst) __RREG32_SOC15_RLC__(reg, AMDGPU_REGS_NO_KIQ, ip##_HWIP, inst)
 
 #define RREG32_SOC15_NO_KIQ(ip, inst, reg) \
 	__RREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, \
@@ -86,8 +86,8 @@
 #define WREG32_SOC15_IP(ip, reg, value) \
 	 __WREG32_SOC15_RLC__(reg, value, 0, ip##_HWIP, 0)
 
-#define WREG32_SOC15_IP_NO_KIQ(ip, reg, value) \
-	 __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_NO_KIQ, ip##_HWIP, 0)
+#define WREG32_SOC15_IP_NO_KIQ(ip, reg, value, inst) \
+	 __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_NO_KIQ, ip##_HWIP, inst)
 
 #define WREG32_SOC15_NO_KIQ(ip, inst, reg, value) \
 	__WREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, \
@@ -140,7 +140,7 @@
 
 /* for GC only */
 #define RREG32_RLC(reg) \
-	__RREG32_SOC15_RLC__(reg, AMDGPU_REGS_RLC, GC_HWIP)
+	__RREG32_SOC15_RLC__(reg, AMDGPU_REGS_RLC, GC_HWIP, 0)
 
 #define WREG32_RLC_NO_KIQ(reg, value, hwip) \
 	__WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_NO_KIQ | AMDGPU_REGS_RLC, hwip, 0)
@@ -204,4 +204,10 @@
 			+ adev->asic_funcs->encode_ext_smn_addressing(ext), \
 			value) \
 
+#define RREG64_MCA(ext, mca_base, idx) \
+	RREG64_PCIE_EXT(adev->asic_funcs->encode_ext_smn_addressing(ext) + mca_base + (idx * 8))
+
+#define WREG64_MCA(ext, mca_base, idx, val) \
+	WREG64_PCIE_EXT(adev->asic_funcs->encode_ext_smn_addressing(ext) + mca_base + (idx * 8), val)
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c
index d5083c549330..48c6efcdeac9 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc21.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc21.c
@@ -381,6 +381,7 @@ soc21_asic_reset_method(struct amdgpu_device *adev)
 		return AMD_RESET_METHOD_MODE1;
 	case IP_VERSION(13, 0, 4):
 	case IP_VERSION(13, 0, 11):
+	case IP_VERSION(14, 0, 0):
 		return AMD_RESET_METHOD_MODE2;
 	default:
 		if (amdgpu_dpm_is_baco_supported(adev))
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
index 770b4b4e3138..e9c2ff74f0bc 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
@@ -88,7 +88,7 @@ static void umc_v12_0_reset_error_count(struct amdgpu_device *adev)
 		umc_v12_0_reset_error_count_per_channel, NULL);
 }
 
-static bool umc_v12_0_is_uncorrectable_error(uint64_t mc_umc_status)
+bool umc_v12_0_is_uncorrectable_error(uint64_t mc_umc_status)
 {
 	return ((REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) &&
 		(REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC) == 1 ||
@@ -96,7 +96,7 @@ static bool umc_v12_0_is_uncorrectable_error(uint64_t mc_umc_status)
 		REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC) == 1));
 }
 
-static bool umc_v12_0_is_correctable_error(uint64_t mc_umc_status)
+bool umc_v12_0_is_correctable_error(uint64_t mc_umc_status)
 {
 	return (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 &&
 		(REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1 ||
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h
index 4885b9fff272..b34b1e358f8b 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h
@@ -117,6 +117,9 @@
 		(pa) |= (UMC_V12_0_CHANNEL_HASH_CH6(channel_idx, pa) << UMC_V12_0_PA_CH6_BIT); \
 	} while (0)
 
+bool umc_v12_0_is_uncorrectable_error(uint64_t mc_umc_status);
+bool umc_v12_0_is_correctable_error(uint64_t mc_umc_status);
+
 extern const uint32_t
 	umc_v12_0_channel_idx_tbl[]
 			[UMC_V12_0_UMC_INSTANCE_NUM]
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c b/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c
index 58a8f78c003c..a6006f231c65 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c
@@ -577,8 +577,6 @@ static int uvd_v3_1_sw_init(void *handle)
 	ptr += ucode_len;
 	memcpy(&adev->uvd.keyselect, ptr, 4);
 
-	r = amdgpu_uvd_entity_init(adev);
-
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
index d3b1e31f5450..1aa09ad7bbe3 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
@@ -127,8 +127,6 @@ static int uvd_v4_2_sw_init(void *handle)
 	if (r)
 		return r;
 
-	r = amdgpu_uvd_entity_init(adev);
-
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
index 5a8116437abf..f8b229b75435 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
@@ -125,8 +125,6 @@ static int uvd_v5_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	r = amdgpu_uvd_entity_init(adev);
-
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
index 74c09230aeb3..a9a6880f44e3 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
@@ -432,8 +432,6 @@ static int uvd_v6_0_sw_init(void *handle)
 		}
 	}
 
-	r = amdgpu_uvd_entity_init(adev);
-
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
index 1c42cf10cc29..6068b784dc69 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
@@ -480,10 +480,6 @@ static int uvd_v7_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	r = amdgpu_uvd_entity_init(adev);
-	if (r)
-		return r;
-
 	r = amdgpu_virt_alloc_mm_table(adev);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
index 67eb01fef789..a08e7abca423 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
@@ -441,8 +441,6 @@ static int vce_v2_0_sw_init(void *handle)
 			return r;
 	}
 
-	r = amdgpu_vce_entity_init(adev);
-
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index 18f6e62af339..f4760748d349 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -450,8 +450,6 @@ static int vce_v3_0_sw_init(void *handle)
 			return r;
 	}
 
-	r = amdgpu_vce_entity_init(adev);
-
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
index e0b70cd3b697..06d787385ad4 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
@@ -486,11 +486,6 @@ static int vce_v4_0_sw_init(void *handle)
 			return r;
 	}
 
-
-	r = amdgpu_vce_entity_init(adev);
-	if (r)
-		return r;
-
 	r = amdgpu_virt_alloc_mm_table(adev);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index fbf053001af9..7a33e06f5c90 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1416,8 +1416,13 @@ bool kfd_process_xnack_mode(struct kfd_process *p, bool supported)
 		 * per-process XNACK mode selection. But let the dev->noretry
 		 * setting still influence the default XNACK mode.
 		 */
-		if (supported && KFD_SUPPORT_XNACK_PER_PROCESS(dev))
+		if (supported && KFD_SUPPORT_XNACK_PER_PROCESS(dev)) {
+			if (!amdgpu_sriov_xnack_support(dev->kfd->adev)) {
+				pr_debug("SRIOV platform xnack not supported\n");
+				return false;
+			}
 			continue;
+		}
 
 		/* GFXv10 and later GPUs do not support shader preemption
 		 * during page faults. This can lead to poor QoS for queue
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index e67d06a42809..f2f3c338fd94 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1255,9 +1255,11 @@ svm_range_get_pte_flags(struct kfd_node *node,
 		}
 		break;
 	case IP_VERSION(9, 4, 3):
-		mtype_local = amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC :
-			      (amdgpu_mtype_local == 2 || ext_coherent ?
-					AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW);
+		if (ext_coherent)
+			mtype_local = node->adev->rev_id ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_UC;
+		else
+			mtype_local = amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC :
+				amdgpu_mtype_local == 2 ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
 		snoop = true;
 		if (uncached) {
 			mapping_flags |= AMDGPU_VM_MTYPE_UC;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index 9cd83b780102..ed784cf27d39 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -1216,6 +1216,9 @@ bool dm_helpers_dp_handle_test_pattern_request(
 
 	}
 
+	pipe_ctx->stream->test_pattern.type = test_pattern;
+	pipe_ctx->stream->test_pattern.color_space = test_pattern_color_space;
+
 	dc_link_dp_set_test_pattern(
 		(struct dc_link *) link,
 		test_pattern,
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
index 0f580ea37576..133af994a08c 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
@@ -37,7 +37,7 @@
 #include <drm/drm_framebuffer.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_atomic.h>
-#include "dcn10/dcn10_optc.h"
+#include "dc/inc/hw/optc.h"
 
 #include "dc/inc/core_types.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
index f80917f6153b..0fa4fcd00de2 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
@@ -111,17 +111,21 @@ static int dcn35_get_active_display_cnt_wa(
 	return display_count;
 }
 
-static void dcn35_disable_otg_wa(struct clk_mgr *clk_mgr_base, struct dc_state *context, bool disable)
+static void dcn35_disable_otg_wa(struct clk_mgr *clk_mgr_base, struct dc_state *context,
+		bool safe_to_lower, bool disable)
 {
 	struct dc *dc = clk_mgr_base->ctx->dc;
 	int i;
 
 	for (i = 0; i < dc->res_pool->pipe_count; ++i) {
-		struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+		struct pipe_ctx *pipe = safe_to_lower
+			? &context->res_ctx.pipe_ctx[i]
+			: &dc->current_state->res_ctx.pipe_ctx[i];
 
 		if (pipe->top_pipe || pipe->prev_odm_pipe)
 			continue;
-		if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) {
+		if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal) ||
+				     !pipe->stream->link_enc)) {
 			struct stream_encoder *stream_enc = pipe->stream_res.stream_enc;
 
 			if (disable) {
@@ -301,11 +305,11 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 	}
 
 	if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)) {
-		dcn35_disable_otg_wa(clk_mgr_base, context, true);
+		dcn35_disable_otg_wa(clk_mgr_base, context, safe_to_lower, true);
 
 		clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz;
 		dcn35_smu_set_dispclk(clk_mgr, clk_mgr_base->clks.dispclk_khz);
-		dcn35_disable_otg_wa(clk_mgr_base, context, false);
+		dcn35_disable_otg_wa(clk_mgr_base, context, safe_to_lower, false);
 
 		update_dispclk = true;
 	}
@@ -814,7 +818,8 @@ static void dcn35_set_idle_state(struct clk_mgr *clk_mgr_base, bool allow_idle)
 	struct dc *dc = clk_mgr_base->ctx->dc;
 	uint32_t val = dcn35_smu_read_ips_scratch(clk_mgr);
 
-	if (dc->config.disable_ips == 0) {
+	if (dc->config.disable_ips == DMUB_IPS_ENABLE ||
+		dc->config.disable_ips == DMUB_IPS_DISABLE_DYNAMIC) {
 		val |= DMUB_IPS1_ALLOW_MASK;
 		val |= DMUB_IPS2_ALLOW_MASK;
 	} else if (dc->config.disable_ips == DMUB_IPS_DISABLE_IPS1) {
@@ -1114,7 +1119,7 @@ void dcn35_clk_mgr_construct(
 		dm_helpers_free_gpu_mem(clk_mgr->base.base.ctx, DC_MEM_ALLOC_TYPE_FRAME_BUFFER,
 				smu_dpm_clks.dpm_clks);
 
-	if (ctx->dc->config.disable_ips == 0) {
+	if (ctx->dc->config.disable_ips != DMUB_IPS_DISABLE_ALL) {
 		bool ips_support = false;
 
 		/*avoid call pmfw at init*/
@@ -1127,7 +1132,7 @@ void dcn35_clk_mgr_construct(
 			ctx->dc->debug.disable_hpo_power_gate = false;
 		} else {
 			/*let's reset the config control flag*/
-			ctx->dc->config.disable_ips = 1; /*pmfw not support it, disable it all*/
+			ctx->dc->config.disable_ips = DMUB_IPS_DISABLE_ALL; /*pmfw not support it, disable it all*/
 		}
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 4360a696f10a..7b9bf5cb4529 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2582,6 +2582,9 @@ static enum surface_update_type det_surface_update(const struct dc *dc,
 	if (u->gamut_remap_matrix)
 		update_flags->bits.gamut_remap_change = 1;
 
+	if (u->blend_tf)
+		update_flags->bits.gamma_change = 1;
+
 	if (u->gamma) {
 		enum surface_pixel_format format = SURFACE_PIXEL_FORMAT_GRPH_BEGIN;
 
@@ -4113,8 +4116,17 @@ static bool commit_minimal_transition_state_for_windowed_mpo_odm(struct dc *dc,
 	bool success = false;
 	struct dc_state *minimal_transition_context;
 	struct pipe_split_policy_backup policy;
+	struct mall_temp_config mall_temp_config;
 
 	/* commit based on new context */
+	/* Since all phantom pipes are removed in full validation,
+	 * we have to save and restore the subvp/mall config when
+	 * we do a minimal transition since the flags marking the
+	 * pipe as subvp/phantom will be cleared (dc copy constructor
+	 * creates a shallow copy).
+	 */
+	if (dc->res_pool->funcs->save_mall_state)
+		dc->res_pool->funcs->save_mall_state(dc, context, &mall_temp_config);
 	minimal_transition_context = create_minimal_transition_state(dc,
 			context, &policy);
 	if (minimal_transition_context) {
@@ -4123,9 +4135,20 @@ static bool commit_minimal_transition_state_for_windowed_mpo_odm(struct dc *dc,
 			dc->hwss.is_pipe_topology_transition_seamless(
 					dc, minimal_transition_context, context)) {
 			DC_LOG_DC("%s base = new state\n", __func__);
+
 			success = dc_commit_state_no_check(dc, minimal_transition_context) == DC_OK;
 		}
 		release_minimal_transition_state(dc, minimal_transition_context, &policy);
+		if (dc->res_pool->funcs->restore_mall_state)
+			dc->res_pool->funcs->restore_mall_state(dc, context, &mall_temp_config);
+		/* If we do a minimal transition with plane removal and the context
+		 * has subvp we also have to retain back the phantom stream / planes
+		 * since the refcount is decremented as part of the min transition
+		 * (we commit a state with no subvp, so the phantom streams / planes
+		 * had to be removed).
+		 */
+		if (dc->res_pool->funcs->retain_phantom_pipes)
+			dc->res_pool->funcs->retain_phantom_pipes(dc, context);
 	}
 
 	if (!success) {
@@ -4884,7 +4907,7 @@ void dc_allow_idle_optimizations(struct dc *dc, bool allow)
 	if (dc->debug.disable_idle_power_optimizations)
 		return;
 
-	if (dc->caps.ips_support && dc->config.disable_ips)
+	if (dc->caps.ips_support && (dc->config.disable_ips == DMUB_IPS_DISABLE_ALL))
 		return;
 
 	if (dc->clk_mgr != NULL && dc->clk_mgr->funcs->is_smu_present)
@@ -4905,7 +4928,7 @@ bool dc_dmub_is_ips_idle_state(struct dc *dc)
 	if (dc->debug.disable_idle_power_optimizations)
 		return false;
 
-	if (!dc->caps.ips_support || dc->config.disable_ips)
+	if (!dc->caps.ips_support || (dc->config.disable_ips == DMUB_IPS_DISABLE_ALL))
 		return false;
 
 	if (dc->hwss.get_idle_state)
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 6e54ca055fcb..9316b737a8ba 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -49,7 +49,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.256"
+#define DC_VER "3.2.259"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index ba142bef626b..e4c007203318 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -120,6 +120,80 @@ void dc_dmub_srv_send_inbox0_cmd(struct dc_dmub_srv *dc_dmub_srv,
 	}
 }
 
+bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
+		unsigned int count,
+		union dmub_rb_cmd *cmd_list)
+{
+	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
+	struct dmub_srv *dmub;
+	enum dmub_status status;
+	int i;
+
+	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
+		return false;
+
+	dmub = dc_dmub_srv->dmub;
+
+	for (i = 0 ; i < count; i++) {
+		// Queue command
+		status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
+
+		if (status == DMUB_STATUS_QUEUE_FULL) {
+			/* Execute and wait for queue to become empty again. */
+			dmub_srv_cmd_execute(dmub);
+			dmub_srv_wait_for_idle(dmub, 100000);
+
+			/* Requeue the command. */
+			status = dmub_srv_cmd_queue(dmub, &cmd_list[i]);
+		}
+
+		if (status != DMUB_STATUS_OK) {
+			DC_ERROR("Error queueing DMUB command: status=%d\n", status);
+			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
+			return false;
+		}
+	}
+
+	status = dmub_srv_cmd_execute(dmub);
+	if (status != DMUB_STATUS_OK) {
+		DC_ERROR("Error starting DMUB execution: status=%d\n", status);
+		dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
+		return false;
+	}
+
+	return true;
+}
+
+bool dc_dmub_srv_wait_for_idle(struct dc_dmub_srv *dc_dmub_srv,
+		enum dm_dmub_wait_type wait_type,
+		union dmub_rb_cmd *cmd_list)
+{
+	struct dmub_srv *dmub;
+	enum dmub_status status;
+
+	if (!dc_dmub_srv || !dc_dmub_srv->dmub)
+		return false;
+
+	dmub = dc_dmub_srv->dmub;
+
+	// Wait for DMUB to process command
+	if (wait_type != DM_DMUB_WAIT_TYPE_NO_WAIT) {
+		status = dmub_srv_wait_for_idle(dmub, 100000);
+
+		if (status != DMUB_STATUS_OK) {
+			DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status);
+			dc_dmub_srv_log_diagnostic_data(dc_dmub_srv);
+			return false;
+		}
+
+		// Copy data back from ring buffer into command
+		if (wait_type == DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)
+			dmub_rb_get_return_data(&dmub->inbox1_rb, cmd_list);
+	}
+
+	return true;
+}
+
 bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type)
 {
 	return dc_dmub_srv_cmd_run_list(dc_dmub_srv, 1, cmd, wait_type);
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
index 31150b214394..d4a60f53faab 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
@@ -56,6 +56,14 @@ void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv);
 
 bool dc_dmub_srv_optimized_init_done(struct dc_dmub_srv *dc_dmub_srv);
 
+bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv,
+		unsigned int count,
+		union dmub_rb_cmd *cmd_list);
+
+bool dc_dmub_srv_wait_for_idle(struct dc_dmub_srv *dc_dmub_srv,
+		enum dm_dmub_wait_type wait_type,
+		union dmub_rb_cmd *cmd_list);
+
 bool dc_dmub_srv_cmd_run(struct dc_dmub_srv *dc_dmub_srv, union dmub_rb_cmd *cmd, enum dm_dmub_wait_type wait_type);
 
 bool dc_dmub_srv_cmd_run_list(struct dc_dmub_srv *dc_dmub_srv, unsigned int count, union dmub_rb_cmd *cmd_list, enum dm_dmub_wait_type wait_type);
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
index 35ae245ef722..eeeeeef4d717 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -142,7 +142,8 @@ enum dp_test_link_rate {
 	DP_TEST_LINK_RATE_HBR3		= 0x1E,
 	DP_TEST_LINK_RATE_UHBR10	= 0x01,
 	DP_TEST_LINK_RATE_UHBR20	= 0x02,
-	DP_TEST_LINK_RATE_UHBR13_5	= 0x03,
+	DP_TEST_LINK_RATE_UHBR13_5_LEGACY	= 0x03, /* For backward compatibility*/
+	DP_TEST_LINK_RATE_UHBR13_5	= 0x04,
 };
 
 struct dc_link_settings {
diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 40dc51853d62..cea666ea66c6 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -1037,7 +1037,9 @@ struct replay_config {
 	bool replay_smu_opt_supported;                  // SMU optimization is supported
 	unsigned int replay_enable_option;              // Replay enablement option
 	uint32_t debug_flags;                           // Replay debug flags
-	bool replay_timing_sync_supported;             // Replay desync is supported
+	bool replay_timing_sync_supported; // Replay desync is supported
+	bool force_disable_desync_error_check;             // Replay desync is supported
+	bool received_desync_error_hpd; //Replay Received Desync Error HPD.
 	union replay_error_status replay_error_status; // Replay error status
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h
index c50aa30614be..051e4c2b4cf2 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.h
@@ -128,21 +128,6 @@
 	SRI(DC_ABM1_ACE_THRES_12, ABM, id), \
 	NBIO_SR(BIOS_SCRATCH_2)
 
-#define ABM_DCN32_REG_LIST(id)\
-	SRI(DC_ABM1_HG_SAMPLE_RATE, ABM, id), \
-	SRI(DC_ABM1_LS_SAMPLE_RATE, ABM, id), \
-	SRI(BL1_PWM_BL_UPDATE_SAMPLE_RATE, ABM, id), \
-	SRI(DC_ABM1_HG_MISC_CTRL, ABM, id), \
-	SRI(DC_ABM1_IPCSC_COEFF_SEL, ABM, id), \
-	SRI(BL1_PWM_CURRENT_ABM_LEVEL, ABM, id), \
-	SRI(BL1_PWM_TARGET_ABM_LEVEL, ABM, id), \
-	SRI(BL1_PWM_USER_LEVEL, ABM, id), \
-	SRI(DC_ABM1_LS_MIN_MAX_PIXEL_VALUE_THRES, ABM, id), \
-	SRI(DC_ABM1_HGLS_REG_READ_PROGRESS, ABM, id), \
-	SRI(DC_ABM1_ACE_OFFSET_SLOPE_0, ABM, id), \
-	SRI(DC_ABM1_ACE_THRES_12, ABM, id), \
-	NBIO_SR(BIOS_SCRATCH_2)
-
 #define ABM_SF(reg_name, field_name, post_fix)\
 	.field_name = reg_name ## __ ## field_name ## post_fix
 
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
index aaf6c981fd9e..ab81594a7fad 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h
@@ -26,7 +26,7 @@
 #ifndef __DC_TIMING_GENERATOR_DCN10_H__
 #define __DC_TIMING_GENERATOR_DCN10_H__
 
-#include "timing_generator.h"
+#include "optc.h"
 
 #define DCN10TG_FROM_TG(tg)\
 	container_of(tg, struct optc, base)
@@ -594,190 +594,6 @@ struct dcn_optc_mask {
 	TG_REG_FIELD_LIST_DCN3_5(uint32_t)
 };
 
-struct optc {
-	struct timing_generator base;
-
-	const struct dcn_optc_registers *tg_regs;
-	const struct dcn_optc_shift *tg_shift;
-	const struct dcn_optc_mask *tg_mask;
-
-	int opp_count;
-
-	uint32_t max_h_total;
-	uint32_t max_v_total;
-
-	uint32_t min_h_blank;
-
-	uint32_t min_h_sync_width;
-	uint32_t min_v_sync_width;
-	uint32_t min_v_blank;
-	uint32_t min_v_blank_interlace;
-
-	int vstartup_start;
-	int vupdate_offset;
-	int vupdate_width;
-	int vready_offset;
-	struct dc_crtc_timing orginal_patched_timing;
-	enum signal_type signal;
-};
-
 void dcn10_timing_generator_init(struct optc *optc);
 
-struct dcn_otg_state {
-	uint32_t v_blank_start;
-	uint32_t v_blank_end;
-	uint32_t v_sync_a_pol;
-	uint32_t v_total;
-	uint32_t v_total_max;
-	uint32_t v_total_min;
-	uint32_t v_total_min_sel;
-	uint32_t v_total_max_sel;
-	uint32_t v_sync_a_start;
-	uint32_t v_sync_a_end;
-	uint32_t h_blank_start;
-	uint32_t h_blank_end;
-	uint32_t h_sync_a_start;
-	uint32_t h_sync_a_end;
-	uint32_t h_sync_a_pol;
-	uint32_t h_total;
-	uint32_t underflow_occurred_status;
-	uint32_t otg_enabled;
-	uint32_t blank_enabled;
-	uint32_t vertical_interrupt1_en;
-	uint32_t vertical_interrupt1_line;
-	uint32_t vertical_interrupt2_en;
-	uint32_t vertical_interrupt2_line;
-};
-
-void optc1_read_otg_state(struct optc *optc1,
-		struct dcn_otg_state *s);
-
-bool optc1_get_hw_timing(struct timing_generator *tg,
-		struct dc_crtc_timing *hw_crtc_timing);
-
-bool optc1_validate_timing(
-	struct timing_generator *optc,
-	const struct dc_crtc_timing *timing);
-
-void optc1_program_timing(
-	struct timing_generator *optc,
-	const struct dc_crtc_timing *dc_crtc_timing,
-	int vready_offset,
-	int vstartup_start,
-	int vupdate_offset,
-	int vupdate_width,
-	const enum signal_type signal,
-	bool use_vbios);
-
-void optc1_setup_vertical_interrupt0(
-		struct timing_generator *optc,
-		uint32_t start_line,
-		uint32_t end_line);
-void optc1_setup_vertical_interrupt1(
-		struct timing_generator *optc,
-		uint32_t start_line);
-void optc1_setup_vertical_interrupt2(
-		struct timing_generator *optc,
-		uint32_t start_line);
-
-void optc1_program_global_sync(
-		struct timing_generator *optc,
-		int vready_offset,
-		int vstartup_start,
-		int vupdate_offset,
-		int vupdate_width);
-
-bool optc1_disable_crtc(struct timing_generator *optc);
-
-bool optc1_is_counter_moving(struct timing_generator *optc);
-
-void optc1_get_position(struct timing_generator *optc,
-		struct crtc_position *position);
-
-uint32_t optc1_get_vblank_counter(struct timing_generator *optc);
-
-void optc1_get_crtc_scanoutpos(
-	struct timing_generator *optc,
-	uint32_t *v_blank_start,
-	uint32_t *v_blank_end,
-	uint32_t *h_position,
-	uint32_t *v_position);
-
-void optc1_set_early_control(
-	struct timing_generator *optc,
-	uint32_t early_cntl);
-
-void optc1_wait_for_state(struct timing_generator *optc,
-		enum crtc_state state);
-
-void optc1_set_blank(struct timing_generator *optc,
-		bool enable_blanking);
-
-bool optc1_is_blanked(struct timing_generator *optc);
-
-void optc1_program_blank_color(
-		struct timing_generator *optc,
-		const struct tg_color *black_color);
-
-bool optc1_did_triggered_reset_occur(
-	struct timing_generator *optc);
-
-void optc1_enable_reset_trigger(struct timing_generator *optc, int source_tg_inst);
-
-void optc1_disable_reset_trigger(struct timing_generator *optc);
-
-void optc1_lock(struct timing_generator *optc);
-
-void optc1_unlock(struct timing_generator *optc);
-
-void optc1_enable_optc_clock(struct timing_generator *optc, bool enable);
-
-void optc1_set_drr(
-	struct timing_generator *optc,
-	const struct drr_params *params);
-
-void optc1_set_vtotal_min_max(struct timing_generator *optc, int vtotal_min, int vtotal_max);
-
-void optc1_set_static_screen_control(
-	struct timing_generator *optc,
-	uint32_t event_triggers,
-	uint32_t num_frames);
-
-void optc1_program_stereo(struct timing_generator *optc,
-	const struct dc_crtc_timing *timing, struct crtc_stereo_flags *flags);
-
-bool optc1_is_stereo_left_eye(struct timing_generator *optc);
-
-void optc1_clear_optc_underflow(struct timing_generator *optc);
-
-void optc1_tg_init(struct timing_generator *optc);
-
-bool optc1_is_tg_enabled(struct timing_generator *optc);
-
-bool optc1_is_optc_underflow_occurred(struct timing_generator *optc);
-
-void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enable);
-
-void optc1_set_timing_double_buffer(struct timing_generator *optc, bool enable);
-
-bool optc1_get_otg_active_size(struct timing_generator *optc,
-		uint32_t *otg_active_width,
-		uint32_t *otg_active_height);
-
-void optc1_enable_crtc_reset(
-		struct timing_generator *optc,
-		int source_tg_inst,
-		struct crtc_trigger_info *crtc_tp);
-
-bool optc1_configure_crc(struct timing_generator *optc,
-			  const struct crc_params *params);
-
-bool optc1_get_crc(struct timing_generator *optc,
-		    uint32_t *r_cr, uint32_t *g_y, uint32_t *b_cb);
-
-bool optc1_is_two_pixels_per_containter(const struct dc_crtc_timing *timing);
-
-void optc1_set_vtg_params(struct timing_generator *optc,
-		const struct dc_crtc_timing *dc_crtc_timing, bool program_fp2);
-
 #endif /* __DC_TIMING_GENERATOR_DCN10_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c
index 5eebe7f03ddc..c9ae2d8f0096 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dsc.c
@@ -137,7 +137,15 @@ void dsc2_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz)
 		dsc_enc_caps->max_total_throughput_mps = DCN20_MAX_DISPLAY_CLOCK_Mhz * 2;
 	}
 
-	// TODO DSC: This is actually image width limitation, not a slice width. This should be added to the criteria to use ODM.
+	/* For pixel clock bigger than a single-pipe limit needing four engines ODM 4:1, which then quardruples our
+	 * throughput and number of slices
+	 */
+	if (pixel_clock_100Hz > DCN20_MAX_PIXEL_CLOCK_Mhz*10000*2) {
+		dsc_enc_caps->slice_caps.bits.NUM_SLICES_12 = 1;
+		dsc_enc_caps->slice_caps.bits.NUM_SLICES_16 = 1;
+		dsc_enc_caps->max_total_throughput_mps = DCN20_MAX_DISPLAY_CLOCK_Mhz * 4;
+	}
+
 	dsc_enc_caps->max_slice_width = 5184; /* (including 64 overlap pixels for eDP MSO mode) */
 	dsc_enc_caps->bpp_increment_div = 16; /* 1/16th of a bit */
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dccg.c
index addedcfd1238..479f3683c0b7 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_dccg.c
@@ -325,6 +325,43 @@ static void dccg35_set_dpstreamclk(
 	}
 }
 
