amd-drm-next-6.8-2023-12-15:

amdgpu:
 - Suspend fixes
 - Misc code cleanups
 - JPEG fix
 - Add AMD specific color management (protected by AMD_PRIVATE_COLOR)
 - UHBR13.5 cable fixes
 - Misc display fixes
 - Display WB fixes
 - PSR fixes
 - XGMI fix
 - ACPI WBRF support for handling potential RF interference from GPU clocks
 - Enable tunneling on high priority compute queues
 - drm_edid.h include cleanup
 - VPE DPM support
 - SMU 13 fixes
 - Fix possible double frees in error paths
 - Misc fixes
 
 amdkfd:
 - Support import and export of dma-bufs using GEM handles
 - MES shader debugger fixes
 - SVM fixes
 
 radeon:
 - drm_edid.h include cleanup
 - Misc code cleanups
 - Fix possible memory leak in error path
 
 drm:
 - Increase max objects to accomodate new color props
 - Make replace_property_blob_from_id a DRM helper
 - Track color management changes per plane
 
 platform-x86:
 - Merge immutable branch from Hans for platform dependencies for WBRF to coordinate
   merge of WBRF feature across wifi, platform, and GPU
 -----BEGIN PGP SIGNATURE-----
 
 iHUEABYKAB0WIQQgO5Idg2tXNTSZAr293/aFa7yZ2AUCZXygTgAKCRC93/aFa7yZ
 2EW1AQCILfGTtDWXzgLSpUBtt9jOooHqaSrah19Cfw0HlA3QIQD+OCohXH1LLZo1
 tYHyfsLv0LsNawI198qABzB1PwptSAI=
 =M1AO
 -----END PGP SIGNATURE-----

Merge tag 'amd-drm-next-6.8-2023-12-15' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.8-2023-12-15:

amdgpu:
- Suspend fixes
- Misc code cleanups
- JPEG fix
- Add AMD specific color management (protected by AMD_PRIVATE_COLOR)
- UHBR13.5 cable fixes
- Misc display fixes
- Display WB fixes
- PSR fixes
- XGMI fix
- ACPI WBRF support for handling potential RF interference from GPU clocks
- Enable tunneling on high priority compute queues
- drm_edid.h include cleanup
- VPE DPM support
- SMU 13 fixes
- Fix possible double frees in error paths
- Misc fixes

amdkfd:
- Support import and export of dma-bufs using GEM handles
- MES shader debugger fixes
- SVM fixes

radeon:
- drm_edid.h include cleanup
- Misc code cleanups
- Fix possible memory leak in error path

drm:
- Increase max objects to accomodate new color props
- Make replace_property_blob_from_id a DRM helper
- Track color management changes per plane

platform-x86:
- Merge immutable branch from Hans for platform dependencies for WBRF to coordinate
  merge of WBRF feature across wifi, platform, and GPU

Signed-off-by: Dave Airlie <airlied@redhat.com>

# -----BEGIN PGP SIGNATURE-----
#
# iHUEABYKAB0WIQQgO5Idg2tXNTSZAr293/aFa7yZ2AUCZXygTgAKCRC93/aFa7yZ
# 2EW1AQCILfGTtDWXzgLSpUBtt9jOooHqaSrah19Cfw0HlA3QIQD+OCohXH1LLZo1
# tYHyfsLv0LsNawI198qABzB1PwptSAI=
# =M1AO
# -----END PGP SIGNATURE-----
# gpg: Signature made Sat 16 Dec 2023 04:51:58 AEST
# gpg:                using EDDSA key 203B921D836B5735349902BDBDDFF6856BBC99D8
# gpg: Can't check signature: No public key
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231215193519.5040-1-alexander.deucher@amd.com

Documentation/driver-api/index.rst

@@ -115,6 +115,7 @@ available subsections can be seen below.
    hte/index
    wmi
    dpll
+   wbrf
 
 .. only::  subproject and html
 

Documentation/driver-api/wbrf.rst

@@ -0,0 +1,78 @@
+.. SPDX-License-Identifier: GPL-2.0-or-later
+
+=================================
+WBRF - Wifi Band RFI Mitigations
+=================================
+
+Due to electrical and mechanical constraints in certain platform designs
+there may be likely interference of relatively high-powered harmonics of
+the GPU memory clocks with local radio module frequency bands used by
+certain Wifi bands.
+
+To mitigate possible RFI interference producers can advertise the
+frequencies in use and consumers can use this information to avoid using
+these frequencies for sensitive features.
+
+When a platform is known to have this issue with any contained devices,
+the platform designer will advertise the availability of this feature via
+ACPI devices with a device specific method (_DSM).
+* Producers with this _DSM will be able to advertise the frequencies in use.
+* Consumers with this _DSM will be able to register for notifications of
+frequencies in use.
+
+Some general terms
+==================
+
+Producer: such component who can produce high-powered radio frequency
+Consumer: such component who can adjust its in-use frequency in
+response to the radio frequencies of other components to mitigate the
+possible RFI.
+
+To make the mechanism function, those producers should notify active use
+of their particular frequencies so that other consumers can make relative
+internal adjustments as necessary to avoid this resonance.
+
+ACPI interface
+==============
+
+Although initially used by for wifi + dGPU use cases, the ACPI interface
+can be scaled to any type of device that a platform designer discovers
+can cause interference.
+
+The GUID used for the _DSM is 7B7656CF-DC3D-4C1C-83E9-66E721DE3070.
+
+3 functions are available in this _DSM:
+
+* 0: discover # of functions available
+* 1: record RF bands in use
+* 2: retrieve RF bands in use
+
+Driver programming interface
+============================
+
+.. kernel-doc:: drivers/platform/x86/amd/wbrf.c
+
+Sample Usage
+=============
+
+The expected flow for the producers:
+1. During probe, call `acpi_amd_wbrf_supported_producer` to check if WBRF
+can be enabled for the device.
+2. On using some frequency band, call `acpi_amd_wbrf_add_remove` with 'add'
+param to get other consumers properly notified.
+3. Or on stopping using some frequency band, call
+`acpi_amd_wbrf_add_remove` with 'remove' param to get other consumers notified.
+
+The expected flow for the consumers:
+1. During probe, call `acpi_amd_wbrf_supported_consumer` to check if WBRF
+can be enabled for the device.
+2. Call `amd_wbrf_register_notifier` to register for notification
+of frequency band change(add or remove) from other producers.
+3. Call the `amd_wbrf_retrieve_freq_band` initally to retrieve
+current active frequency bands considering some producers may broadcast
+such information before the consumer is up.
+4. On receiving a notification for frequency band change, run
+`amd_wbrf_retrieve_freq_band` again to retrieve the latest
+active frequency bands.
+5. During driver cleanup, call `amd_wbrf_unregister_notifier` to
+unregister the notifier.

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -252,6 +252,8 @@ extern int amdgpu_seamless;
 extern int amdgpu_user_partt_mode;
 extern int amdgpu_agp;
 
+extern int amdgpu_wbrf;
+
 #define AMDGPU_VM_MAX_NUM_CTX			4096
 #define AMDGPU_SG_THRESHOLD			(256*1024*1024)
 #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS	        3000
@@ -789,6 +791,7 @@ struct amdgpu_mqd_prop {
 	uint64_t eop_gpu_addr;
 	uint32_t hqd_pipe_priority;
 	uint32_t hqd_queue_priority;
+	bool allow_tunneling;
 	bool hqd_active;
 };
 

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c

@@ -142,6 +142,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 {
 	int i;
 	int last_valid_bit;
+	int ret;
 
 	amdgpu_amdkfd_gpuvm_init_mem_limits();
 
@@ -160,6 +161,12 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 			.enable_mes = adev->enable_mes,
 		};
 
+		ret = drm_client_init(&adev->ddev, &adev->kfd.client, "kfd", NULL);
+		if (ret) {
+			dev_err(adev->dev, "Failed to init DRM client: %d\n", ret);
+			return;
+		}
+
 		/* this is going to have a few of the MSBs set that we need to
 		 * clear
 		 */
@@ -198,6 +205,10 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 
 		adev->kfd.init_complete = kgd2kfd_device_init(adev->kfd.dev,
 							&gpu_resources);
+		if (adev->kfd.init_complete)
+			drm_client_register(&adev->kfd.client);
+		else
+			drm_client_release(&adev->kfd.client);
 
 		amdgpu_amdkfd_total_mem_size += adev->gmc.real_vram_size;
 

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h

@@ -33,6 +33,7 @@
 #include <linux/mmu_notifier.h>
 #include <linux/memremap.h>
 #include <kgd_kfd_interface.h>
+#include <drm/drm_client.h>
 #include "amdgpu_sync.h"
 #include "amdgpu_vm.h"
 #include "amdgpu_xcp.h"
@@ -83,6 +84,7 @@ struct kgd_mem {
 
 	struct amdgpu_sync sync;
 
+	uint32_t gem_handle;
 	bool aql_queue;
 	bool is_imported;
 };
@@ -105,6 +107,9 @@ struct amdgpu_kfd_dev {
 
 	/* HMM page migration MEMORY_DEVICE_PRIVATE mapping */
 	struct dev_pagemap pgmap;
+
+	/* Client for KFD BO GEM handle allocations */
+	struct drm_client_dev client;
 };
 
 enum kgd_engine_type {
@@ -309,11 +314,10 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *process_info,
 					    struct dma_fence **ef);
 int amdgpu_amdkfd_gpuvm_get_vm_fault_info(struct amdgpu_device *adev,
 					      struct kfd_vm_fault_info *info);
-int amdgpu_amdkfd_gpuvm_import_dmabuf(struct amdgpu_device *adev,
+int amdgpu_amdkfd_gpuvm_import_dmabuf_fd(struct amdgpu_device *adev, int fd,
-				      struct dma_buf *dmabuf,
+					 uint64_t va, void *drm_priv,
-				      uint64_t va, void *drm_priv,
+					 struct kgd_mem **mem, uint64_t *size,
-				      struct kgd_mem **mem, uint64_t *size,
+					 uint64_t *mmap_offset);
-				      uint64_t *mmap_offset);
 int amdgpu_amdkfd_gpuvm_export_dmabuf(struct kgd_mem *mem,
 				      struct dma_buf **dmabuf);
 void amdgpu_amdkfd_debug_mem_fence(struct amdgpu_device *adev);

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c

@@ -25,6 +25,7 @@
 #include <linux/pagemap.h>
 #include <linux/sched/mm.h>
 #include <linux/sched/task.h>
+#include <linux/fdtable.h>
 #include <drm/ttm/ttm_tt.h>
 
 #include <drm/drm_exec.h>
@@ -806,13 +807,22 @@ kfd_mem_dmaunmap_attachment(struct kgd_mem *mem,
 static int kfd_mem_export_dmabuf(struct kgd_mem *mem)
 {
 	if (!mem->dmabuf) {
-		struct dma_buf *ret = amdgpu_gem_prime_export(
+		struct amdgpu_device *bo_adev;
-			&mem->bo->tbo.base,
+		struct dma_buf *dmabuf;
+		int r, fd;
+
+		bo_adev = amdgpu_ttm_adev(mem->bo->tbo.bdev);
+		r = drm_gem_prime_handle_to_fd(&bo_adev->ddev, bo_adev->kfd.client.file,
+					       mem->gem_handle,
 			mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE ?
-				DRM_RDWR : 0);
+					       DRM_RDWR : 0, &fd);
-		if (IS_ERR(ret))
+		if (r)
-			return PTR_ERR(ret);
+			return r;
-		mem->dmabuf = ret;
+		dmabuf = dma_buf_get(fd);
+		close_fd(fd);
+		if (WARN_ON_ONCE(IS_ERR(dmabuf)))
+			return PTR_ERR(dmabuf);
+		mem->dmabuf = dmabuf;
 	}
 
 	return 0;
@@ -1778,6 +1788,9 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 		pr_debug("Failed to allow vma node access. ret %d\n", ret);
 		goto err_node_allow;
 	}
+	ret = drm_gem_handle_create(adev->kfd.client.file, gobj, &(*mem)->gem_handle);
+	if (ret)
+		goto err_gem_handle_create;
 	bo = gem_to_amdgpu_bo(gobj);
 	if (bo_type == ttm_bo_type_sg) {
 		bo->tbo.sg = sg;
@@ -1829,6 +1842,8 @@ allocate_init_user_pages_failed:
 err_pin_bo:
 err_validate_bo:
 	remove_kgd_mem_from_kfd_bo_list(*mem, avm->process_info);
+	drm_gem_handle_delete(adev->kfd.client.file, (*mem)->gem_handle);
+err_gem_handle_create:
 	drm_vma_node_revoke(&gobj->vma_node, drm_priv);
 err_node_allow:
 	/* Don't unreserve system mem limit twice */
@@ -1941,8 +1956,11 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu(
 
 	/* Free the BO*/
 	drm_vma_node_revoke(&mem->bo->tbo.base.vma_node, drm_priv);
-	if (mem->dmabuf)
+	drm_gem_handle_delete(adev->kfd.client.file, mem->gem_handle);
+	if (mem->dmabuf) {
 		dma_buf_put(mem->dmabuf);
+		mem->dmabuf = NULL;
+	}
 	mutex_destroy(&mem->lock);
 
 	/* If this releases the last reference, it will end up calling
@@ -2294,34 +2312,26 @@ int amdgpu_amdkfd_gpuvm_get_vm_fault_info(struct amdgpu_device *adev,
 	return 0;
 }
 
-int amdgpu_amdkfd_gpuvm_import_dmabuf(struct amdgpu_device *adev,
+static int import_obj_create(struct amdgpu_device *adev,
-				      struct dma_buf *dma_buf,
+			     struct dma_buf *dma_buf,
-				      uint64_t va, void *drm_priv,
+			     struct drm_gem_object *obj,
-				      struct kgd_mem **mem, uint64_t *size,
+			     uint64_t va, void *drm_priv,
-				      uint64_t *mmap_offset)
+			     struct kgd_mem **mem, uint64_t *size,
+			     uint64_t *mmap_offset)
 {
 	struct amdgpu_vm *avm = drm_priv_to_vm(drm_priv);
-	struct drm_gem_object *obj;
 	struct amdgpu_bo *bo;
 	int ret;
 
-	obj = amdgpu_gem_prime_import(adev_to_drm(adev), dma_buf);
-	if (IS_ERR(obj))
-		return PTR_ERR(obj);
-
 	bo = gem_to_amdgpu_bo(obj);
 	if (!(bo->preferred_domains & (AMDGPU_GEM_DOMAIN_VRAM |
-				    AMDGPU_GEM_DOMAIN_GTT))) {
+				    AMDGPU_GEM_DOMAIN_GTT)))
 		/* Only VRAM and GTT BOs are supported */
-		ret = -EINVAL;
+		return -EINVAL;
-		goto err_put_obj;
-	}
 
 	*mem = kzalloc(sizeof(struct kgd_mem), GFP_KERNEL);
-	if (!*mem) {
+	if (!*mem)
-		ret = -ENOMEM;
+		return -ENOMEM;
-		goto err_put_obj;
-	}
 
 	ret = drm_vma_node_allow(&obj->vma_node, drm_priv);
 	if (ret)
@@ -2371,8 +2381,41 @@ err_remove_mem:
 	drm_vma_node_revoke(&obj->vma_node, drm_priv);
 err_free_mem:
 	kfree(*mem);
+	return ret;
+}
+
+int amdgpu_amdkfd_gpuvm_import_dmabuf_fd(struct amdgpu_device *adev, int fd,
+					 uint64_t va, void *drm_priv,
+					 struct kgd_mem **mem, uint64_t *size,
+					 uint64_t *mmap_offset)
+{
+	struct drm_gem_object *obj;
+	uint32_t handle;
+	int ret;
+
+	ret = drm_gem_prime_fd_to_handle(&adev->ddev, adev->kfd.client.file, fd,
+					 &handle);
+	if (ret)
+		return ret;
+	obj = drm_gem_object_lookup(adev->kfd.client.file, handle);
+	if (!obj) {
+		ret = -EINVAL;
+		goto err_release_handle;
+	}
+
+	ret = import_obj_create(adev, obj->dma_buf, obj, va, drm_priv, mem, size,
+				mmap_offset);
+	if (ret)
+		goto err_put_obj;
+
+	(*mem)->gem_handle = handle;
+
+	return 0;
+
 err_put_obj:
 	drm_gem_object_put(obj);
+err_release_handle:
+	drm_gem_handle_delete(adev->kfd.client.file, handle);
 	return ret;
 }
 

drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c

@@ -755,7 +755,7 @@ static ssize_t amdgpu_debugfs_regs_smc_read(struct file *f, char __user *buf,
 	int r;
 
 	if (!adev->smc_rreg)
-		return -EPERM;
+		return -EOPNOTSUPP;
 
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;
@@ -814,7 +814,7 @@ static ssize_t amdgpu_debugfs_regs_smc_write(struct file *f, const char __user *
 	int r;
 
 	if (!adev->smc_wreg)
-		return -EPERM;
+		return -EOPNOTSUPP;
 
 	if (size & 0x3 || *pos & 0x3)
 		return -EINVAL;

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

@@ -1599,7 +1599,7 @@ bool amdgpu_device_seamless_boot_supported(struct amdgpu_device *adev)
 	if (adev->mman.keep_stolen_vga_memory)
 		return false;
 
-	return adev->ip_versions[DCE_HWIP][0] >= IP_VERSION(3, 0, 0);
+	return amdgpu_ip_version(adev, DCE_HWIP, 0) >= IP_VERSION(3, 0, 0);
 }
 
 /*
@@ -4589,8 +4589,6 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 
 	amdgpu_ras_suspend(adev);
 
-	amdgpu_ttm_set_buffer_funcs_status(adev, false);
-
 	amdgpu_device_ip_suspend_phase1(adev);
 
 	if (!adev->in_s0ix)

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c

@@ -115,9 +115,10 @@
  *   3.54.0 - Add AMDGPU_CTX_QUERY2_FLAGS_RESET_IN_PROGRESS support
  * - 3.55.0 - Add AMDGPU_INFO_GPUVM_FAULT query
  * - 3.56.0 - Update IB start address and size alignment for decode and encode
+ * - 3.57.0 - Compute tunneling on GFX10+
  */
 #define KMS_DRIVER_MAJOR	3
-#define KMS_DRIVER_MINOR	56
+#define KMS_DRIVER_MINOR	57
 #define KMS_DRIVER_PATCHLEVEL	0
 
 /*
@@ -208,6 +209,7 @@ int amdgpu_umsch_mm;
 int amdgpu_seamless = -1; /* auto */
 uint amdgpu_debug_mask;
 int amdgpu_agp = -1; /* auto */
+int amdgpu_wbrf = -1;
 
 static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work);
 
@@ -971,6 +973,22 @@ module_param_named(debug_mask, amdgpu_debug_mask, uint, 0444);
 MODULE_PARM_DESC(agp, "AGP (-1 = auto (default), 0 = disable, 1 = enable)");
 module_param_named(agp, amdgpu_agp, int, 0444);
 
+/**
+ * DOC: wbrf (int)
+ * Enable Wifi RFI interference mitigation feature.
+ * Due to electrical and mechanical constraints there may be likely interference of
+ * relatively high-powered harmonics of the (G-)DDR memory clocks with local radio
+ * module frequency bands used by Wifi 6/6e/7. To mitigate the possible RFI interference,
+ * with this feature enabled, PMFW will use either “shadowed P-State” or “P-State” based
+ * on active list of frequencies in-use (to be avoided) as part of initial setting or
+ * P-state transition. However, there may be potential performance impact with this
+ * feature enabled.
+ * (0 = disabled, 1 = enabled, -1 = auto (default setting, will be enabled if supported))
+ */
+MODULE_PARM_DESC(wbrf,
+	"Enable Wifi RFI interference mitigation (0 = disabled, 1 = enabled, -1 = auto(default)");
+module_param_named(wbrf, amdgpu_wbrf, int, 0444);
+
 /* These devices are not supported by amdgpu.
  * They are supported by the mach64, r128, radeon drivers
  */

drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c

@@ -190,8 +190,8 @@ int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier,
 		pr_debug("hmm range: start = 0x%lx, end = 0x%lx",
 			hmm_range->start, hmm_range->end);
 
-		/* Assuming 128MB takes maximum 1 second to fault page address */
+		/* Assuming 64MB takes maximum 1 second to fault page address */
-		timeout = max((hmm_range->end - hmm_range->start) >> 27, 1UL);
+		timeout = max((hmm_range->end - hmm_range->start) >> 26, 1UL);
 		timeout *= HMM_RANGE_DEFAULT_TIMEOUT;
 		timeout = jiffies + msecs_to_jiffies(timeout);
 
@@ -199,6 +199,7 @@ retry:
 		hmm_range->notifier_seq = mmu_interval_read_begin(notifier);
 		r = hmm_range_fault(hmm_range);
 		if (unlikely(r)) {
+			schedule();
 			/*
 			 * FIXME: This timeout should encompass the retry from
 			 * mmu_interval_read_retry() as well.
@@ -212,7 +213,6 @@ retry:
 			break;
 		hmm_range->hmm_pfns += MAX_WALK_BYTE >> PAGE_SHIFT;
 		hmm_range->start = hmm_range->end;
-		schedule();
 	} while (hmm_range->end < end);
 
 	hmm_range->start = start;

drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h

@@ -46,6 +46,8 @@
 #define MCA_REG__STATUS__ERRORCODEEXT(x)	MCA_REG_FIELD(x, 21, 16)
 #define MCA_REG__STATUS__ERRORCODE(x)		MCA_REG_FIELD(x, 15, 0)
 
+#define MCA_REG__MISC0__ERRCNT(x)		MCA_REG_FIELD(x, 43, 32)
+
 #define MCA_REG__SYND__ERRORINFORMATION(x)	MCA_REG_FIELD(x, 17, 0)
 
 enum amdgpu_mca_ip {

drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c

@@ -916,6 +916,11 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 	op_input.op = MES_MISC_OP_SET_SHADER_DEBUGGER;
 	op_input.set_shader_debugger.process_context_addr = process_context_addr;
 	op_input.set_shader_debugger.flags.u32all = flags;
+
+	/* use amdgpu mes_flush_shader_debugger instead */
+	if (op_input.set_shader_debugger.flags.process_ctx_flush)
+		return -EINVAL;
+
 	op_input.set_shader_debugger.spi_gdbg_per_vmid_cntl = spi_gdbg_per_vmid_cntl;
 	memcpy(op_input.set_shader_debugger.tcp_watch_cntl, tcp_watch_cntl,
 			sizeof(op_input.set_shader_debugger.tcp_watch_cntl));
@@ -935,6 +940,32 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 	return r;
 }
 
+int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev,
+				     uint64_t process_context_addr)
+{
+	struct mes_misc_op_input op_input = {0};
+	int r;
+
+	if (!adev->mes.funcs->misc_op) {
+		DRM_ERROR("mes flush shader debugger is not supported!\n");
+		return -EINVAL;
+	}
+
+	op_input.op = MES_MISC_OP_SET_SHADER_DEBUGGER;
+	op_input.set_shader_debugger.process_context_addr = process_context_addr;
+	op_input.set_shader_debugger.flags.process_ctx_flush = true;
+
+	amdgpu_mes_lock(&adev->mes);
+
+	r = adev->mes.funcs->misc_op(&adev->mes, &op_input);
+	if (r)
+		DRM_ERROR("failed to set_shader_debugger\n");
+
+	amdgpu_mes_unlock(&adev->mes);
+
+	return r;
+}
+
 static void
 amdgpu_mes_ring_to_queue_props(struct amdgpu_device *adev,
 			       struct amdgpu_ring *ring,

drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h

@@ -296,9 +296,10 @@ struct mes_misc_op_input {
 			uint64_t process_context_addr;
 			union {
 				struct {
-					uint64_t single_memop : 1;
+					uint32_t single_memop : 1;
-					uint64_t single_alu_op : 1;
+					uint32_t single_alu_op : 1;
-					uint64_t reserved: 30;
+					uint32_t reserved: 29;
+					uint32_t process_ctx_flush: 1;
 				};
 				uint32_t u32all;
 			} flags;
@@ -374,7 +375,8 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev,
 				const uint32_t *tcp_watch_cntl,
 				uint32_t flags,
 				bool trap_en);
-
+int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev,
+				uint64_t process_context_addr);
 int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id,
 			int queue_type, int idx,
 			struct amdgpu_mes_ctx_data *ctx_data,

drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h

@@ -32,7 +32,6 @@
 
 #include <drm/display/drm_dp_helper.h>
 #include <drm/drm_crtc.h>
-#include <drm/drm_edid.h>
 #include <drm/drm_encoder.h>
 #include <drm/drm_fixed.h>
 #include <drm/drm_framebuffer.h>
@@ -51,6 +50,7 @@ struct amdgpu_device;
 struct amdgpu_encoder;
 struct amdgpu_router;
 struct amdgpu_hpd;
+struct edid;
 
 #define to_amdgpu_crtc(x) container_of(x, struct amdgpu_crtc, base)
 #define to_amdgpu_connector(x) container_of(x, struct amdgpu_connector, base)
@@ -343,6 +343,97 @@ struct amdgpu_mode_info {
 	int			disp_priority;
 	const struct amdgpu_display_funcs *funcs;
 	const enum drm_plane_type *plane_type;
+
+	/* Driver-private color mgmt props */
+
+	/* @plane_degamma_lut_property: Plane property to set a degamma LUT to
+	 * convert encoded values to light linear values before sampling or
+	 * blending.
+	 */
+	struct drm_property *plane_degamma_lut_property;
+	/* @plane_degamma_lut_size_property: Plane property to define the max
+	 * size of degamma LUT as supported by the driver (read-only).
+	 */
+	struct drm_property *plane_degamma_lut_size_property;
+	/**
+	 * @plane_degamma_tf_property: Plane pre-defined transfer function to
+	 * to go from scanout/encoded values to linear values.
+	 */
+	struct drm_property *plane_degamma_tf_property;
+	/**
+	 * @plane_hdr_mult_property:
+	 */
+	struct drm_property *plane_hdr_mult_property;
+
+	struct drm_property *plane_ctm_property;
+	/**
+	 * @shaper_lut_property: Plane property to set pre-blending shaper LUT
+	 * that converts color content before 3D LUT. If
+	 * plane_shaper_tf_property != Identity TF, AMD color module will
+	 * combine the user LUT values with pre-defined TF into the LUT
+	 * parameters to be programmed.
+	 */
+	struct drm_property *plane_shaper_lut_property;
+	/**
+	 * @shaper_lut_size_property: Plane property for the size of
+	 * pre-blending shaper LUT as supported by the driver (read-only).
+	 */
+	struct drm_property *plane_shaper_lut_size_property;
+	/**
+	 * @plane_shaper_tf_property: Plane property to set a predefined
+	 * transfer function for pre-blending shaper (before applying 3D LUT)
+	 * with or without LUT. There is no shaper ROM, but we can use AMD
+	 * color modules to program LUT parameters from predefined TF (or
+	 * from a combination of pre-defined TF and the custom 1D LUT).
+	 */
+	struct drm_property *plane_shaper_tf_property;
+	/**
+	 * @plane_lut3d_property: Plane property for color transformation using
+	 * a 3D LUT (pre-blending), a three-dimensional array where each
+	 * element is an RGB triplet. Each dimension has the size of
+	 * lut3d_size. The array contains samples from the approximated
+	 * function. On AMD, values between samples are estimated by
+	 * tetrahedral interpolation. The array is accessed with three indices,
+	 * one for each input dimension (color channel), blue being the
+	 * outermost dimension, red the innermost.
+	 */
+	struct drm_property *plane_lut3d_property;
+	/**
+	 * @plane_degamma_lut_size_property: Plane property to define the max
+	 * size of 3D LUT as supported by the driver (read-only). The max size
+	 * is the max size of one dimension and, therefore, the max number of
+	 * entries for 3D LUT array is the 3D LUT size cubed;
+	 */
+	struct drm_property *plane_lut3d_size_property;
+	/**
+	 * @plane_blend_lut_property: Plane property for output gamma before
+	 * blending. Userspace set a blend LUT to convert colors after 3D LUT
+	 * conversion. It works as a post-3DLUT 1D LUT. With shaper LUT, they
+	 * are sandwiching 3D LUT with two 1D LUT. If plane_blend_tf_property
+	 * != Identity TF, AMD color module will combine the user LUT values
+	 * with pre-defined TF into the LUT parameters to be programmed.
+	 */
+	struct drm_property *plane_blend_lut_property;
+	/**
+	 * @plane_blend_lut_size_property: Plane property to define the max
+	 * size of blend LUT as supported by the driver (read-only).
+	 */
+	struct drm_property *plane_blend_lut_size_property;
+	/**
+	 * @plane_blend_tf_property: Plane property to set a predefined
+	 * transfer function for pre-blending blend/out_gamma (after applying
+	 * 3D LUT) with or without LUT. There is no blend ROM, but we can use
+	 * AMD color modules to program LUT parameters from predefined TF (or
+	 * from a combination of pre-defined TF and the custom 1D LUT).
+	 */
+	struct drm_property *plane_blend_tf_property;
+	/* @regamma_tf_property: Transfer function for CRTC regamma
+	 * (post-blending). Possible values are defined by `enum
+	 * amdgpu_transfer_function`. There is no regamma ROM, but we can use
+	 * AMD color modules to program LUT parameters from predefined TF (or
+	 * from a combination of pre-defined TF and the custom 1D LUT).
+	 */
+	struct drm_property *regamma_tf_property;
 };
 