+static void dccg35_set_physymclk_root_clock_gating(
+		struct dccg *dccg,
+		int phy_inst,
+		bool enable)
+{
+	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+
+	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
+		return;
+
+	switch (phy_inst) {
+	case 0:
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
+				PHYASYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+		break;
+	case 1:
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
+				PHYBSYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+		break;
+	case 2:
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
+				PHYCSYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+		break;
+	case 3:
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
+				PHYDSYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+		break;
+	case 4:
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
+				PHYESYMCLK_ROOT_GATE_DISABLE, enable ? 1 : 0);
+		break;
+	default:
+		BREAK_TO_DEBUGGER();
+		return;
+	}
+}
+
 static void dccg35_set_physymclk(
 		struct dccg *dccg,
 		int phy_inst,
@@ -340,16 +377,10 @@ static void dccg35_set_physymclk(
 			REG_UPDATE_2(PHYASYMCLK_CLOCK_CNTL,
 					PHYASYMCLK_EN, 1,
 					PHYASYMCLK_SRC_SEL, clk_src);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYASYMCLK_ROOT_GATE_DISABLE, 1);
 		} else {
 			REG_UPDATE_2(PHYASYMCLK_CLOCK_CNTL,
 					PHYASYMCLK_EN, 0,
 					PHYASYMCLK_SRC_SEL, 0);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYASYMCLK_ROOT_GATE_DISABLE, 0);
 		}
 		break;
 	case 1:
@@ -357,16 +388,10 @@ static void dccg35_set_physymclk(
 			REG_UPDATE_2(PHYBSYMCLK_CLOCK_CNTL,
 					PHYBSYMCLK_EN, 1,
 					PHYBSYMCLK_SRC_SEL, clk_src);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYBSYMCLK_ROOT_GATE_DISABLE, 1);
 		} else {
 			REG_UPDATE_2(PHYBSYMCLK_CLOCK_CNTL,
 					PHYBSYMCLK_EN, 0,
 					PHYBSYMCLK_SRC_SEL, 0);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYBSYMCLK_ROOT_GATE_DISABLE, 0);
 		}
 		break;
 	case 2:
@@ -374,16 +399,10 @@ static void dccg35_set_physymclk(
 			REG_UPDATE_2(PHYCSYMCLK_CLOCK_CNTL,
 					PHYCSYMCLK_EN, 1,
 					PHYCSYMCLK_SRC_SEL, clk_src);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYCSYMCLK_ROOT_GATE_DISABLE, 1);
 		} else {
 			REG_UPDATE_2(PHYCSYMCLK_CLOCK_CNTL,
 					PHYCSYMCLK_EN, 0,
 					PHYCSYMCLK_SRC_SEL, 0);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYCSYMCLK_ROOT_GATE_DISABLE, 0);
 		}
 		break;
 	case 3:
@@ -391,16 +410,10 @@ static void dccg35_set_physymclk(
 			REG_UPDATE_2(PHYDSYMCLK_CLOCK_CNTL,
 					PHYDSYMCLK_EN, 1,
 					PHYDSYMCLK_SRC_SEL, clk_src);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYDSYMCLK_ROOT_GATE_DISABLE, 1);
 		} else {
 			REG_UPDATE_2(PHYDSYMCLK_CLOCK_CNTL,
 					PHYDSYMCLK_EN, 0,
 					PHYDSYMCLK_SRC_SEL, 0);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYDSYMCLK_ROOT_GATE_DISABLE, 0);
 		}
 		break;
 	case 4:
@@ -408,16 +421,10 @@ static void dccg35_set_physymclk(
 			REG_UPDATE_2(PHYESYMCLK_CLOCK_CNTL,
 					PHYESYMCLK_EN, 1,
 					PHYESYMCLK_SRC_SEL, clk_src);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYESYMCLK_ROOT_GATE_DISABLE, 1);
 		} else {
 			REG_UPDATE_2(PHYESYMCLK_CLOCK_CNTL,
 					PHYESYMCLK_EN, 0,
 					PHYESYMCLK_SRC_SEL, 0);
-			if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
-				REG_UPDATE(DCCG_GATE_DISABLE_CNTL2,
-					PHYESYMCLK_ROOT_GATE_DISABLE, 0);
 		}
 		break;
 	default:
@@ -490,8 +497,8 @@ void dccg35_init(struct dccg *dccg)
 
 	if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk)
 		for (otg_inst = 0; otg_inst < 5; otg_inst++)
-			dccg35_set_physymclk(dccg, otg_inst,
-					     PHYSYMCLK_FORCE_SRC_SYMCLK, false);
+			dccg35_set_physymclk_root_clock_gating(dccg, otg_inst,
+					false);
 /*
 	dccg35_enable_global_fgcg_rep(
 		dccg, dccg->ctx->dc->debug.enable_fine_grain_clock_gating.bits
@@ -754,7 +761,9 @@ static const struct dccg_funcs dccg35_funcs = {
 	.disable_symclk32_se = dccg31_disable_symclk32_se,
 	.enable_symclk32_le = dccg31_enable_symclk32_le,
 	.disable_symclk32_le = dccg31_disable_symclk32_le,
+	.set_symclk32_le_root_clock_gating = dccg31_set_symclk32_le_root_clock_gating,
 	.set_physymclk = dccg35_set_physymclk,
+	.set_physymclk_root_clock_gating = dccg35_set_physymclk_root_clock_gating,
 	.set_dtbclk_dto = dccg35_set_dtbclk_dto,
 	.set_audio_dtbclk_dto = dccg31_set_audio_dtbclk_dto,
 	.set_fifo_errdet_ovr_en = dccg2_set_fifo_errdet_ovr_en,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_pg_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_pg_cntl.c
index 0f60c40e1fc5..46f71ff08fd1 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_pg_cntl.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_pg_cntl.c
@@ -332,6 +332,13 @@ void pg_cntl35_io_clk_pg_control(struct pg_cntl *pg_cntl, bool power_on)
 	pg_cntl->pg_res_enable[PG_DCIO] = power_on;
 }
 
+void pg_cntl35_set_force_poweron_domain22(struct pg_cntl *pg_cntl, bool power_on)
+{
+	struct dcn_pg_cntl *pg_cntl_dcn = TO_DCN_PG_CNTL(pg_cntl);
+
+	REG_UPDATE(DOMAIN22_PG_CONFIG, DOMAIN_POWER_FORCEON, power_on ? 1 : 0);
+}
+
 static bool pg_cntl35_plane_otg_status(struct pg_cntl *pg_cntl)
 {
 	struct dcn_pg_cntl *pg_cntl_dcn = TO_DCN_PG_CNTL(pg_cntl);
@@ -501,7 +508,8 @@ static const struct pg_cntl_funcs pg_cntl35_funcs = {
 	.mpcc_pg_control = pg_cntl35_mpcc_pg_control,
 	.opp_pg_control = pg_cntl35_opp_pg_control,
 	.optc_pg_control = pg_cntl35_optc_pg_control,
-	.dwb_pg_control = pg_cntl35_dwb_pg_control
+	.dwb_pg_control = pg_cntl35_dwb_pg_control,
+	.set_force_poweron_domain22 = pg_cntl35_set_force_poweron_domain22
 };
 
 struct pg_cntl *pg_cntl35_create(
diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_pg_cntl.h b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_pg_cntl.h
index 3de240884d22..069dae08e222 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_pg_cntl.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_pg_cntl.h
@@ -183,6 +183,7 @@ void pg_cntl35_optc_pg_control(struct pg_cntl *pg_cntl,
 	unsigned int optc_inst, bool power_on);
 void pg_cntl35_dwb_pg_control(struct pg_cntl *pg_cntl, bool power_on);
 void pg_cntl35_init_pg_status(struct pg_cntl *pg_cntl);
+void pg_cntl35_set_force_poweron_domain22(struct pg_cntl *pg_cntl, bool power_on);
 
 struct pg_cntl *pg_cntl35_create(
 	struct dc_context *ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c
index 3c7c810bab1f..c7e011d26d41 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn35/dcn35_resource.c
@@ -610,7 +610,23 @@ static struct dce_hwseq_registers hwseq_reg;
 	HWS_SF(, DMU_CLK_CNTL, LONO_FGCG_REP_DIS, mask_sh),\
 	HWS_SF(, DMU_CLK_CNTL, LONO_DISPCLK_GATE_DISABLE, mask_sh),\
 	HWS_SF(, DMU_CLK_CNTL, LONO_SOCCLK_GATE_DISABLE, mask_sh),\
-	HWS_SF(, DMU_CLK_CNTL, LONO_DMCUBCLK_GATE_DISABLE, mask_sh)
+	HWS_SF(, DMU_CLK_CNTL, LONO_DMCUBCLK_GATE_DISABLE, mask_sh),\
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKA_FE_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKB_FE_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKC_FE_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKD_FE_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKE_FE_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, HDMICHARCLK0_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKA_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKB_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKC_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKD_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, SYMCLKE_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, PHYASYMCLK_ROOT_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, PHYBSYMCLK_ROOT_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, PHYCSYMCLK_ROOT_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, PHYDSYMCLK_ROOT_GATE_DISABLE, mask_sh), \
+	HWS_SF(, DCCG_GATE_DISABLE_CNTL2, PHYESYMCLK_ROOT_GATE_DISABLE, mask_sh)
 
 static const struct dce_hwseq_shift hwseq_shift = {
 		HWSEQ_DCN35_MASK_SH_LIST(__SHIFT)
@@ -708,7 +724,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 			.i2c = true,
 			.dmcu = false, // This is previously known to cause hang on S3 cycles if enabled
 			.dscl = true,
-			.cm = true,
+			.cm = false,
 			.mpc = true,
 			.optc = true,
 			.vpg = true,
@@ -719,14 +735,14 @@ static const struct dc_debug_options debug_defaults_drv = {
 		.bits = {
 			.dpp = true,
 			.dsc = true,/*dscclk and dsc pg*/
-			.hdmistream = false,
-			.hdmichar = false,
-			.dpstream = false,
-			.symclk32_se = false,
-			.symclk32_le = false,
-			.symclk_fe = false,
-			.physymclk = false,
-			.dpiasymclk = false,
+			.hdmistream = true,
+			.hdmichar = true,
+			.dpstream = true,
+			.symclk32_se = true,
+			.symclk32_le = true,
+			.symclk_fe = true,
+			.physymclk = true,
+			.dpiasymclk = true,
 		}
 	},
 	.seamless_boot_odm_combine = DML_FAIL_SOURCE_PIXEL_FORMAT,
@@ -741,7 +757,6 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_boot_optimizations = false,
 	.disable_unbounded_requesting = false,
 	.disable_mem_low_power = false,
-	.enable_hpo_pg_support = false,
 	//must match enable_single_display_2to1_odm_policy to support dynamic ODM transitions
 	.enable_double_buffered_dsc_pg_support = true,
 	.enable_dp_dig_pixel_rate_div_policy = 1,
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
index ad741a723c0e..3686f1e7de3a 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c
@@ -5128,7 +5128,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 			ViewportExceedsSurface = true;
 
 		if (v->SourcePixelFormat[k] != dm_444_64 && v->SourcePixelFormat[k] != dm_444_32 && v->SourcePixelFormat[k] != dm_444_16
-				&& v->SourcePixelFormat[k] != dm_444_16 && v->SourcePixelFormat[k] != dm_444_8 && v->SourcePixelFormat[k] != dm_rgbe) {
+				&& v->SourcePixelFormat[k] != dm_444_8 && v->SourcePixelFormat[k] != dm_rgbe) {
 			if (v->ViewportWidthChroma[k] > v->SurfaceWidthC[k] || v->ViewportHeightChroma[k] > v->SurfaceHeightC[k]) {
 				ViewportExceedsSurface = true;
 			}
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
index d2046e770c50..1a2b24cc6b61 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
@@ -55,10 +55,11 @@ struct dc_pipe_mapping_scratch {
 	struct dc_plane_pipe_pool pipe_pool;
 };
 
-static bool get_plane_id(const struct dc_state *state, const struct dc_plane_state *plane,
-	unsigned int stream_id, unsigned int *plane_id)
+static bool get_plane_id(struct dml2_context *dml2, const struct dc_state *state, const struct dc_plane_state *plane,
+	unsigned int stream_id, unsigned int plane_index, unsigned int *plane_id)
 {
 	int i, j;
+	bool is_plane_duplicate = dml2->v20.scratch.plane_duplicate_exists;
 
 	if (!plane_id)
 		return false;
@@ -66,7 +67,8 @@ static bool get_plane_id(const struct dc_state *state, const struct dc_plane_sta
 	for (i = 0; i < state->stream_count; i++) {
 		if (state->streams[i]->stream_id == stream_id) {
 			for (j = 0; j < state->stream_status[i].plane_count; j++) {
-				if (state->stream_status[i].plane_states[j] == plane) {
+				if (state->stream_status[i].plane_states[j] == plane &&
+					(!is_plane_duplicate || (is_plane_duplicate && (j == plane_index)))) {
 					*plane_id = (i << 16) | j;
 					return true;
 				}
@@ -123,8 +125,9 @@ static struct pipe_ctx *find_master_pipe_of_plane(struct dml2_context *ctx,
 	unsigned int plane_id_assigned_to_pipe;
 
 	for (i = 0; i < ctx->config.dcn_pipe_count; i++) {
-		if (state->res_ctx.pipe_ctx[i].plane_state && get_plane_id(state, state->res_ctx.pipe_ctx[i].plane_state,
-			state->res_ctx.pipe_ctx[i].stream->stream_id, &plane_id_assigned_to_pipe)) {
+		if (state->res_ctx.pipe_ctx[i].plane_state && get_plane_id(ctx, state, state->res_ctx.pipe_ctx[i].plane_state,
+			state->res_ctx.pipe_ctx[i].stream->stream_id,
+			ctx->v20.scratch.dml_to_dc_pipe_mapping.dml_pipe_idx_to_plane_index[state->res_ctx.pipe_ctx[i].pipe_idx], &plane_id_assigned_to_pipe)) {
 			if (plane_id_assigned_to_pipe == plane_id)
 				return &state->res_ctx.pipe_ctx[i];
 		}
@@ -141,8 +144,9 @@ static unsigned int find_pipes_assigned_to_plane(struct dml2_context *ctx,
 	unsigned int plane_id_assigned_to_pipe;
 
 	for (i = 0; i < ctx->config.dcn_pipe_count; i++) {
-		if (state->res_ctx.pipe_ctx[i].plane_state && get_plane_id(state, state->res_ctx.pipe_ctx[i].plane_state,
-			state->res_ctx.pipe_ctx[i].stream->stream_id, &plane_id_assigned_to_pipe)) {
+		if (state->res_ctx.pipe_ctx[i].plane_state && get_plane_id(ctx, state, state->res_ctx.pipe_ctx[i].plane_state,
+			state->res_ctx.pipe_ctx[i].stream->stream_id,
+			ctx->v20.scratch.dml_to_dc_pipe_mapping.dml_pipe_idx_to_plane_index[state->res_ctx.pipe_ctx[i].pipe_idx], &plane_id_assigned_to_pipe)) {
 			if (plane_id_assigned_to_pipe == plane_id)
 				pipes[num_found++] = i;
 		}
@@ -609,6 +613,7 @@ static struct pipe_ctx *assign_pipes_to_plane(struct dml2_context *ctx, struct d
 		const struct dc_plane_state *plane,
 		int odm_factor,
 		int mpc_factor,
+		int plane_index,
 		struct dc_plane_pipe_pool *pipe_pool,
 		const struct dc_state *existing_state)
 {
@@ -620,7 +625,7 @@ static struct pipe_ctx *assign_pipes_to_plane(struct dml2_context *ctx, struct d
 	unsigned int next_pipe_to_assign;
 	int odm_slice, mpc_slice;
 
-	if (!get_plane_id(state, plane, stream->stream_id, &plane_id)) {
+	if (!get_plane_id(ctx, state, plane, stream->stream_id, plane_index, &plane_id)) {
 		ASSERT(false);
 		return master_pipe;
 	}
@@ -667,12 +672,16 @@ static void free_pipe(struct pipe_ctx *pipe)
 }
 
 static void free_unused_pipes_for_plane(struct dml2_context *ctx, struct dc_state *state,
-	const struct dc_plane_state *plane, const struct dc_plane_pipe_pool *pool, unsigned int stream_id)
+	const struct dc_plane_state *plane, const struct dc_plane_pipe_pool *pool, unsigned int stream_id, int plane_index)
 {
 	int i;
+	bool is_plane_duplicate = ctx->v20.scratch.plane_duplicate_exists;
+
 	for (i = 0; i < ctx->config.dcn_pipe_count; i++) {
 		if (state->res_ctx.pipe_ctx[i].plane_state == plane &&
 			state->res_ctx.pipe_ctx[i].stream->stream_id == stream_id &&
+			(!is_plane_duplicate || (is_plane_duplicate &&
+			ctx->v20.scratch.dml_to_dc_pipe_mapping.dml_pipe_idx_to_plane_index[state->res_ctx.pipe_ctx[i].pipe_idx] == plane_index)) &&
 			!is_pipe_used(pool, state->res_ctx.pipe_ctx[i].pipe_idx)) {
 			free_pipe(&state->res_ctx.pipe_ctx[i]);
 		}
@@ -717,19 +726,20 @@ static void map_pipes_for_stream(struct dml2_context *ctx, struct dc_state *stat
 }
 
 static void map_pipes_for_plane(struct dml2_context *ctx, struct dc_state *state, const struct dc_stream_state *stream, const struct dc_plane_state *plane,
-		struct dc_pipe_mapping_scratch *scratch, const struct dc_state *existing_state)
+		int plane_index, struct dc_pipe_mapping_scratch *scratch, const struct dc_state *existing_state)
 {
 	int odm_slice_index;
 	unsigned int plane_id;
 	struct pipe_ctx *master_pipe = NULL;
 	int i;
 
-	if (!get_plane_id(state, plane, stream->stream_id, &plane_id)) {
+	if (!get_plane_id(ctx, state, plane, stream->stream_id, plane_index, &plane_id)) {
 		ASSERT(false);
 		return;
 	}
 
-	master_pipe = assign_pipes_to_plane(ctx, state, stream, plane, scratch->odm_info.odm_factor, scratch->mpc_info.mpc_factor, &scratch->pipe_pool, existing_state);
+	master_pipe = assign_pipes_to_plane(ctx, state, stream, plane, scratch->odm_info.odm_factor,
+			scratch->mpc_info.mpc_factor, plane_index, &scratch->pipe_pool, existing_state);
 	sort_pipes_for_splitting(&scratch->pipe_pool);
 
 	for (odm_slice_index = 0; odm_slice_index < scratch->odm_info.odm_factor; odm_slice_index++) {
@@ -755,7 +765,7 @@ static void map_pipes_for_plane(struct dml2_context *ctx, struct dc_state *state
 		}
 	}
 
-	free_unused_pipes_for_plane(ctx, state, plane, &scratch->pipe_pool, stream->stream_id);
+	free_unused_pipes_for_plane(ctx, state, plane, &scratch->pipe_pool, stream->stream_id, plane_index);
 }
 
 static unsigned int get_mpc_factor(struct dml2_context *ctx,
@@ -768,7 +778,7 @@ static unsigned int get_mpc_factor(struct dml2_context *ctx,
 	unsigned int plane_id;
 	unsigned int cfg_idx;
 
-	get_plane_id(state, status->plane_states[plane_idx], stream_id, &plane_id);
+	get_plane_id(ctx, state, status->plane_states[plane_idx], stream_id, plane_idx, &plane_id);
 	cfg_idx = find_disp_cfg_idx_by_plane_id(mapping, plane_id);
 	if (ctx->architecture == dml2_architecture_20)
 		return (unsigned int)disp_cfg->hw.DPPPerSurface[cfg_idx];
@@ -911,26 +921,14 @@ bool dml2_map_dc_pipes(struct dml2_context *ctx, struct dc_state *state, const s
 	unsigned int stream_id;
 
 	const unsigned int *ODMMode, *DPPPerSurface;
-	unsigned int odm_mode_array[__DML2_WRAPPER_MAX_STREAMS_PLANES__] = {0}, dpp_per_surface_array[__DML2_WRAPPER_MAX_STREAMS_PLANES__] = {0};
 	struct dc_pipe_mapping_scratch scratch;
 
 	if (ctx->config.map_dc_pipes_with_callbacks)
 		return map_dc_pipes_with_callbacks(
 				ctx, state, disp_cfg, mapping, existing_state);
 
-	if (ctx->architecture == dml2_architecture_21) {
-		/*
-		 * Extract ODM and DPP outputs from DML2.1 and map them in an array as required for pipe mapping in dml2_map_dc_pipes.
-		 * As data cannot be directly extracted in const pointers, assign these arrays to const pointers before proceeding to
-		 * maximize the reuse of existing code. Const pointers are required because dml2.0 dml_display_cfg_st is const.
-		 *
-		 */
-		ODMMode = (const unsigned int *)odm_mode_array;
-		DPPPerSurface = (const unsigned int *)dpp_per_surface_array;
-	} else {
-		ODMMode = (unsigned int *)disp_cfg->hw.ODMMode;
-		DPPPerSurface = disp_cfg->hw.DPPPerSurface;
-	}
+	ODMMode = (unsigned int *)disp_cfg->hw.ODMMode;
+	DPPPerSurface = disp_cfg->hw.DPPPerSurface;
 
 	for (stream_index = 0; stream_index < state->stream_count; stream_index++) {
 		memset(&scratch, 0, sizeof(struct dc_pipe_mapping_scratch));
@@ -958,8 +956,8 @@ bool dml2_map_dc_pipes(struct dml2_context *ctx, struct dc_state *state, const s
 
 		for (plane_index = 0; plane_index < state->stream_status[stream_index].plane_count; plane_index++) {
 			// Planes are ordered top to bottom.
-			if (get_plane_id(state, state->stream_status[stream_index].plane_states[plane_index],
-				stream_id, &plane_id)) {
+			if (get_plane_id(ctx, state, state->stream_status[stream_index].plane_states[plane_index],
+				stream_id, plane_index, &plane_id)) {
 				plane_disp_cfg_index = find_disp_cfg_idx_by_plane_id(mapping, plane_id);
 
 				// Setup mpc_info for this plane
@@ -983,7 +981,8 @@ bool dml2_map_dc_pipes(struct dml2_context *ctx, struct dc_state *state, const s
 				// Clear the pool assignment scratch (which is per plane)
 				memset(&scratch.pipe_pool, 0, sizeof(struct dc_plane_pipe_pool));
 
-				map_pipes_for_plane(ctx, state, state->streams[stream_index], state->stream_status[stream_index].plane_states[plane_index], &scratch, existing_state);
+				map_pipes_for_plane(ctx, state, state->streams[stream_index],
+					state->stream_status[stream_index].plane_states[plane_index], plane_index, &scratch, existing_state);
 			} else {
 				// Plane ID cannot be generated, therefore no DML mapping can be performed.
 				ASSERT(false);
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_internal_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml2_internal_types.h
index ed5b767d46e0..1cf8a884c0fb 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_internal_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_internal_types.h
@@ -75,6 +75,8 @@ struct dml2_dml_to_dc_pipe_mapping {
 	bool dml_pipe_idx_to_stream_id_valid[__DML2_WRAPPER_MAX_STREAMS_PLANES__];
 	unsigned int dml_pipe_idx_to_plane_id[__DML2_WRAPPER_MAX_STREAMS_PLANES__];
 	bool dml_pipe_idx_to_plane_id_valid[__DML2_WRAPPER_MAX_STREAMS_PLANES__];
+	unsigned int dml_pipe_idx_to_plane_index[__DML2_WRAPPER_MAX_STREAMS_PLANES__];
+	bool dml_pipe_idx_to_plane_index_valid[__DML2_WRAPPER_MAX_STREAMS_PLANES__];
 };
 
 struct dml2_wrapper_scratch {
@@ -96,6 +98,7 @@ struct dml2_wrapper_scratch {
 
 	struct dml2_dml_to_dc_pipe_mapping dml_to_dc_pipe_mapping;
 	bool enable_flexible_pipe_mapping;
+	bool plane_duplicate_exists;
 };
 
 struct dml2_helper_det_policy_scratch {
@@ -104,7 +107,6 @@ struct dml2_helper_det_policy_scratch {
 
 enum dml2_architecture {
 	dml2_architecture_20,
-	dml2_architecture_21
 };
 
 struct dml2_context {
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c
index 89836f175a13..75171bee6f71 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c
@@ -231,6 +231,7 @@ void dml2_init_socbb_params(struct dml2_context *dml2, const struct dc *in_dc, s
 		out->num_chans = 4;
 		out->round_trip_ping_latency_dcfclk_cycles = 106;
 		out->smn_latency_us = 2;
+		out->dispclk_dppclk_vco_speed_mhz = 3600;
 		break;
 
 	case dml_project_dcn351:
@@ -930,10 +931,11 @@ static unsigned int map_stream_to_dml_display_cfg(const struct dml2_context *dml
 	return location;
 }
 
-static bool get_plane_id(const struct dc_state *context, const struct dc_plane_state *plane,
-		unsigned int stream_id, unsigned int *plane_id)
+static bool get_plane_id(struct dml2_context *dml2, const struct dc_state *context, const struct dc_plane_state *plane,
+		unsigned int stream_id, unsigned int plane_index, unsigned int *plane_id)
 {
 	int i, j;
+	bool is_plane_duplicate = dml2->v20.scratch.plane_duplicate_exists;
 
 	if (!plane_id)
 		return false;
@@ -941,7 +943,8 @@ static bool get_plane_id(const struct dc_state *context, const struct dc_plane_s
 	for (i = 0; i < context->stream_count; i++) {
 		if (context->streams[i]->stream_id == stream_id) {
 			for (j = 0; j < context->stream_status[i].plane_count; j++) {
-				if (context->stream_status[i].plane_states[j] == plane) {
+				if (context->stream_status[i].plane_states[j] == plane &&
+					(!is_plane_duplicate || (is_plane_duplicate && (j == plane_index)))) {
 					*plane_id = (i << 16) | j;
 					return true;
 				}
@@ -953,13 +956,13 @@ static bool get_plane_id(const struct dc_state *context, const struct dc_plane_s
 }
 
 static unsigned int map_plane_to_dml_display_cfg(const struct dml2_context *dml2, const struct dc_plane_state *plane,
-		const struct dc_state *context, const struct dml_display_cfg_st *dml_dispcfg, unsigned int stream_id)
+		const struct dc_state *context, const struct dml_display_cfg_st *dml_dispcfg, unsigned int stream_id, int plane_index)
 {
 	unsigned int plane_id;
 	int i = 0;
 	int location = -1;
 
-	if (!get_plane_id(context, plane, stream_id, &plane_id)) {
+	if (!get_plane_id(context->bw_ctx.dml2, context, plane, stream_id, plane_index, &plane_id)) {
 		ASSERT(false);
 		return -1;
 	}
@@ -990,7 +993,41 @@ static void apply_legacy_svp_drr_settings(struct dml2_context *dml2, const struc
 	}
 }
 
-void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, const struct dc_state *context, struct dml_display_cfg_st *dml_dispcfg)
+static void dml2_populate_pipe_to_plane_index_mapping(struct dml2_context *dml2, struct dc_state *state)
+{
+	unsigned int i;
+	unsigned int pipe_index = 0;
+	unsigned int plane_index = 0;
+	struct dml2_dml_to_dc_pipe_mapping *dml_to_dc_pipe_mapping = &dml2->v20.scratch.dml_to_dc_pipe_mapping;
+
+	for (i = 0; i < __DML2_WRAPPER_MAX_STREAMS_PLANES__; i++) {
+		dml_to_dc_pipe_mapping->dml_pipe_idx_to_plane_index_valid[i] = false;
+		dml_to_dc_pipe_mapping->dml_pipe_idx_to_plane_index[i] = 0;
+	}
+
+	for (i = 0; i < __DML2_WRAPPER_MAX_STREAMS_PLANES__; i++) {
+		struct pipe_ctx *pipe = &state->res_ctx.pipe_ctx[i];
+
+		if (!pipe || !pipe->stream || !pipe->plane_state)
+			continue;
+
+		while (pipe) {
+			pipe_index = pipe->pipe_idx;
+
+			if (pipe->stream && dml_to_dc_pipe_mapping->dml_pipe_idx_to_plane_index_valid[pipe_index] == false) {
+				dml_to_dc_pipe_mapping->dml_pipe_idx_to_plane_index[pipe_index] = plane_index;
+				plane_index++;
+				dml_to_dc_pipe_mapping->dml_pipe_idx_to_plane_index_valid[pipe_index] = true;
+			}
+
+			pipe = pipe->bottom_pipe;
+		}
+
+		plane_index = 0;
+	}
+}
+
+void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_state *context, struct dml_display_cfg_st *dml_dispcfg)
 {
 	int i = 0, j = 0;
 	int disp_cfg_stream_location, disp_cfg_plane_location;
@@ -1007,6 +1044,8 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, const struct d
 	dml_dispcfg->plane.GPUVMMaxPageTableLevels = 4;
 	dml_dispcfg->plane.HostVMEnable = false;
 
+	dml2_populate_pipe_to_plane_index_mapping(dml2, context);
+
 	for (i = 0; i < context->stream_count; i++) {
 		disp_cfg_stream_location = map_stream_to_dml_display_cfg(dml2, context->streams[i], dml_dispcfg);
 
@@ -1043,7 +1082,7 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, const struct d
 		} else {
 			for (j = 0; j < context->stream_status[i].plane_count; j++) {
 				disp_cfg_plane_location = map_plane_to_dml_display_cfg(dml2,
-					context->stream_status[i].plane_states[j], context, dml_dispcfg, context->streams[i]->stream_id);
+					context->stream_status[i].plane_states[j], context, dml_dispcfg, context->streams[i]->stream_id, j);
 
 				if (disp_cfg_plane_location < 0)
 					disp_cfg_plane_location = dml_dispcfg->num_surfaces++;
@@ -1067,7 +1106,7 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, const struct d
 
 				dml_dispcfg->plane.BlendingAndTiming[disp_cfg_plane_location] = disp_cfg_stream_location;
 