 #define AMDGPU_MAX_BL_LEVEL 0xFF

drivers/gpu/drm/amd/amdgpu/amdgpu_object.c

@@ -1245,19 +1245,15 @@ int amdgpu_bo_get_metadata(struct amdgpu_bo *bo, void *buffer,
  * amdgpu_bo_move_notify - notification about a memory move
  * @bo: pointer to a buffer object
  * @evict: if this move is evicting the buffer from the graphics address space
- * @new_mem: new information of the bufer object
  *
  * Marks the corresponding &amdgpu_bo buffer object as invalid, also performs
  * bookkeeping.
  * TTM driver callback which is called when ttm moves a buffer.
  */
-void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
+void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, bool evict)
-			   bool evict,
-			   struct ttm_resource *new_mem)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->bdev);
 	struct amdgpu_bo *abo;
-	struct ttm_resource *old_mem = bo->resource;
 
 	if (!amdgpu_bo_is_amdgpu_bo(bo))
 		return;
@@ -1274,13 +1270,6 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
 	/* remember the eviction */
 	if (evict)
 		atomic64_inc(&adev->num_evictions);
-
-	/* update statistics */
-	if (!new_mem)
-		return;
-
-	/* move_notify is called before move happens */
-	trace_amdgpu_bo_move(abo, new_mem->mem_type, old_mem->mem_type);
 }
 
 void amdgpu_bo_get_memory(struct amdgpu_bo *bo,
@@ -1343,6 +1332,8 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo)
 
 	abo = ttm_to_amdgpu_bo(bo);
 
+	WARN_ON(abo->vm_bo);
+
 	if (abo->kfd_bo)
 		amdgpu_amdkfd_release_notify(abo);
 

drivers/gpu/drm/amd/amdgpu/amdgpu_object.h

@@ -344,9 +344,7 @@ int amdgpu_bo_set_metadata (struct amdgpu_bo *bo, void *metadata,
 int amdgpu_bo_get_metadata(struct amdgpu_bo *bo, void *buffer,
 			   size_t buffer_size, uint32_t *metadata_size,
 			   uint64_t *flags);
-void amdgpu_bo_move_notify(struct ttm_buffer_object *bo,
+void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, bool evict);
-			   bool evict,
-			   struct ttm_resource *new_mem);
 void amdgpu_bo_release_notify(struct ttm_buffer_object *bo);
 vm_fault_t amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo);
 void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence,

drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c

@@ -1433,8 +1433,8 @@ int psp_xgmi_get_topology_info(struct psp_context *psp,
 			 get_extended_data) ||
 			amdgpu_ip_version(psp->adev, MP0_HWIP, 0) ==
 				IP_VERSION(13, 0, 6);
-		bool ta_port_num_support = psp->xgmi_context.xgmi_ta_caps &
+		bool ta_port_num_support = amdgpu_sriov_vf(psp->adev) ? 0 :
-						EXTEND_PEER_LINK_INFO_CMD_FLAG;
+				psp->xgmi_context.xgmi_ta_caps & EXTEND_PEER_LINK_INFO_CMD_FLAG;
 
 		/* popluate the shared output buffer rather than the cmd input buffer
 		 * with node_ids as the input for GET_PEER_LINKS command execution.

drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c

@@ -642,6 +642,10 @@ static void amdgpu_ring_to_mqd_prop(struct amdgpu_ring *ring,
 				    struct amdgpu_mqd_prop *prop)
 {
 	struct amdgpu_device *adev = ring->adev;
+	bool is_high_prio_compute = ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE &&
+				    amdgpu_gfx_is_high_priority_compute_queue(adev, ring);
+	bool is_high_prio_gfx = ring->funcs->type == AMDGPU_RING_TYPE_GFX &&
+				amdgpu_gfx_is_high_priority_graphics_queue(adev, ring);
 
 	memset(prop, 0, sizeof(*prop));
 
@@ -659,10 +663,8 @@ static void amdgpu_ring_to_mqd_prop(struct amdgpu_ring *ring,
 	 */
 	prop->hqd_active = ring->funcs->type == AMDGPU_RING_TYPE_KIQ;
 
-	if ((ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE &&
+	prop->allow_tunneling = is_high_prio_compute;
-	     amdgpu_gfx_is_high_priority_compute_queue(adev, ring)) ||
+	if (is_high_prio_compute || is_high_prio_gfx) {
-	    (ring->funcs->type == AMDGPU_RING_TYPE_GFX &&
-	     amdgpu_gfx_is_high_priority_graphics_queue(adev, ring))) {
 		prop->hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_HIGH;
 		prop->hqd_queue_priority = AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM;
 	}

drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c

@@ -545,10 +545,11 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 			return r;
 	}
 
+	trace_amdgpu_bo_move(abo, new_mem->mem_type, old_mem->mem_type);
 out:
 	/* update statistics */
 	atomic64_add(bo->base.size, &adev->num_bytes_moved);
-	amdgpu_bo_move_notify(bo, evict, new_mem);
+	amdgpu_bo_move_notify(bo, evict);
 	return 0;
 }
 
@@ -1553,7 +1554,7 @@ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo,
 static void
 amdgpu_bo_delete_mem_notify(struct ttm_buffer_object *bo)
 {
-	amdgpu_bo_move_notify(bo, false, NULL);
+	amdgpu_bo_move_notify(bo, false);
 }
 
 static struct ttm_device_funcs amdgpu_bo_driver = {

drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c

@@ -1099,7 +1099,8 @@ 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))
+	if (amdgpu_sriov_vf(adev) &&
+	    amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 2))
 		xnack_mode = false;
 
 	return xnack_mode;

drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c

@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0+
 
 #include <drm/drm_atomic_helper.h>
+#include <drm/drm_edid.h>
 #include <drm/drm_simple_kms_helper.h>
 #include <drm/drm_vblank.h>
 

drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c

@@ -642,13 +642,14 @@ static void amdgpu_vm_pt_free(struct amdgpu_vm_bo_base *entry)
 
 	if (!entry->bo)
 		return;
+
+	entry->bo->vm_bo = NULL;
 	shadow = amdgpu_bo_shadowed(entry->bo);
 	if (shadow) {
 		ttm_bo_set_bulk_move(&shadow->tbo, NULL);
 		amdgpu_bo_unref(&shadow);
 	}
 	ttm_bo_set_bulk_move(&entry->bo->tbo, NULL);
-	entry->bo->vm_bo = NULL;
 
 	spin_lock(&entry->vm->status_lock);
 	list_del(&entry->vm_status);

drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c

@@ -26,6 +26,7 @@
 #include "amdgpu.h"
 #include "amdgpu_ucode.h"
 #include "amdgpu_vpe.h"
+#include "amdgpu_smu.h"
 #include "soc15_common.h"
 #include "vpe_v6_1.h"
 
@@ -33,8 +34,186 @@
 /* VPE CSA resides in the 4th page of CSA */
 #define AMDGPU_CSA_VPE_OFFSET 	(4096 * 3)
 
+/* 1 second timeout */
+#define VPE_IDLE_TIMEOUT	msecs_to_jiffies(1000)
+
+#define VPE_MAX_DPM_LEVEL			4
+#define FIXED1_8_BITS_PER_FRACTIONAL_PART	8
+#define GET_PRATIO_INTEGER_PART(x)		((x) >> FIXED1_8_BITS_PER_FRACTIONAL_PART)
+
 static void vpe_set_ring_funcs(struct amdgpu_device *adev);
 
+static inline uint16_t div16_u16_rem(uint16_t dividend, uint16_t divisor, uint16_t *remainder)
+{
+	*remainder = dividend % divisor;
+	return dividend / divisor;
+}
+
+static inline uint16_t complete_integer_division_u16(
+	uint16_t dividend,
+	uint16_t divisor,
+	uint16_t *remainder)
+{
+	return div16_u16_rem(dividend, divisor, (uint16_t *)remainder);
+}
+
+static uint16_t vpe_u1_8_from_fraction(uint16_t numerator, uint16_t denominator)
+{
+	bool arg1_negative = numerator < 0;
+	bool arg2_negative = denominator < 0;
+
+	uint16_t arg1_value = (uint16_t)(arg1_negative ? -numerator : numerator);
+	uint16_t arg2_value = (uint16_t)(arg2_negative ? -denominator : denominator);
+
+	uint16_t remainder;
+
+	/* determine integer part */
+	uint16_t res_value = complete_integer_division_u16(
+		arg1_value, arg2_value, &remainder);
+
+	if (res_value > 127 /* CHAR_MAX */)
+		return 0;
+
+	/* determine fractional part */
+	{
+		unsigned int i = FIXED1_8_BITS_PER_FRACTIONAL_PART;
+
+		do {
+			remainder <<= 1;
+
+			res_value <<= 1;
+
+			if (remainder >= arg2_value) {
+				res_value |= 1;
+				remainder -= arg2_value;
+			}
+		} while (--i != 0);
+	}
+
+	/* round up LSB */
+	{
+		uint16_t summand = (remainder << 1) >= arg2_value;
+
+		if ((res_value + summand) > 32767 /* SHRT_MAX */)
+			return 0;
+
+		res_value += summand;
+	}
+
+	if (arg1_negative ^ arg2_negative)
+		res_value = -res_value;
+
+	return res_value;
+}
+
+static uint16_t vpe_internal_get_pratio(uint16_t from_frequency, uint16_t to_frequency)
+{
+	uint16_t pratio = vpe_u1_8_from_fraction(from_frequency, to_frequency);
+
+	if (GET_PRATIO_INTEGER_PART(pratio) > 1)
+		pratio = 0;
+
+	return pratio;
+}
+
+/*
+ * VPE has 4 DPM levels from level 0 (lowerest) to 3 (highest),
+ * VPE FW will dynamically decide which level should be used according to current loading.
+ *
+ * Get VPE and SOC clocks from PM, and select the appropriate four clock values,
+ * calculate the ratios of adjusting from one clock to another.
+ * The VPE FW can then request the appropriate frequency from the PMFW.
+ */
+int amdgpu_vpe_configure_dpm(struct amdgpu_vpe *vpe)
+{
+	struct amdgpu_device *adev = vpe->ring.adev;
+	uint32_t dpm_ctl;
+
+	if (adev->pm.dpm_enabled) {
+		struct dpm_clocks clock_table = { 0 };
+		struct dpm_clock *VPEClks;
+		struct dpm_clock *SOCClks;
+		uint32_t idx;
+		uint32_t pratio_vmax_vnorm = 0, pratio_vnorm_vmid = 0, pratio_vmid_vmin = 0;
+		uint16_t pratio_vmin_freq = 0, pratio_vmid_freq = 0, pratio_vnorm_freq = 0, pratio_vmax_freq = 0;
+
+		dpm_ctl = RREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_enable));
+		dpm_ctl |= 1; /* DPM enablement */
+		WREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_enable), dpm_ctl);
+
+		/* Get VPECLK and SOCCLK */
+		if (amdgpu_dpm_get_dpm_clock_table(adev, &clock_table)) {
+			dev_dbg(adev->dev, "%s: get clock failed!\n", __func__);
+			goto disable_dpm;
+		}
+
+		SOCClks = clock_table.SocClocks;
+		VPEClks = clock_table.VPEClocks;
+
+		/* vpe dpm only cares 4 levels. */
+		for (idx = 0; idx < VPE_MAX_DPM_LEVEL; idx++) {
+			uint32_t soc_dpm_level;
+			uint32_t min_freq;
+
+			if (idx == 0)
+				soc_dpm_level = 0;
+			else
+				soc_dpm_level = (idx * 2) + 1;
+
+			/* clamp the max level */
+			if (soc_dpm_level > PP_SMU_NUM_VPECLK_DPM_LEVELS - 1)
+				soc_dpm_level = PP_SMU_NUM_VPECLK_DPM_LEVELS - 1;
+
+			min_freq = (SOCClks[soc_dpm_level].Freq < VPEClks[soc_dpm_level].Freq) ?
+				   SOCClks[soc_dpm_level].Freq : VPEClks[soc_dpm_level].Freq;
+
+			switch (idx) {
+			case 0:
+				pratio_vmin_freq = min_freq;
+				break;
+			case 1:
+				pratio_vmid_freq = min_freq;
+				break;
+			case 2:
+				pratio_vnorm_freq = min_freq;
+				break;
+			case 3:
+				pratio_vmax_freq = min_freq;
+				break;
+			default:
+				break;
+			}
+		}
+
+		if (pratio_vmin_freq && pratio_vmid_freq && pratio_vnorm_freq && pratio_vmax_freq) {
+			uint32_t pratio_ctl;
+
+			pratio_vmax_vnorm = (uint32_t)vpe_internal_get_pratio(pratio_vmax_freq, pratio_vnorm_freq);
+			pratio_vnorm_vmid = (uint32_t)vpe_internal_get_pratio(pratio_vnorm_freq, pratio_vmid_freq);
+			pratio_vmid_vmin = (uint32_t)vpe_internal_get_pratio(pratio_vmid_freq, pratio_vmin_freq);
+
+			pratio_ctl = pratio_vmax_vnorm | (pratio_vnorm_vmid << 9) | (pratio_vmid_vmin << 18);
+			WREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_pratio), pratio_ctl);		/* PRatio */
+			WREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_request_interval), 24000);	/* 1ms, unit=1/24MHz */
+			WREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_decision_threshold), 1200000);	/* 50ms */
+			WREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_busy_clamp_threshold), 1200000);/* 50ms */
+			WREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_idle_clamp_threshold), 1200000);/* 50ms */
+			dev_dbg(adev->dev, "%s: configure vpe dpm pratio done!\n", __func__);
+		} else {
+			dev_dbg(adev->dev, "%s: invalid pratio parameters!\n", __func__);
+			goto disable_dpm;
+		}
+	}
+	return 0;
+
+disable_dpm:
+	dpm_ctl = RREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_enable));
+	dpm_ctl &= 0xfffffffe; /* Disable DPM */
+	WREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.dpm_enable), dpm_ctl);
+	dev_dbg(adev->dev, "%s: disable vpe dpm\n", __func__);
+	return 0;
+}
+
 int amdgpu_vpe_psp_update_sram(struct amdgpu_device *adev)
 {
 	struct amdgpu_firmware_info ucode = {
@@ -134,6 +313,19 @@ static int vpe_early_init(void *handle)
 	return 0;
 }
 
+static void vpe_idle_work_handler(struct work_struct *work)
+{
+	struct amdgpu_device *adev =
+		container_of(work, struct amdgpu_device, vpe.idle_work.work);
+	unsigned int fences = 0;
+
+	fences += amdgpu_fence_count_emitted(&adev->vpe.ring);
+
+	if (fences == 0)
+		amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VPE, AMD_PG_STATE_GATE);
+	else
+		schedule_delayed_work(&adev->vpe.idle_work, VPE_IDLE_TIMEOUT);
+}
 
 static int vpe_common_init(struct amdgpu_vpe *vpe)
 {
@@ -150,6 +342,9 @@ static int vpe_common_init(struct amdgpu_vpe *vpe)
 		return r;
 	}
 
+	vpe->context_started = false;
+	INIT_DELAYED_WORK(&adev->vpe.idle_work, vpe_idle_work_handler);
+
 	return 0;
 }
 
@@ -219,6 +414,9 @@ static int vpe_hw_fini(void *handle)
 
 	vpe_ring_stop(vpe);
 
+	/* Power off VPE */
+	amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VPE, AMD_PG_STATE_GATE);
+
 	return 0;
 }
 
@@ -226,6 +424,8 @@ static int vpe_suspend(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	cancel_delayed_work_sync(&adev->vpe.idle_work);
+
 	return vpe_hw_fini(adev);
 }
 
@@ -430,6 +630,21 @@ static int vpe_set_clockgating_state(void *handle,
 static int vpe_set_powergating_state(void *handle,
 				     enum amd_powergating_state state)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct amdgpu_vpe *vpe = &adev->vpe;
+
+	if (!adev->pm.dpm_enabled)
+		dev_err(adev->dev, "Without PM, cannot support powergating\n");
+
+	dev_dbg(adev->dev, "%s: %s!\n", __func__, (state == AMD_PG_STATE_GATE) ? "GATE":"UNGATE");
+
+	if (state == AMD_PG_STATE_GATE) {
+		amdgpu_dpm_enable_vpe(adev, false);
+		vpe->context_started = false;
+	} else {
+		amdgpu_dpm_enable_vpe(adev, true);
+	}
+
 	return 0;
 }
 
@@ -595,6 +810,38 @@ err0:
 	return ret;
 }
 
+static void vpe_ring_begin_use(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_vpe *vpe = &adev->vpe;
+
+	cancel_delayed_work_sync(&adev->vpe.idle_work);
+
+	/* Power on VPE and notify VPE of new context  */
+	if (!vpe->context_started) {
+		uint32_t context_notify;
+
+		/* Power on VPE */
+		amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VPE, AMD_PG_STATE_UNGATE);
+
+		/* Indicates that a job from a new context has been submitted. */
+		context_notify = RREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.context_indicator));
+		if ((context_notify & 0x1) == 0)
+			context_notify |= 0x1;
+		else
+			context_notify &= ~(0x1);
+		WREG32(vpe_get_reg_offset(vpe, 0, vpe->regs.context_indicator), context_notify);
+		vpe->context_started = true;
+	}
+}
+
+static void vpe_ring_end_use(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	schedule_delayed_work(&adev->vpe.idle_work, VPE_IDLE_TIMEOUT);
+}
+
 static const struct amdgpu_ring_funcs vpe_ring_funcs = {
 	.type = AMDGPU_RING_TYPE_VPE,
 	.align_mask = 0xf,
@@ -625,6 +872,8 @@ static const struct amdgpu_ring_funcs vpe_ring_funcs = {
 	.init_cond_exec = vpe_ring_init_cond_exec,
 	.patch_cond_exec = vpe_ring_patch_cond_exec,
 	.preempt_ib = vpe_ring_preempt_ib,
+	.begin_use = vpe_ring_begin_use,
+	.end_use = vpe_ring_end_use,
 };
 
 static void vpe_set_ring_funcs(struct amdgpu_device *adev)

drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.h

@@ -47,6 +47,15 @@ struct vpe_regs {
 	uint32_t queue0_rb_wptr_lo;
 	uint32_t queue0_rb_wptr_hi;
 	uint32_t queue0_preempt;
+
+	uint32_t dpm_enable;
+	uint32_t dpm_pratio;
+	uint32_t dpm_request_interval;
+	uint32_t dpm_decision_threshold;
+	uint32_t dpm_busy_clamp_threshold;
+	uint32_t dpm_idle_clamp_threshold;
+	uint32_t dpm_request_lv;
+	uint32_t context_indicator;
 };
 
 struct amdgpu_vpe {
@@ -63,12 +72,15 @@ struct amdgpu_vpe {
 	struct amdgpu_bo		*cmdbuf_obj;
 	uint64_t			cmdbuf_gpu_addr;
 	uint32_t			*cmdbuf_cpu_addr;
+	struct delayed_work		idle_work;
+	bool				context_started;
 };
 
 int amdgpu_vpe_psp_update_sram(struct amdgpu_device *adev);
 int amdgpu_vpe_init_microcode(struct amdgpu_vpe *vpe);
 int amdgpu_vpe_ring_init(struct amdgpu_vpe *vpe);
 int amdgpu_vpe_ring_fini(struct amdgpu_vpe *vpe);
+int amdgpu_vpe_configure_dpm(struct amdgpu_vpe *vpe);
 
 #define vpe_ring_init(vpe) ((vpe)->funcs->ring_init ? (vpe)->funcs->ring_init((vpe)) : 0)
 #define vpe_ring_start(vpe) ((vpe)->funcs->ring_start ? (vpe)->funcs->ring_start((vpe)) : 0)

drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c

@@ -823,6 +823,28 @@ static int amdgpu_xgmi_initialize_hive_get_data_partition(struct amdgpu_hive_inf
 	return 0;
 }
 
+static void amdgpu_xgmi_fill_topology_info(struct amdgpu_device *adev,
+	struct amdgpu_device *peer_adev)
+{
+	struct psp_xgmi_topology_info *top_info = &adev->psp.xgmi_context.top_info;
+	struct psp_xgmi_topology_info *peer_info = &peer_adev->psp.xgmi_context.top_info;
+
+	for (int i = 0; i < peer_info->num_nodes; i++) {
+		if (peer_info->nodes[i].node_id == adev->gmc.xgmi.node_id) {
+			for (int j = 0; j < top_info->num_nodes; j++) {
+				if (top_info->nodes[j].node_id == peer_adev->gmc.xgmi.node_id) {
+					peer_info->nodes[i].num_hops = top_info->nodes[j].num_hops;
+					peer_info->nodes[i].is_sharing_enabled =
+							top_info->nodes[j].is_sharing_enabled;
+					peer_info->nodes[i].num_links =
+							top_info->nodes[j].num_links;
+					return;
+				}
+			}
+		}
+	}
+}
+
 int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 {
 	struct psp_xgmi_topology_info *top_info;
@@ -897,18 +919,38 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
 				goto exit_unlock;
 		}
 
-		/* get latest topology info for each device from psp */
+		if (amdgpu_sriov_vf(adev) &&
-		list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
+			adev->psp.xgmi_context.xgmi_ta_caps & EXTEND_PEER_LINK_INFO_CMD_FLAG) {
-			ret = psp_xgmi_get_topology_info(&tmp_adev->psp, count,
+			/* only get topology for VF being init if it can support full duplex */
-					&tmp_adev->psp.xgmi_context.top_info, false);
+			ret = psp_xgmi_get_topology_info(&adev->psp, count,
+						&adev->psp.xgmi_context.top_info, false);
 			if (ret) {
-				dev_err(tmp_adev->dev,
+				dev_err(adev->dev,
 					"XGMI: Get topology failure on device %llx, hive %llx, ret %d",
-					tmp_adev->gmc.xgmi.node_id,
+					adev->gmc.xgmi.node_id,
-					tmp_adev->gmc.xgmi.hive_id, ret);
+					adev->gmc.xgmi.hive_id, ret);
-				/* To do : continue with some node failed or disable the whole hive */
+				/* To do: continue with some node failed or disable the whole hive*/
 				goto exit_unlock;
 			}
+
+			/* fill the topology info for peers instead of getting from PSP */
+			list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
+				amdgpu_xgmi_fill_topology_info(adev, tmp_adev);
+			}
+		} else {
+			/* get latest topology info for each device from psp */
+			list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
+				ret = psp_xgmi_get_topology_info(&tmp_adev->psp, count,
+					&tmp_adev->psp.xgmi_context.top_info, false);
+				if (ret) {
+					dev_err(tmp_adev->dev,
+						"XGMI: Get topology failure on device %llx, hive %llx, ret %d",
+						tmp_adev->gmc.xgmi.node_id,
+						tmp_adev->gmc.xgmi.hive_id, ret);
+					/* To do : continue with some node failed or disable the whole hive */
+					goto exit_unlock;
+				}
+			}
 		}
 
 		/* get topology again for hives that support extended data */

drivers/gpu/drm/amd/amdgpu/atombios_encoders.c

@@ -28,6 +28,7 @@
 
 #include <acpi/video.h>
 
+#include <drm/drm_edid.h>
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
 #include "amdgpu_connectors.h"

drivers/gpu/drm/amd/amdgpu/dce_v10_0.c

@@ -21,6 +21,7 @@
  *
  */
 
+#include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_modeset_helper.h>
 #include <drm/drm_modeset_helper_vtables.h>

drivers/gpu/drm/amd/amdgpu/dce_v11_0.c

@@ -21,6 +21,7 @@
  *
  */
 
+#include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_modeset_helper.h>
 #include <drm/drm_modeset_helper_vtables.h>

drivers/gpu/drm/amd/amdgpu/dce_v6_0.c

@@ -23,6 +23,7 @@
 
 #include <linux/pci.h>
 
+#include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_modeset_helper.h>
 #include <drm/drm_modeset_helper_vtables.h>

drivers/gpu/drm/amd/amdgpu/dce_v8_0.c

@@ -21,6 +21,7 @@
  *
  */
 
+#include <drm/drm_edid.h>
 #include <drm/drm_fourcc.h>
 #include <drm/drm_modeset_helper.h>
 #include <drm/drm_modeset_helper_vtables.h>

drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c

@@ -6593,7 +6593,8 @@ static int gfx_v10_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, ENDIAN_SWAP, 1);
 #endif
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 0);
-	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH, 0);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH,
+			    prop->allow_tunneling);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
 	mqd->cp_hqd_pq_control = tmp;

drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c

@@ -3847,7 +3847,8 @@ static int gfx_v11_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, RPTR_BLOCK_SIZE,
 			    (order_base_2(AMDGPU_GPU_PAGE_SIZE / 4) - 1));
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 0);
-	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH, 0);
+	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH,
+			    prop->allow_tunneling);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
 	mqd->cp_hqd_pq_control = tmp;

drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c

@@ -883,7 +883,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	 * GRBM interface.
 	 */
 	if ((vmhub == AMDGPU_GFXHUB(0)) &&
-	    (adev->ip_versions[GC_HWIP][0] < IP_VERSION(9, 4, 2)))
+	    (amdgpu_ip_version(adev, GC_HWIP, 0) < IP_VERSION(9, 4, 2)))
 		RREG32_NO_KIQ(req);
 
 	for (j = 0; j < adev->usec_timeout; j++) {

drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c

@@ -155,13 +155,6 @@ static int jpeg_v4_0_5_hw_init(void *handle)
 	struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec;
 	int r;
 
-	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-				(adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0);
-
-	WREG32_SOC15(VCN, 0, regVCN_JPEG_DB_CTRL,
-		ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
-		VCN_JPEG_DB_CTRL__EN_MASK);
-
 	r = amdgpu_ring_test_helper(ring);
 	if (r)
 		return r;
@@ -336,6 +329,14 @@ static int jpeg_v4_0_5_start(struct amdgpu_device *adev)
 	if (adev->pm.dpm_enabled)
 		amdgpu_dpm_enable_jpeg(adev, true);
 