-				if (get_plane_id(context, context->stream_status[i].plane_states[j], context->streams[i]->stream_id,
+				if (get_plane_id(dml2, context, context->stream_status[i].plane_states[j], context->streams[i]->stream_id, j,
 					&dml2->v20.scratch.dml_to_dc_pipe_mapping.disp_cfg_to_plane_id[disp_cfg_plane_location]))
 					dml2->v20.scratch.dml_to_dc_pipe_mapping.disp_cfg_to_plane_id_valid[disp_cfg_plane_location] = true;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.h
index dac6d27b14cd..d764773938f4 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.h
@@ -34,7 +34,7 @@ void dml2_init_soc_states(struct dml2_context *dml2, const struct dc *in_dc,
 void dml2_translate_ip_params(const struct dc *in_dc, struct ip_params_st *out);
 void dml2_translate_socbb_params(const struct dc *in_dc, struct soc_bounding_box_st *out);
 void dml2_translate_soc_states(const struct dc *in_dc, struct soc_states_st *out, int num_states);
-void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, const struct dc_state *context, struct dml_display_cfg_st *dml_dispcfg);
+void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_state *context, struct dml_display_cfg_st *dml_dispcfg);
 void dml2_update_pipe_ctx_dchub_regs(struct _vcs_dpi_dml_display_rq_regs_st *rq_regs, struct _vcs_dpi_dml_display_dlg_regs_st *disp_dlg_regs, struct _vcs_dpi_dml_display_ttu_regs_st *disp_ttu_regs, struct pipe_ctx *out);
 bool is_dp2p0_output_encoder(const struct pipe_ctx *pipe);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c
index 69fd96f4f3b0..2498b8341199 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c
@@ -209,10 +209,11 @@ static int find_dml_pipe_idx_by_plane_id(struct dml2_context *ctx, unsigned int
 	return -1;
 }
 
-static bool get_plane_id(const struct dc_state *state, const struct dc_plane_state *plane,
-	unsigned int stream_id, unsigned int *plane_id)
+static bool get_plane_id(struct dml2_context *dml2, const struct dc_state *state, const struct dc_plane_state *plane,
+	unsigned int stream_id, unsigned int plane_index, unsigned int *plane_id)
 {
 	int i, j;
+	bool is_plane_duplicate = dml2->v20.scratch.plane_duplicate_exists;
 
 	if (!plane_id)
 		return false;
@@ -220,7 +221,8 @@ static bool get_plane_id(const struct dc_state *state, const struct dc_plane_sta
 	for (i = 0; i < state->stream_count; i++) {
 		if (state->streams[i]->stream_id == stream_id) {
 			for (j = 0; j < state->stream_status[i].plane_count; j++) {
-				if (state->stream_status[i].plane_states[j] == plane) {
+				if (state->stream_status[i].plane_states[j] == plane &&
+					(!is_plane_duplicate || (is_plane_duplicate && (j == plane_index)))) {
 					*plane_id = (i << 16) | j;
 					return true;
 				}
@@ -304,8 +306,9 @@ void dml2_calculate_rq_and_dlg_params(const struct dc *dc, struct dc_state *cont
 		 * there is a need to know which DML pipe index maps to which DC pipe. The code below
 		 * finds a dml_pipe_index from the plane id if a plane is valid. If a plane is not valid then
 		 * it finds a dml_pipe_index from the stream id. */
-		if (get_plane_id(context, context->res_ctx.pipe_ctx[dc_pipe_ctx_index].plane_state,
-			context->res_ctx.pipe_ctx[dc_pipe_ctx_index].stream->stream_id, &plane_id)) {
+		if (get_plane_id(in_ctx, context, context->res_ctx.pipe_ctx[dc_pipe_ctx_index].plane_state,
+			context->res_ctx.pipe_ctx[dc_pipe_ctx_index].stream->stream_id,
+			in_ctx->v20.scratch.dml_to_dc_pipe_mapping.dml_pipe_idx_to_plane_index[context->res_ctx.pipe_ctx[dc_pipe_ctx_index].pipe_idx], &plane_id)) {
 			dml_pipe_idx = find_dml_pipe_idx_by_plane_id(in_ctx, plane_id);
 		} else {
 			dml_pipe_idx = dml2_helper_find_dml_pipe_idx_by_stream_id(in_ctx, context->res_ctx.pipe_ctx[dc_pipe_ctx_index].stream->stream_id);
@@ -445,8 +448,9 @@ bool dml2_verify_det_buffer_configuration(struct dml2_context *in_ctx, struct dc
 	for (i = 0; i < MAX_PIPES; i++) {
 		if (!display_state->res_ctx.pipe_ctx[i].stream)
 			continue;
-		if (get_plane_id(display_state, display_state->res_ctx.pipe_ctx[i].plane_state,
-			display_state->res_ctx.pipe_ctx[i].stream->stream_id, &plane_id))
+		if (get_plane_id(in_ctx, display_state, display_state->res_ctx.pipe_ctx[i].plane_state,
+			display_state->res_ctx.pipe_ctx[i].stream->stream_id,
+			in_ctx->v20.scratch.dml_to_dc_pipe_mapping.dml_pipe_idx_to_plane_index[display_state->res_ctx.pipe_ctx[i].pipe_idx], &plane_id))
 			dml_pipe_idx = find_dml_pipe_idx_by_plane_id(in_ctx, plane_id);
 		else
 			dml_pipe_idx = dml2_helper_find_dml_pipe_idx_by_stream_id(in_ctx, display_state->res_ctx.pipe_ctx[i].stream->stream_id);
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
index 0a06bf3b135a..8f231418870f 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
@@ -639,7 +639,7 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s
 	return result;
 }
 
-static bool dml2_validate_only(const struct dc_state *context)
+static bool dml2_validate_only(struct dc_state *context)
 {
 	struct dml2_context *dml2 = context->bw_ctx.dml2;
 	unsigned int result = 0;
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
index 3966845c7694..e8b5f17beb96 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
@@ -512,6 +512,11 @@ static bool intersect_dsc_caps(
 		dsc_sink_caps->slice_caps1.bits.NUM_SLICES_4 && dsc_enc_caps->slice_caps.bits.NUM_SLICES_4;
 	dsc_common_caps->slice_caps.bits.NUM_SLICES_8 =
 		dsc_sink_caps->slice_caps1.bits.NUM_SLICES_8 && dsc_enc_caps->slice_caps.bits.NUM_SLICES_8;
+	dsc_common_caps->slice_caps.bits.NUM_SLICES_12 =
+		dsc_sink_caps->slice_caps1.bits.NUM_SLICES_12 && dsc_enc_caps->slice_caps.bits.NUM_SLICES_12;
+	dsc_common_caps->slice_caps.bits.NUM_SLICES_16 =
+		dsc_sink_caps->slice_caps2.bits.NUM_SLICES_16 && dsc_enc_caps->slice_caps.bits.NUM_SLICES_16;
+
 	if (!dsc_common_caps->slice_caps.raw)
 		return false;
 
@@ -703,6 +708,12 @@ static int get_available_dsc_slices(union dsc_enc_slice_caps slice_caps, int *av
 	if (slice_caps.bits.NUM_SLICES_8)
 		available_slices[idx++] = 8;
 
+	if (slice_caps.bits.NUM_SLICES_12)
+		available_slices[idx++] = 12;
+
+	if (slice_caps.bits.NUM_SLICES_16)
+		available_slices[idx++] = 16;
+
 	return idx;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce/dce_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dce/dce_hwseq.h
index 2fefdf40612d..44b4df6469d1 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce/dce_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce/dce_hwseq.h
@@ -1183,7 +1183,23 @@ struct dce_hwseq_registers {
 	type LONO_FGCG_REP_DIS;\
 	type LONO_DISPCLK_GATE_DISABLE;\
 	type LONO_SOCCLK_GATE_DISABLE;\
-	type LONO_DMCUBCLK_GATE_DISABLE;
+	type LONO_DMCUBCLK_GATE_DISABLE;\
+	type SYMCLKA_FE_GATE_DISABLE;\
+	type SYMCLKB_FE_GATE_DISABLE;\
+	type SYMCLKC_FE_GATE_DISABLE;\
+	type SYMCLKD_FE_GATE_DISABLE;\
+	type SYMCLKE_FE_GATE_DISABLE;\
+	type HDMICHARCLK0_GATE_DISABLE;\
+	type SYMCLKA_GATE_DISABLE;\
+	type SYMCLKB_GATE_DISABLE;\
+	type SYMCLKC_GATE_DISABLE;\
+	type SYMCLKD_GATE_DISABLE;\
+	type SYMCLKE_GATE_DISABLE;\
+	type PHYASYMCLK_ROOT_GATE_DISABLE;\
+	type PHYBSYMCLK_ROOT_GATE_DISABLE;\
+	type PHYCSYMCLK_ROOT_GATE_DISABLE;\
+	type PHYDSYMCLK_ROOT_GATE_DISABLE;\
+	type PHYESYMCLK_ROOT_GATE_DISABLE;
 
 struct dce_hwseq_shift {
 	HWSEQ_REG_FIELD_LIST(uint8_t)
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
index 1b9f21fd4f17..6a65af8c36b9 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
@@ -615,12 +615,6 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context)
 		    pipe->stream->fpo_in_use)) {
 			if (hubp && hubp->funcs->hubp_update_force_pstate_disallow)
 				hubp->funcs->hubp_update_force_pstate_disallow(hubp, false);
-		}
-
-		/* Today only FPO uses cursor P-State force. Only clear cursor P-State force
-		 * if it's not FPO.
-		 */
-		if (!pipe->stream || !pipe->stream->fpo_in_use) {
 			if (hubp && hubp->funcs->hubp_update_force_cursor_pstate_disallow)
 				hubp->funcs->hubp_update_force_cursor_pstate_disallow(hubp, false);
 		}
@@ -632,17 +626,10 @@ void dcn32_update_force_pstate(struct dc *dc, struct dc_state *context)
 		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
 		struct hubp *hubp = pipe->plane_res.hubp;
 
-		if (pipe->stream && pipe->plane_state && pipe->stream->mall_stream_config.type == SUBVP_MAIN) {
+		if (pipe->stream && (pipe->stream->mall_stream_config.type == SUBVP_MAIN ||
+				pipe->stream->fpo_in_use)) {
 			if (hubp && hubp->funcs->hubp_update_force_pstate_disallow)
 				hubp->funcs->hubp_update_force_pstate_disallow(hubp, true);
-		}
-
-		if (pipe->stream && pipe->stream->fpo_in_use) {
-			if (hubp && hubp->funcs->hubp_update_force_pstate_disallow)
-				hubp->funcs->hubp_update_force_pstate_disallow(hubp, true);
-			/* For now only force cursor p-state disallow for FPO
-			 * Needs to be added for subvp once FW side gets updated
-			 */
 			if (hubp && hubp->funcs->hubp_update_force_cursor_pstate_disallow)
 				hubp->funcs->hubp_update_force_cursor_pstate_disallow(hubp, true);
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
index 34737d60b965..5a8258287438 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
@@ -138,16 +138,25 @@ void dcn35_init_hw(struct dc *dc)
 	if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks)
 		dc->clk_mgr->funcs->init_clocks(dc->clk_mgr);
 
-	REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0);
-	REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0x3F000000);
-	REG_WRITE(DCCG_GATE_DISABLE_CNTL5, 0x1f7c3fcf);
-
 	//dcn35_set_dmu_fgcg(hws, dc->debug.enable_fine_grain_clock_gating.bits.dmu);
 
 	if (!dcb->funcs->is_accelerated_mode(dcb)) {
 		/*this calls into dmubfw to do the init*/
 		hws->funcs.bios_golden_init(dc);
 	}
+
+	REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0);
+	REG_WRITE(DCCG_GATE_DISABLE_CNTL2,  0);
+
+	/* Disable gating for PHYASYMCLK. This will be enabled in dccg if needed */
+	REG_UPDATE_5(DCCG_GATE_DISABLE_CNTL2, PHYASYMCLK_ROOT_GATE_DISABLE, 1,
+			PHYBSYMCLK_ROOT_GATE_DISABLE, 1,
+			PHYCSYMCLK_ROOT_GATE_DISABLE, 1,
+			PHYDSYMCLK_ROOT_GATE_DISABLE, 1,
+			PHYESYMCLK_ROOT_GATE_DISABLE, 1);
+
+	REG_WRITE(DCCG_GATE_DISABLE_CNTL5, 0x1f7c3fcf);
+
 	// Initialize the dccg
 	if (res_pool->dccg->funcs->dccg_init)
 		res_pool->dccg->funcs->dccg_init(res_pool->dccg);
@@ -274,7 +283,19 @@ void dcn35_init_hw(struct dc *dc)
 	if (!dc->debug.disable_clock_gate) {
 		/* enable all DCN clock gating */
 		REG_WRITE(DCCG_GATE_DISABLE_CNTL, 0);
-		REG_WRITE(DCCG_GATE_DISABLE_CNTL2, 0);
+
+		REG_UPDATE_5(DCCG_GATE_DISABLE_CNTL2, SYMCLKA_FE_GATE_DISABLE, 0,
+				SYMCLKB_FE_GATE_DISABLE, 0,
+				SYMCLKC_FE_GATE_DISABLE, 0,
+				SYMCLKD_FE_GATE_DISABLE, 0,
+				SYMCLKE_FE_GATE_DISABLE, 0);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, HDMICHARCLK0_GATE_DISABLE, 0);
+		REG_UPDATE_5(DCCG_GATE_DISABLE_CNTL2, SYMCLKA_GATE_DISABLE, 0,
+				SYMCLKB_GATE_DISABLE, 0,
+				SYMCLKC_GATE_DISABLE, 0,
+				SYMCLKD_GATE_DISABLE, 0,
+				SYMCLKE_GATE_DISABLE, 0);
+
 		REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0);
 	}
 
@@ -311,6 +332,9 @@ void dcn35_init_hw(struct dc *dc)
 	if (dc->res_pool->pg_cntl) {
 		if (dc->res_pool->pg_cntl->funcs->init_pg_status)
 			dc->res_pool->pg_cntl->funcs->init_pg_status(dc->res_pool->pg_cntl);
+
+		if (dc->res_pool->pg_cntl->funcs->set_force_poweron_domain22)
+			dc->res_pool->pg_cntl->funcs->set_force_poweron_domain22(dc->res_pool->pg_cntl, false);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
index 13f12f2a3f81..ce2f0c0e82bd 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
@@ -141,6 +141,11 @@ struct dccg_funcs {
 			enum physymclk_clock_source clk_src,
 			bool force_enable);
 