+	/* doorbell programming is done for every playback */
+	adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+				(adev->doorbell_index.vcn.vcn_ring0_1 << 1), 0);
+
+	WREG32_SOC15(VCN, 0, regVCN_JPEG_DB_CTRL,
+		ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT |
+		VCN_JPEG_DB_CTRL__EN_MASK);
+
 	/* disable power gating */
 	r = jpeg_v4_0_5_disable_static_power_gating(adev);
 	if (r)

drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c

@@ -813,12 +813,12 @@ static int sdma_v2_4_early_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	int r;
 
+	adev->sdma.num_instances = SDMA_MAX_INSTANCE;
+
 	r = sdma_v2_4_init_microcode(adev);
 	if (r)
 		return r;
 
-	adev->sdma.num_instances = SDMA_MAX_INSTANCE;
-
 	sdma_v2_4_set_ring_funcs(adev);
 	sdma_v2_4_set_buffer_funcs(adev);
 	sdma_v2_4_set_vm_pte_funcs(adev);

drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c

@@ -1643,6 +1643,32 @@ static void sdma_v5_2_get_clockgating_state(void *handle, u64 *flags)
 		*flags |= AMD_CG_SUPPORT_SDMA_LS;
 }
 
+static void sdma_v5_2_ring_begin_use(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	/* SDMA 5.2.3 (RMB) FW doesn't seem to properly
+	 * disallow GFXOFF in some cases leading to
+	 * hangs in SDMA.  Disallow GFXOFF while SDMA is active.
+	 * We can probably just limit this to 5.2.3,
+	 * but it shouldn't hurt for other parts since
+	 * this GFXOFF will be disallowed anyway when SDMA is
+	 * active, this just makes it explicit.
+	 */
+	amdgpu_gfx_off_ctrl(adev, false);
+}
+
+static void sdma_v5_2_ring_end_use(struct amdgpu_ring *ring)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	/* SDMA 5.2.3 (RMB) FW doesn't seem to properly
+	 * disallow GFXOFF in some cases leading to
+	 * hangs in SDMA.  Allow GFXOFF when SDMA is complete.
+	 */
+	amdgpu_gfx_off_ctrl(adev, true);
+}
+
 const struct amd_ip_funcs sdma_v5_2_ip_funcs = {
 	.name = "sdma_v5_2",
 	.early_init = sdma_v5_2_early_init,
@@ -1690,6 +1716,8 @@ static const struct amdgpu_ring_funcs sdma_v5_2_ring_funcs = {
 	.test_ib = sdma_v5_2_ring_test_ib,
 	.insert_nop = sdma_v5_2_ring_insert_nop,
 	.pad_ib = sdma_v5_2_ring_pad_ib,
+	.begin_use = sdma_v5_2_ring_begin_use,
+	.end_use = sdma_v5_2_ring_end_use,
 	.emit_wreg = sdma_v5_2_ring_emit_wreg,
 	.emit_reg_wait = sdma_v5_2_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = sdma_v5_2_ring_emit_reg_write_reg_wait,

drivers/gpu/drm/amd/amdgpu/vpe_v6_1.c

@@ -96,6 +96,10 @@ static int vpe_v6_1_load_microcode(struct amdgpu_vpe *vpe)
 		adev->vpe.cmdbuf_cpu_addr[1] = f32_cntl;
 
 		amdgpu_vpe_psp_update_sram(adev);
+
+		/* Config DPM */
+		amdgpu_vpe_configure_dpm(vpe);
+
 		return 0;
 	}
 
@@ -128,6 +132,8 @@ static int vpe_v6_1_load_microcode(struct amdgpu_vpe *vpe)
 	}
 
 	vpe_v6_1_halt(vpe, false);
+	/* Config DPM */
+	amdgpu_vpe_configure_dpm(vpe);
 
 	return 0;
 }
@@ -264,6 +270,15 @@ static int vpe_v6_1_set_regs(struct amdgpu_vpe *vpe)
 	vpe->regs.queue0_rb_wptr_hi = regVPEC_QUEUE0_RB_WPTR_HI;
 	vpe->regs.queue0_preempt = regVPEC_QUEUE0_PREEMPT;
 
+	vpe->regs.dpm_enable = regVPEC_PUB_DUMMY2;
+	vpe->regs.dpm_pratio = regVPEC_QUEUE6_DUMMY4;
+	vpe->regs.dpm_request_interval = regVPEC_QUEUE5_DUMMY3;
+	vpe->regs.dpm_decision_threshold = regVPEC_QUEUE5_DUMMY4;
+	vpe->regs.dpm_busy_clamp_threshold = regVPEC_QUEUE7_DUMMY2;
+	vpe->regs.dpm_idle_clamp_threshold = regVPEC_QUEUE7_DUMMY3;
+	vpe->regs.dpm_request_lv = regVPEC_QUEUE7_DUMMY1;
+	vpe->regs.context_indicator = regVPEC_QUEUE6_DUMMY3;
+
 	return 0;
 }
 

drivers/gpu/drm/amd/amdkfd/kfd_chardev.c

@@ -1564,16 +1564,11 @@ static int kfd_ioctl_import_dmabuf(struct file *filep,
 {
 	struct kfd_ioctl_import_dmabuf_args *args = data;
 	struct kfd_process_device *pdd;
-	struct dma_buf *dmabuf;
 	int idr_handle;
 	uint64_t size;
 	void *mem;
 	int r;
 
-	dmabuf = dma_buf_get(args->dmabuf_fd);
-	if (IS_ERR(dmabuf))
-		return PTR_ERR(dmabuf);
-
 	mutex_lock(&p->mutex);
 	pdd = kfd_process_device_data_by_id(p, args->gpu_id);
 	if (!pdd) {
@@ -1587,10 +1582,10 @@ static int kfd_ioctl_import_dmabuf(struct file *filep,
 		goto err_unlock;
 	}
 
-	r = amdgpu_amdkfd_gpuvm_import_dmabuf(pdd->dev->adev, dmabuf,
+	r = amdgpu_amdkfd_gpuvm_import_dmabuf_fd(pdd->dev->adev, args->dmabuf_fd,
-					      args->va_addr, pdd->drm_priv,
+						 args->va_addr, pdd->drm_priv,
-					      (struct kgd_mem **)&mem, &size,
+						 (struct kgd_mem **)&mem, &size,
-					      NULL);
+						 NULL);
 	if (r)
 		goto err_unlock;
 
@@ -1601,7 +1596,6 @@ static int kfd_ioctl_import_dmabuf(struct file *filep,
 	}
 
 	mutex_unlock(&p->mutex);
-	dma_buf_put(dmabuf);
 
 	args->handle = MAKE_HANDLE(args->gpu_id, idr_handle);
 
@@ -1612,7 +1606,6 @@ err_free:
 					       pdd->drm_priv, NULL);
 err_unlock:
 	mutex_unlock(&p->mutex);
-	dma_buf_put(dmabuf);
 	return r;
 }
 
@@ -1855,8 +1848,8 @@ static uint32_t get_process_num_bos(struct kfd_process *p)
 	return num_of_bos;
 }
 
-static int criu_get_prime_handle(struct kgd_mem *mem, int flags,
+static int criu_get_prime_handle(struct kgd_mem *mem,
-				      u32 *shared_fd)
+				 int flags, u32 *shared_fd)
 {
 	struct dma_buf *dmabuf;
 	int ret;

drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c

@@ -87,6 +87,8 @@ void kfd_process_dequeue_from_device(struct kfd_process_device *pdd)
 		return;
 
 	dev->dqm->ops.process_termination(dev->dqm, &pdd->qpd);
+	if (dev->kfd->shared_resources.enable_mes)
+		amdgpu_mes_flush_shader_debugger(dev->adev, pdd->proc_ctx_gpu_addr);
 	pdd->already_dequeued = true;
 }
 

drivers/gpu/drm/amd/amdkfd/kfd_svm.c

@@ -1607,18 +1607,24 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 			if (test_bit(gpuidx, prange->bitmap_access))
 				bitmap_set(ctx->bitmap, gpuidx, 1);
 		}
+
+		/*
+		 * If prange is already mapped or with always mapped flag,
+		 * update mapping on GPUs with ACCESS attribute
+		 */
+		if (bitmap_empty(ctx->bitmap, MAX_GPU_INSTANCE)) {
+			if (prange->mapped_to_gpu ||
+			    prange->flags & KFD_IOCTL_SVM_FLAG_GPU_ALWAYS_MAPPED)
+				bitmap_copy(ctx->bitmap, prange->bitmap_access, MAX_GPU_INSTANCE);
+		}
 	} else {
 		bitmap_or(ctx->bitmap, prange->bitmap_access,
 			  prange->bitmap_aip, MAX_GPU_INSTANCE);
 	}
 
 	if (bitmap_empty(ctx->bitmap, MAX_GPU_INSTANCE)) {
-		bitmap_copy(ctx->bitmap, prange->bitmap_access, MAX_GPU_INSTANCE);
+		r = 0;
-		if (!prange->mapped_to_gpu ||
+		goto free_ctx;
-		    bitmap_empty(ctx->bitmap, MAX_GPU_INSTANCE)) {
-			r = 0;
-			goto free_ctx;
-		}
 	}
 
 	if (prange->actual_loc && !prange->ttm_res) {

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

@@ -1712,7 +1712,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	init_data.clk_reg_offsets = adev->reg_offset[CLK_HWIP][0];
 
 	/* Enable DWB for tested platforms only */
-	if (adev->ip_versions[DCE_HWIP][0] >= IP_VERSION(3, 0, 0))
+	if (amdgpu_ip_version(adev, DCE_HWIP, 0) >= IP_VERSION(3, 0, 0))
 		init_data.num_virtual_links = 1;
 
 	INIT_LIST_HEAD(&adev->dm.da_list);
@@ -2687,6 +2687,7 @@ static int dm_suspend(void *handle)
 	hpd_rx_irq_work_suspend(dm);
 
 	dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D3);
+	dc_dmub_srv_set_power_state(dm->dc->ctx->dmub_srv, DC_ACPI_CM_POWER_STATE_D3);
 
 	return 0;
 }
@@ -2882,6 +2883,7 @@ static int dm_resume(void *handle)
 		if (r)
 			DRM_ERROR("DMUB interface failed to initialize: status=%d\n", r);
 
+		dc_dmub_srv_set_power_state(dm->dc->ctx->dmub_srv, DC_ACPI_CM_POWER_STATE_D0);
 		dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D0);
 
 		dc_resume(dm->dc);
@@ -2932,6 +2934,7 @@ static int dm_resume(void *handle)
 	}
 
 	/* power on hardware */
+	dc_dmub_srv_set_power_state(dm->dc->ctx->dmub_srv, DC_ACPI_CM_POWER_STATE_D0);
 	dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D0);
 
 	/* program HPD filter */
@@ -4067,6 +4070,11 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev)
 		return r;
 	}
 
+#ifdef AMD_PRIVATE_COLOR
+	if (amdgpu_dm_create_color_properties(adev))
+		return -ENOMEM;
+#endif
+
 	r = amdgpu_dm_audio_init(adev);
 	if (r) {
 		dc_release_state(state->context);
@@ -5164,7 +5172,9 @@ static int fill_dc_plane_attributes(struct amdgpu_device *adev,
 	 * Always set input transfer function, since plane state is refreshed
 	 * every time.
 	 */
-	ret = amdgpu_dm_update_plane_color_mgmt(dm_crtc_state, dc_plane_state);
+	ret = amdgpu_dm_update_plane_color_mgmt(dm_crtc_state,
+						plane_state,
+						dc_plane_state);
 	if (ret)
 		return ret;
 
@@ -8261,6 +8271,10 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 			bundle->surface_updates[planes_count].gamma = dc_plane->gamma_correction;
 			bundle->surface_updates[planes_count].in_transfer_func = dc_plane->in_transfer_func;
 			bundle->surface_updates[planes_count].gamut_remap_matrix = &dc_plane->gamut_remap_matrix;
+			bundle->surface_updates[planes_count].hdr_mult = dc_plane->hdr_mult;
+			bundle->surface_updates[planes_count].func_shaper = dc_plane->in_shaper_func;
+			bundle->surface_updates[planes_count].lut3d_func = dc_plane->lut3d_func;
+			bundle->surface_updates[planes_count].blend_tf = dc_plane->blend_tf;
 		}
 
 		amdgpu_dm_plane_fill_dc_scaling_info(dm->adev, new_plane_state,
@@ -8472,6 +8486,10 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 				&acrtc_state->stream->csc_color_matrix;
 			bundle->stream_update.out_transfer_func =
 				acrtc_state->stream->out_transfer_func;
+			bundle->stream_update.lut3d_func =
+				(struct dc_3dlut *) acrtc_state->stream->lut3d_func;
+			bundle->stream_update.func_shaper =
+				(struct dc_transfer_func *) acrtc_state->stream->func_shaper;
 		}
 
 		acrtc_state->stream->abm_level = acrtc_state->abm_level;
@@ -8874,12 +8892,14 @@ static void dm_set_writeback(struct amdgpu_display_manager *dm,
 	acrtc = to_amdgpu_crtc(wb_conn->encoder.crtc);
 	if (!acrtc) {
 		DRM_ERROR("no amdgpu_crtc found\n");
+		kfree(wb_info);
 		return;
 	}
 
 	afb = to_amdgpu_framebuffer(new_con_state->writeback_job->fb);
 	if (!afb) {
 		DRM_ERROR("No amdgpu_framebuffer found\n");
+		kfree(wb_info);
 		return;
 	}
 
@@ -8934,7 +8954,7 @@ static void dm_set_writeback(struct amdgpu_display_manager *dm,
 	}
 
 	wb_info->mcif_buf_params.p_vmid = 1;
-	if (adev->ip_versions[DCE_HWIP][0] >= IP_VERSION(3, 0, 0)) {
+	if (amdgpu_ip_version(adev, DCE_HWIP, 0) >= IP_VERSION(3, 0, 0)) {
 		wb_info->mcif_warmup_params.start_address.quad_part = afb->address;
 		wb_info->mcif_warmup_params.region_size =
 			wb_info->mcif_buf_params.luma_pitch * wb_info->dwb_params.dest_height;
@@ -9853,6 +9873,7 @@ skip_modeset:
 	 * when a modeset is needed, to ensure it gets reprogrammed.
 	 */
 	if (dm_new_crtc_state->base.color_mgmt_changed ||
+	    dm_old_crtc_state->regamma_tf != dm_new_crtc_state->regamma_tf ||
 	    drm_atomic_crtc_needs_modeset(new_crtc_state)) {
 		ret = amdgpu_dm_update_crtc_color_mgmt(dm_new_crtc_state);
 		if (ret)
@@ -9886,7 +9907,8 @@ static bool should_reset_plane(struct drm_atomic_state *state,
 	 * TODO: Remove this hack for all asics once it proves that the
 	 * fast updates works fine on DCN3.2+.
 	 */
-	if (adev->ip_versions[DCE_HWIP][0] < IP_VERSION(3, 2, 0) && state->allow_modeset)
+	if (amdgpu_ip_version(adev, DCE_HWIP, 0) < IP_VERSION(3, 2, 0) &&
+	    state->allow_modeset)
 		return true;
 
 	/* Exit early if we know that we're adding or removing the plane. */
@@ -9920,6 +9942,10 @@ static bool should_reset_plane(struct drm_atomic_state *state,
 	 */
 	for_each_oldnew_plane_in_state(state, other, old_other_state, new_other_state, i) {
 		struct amdgpu_framebuffer *old_afb, *new_afb;
+		struct dm_plane_state *dm_new_other_state, *dm_old_other_state;
+
+		dm_new_other_state = to_dm_plane_state(new_other_state);
+		dm_old_other_state = to_dm_plane_state(old_other_state);
 
 		if (other->type == DRM_PLANE_TYPE_CURSOR)
 			continue;
@@ -9956,6 +9982,18 @@ static bool should_reset_plane(struct drm_atomic_state *state,
 		    old_other_state->color_encoding != new_other_state->color_encoding)
 			return true;
 
+		/* HDR/Transfer Function changes. */
+		if (dm_old_other_state->degamma_tf != dm_new_other_state->degamma_tf ||
+		    dm_old_other_state->degamma_lut != dm_new_other_state->degamma_lut ||
+		    dm_old_other_state->hdr_mult != dm_new_other_state->hdr_mult ||
+		    dm_old_other_state->ctm != dm_new_other_state->ctm ||
+		    dm_old_other_state->shaper_lut != dm_new_other_state->shaper_lut ||
+		    dm_old_other_state->shaper_tf != dm_new_other_state->shaper_tf ||
+		    dm_old_other_state->lut3d != dm_new_other_state->lut3d ||
+		    dm_old_other_state->blend_lut != dm_new_other_state->blend_lut ||
+		    dm_old_other_state->blend_tf != dm_new_other_state->blend_tf)
+			return true;
+
 		/* Framebuffer checks fall at the end. */
 		if (!old_other_state->fb || !new_other_state->fb)
 			continue;

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h

@@ -55,6 +55,9 @@
 #define HDMI_AMD_VENDOR_SPECIFIC_DATA_BLOCK_IEEE_REGISTRATION_ID 0x00001A
 #define AMD_VSDB_VERSION_3_FEATURECAP_REPLAYMODE 0x40
 #define HDMI_AMD_VENDOR_SPECIFIC_DATA_BLOCK_VERSION_3 0x3
+
+#define AMDGPU_HDR_MULT_DEFAULT (0x100000000LL)
+
 /*
 #include "include/amdgpu_dal_power_if.h"
 #include "amdgpu_dm_irq.h"
@@ -724,9 +727,98 @@ struct amdgpu_dm_wb_connector {
 
 extern const struct amdgpu_ip_block_version dm_ip_block;
 
+/* enum amdgpu_transfer_function: pre-defined transfer function supported by AMD.
+ *
+ * It includes standardized transfer functions and pure power functions. The
+ * transfer function coefficients are available at modules/color/color_gamma.c
+ */
+enum amdgpu_transfer_function {
+	AMDGPU_TRANSFER_FUNCTION_DEFAULT,
+	AMDGPU_TRANSFER_FUNCTION_SRGB_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_BT709_INV_OETF,
+	AMDGPU_TRANSFER_FUNCTION_PQ_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_IDENTITY,
+	AMDGPU_TRANSFER_FUNCTION_GAMMA22_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_GAMMA24_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_GAMMA26_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_SRGB_INV_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_BT709_OETF,
+	AMDGPU_TRANSFER_FUNCTION_PQ_INV_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_GAMMA22_INV_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_GAMMA24_INV_EOTF,
+	AMDGPU_TRANSFER_FUNCTION_GAMMA26_INV_EOTF,
+        AMDGPU_TRANSFER_FUNCTION_COUNT
+};
+
 struct dm_plane_state {
 	struct drm_plane_state base;
 	struct dc_plane_state *dc_state;
+
+	/* Plane color mgmt */
+	/**
+	 * @degamma_lut:
+	 *
+	 * 1D LUT for mapping framebuffer/plane pixel data before sampling or
+	 * blending operations. It's usually applied to linearize input space.
+	 * The blob (if not NULL) is an array of &struct drm_color_lut.
+	 */
+	struct drm_property_blob *degamma_lut;
+	/**
+	 * @degamma_tf:
+	 *
+	 * Predefined transfer function to tell DC driver the input space to
+	 * linearize.
+	 */
+	enum amdgpu_transfer_function degamma_tf;
+	/**
+	 * @hdr_mult:
+	 *
+	 * Multiplier to 'gain' the plane.  When PQ is decoded using the fixed
+	 * func transfer function to the internal FP16 fb, 1.0 -> 80 nits (on
+	 * AMD at least). When sRGB is decoded, 1.0 -> 1.0, obviously.
+	 * Therefore, 1.0 multiplier = 80 nits for SDR content.  So if you
+	 * want, 203 nits for SDR content, pass in (203.0 / 80.0).  Format is
+	 * S31.32 sign-magnitude.
+	 *
+	 * HDR multiplier can wide range beyond [0.0, 1.0]. This means that PQ
+	 * TF is needed for any subsequent linear-to-non-linear transforms.
+	 */
+	__u64 hdr_mult;
+	/**
+	 * @ctm:
+	 *
+	 * Color transformation matrix. The blob (if not NULL) is a &struct
+	 * drm_color_ctm_3x4.
+	 */
+	struct drm_property_blob *ctm;
+	/**
+	 * @shaper_lut: shaper lookup table blob. The blob (if not NULL) is an
+	 * array of &struct drm_color_lut.
+	 */
+	struct drm_property_blob *shaper_lut;
+	/**
+	 * @shaper_tf:
+	 *
+	 * Predefined transfer function to delinearize color space.
+	 */
+	enum amdgpu_transfer_function shaper_tf;
+	/**
+	 * @lut3d: 3D lookup table blob. The blob (if not NULL) is an array of
+	 * &struct drm_color_lut.
+	 */
+	struct drm_property_blob *lut3d;
+	/**
+	 * @blend_lut: blend lut lookup table blob. The blob (if not NULL) is an
+	 * array of &struct drm_color_lut.
+	 */
+	struct drm_property_blob *blend_lut;
+	/**
+	 * @blend_tf:
+	 *
+	 * Pre-defined transfer function for converting plane pixel data before
+	 * applying blend LUT.
+	 */
+	enum amdgpu_transfer_function blend_tf;
 };
 
 struct dm_crtc_state {
@@ -751,6 +843,14 @@ struct dm_crtc_state {
 	struct dc_info_packet vrr_infopacket;
 
 	int abm_level;
+
+        /**
+	 * @regamma_tf:
+	 *
+	 * Pre-defined transfer function for converting internal FB -> wire
+	 * encoding.
+	 */
+	enum amdgpu_transfer_function regamma_tf;
 };
 
 #define to_dm_crtc_state(x) container_of(x, struct dm_crtc_state, base)
@@ -812,14 +912,22 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector,
 
 void amdgpu_dm_trigger_timing_sync(struct drm_device *dev);
 
+/* 3D LUT max size is 17x17x17 (4913 entries) */
+#define MAX_COLOR_3DLUT_SIZE 17
+#define MAX_COLOR_3DLUT_BITDEPTH 12
+int amdgpu_dm_verify_lut3d_size(struct amdgpu_device *adev,
+				struct drm_plane_state *plane_state);
+/* 1D LUT size */
 #define MAX_COLOR_LUT_ENTRIES 4096
 /* Legacy gamm LUT users such as X doesn't like large LUT sizes */
 #define MAX_COLOR_LEGACY_LUT_ENTRIES 256
 
 void amdgpu_dm_init_color_mod(void);
+int amdgpu_dm_create_color_properties(struct amdgpu_device *adev);
 int amdgpu_dm_verify_lut_sizes(const struct drm_crtc_state *crtc_state);
 int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc);
 int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc,
+				      struct drm_plane_state *plane_state,
 				      struct dc_plane_state *dc_plane_state);
 
 void amdgpu_dm_update_connector_after_detect(

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c

@@ -72,6 +72,7 @@
  */
 
 #define MAX_DRM_LUT_VALUE 0xFFFF
+#define SDR_WHITE_LEVEL_INIT_VALUE 80
 
 /**
  * amdgpu_dm_init_color_mod - Initialize the color module.
@@ -84,6 +85,235 @@ void amdgpu_dm_init_color_mod(void)
 	setup_x_points_distribution();
 }
 