+	void (*set_physymclk_root_clock_gating)(
+			struct dccg *dccg,
+			int phy_inst,
+			bool enable);
+
 	void (*set_dtbclk_dto)(
 			struct dccg *dccg,
 			const struct dtbclk_dto_params *params);
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h
index d7b8d586b523..4b27f29d0d80 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dsc.h
@@ -76,6 +76,8 @@ union dsc_enc_slice_caps {
 		uint8_t NUM_SLICES_3 : 1; /* This one is not per DSC spec, but our encoder supports it */
 		uint8_t NUM_SLICES_4 : 1;
 		uint8_t NUM_SLICES_8 : 1;
+		uint8_t NUM_SLICES_12 : 1;
+		uint8_t NUM_SLICES_16 : 1;
 	} bits;
 	uint8_t raw;
 };
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/optc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/optc.h
new file mode 100644
index 000000000000..9a8bf6ec70ea
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/optc.h
@@ -0,0 +1,219 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2023 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+/**
+ * DOC: overview
+ *
+ * Output Pipe Timing Combiner (OPTC) includes two major functional blocks:
+ * Output Data Mapper (ODM) and Output Timing Generator (OTG).
+ *
+ * - ODM: It is Output Data Mapping block. It can combine input data from
+ *   multiple OPP data pipes into one single data stream or split data from one
+ *   OPP data pipe into multiple data streams or just bypass OPP data to DIO.
+ * - OTG: It is Output Timing Generator. It generates display timing signals to
+ *   drive the display output.
+ */
+
+#ifndef __DC_OPTC_H__
+#define __DC_OPTC_H__
+
+#include "timing_generator.h"
+
+struct optc {
+	struct timing_generator base;
+
+	const struct dcn_optc_registers *tg_regs;
+	const struct dcn_optc_shift *tg_shift;
+	const struct dcn_optc_mask *tg_mask;
+
+	int opp_count;
+
+	uint32_t max_h_total;
+	uint32_t max_v_total;
+
+	uint32_t min_h_blank;
+
+	uint32_t min_h_sync_width;
+	uint32_t min_v_sync_width;
+	uint32_t min_v_blank;
+	uint32_t min_v_blank_interlace;
+
+	int vstartup_start;
+	int vupdate_offset;
+	int vupdate_width;
+	int vready_offset;
+	struct dc_crtc_timing orginal_patched_timing;
+	enum signal_type signal;
+};
+
+struct dcn_otg_state {
+	uint32_t v_blank_start;
+	uint32_t v_blank_end;
+	uint32_t v_sync_a_pol;
+	uint32_t v_total;
+	uint32_t v_total_max;
+	uint32_t v_total_min;
+	uint32_t v_total_min_sel;
+	uint32_t v_total_max_sel;
+	uint32_t v_sync_a_start;
+	uint32_t v_sync_a_end;
+	uint32_t h_blank_start;
+	uint32_t h_blank_end;
+	uint32_t h_sync_a_start;
+	uint32_t h_sync_a_end;
+	uint32_t h_sync_a_pol;
+	uint32_t h_total;
+	uint32_t underflow_occurred_status;
+	uint32_t otg_enabled;
+	uint32_t blank_enabled;
+	uint32_t vertical_interrupt1_en;
+	uint32_t vertical_interrupt1_line;
+	uint32_t vertical_interrupt2_en;
+	uint32_t vertical_interrupt2_line;
+};
+
+void optc1_read_otg_state(struct optc *optc1, struct dcn_otg_state *s);
+
+bool optc1_get_hw_timing(struct timing_generator *tg, struct dc_crtc_timing *hw_crtc_timing);
+
+bool optc1_validate_timing(struct timing_generator *optc,
+			   const struct dc_crtc_timing *timing);
+
+void optc1_program_timing(struct timing_generator *optc,
+			  const struct dc_crtc_timing *dc_crtc_timing,
+			  int vready_offset,
+			  int vstartup_start,
+			  int vupdate_offset,
+			  int vupdate_width,
+			  const enum signal_type signal,
+			  bool use_vbios);
+
+void optc1_setup_vertical_interrupt0(struct timing_generator *optc,
+				     uint32_t start_line,
+				     uint32_t end_line);
+
+void optc1_setup_vertical_interrupt1(struct timing_generator *optc,
+				     uint32_t start_line);
+
+void optc1_setup_vertical_interrupt2(struct timing_generator *optc,
+				     uint32_t start_line);
+
+void optc1_program_global_sync(struct timing_generator *optc,
+			       int vready_offset,
+			       int vstartup_start,
+			       int vupdate_offset,
+			       int vupdate_width);
+
+bool optc1_disable_crtc(struct timing_generator *optc);
+
+bool optc1_is_counter_moving(struct timing_generator *optc);
+
+void optc1_get_position(struct timing_generator *optc,
+			struct crtc_position *position);
+
+uint32_t optc1_get_vblank_counter(struct timing_generator *optc);
+
+void optc1_get_crtc_scanoutpos(struct timing_generator *optc,
+			       uint32_t *v_blank_start,
+			       uint32_t *v_blank_end,
+			       uint32_t *h_position,
+			       uint32_t *v_position);
+
+void optc1_set_early_control(struct timing_generator *optc,
+			     uint32_t early_cntl);
+
+void optc1_wait_for_state(struct timing_generator *optc,
+			  enum crtc_state state);
+
+void optc1_set_blank(struct timing_generator *optc,
+		     bool enable_blanking);
+
+bool optc1_is_blanked(struct timing_generator *optc);
+
+void optc1_program_blank_color(struct timing_generator *optc,
+			       const struct tg_color *black_color);
+
+bool optc1_did_triggered_reset_occur(struct timing_generator *optc);
+
+void optc1_enable_reset_trigger(struct timing_generator *optc, int source_tg_inst);
+
+void optc1_disable_reset_trigger(struct timing_generator *optc);
+
+void optc1_lock(struct timing_generator *optc);
+
+void optc1_unlock(struct timing_generator *optc);
+
+void optc1_enable_optc_clock(struct timing_generator *optc, bool enable);
+
+void optc1_set_drr(struct timing_generator *optc,
+		   const struct drr_params *params);
+
+void optc1_set_vtotal_min_max(struct timing_generator *optc, int vtotal_min, int vtotal_max);
+
+void optc1_set_static_screen_control(struct timing_generator *optc,
+				     uint32_t event_triggers,
+				     uint32_t num_frames);
+
+void optc1_program_stereo(struct timing_generator *optc,
+			  const struct dc_crtc_timing *timing,
+			  struct crtc_stereo_flags *flags);
+
+bool optc1_is_stereo_left_eye(struct timing_generator *optc);
+
+void optc1_clear_optc_underflow(struct timing_generator *optc);
+
+void optc1_tg_init(struct timing_generator *optc);
+
+bool optc1_is_tg_enabled(struct timing_generator *optc);
+
+bool optc1_is_optc_underflow_occurred(struct timing_generator *optc);
+
+void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enable);
+
+void optc1_set_timing_double_buffer(struct timing_generator *optc, bool enable);
+
+bool optc1_get_otg_active_size(struct timing_generator *optc,
+			       uint32_t *otg_active_width,
+			       uint32_t *otg_active_height);
+
+void optc1_enable_crtc_reset(struct timing_generator *optc,
+			     int source_tg_inst,
+			     struct crtc_trigger_info *crtc_tp);
+
+bool optc1_configure_crc(struct timing_generator *optc, const struct crc_params *params);
+
+bool optc1_get_crc(struct timing_generator *optc,
+		   uint32_t *r_cr,
+		   uint32_t *g_y,
+		   uint32_t *b_cb);
+
+bool optc1_is_two_pixels_per_containter(const struct dc_crtc_timing *timing);
+
+void optc1_set_vtg_params(struct timing_generator *optc,
+			  const struct dc_crtc_timing *dc_crtc_timing,
+			  bool program_fp2);
+
+#endif
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/pg_cntl.h b/drivers/gpu/drm/amd/display/dc/inc/hw/pg_cntl.h
index 00ea3864dd4d..b9812afb886b 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/pg_cntl.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/pg_cntl.h
@@ -47,6 +47,8 @@ struct pg_cntl_funcs {
 	void (*optc_pg_control)(struct pg_cntl *pg_cntl, unsigned int optc_inst, bool power_on);
 	void (*dwb_pg_control)(struct pg_cntl *pg_cntl, bool power_on);
 	void (*init_pg_status)(struct pg_cntl *pg_cntl);
+
+	void (*set_force_poweron_domain22)(struct pg_cntl *pg_cntl, bool power_on);
 };
 
 #endif //__DC_PG_CNTL_H__
diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
index 21a39afd274b..2d152b68a501 100644
--- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
+++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
@@ -53,6 +53,7 @@ static enum dc_link_rate get_link_rate_from_test_link_rate(uint8_t test_rate)
 		return LINK_RATE_UHBR10;
 	case DP_TEST_LINK_RATE_UHBR20:
 		return LINK_RATE_UHBR20;
+	case DP_TEST_LINK_RATE_UHBR13_5_LEGACY:
 	case DP_TEST_LINK_RATE_UHBR13_5:
 		return LINK_RATE_UHBR13_5;
 	default:
@@ -119,6 +120,11 @@ static void dp_test_send_link_training(struct dc_link *link)
 			1);
 	link_settings.link_rate = get_link_rate_from_test_link_rate(test_rate);
 
+	if (link_settings.link_rate == LINK_RATE_UNKNOWN) {
+		DC_LOG_ERROR("%s: Invalid test link rate.", __func__);
+		ASSERT(0);
+	}
+
 	/* Set preferred link settings */
 	link->verified_link_cap.lane_count = link_settings.lane_count;
 	link->verified_link_cap.link_rate = link_settings.link_rate;
@@ -457,7 +463,7 @@ static void set_crtc_test_pattern(struct dc_link *link,
 			controller_color_space = pipe_ctx->stream_res.test_pattern_params.color_space;
 
 			if (controller_color_space == CONTROLLER_DP_COLOR_SPACE_UDEFINED) {
-				DC_LOG_WARNING("%s: Color space must be defined for test pattern", __func__);
+				DC_LOG_ERROR("%s: Color space must be defined for test pattern", __func__);
 				ASSERT(0);
 			}
 
@@ -592,6 +598,7 @@ bool dp_set_test_pattern(
 	const unsigned char *p_custom_pattern,
 	unsigned int cust_pattern_size)
 {
+	const struct link_hwss *link_hwss;
 	struct pipe_ctx *pipes = link->dc->current_state->res_ctx.pipe_ctx;
 	struct pipe_ctx *pipe_ctx = NULL;
 	unsigned int lane;
@@ -828,11 +835,9 @@ bool dp_set_test_pattern(
 
 		pipe_ctx->stream_res.tg->funcs->lock(pipe_ctx->stream_res.tg);
 		/* update MSA to requested color space */
-		pipe_ctx->stream_res.stream_enc->funcs->dp_set_stream_attribute(pipe_ctx->stream_res.stream_enc,
-				&pipe_ctx->stream->timing,
-				color_space,
-				pipe_ctx->stream->use_vsc_sdp_for_colorimetry,
-				link->dpcd_caps.dprx_feature.bits.SST_SPLIT_SDP_CAP);
+		link_hwss = get_link_hwss(link, &pipe_ctx->link_res);
+		pipe_ctx->stream->output_color_space = color_space;
+		link_hwss->setup_stream_attribute(pipe_ctx);
 
 		if (pipe_ctx->stream->use_vsc_sdp_for_colorimetry) {
 			if (test_pattern == DP_TEST_PATTERN_COLOR_SQUARES_CEA)
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
index 34bf8a9ef738..0c00e94e90b1 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c
@@ -184,14 +184,17 @@ static bool handle_hpd_irq_psr_sink(struct dc_link *link)
 	return false;
 }
 
-static bool handle_hpd_irq_replay_sink(struct dc_link *link)
+static void handle_hpd_irq_replay_sink(struct dc_link *link)
 {
 	union dpcd_replay_configuration replay_configuration;
 	/*AMD Replay version reuse DP_PSR_ERROR_STATUS for REPLAY_ERROR status.*/
 	union psr_error_status replay_error_status;
 
+	if (link->replay_settings.config.force_disable_desync_error_check)
+		return;
+
 	if (!link->replay_settings.replay_feature_enabled)
-		return false;
+		return;
 
 	dm_helpers_dp_read_dpcd(
 		link->ctx,
@@ -207,6 +210,9 @@ static bool handle_hpd_irq_replay_sink(struct dc_link *link)
 		&replay_error_status.raw,
 		sizeof(replay_error_status.raw));
 
+	if (replay_configuration.bits.DESYNC_ERROR_STATUS)
+		link->replay_settings.config.received_desync_error_hpd = 1;
+
 	link->replay_settings.config.replay_error_status.bits.LINK_CRC_ERROR =
 		replay_error_status.bits.LINK_CRC_ERROR;
 	link->replay_settings.config.replay_error_status.bits.DESYNC_ERROR =
@@ -243,7 +249,6 @@ static bool handle_hpd_irq_replay_sink(struct dc_link *link)
 			edp_set_replay_allow_active(link, &allow_active, true, false, NULL);
 		}
 	}
-	return true;
 }
 
 void dp_handle_link_loss(struct dc_link *link)
@@ -424,9 +429,7 @@ bool dp_handle_hpd_rx_irq(struct dc_link *link,
 		/* PSR-related error was detected and handled */
 		return true;
 
-	if (handle_hpd_irq_replay_sink(link))
-		/* Replay-related error was detected and handled */
-		return true;
+	handle_hpd_irq_replay_sink(link);
 
 	/* If PSR-related error handled, Main link may be off,
 	 * so do not handle as a normal sink status change interrupt.
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index bc907ae2052d..ed4379c04715 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -583,6 +583,7 @@ union dmub_fw_boot_status {
 		uint32_t fams_enabled : 1; /**< 1 if VBIOS data is deferred programmed */
 		uint32_t detection_required: 1; /**<  if detection need to be triggered by driver */
 		uint32_t hw_power_init_done: 1; /**< 1 if hw power init is completed */
+		uint32_t ono_regions_enabled: 1; /**< 1 if ONO regions are enabled */
 	} bits; /**< status bits */
 	uint32_t all; /**< 32-bit access to status bits */
 };
@@ -599,6 +600,7 @@ enum dmub_fw_boot_status_bit {
 	DMUB_FW_BOOT_STATUS_BIT_FAMS_ENABLED = (1 << 5), /**< 1 if FAMS is enabled*/
 	DMUB_FW_BOOT_STATUS_BIT_DETECTION_REQUIRED = (1 << 6), /**< 1 if detection need to be triggered by driver*/
 	DMUB_FW_BOOT_STATUS_BIT_HW_POWER_INIT_DONE = (1 << 7), /**< 1 if hw power init is completed */
+	DMUB_FW_BOOT_STATUS_BIT_ONO_REGIONS_ENABLED = (1 << 8), /**< 1 if ONO regions are enabled */
 };
 
 /* Register bit definition for SCRATCH5 */
@@ -617,9 +619,12 @@ enum dmub_lvtma_status_bit {
 };
 
 enum dmub_ips_disable_type {
-	DMUB_IPS_DISABLE_IPS1 = 1,
-	DMUB_IPS_DISABLE_IPS2 = 2,
-	DMUB_IPS_DISABLE_IPS2_Z10 = 3,
+	DMUB_IPS_ENABLE = 0,
+	DMUB_IPS_DISABLE_ALL = 1,
+	DMUB_IPS_DISABLE_IPS1 = 2,
+	DMUB_IPS_DISABLE_IPS2 = 3,
+	DMUB_IPS_DISABLE_IPS2_Z10 = 4,
+	DMUB_IPS_DISABLE_DYNAMIC = 5,
 };
 