+#ifdef AMD_PRIVATE_COLOR
+/* Pre-defined Transfer Functions (TF)
+ *
+ * AMD driver supports pre-defined mathematical functions for transferring
+ * between encoded values and optical/linear space. Depending on HW color caps,
+ * ROMs and curves built by the AMD color module support these transforms.
+ *
+ * The driver-specific color implementation exposes properties for pre-blending
+ * degamma TF, shaper TF (before 3D LUT), and blend(dpp.ogam) TF and
+ * post-blending regamma (mpc.ogam) TF. However, only pre-blending degamma
+ * supports ROM curves. AMD color module uses pre-defined coefficients to build
+ * curves for the other blocks. What can be done by each color block is
+ * described by struct dpp_color_capsand struct mpc_color_caps.
+ *
+ * AMD driver-specific color API exposes the following pre-defined transfer
+ * functions:
+ *
+ * - Identity: linear/identity relationship between pixel value and
+ *   luminance value;
+ * - Gamma 2.2, Gamma 2.4, Gamma 2.6: pure power functions;
+ * - sRGB: 2.4: The piece-wise transfer function from IEC 61966-2-1:1999;
+ * - BT.709: has a linear segment in the bottom part and then a power function
+ *   with a 0.45 (~1/2.22) gamma for the rest of the range; standardized by
+ *   ITU-R BT.709-6;
+ * - PQ (Perceptual Quantizer): used for HDR display, allows luminance range
+ *   capability of 0 to 10,000 nits; standardized by SMPTE ST 2084.
+ *
+ * The AMD color model is designed with an assumption that SDR (sRGB, BT.709,
+ * Gamma 2.2, etc.) peak white maps (normalized to 1.0 FP) to 80 nits in the PQ
+ * system. This has the implication that PQ EOTF (non-linear to linear) maps to
+ * [0.0..125.0] where 125.0 = 10,000 nits / 80 nits.
+ *
+ * Non-linear and linear forms are described in the table below:
+ *
+ * ┌───────────┬─────────────────────┬──────────────────────┐
+ * │           │     Non-linear      │   Linear             │
+ * ├───────────┼─────────────────────┼──────────────────────┤
+ * │      sRGB │ UNORM or [0.0, 1.0] │ [0.0, 1.0]           │
+ * ├───────────┼─────────────────────┼──────────────────────┤
+ * │     BT709 │ UNORM or [0.0, 1.0] │ [0.0, 1.0]           │
+ * ├───────────┼─────────────────────┼──────────────────────┤
+ * │ Gamma 2.x │ UNORM or [0.0, 1.0] │ [0.0, 1.0]           │
+ * ├───────────┼─────────────────────┼──────────────────────┤
+ * │        PQ │ UNORM or FP16 CCCS* │ [0.0, 125.0]         │
+ * ├───────────┼─────────────────────┼──────────────────────┤
+ * │  Identity │ UNORM or FP16 CCCS* │ [0.0, 1.0] or CCCS** │
+ * └───────────┴─────────────────────┴──────────────────────┘
+ * * CCCS: Windows canonical composition color space
+ * ** Respectively
+ *
+ * In the driver-specific API, color block names attached to TF properties
+ * suggest the intention regarding non-linear encoding pixel's luminance
+ * values. As some newer encodings don't use gamma curve, we make encoding and
+ * decoding explicit by defining an enum list of transfer functions supported
+ * in terms of EOTF and inverse EOTF, where:
+ *
+ * - EOTF (electro-optical transfer function): is the transfer function to go
+ *   from the encoded value to an optical (linear) value. De-gamma functions
+ *   traditionally do this.
+ * - Inverse EOTF (simply the inverse of the EOTF): is usually intended to go
+ *   from an optical/linear space (which might have been used for blending)
+ *   back to the encoded values. Gamma functions traditionally do this.
+ */
+static const char * const
+amdgpu_transfer_function_names[] = {
+	[AMDGPU_TRANSFER_FUNCTION_DEFAULT]		= "Default",
+	[AMDGPU_TRANSFER_FUNCTION_IDENTITY]		= "Identity",
+	[AMDGPU_TRANSFER_FUNCTION_SRGB_EOTF]		= "sRGB EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_BT709_INV_OETF]	= "BT.709 inv_OETF",
+	[AMDGPU_TRANSFER_FUNCTION_PQ_EOTF]		= "PQ EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_GAMMA22_EOTF]		= "Gamma 2.2 EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_GAMMA24_EOTF]		= "Gamma 2.4 EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_GAMMA26_EOTF]		= "Gamma 2.6 EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_SRGB_INV_EOTF]	= "sRGB inv_EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_BT709_OETF]		= "BT.709 OETF",
+	[AMDGPU_TRANSFER_FUNCTION_PQ_INV_EOTF]		= "PQ inv_EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_GAMMA22_INV_EOTF]	= "Gamma 2.2 inv_EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_GAMMA24_INV_EOTF]	= "Gamma 2.4 inv_EOTF",
+	[AMDGPU_TRANSFER_FUNCTION_GAMMA26_INV_EOTF]	= "Gamma 2.6 inv_EOTF",
+};
+
+static const u32 amdgpu_eotf =
+	BIT(AMDGPU_TRANSFER_FUNCTION_SRGB_EOTF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_BT709_INV_OETF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_PQ_EOTF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_GAMMA22_EOTF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_GAMMA24_EOTF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_GAMMA26_EOTF);
+
+static const u32 amdgpu_inv_eotf =
+	BIT(AMDGPU_TRANSFER_FUNCTION_SRGB_INV_EOTF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_BT709_OETF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_PQ_INV_EOTF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_GAMMA22_INV_EOTF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_GAMMA24_INV_EOTF) |
+	BIT(AMDGPU_TRANSFER_FUNCTION_GAMMA26_INV_EOTF);
+
+static struct drm_property *
+amdgpu_create_tf_property(struct drm_device *dev,
+			  const char *name,
+			  u32 supported_tf)
+{
+	u32 transfer_functions = supported_tf |
+				 BIT(AMDGPU_TRANSFER_FUNCTION_DEFAULT) |
+				 BIT(AMDGPU_TRANSFER_FUNCTION_IDENTITY);
+	struct drm_prop_enum_list enum_list[AMDGPU_TRANSFER_FUNCTION_COUNT];
+	int i, len;
+
+	len = 0;
+	for (i = 0; i < AMDGPU_TRANSFER_FUNCTION_COUNT; i++) {
+		if ((transfer_functions & BIT(i)) == 0)
+			continue;
+
+		enum_list[len].type = i;
+		enum_list[len].name = amdgpu_transfer_function_names[i];
+		len++;
+	}
+
+	return drm_property_create_enum(dev, DRM_MODE_PROP_ENUM,
+					name, enum_list, len);
+}
+
+int
+amdgpu_dm_create_color_properties(struct amdgpu_device *adev)
+{
+	struct drm_property *prop;
+
+	prop = drm_property_create(adev_to_drm(adev),
+				   DRM_MODE_PROP_BLOB,
+				   "AMD_PLANE_DEGAMMA_LUT", 0);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_degamma_lut_property = prop;
+
+	prop = drm_property_create_range(adev_to_drm(adev),
+					 DRM_MODE_PROP_IMMUTABLE,
+					 "AMD_PLANE_DEGAMMA_LUT_SIZE",
+					 0, UINT_MAX);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_degamma_lut_size_property = prop;
+
+	prop = amdgpu_create_tf_property(adev_to_drm(adev),
+					 "AMD_PLANE_DEGAMMA_TF",
+					 amdgpu_eotf);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_degamma_tf_property = prop;
+
+	prop = drm_property_create_range(adev_to_drm(adev),
+					 0, "AMD_PLANE_HDR_MULT", 0, U64_MAX);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_hdr_mult_property = prop;
+
+	prop = drm_property_create(adev_to_drm(adev),
+				   DRM_MODE_PROP_BLOB,
+				   "AMD_PLANE_CTM", 0);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_ctm_property = prop;
+
+	prop = drm_property_create(adev_to_drm(adev),
+				   DRM_MODE_PROP_BLOB,
+				   "AMD_PLANE_SHAPER_LUT", 0);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_shaper_lut_property = prop;
+
+	prop = drm_property_create_range(adev_to_drm(adev),
+					 DRM_MODE_PROP_IMMUTABLE,
+					 "AMD_PLANE_SHAPER_LUT_SIZE", 0, UINT_MAX);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_shaper_lut_size_property = prop;
+
+	prop = amdgpu_create_tf_property(adev_to_drm(adev),
+					 "AMD_PLANE_SHAPER_TF",
+					 amdgpu_inv_eotf);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_shaper_tf_property = prop;
+
+	prop = drm_property_create(adev_to_drm(adev),
+				   DRM_MODE_PROP_BLOB,
+				   "AMD_PLANE_LUT3D", 0);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_lut3d_property = prop;
+
+	prop = drm_property_create_range(adev_to_drm(adev),
+					 DRM_MODE_PROP_IMMUTABLE,
+					 "AMD_PLANE_LUT3D_SIZE", 0, UINT_MAX);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_lut3d_size_property = prop;
+
+	prop = drm_property_create(adev_to_drm(adev),
+				   DRM_MODE_PROP_BLOB,
+				   "AMD_PLANE_BLEND_LUT", 0);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_blend_lut_property = prop;
+
+	prop = drm_property_create_range(adev_to_drm(adev),
+					 DRM_MODE_PROP_IMMUTABLE,
+					 "AMD_PLANE_BLEND_LUT_SIZE", 0, UINT_MAX);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_blend_lut_size_property = prop;
+
+	prop = amdgpu_create_tf_property(adev_to_drm(adev),
+					 "AMD_PLANE_BLEND_TF",
+					 amdgpu_eotf);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.plane_blend_tf_property = prop;
+
+	prop = amdgpu_create_tf_property(adev_to_drm(adev),
+					 "AMD_CRTC_REGAMMA_TF",
+					 amdgpu_inv_eotf);
+	if (!prop)
+		return -ENOMEM;
+	adev->mode_info.regamma_tf_property = prop;
+
+	return 0;
+}
+#endif
+
 /**
  * __extract_blob_lut - Extracts the DRM lut and lut size from a blob.
  * @blob: DRM color mgmt property blob
@@ -182,7 +412,6 @@ static void __drm_lut_to_dc_gamma(const struct drm_color_lut *lut,
 static void __drm_ctm_to_dc_matrix(const struct drm_color_ctm *ctm,
 				   struct fixed31_32 *matrix)
 {
-	int64_t val;
 	int i;
 
 	/*
@@ -201,12 +430,29 @@ static void __drm_ctm_to_dc_matrix(const struct drm_color_ctm *ctm,
 		}
 
 		/* gamut_remap_matrix[i] = ctm[i - floor(i/4)] */
-		val = ctm->matrix[i - (i / 4)];
+		matrix[i] = dc_fixpt_from_s3132(ctm->matrix[i - (i / 4)]);
-		/* If negative, convert to 2's complement. */
+	}
-		if (val & (1ULL << 63))
+}
-			val = -(val & ~(1ULL << 63));
 
-		matrix[i].value = val;
+/**
+ * __drm_ctm_3x4_to_dc_matrix - converts a DRM CTM 3x4 to a DC CSC float matrix
+ * @ctm: DRM color transformation matrix with 3x4 dimensions
+ * @matrix: DC CSC float matrix
+ *
+ * The matrix needs to be a 3x4 (12 entry) matrix.
+ */
+static void __drm_ctm_3x4_to_dc_matrix(const struct drm_color_ctm_3x4 *ctm,
+				       struct fixed31_32 *matrix)
+{
+	int i;
+
+	/* The format provided is S31.32, using signed-magnitude representation.
+	 * Our fixed31_32 is also S31.32, but is using 2's complement. We have
+	 * to convert from signed-magnitude to 2's complement.
+	 */
+	for (i = 0; i < 12; i++) {
+		/* gamut_remap_matrix[i] = ctm[i - floor(i/4)] */
+		matrix[i] = dc_fixpt_from_s3132(ctm->matrix[i]);
 	}
 }
 
@@ -268,16 +514,18 @@ static int __set_output_tf(struct dc_transfer_func *func,
 	struct calculate_buffer cal_buffer = {0};
 	bool res;
 
-	ASSERT(lut && lut_size == MAX_COLOR_LUT_ENTRIES);
-
 	cal_buffer.buffer_index = -1;
 
-	gamma = dc_create_gamma();
+	if (lut_size) {
-	if (!gamma)
+		ASSERT(lut && lut_size == MAX_COLOR_LUT_ENTRIES);
-		return -ENOMEM;
 
-	gamma->num_entries = lut_size;
+		gamma = dc_create_gamma();
-	__drm_lut_to_dc_gamma(lut, gamma, false);
+		if (!gamma)
+			return -ENOMEM;
+
+		gamma->num_entries = lut_size;
+		__drm_lut_to_dc_gamma(lut, gamma, false);
+	}
 
 	if (func->tf == TRANSFER_FUNCTION_LINEAR) {
 		/*
@@ -285,27 +533,68 @@ static int __set_output_tf(struct dc_transfer_func *func,
 		 * on top of a linear input. But degamma params can be used
 		 * instead to simulate this.
 		 */
-		gamma->type = GAMMA_CUSTOM;
+		if (gamma)
+			gamma->type = GAMMA_CUSTOM;
 		res = mod_color_calculate_degamma_params(NULL, func,
-							gamma, true);
+							 gamma, gamma != NULL);
 	} else {
 		/*
 		 * Assume sRGB. The actual mapping will depend on whether the
 		 * input was legacy or not.
 		 */
-		gamma->type = GAMMA_CS_TFM_1D;
+		if (gamma)
-		res = mod_color_calculate_regamma_params(func, gamma, false,
+			gamma->type = GAMMA_CS_TFM_1D;
+		res = mod_color_calculate_regamma_params(func, gamma, gamma != NULL,
 							 has_rom, NULL, &cal_buffer);
 	}
 
-	dc_gamma_release(&gamma);
+	if (gamma)
+		dc_gamma_release(&gamma);
 
 	return res ? 0 : -ENOMEM;
 }
 
+static int amdgpu_dm_set_atomic_regamma(struct dc_stream_state *stream,
+					const struct drm_color_lut *regamma_lut,
+					uint32_t regamma_size, bool has_rom,
+					enum dc_transfer_func_predefined tf)
+{
+	struct dc_transfer_func *out_tf = stream->out_transfer_func;
+	int ret = 0;
+
+	if (regamma_size || tf != TRANSFER_FUNCTION_LINEAR) {
+		/*
+		 * CRTC RGM goes into RGM LUT.
+		 *
+		 * Note: there is no implicit sRGB regamma here. We are using
+		 * degamma calculation from color module to calculate the curve
+		 * from a linear base if gamma TF is not set. However, if gamma
+		 * TF (!= Linear) and LUT are set at the same time, we will use
+		 * regamma calculation, and the color module will combine the
+		 * pre-defined TF and the custom LUT values into the LUT that's
+		 * actually programmed.
+		 */
+		out_tf->type = TF_TYPE_DISTRIBUTED_POINTS;
+		out_tf->tf = tf;
+		out_tf->sdr_ref_white_level = SDR_WHITE_LEVEL_INIT_VALUE;
+
+		ret = __set_output_tf(out_tf, regamma_lut, regamma_size, has_rom);
+	} else {
+		/*
+		 * No CRTC RGM means we can just put the block into bypass
+		 * since we don't have any plane level adjustments using it.
+		 */
+		out_tf->type = TF_TYPE_BYPASS;
+		out_tf->tf = TRANSFER_FUNCTION_LINEAR;
+	}
+
+	return ret;
+}
+
 /**
  * __set_input_tf - calculates the input transfer function based on expected
  * input space.
+ * @caps: dc color capabilities
  * @func: transfer function
  * @lut: lookup table that defines the color space
  * @lut_size: size of respective lut.
@@ -313,27 +602,241 @@ static int __set_output_tf(struct dc_transfer_func *func,
  * Returns:
  * 0 in case of success. -ENOMEM if fails.
  */
-static int __set_input_tf(struct dc_transfer_func *func,
+static int __set_input_tf(struct dc_color_caps *caps, struct dc_transfer_func *func,
 			  const struct drm_color_lut *lut, uint32_t lut_size)
 {
 	struct dc_gamma *gamma = NULL;
 	bool res;
 
-	gamma = dc_create_gamma();
+	if (lut_size) {
-	if (!gamma)
+		gamma = dc_create_gamma();
-		return -ENOMEM;
+		if (!gamma)
+			return -ENOMEM;
 
-	gamma->type = GAMMA_CUSTOM;
+		gamma->type = GAMMA_CUSTOM;
-	gamma->num_entries = lut_size;
+		gamma->num_entries = lut_size;
 
-	__drm_lut_to_dc_gamma(lut, gamma, false);
+		__drm_lut_to_dc_gamma(lut, gamma, false);
+	}
 
-	res = mod_color_calculate_degamma_params(NULL, func, gamma, true);
+	res = mod_color_calculate_degamma_params(caps, func, gamma, gamma != NULL);
-	dc_gamma_release(&gamma);
+
+	if (gamma)
+		dc_gamma_release(&gamma);
 
 	return res ? 0 : -ENOMEM;
 }
 
+static enum dc_transfer_func_predefined
+amdgpu_tf_to_dc_tf(enum amdgpu_transfer_function tf)
+{
+	switch (tf)
+	{
+	default:
+	case AMDGPU_TRANSFER_FUNCTION_DEFAULT:
+	case AMDGPU_TRANSFER_FUNCTION_IDENTITY:
+		return TRANSFER_FUNCTION_LINEAR;
+	case AMDGPU_TRANSFER_FUNCTION_SRGB_EOTF:
+	case AMDGPU_TRANSFER_FUNCTION_SRGB_INV_EOTF:
+		return TRANSFER_FUNCTION_SRGB;
+	case AMDGPU_TRANSFER_FUNCTION_BT709_OETF:
+	case AMDGPU_TRANSFER_FUNCTION_BT709_INV_OETF:
+		return TRANSFER_FUNCTION_BT709;
+	case AMDGPU_TRANSFER_FUNCTION_PQ_EOTF:
+	case AMDGPU_TRANSFER_FUNCTION_PQ_INV_EOTF:
+		return TRANSFER_FUNCTION_PQ;
+	case AMDGPU_TRANSFER_FUNCTION_GAMMA22_EOTF:
+	case AMDGPU_TRANSFER_FUNCTION_GAMMA22_INV_EOTF:
+		return TRANSFER_FUNCTION_GAMMA22;
+	case AMDGPU_TRANSFER_FUNCTION_GAMMA24_EOTF:
+	case AMDGPU_TRANSFER_FUNCTION_GAMMA24_INV_EOTF:
+		return TRANSFER_FUNCTION_GAMMA24;
+	case AMDGPU_TRANSFER_FUNCTION_GAMMA26_EOTF:
+	case AMDGPU_TRANSFER_FUNCTION_GAMMA26_INV_EOTF:
+		return TRANSFER_FUNCTION_GAMMA26;
+	}
+}
+
+static void __to_dc_lut3d_color(struct dc_rgb *rgb,
+				const struct drm_color_lut lut,
+				int bit_precision)
+{
+	rgb->red = drm_color_lut_extract(lut.red, bit_precision);
+	rgb->green = drm_color_lut_extract(lut.green, bit_precision);
+	rgb->blue  = drm_color_lut_extract(lut.blue, bit_precision);
+}
+
+static void __drm_3dlut_to_dc_3dlut(const struct drm_color_lut *lut,
+				    uint32_t lut3d_size,
+				    struct tetrahedral_params *params,
+				    bool use_tetrahedral_9,
+				    int bit_depth)
+{
+	struct dc_rgb *lut0;
+	struct dc_rgb *lut1;
+	struct dc_rgb *lut2;
+	struct dc_rgb *lut3;
+	int lut_i, i;
+
+
+	if (use_tetrahedral_9) {
+		lut0 = params->tetrahedral_9.lut0;
+		lut1 = params->tetrahedral_9.lut1;
+		lut2 = params->tetrahedral_9.lut2;
+		lut3 = params->tetrahedral_9.lut3;
+	} else {
+		lut0 = params->tetrahedral_17.lut0;
+		lut1 = params->tetrahedral_17.lut1;
+		lut2 = params->tetrahedral_17.lut2;
+		lut3 = params->tetrahedral_17.lut3;
+	}
+
+	for (lut_i = 0, i = 0; i < lut3d_size - 4; lut_i++, i += 4) {
+		/*
+		 * We should consider the 3D LUT RGB values are distributed
+		 * along four arrays lut0-3 where the first sizes 1229 and the
+		 * other 1228. The bit depth supported for 3dlut channel is
+		 * 12-bit, but DC also supports 10-bit.
+		 *
+		 * TODO: improve color pipeline API to enable the userspace set
+		 * bit depth and 3D LUT size/stride, as specified by VA-API.
+		 */
+		__to_dc_lut3d_color(&lut0[lut_i], lut[i], bit_depth);
+		__to_dc_lut3d_color(&lut1[lut_i], lut[i + 1], bit_depth);
+		__to_dc_lut3d_color(&lut2[lut_i], lut[i + 2], bit_depth);
+		__to_dc_lut3d_color(&lut3[lut_i], lut[i + 3], bit_depth);
+	}
+	/* lut0 has 1229 points (lut_size/4 + 1) */
+	__to_dc_lut3d_color(&lut0[lut_i], lut[i], bit_depth);
+}
+
+/* amdgpu_dm_atomic_lut3d - set DRM 3D LUT to DC stream
+ * @drm_lut3d: user 3D LUT
+ * @drm_lut3d_size: size of 3D LUT
+ * @lut3d: DC 3D LUT
+ *
+ * Map user 3D LUT data to DC 3D LUT and all necessary bits to program it
+ * on DCN accordingly.
+ */
+static void amdgpu_dm_atomic_lut3d(const struct drm_color_lut *drm_lut3d,
+				   uint32_t drm_lut3d_size,
+				   struct dc_3dlut *lut)
+{
+	if (!drm_lut3d_size) {
+		lut->state.bits.initialized = 0;
+	} else {
+		/* Stride and bit depth are not programmable by API yet.
+		 * Therefore, only supports 17x17x17 3D LUT (12-bit).
+		 */
+		lut->lut_3d.use_tetrahedral_9 = false;
+		lut->lut_3d.use_12bits = true;
+		lut->state.bits.initialized = 1;
+		__drm_3dlut_to_dc_3dlut(drm_lut3d, drm_lut3d_size, &lut->lut_3d,
+					lut->lut_3d.use_tetrahedral_9,
+					MAX_COLOR_3DLUT_BITDEPTH);
+	}
+}
+
+static int amdgpu_dm_atomic_shaper_lut(const struct drm_color_lut *shaper_lut,
+				       bool has_rom,
+				       enum dc_transfer_func_predefined tf,
+				       uint32_t shaper_size,
+				       struct dc_transfer_func *func_shaper)
+{
+	int ret = 0;
+
+	if (shaper_size || tf != TRANSFER_FUNCTION_LINEAR) {
+		/*
+		 * If user shaper LUT is set, we assume a linear color space
+		 * (linearized by degamma 1D LUT or not).
+		 */
+		func_shaper->type = TF_TYPE_DISTRIBUTED_POINTS;
+		func_shaper->tf = tf;
+		func_shaper->sdr_ref_white_level = SDR_WHITE_LEVEL_INIT_VALUE;
+
+		ret = __set_output_tf(func_shaper, shaper_lut, shaper_size, has_rom);
+	} else {
+		func_shaper->type = TF_TYPE_BYPASS;
+		func_shaper->tf = TRANSFER_FUNCTION_LINEAR;
+	}
+
+	return ret;
+}
+
+static int amdgpu_dm_atomic_blend_lut(const struct drm_color_lut *blend_lut,
+				       bool has_rom,
+				       enum dc_transfer_func_predefined tf,
+				       uint32_t blend_size,
+				       struct dc_transfer_func *func_blend)
+{
+	int ret = 0;
+
+	if (blend_size || tf != TRANSFER_FUNCTION_LINEAR) {
+		/*
+		 * DRM plane gamma LUT or TF means we are linearizing color
+		 * space before blending (similar to degamma programming). As
+		 * we don't have hardcoded curve support, or we use AMD color
+		 * module to fill the parameters that will be translated to HW
+		 * points.
+		 */
+		func_blend->type = TF_TYPE_DISTRIBUTED_POINTS;
+		func_blend->tf = tf;
+		func_blend->sdr_ref_white_level = SDR_WHITE_LEVEL_INIT_VALUE;
+
+		ret = __set_input_tf(NULL, func_blend, blend_lut, blend_size);
+	} else {
+		func_blend->type = TF_TYPE_BYPASS;
+		func_blend->tf = TRANSFER_FUNCTION_LINEAR;
+	}
+
+	return ret;
+}
+
+/**
+ * amdgpu_dm_verify_lut3d_size - verifies if 3D LUT is supported and if user
+ * shaper and 3D LUTs match the hw supported size
+ * @adev: amdgpu device
+ * @plane_state: the DRM plane state
+ *
+ * Verifies if pre-blending (DPP) 3D LUT is supported by the HW (DCN 2.0 or
+ * newer) and if the user shaper and 3D LUTs match the supported size.
+ *
+ * Returns:
+ * 0 on success. -EINVAL if lut size are invalid.
+ */
+int amdgpu_dm_verify_lut3d_size(struct amdgpu_device *adev,
+				struct drm_plane_state *plane_state)
+{
+	struct dm_plane_state *dm_plane_state = to_dm_plane_state(plane_state);
+	const struct drm_color_lut *shaper = NULL, *lut3d = NULL;
+	uint32_t exp_size, size, dim_size = MAX_COLOR_3DLUT_SIZE;
+	bool has_3dlut = adev->dm.dc->caps.color.dpp.hw_3d_lut;
+
+	/* shaper LUT is only available if 3D LUT color caps */
+	exp_size = has_3dlut ? MAX_COLOR_LUT_ENTRIES : 0;
+	shaper = __extract_blob_lut(dm_plane_state->shaper_lut, &size);
+
+	if (shaper && size != exp_size) {
+		drm_dbg(&adev->ddev,
+			"Invalid Shaper LUT size. Should be %u but got %u.\n",
+			exp_size, size);
+		return -EINVAL;
+	}
+
+	/* The number of 3D LUT entries is the dimension size cubed */
+	exp_size = has_3dlut ? dim_size * dim_size * dim_size : 0;
+	lut3d = __extract_blob_lut(dm_plane_state->lut3d, &size);
+
+	if (lut3d && size != exp_size) {
+		drm_dbg(&adev->ddev,
+			"Invalid 3D LUT size. Should be %u but got %u.\n",
+			exp_size, size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 /**
  * amdgpu_dm_verify_lut_sizes - verifies if DRM luts match the hw supported sizes
  * @crtc_state: the DRM CRTC state
@@ -401,9 +904,12 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
 	const struct drm_color_lut *degamma_lut, *regamma_lut;
 	uint32_t degamma_size, regamma_size;
 	bool has_regamma, has_degamma;
+	enum dc_transfer_func_predefined tf = TRANSFER_FUNCTION_LINEAR;
 	bool is_legacy;
 	int r;
 
+	tf = amdgpu_tf_to_dc_tf(crtc->regamma_tf);
+
 	r = amdgpu_dm_verify_lut_sizes(&crtc->base);
 	if (r)
 		return r;
@@ -439,27 +945,23 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
 		crtc->cm_is_degamma_srgb = true;
 		stream->out_transfer_func->type = TF_TYPE_DISTRIBUTED_POINTS;
 		stream->out_transfer_func->tf = TRANSFER_FUNCTION_SRGB;
-
+		/*
+		 * Note: although we pass has_rom as parameter here, we never
+		 * actually use ROM because the color module only takes the ROM
+		 * path if transfer_func->type == PREDEFINED.
+		 *
+		 * See more in mod_color_calculate_regamma_params()
+		 */
 		r = __set_legacy_tf(stream->out_transfer_func, regamma_lut,
 				    regamma_size, has_rom);
 		if (r)
 			return r;
-	} else if (has_regamma) {
+	} else {
-		/* If atomic regamma, CRTC RGM goes into RGM LUT. */
+		regamma_size = has_regamma ? regamma_size : 0;
-		stream->out_transfer_func->type = TF_TYPE_DISTRIBUTED_POINTS;
+		r = amdgpu_dm_set_atomic_regamma(stream, regamma_lut,
-		stream->out_transfer_func->tf = TRANSFER_FUNCTION_LINEAR;
+						 regamma_size, has_rom, tf);
-
-		r = __set_output_tf(stream->out_transfer_func, regamma_lut,
-				    regamma_size, has_rom);
 		if (r)
 			return r;
-	} else {
-		/*
-		 * No CRTC RGM means we can just put the block into bypass
-		 * since we don't have any plane level adjustments using it.
-		 */
-		stream->out_transfer_func->type = TF_TYPE_BYPASS;
-		stream->out_transfer_func->tf = TRANSFER_FUNCTION_LINEAR;
 	}
 
 	/*
@@ -495,20 +997,10 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
 	return 0;
 }
 
-/**
+static int
- * amdgpu_dm_update_plane_color_mgmt: Maps DRM color management to DC plane.
+map_crtc_degamma_to_dc_plane(struct dm_crtc_state *crtc,
- * @crtc: amdgpu_dm crtc state
+			     struct dc_plane_state *dc_plane_state,
- * @dc_plane_state: target DC surface
+			     struct dc_color_caps *caps)
- *
- * Update the underlying dc_stream_state's input transfer function (ITF) in
- * preparation for hardware commit. The transfer function used depends on
- * the preparation done on the stream for color management.
- *
- * Returns:
- * 0 on success. -ENOMEM if mem allocation fails.
- */
-int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc,
-				      struct dc_plane_state *dc_plane_state)
 {
 	const struct drm_color_lut *degamma_lut;
 	enum dc_transfer_func_predefined tf = TRANSFER_FUNCTION_SRGB;
@@ -531,8 +1023,7 @@ int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc,
 						 &degamma_size);
 		ASSERT(degamma_size == MAX_COLOR_LUT_ENTRIES);
 
-		dc_plane_state->in_transfer_func->type =
+		dc_plane_state->in_transfer_func->type = TF_TYPE_DISTRIBUTED_POINTS;
-			TF_TYPE_DISTRIBUTED_POINTS;
 
 		/*
 		 * This case isn't fully correct, but also fairly
@@ -564,11 +1055,11 @@ int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc,
 			dc_plane_state->in_transfer_func->tf =
 				TRANSFER_FUNCTION_LINEAR;
 
-		r = __set_input_tf(dc_plane_state->in_transfer_func,
+		r = __set_input_tf(caps, dc_plane_state->in_transfer_func,
 				   degamma_lut, degamma_size);
 		if (r)
 			return r;
-	} else if (crtc->cm_is_degamma_srgb) {
+	} else {
 		/*
 		 * For legacy gamma support we need the regamma input
 		 * in linear space. Assume that the input is sRGB.
@@ -577,14 +1068,209 @@ int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc,
 		dc_plane_state->in_transfer_func->tf = tf;
 
 		if (tf != TRANSFER_FUNCTION_SRGB &&
-		    !mod_color_calculate_degamma_params(NULL,
+		    !mod_color_calculate_degamma_params(caps,
-			    dc_plane_state->in_transfer_func, NULL, false))
+							dc_plane_state->in_transfer_func,
+							NULL, false))
 			return -ENOMEM;
-	} else {
-		/* ...Otherwise we can just bypass the DGM block. */
-		dc_plane_state->in_transfer_func->type = TF_TYPE_BYPASS;
-		dc_plane_state->in_transfer_func->tf = TRANSFER_FUNCTION_LINEAR;
 	}
 
 	return 0;
 }
+
+static int
+__set_dm_plane_degamma(struct drm_plane_state *plane_state,
+		       struct dc_plane_state *dc_plane_state,
+		       struct dc_color_caps *color_caps)
+{
+	struct dm_plane_state *dm_plane_state = to_dm_plane_state(plane_state);
+	const struct drm_color_lut *degamma_lut;
+	enum amdgpu_transfer_function tf = AMDGPU_TRANSFER_FUNCTION_DEFAULT;
+	uint32_t degamma_size;
+	bool has_degamma_lut;
+	int ret;
+
+	degamma_lut = __extract_blob_lut(dm_plane_state->degamma_lut,
+					 &degamma_size);
+
+	has_degamma_lut = degamma_lut &&
+			  !__is_lut_linear(degamma_lut, degamma_size);
+
+	tf = dm_plane_state->degamma_tf;
+
+	/* If we don't have plane degamma LUT nor TF to set on DC, we have
+	 * nothing to do here, return.
+	 */
+	if (!has_degamma_lut && tf == AMDGPU_TRANSFER_FUNCTION_DEFAULT)
+		return -EINVAL;
+
+	dc_plane_state->in_transfer_func->tf = amdgpu_tf_to_dc_tf(tf);
+
+	if (has_degamma_lut) {
+		ASSERT(degamma_size == MAX_COLOR_LUT_ENTRIES);
+
+		dc_plane_state->in_transfer_func->type =
+			TF_TYPE_DISTRIBUTED_POINTS;
+
+		ret = __set_input_tf(color_caps, dc_plane_state->in_transfer_func,
+				     degamma_lut, degamma_size);
+		if (ret)
+			return ret;
+       } else {
+		dc_plane_state->in_transfer_func->type =
+			TF_TYPE_PREDEFINED;
+
+		if (!mod_color_calculate_degamma_params(color_caps,
+		    dc_plane_state->in_transfer_func, NULL, false))
+			return -ENOMEM;
+	}
+	return 0;
+}
+
+static int
+amdgpu_dm_plane_set_color_properties(struct drm_plane_state *plane_state,
+				     struct dc_plane_state *dc_plane_state)
+{
+	struct dm_plane_state *dm_plane_state = to_dm_plane_state(plane_state);
+	enum amdgpu_transfer_function shaper_tf = AMDGPU_TRANSFER_FUNCTION_DEFAULT;
+	enum amdgpu_transfer_function blend_tf = AMDGPU_TRANSFER_FUNCTION_DEFAULT;
+	const struct drm_color_lut *shaper_lut, *lut3d, *blend_lut;
+	uint32_t shaper_size, lut3d_size, blend_size;
+	int ret;
+
+	dc_plane_state->hdr_mult = dc_fixpt_from_s3132(dm_plane_state->hdr_mult);
+
+	shaper_lut = __extract_blob_lut(dm_plane_state->shaper_lut, &shaper_size);
+	shaper_size = shaper_lut != NULL ? shaper_size : 0;
+	shaper_tf = dm_plane_state->shaper_tf;
+	lut3d = __extract_blob_lut(dm_plane_state->lut3d, &lut3d_size);
+	lut3d_size = lut3d != NULL ? lut3d_size : 0;
+
+	amdgpu_dm_atomic_lut3d(lut3d, lut3d_size, dc_plane_state->lut3d_func);
+	ret = amdgpu_dm_atomic_shaper_lut(shaper_lut, false,
+					  amdgpu_tf_to_dc_tf(shaper_tf),
+					  shaper_size,
+					  dc_plane_state->in_shaper_func);
+	if (ret) {
+		drm_dbg_kms(plane_state->plane->dev,
+			    "setting plane %d shaper LUT failed.\n",
+			    plane_state->plane->index);
+
+		return ret;
+	}
+
+	blend_tf = dm_plane_state->blend_tf;
+	blend_lut = __extract_blob_lut(dm_plane_state->blend_lut, &blend_size);
+	blend_size = blend_lut != NULL ? blend_size : 0;
+
+	ret = amdgpu_dm_atomic_blend_lut(blend_lut, false,
+					 amdgpu_tf_to_dc_tf(blend_tf),
+					 blend_size, dc_plane_state->blend_tf);
+	if (ret) {
+		drm_dbg_kms(plane_state->plane->dev,
+			    "setting plane %d gamma lut failed.\n",
+			    plane_state->plane->index);
+
+		return ret;
+	}
+
+	return 0;
+}
+
+/**
+ * amdgpu_dm_update_plane_color_mgmt: Maps DRM color management to DC plane.
+ * @crtc: amdgpu_dm crtc state
+ * @plane_state: DRM plane state
+ * @dc_plane_state: target DC surface
+ *
+ * Update the underlying dc_stream_state's input transfer function (ITF) in
+ * preparation for hardware commit. The transfer function used depends on
+ * the preparation done on the stream for color management.
+ *
+ * Returns:
+ * 0 on success. -ENOMEM if mem allocation fails.
+ */
+int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc,
+				      struct drm_plane_state *plane_state,
+				      struct dc_plane_state *dc_plane_state)
+{
+	struct amdgpu_device *adev = drm_to_adev(crtc->base.state->dev);
+	struct dm_plane_state *dm_plane_state = to_dm_plane_state(plane_state);
+	struct drm_color_ctm_3x4 *ctm = NULL;
+	struct dc_color_caps *color_caps = NULL;
+	bool has_crtc_cm_degamma;
+	int ret;
+
+	ret = amdgpu_dm_verify_lut3d_size(adev, plane_state);
+	if (ret) {
+		drm_dbg_driver(&adev->ddev, "amdgpu_dm_verify_lut3d_size() failed\n");
+		return ret;
+	}
+
+	if (dc_plane_state->ctx && dc_plane_state->ctx->dc)
+		color_caps = &dc_plane_state->ctx->dc->caps.color;
+
+	/* Initially, we can just bypass the DGM block. */
+	dc_plane_state->in_transfer_func->type = TF_TYPE_BYPASS;
+	dc_plane_state->in_transfer_func->tf = TRANSFER_FUNCTION_LINEAR;
+
+	/* After, we start to update values according to color props */
+	has_crtc_cm_degamma = (crtc->cm_has_degamma || crtc->cm_is_degamma_srgb);
+
+	ret = __set_dm_plane_degamma(plane_state, dc_plane_state, color_caps);
+	if (ret == -ENOMEM)
+		return ret;
+
+	/* We only have one degamma block available (pre-blending) for the
+	 * whole color correction pipeline, so that we can't actually perform
+	 * plane and CRTC degamma at the same time. Explicitly reject atomic
+	 * updates when userspace sets both plane and CRTC degamma properties.
+	 */
+	if (has_crtc_cm_degamma && ret != -EINVAL){
+		drm_dbg_kms(crtc->base.crtc->dev,
+			    "doesn't support plane and CRTC degamma at the same time\n");
+			return -EINVAL;
+	}
+
+	/* If we are here, it means we don't have plane degamma settings, check
+	 * if we have CRTC degamma waiting for mapping to pre-blending degamma
+	 * block
+	 */
+	if (has_crtc_cm_degamma) {
+		/*
+		 * AMD HW doesn't have post-blending degamma caps. When DRM
+		 * CRTC atomic degamma is set, we maps it to DPP degamma block
+		 * (pre-blending) or, on legacy gamma, we use DPP degamma to
+		 * linearize (implicit degamma) from sRGB/BT709 according to
+		 * the input space.
+		 */
+		ret = map_crtc_degamma_to_dc_plane(crtc, dc_plane_state, color_caps);
+		if (ret)
+			return ret;
+	}
+
+	/* Setup CRTC CTM. */
+	if (dm_plane_state->ctm) {
+		ctm = (struct drm_color_ctm_3x4 *)dm_plane_state->ctm->data;
+		/*
+		 * DCN2 and older don't support both pre-blending and
+		 * post-blending gamut remap. For this HW family, if we have
+		 * the plane and CRTC CTMs simultaneously, CRTC CTM takes
+		 * priority, and we discard plane CTM, as implemented in
+		 * dcn10_program_gamut_remap(). However, DCN3+ has DPP
+		 * (pre-blending) and MPC (post-blending) `gamut remap` blocks;
+		 * therefore, we can program plane and CRTC CTMs together by
+		 * mapping CRTC CTM to MPC and keeping plane CTM setup at DPP,
+		 * as it's done by dcn30_program_gamut_remap().
+		 */
+		__drm_ctm_3x4_to_dc_matrix(ctm, dc_plane_state->gamut_remap_matrix.matrix);
+
+		dc_plane_state->gamut_remap_matrix.enable_remap = true;
+		dc_plane_state->input_csc_color_matrix.enable_adjustment = false;
+	} else {
+		/* Bypass CTM. */
+		dc_plane_state->gamut_remap_matrix.enable_remap = false;
+		dc_plane_state->input_csc_color_matrix.enable_adjustment = false;
+	}
+
+	return amdgpu_dm_plane_set_color_properties(plane_state, dc_plane_state);
+}

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c

@@ -260,6 +260,7 @@ static struct drm_crtc_state *amdgpu_dm_crtc_duplicate_state(struct drm_crtc *cr
 	state->freesync_config = cur->freesync_config;
 	state->cm_has_degamma = cur->cm_has_degamma;
 	state->cm_is_degamma_srgb = cur->cm_is_degamma_srgb;
+	state->regamma_tf = cur->regamma_tf;
 	state->crc_skip_count = cur->crc_skip_count;
 	state->mpo_requested = cur->mpo_requested;
 	/* TODO Duplicate dc_stream after objects are stream object is flattened */
@@ -296,6 +297,70 @@ static int amdgpu_dm_crtc_late_register(struct drm_crtc *crtc)
 }
 #endif
 
+#ifdef AMD_PRIVATE_COLOR
+/**
+ * dm_crtc_additional_color_mgmt - enable additional color properties
+ * @crtc: DRM CRTC
+ *
+ * This function lets the driver enable post-blending CRTC regamma transfer
+ * function property in addition to DRM CRTC gamma LUT. Default value means
+ * linear transfer function, which is the default CRTC gamma LUT behaviour
+ * without this property.
+ */
+static void
+dm_crtc_additional_color_mgmt(struct drm_crtc *crtc)
+{
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
+
+	if(adev->dm.dc->caps.color.mpc.ogam_ram)
+		drm_object_attach_property(&crtc->base,
+					   adev->mode_info.regamma_tf_property,
+					   AMDGPU_TRANSFER_FUNCTION_DEFAULT);
+}
+
+static int
+amdgpu_dm_atomic_crtc_set_property(struct drm_crtc *crtc,
+				   struct drm_crtc_state *state,
+				   struct drm_property *property,
+				   uint64_t val)
+{
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
+	struct dm_crtc_state *acrtc_state = to_dm_crtc_state(state);
+
+	if (property == adev->mode_info.regamma_tf_property) {
+		if (acrtc_state->regamma_tf != val) {
+			acrtc_state->regamma_tf = val;
+			acrtc_state->base.color_mgmt_changed |= 1;
+		}
+	} else {
+		drm_dbg_atomic(crtc->dev,
+			       "[CRTC:%d:%s] unknown property [PROP:%d:%s]]\n",
+			       crtc->base.id, crtc->name,
+			       property->base.id, property->name);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+amdgpu_dm_atomic_crtc_get_property(struct drm_crtc *crtc,
+				   const struct drm_crtc_state *state,
+				   struct drm_property *property,
+				   uint64_t *val)
+{
+	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
+	struct dm_crtc_state *acrtc_state = to_dm_crtc_state(state);
+
+	if (property == adev->mode_info.regamma_tf_property)
+		*val = acrtc_state->regamma_tf;
+	else
+		return -EINVAL;
+
+	return 0;
+}
+#endif
+
 /* Implemented only the options currently available for the driver */
 static const struct drm_crtc_funcs amdgpu_dm_crtc_funcs = {
 	.reset = amdgpu_dm_crtc_reset_state,
@@ -314,6 +379,10 @@ static const struct drm_crtc_funcs amdgpu_dm_crtc_funcs = {
 #if defined(CONFIG_DEBUG_FS)
 	.late_register = amdgpu_dm_crtc_late_register,
 #endif
+#ifdef AMD_PRIVATE_COLOR
+	.atomic_set_property = amdgpu_dm_atomic_crtc_set_property,
+	.atomic_get_property = amdgpu_dm_atomic_crtc_get_property,
+#endif
 };
 
 static void amdgpu_dm_crtc_helper_disable(struct drm_crtc *crtc)
@@ -489,6 +558,9 @@ int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm,
 
 	drm_mode_crtc_set_gamma_size(&acrtc->base, MAX_COLOR_LEGACY_LUT_ENTRIES);
 
+#ifdef AMD_PRIVATE_COLOR
+	dm_crtc_additional_color_mgmt(&acrtc->base);
+#endif
 	return 0;
 
 fail:

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c

@@ -28,6 +28,7 @@
 #include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fixed.h>
+#include <drm/drm_edid.h>
 #include "dm_services.h"
 #include "amdgpu.h"
 #include "amdgpu_dm.h"

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c

@@ -1337,8 +1337,14 @@ static void amdgpu_dm_plane_drm_plane_reset(struct drm_plane *plane)
 	amdgpu_state = kzalloc(sizeof(*amdgpu_state), GFP_KERNEL);
 	WARN_ON(amdgpu_state == NULL);
 
-	if (amdgpu_state)
+	if (!amdgpu_state)
-		__drm_atomic_helper_plane_reset(plane, &amdgpu_state->base);
+		return;
+
+	__drm_atomic_helper_plane_reset(plane, &amdgpu_state->base);
+	amdgpu_state->degamma_tf = AMDGPU_TRANSFER_FUNCTION_DEFAULT;
+	amdgpu_state->hdr_mult = AMDGPU_HDR_MULT_DEFAULT;
+	amdgpu_state->shaper_tf = AMDGPU_TRANSFER_FUNCTION_DEFAULT;
+	amdgpu_state->blend_tf = AMDGPU_TRANSFER_FUNCTION_DEFAULT;
 }
 
 static struct drm_plane_state *amdgpu_dm_plane_drm_plane_duplicate_state(struct drm_plane *plane)
@@ -1357,6 +1363,27 @@ static struct drm_plane_state *amdgpu_dm_plane_drm_plane_duplicate_state(struct
 		dc_plane_state_retain(dm_plane_state->dc_state);
 	}
 
+	if (old_dm_plane_state->degamma_lut)
+		dm_plane_state->degamma_lut =
+			drm_property_blob_get(old_dm_plane_state->degamma_lut);
+	if (old_dm_plane_state->ctm)
+		dm_plane_state->ctm =
+			drm_property_blob_get(old_dm_plane_state->ctm);
+	if (old_dm_plane_state->shaper_lut)
+		dm_plane_state->shaper_lut =
+			drm_property_blob_get(old_dm_plane_state->shaper_lut);
+	if (old_dm_plane_state->lut3d)
+		dm_plane_state->lut3d =
+			drm_property_blob_get(old_dm_plane_state->lut3d);
+	if (old_dm_plane_state->blend_lut)
+		dm_plane_state->blend_lut =
+			drm_property_blob_get(old_dm_plane_state->blend_lut);
+
+	dm_plane_state->degamma_tf = old_dm_plane_state->degamma_tf;
+	dm_plane_state->hdr_mult = old_dm_plane_state->hdr_mult;
+	dm_plane_state->shaper_tf = old_dm_plane_state->shaper_tf;
+	dm_plane_state->blend_tf = old_dm_plane_state->blend_tf;
+
 	return &dm_plane_state->base;
 }
 
@@ -1424,12 +1451,206 @@ static void amdgpu_dm_plane_drm_plane_destroy_state(struct drm_plane *plane,
 {
 	struct dm_plane_state *dm_plane_state = to_dm_plane_state(state);
 
+	if (dm_plane_state->degamma_lut)
+		drm_property_blob_put(dm_plane_state->degamma_lut);
+	if (dm_plane_state->ctm)
+		drm_property_blob_put(dm_plane_state->ctm);
+	if (dm_plane_state->lut3d)
+		drm_property_blob_put(dm_plane_state->lut3d);
+	if (dm_plane_state->shaper_lut)
+		drm_property_blob_put(dm_plane_state->shaper_lut);
+	if (dm_plane_state->blend_lut)
+		drm_property_blob_put(dm_plane_state->blend_lut);
+
 	if (dm_plane_state->dc_state)
 		dc_plane_state_release(dm_plane_state->dc_state);
 
 	drm_atomic_helper_plane_destroy_state(plane, state);
 }
 
+#ifdef AMD_PRIVATE_COLOR
+static void
+dm_atomic_plane_attach_color_mgmt_properties(struct amdgpu_display_manager *dm,
+					     struct drm_plane *plane)
+{
+	struct amdgpu_mode_info mode_info = dm->adev->mode_info;
+	struct dpp_color_caps dpp_color_caps = dm->dc->caps.color.dpp;
+
+	/* Check HW color pipeline capabilities on DPP block (pre-blending)
+	 * before exposing related properties.
+	 */
+	if (dpp_color_caps.dgam_ram || dpp_color_caps.gamma_corr) {
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_degamma_lut_property,
+					   0);
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_degamma_lut_size_property,
+					   MAX_COLOR_LUT_ENTRIES);
+		drm_object_attach_property(&plane->base,
+					   dm->adev->mode_info.plane_degamma_tf_property,
+					   AMDGPU_TRANSFER_FUNCTION_DEFAULT);
+	}
+	/* HDR MULT is always available */
+	drm_object_attach_property(&plane->base,
+				   dm->adev->mode_info.plane_hdr_mult_property,
+				   AMDGPU_HDR_MULT_DEFAULT);
+
+	/* Only enable plane CTM if both DPP and MPC gamut remap is available. */
+	if (dm->dc->caps.color.mpc.gamut_remap)
+		drm_object_attach_property(&plane->base,
+					   dm->adev->mode_info.plane_ctm_property, 0);
+
+	if (dpp_color_caps.hw_3d_lut) {
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_shaper_lut_property, 0);
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_shaper_lut_size_property,
+					   MAX_COLOR_LUT_ENTRIES);
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_shaper_tf_property,
+					   AMDGPU_TRANSFER_FUNCTION_DEFAULT);
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_lut3d_property, 0);
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_lut3d_size_property,
+					   MAX_COLOR_3DLUT_SIZE);
+	}
+
+	if (dpp_color_caps.ogam_ram) {
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_blend_lut_property, 0);
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_blend_lut_size_property,
+					   MAX_COLOR_LUT_ENTRIES);
+		drm_object_attach_property(&plane->base,
+					   mode_info.plane_blend_tf_property,
+					   AMDGPU_TRANSFER_FUNCTION_DEFAULT);
+	}
+}
+
+static int
+dm_atomic_plane_set_property(struct drm_plane *plane,
+			     struct drm_plane_state *state,
+			     struct drm_property *property,
+			     uint64_t val)
+{
+	struct dm_plane_state *dm_plane_state = to_dm_plane_state(state);
+	struct amdgpu_device *adev = drm_to_adev(plane->dev);
+	bool replaced = false;
+	int ret;
+
+	if (property == adev->mode_info.plane_degamma_lut_property) {
+		ret = drm_property_replace_blob_from_id(plane->dev,
+							&dm_plane_state->degamma_lut,
+							val, -1,
+							sizeof(struct drm_color_lut),
+							&replaced);
+		dm_plane_state->base.color_mgmt_changed |= replaced;
+		return ret;
+	} else if (property == adev->mode_info.plane_degamma_tf_property) {
+		if (dm_plane_state->degamma_tf != val) {
+			dm_plane_state->degamma_tf = val;
+			dm_plane_state->base.color_mgmt_changed = 1;
+		}
+	} else if (property == adev->mode_info.plane_hdr_mult_property) {
+		if (dm_plane_state->hdr_mult != val) {
+			dm_plane_state->hdr_mult = val;
+			dm_plane_state->base.color_mgmt_changed = 1;
+		}
+	} else if (property == adev->mode_info.plane_ctm_property) {
+		ret = drm_property_replace_blob_from_id(plane->dev,
+							&dm_plane_state->ctm,
+							val,
+							sizeof(struct drm_color_ctm_3x4), -1,
+							&replaced);
+		dm_plane_state->base.color_mgmt_changed |= replaced;
+		return ret;
+	} else if (property == adev->mode_info.plane_shaper_lut_property) {
+		ret = drm_property_replace_blob_from_id(plane->dev,
+							&dm_plane_state->shaper_lut,
+							val, -1,
+							sizeof(struct drm_color_lut),
+							&replaced);
+		dm_plane_state->base.color_mgmt_changed |= replaced;
+		return ret;
+	} else if (property == adev->mode_info.plane_shaper_tf_property) {
+		if (dm_plane_state->shaper_tf != val) {
+			dm_plane_state->shaper_tf = val;
+			dm_plane_state->base.color_mgmt_changed = 1;
+		}
+	} else if (property == adev->mode_info.plane_lut3d_property) {
+		ret = drm_property_replace_blob_from_id(plane->dev,
+							&dm_plane_state->lut3d,
+							val, -1,
+							sizeof(struct drm_color_lut),
+							&replaced);
+		dm_plane_state->base.color_mgmt_changed |= replaced;
+		return ret;
+	} else if (property == adev->mode_info.plane_blend_lut_property) {
+		ret = drm_property_replace_blob_from_id(plane->dev,
+							&dm_plane_state->blend_lut,
+							val, -1,
+							sizeof(struct drm_color_lut),
+							&replaced);
+		dm_plane_state->base.color_mgmt_changed |= replaced;
+		return ret;
+	} else if (property == adev->mode_info.plane_blend_tf_property) {
+		if (dm_plane_state->blend_tf != val) {
+			dm_plane_state->blend_tf = val;
+			dm_plane_state->base.color_mgmt_changed = 1;
+		}
+	} else {
+		drm_dbg_atomic(plane->dev,
+			       "[PLANE:%d:%s] unknown property [PROP:%d:%s]]\n",
+			       plane->base.id, plane->name,
+			       property->base.id, property->name);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int
+dm_atomic_plane_get_property(struct drm_plane *plane,
+			     const struct drm_plane_state *state,
+			     struct drm_property *property,
+			     uint64_t *val)
+{
+	struct dm_plane_state *dm_plane_state = to_dm_plane_state(state);
+	struct amdgpu_device *adev = drm_to_adev(plane->dev);
+
+	if (property == adev->mode_info.plane_degamma_lut_property) {
+		*val = (dm_plane_state->degamma_lut) ?
+			dm_plane_state->degamma_lut->base.id : 0;
+	} else if (property == adev->mode_info.plane_degamma_tf_property) {
+		*val = dm_plane_state->degamma_tf;
+	} else if (property == adev->mode_info.plane_hdr_mult_property) {
+		*val = dm_plane_state->hdr_mult;
+	} else if (property == adev->mode_info.plane_ctm_property) {
+		*val = (dm_plane_state->ctm) ?
+			dm_plane_state->ctm->base.id : 0;
+	} else 	if (property == adev->mode_info.plane_shaper_lut_property) {
+		*val = (dm_plane_state->shaper_lut) ?
+			dm_plane_state->shaper_lut->base.id : 0;
+	} else if (property == adev->mode_info.plane_shaper_tf_property) {
+		*val = dm_plane_state->shaper_tf;
+	} else 	if (property == adev->mode_info.plane_lut3d_property) {
+		*val = (dm_plane_state->lut3d) ?
+			dm_plane_state->lut3d->base.id : 0;
+	} else 	if (property == adev->mode_info.plane_blend_lut_property) {
+		*val = (dm_plane_state->blend_lut) ?
+			dm_plane_state->blend_lut->base.id : 0;
+	} else if (property == adev->mode_info.plane_blend_tf_property) {
+		*val = dm_plane_state->blend_tf;
+
+	} else {
+		return -EINVAL;
+	}
+
+	return 0;
+}
+#endif
+
 static const struct drm_plane_funcs dm_plane_funcs = {
 	.update_plane	= drm_atomic_helper_update_plane,
 	.disable_plane	= drm_atomic_helper_disable_plane,
@@ -1438,6 +1659,10 @@ static const struct drm_plane_funcs dm_plane_funcs = {
 	.atomic_duplicate_state = amdgpu_dm_plane_drm_plane_duplicate_state,
 	.atomic_destroy_state = amdgpu_dm_plane_drm_plane_destroy_state,
 	.format_mod_supported = amdgpu_dm_plane_format_mod_supported,
+#ifdef AMD_PRIVATE_COLOR
+	.atomic_set_property = dm_atomic_plane_set_property,
+	.atomic_get_property = dm_atomic_plane_get_property,
+#endif
 };
 
 int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
@@ -1517,6 +1742,9 @@ int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
 
 	drm_plane_helper_add(plane, &dm_plane_helper_funcs);
 
+#ifdef AMD_PRIVATE_COLOR
+	dm_atomic_plane_attach_color_mgmt_properties(dm, plane);
+#endif
 	/* Create (reset) the plane state */
 	if (plane->funcs->reset)
 		plane->funcs->reset(plane);

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_wb.c

@@ -32,6 +32,7 @@
 #include "amdgpu_display.h"
 #include "dc.h"
 
+#include <drm/drm_edid.h>
 #include <drm/drm_atomic_state_helper.h>
 #include <drm/drm_modeset_helper_vtables.h>
 

drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c

@@ -1747,7 +1747,6 @@ static enum bp_result bios_parser_get_firmware_info(
 				result = get_firmware_info_v3_2(bp, info);
 				break;
 			case 4:
-			case 5:
 				result = get_firmware_info_v3_4(bp, info);
 				break;
 			default:
@@ -2387,18 +2386,11 @@ static enum bp_result get_vram_info_v30(
 		return BP_RESULT_BADBIOSTABLE;
 
 	info->num_chans = info_v30->channel_num;
-	/* As suggested by VBIOS we should always use
+	info->dram_channel_width_bytes = (1 << info_v30->channel_width) / 8;
-	 * dram_channel_width_bytes = 2 when using VRAM
-	 * table version 3.0. This is because the channel_width
-	 * param in the VRAM info table is changed in 7000 series and
-	 * no longer represents the memory channel width.
-	 */
-	info->dram_channel_width_bytes = 2;
 
 	return result;
 }
 
-
 /*
  * get_integrated_info_v11
  *

drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c

@@ -368,7 +368,7 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p
 	}
 	break;
 
-#endif /* CONFIG_DRM_AMD_DC_FP - Family RV */
+#endif	/* CONFIG_DRM_AMD_DC_FP */
 	default:
 		ASSERT(0); /* Unknown Asic */
 		break;

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c

@@ -361,26 +361,26 @@ void dcn35_smu_set_zstate_support(struct clk_mgr_internal *clk_mgr, enum dcn_zst
 	case DCN_ZSTATE_SUPPORT_ALLOW:
 		msg_id = VBIOSSMC_MSG_AllowZstatesEntry;
 		param = (1 << 10) | (1 << 9) | (1 << 8);
-		smu_print("%s: SMC_MSG_AllowZstatesEntr msg = ALLOW, param = %d\n", __func__, param);
+		smu_print("%s: SMC_MSG_AllowZstatesEntry msg = ALLOW, param = %d\n", __func__, param);
 		break;
 
 	case DCN_ZSTATE_SUPPORT_DISALLOW:
 		msg_id = VBIOSSMC_MSG_AllowZstatesEntry;
 		param = 0;
-		smu_print("%s: SMC_MSG_AllowZstatesEntr msg_id = DISALLOW, param = %d\n",  __func__, param);
+		smu_print("%s: SMC_MSG_AllowZstatesEntry msg_id = DISALLOW, param = %d\n",  __func__, param);
 		break;
 
 
 	case DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY:
 		msg_id = VBIOSSMC_MSG_AllowZstatesEntry;
 		param = (1 << 10);
-		smu_print("%s: SMC_MSG_AllowZstatesEntr msg = ALLOW_Z10_ONLY, param = %d\n", __func__, param);
+		smu_print("%s: SMC_MSG_AllowZstatesEntry msg = ALLOW_Z10_ONLY, param = %d\n", __func__, param);
 		break;
 
 	case DCN_ZSTATE_SUPPORT_ALLOW_Z8_Z10_ONLY:
 		msg_id = VBIOSSMC_MSG_AllowZstatesEntry;
 		param = (1 << 10) | (1 << 8);
-		smu_print("%s: SMC_MSG_AllowZstatesEntr msg = ALLOW_Z8_Z10_ONLY, param = %d\n", __func__, param);
+		smu_print("%s: SMC_MSG_AllowZstatesEntry msg = ALLOW_Z8_Z10_ONLY, param = %d\n", __func__, param);
 		break;
 
 	case DCN_ZSTATE_SUPPORT_ALLOW_Z8_ONLY:

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 "3.2.263"
+#define DC_VER "3.2.264"
 
 #define MAX_SURFACES 3
 #define MAX_PLANES 6

drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c

@@ -1268,3 +1268,17 @@ void dc_dmub_srv_exit_low_power_state(const struct dc *dc)
 		ASSERT(0);
 }
 
+void dc_dmub_srv_set_power_state(struct dc_dmub_srv *dc_dmub_srv, enum dc_acpi_cm_power_state powerState)
+{
+	struct dmub_srv *dmub;
+
+	if (!dc_dmub_srv)
+		return;
+
+	dmub = dc_dmub_srv->dmub;
+
+	if (powerState == DC_ACPI_CM_POWER_STATE_D0)
+		dmub_srv_set_power_state(dmub, DMUB_POWER_STATE_D0);
+	else
+		dmub_srv_set_power_state(dmub, DMUB_POWER_STATE_D3);
+}

drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h

@@ -102,4 +102,6 @@ void dc_dmub_srv_subvp_save_surf_addr(const struct dc_dmub_srv *dc_dmub_srv, con
 bool dc_dmub_srv_is_hw_pwr_up(struct dc_dmub_srv *dc_dmub_srv, bool wait);
 void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle);
 void dc_dmub_srv_exit_low_power_state(const struct dc *dc);
+
+void dc_dmub_srv_set_power_state(struct dc_dmub_srv *dc_dmub_srv, enum dc_acpi_cm_power_state powerState);
 #endif /* _DMUB_DC_SRV_H_ */

drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c

@@ -64,11 +64,15 @@ static void dmub_abm_init_ex(struct abm *abm, uint32_t backlight)
 
 static unsigned int dmub_abm_get_current_backlight_ex(struct abm *abm)
 {
+	dc_allow_idle_optimizations(abm->ctx->dc, false);
+
 	return dmub_abm_get_current_backlight(abm);
 }
 
 static unsigned int dmub_abm_get_target_backlight_ex(struct abm *abm)
 {
+	dc_allow_idle_optimizations(abm->ctx->dc, false);
+
 	return dmub_abm_get_target_backlight(abm);
 }
 

drivers/gpu/drm/amd/display/dc/dcn35/dcn35_init.c

@@ -118,7 +118,8 @@ static const struct hw_sequencer_funcs dcn35_funcs = {
 	.update_dsc_pg = dcn32_update_dsc_pg,
 	.calc_blocks_to_gate = dcn35_calc_blocks_to_gate,
 	.calc_blocks_to_ungate = dcn35_calc_blocks_to_ungate,
-	.block_power_control = dcn35_block_power_control,
+	.hw_block_power_up = dcn35_hw_block_power_up,
+	.hw_block_power_down = dcn35_hw_block_power_down,
 	.root_clock_control = dcn35_root_clock_control,
 	.set_idle_state = dcn35_set_idle_state,
 	.get_idle_state = dcn35_get_idle_state

drivers/gpu/drm/amd/display/dc/dm_pp_smu.h

@@ -247,6 +247,7 @@ struct pp_smu_funcs_nv {
 #define PP_SMU_NUM_MEMCLK_DPM_LEVELS  4
 #define PP_SMU_NUM_DCLK_DPM_LEVELS    8
 #define PP_SMU_NUM_VCLK_DPM_LEVELS    8
+#define PP_SMU_NUM_VPECLK_DPM_LEVELS  8
 
 struct dpm_clock {
   uint32_t  Freq;    // In MHz
@@ -262,6 +263,7 @@ struct dpm_clocks {
 	struct dpm_clock MemClocks[PP_SMU_NUM_MEMCLK_DPM_LEVELS];
 	struct dpm_clock VClocks[PP_SMU_NUM_VCLK_DPM_LEVELS];
 	struct dpm_clock DClocks[PP_SMU_NUM_DCLK_DPM_LEVELS];
+	struct dpm_clock VPEClocks[PP_SMU_NUM_VPECLK_DPM_LEVELS];
 };
 
 

drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c

@@ -813,6 +813,8 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman
 					(v->DRAMSpeedPerState[mode_lib->vba.VoltageLevel] <= MEM_STROBE_FREQ_MHZ ||
 						v->DCFCLKPerState[mode_lib->vba.VoltageLevel] <= DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ) ?
 							mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
+					mode_lib->vba.PrefetchModePerState[mode_lib->vba.VoltageLevel][mode_lib->vba.maxMpcComb] > 0 || mode_lib->vba.DRAMClockChangeRequirementFinal == false,
+
 					/* Output */
 					&v->DSTXAfterScaler[k],
 					&v->DSTYAfterScaler[k],
@@ -3317,6 +3319,7 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 							v->SwathHeightCThisState[k], v->TWait,
 							(v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ || v->DCFCLKState[i][j] <= DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ) ?
 									mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
+							mode_lib->vba.PrefetchModePerState[i][j] > 0 || mode_lib->vba.DRAMClockChangeRequirementFinal == false,
 
 							/* Output */
 							&v->dummy_vars.dml32_ModeSupportAndSystemConfigurationFull.DSTXAfterScaler[k],

drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c

@@ -3423,6 +3423,7 @@ bool dml32_CalculatePrefetchSchedule(
 		unsigned int SwathHeightC,
 		double TWait,
 		double TPreReq,
+		bool ExtendPrefetchIfPossible,
 		/* Output */
 		double   *DSTXAfterScaler,
 		double   *DSTYAfterScaler,
@@ -3892,12 +3893,32 @@ bool dml32_CalculatePrefetchSchedule(
 			/* Clamp to oto for bandwidth calculation */
 			LinesForPrefetchBandwidth = dst_y_prefetch_oto;
 		} else {
-			*DestinationLinesForPrefetch = dst_y_prefetch_equ;
+			/* For mode programming we want to extend the prefetch as much as possible
-			TimeForFetchingMetaPTE = Tvm_equ;
+			 * (up to oto, or as long as we can for equ) if we're not already applying
-			TimeForFetchingRowInVBlank = Tr0_equ;
+			 * the 60us prefetch requirement. This is to avoid intermittent underflow
-			*PrefetchBandwidth = prefetch_bw_equ;
+			 * issues during prefetch.
-			/* Clamp to equ for bandwidth calculation */
+			 *
-			LinesForPrefetchBandwidth = dst_y_prefetch_equ;
+			 * The prefetch extension is applied under the following scenarios:
+			 * 1. We're in prefetch mode > 0 (i.e. we don't support MCLK switch in blank)
+			 * 2. We're using subvp or drr methods of p-state switch, in which case we
+			 *    we don't care if prefetch takes up more of the blanking time
+			 *
+			 * Mode programming typically chooses the smallest prefetch time possible
+			 * (i.e. highest bandwidth during prefetch) presumably to create margin between
+			 * p-states / c-states that happen in vblank and prefetch. Therefore we only
+			 * apply this prefetch extension when p-state in vblank is not required (UCLK
+			 * p-states take up the most vblank time).
+			 */
+			if (ExtendPrefetchIfPossible && TPreReq == 0 && VStartup < MaxVStartup) {
+				MyError = true;
+			} else {
+				*DestinationLinesForPrefetch = dst_y_prefetch_equ;
+				TimeForFetchingMetaPTE = Tvm_equ;
+				TimeForFetchingRowInVBlank = Tr0_equ;
+				*PrefetchBandwidth = prefetch_bw_equ;
+				/* Clamp to equ for bandwidth calculation */
+				LinesForPrefetchBandwidth = dst_y_prefetch_equ;
+			}
 		}
 
 		*DestinationLinesToRequestVMInVBlank = dml_ceil(4.0 * TimeForFetchingMetaPTE / LineTime, 1.0) / 4.0;

drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.h

@@ -747,6 +747,7 @@ bool dml32_CalculatePrefetchSchedule(
 		unsigned int SwathHeightC,
 		double TWait,
 		double TPreReq,
+		bool ExtendPrefetchIfPossible,
 		/* Output */
 		double   *DSTXAfterScaler,
 		double   *DSTYAfterScaler,

drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c

@@ -124,7 +124,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = {
 			.phyclk_mhz = 600.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 186.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 		{
 			.state = 1,
@@ -133,7 +133,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = {
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 209.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 		{
 			.state = 2,
@@ -142,7 +142,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = {
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 209.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 		{
 			.state = 3,
@@ -151,7 +151,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = {
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 371.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 		{
 			.state = 4,
@@ -160,7 +160,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_5_soc = {
 			.phyclk_mhz = 810.0,
 			.phyclk_d18_mhz = 667.0,
 			.dscclk_mhz = 417.0,
-			.dtbclk_mhz = 625.0,
+			.dtbclk_mhz = 600.0,
 		},
 	},
 	.num_states = 5,
@@ -367,6 +367,8 @@ void dcn35_update_bw_bounding_box_fpu(struct dc *dc,
 				clock_limits[i].socclk_mhz;
 			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].memclk_mhz =
 				clk_table->entries[i].memclk_mhz * clk_table->entries[i].wck_ratio;
+			dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz =
+				clock_limits[i].dtbclk_mhz;
 			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dcfclk_levels =
 				clk_table->num_entries;
 			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_fclk_levels =
@@ -379,6 +381,8 @@ void dcn35_update_bw_bounding_box_fpu(struct dc *dc,
 				clk_table->num_entries;
 			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_memclk_levels =
 				clk_table->num_entries;
+			dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dtbclk_levels =
+				clk_table->num_entries;
 		}
 	}
 

drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c

@@ -6329,7 +6329,7 @@ static void dml_prefetch_check(struct display_mode_lib_st *mode_lib)
 				mode_lib->ms.NoOfDPPThisState,
 				mode_lib->ms.dpte_group_bytes,
 				s->HostVMInefficiencyFactor,
-				mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024,
+				mode_lib->ms.soc.hostvm_min_page_size_kbytes,
 				mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels);
 
 		s->NextMaxVStartup = s->MaxVStartupAllPlanes[j];
@@ -6542,7 +6542,7 @@ static void dml_prefetch_check(struct display_mode_lib_st *mode_lib)
 						mode_lib->ms.cache_display_cfg.plane.HostVMEnable,
 						mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels,
 						mode_lib->ms.cache_display_cfg.plane.GPUVMEnable,
-						mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024,
+						mode_lib->ms.soc.hostvm_min_page_size_kbytes,
 						mode_lib->ms.PDEAndMetaPTEBytesPerFrame[j][k],
 						mode_lib->ms.MetaRowBytes[j][k],
 						mode_lib->ms.DPTEBytesPerRow[j][k],
@@ -7687,7 +7687,7 @@ dml_bool_t dml_core_mode_support(struct display_mode_lib_st *mode_lib)
 		CalculateVMRowAndSwath_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels;
 		CalculateVMRowAndSwath_params->GPUVMMaxPageTableLevels = mode_lib->ms.cache_display_cfg.plane.GPUVMMaxPageTableLevels;
 		CalculateVMRowAndSwath_params->GPUVMMinPageSizeKBytes = mode_lib->ms.cache_display_cfg.plane.GPUVMMinPageSizeKBytes;
-		CalculateVMRowAndSwath_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024;
+		CalculateVMRowAndSwath_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes;
 		CalculateVMRowAndSwath_params->PTEBufferModeOverrideEn = mode_lib->ms.cache_display_cfg.plane.PTEBufferModeOverrideEn;
 		CalculateVMRowAndSwath_params->PTEBufferModeOverrideVal = mode_lib->ms.cache_display_cfg.plane.PTEBufferMode;
 		CalculateVMRowAndSwath_params->PTEBufferSizeNotExceeded = mode_lib->ms.PTEBufferSizeNotExceededPerState;
@@ -7957,7 +7957,7 @@ dml_bool_t dml_core_mode_support(struct display_mode_lib_st *mode_lib)
 		UseMinimumDCFCLK_params->GPUVMMaxPageTableLevels = mode_lib->ms.cache_display_cfg.plane.GPUVMMaxPageTableLevels;
 		UseMinimumDCFCLK_params->HostVMEnable = mode_lib->ms.cache_display_cfg.plane.HostVMEnable;
 		UseMinimumDCFCLK_params->NumberOfActiveSurfaces = mode_lib->ms.num_active_planes;
-		UseMinimumDCFCLK_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024;
+		UseMinimumDCFCLK_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes;
 		UseMinimumDCFCLK_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels;
 		UseMinimumDCFCLK_params->DynamicMetadataVMEnabled = mode_lib->ms.ip.dynamic_metadata_vm_enabled;
 		UseMinimumDCFCLK_params->ImmediateFlipRequirement = s->ImmediateFlipRequiredFinal;
@@ -8699,7 +8699,7 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc
 	CalculateVMRowAndSwath_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels;
 	CalculateVMRowAndSwath_params->GPUVMMaxPageTableLevels = mode_lib->ms.cache_display_cfg.plane.GPUVMMaxPageTableLevels;
 	CalculateVMRowAndSwath_params->GPUVMMinPageSizeKBytes = mode_lib->ms.cache_display_cfg.plane.GPUVMMinPageSizeKBytes;
-	CalculateVMRowAndSwath_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024;
+	CalculateVMRowAndSwath_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes;
 	CalculateVMRowAndSwath_params->PTEBufferModeOverrideEn = mode_lib->ms.cache_display_cfg.plane.PTEBufferModeOverrideEn;
 	CalculateVMRowAndSwath_params->PTEBufferModeOverrideVal = mode_lib->ms.cache_display_cfg.plane.PTEBufferMode;
 	CalculateVMRowAndSwath_params->PTEBufferSizeNotExceeded = s->dummy_boolean_array[0];
@@ -8805,7 +8805,7 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc
 			mode_lib->ms.cache_display_cfg.hw.DPPPerSurface,
 			locals->dpte_group_bytes,
 			s->HostVMInefficiencyFactor,
-			mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024,
+			mode_lib->ms.soc.hostvm_min_page_size_kbytes,
 			mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels);
 
 	locals->TCalc = 24.0 / locals->DCFCLKDeepSleep;
@@ -8995,7 +8995,7 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc
 			CalculatePrefetchSchedule_params->GPUVMEnable = mode_lib->ms.cache_display_cfg.plane.GPUVMEnable;
 			CalculatePrefetchSchedule_params->HostVMEnable = mode_lib->ms.cache_display_cfg.plane.HostVMEnable;
 			CalculatePrefetchSchedule_params->HostVMMaxNonCachedPageTableLevels = mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels;
-			CalculatePrefetchSchedule_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024;
+			CalculatePrefetchSchedule_params->HostVMMinPageSize = mode_lib->ms.soc.hostvm_min_page_size_kbytes;
 			CalculatePrefetchSchedule_params->DynamicMetadataEnable = mode_lib->ms.cache_display_cfg.plane.DynamicMetadataEnable[k];
 			CalculatePrefetchSchedule_params->DynamicMetadataVMEnabled = mode_lib->ms.ip.dynamic_metadata_vm_enabled;
 			CalculatePrefetchSchedule_params->DynamicMetadataLinesBeforeActiveRequired = mode_lib->ms.cache_display_cfg.plane.DynamicMetadataLinesBeforeActiveRequired[k];
@@ -9240,7 +9240,7 @@ void dml_core_mode_programming(struct display_mode_lib_st *mode_lib, const struc
 						mode_lib->ms.cache_display_cfg.plane.HostVMEnable,
 						mode_lib->ms.cache_display_cfg.plane.HostVMMaxPageTableLevels,
 						mode_lib->ms.cache_display_cfg.plane.GPUVMEnable,
-						mode_lib->ms.soc.hostvm_min_page_size_kbytes * 1024,
+						mode_lib->ms.soc.hostvm_min_page_size_kbytes,
 						locals->PDEAndMetaPTEBytesFrame[k],
 						locals->MetaRowByte[k],
 						locals->PixelPTEBytesPerRow[k],

drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c

@@ -425,8 +425,9 @@ void dml2_init_soc_states(struct dml2_context *dml2, const struct dc *in_dc,
 		}
 
 		for (i = 0; i < dml2->config.bbox_overrides.clks_table.num_entries_per_clk.num_dtbclk_levels; i++) {
-			p->in_states->state_array[i].dtbclk_mhz =
+			if (dml2->config.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz > 0)
-				dml2->config.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz;
+				p->in_states->state_array[i].dtbclk_mhz =
+					dml2->config.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz;
 		}
 
 		for (i = 0; i < dml2->config.bbox_overrides.clks_table.num_entries_per_clk.num_dispclk_levels; i++) {

drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c

@@ -160,13 +160,6 @@ bool is_dp2p0_output_encoder(const struct pipe_ctx *pipe_ctx)
 
 	if (pipe_ctx->stream == NULL)
 		return false;
-	/* Count MST hubs once by treating only 1st remote sink in topology as an encoder */
-	if (pipe_ctx->stream->link && pipe_ctx->stream->link->remote_sinks[0]) {
-			return (pipe_ctx->stream_res.hpo_dp_stream_enc &&
-				pipe_ctx->link_res.hpo_dp_link_enc &&
-				dc_is_dp_signal(pipe_ctx->stream->signal) &&
-				(pipe_ctx->stream->link->remote_sinks[0] == pipe_ctx->stream->sink));
-	}
 
 	return (pipe_ctx->stream_res.hpo_dp_stream_enc &&
 		pipe_ctx->link_res.hpo_dp_link_enc &&

drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c

@@ -1877,6 +1877,8 @@ void dcn20_program_front_end_for_ctx(
 	int i;
 	struct dce_hwseq *hws = dc->hwseq;
 	DC_LOGGER_INIT(dc->ctx->logger);
+	unsigned int prev_hubp_count = 0;
+	unsigned int hubp_count = 0;
 
 	if (resource_is_pipe_topology_changed(dc->current_state, context))
 		resource_log_pipe_topology_update(dc, context);
@@ -1894,6 +1896,20 @@ void dcn20_program_front_end_for_ctx(
 		}
 	}
 
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (dc->current_state->res_ctx.pipe_ctx[i].plane_state)
+			prev_hubp_count++;
+		if (context->res_ctx.pipe_ctx[i].plane_state)
+			hubp_count++;
+	}
+
+	if (prev_hubp_count == 0 && hubp_count > 0) {
+		if (dc->res_pool->hubbub->funcs->force_pstate_change_control)
+			dc->res_pool->hubbub->funcs->force_pstate_change_control(
+					dc->res_pool->hubbub, true, false);
+		udelay(500);
+	}
+
 	/* Set pipe update flags and lock pipes */
 	for (i = 0; i < dc->res_pool->pipe_count; i++)
 		dcn20_detect_pipe_changes(&dc->current_state->res_ctx.pipe_ctx[i],
@@ -2039,6 +2055,10 @@ void dcn20_post_unlock_program_front_end(
 		}
 	}
 
+	if (dc->res_pool->hubbub->funcs->force_pstate_change_control)
+		dc->res_pool->hubbub->funcs->force_pstate_change_control(
+				dc->res_pool->hubbub, false, false);
+
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i];
 

drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c

@@ -1123,9 +1123,23 @@ void dcn35_calc_blocks_to_ungate(struct dc *dc, struct dc_state *context,
 		update_state->pg_res_update[PG_HPO] = true;
 
 }
-
+/**
-void dcn35_block_power_control(struct dc *dc,
+	 * power down sequence
-	struct pg_block_update *update_state, bool power_on)
+	 * ONO Region 3, DCPG 25: hpo - SKIPPED
+	 * ONO Region 4, DCPG 0: dchubp0, dpp0
+	 * ONO Region 6, DCPG 1: dchubp1, dpp1
+	 * ONO Region 8, DCPG 2: dchubp2, dpp2
+	 * ONO Region 10, DCPG 3: dchubp3, dpp3
+	 * ONO Region 1, DCPG 23: dchubbub dchvm dchubbubmem - SKIPPED. PMFW will pwr dwn at IPS2 entry
+	 * ONO Region 5, DCPG 16: dsc0
+	 * ONO Region 7, DCPG 17: dsc1
+	 * ONO Region 9, DCPG 18: dsc2
+	 * ONO Region 11, DCPG 19: dsc3
+	 * ONO Region 2, DCPG 24: mpc opp optc dwb
+	 * ONO Region 0, DCPG 22: dccg dio dcio - SKIPPED. will be pwr dwn after lono timer is armed
+*/
+void dcn35_hw_block_power_down(struct dc *dc,
+	struct pg_block_update *update_state)
 {
 	int i = 0;
 	struct pg_cntl *pg_cntl = dc->res_pool->pg_cntl;
@@ -1134,50 +1148,81 @@ void dcn35_block_power_control(struct dc *dc,
 		return;
 	if (dc->debug.ignore_pg)
 		return;
+
 	if (update_state->pg_res_update[PG_HPO]) {
 		if (pg_cntl->funcs->hpo_pg_control)
-			pg_cntl->funcs->hpo_pg_control(pg_cntl, power_on);
+			pg_cntl->funcs->hpo_pg_control(pg_cntl, false);
 	}
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
 		if (update_state->pg_pipe_res_update[PG_HUBP][i] &&
 			update_state->pg_pipe_res_update[PG_DPP][i]) {
 			if (pg_cntl->funcs->hubp_dpp_pg_control)
-				pg_cntl->funcs->hubp_dpp_pg_control(pg_cntl, i, power_on);
+				pg_cntl->funcs->hubp_dpp_pg_control(pg_cntl, i, false);
 		}
-
+	}
+	for (i = 0; i < dc->res_pool->res_cap->num_dsc; i++)
 		if (update_state->pg_pipe_res_update[PG_DSC][i]) {
 			if (pg_cntl->funcs->dsc_pg_control)
-				pg_cntl->funcs->dsc_pg_control(pg_cntl, i, power_on);
+				pg_cntl->funcs->dsc_pg_control(pg_cntl, i, false);
 		}
 
-		if (update_state->pg_pipe_res_update[PG_MPCC][i]) {
-			if (pg_cntl->funcs->mpcc_pg_control)
-				pg_cntl->funcs->mpcc_pg_control(pg_cntl, i, power_on);
-		}
-
-		if (update_state->pg_pipe_res_update[PG_OPP][i]) {
-			if (pg_cntl->funcs->opp_pg_control)
-				pg_cntl->funcs->opp_pg_control(pg_cntl, i, power_on);
-		}
-
-		if (update_state->pg_pipe_res_update[PG_OPTC][i]) {
-			if (pg_cntl->funcs->optc_pg_control)
-				pg_cntl->funcs->optc_pg_control(pg_cntl, i, power_on);
-		}
-	}
-
-	if (update_state->pg_res_update[PG_DWB]) {
-		if (pg_cntl->funcs->dwb_pg_control)
-			pg_cntl->funcs->dwb_pg_control(pg_cntl, power_on);
-	}
 
 	/*this will need all the clients to unregister optc interruts let dmubfw handle this*/
 	if (pg_cntl->funcs->plane_otg_pg_control)
-		pg_cntl->funcs->plane_otg_pg_control(pg_cntl, power_on);
+		pg_cntl->funcs->plane_otg_pg_control(pg_cntl, false);
+
+	//domain22, 23, 25 currently always on.
 
 }
+/**
+	 * power up sequence
+	 * ONO Region 0, DCPG 22: dccg dio dcio - SKIPPED
+	 * ONO Region 2, DCPG 24: mpc opp optc dwb
+	 * ONO Region 5, DCPG 16: dsc0
+	 * ONO Region 7, DCPG 17: dsc1
+	 * ONO Region 9, DCPG 18: dsc2
+	 * ONO Region 11, DCPG 19: dsc3
+	 * ONO Region 1, DCPG 23: dchubbub dchvm dchubbubmem - SKIPPED. PMFW will power up at IPS2 exit
+	 * ONO Region 4, DCPG 0: dchubp0, dpp0
+	 * ONO Region 6, DCPG 1: dchubp1, dpp1
+	 * ONO Region 8, DCPG 2: dchubp2, dpp2
+	 * ONO Region 10, DCPG 3: dchubp3, dpp3
+	 * ONO Region 3, DCPG 25: hpo - SKIPPED
+ */
+void dcn35_hw_block_power_up(struct dc *dc,
+	struct pg_block_update *update_state)
+{
+	int i = 0;
+	struct pg_cntl *pg_cntl = dc->res_pool->pg_cntl;
 
+	if (!pg_cntl)
+		return;
+	if (dc->debug.ignore_pg)
+		return;
+	//domain22, 23, 25 currently always on.
+	/*this will need all the clients to unregister optc interruts let dmubfw handle this*/
+	if (pg_cntl->funcs->plane_otg_pg_control)
+		pg_cntl->funcs->plane_otg_pg_control(pg_cntl, true);
+
+	for (i = 0; i < dc->res_pool->res_cap->num_dsc; i++)
+		if (update_state->pg_pipe_res_update[PG_DSC][i]) {
+			if (pg_cntl->funcs->dsc_pg_control)
+				pg_cntl->funcs->dsc_pg_control(pg_cntl, i, true);
+		}
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (update_state->pg_pipe_res_update[PG_HUBP][i] &&
+			update_state->pg_pipe_res_update[PG_DPP][i]) {
+			if (pg_cntl->funcs->hubp_dpp_pg_control)
+				pg_cntl->funcs->hubp_dpp_pg_control(pg_cntl, i, true);
+		}
+	}
+	if (update_state->pg_res_update[PG_HPO]) {
+		if (pg_cntl->funcs->hpo_pg_control)
+			pg_cntl->funcs->hpo_pg_control(pg_cntl, true);
+	}
+}
 void dcn35_root_clock_control(struct dc *dc,
 	struct pg_block_update *update_state, bool power_on)
 {
@@ -1186,14 +1231,16 @@ void dcn35_root_clock_control(struct dc *dc,
 
 	if (!pg_cntl)
 		return;
-
+	/*enable root clock first when power up*/
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+	if (power_on)
-		if (update_state->pg_pipe_res_update[PG_HUBP][i] &&
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
-			update_state->pg_pipe_res_update[PG_DPP][i]) {
+			if (update_state->pg_pipe_res_update[PG_HUBP][i] &&
-			if (dc->hwseq->funcs.dpp_root_clock_control)
+				update_state->pg_pipe_res_update[PG_DPP][i]) {
-				dc->hwseq->funcs.dpp_root_clock_control(dc->hwseq, i, power_on);
+				if (dc->hwseq->funcs.dpp_root_clock_control)
+					dc->hwseq->funcs.dpp_root_clock_control(dc->hwseq, i, power_on);
+			}
 		}
-
+	for (i = 0; i < dc->res_pool->res_cap->num_dsc; i++) {
 		if (update_state->pg_pipe_res_update[PG_DSC][i]) {
 			if (power_on) {
 				if (dc->res_pool->dccg->funcs->enable_dsc)
@@ -1204,6 +1251,15 @@ void dcn35_root_clock_control(struct dc *dc,
 			}
 		}
 	}
+	/*disable root clock first when power down*/
+	if (!power_on)
+		for (i = 0; i < dc->res_pool->pipe_count; i++) {
+			if (update_state->pg_pipe_res_update[PG_HUBP][i] &&
+				update_state->pg_pipe_res_update[PG_DPP][i]) {
+				if (dc->hwseq->funcs.dpp_root_clock_control)
+					dc->hwseq->funcs.dpp_root_clock_control(dc->hwseq, i, power_on);
+			}
+		}
 }
 
 void dcn35_prepare_bandwidth(
@@ -1217,9 +1273,9 @@ void dcn35_prepare_bandwidth(
 
 		if (dc->hwss.root_clock_control)
 			dc->hwss.root_clock_control(dc, &pg_update_state, true);
-
+		/*power up required HW block*/
-		if (dc->hwss.block_power_control)
+		if (dc->hwss.hw_block_power_up)
-			dc->hwss.block_power_control(dc, &pg_update_state, true);
+			dc->hwss.hw_block_power_up(dc, &pg_update_state);
 	}
 
 	dcn20_prepare_bandwidth(dc, context);
@@ -1235,9 +1291,9 @@ void dcn35_optimize_bandwidth(
 
 	if (dc->hwss.calc_blocks_to_gate) {
 		dc->hwss.calc_blocks_to_gate(dc, context, &pg_update_state);
-
+		/*try to power down unused block*/
-		if (dc->hwss.block_power_control)
+		if (dc->hwss.hw_block_power_down)
-			dc->hwss.block_power_control(dc, &pg_update_state, false);
+			dc->hwss.hw_block_power_down(dc, &pg_update_state);
 
 		if (dc->hwss.root_clock_control)
 			dc->hwss.root_clock_control(dc, &pg_update_state, false);

drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.h

@@ -63,8 +63,10 @@ void dcn35_calc_blocks_to_gate(struct dc *dc, struct dc_state *context,
 	struct pg_block_update *update_state);
 void dcn35_calc_blocks_to_ungate(struct dc *dc, struct dc_state *context,
 	struct pg_block_update *update_state);
-void dcn35_block_power_control(struct dc *dc,
+void dcn35_hw_block_power_up(struct dc *dc,
-	struct pg_block_update *update_state, bool power_on);
+	struct pg_block_update *update_state);
+void dcn35_hw_block_power_down(struct dc *dc,
+	struct pg_block_update *update_state);
 void dcn35_root_clock_control(struct dc *dc,
 	struct pg_block_update *update_state, bool power_on);
 

drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h

@@ -414,8 +414,10 @@ struct hw_sequencer_funcs {
 		struct pg_block_update *update_state);
 	void (*calc_blocks_to_ungate)(struct dc *dc, struct dc_state *context,
 		struct pg_block_update *update_state);
-	void (*block_power_control)(struct dc *dc,
+	void (*hw_block_power_up)(struct dc *dc,
-		struct pg_block_update *update_state, bool power_on);
+		struct pg_block_update *update_state);
+	void (*hw_block_power_down)(struct dc *dc,
+		struct pg_block_update *update_state);
 	void (*root_clock_control)(struct dc *dc,
 		struct pg_block_update *update_state, bool power_on);
 	void (*set_idle_state)(const struct dc *dc, bool allow_idle);

drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h

@@ -59,8 +59,8 @@ enum dentist_dispclk_change_mode {
 struct dp_dto_params {
 	int otg_inst;
 	enum signal_type signal;
-	long long pixclk_hz;
+	uint64_t pixclk_hz;
-	long long refclk_hz;
+	uint64_t refclk_hz;
 };
 
 enum pixel_rate_div {

drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c

@@ -412,12 +412,18 @@ static enum dc_link_rate get_cable_max_link_rate(struct dc_link *link)
 {
 	enum dc_link_rate cable_max_link_rate = LINK_RATE_UNKNOWN;
 