 #define DMUB_IPS1_ALLOW_MASK 0x00000001
@@ -653,8 +658,8 @@ union dmub_fw_boot_options {
 		uint32_t disable_clk_ds: 1; /* 1 if disallow dispclk_ds and dppclk_ds*/
 		uint32_t disable_timeout_recovery : 1; /* 1 if timeout recovery should be disabled */
 		uint32_t ips_pg_disable: 1; /* 1 to disable ONO domains power gating*/
-		uint32_t ips_disable: 2; /* options to disable ips support*/
-		uint32_t reserved : 10; /**< reserved */
+		uint32_t ips_disable: 3; /* options to disable ips support*/
+		uint32_t reserved : 9; /**< reserved */
 	} bits; /**< boot bits */
 	uint32_t all; /**< 32-bit access to bits */
 };
@@ -2098,7 +2103,7 @@ enum psr_version {
 	/**
 	 * PSR not supported.
 	 */
-	PSR_VERSION_UNSUPPORTED			= 0xFFFFFFFF,
+	PSR_VERSION_UNSUPPORTED			= 0xFF,	// psr_version field is only 8 bits wide
 };
 
 /**
@@ -3620,7 +3625,6 @@ struct dmub_cmd_abm_pause_data {
 	uint8_t pad[1];
 };
 
-
 /**
  * Definition of a DMUB_CMD__ABM_PAUSE command.
  */
@@ -4046,6 +4050,7 @@ union dmub_rb_cmd {
 	 * Definition of a DMUB_CMD__MALL command.
 	 */
 	struct dmub_rb_cmd_mall mall;
+
 	/**
 	 * Definition of a DMUB_CMD__CAB command.
 	 */
@@ -4067,6 +4072,7 @@ union dmub_rb_cmd {
 	 * Definition of DMUB_CMD__PANEL_CNTL commands.
 	 */
 	struct dmub_rb_cmd_panel_cntl panel_cntl;
+
 	/**
 	 * Definition of a DMUB_CMD__ABM_SET_PIPE command.
 	 */
@@ -4470,10 +4476,6 @@ static inline void dmub_rb_flush_pending(const struct dmub_rb *rb)
 		uint64_t *data = (uint64_t *)((uint8_t *)(rb->base_address) + rptr);
 		uint8_t i;
 
-		/* Don't remove this.
-		 * The contents need to actually be read from the ring buffer
-		 * for this function to be effective.
-		 */
 		for (i = 0; i < DMUB_RB_CMD_SIZE / sizeof(uint64_t); i++)
 			(void)READ_ONCE(*data++);
 
@@ -4522,5 +4524,4 @@ static inline void dmub_rb_get_return_data(struct dmub_rb *rb,
 //==============================================================================
 //</DMUB_RB>====================================================================
 //==============================================================================
-
 #endif /* _DMUB_CMD_H_ */
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
index acf3527fff2d..08cb79401410 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
@@ -491,7 +491,7 @@ int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors senso
 int amdgpu_dpm_get_apu_thermal_limit(struct amdgpu_device *adev, uint32_t *limit)
 {
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
-	int ret = -EINVAL;
+	int ret = -EOPNOTSUPP;
 
 	if (pp_funcs && pp_funcs->get_apu_thermal_limit) {
 		mutex_lock(&adev->pm.mutex);
@@ -505,7 +505,7 @@ int amdgpu_dpm_get_apu_thermal_limit(struct amdgpu_device *adev, uint32_t *limit
 int amdgpu_dpm_set_apu_thermal_limit(struct amdgpu_device *adev, uint32_t limit)
 {
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
-	int ret = -EINVAL;
+	int ret = -EOPNOTSUPP;
 
 	if (pp_funcs && pp_funcs->set_apu_thermal_limit) {
 		mutex_lock(&adev->pm.mutex);
@@ -1182,7 +1182,7 @@ int amdgpu_dpm_get_sclk_od(struct amdgpu_device *adev)
 	int ret = 0;
 
 	if (!pp_funcs->get_sclk_od)
-		return 0;
+		return -EOPNOTSUPP;
 
 	mutex_lock(&adev->pm.mutex);
 	ret = pp_funcs->get_sclk_od(adev->powerplay.pp_handle);
@@ -1196,7 +1196,7 @@ int amdgpu_dpm_set_sclk_od(struct amdgpu_device *adev, uint32_t value)
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
 
 	if (is_support_sw_smu(adev))
-		return 0;
+		return -EOPNOTSUPP;
 
 	mutex_lock(&adev->pm.mutex);
 	if (pp_funcs->set_sclk_od)
@@ -1219,7 +1219,7 @@ int amdgpu_dpm_get_mclk_od(struct amdgpu_device *adev)
 	int ret = 0;
 
 	if (!pp_funcs->get_mclk_od)
-		return 0;
+		return -EOPNOTSUPP;
 
 	mutex_lock(&adev->pm.mutex);
 	ret = pp_funcs->get_mclk_od(adev->powerplay.pp_handle);
@@ -1233,7 +1233,7 @@ int amdgpu_dpm_set_mclk_od(struct amdgpu_device *adev, uint32_t value)
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
 
 	if (is_support_sw_smu(adev))
-		return 0;
+		return -EOPNOTSUPP;
 
 	mutex_lock(&adev->pm.mutex);
 	if (pp_funcs->set_mclk_od)
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 4ba9195c83c5..ca2ece24e1e0 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -989,12 +989,13 @@ static ssize_t amdgpu_get_pp_features(struct device *dev,
  * Reading back the files will show you the available power levels within
  * the power state and the clock information for those levels. If deep sleep is
  * applied to a clock, the level will be denoted by a special level 'S:'
- * E.g.,
- *	S: 19Mhz *
- *	0: 615Mhz
- *	1: 800Mhz
- *	2: 888Mhz
- *	3: 1000Mhz
+ * E.g., ::
+ *
+ *  S: 19Mhz *
+ *  0: 615Mhz
+ *  1: 800Mhz
+ *  2: 888Mhz
+ *  3: 1000Mhz
  *
  *
  * To manually adjust these states, first select manual using
@@ -2197,6 +2198,22 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_
 	} else if (DEVICE_ATTR_IS(xgmi_plpd_policy)) {
 		if (amdgpu_dpm_get_xgmi_plpd_mode(adev, NULL) == XGMI_PLPD_NONE)
 			*states = ATTR_STATE_UNSUPPORTED;
+	} else if (DEVICE_ATTR_IS(pp_dpm_mclk_od)) {
+		if (amdgpu_dpm_get_mclk_od(adev) == -EOPNOTSUPP)
+			*states = ATTR_STATE_UNSUPPORTED;
+	} else if (DEVICE_ATTR_IS(pp_dpm_sclk_od)) {
+		if (amdgpu_dpm_get_sclk_od(adev) == -EOPNOTSUPP)
+			*states = ATTR_STATE_UNSUPPORTED;
+	} else if (DEVICE_ATTR_IS(apu_thermal_cap)) {
+		u32 limit;
+
+		if (amdgpu_dpm_get_apu_thermal_limit(adev, &limit) ==
+		    -EOPNOTSUPP)
+			*states = ATTR_STATE_UNSUPPORTED;
+	} else if (DEVICE_ATTR_IS(pp_dpm_pcie)) {
+		if (gc_ver == IP_VERSION(9, 4, 2) ||
+		    gc_ver == IP_VERSION(9, 4, 3))
+			*states = ATTR_STATE_UNSUPPORTED;
 	}
 
 	switch (gc_ver) {
diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 23b00eddc1af..1ead323f1c78 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -1711,6 +1711,7 @@ static int smu_disable_dpms(struct smu_context *smu)
 	}
 
 	if (amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(9, 4, 2) &&
+	    !((adev->flags & AMD_IS_APU) && adev->gfx.imu.funcs) &&
 	    !amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs->stop)
 		adev->gfx.rlc.funcs->stop(adev);
 
@@ -2747,7 +2748,7 @@ unlock:
 
 static int smu_get_apu_thermal_limit(void *handle, uint32_t *limit)
 {
-	int ret = -EINVAL;
+	int ret = -EOPNOTSUPP;
 	struct smu_context *smu = handle;
 
 	if (smu->ppt_funcs && smu->ppt_funcs->get_apu_thermal_limit)
@@ -2758,7 +2759,7 @@ static int smu_get_apu_thermal_limit(void *handle, uint32_t *limit)
 
 static int smu_set_apu_thermal_limit(void *handle, uint32_t limit)
 {
-	int ret = -EINVAL;
+	int ret = -EOPNOTSUPP;
 	struct smu_context *smu = handle;
 
 	if (smu->ppt_funcs && smu->ppt_funcs->set_apu_thermal_limit)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 20f66e696f87..891605d4975f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -48,6 +48,7 @@
 #include "smu_cmn.h"
 #include "mp/mp_13_0_6_offset.h"
 #include "mp/mp_13_0_6_sh_mask.h"
+#include "umc_v12_0.h"
 
 #undef MP1_Public
 #undef smnMP1_FIRMWARE_FLAGS
@@ -94,22 +95,11 @@ MODULE_FIRMWARE("amdgpu/smu_13_0_6.bin");
 #define PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT 0x5
 #define LINK_SPEED_MAX				4
 
-#define SMU_13_0_6_DSCLK_THRESHOLD 100
+#define SMU_13_0_6_DSCLK_THRESHOLD 140
 
 #define MCA_BANK_IPID(_ip, _hwid, _type) \
 	[AMDGPU_MCA_IP_##_ip] = { .hwid = _hwid, .mcatype = _type, }
 
-enum mca_reg_idx {
-	MCA_REG_IDX_CONTROL		= 0,
-	MCA_REG_IDX_STATUS		= 1,
-	MCA_REG_IDX_ADDR		= 2,
-	MCA_REG_IDX_MISC0		= 3,
-	MCA_REG_IDX_CONFIG		= 4,
-	MCA_REG_IDX_IPID		= 5,
-	MCA_REG_IDX_SYND		= 6,
-	MCA_REG_IDX_COUNT		= 16,
-};
-
 struct mca_bank_ipid {
 	enum amdgpu_mca_ip ip;
 	uint16_t hwid;
@@ -122,7 +112,9 @@ struct mca_ras_info {
 	int *err_code_array;
 	int err_code_count;
 	int (*get_err_count)(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
-			     enum amdgpu_mca_error_type type, int idx, uint32_t *count);
+			     enum amdgpu_mca_error_type type, struct mca_bank_entry *entry, uint32_t *count);
+	bool (*bank_is_valid)(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
+			      enum amdgpu_mca_error_type type, struct mca_bank_entry *entry);
 };
 
 #define P2S_TABLE_ID_A 0x50325341
@@ -2305,7 +2297,7 @@ static int smu_v13_0_6_post_init(struct smu_context *smu)
 	struct amdgpu_device *adev = smu->adev;
 
 	if (!amdgpu_sriov_vf(adev) && adev->ras_enabled)
-		return smu_v13_0_6_mca_set_debug_mode(smu, true);
+		return smu_v13_0_6_mca_set_debug_mode(smu, false);
 
 	return 0;
 }
@@ -2387,6 +2379,7 @@ static const struct mca_bank_ipid smu_v13_0_6_mca_ipid_table[AMDGPU_MCA_IP_COUNT
 	MCA_BANK_IPID(UMC, 0x96, 0x0),
 	MCA_BANK_IPID(SMU, 0x01, 0x1),
 	MCA_BANK_IPID(MP5, 0x01, 0x2),
+	MCA_BANK_IPID(PCS_XGMI, 0x50, 0x0),
 };
 
 static void mca_bank_entry_info_decode(struct mca_bank_entry *entry, struct mca_bank_info *info)
@@ -2448,53 +2441,60 @@ static int mca_get_mca_entry(struct amdgpu_device *adev, enum amdgpu_mca_error_t
 	return 0;
 }
 
-static int mca_decode_mca_ipid(struct amdgpu_device *adev, enum amdgpu_mca_error_type type, int idx, int *ip)
+static int mca_decode_ipid_to_hwip(uint64_t val)
 {
 	const struct mca_bank_ipid *ipid;
-	uint64_t val;
 	uint16_t hwid, mcatype;
-	int i, ret;
-
-	ret = mca_bank_read_reg(adev, type, idx, MCA_REG_IDX_IPID, &val);
-	if (ret)
-		return ret;
+	int i;
 
 	hwid = REG_GET_FIELD(val, MCMP1_IPIDT0, HardwareID);
 	mcatype = REG_GET_FIELD(val, MCMP1_IPIDT0, McaType);
 
-	if (hwid) {
-		for (i = 0; i < ARRAY_SIZE(smu_v13_0_6_mca_ipid_table); i++) {
-			ipid = &smu_v13_0_6_mca_ipid_table[i];
+	for (i = 0; i < ARRAY_SIZE(smu_v13_0_6_mca_ipid_table); i++) {
+		ipid = &smu_v13_0_6_mca_ipid_table[i];
 
-			if (!ipid->hwid)
-				continue;
+		if (!ipid->hwid)
+			continue;
 
-			if (ipid->hwid == hwid && ipid->mcatype == mcatype) {
-				*ip = i;
-				return 0;
-			}
-		}
+		if (ipid->hwid == hwid && ipid->mcatype == mcatype)
+			return i;
 	}
 
-	*ip = AMDGPU_MCA_IP_UNKNOW;
+	return AMDGPU_MCA_IP_UNKNOW;
+}
+
+static int mca_umc_mca_get_err_count(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
+				     enum amdgpu_mca_error_type type, struct mca_bank_entry *entry, uint32_t *count)
+{
+	uint64_t status0;
+
+	status0 = entry->regs[MCA_REG_IDX_STATUS];
+
+	if (!REG_GET_FIELD(status0, MCMP1_STATUST0, Val)) {
+		*count = 0;
+		return 0;
+	}
+
+	if (type == AMDGPU_MCA_ERROR_TYPE_UE && umc_v12_0_is_uncorrectable_error(status0))
+		*count = 1;
+	else if (type == AMDGPU_MCA_ERROR_TYPE_CE && umc_v12_0_is_correctable_error(status0))
+		*count = 1;
 
 	return 0;
 }
 
-static int mca_normal_mca_get_err_count(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
-					enum amdgpu_mca_error_type type, int idx, uint32_t *count)
+static int mca_pcs_xgmi_mca_get_err_count(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
+					  enum amdgpu_mca_error_type type, struct mca_bank_entry *entry,
+					  uint32_t *count)
 {
-	uint64_t status0;
-	int ret;
+	u32 ext_error_code;
 
-	ret = mca_bank_read_reg(adev, type, idx, MCA_REG_IDX_STATUS, &status0);
-	if (ret)
-		return ret;
+	ext_error_code = MCA_REG__STATUS__ERRORCODEEXT(entry->regs[MCA_REG_IDX_STATUS]);
 
-	if (REG_GET_FIELD(status0, MCMP1_STATUST0, Val))
+	if (type == AMDGPU_MCA_ERROR_TYPE_UE && ext_error_code == 0)
+		*count = 1;
+	else if (type == AMDGPU_MCA_ERROR_TYPE_CE && ext_error_code == 6)
 		*count = 1;
-	else
-		*count = 0;
 
 	return 0;
 }
@@ -2515,70 +2515,41 @@ static bool mca_smu_check_error_code(struct amdgpu_device *adev, const struct mc
 	return false;
 }
 