-	if (link->dpcd_caps.cable_id.bits.UHBR10_20_CAPABILITY & DP_UHBR20)
+	if (link->dpcd_caps.cable_id.bits.UHBR10_20_CAPABILITY & DP_UHBR20) {
 		cable_max_link_rate = LINK_RATE_UHBR20;
-	else if (link->dpcd_caps.cable_id.bits.UHBR13_5_CAPABILITY)
+	} else if (link->dpcd_caps.cable_id.bits.UHBR13_5_CAPABILITY) {
 		cable_max_link_rate = LINK_RATE_UHBR13_5;
-	else if (link->dpcd_caps.cable_id.bits.UHBR10_20_CAPABILITY & DP_UHBR10)
+	} else if (link->dpcd_caps.cable_id.bits.UHBR10_20_CAPABILITY & DP_UHBR10) {
-		cable_max_link_rate = LINK_RATE_UHBR10;
+		// allow DP40 cables to do UHBR13.5 for passive or unknown cable type
+		if (link->dpcd_caps.cable_id.bits.CABLE_TYPE < 2) {
+			cable_max_link_rate = LINK_RATE_UHBR13_5;
+		} else {
+			cable_max_link_rate = LINK_RATE_UHBR10;
+		}
+	}
 
 	return cable_max_link_rate;
 }

drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c

@@ -717,6 +717,8 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.disable_dcc = DCC_ENABLE,
 	.disable_dpp_power_gate = true,
 	.disable_hubp_power_gate = true,
+	.disable_optc_power_gate = true, /*should the same as above two*/
+	.disable_hpo_power_gate = true, /*dmubfw force domain25 on*/
 	.disable_clock_gate = false,
 	.disable_dsc_power_gate = true,
 	.vsr_support = true,

drivers/gpu/drm/amd/display/dmub/dmub_srv.h

@@ -150,6 +150,13 @@ enum dmub_memory_access_type {
 	DMUB_MEMORY_ACCESS_DMA
 };
 
+/* enum dmub_power_state type - to track DC power state in dmub_srv */
+enum dmub_srv_power_state_type {
+	DMUB_POWER_STATE_UNDEFINED = 0,
+	DMUB_POWER_STATE_D0 = 1,
+	DMUB_POWER_STATE_D3 = 8
+};
+
 /**
  * struct dmub_region - dmub hw memory region
  * @base: base address for region, must be 256 byte aligned
@@ -485,6 +492,8 @@ struct dmub_srv {
 	/* Feature capabilities reported by fw */
 	struct dmub_feature_caps feature_caps;
 	struct dmub_visual_confirm_color visual_confirm_color;
+
+	enum dmub_srv_power_state_type power_state;
 };
 
 /**
@@ -889,6 +898,18 @@ enum dmub_status dmub_srv_clear_inbox0_ack(struct dmub_srv *dmub);
  */
 void dmub_srv_subvp_save_surf_addr(struct dmub_srv *dmub, const struct dc_plane_address *addr, uint8_t subvp_index);
 
+/**
+ * dmub_srv_set_power_state() - Track DC power state in dmub_srv
+ * @dmub: The dmub service
+ * @power_state: DC power state setting
+ *
+ * Store DC power state in dmub_srv.  If dmub_srv is in D3, then don't send messages to DMUB
+ *
+ * Return:
+ *   void
+ */
+void dmub_srv_set_power_state(struct dmub_srv *dmub, enum dmub_srv_power_state_type dmub_srv_power_state);
+
 #if defined(__cplusplus)
 }
 #endif

drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c

@@ -713,6 +713,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub,
 		dmub->hw_funcs.reset_release(dmub);
 
 	dmub->hw_init = true;
+	dmub->power_state = DMUB_POWER_STATE_D0;
 
 	return DMUB_STATUS_OK;
 }
@@ -766,6 +767,9 @@ enum dmub_status dmub_srv_cmd_queue(struct dmub_srv *dmub,
 	if (!dmub->hw_init)
 		return DMUB_STATUS_INVALID;
 
+	if (dmub->power_state != DMUB_POWER_STATE_D0)
+		return DMUB_STATUS_INVALID;
+
 	if (dmub->inbox1_rb.rptr > dmub->inbox1_rb.capacity ||
 	    dmub->inbox1_rb.wrpt > dmub->inbox1_rb.capacity) {
 		return DMUB_STATUS_HW_FAILURE;
@@ -784,6 +788,9 @@ enum dmub_status dmub_srv_cmd_execute(struct dmub_srv *dmub)
 	if (!dmub->hw_init)
 		return DMUB_STATUS_INVALID;
 
+	if (dmub->power_state != DMUB_POWER_STATE_D0)
+		return DMUB_STATUS_INVALID;
+
 	/**
 	 * Read back all the queued commands to ensure that they've
 	 * been flushed to framebuffer memory. Otherwise DMCUB might
@@ -1100,3 +1107,11 @@ void dmub_srv_subvp_save_surf_addr(struct dmub_srv *dmub, const struct dc_plane_
 				subvp_index);
 	}
 }
+
+void dmub_srv_set_power_state(struct dmub_srv *dmub, enum dmub_srv_power_state_type dmub_srv_power_state)
+{
+	if (!dmub || !dmub->hw_init)
+		return;
+
+	dmub->power_state = dmub_srv_power_state;
+}

drivers/gpu/drm/amd/display/include/fixed31_32.h

@@ -69,6 +69,18 @@ static const struct fixed31_32 dc_fixpt_epsilon = { 1LL };
 static const struct fixed31_32 dc_fixpt_half = { 0x80000000LL };
 static const struct fixed31_32 dc_fixpt_one = { 0x100000000LL };
 
+static inline struct fixed31_32 dc_fixpt_from_s3132(__u64 x)
+{
+	struct fixed31_32 val;
+
+	/* If negative, convert to 2's complement. */
+	if (x & (1ULL << 63))
+		x = -(x & ~(1ULL << 63));
+
+	val.value = x;
+	return val;
+}
+
 /*
  * @brief
  * Initialization routines

drivers/gpu/drm/amd/display/modules/power/power_helpers.c

@@ -839,6 +839,8 @@ bool is_psr_su_specific_panel(struct dc_link *link)
 				((dpcd_caps->sink_dev_id_str[1] == 0x08 && dpcd_caps->sink_dev_id_str[0] == 0x08) ||
 				(dpcd_caps->sink_dev_id_str[1] == 0x08 && dpcd_caps->sink_dev_id_str[0] == 0x07)))
 				isPSRSUSupported = false;
+			else if (dpcd_caps->sink_dev_id_str[1] == 0x08 && dpcd_caps->sink_dev_id_str[0] == 0x03)
+				isPSRSUSupported = false;
 			else if (dpcd_caps->psr_info.force_psrsu_cap == 0x1)
 				isPSRSUSupported = true;
 		}

drivers/gpu/drm/amd/include/mes_v11_api_def.h

@@ -572,7 +572,8 @@ struct SET_SHADER_DEBUGGER {
 		struct {
 			uint32_t single_memop : 1;  /* SQ_DEBUG.single_memop */
 			uint32_t single_alu_op : 1; /* SQ_DEBUG.single_alu_op */
-			uint32_t reserved : 30;
+			uint32_t reserved : 29;
+			uint32_t process_ctx_flush : 1;
 		};
 		uint32_t u32all;
 	} flags;

drivers/gpu/drm/amd/pm/amdgpu_dpm.c

@@ -616,6 +616,16 @@ void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable)
 			  enable ? "enable" : "disable", ret);
 }
 
+void amdgpu_dpm_enable_vpe(struct amdgpu_device *adev, bool enable)
+{
+	int ret = 0;
+
+	ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_VPE, !enable);
+	if (ret)
+		DRM_ERROR("Dpm %s vpe failed, ret = %d.\n",
+			  enable ? "enable" : "disable", ret);
+}
+
 int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version)
 {
 	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;

drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h

@@ -445,6 +445,7 @@ void amdgpu_dpm_compute_clocks(struct amdgpu_device *adev);
 void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable);
 void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable);
 void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable);
+void amdgpu_dpm_enable_vpe(struct amdgpu_device *adev, bool enable);
 int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version);
 int amdgpu_dpm_handle_passthrough_sbr(struct amdgpu_device *adev, bool enable);
 int amdgpu_dpm_send_hbm_bad_pages_num(struct amdgpu_device *adev, uint32_t size);

drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c

@@ -2735,10 +2735,8 @@ static int kv_parse_power_table(struct amdgpu_device *adev)
 		non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
 			&non_clock_info_array->nonClockInfo[non_clock_array_index];
 		ps = kzalloc(sizeof(struct kv_ps), GFP_KERNEL);
-		if (ps == NULL) {
+		if (ps == NULL)
-			kfree(adev->pm.dpm.ps);
 			return -ENOMEM;
-		}
 		adev->pm.dpm.ps[i].ps_priv = ps;
 		k = 0;
 		idx = (u8 *)&power_state->v2.clockInfoIndex[0];

drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c

@@ -272,10 +272,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				 le16_to_cpu(power_info->pplib4.usVddcDependencyOnSCLKOffset));
 			ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
 								 dep_table);
-			if (ret) {
+			if (ret)
-				amdgpu_free_extended_power_table(adev);
 				return ret;
-			}
 		}
 		if (power_info->pplib4.usVddciDependencyOnMCLKOffset) {
 			dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *)
@@ -283,10 +281,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				 le16_to_cpu(power_info->pplib4.usVddciDependencyOnMCLKOffset));
 			ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
 								 dep_table);
-			if (ret) {
+			if (ret)
-				amdgpu_free_extended_power_table(adev);
 				return ret;
-			}
 		}
 		if (power_info->pplib4.usVddcDependencyOnMCLKOffset) {
 			dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *)
@@ -294,10 +290,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				 le16_to_cpu(power_info->pplib4.usVddcDependencyOnMCLKOffset));
 			ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
 								 dep_table);
-			if (ret) {
+			if (ret)
-				amdgpu_free_extended_power_table(adev);
 				return ret;
-			}
 		}
 		if (power_info->pplib4.usMvddDependencyOnMCLKOffset) {
 			dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *)
@@ -305,10 +299,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				 le16_to_cpu(power_info->pplib4.usMvddDependencyOnMCLKOffset));
 			ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk,
 								 dep_table);
-			if (ret) {
+			if (ret)
-				amdgpu_free_extended_power_table(adev);
 				return ret;
-			}
 		}
 		if (power_info->pplib4.usMaxClockVoltageOnDCOffset) {
 			ATOM_PPLIB_Clock_Voltage_Limit_Table *clk_v =
@@ -339,10 +331,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				kcalloc(psl->ucNumEntries,
 					sizeof(struct amdgpu_phase_shedding_limits_entry),
 					GFP_KERNEL);
-			if (!adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries) {
+			if (!adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries)
-				amdgpu_free_extended_power_table(adev);
 				return -ENOMEM;
-			}
 
 			entry = &psl->entries[0];
 			for (i = 0; i < psl->ucNumEntries; i++) {
@@ -383,10 +373,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 			ATOM_PPLIB_CAC_Leakage_Record *entry;
 			u32 size = cac_table->ucNumEntries * sizeof(struct amdgpu_cac_leakage_table);
 			adev->pm.dpm.dyn_state.cac_leakage_table.entries = kzalloc(size, GFP_KERNEL);
-			if (!adev->pm.dpm.dyn_state.cac_leakage_table.entries) {
+			if (!adev->pm.dpm.dyn_state.cac_leakage_table.entries)
-				amdgpu_free_extended_power_table(adev);
 				return -ENOMEM;
-			}
 			entry = &cac_table->entries[0];
 			for (i = 0; i < cac_table->ucNumEntries; i++) {
 				if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_EVV) {
@@ -438,10 +426,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				sizeof(struct amdgpu_vce_clock_voltage_dependency_entry);
 			adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries =
 				kzalloc(size, GFP_KERNEL);
-			if (!adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries) {
+			if (!adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries)
-				amdgpu_free_extended_power_table(adev);
 				return -ENOMEM;
-			}
 			adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count =
 				limits->numEntries;
 			entry = &limits->entries[0];
@@ -493,10 +479,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				sizeof(struct amdgpu_uvd_clock_voltage_dependency_entry);
 			adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries =
 				kzalloc(size, GFP_KERNEL);
-			if (!adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries) {
+			if (!adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries)
-				amdgpu_free_extended_power_table(adev);
 				return -ENOMEM;
-			}
 			adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count =
 				limits->numEntries;
 			entry = &limits->entries[0];
@@ -525,10 +509,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				sizeof(struct amdgpu_clock_voltage_dependency_entry);
 			adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries =
 				kzalloc(size, GFP_KERNEL);
-			if (!adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries) {
+			if (!adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries)
-				amdgpu_free_extended_power_table(adev);
 				return -ENOMEM;
-			}
 			adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.count =
 				limits->numEntries;
 			entry = &limits->entries[0];
@@ -548,10 +530,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				 le16_to_cpu(ext_hdr->usPPMTableOffset));
 			adev->pm.dpm.dyn_state.ppm_table =
 				kzalloc(sizeof(struct amdgpu_ppm_table), GFP_KERNEL);
-			if (!adev->pm.dpm.dyn_state.ppm_table) {
+			if (!adev->pm.dpm.dyn_state.ppm_table)
-				amdgpu_free_extended_power_table(adev);
 				return -ENOMEM;
-			}
 			adev->pm.dpm.dyn_state.ppm_table->ppm_design = ppm->ucPpmDesign;
 			adev->pm.dpm.dyn_state.ppm_table->cpu_core_number =
 				le16_to_cpu(ppm->usCpuCoreNumber);
@@ -583,10 +563,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 				sizeof(struct amdgpu_clock_voltage_dependency_entry);
 			adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries =
 				kzalloc(size, GFP_KERNEL);
-			if (!adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries) {
+			if (!adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries)
-				amdgpu_free_extended_power_table(adev);
 				return -ENOMEM;
-			}
 			adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.count =
 				limits->numEntries;
 			entry = &limits->entries[0];
@@ -606,10 +584,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 			ATOM_PowerTune_Table *pt;
 			adev->pm.dpm.dyn_state.cac_tdp_table =
 				kzalloc(sizeof(struct amdgpu_cac_tdp_table), GFP_KERNEL);
-			if (!adev->pm.dpm.dyn_state.cac_tdp_table) {
+			if (!adev->pm.dpm.dyn_state.cac_tdp_table)
-				amdgpu_free_extended_power_table(adev);
 				return -ENOMEM;
-			}
 			if (rev > 0) {
 				ATOM_PPLIB_POWERTUNE_Table_V1 *ppt = (ATOM_PPLIB_POWERTUNE_Table_V1 *)
 					(mode_info->atom_context->bios + data_offset +
@@ -645,10 +621,8 @@ int amdgpu_parse_extended_power_table(struct amdgpu_device *adev)
 			ret = amdgpu_parse_clk_voltage_dep_table(
 					&adev->pm.dpm.dyn_state.vddgfx_dependency_on_sclk,
 					dep_table);
-			if (ret) {
+			if (ret)
-				kfree(adev->pm.dpm.dyn_state.vddgfx_dependency_on_sclk.entries);
 				return ret;
-			}
 		}
 	}
 

drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c

@@ -7379,10 +7379,9 @@ static int si_dpm_init(struct amdgpu_device *adev)
 		kcalloc(4,
 			sizeof(struct amdgpu_clock_voltage_dependency_entry),
 			GFP_KERNEL);
-	if (!adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries) {
+	if (!adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries)
-		amdgpu_free_extended_power_table(adev);
 		return -ENOMEM;
-	}
+
 	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.count = 4;
 	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[0].clk = 0;
 	adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[0].v = 0;

drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c

@@ -1322,6 +1322,187 @@ static int smu_get_thermal_temperature_range(struct smu_context *smu)
 	return ret;
 }
 
+/**
+ * smu_wbrf_handle_exclusion_ranges - consume the wbrf exclusion ranges
+ *
+ * @smu: smu_context pointer
+ *
+ * Retrieve the wbrf exclusion ranges and send them to PMFW for proper handling.
+ * Returns 0 on success, error on failure.
+ */
+static int smu_wbrf_handle_exclusion_ranges(struct smu_context *smu)
+{
+	struct wbrf_ranges_in_out wbrf_exclusion = {0};
+	struct freq_band_range *wifi_bands = wbrf_exclusion.band_list;
+	struct amdgpu_device *adev = smu->adev;
+	uint32_t num_of_wbrf_ranges = MAX_NUM_OF_WBRF_RANGES;
+	uint64_t start, end;
+	int ret, i, j;
+
+	ret = amd_wbrf_retrieve_freq_band(adev->dev, &wbrf_exclusion);
+	if (ret) {
+		dev_err(adev->dev, "Failed to retrieve exclusion ranges!\n");
+		return ret;
+	}
+
+	/*
+	 * The exclusion ranges array we got might be filled with holes and duplicate
+	 * entries. For example:
+	 * {(2400, 2500), (0, 0), (6882, 6962), (2400, 2500), (0, 0), (6117, 6189), (0, 0)...}
+	 * We need to do some sortups to eliminate those holes and duplicate entries.
+	 * Expected output: {(2400, 2500), (6117, 6189), (6882, 6962), (0, 0)...}
+	 */
+	for (i = 0; i < num_of_wbrf_ranges; i++) {
+		start = wifi_bands[i].start;
+		end = wifi_bands[i].end;
+
+		/* get the last valid entry to fill the intermediate hole */
+		if (!start && !end) {
+			for (j = num_of_wbrf_ranges - 1; j > i; j--)
+				if (wifi_bands[j].start && wifi_bands[j].end)
+					break;
+
+			/* no valid entry left */
+			if (j <= i)
+				break;
+
+			start = wifi_bands[i].start = wifi_bands[j].start;
+			end = wifi_bands[i].end = wifi_bands[j].end;
+			wifi_bands[j].start = 0;
+			wifi_bands[j].end = 0;
+			num_of_wbrf_ranges = j;
+		}
+
+		/* eliminate duplicate entries */
+		for (j = i + 1; j < num_of_wbrf_ranges; j++) {
+			if ((wifi_bands[j].start == start) && (wifi_bands[j].end == end)) {
+				wifi_bands[j].start = 0;
+				wifi_bands[j].end = 0;
+			}
+		}
+	}
+
+	/* Send the sorted wifi_bands to PMFW */
+	ret = smu_set_wbrf_exclusion_ranges(smu, wifi_bands);
+	/* Try to set the wifi_bands again */
+	if (unlikely(ret == -EBUSY)) {
+		mdelay(5);
+		ret = smu_set_wbrf_exclusion_ranges(smu, wifi_bands);
+	}
+
+	return ret;
+}
+
+/**
+ * smu_wbrf_event_handler - handle notify events
+ *
+ * @nb: notifier block
+ * @action: event type
+ * @_arg: event data
+ *
+ * Calls relevant amdgpu function in response to wbrf event
+ * notification from kernel.
+ */
+static int smu_wbrf_event_handler(struct notifier_block *nb,
+				  unsigned long action, void *_arg)
+{
+	struct smu_context *smu = container_of(nb, struct smu_context, wbrf_notifier);
+
+	switch (action) {
+	case WBRF_CHANGED:
+		schedule_delayed_work(&smu->wbrf_delayed_work,
+				      msecs_to_jiffies(SMU_WBRF_EVENT_HANDLING_PACE));
+		break;
+	default:
+		return NOTIFY_DONE;
+	}
+
+	return NOTIFY_OK;
+}
+
+/**
+ * smu_wbrf_delayed_work_handler - callback on delayed work timer expired
+ *
+ * @work: struct work_struct pointer
+ *
+ * Flood is over and driver will consume the latest exclusion ranges.
+ */
+static void smu_wbrf_delayed_work_handler(struct work_struct *work)
+{
+	struct smu_context *smu = container_of(work, struct smu_context, wbrf_delayed_work.work);
+
+	smu_wbrf_handle_exclusion_ranges(smu);
+}
+
+/**
+ * smu_wbrf_support_check - check wbrf support
+ *
+ * @smu: smu_context pointer
+ *
+ * Verifies the ACPI interface whether wbrf is supported.
+ */
+static void smu_wbrf_support_check(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	smu->wbrf_supported = smu_is_asic_wbrf_supported(smu) && amdgpu_wbrf &&
+							acpi_amd_wbrf_supported_consumer(adev->dev);
+
+	if (smu->wbrf_supported)
+		dev_info(adev->dev, "RF interference mitigation is supported\n");
+}
+
+/**
+ * smu_wbrf_init - init driver wbrf support
+ *
+ * @smu: smu_context pointer
+ *
+ * Verifies the AMD ACPI interfaces and registers with the wbrf
+ * notifier chain if wbrf feature is supported.
+ * Returns 0 on success, error on failure.
+ */
+static int smu_wbrf_init(struct smu_context *smu)
+{
+	int ret;
+
+	if (!smu->wbrf_supported)
+		return 0;
+
+	INIT_DELAYED_WORK(&smu->wbrf_delayed_work, smu_wbrf_delayed_work_handler);
+
+	smu->wbrf_notifier.notifier_call = smu_wbrf_event_handler;
+	ret = amd_wbrf_register_notifier(&smu->wbrf_notifier);
+	if (ret)
+		return ret;
+
+	/*
+	 * Some wifiband exclusion ranges may be already there
+	 * before our driver loaded. To make sure our driver
+	 * is awared of those exclusion ranges.
+	 */
+	schedule_delayed_work(&smu->wbrf_delayed_work,
+			      msecs_to_jiffies(SMU_WBRF_EVENT_HANDLING_PACE));
+
+	return 0;
+}
+
+/**
+ * smu_wbrf_fini - tear down driver wbrf support
+ *
+ * @smu: smu_context pointer
+ *
+ * Unregisters with the wbrf notifier chain.
+ */
+static void smu_wbrf_fini(struct smu_context *smu)
+{
+	if (!smu->wbrf_supported)
+		return;
+
+	amd_wbrf_unregister_notifier(&smu->wbrf_notifier);
+
+	cancel_delayed_work_sync(&smu->wbrf_delayed_work);
+}
+
 static int smu_smc_hw_setup(struct smu_context *smu)
 {
 	struct smu_feature *feature = &smu->smu_feature;
@@ -1414,6 +1595,15 @@ static int smu_smc_hw_setup(struct smu_context *smu)
 	if (ret)
 		return ret;
 
+	/* Enable UclkShadow on wbrf supported */
+	if (smu->wbrf_supported) {
+		ret = smu_enable_uclk_shadow(smu, true);
+		if (ret) {
+			dev_err(adev->dev, "Failed to enable UclkShadow feature to support wbrf!\n");
+			return ret;
+		}
+	}
+
 	/*
 	 * With SCPM enabled, these actions(and relevant messages) are
 	 * not needed and permitted.
@@ -1512,6 +1702,15 @@ static int smu_smc_hw_setup(struct smu_context *smu)
 	 */
 	ret = smu_set_min_dcef_deep_sleep(smu,
 					  smu->smu_table.boot_values.dcefclk / 100);
+	if (ret) {
+		dev_err(adev->dev, "Error setting min deepsleep dcefclk\n");
+		return ret;
+	}
+
+	/* Init wbrf support. Properly setup the notifier */
+	ret = smu_wbrf_init(smu);
+	if (ret)
+		dev_err(adev->dev, "Error during wbrf init call\n");
 
 	return ret;
 }
@@ -1567,6 +1766,13 @@ static int smu_hw_init(void *handle)
 		return ret;
 	}
 
+	/*
+	 * Check whether wbrf is supported. This needs to be done
+	 * before SMU setup starts since part of SMU configuration
+	 * relies on this.
+	 */
+	smu_wbrf_support_check(smu);
+
 	if (smu->is_apu) {
 		ret = smu_set_gfx_imu_enable(smu);
 		if (ret)
@@ -1733,6 +1939,8 @@ static int smu_smc_hw_cleanup(struct smu_context *smu)
 	struct amdgpu_device *adev = smu->adev;
 	int ret = 0;
 
+	smu_wbrf_fini(smu);
+
 	cancel_work_sync(&smu->throttling_logging_work);
 	cancel_work_sync(&smu->interrupt_work);
 

drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h

@@ -22,6 +22,9 @@
 #ifndef __AMDGPU_SMU_H__
 #define __AMDGPU_SMU_H__
 
+#include <linux/acpi_amd_wbrf.h>
+#include <linux/units.h>
+
 #include "amdgpu.h"
 #include "kgd_pp_interface.h"
 #include "dm_pp_interface.h"
@@ -318,6 +321,7 @@ enum smu_table_id {
 	SMU_TABLE_PACE,
 	SMU_TABLE_ECCINFO,
 	SMU_TABLE_COMBO_PPTABLE,
+	SMU_TABLE_WIFIBAND,
 	SMU_TABLE_COUNT,
 };
 
@@ -471,6 +475,12 @@ struct stb_context {
 
 #define WORKLOAD_POLICY_MAX 7
 
+/*
+ * Configure wbrf event handling pace as there can be only one
+ * event processed every SMU_WBRF_EVENT_HANDLING_PACE ms.
+ */
+#define SMU_WBRF_EVENT_HANDLING_PACE	10
+
 struct smu_context {
 	struct amdgpu_device            *adev;
 	struct amdgpu_irq_src		irq_source;
@@ -570,6 +580,11 @@ struct smu_context {
 	struct delayed_work		swctf_delayed_work;
 
 	enum pp_xgmi_plpd_mode plpd_mode;
+
+	/* data structures for wbrf feature support */
+	bool				wbrf_supported;
+	struct notifier_block		wbrf_notifier;
+	struct delayed_work		wbrf_delayed_work;
 };
 
 struct i2c_adapter;
@@ -1375,6 +1390,22 @@ struct pptable_funcs {
 	 * @notify_rlc_state: Notify RLC power state to SMU.
 	 */
 	int (*notify_rlc_state)(struct smu_context *smu, bool en);
+
+	/**
+	 * @is_asic_wbrf_supported: check whether PMFW supports the wbrf feature
+	 */
+	bool (*is_asic_wbrf_supported)(struct smu_context *smu);
+
+	/**
+	 * @enable_uclk_shadow: Enable the uclk shadow feature on wbrf supported
+	 */
+	int (*enable_uclk_shadow)(struct smu_context *smu, bool enable);
+
+	/**
+	 * @set_wbrf_exclusion_ranges: notify SMU the wifi bands occupied
+	 */
+	int (*set_wbrf_exclusion_ranges)(struct smu_context *smu,
+					struct freq_band_range *exclusion_ranges);
 };
 
 typedef enum {
@@ -1501,6 +1532,17 @@ enum smu_baco_seq {
 			 __dst_size);					   \
 })
 
+typedef struct {
+	uint16_t     LowFreq;
+	uint16_t     HighFreq;
+} WifiOneBand_t;
+
+typedef struct {
+	uint32_t		WifiBandEntryNum;
+	WifiOneBand_t	WifiBandEntry[11];
+	uint32_t		MmHubPadding[8];
+} WifiBandEntryTable_t;
+
 #if !defined(SWSMU_CODE_LAYER_L2) && !defined(SWSMU_CODE_LAYER_L3) && !defined(SWSMU_CODE_LAYER_L4)
 int smu_get_power_limit(void *handle,
 			uint32_t *limit,

drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_0.h

@@ -1615,7 +1615,8 @@ typedef struct {
 #define TABLE_I2C_COMMANDS            9
 #define TABLE_DRIVER_INFO             10
 #define TABLE_ECCINFO                 11
-#define TABLE_COUNT                   12
+#define TABLE_WIFIBAND                12
+#define TABLE_COUNT                   13
 
 //IH Interupt ID
 #define IH_INTERRUPT_ID_TO_DRIVER                   0xFE

drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_7.h

@@ -1605,7 +1605,8 @@ typedef struct {
 #define TABLE_I2C_COMMANDS            9
 #define TABLE_DRIVER_INFO             10
 #define TABLE_ECCINFO                 11
-#define TABLE_COUNT                   12
+#define TABLE_WIFIBAND                12
+#define TABLE_COUNT                   13
 
 //IH Interupt ID
 #define IH_INTERRUPT_ID_TO_DRIVER                   0xFE

drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h

@@ -24,11 +24,6 @@
 #ifndef SMU14_DRIVER_IF_V14_0_0_H
 #define SMU14_DRIVER_IF_V14_0_0_H
 
-// *** IMPORTANT ***
-// SMU TEAM: Always increment the interface version if
-// any structure is changed in this file
-#define PMFW_DRIVER_IF_VERSION 7
-
 typedef struct {
   int32_t value;
   uint32_t numFractionalBits;

drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_0_ppsmc.h

@@ -138,10 +138,9 @@
 #define PPSMC_MSG_SetBadMemoryPagesRetiredFlagsPerChannel 0x4A
 #define PPSMC_MSG_SetPriorityDeltaGain           0x4B
 #define PPSMC_MSG_AllowIHHostInterrupt           0x4C
-
 #define PPSMC_MSG_DALNotPresent                  0x4E
-
+#define PPSMC_MSG_EnableUCLKShadow               0x51
-#define PPSMC_Message_Count                      0x4F
+#define PPSMC_Message_Count                      0x52
 
 //Debug Dump Message
 #define DEBUGSMC_MSG_TestMessage                    0x1

drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_7_ppsmc.h

@@ -134,6 +134,7 @@
 #define PPSMC_MSG_SetBadMemoryPagesRetiredFlagsPerChannel 0x4A
 #define PPSMC_MSG_SetPriorityDeltaGain           0x4B
 #define PPSMC_MSG_AllowIHHostInterrupt           0x4C
-#define PPSMC_Message_Count                      0x4D
+#define PPSMC_MSG_EnableUCLKShadow               0x51
+#define PPSMC_Message_Count                      0x52
 
 #endif

drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h

@@ -260,7 +260,8 @@
 	__SMU_DUMMY_MAP(PowerDownUmsch),	\
 	__SMU_DUMMY_MAP(SetSoftMaxVpe),	\
 	__SMU_DUMMY_MAP(SetSoftMinVpe), \
-	__SMU_DUMMY_MAP(GetMetricsVersion),
+	__SMU_DUMMY_MAP(GetMetricsVersion), \
+	__SMU_DUMMY_MAP(EnableUCLKShadow),
 
 #undef __SMU_DUMMY_MAP
 #define __SMU_DUMMY_MAP(type)	SMU_MSG_##type

drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h

@@ -212,10 +212,6 @@ int smu_v13_0_get_max_sustainable_clocks_by_dc(struct smu_context *smu,
 
 bool smu_v13_0_baco_is_support(struct smu_context *smu);
 
-enum smu_baco_state smu_v13_0_baco_get_state(struct smu_context *smu);
-
-int smu_v13_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state);
-
 int smu_v13_0_baco_enter(struct smu_context *smu);
 int smu_v13_0_baco_exit(struct smu_context *smu);
 
@@ -298,5 +294,9 @@ int smu_v13_0_update_pcie_parameters(struct smu_context *smu,
 
 int smu_v13_0_disable_pmfw_state(struct smu_context *smu);
 
+int smu_v13_0_enable_uclk_shadow(struct smu_context *smu, bool enable);
+
+int smu_v13_0_set_wbrf_exclusion_ranges(struct smu_context *smu,
+						 struct freq_band_range *exclusion_ranges);
 #endif
 #endif

drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h

@@ -26,8 +26,8 @@
 #include "amdgpu_smu.h"
 
 #define SMU14_DRIVER_IF_VERSION_INV 0xFFFFFFFF
+#define SMU14_DRIVER_IF_VERSION_SMU_V14_0_0 0x7
 #define SMU14_DRIVER_IF_VERSION_SMU_V14_0_2 0x1
-#define SMU14_DRIVER_IF_VERSION_SMU_V14_0_0 0x6
 
 #define FEATURE_MASK(feature) (1ULL << feature)
 

drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c

@@ -2407,8 +2407,6 @@ static const struct pptable_funcs arcturus_ppt_funcs = {
 	.set_azalia_d3_pme = smu_v11_0_set_azalia_d3_pme,
 	.get_max_sustainable_clocks_by_dc = smu_v11_0_get_max_sustainable_clocks_by_dc,
 	.baco_is_support = smu_v11_0_baco_is_support,
-	.baco_get_state = smu_v11_0_baco_get_state,
-	.baco_set_state = smu_v11_0_baco_set_state,
 	.baco_enter = smu_v11_0_baco_enter,
 	.baco_exit = smu_v11_0_baco_exit,
 	.get_dpm_ultimate_freq = smu_v11_0_get_dpm_ultimate_freq,

drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c

@@ -3537,8 +3537,6 @@ static const struct pptable_funcs navi10_ppt_funcs = {
 	.set_azalia_d3_pme = smu_v11_0_set_azalia_d3_pme,
 	.get_max_sustainable_clocks_by_dc = smu_v11_0_get_max_sustainable_clocks_by_dc,
 	.baco_is_support = smu_v11_0_baco_is_support,
-	.baco_get_state = smu_v11_0_baco_get_state,
-	.baco_set_state = smu_v11_0_baco_set_state,
 	.baco_enter = navi10_baco_enter,
 	.baco_exit = navi10_baco_exit,
 	.get_dpm_ultimate_freq = smu_v11_0_get_dpm_ultimate_freq,

drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c

@@ -4428,8 +4428,6 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = {
 	.set_azalia_d3_pme = smu_v11_0_set_azalia_d3_pme,
 	.get_max_sustainable_clocks_by_dc = smu_v11_0_get_max_sustainable_clocks_by_dc,
 	.baco_is_support = smu_v11_0_baco_is_support,
-	.baco_get_state = smu_v11_0_baco_get_state,
-	.baco_set_state = smu_v11_0_baco_set_state,
 	.baco_enter = sienna_cichlid_baco_enter,
 	.baco_exit = sienna_cichlid_baco_exit,
 	.mode1_reset_is_support = sienna_cichlid_is_mode1_reset_supported,

drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c

@@ -2221,33 +2221,14 @@ static int smu_v13_0_baco_set_armd3_sequence(struct smu_context *smu,
 	return 0;
 }
 
-bool smu_v13_0_baco_is_support(struct smu_context *smu)
+static enum smu_baco_state smu_v13_0_baco_get_state(struct smu_context *smu)
-{
-	struct smu_baco_context *smu_baco = &smu->smu_baco;
-
-	if (amdgpu_sriov_vf(smu->adev) ||
-	    !smu_baco->platform_support)
-		return false;
-
-	/* return true if ASIC is in BACO state already */
-	if (smu_v13_0_baco_get_state(smu) == SMU_BACO_STATE_ENTER)
-		return true;
-
-	if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_BACO_BIT) &&
-	    !smu_cmn_feature_is_enabled(smu, SMU_FEATURE_BACO_BIT))
-		return false;
-
-	return true;
-}
-
-enum smu_baco_state smu_v13_0_baco_get_state(struct smu_context *smu)
 {
 	struct smu_baco_context *smu_baco = &smu->smu_baco;
 
 	return smu_baco->state;
 }
 
-int smu_v13_0_baco_set_state(struct smu_context *smu,
+static int smu_v13_0_baco_set_state(struct smu_context *smu,
 			     enum smu_baco_state state)
 {
 	struct smu_baco_context *smu_baco = &smu->smu_baco;
@@ -2281,6 +2262,24 @@ int smu_v13_0_baco_set_state(struct smu_context *smu,
 	return ret;
 }
 
+bool smu_v13_0_baco_is_support(struct smu_context *smu)
+{
+	struct smu_baco_context *smu_baco = &smu->smu_baco;
+
+	if (amdgpu_sriov_vf(smu->adev) || !smu_baco->platform_support)
+		return false;
+
+	/* return true if ASIC is in BACO state already */
+	if (smu_v13_0_baco_get_state(smu) == SMU_BACO_STATE_ENTER)
+		return true;
+
+	if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_BACO_BIT) &&
+	    !smu_cmn_feature_is_enabled(smu, SMU_FEATURE_BACO_BIT))
+		return false;
+
+	return true;
+}
+
 int smu_v13_0_baco_enter(struct smu_context *smu)
 {
 	struct smu_baco_context *smu_baco = &smu->smu_baco;
@@ -2508,3 +2507,51 @@ int smu_v13_0_disable_pmfw_state(struct smu_context *smu)
 
 	return ret == 0 ? 0 : -EINVAL;
 }
+
+int smu_v13_0_enable_uclk_shadow(struct smu_context *smu, bool enable)
+{
+	return smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_EnableUCLKShadow, enable, NULL);
+}
+
+int smu_v13_0_set_wbrf_exclusion_ranges(struct smu_context *smu,
+						 struct freq_band_range *exclusion_ranges)
+{
+	WifiBandEntryTable_t wifi_bands;
+	int valid_entries = 0;
+	int ret, i;
+
+	memset(&wifi_bands, 0, sizeof(wifi_bands));
+	for (i = 0; i < ARRAY_SIZE(wifi_bands.WifiBandEntry); i++) {
+		if (!exclusion_ranges[i].start && !exclusion_ranges[i].end)
+			break;
+
+		/* PMFW expects the inputs to be in Mhz unit */
+		wifi_bands.WifiBandEntry[valid_entries].LowFreq =
+			DIV_ROUND_DOWN_ULL(exclusion_ranges[i].start, HZ_PER_MHZ);
+		wifi_bands.WifiBandEntry[valid_entries++].HighFreq =
+			DIV_ROUND_UP_ULL(exclusion_ranges[i].end, HZ_PER_MHZ);
+	}
+	wifi_bands.WifiBandEntryNum = valid_entries;
+
+	/*
+	 * Per confirm with PMFW team, WifiBandEntryNum = 0
+	 * is a valid setting.
+	 *
+	 * Considering the scenarios below:
+	 * - At first the wifi device adds an exclusion range e.g. (2400,2500) to
+	 *   BIOS and our driver gets notified. We will set WifiBandEntryNum = 1
+	 *   and pass the WifiBandEntry (2400, 2500) to PMFW.
+	 *
+	 * - Later the wifi device removes the wifiband list added above and
+	 *   our driver gets notified again. At this time, driver will set
+	 *   WifiBandEntryNum = 0 and pass an empty WifiBandEntry list to PMFW.
+	 *
+	 * - PMFW may still need to do some uclk shadow update(e.g. switching
+	 *   from shadow clock back to primary clock) on receiving this.
+	 */
+	ret = smu_cmn_update_table(smu, SMU_TABLE_WIFIBAND, 0, &wifi_bands, true);
+	if (ret)
+		dev_warn(smu->adev->dev, "Failed to set wifiband!");
+
+	return ret;
+}

drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c

@@ -169,6 +169,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_0_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(AllowIHHostInterrupt,		PPSMC_MSG_AllowIHHostInterrupt,       0),
 	MSG_MAP(ReenableAcDcInterrupt,		PPSMC_MSG_ReenableAcDcInterrupt,       0),
 	MSG_MAP(DALNotPresent,		PPSMC_MSG_DALNotPresent,       0),
+	MSG_MAP(EnableUCLKShadow,		PPSMC_MSG_EnableUCLKShadow,            0),
 };
 
 static struct cmn2asic_mapping smu_v13_0_0_clk_map[SMU_CLK_COUNT] = {
@@ -253,6 +254,7 @@ static struct cmn2asic_mapping smu_v13_0_0_table_map[SMU_TABLE_COUNT] = {
 	TAB_MAP(I2C_COMMANDS),
 	TAB_MAP(ECCINFO),
 	TAB_MAP(OVERDRIVE),
+	TAB_MAP(WIFIBAND),
 };
 
 static struct cmn2asic_mapping smu_v13_0_0_pwr_src_map[SMU_POWER_SOURCE_COUNT] = {
@@ -498,6 +500,9 @@ static int smu_v13_0_0_tables_init(struct smu_context *smu)
 			PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_ECCINFO, sizeof(EccInfoTable_t),
 			PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+	SMU_TABLE_INIT(tables, SMU_TABLE_WIFIBAND,
+		       sizeof(WifiBandEntryTable_t), PAGE_SIZE,
+		       AMDGPU_GEM_DOMAIN_VRAM);
 
 	smu_table->metrics_table = kzalloc(sizeof(SmuMetricsExternal_t), GFP_KERNEL);
 	if (!smu_table->metrics_table)
@@ -2540,16 +2545,19 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu,
 
 	workload_mask = 1 << workload_type;
 
-	/* Add optimizations for SMU13.0.0.  Reuse the power saving profile */
+	/* Add optimizations for SMU13.0.0/10.  Reuse the power saving profile */
-	if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_COMPUTE &&
+	if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_COMPUTE) {
-	    (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 0)) &&
+		if ((amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 0) &&
-	    ((smu->adev->pm.fw_version == 0x004e6601) ||
+			((smu->adev->pm.fw_version == 0x004e6601) ||
-	     (smu->adev->pm.fw_version >= 0x004e7400))) {
+			(smu->adev->pm.fw_version >= 0x004e7300))) ||
-		workload_type = smu_cmn_to_asic_specific_index(smu,
+			(amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 10) &&
-							       CMN2ASIC_MAPPING_WORKLOAD,
+			 smu->adev->pm.fw_version >= 0x00504500)) {
-							       PP_SMC_POWER_PROFILE_POWERSAVING);
+			workload_type = smu_cmn_to_asic_specific_index(smu,
-		if (workload_type >= 0)
+														   CMN2ASIC_MAPPING_WORKLOAD,
-			workload_mask |= 1 << workload_type;
+														   PP_SMC_POWER_PROFILE_POWERSAVING);
+			if (workload_type >= 0)
+				workload_mask |= 1 << workload_type;
+		}
 	}
 
 	return smu_cmn_send_smc_msg_with_param(smu,
@@ -2938,6 +2946,20 @@ static ssize_t smu_v13_0_0_get_ecc_info(struct smu_context *smu,
 	return ret;
 }
 
+static bool smu_v13_0_0_wbrf_support_check(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	switch (adev->ip_versions[MP1_HWIP][0]) {
+	case IP_VERSION(13, 0, 0):
+		return smu->smc_fw_version >= 0x004e6300;
+	case IP_VERSION(13, 0, 10):
+		return smu->smc_fw_version >= 0x00503300;
+	default:
+		return false;
+	}
+}
+
 static const struct pptable_funcs smu_v13_0_0_ppt_funcs = {
 	.get_allowed_feature_mask = smu_v13_0_0_get_allowed_feature_mask,
 	.set_default_dpm_table = smu_v13_0_0_set_default_dpm_table,
@@ -3003,8 +3025,6 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = {
 	.deep_sleep_control = smu_v13_0_deep_sleep_control,
 	.gfx_ulv_control = smu_v13_0_gfx_ulv_control,
 	.baco_is_support = smu_v13_0_baco_is_support,
-	.baco_get_state = smu_v13_0_baco_get_state,
-	.baco_set_state = smu_v13_0_baco_set_state,
 	.baco_enter = smu_v13_0_baco_enter,
 	.baco_exit = smu_v13_0_baco_exit,
 	.mode1_reset_is_support = smu_v13_0_0_is_mode1_reset_supported,
@@ -3018,6 +3038,9 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = {
 	.gpo_control = smu_v13_0_gpo_control,
 	.get_ecc_info = smu_v13_0_0_get_ecc_info,
 	.notify_display_change = smu_v13_0_notify_display_change,
+	.is_asic_wbrf_supported = smu_v13_0_0_wbrf_support_check,
+	.enable_uclk_shadow = smu_v13_0_enable_uclk_shadow,
+	.set_wbrf_exclusion_ranges = smu_v13_0_set_wbrf_exclusion_ranges,
 };
 
 void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu)

drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c

@@ -2537,13 +2537,15 @@ static int mca_pcs_xgmi_mca_get_err_count(const struct mca_ras_info *mca_ras, st
 					  uint32_t *count)
 {
 	u32 ext_error_code;
+	u32 err_cnt;
 
 	ext_error_code = MCA_REG__STATUS__ERRORCODEEXT(entry->regs[MCA_REG_IDX_STATUS]);
+	err_cnt = MCA_REG__MISC0__ERRCNT(entry->regs[MCA_REG_IDX_MISC0]);
 
 	if (type == AMDGPU_MCA_ERROR_TYPE_UE && ext_error_code == 0)
-		*count = 1;
+		*count = err_cnt;
 	else if (type == AMDGPU_MCA_ERROR_TYPE_CE && ext_error_code == 6)
-		*count = 1;
+		*count = err_cnt;
 
 	return 0;
 }

drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c

@@ -140,6 +140,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_7_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(AllowGpo,			PPSMC_MSG_SetGpoAllow,           0),
 	MSG_MAP(GetPptLimit,			PPSMC_MSG_GetPptLimit,                 0),
 	MSG_MAP(NotifyPowerSource,		PPSMC_MSG_NotifyPowerSource,           0),
+	MSG_MAP(EnableUCLKShadow,		PPSMC_MSG_EnableUCLKShadow,            0),
 };
 
 static struct cmn2asic_mapping smu_v13_0_7_clk_map[SMU_CLK_COUNT] = {
@@ -222,6 +223,7 @@ static struct cmn2asic_mapping smu_v13_0_7_table_map[SMU_TABLE_COUNT] = {
 	TAB_MAP(ACTIVITY_MONITOR_COEFF),
 	[SMU_TABLE_COMBO_PPTABLE] = {1, TABLE_COMBO_PPTABLE},
 	TAB_MAP(OVERDRIVE),
+	TAB_MAP(WIFIBAND),
 };
 
 static struct cmn2asic_mapping smu_v13_0_7_pwr_src_map[SMU_POWER_SOURCE_COUNT] = {
@@ -512,6 +514,9 @@ static int smu_v13_0_7_tables_init(struct smu_context *smu)
 		       AMDGPU_GEM_DOMAIN_VRAM);
 	SMU_TABLE_INIT(tables, SMU_TABLE_COMBO_PPTABLE, MP0_MP1_DATA_REGION_SIZE_COMBOPPTABLE,
 			PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM);
+	SMU_TABLE_INIT(tables, SMU_TABLE_WIFIBAND,
+		       sizeof(WifiBandEntryTable_t), PAGE_SIZE,
+		       AMDGPU_GEM_DOMAIN_VRAM);
 
 	smu_table->metrics_table = kzalloc(sizeof(SmuMetricsExternal_t), GFP_KERNEL);
 	if (!smu_table->metrics_table)
@@ -2535,6 +2540,11 @@ static int smu_v13_0_7_set_df_cstate(struct smu_context *smu,
 					       NULL);
 }
 
+static bool smu_v13_0_7_wbrf_support_check(struct smu_context *smu)
+{
+	return smu->smc_fw_version > 0x00524600;
+}
+
 static const struct pptable_funcs smu_v13_0_7_ppt_funcs = {
 	.get_allowed_feature_mask = smu_v13_0_7_get_allowed_feature_mask,
 	.set_default_dpm_table = smu_v13_0_7_set_default_dpm_table,
@@ -2594,8 +2604,6 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = {
 	.get_pp_feature_mask = smu_cmn_get_pp_feature_mask,
 	.set_pp_feature_mask = smu_cmn_set_pp_feature_mask,
 	.baco_is_support = smu_v13_0_baco_is_support,
-	.baco_get_state = smu_v13_0_baco_get_state,
-	.baco_set_state = smu_v13_0_baco_set_state,
 	.baco_enter = smu_v13_0_baco_enter,
 	.baco_exit = smu_v13_0_baco_exit,
 	.mode1_reset_is_support = smu_v13_0_7_is_mode1_reset_supported,
@@ -2603,6 +2611,9 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = {
 	.set_mp1_state = smu_v13_0_7_set_mp1_state,
 	.set_df_cstate = smu_v13_0_7_set_df_cstate,
 	.gpo_control = smu_v13_0_gpo_control,
+	.is_asic_wbrf_supported = smu_v13_0_7_wbrf_support_check,
+	.enable_uclk_shadow = smu_v13_0_enable_uclk_shadow,
+	.set_wbrf_exclusion_ranges = smu_v13_0_set_wbrf_exclusion_ranges,
 };
 
 void smu_v13_0_7_set_ppt_funcs(struct smu_context *smu)

drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c

@@ -224,7 +224,7 @@ int smu_v14_0_check_fw_version(struct smu_context *smu)
 	if (smu->is_apu)
 		adev->pm.fw_version = smu_version;
 
-	switch (adev->ip_versions[MP1_HWIP][0]) {
+	switch (amdgpu_ip_version(adev, MP1_HWIP, 0)) {
 	case IP_VERSION(14, 0, 2):
 		smu->smc_driver_if_version = SMU14_DRIVER_IF_VERSION_SMU_V14_0_2;
 		break;
@@ -235,7 +235,7 @@ int smu_v14_0_check_fw_version(struct smu_context *smu)
 		break;
 	default:
 		dev_err(adev->dev, "smu unsupported IP version: 0x%x.\n",
-			adev->ip_versions[MP1_HWIP][0]);
+			amdgpu_ip_version(adev, MP1_HWIP, 0));
 		smu->smc_driver_if_version = SMU14_DRIVER_IF_VERSION_INV;
 		break;
 	}
@@ -733,7 +733,7 @@ int smu_v14_0_gfx_off_control(struct smu_context *smu, bool enable)
 	int ret = 0;
 	struct amdgpu_device *adev = smu->adev;
 
-	switch (adev->ip_versions[MP1_HWIP][0]) {
+	switch (amdgpu_ip_version(adev, MP1_HWIP, 0)) {
 	case IP_VERSION(14, 0, 2):
 	case IP_VERSION(14, 0, 0):
 		if (!(adev->pm.pp_feature & PP_GFXOFF_MASK))

drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c

@@ -1085,6 +1085,25 @@ static int smu_v14_0_0_set_umsch_mm_enable(struct smu_context *smu,
 					       0, NULL);
 }
 
+static int smu_14_0_0_get_dpm_table(struct smu_context *smu, struct dpm_clocks *clock_table)
+{
+	DpmClocks_t *clk_table = smu->smu_table.clocks_table;
+	uint8_t idx;
+
+	/* Only the Clock information of SOC and VPE is copied to provide VPE DPM settings for use. */
+	for (idx = 0; idx < NUM_SOCCLK_DPM_LEVELS; idx++) {
+		clock_table->SocClocks[idx].Freq = (idx < clk_table->NumSocClkLevelsEnabled) ? clk_table->SocClocks[idx]:0;
+		clock_table->SocClocks[idx].Vol = 0;
+	}
+
+	for (idx = 0; idx < NUM_VPE_DPM_LEVELS; idx++) {
+		clock_table->VPEClocks[idx].Freq = (idx < clk_table->VpeClkLevelsEnabled) ? clk_table->VPEClocks[idx]:0;
+		clock_table->VPEClocks[idx].Vol = 0;
+	}
+
+	return 0;
+}
+
 static const struct pptable_funcs smu_v14_0_0_ppt_funcs = {
 	.check_fw_status = smu_v14_0_check_fw_status,
 	.check_fw_version = smu_v14_0_check_fw_version,
@@ -1115,6 +1134,7 @@ static const struct pptable_funcs smu_v14_0_0_ppt_funcs = {
 	.set_gfx_power_up_by_imu = smu_v14_0_set_gfx_power_up_by_imu,
 	.dpm_set_vpe_enable = smu_v14_0_0_set_vpe_enable,
 	.dpm_set_umsch_mm_enable = smu_v14_0_0_set_umsch_mm_enable,
+	.get_dpm_clock_table = smu_14_0_0_get_dpm_table,
 };
 
 static void smu_v14_0_0_set_smu_mailbox_registers(struct smu_context *smu)

drivers/gpu/drm/amd/pm/swsmu/smu_internal.h

@@ -98,6 +98,9 @@
 #define smu_set_config_table(smu, config_table)				smu_ppt_funcs(set_config_table, -EOPNOTSUPP, smu, config_table)
 #define smu_init_pptable_microcode(smu)					smu_ppt_funcs(init_pptable_microcode, 0, smu)
 #define smu_notify_rlc_state(smu, en)					smu_ppt_funcs(notify_rlc_state, 0, smu, en)
+#define smu_is_asic_wbrf_supported(smu)			smu_ppt_funcs(is_asic_wbrf_supported, false, smu)
+#define smu_enable_uclk_shadow(smu, enable)		smu_ppt_funcs(enable_uclk_shadow, 0, smu, enable)
+#define smu_set_wbrf_exclusion_ranges(smu, freq_band_range)		smu_ppt_funcs(set_wbrf_exclusion_ranges, -EOPNOTSUPP, smu, freq_band_range)
 
 #endif
 #endif