-static int mca_mp5_mca_get_err_count(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
-				     enum amdgpu_mca_error_type type, int idx, uint32_t *count)
+static int mca_gfx_mca_get_err_count(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
+				     enum amdgpu_mca_error_type type, struct mca_bank_entry *entry, uint32_t *count)
 {
-	uint64_t status0 = 0, misc0 = 0;
-	uint32_t errcode;
-	int ret;
-
-	if (mca_ras->ip != AMDGPU_MCA_IP_MP5)
-		return -EINVAL;
-
-	ret = mca_bank_read_reg(adev, type, idx, MCA_REG_IDX_STATUS, &status0);
-	if (ret)
-		return ret;
+	uint64_t status0, misc0;
 
+	status0 = entry->regs[MCA_REG_IDX_STATUS];
 	if (!REG_GET_FIELD(status0, MCMP1_STATUST0, Val)) {
 		*count = 0;
 		return 0;
 	}
 
-	errcode = REG_GET_FIELD(status0, MCMP1_STATUST0, ErrorCode);
-	if (!mca_smu_check_error_code(adev, mca_ras, errcode))
-		return 0;
-
 	if (type == AMDGPU_MCA_ERROR_TYPE_UE &&
 	    REG_GET_FIELD(status0, MCMP1_STATUST0, UC) == 1 &&
 	    REG_GET_FIELD(status0, MCMP1_STATUST0, PCC) == 1) {
-		if (count)
-			*count = 1;
+		*count = 1;
 		return 0;
-	}
-
-	ret = mca_bank_read_reg(adev, type, idx, MCA_REG_IDX_MISC0, &misc0);
-	if (ret)
-		return ret;
-
-	if (count)
+	} else {
+		misc0 = entry->regs[MCA_REG_IDX_MISC0];
 		*count = REG_GET_FIELD(misc0, MCMP1_MISC0T0, ErrCnt);
+	}
 
 	return 0;
 }
 
 static int mca_smu_mca_get_err_count(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
-				     enum amdgpu_mca_error_type type, int idx, uint32_t *count)
+				     enum amdgpu_mca_error_type type, struct mca_bank_entry *entry, uint32_t *count)
 {
-	uint64_t status0 = 0, misc0 = 0;
-	uint32_t errcode;
-	int ret;
-
-	if (mca_ras->ip != AMDGPU_MCA_IP_SMU)
-		return -EINVAL;
-
-	ret = mca_bank_read_reg(adev, type, idx, MCA_REG_IDX_STATUS, &status0);
-	if (ret)
-		return ret;
+	uint64_t status0, misc0;
 
+	status0 = entry->regs[MCA_REG_IDX_STATUS];
 	if (!REG_GET_FIELD(status0, MCMP1_STATUST0, Val)) {
 		*count = 0;
 		return 0;
 	}
 
-	errcode = REG_GET_FIELD(status0, MCMP1_STATUST0, ErrorCode);
-	if (!mca_smu_check_error_code(adev, mca_ras, errcode))
-		return 0;
-
 	if (type == AMDGPU_MCA_ERROR_TYPE_UE &&
 	    REG_GET_FIELD(status0, MCMP1_STATUST0, UC) == 1 &&
 	    REG_GET_FIELD(status0, MCMP1_STATUST0, PCC) == 1) {
@@ -2587,16 +2558,43 @@ static int mca_smu_mca_get_err_count(const struct mca_ras_info *mca_ras, struct
 		return 0;
 	}
 
-	ret = mca_bank_read_reg(adev, type, idx, MCA_REG_IDX_MISC0, &misc0);
-	if (ret)
-		return ret;
-
-	if (count)
-		*count = REG_GET_FIELD(misc0, MCMP1_MISC0T0, ErrCnt);
+	misc0 = entry->regs[MCA_REG_IDX_MISC0];
+	*count = REG_GET_FIELD(misc0, MCMP1_MISC0T0, ErrCnt);
 
 	return 0;
 }
 
+static bool mca_gfx_smu_bank_is_valid(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
+				      enum amdgpu_mca_error_type type, struct mca_bank_entry *entry)
+{
+	uint32_t instlo;
+
+	instlo = REG_GET_FIELD(entry->regs[MCA_REG_IDX_IPID], MCMP1_IPIDT0, InstanceIdLo);
+	switch (instlo) {
+	case 0x36430400: /* SMNAID XCD 0 */
+	case 0x38430400: /* SMNAID XCD 1 */
+	case 0x40430400: /* SMNXCD XCD 0, NOTE: FIXME: fix this error later */
+		return true;
+	default:
+		return false;
+	}
+
+	return false;
+};
+
+static bool mca_smu_bank_is_valid(const struct mca_ras_info *mca_ras, struct amdgpu_device *adev,
+				  enum amdgpu_mca_error_type type, struct mca_bank_entry *entry)
+{
+	uint32_t errcode, instlo;
+
+	instlo = REG_GET_FIELD(entry->regs[MCA_REG_IDX_IPID], MCMP1_IPIDT0, InstanceIdLo);
+	if (instlo != 0x03b30400)
+		return false;
+
+	errcode = REG_GET_FIELD(entry->regs[MCA_REG_IDX_STATUS], MCMP1_STATUST0, ErrorCode);
+	return mca_smu_check_error_code(adev, mca_ras, errcode);
+}
+
 static int sdma_err_codes[] = { CODE_SDMA0, CODE_SDMA1, CODE_SDMA2, CODE_SDMA3 };
 static int mmhub_err_codes[] = {
 	CODE_DAGB0, CODE_DAGB0 + 1, CODE_DAGB0 + 2, CODE_DAGB0 + 3, CODE_DAGB0 + 4, /* DAGB0-4 */
@@ -2608,23 +2606,30 @@ static const struct mca_ras_info mca_ras_table[] = {
 	{
 		.blkid = AMDGPU_RAS_BLOCK__UMC,
 		.ip = AMDGPU_MCA_IP_UMC,
-		.get_err_count = mca_normal_mca_get_err_count,
+		.get_err_count = mca_umc_mca_get_err_count,
 	}, {
 		.blkid = AMDGPU_RAS_BLOCK__GFX,
-		.ip = AMDGPU_MCA_IP_MP5,
-		.get_err_count = mca_mp5_mca_get_err_count,
+		.ip = AMDGPU_MCA_IP_SMU,
+		.get_err_count = mca_gfx_mca_get_err_count,
+		.bank_is_valid = mca_gfx_smu_bank_is_valid,
 	}, {
 		.blkid = AMDGPU_RAS_BLOCK__SDMA,
 		.ip = AMDGPU_MCA_IP_SMU,
 		.err_code_array = sdma_err_codes,
 		.err_code_count = ARRAY_SIZE(sdma_err_codes),
 		.get_err_count = mca_smu_mca_get_err_count,
+		.bank_is_valid = mca_smu_bank_is_valid,
 	}, {
 		.blkid = AMDGPU_RAS_BLOCK__MMHUB,
 		.ip = AMDGPU_MCA_IP_SMU,
 		.err_code_array = mmhub_err_codes,
 		.err_code_count = ARRAY_SIZE(mmhub_err_codes),
 		.get_err_count = mca_smu_mca_get_err_count,
+		.bank_is_valid = mca_smu_bank_is_valid,
+	}, {
+		.blkid = AMDGPU_RAS_BLOCK__XGMI_WAFL,
+		.ip = AMDGPU_MCA_IP_PCS_XGMI,
+		.get_err_count = mca_pcs_xgmi_mca_get_err_count,
 	},
 };
 
@@ -2659,130 +2664,84 @@ static int mca_get_valid_mca_count(struct amdgpu_device *adev, enum amdgpu_mca_e
 }
 
 static bool mca_bank_is_valid(struct amdgpu_device *adev, const struct mca_ras_info *mca_ras,
-			      enum amdgpu_mca_error_type type, int idx)
+			      enum amdgpu_mca_error_type type, struct mca_bank_entry *entry)
 {
-	int ret, ip = AMDGPU_MCA_IP_UNKNOW;
-
-	ret = mca_decode_mca_ipid(adev, type, idx, &ip);
-	if (ret)
-		return false;
-
-	if (ip == AMDGPU_MCA_IP_UNKNOW)
+	if (mca_decode_ipid_to_hwip(entry->regs[MCA_REG_IDX_IPID]) != mca_ras->ip)
 		return false;
 
-	return ip == mca_ras->ip;
-}
+	if (mca_ras->bank_is_valid)
+		return mca_ras->bank_is_valid(mca_ras, adev, type, entry);
 
-static int mca_get_valid_mca_idx(struct amdgpu_device *adev, const struct mca_ras_info *mca_ras,
-				 enum amdgpu_mca_error_type type,
-				 uint32_t mca_cnt, int *idx_array, int idx_array_size)
-{
-	int i, idx_cnt = 0;
-
-	for (i = 0; i < mca_cnt; i++) {
-		if (!mca_bank_is_valid(adev, mca_ras, type, i))
-			continue;
-
-		if (idx_array) {
-			if (idx_cnt < idx_array_size)
-				idx_array[idx_cnt] = i;
-			else
-				return -EINVAL;
-		}
-
-		idx_cnt++;
-	}
-
-	return idx_cnt;
+	return true;
 }
 
-static int __mca_smu_get_error_count(struct amdgpu_device *adev, const struct mca_ras_info *mca_ras, enum amdgpu_mca_error_type type, uint32_t *count)
+static int __mca_smu_get_ras_mca_set(struct amdgpu_device *adev, const struct mca_ras_info *mca_ras,
+				     enum amdgpu_mca_error_type type, struct mca_bank_set *mca_set)
 {
-	uint32_t result, mca_cnt, total = 0;
-	int idx_array[16];
-	int i, ret, idx_cnt = 0;
+	struct mca_bank_entry entry;
+	uint32_t mca_cnt;
+	int i, ret;
 
 	ret = mca_get_valid_mca_count(adev, type, &mca_cnt);
 	if (ret)
 		return ret;
 
 	/* if valid mca bank count is 0, the driver can return 0 directly */
-	if (!mca_cnt) {
-		*count = 0;
+	if (!mca_cnt)
 		return 0;
-	}
 
-	if (!mca_ras->get_err_count)
-		return -EINVAL;
+	for (i = 0; i < mca_cnt; i++) {
+		memset(&entry, 0, sizeof(entry));
+		ret = mca_get_mca_entry(adev, type, i, &entry);
+		if (ret)
+			return ret;
 
-	idx_cnt = mca_get_valid_mca_idx(adev, mca_ras, type, mca_cnt, idx_array, ARRAY_SIZE(idx_array));
-	if (idx_cnt < 0)
-		return -EINVAL;
+		if (mca_ras && !mca_bank_is_valid(adev, mca_ras, type, &entry))
+			continue;
 
-	for (i = 0; i < idx_cnt; i++) {
-		result = 0;
-		ret = mca_ras->get_err_count(mca_ras, adev, type, idx_array[i], &result);
+		ret = amdgpu_mca_bank_set_add_entry(mca_set, &entry);
 		if (ret)
 			return ret;
-
-		total += result;
 	}
 
-	*count = total;
-
 	return 0;
 }
 
-static int mca_smu_get_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
-				   enum amdgpu_mca_error_type type, uint32_t *count)
+static int mca_smu_get_ras_mca_set(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
+				   enum amdgpu_mca_error_type type, struct mca_bank_set *mca_set)
 {
-	const struct mca_ras_info *mca_ras;
+	const struct mca_ras_info *mca_ras = NULL;
 
-	if (!count)
+	if (!mca_set)
 		return -EINVAL;
 
-	mca_ras = mca_get_mca_ras_info(adev, blk);
-	if (!mca_ras)
-		return -EOPNOTSUPP;
-
-	return __mca_smu_get_error_count(adev, mca_ras, type, count);
-}
-
-static int __mca_smu_get_ras_mca_idx_array(struct amdgpu_device *adev, const struct mca_ras_info *mca_ras,
-					   enum amdgpu_mca_error_type type, int *idx_array, int *idx_array_size)
-{
-	uint32_t mca_cnt = 0;
-	int ret, idx_cnt = 0;
-
-	ret = mca_get_valid_mca_count(adev, type, &mca_cnt);
-	if (ret)
-		return ret;
-
-	/* if valid mca bank count is 0, the driver can return 0 directly */
-	if (!mca_cnt) {
-		*idx_array_size = 0;
-		return 0;
+	if (blk != AMDGPU_RAS_BLOCK_COUNT) {
+		mca_ras = mca_get_mca_ras_info(adev, blk);
+		if (!mca_ras)
+			return -EOPNOTSUPP;
 	}
 
-	idx_cnt = mca_get_valid_mca_idx(adev, mca_ras, type, mca_cnt, idx_array, *idx_array_size);
-	if (idx_cnt < 0)
-		return -EINVAL;
-
-	*idx_array_size = idx_cnt;
-
-	return 0;
+	return __mca_smu_get_ras_mca_set(adev, mca_ras, type, mca_set);
 }
 
-static int mca_smu_get_ras_mca_idx_array(struct amdgpu_device *adev, enum amdgpu_ras_block blk,
-					 enum amdgpu_mca_error_type type, int *idx_array, int *idx_array_size)
+static int mca_smu_parse_mca_error_count(struct amdgpu_device *adev, enum amdgpu_ras_block blk, enum amdgpu_mca_error_type type,
+					 struct mca_bank_entry *entry, uint32_t *count)
 {
 	const struct mca_ras_info *mca_ras;
 
+	if (!entry || !count)
+		return -EINVAL;
+
 	mca_ras = mca_get_mca_ras_info(adev, blk);
 	if (!mca_ras)
 		return -EOPNOTSUPP;
 
-	return __mca_smu_get_ras_mca_idx_array(adev, mca_ras, type, idx_array, idx_array_size);
+	if (!mca_bank_is_valid(adev, mca_ras, type, entry)) {
+		*count = 0;
+		return 0;
+	}
+
+	return mca_ras->get_err_count(mca_ras, adev, type, entry, count);
 }
 
 static int mca_smu_get_mca_entry(struct amdgpu_device *adev,
@@ -2801,10 +2760,10 @@ static const struct amdgpu_mca_smu_funcs smu_v13_0_6_mca_smu_funcs = {
 	.max_ue_count = 12,
 	.max_ce_count = 12,
 	.mca_set_debug_mode = mca_smu_set_debug_mode,
-	.mca_get_error_count = mca_smu_get_error_count,
+	.mca_get_ras_mca_set = mca_smu_get_ras_mca_set,
+	.mca_parse_mca_error_count = mca_smu_parse_mca_error_count,
 	.mca_get_mca_entry = mca_smu_get_mca_entry,
 	.mca_get_valid_mca_count = mca_smu_get_valid_mca_count,
-	.mca_get_ras_mca_idx_array = mca_smu_get_ras_mca_idx_array,
 };
 
 static int smu_v13_0_6_select_xgmi_plpd_policy(struct smu_context *smu,
author	Daniel Vetter <daniel.vetter@ffwll.ch>	2023-11-10 20:51:37 +0100
committer	Daniel Vetter <daniel.vetter@ffwll.ch>	2023-11-10 20:51:38 +0100
commit	03df0fc007ca4713fa1e716329af556f981807e4 (patch)
tree	afb3271fe3d62fb9516ab9d468f66e42affc6858 /drivers
parent	aec3e2e23b08f188c22f36c4108467f80e980b15 (diff)
parent	8473bfdcb5b1a32fd05629c4535ccacd73bc5567 (diff)