Merge tag 'amd-drm-next-6.18-2025-09-26' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.18-2025-09-26:

amdgpu:
- Misc fixes
- Misc cleanups
- SMU 13.x fixes
- MES fix
- VCN 5.0.1 reset fixes
- DCN 3.2 watermark fixes
- AVI infoframe fixes
- PSR fix
- Brightness fixes
- DCN 3.1.4 fixes
- DCN 3.1+ DTM fixes
- DCN powergating fixes
- DMUB fixes
- DCN/SMU cleanup
- DCN stutter fixes
- DCN 3.5 fixes
- GAMMA_LUT fixes
- Add UserQ documentation
- GC 9.4 reset fixes
- Enforce isolation cleanups
- UserQ fixes
- DC/non-DC common modes cleanup
- DCE6-10 fixes

amdkfd:
- Fix a race in sw_fini
- Switch partition fix

Signed-off-by: Dave Airlie <[e-mail]>

From: Alex Deucher <[e-mail]>
Link: https://lore.kernel.org/r/[e-mail]

Documentation/gpu/amdgpu/index.rst

···
    module-parameters
    gc/index
    display/index
+   userq
    flashing
    xgmi
    ras

Documentation/gpu/amdgpu/userq.rst

···
+==================
+ User Mode Queues
+==================
+
+Introduction
+============
+
+Similar to the KFD, GPU engine queues move into userspace.  The idea is to let
+user processes manage their submissions to the GPU engines directly, bypassing
+IOCTL calls to the driver to submit work.  This reduces overhead and also allows
+the GPU to submit work to itself.  Applications can set up work graphs of jobs
+across multiple GPU engines without needing trips through the CPU.
+
+UMDs directly interface with firmware via per application shared memory areas.
+The main vehicle for this is queue.  A queue is a ring buffer with a read
+pointer (rptr) and a write pointer (wptr).  The UMD writes IP specific packets
+into the queue and the firmware processes those packets, kicking off work on the
+GPU engines.  The CPU in the application (or another queue or device) updates
+the wptr to tell the firmware how far into the ring buffer to process packets
+and the rtpr provides feedback to the UMD on how far the firmware has progressed
+in executing those packets.  When the wptr and the rptr are equal, the queue is
+idle.
+
+Theory of Operation
+===================
+
+The various engines on modern AMD GPUs support multiple queues per engine with a
+scheduling firmware which handles dynamically scheduling user queues on the
+available hardware queue slots.  When the number of user queues outnumbers the
+available hardware queue slots, the scheduling firmware dynamically maps and
+unmaps queues based on priority and time quanta.  The state of each user queue
+is managed in the kernel driver in an MQD (Memory Queue Descriptor).  This is a
+buffer in GPU accessible memory that stores the state of a user queue.  The
+scheduling firmware uses the MQD to load the queue state into an HQD (Hardware
+Queue Descriptor) when a user queue is mapped.  Each user queue requires a
+number of additional buffers which represent the ring buffer and any metadata
+needed by the engine for runtime operation.  On most engines this consists of
+the ring buffer itself, a rptr buffer (where the firmware will shadow the rptr
+to userspace), a wptr buffer (where the application will write the wptr for the
+firmware to fetch it), and a doorbell.  A doorbell is a piece of one of the
+device's MMIO BARs which can be mapped to specific user queues.  When the
+application writes to the doorbell, it will signal the firmware to take some
+action. Writing to the doorbell wakes the firmware and causes it to fetch the
+wptr and start processing the packets in the queue. Each 4K page of the doorbell
+BAR supports specific offset ranges for specific engines.  The doorbell of a
+queue must be mapped into the aperture aligned to the IP used by the queue
+(e.g., GFX, VCN, SDMA, etc.).  These doorbell apertures are set up via NBIO
+registers.  Doorbells are 32 bit or 64 bit (depending on the engine) chunks of
+the doorbell BAR.  A 4K doorbell page provides 512 64-bit doorbells for up to
+512 user queues.  A subset of each page is reserved for each IP type supported
+on the device.  The user can query the doorbell ranges for each IP via the INFO
+IOCTL.  See the IOCTL Interfaces section for more information.
+
+When an application wants to create a user queue, it allocates the necessary
+buffers for the queue (ring buffer, wptr and rptr, context save areas, etc.).
+These can be separate buffers or all part of one larger buffer.  The application
+would map the buffer(s) into its GPUVM and use the GPU virtual addresses of for
+the areas of memory they want to use for the user queue.  They would also
+allocate a doorbell page for the doorbells used by the user queues.  The
+application would then populate the MQD in the USERQ IOCTL structure with the
+GPU virtual addresses and doorbell index they want to use.  The user can also
+specify the attributes for the user queue (priority, whether the queue is secure
+for protected content, etc.).  The application would then call the USERQ
+CREATE IOCTL to create the queue using the specified MQD details in the IOCTL.
+The kernel driver then validates the MQD provided by the application and
+translates the MQD into the engine specific MQD format for the IP.  The IP
+specific MQD would be allocated and the queue would be added to the run list
+maintained by the scheduling firmware.  Once the queue has been created, the
+application can write packets directly into the queue, update the wptr, and
+write to the doorbell offset to kick off work in the user queue.
+
+When the application is done with the user queue, it would call the USERQ
+FREE IOCTL to destroy it.  The kernel driver would preempt the queue and
+remove it from the scheduling firmware's run list.  Then the IP specific MQD
+would be freed and the user queue state would be cleaned up.
+
+Some engines may require the aggregated doorbell too if the engine does not
+support doorbells from unmapped queues.  The aggregated doorbell is a special
+page of doorbell space which wakes the scheduler.  In cases where the engine may
+be oversubscribed, some queues may not be mapped.  If the doorbell is rung when
+the queue is not mapped, the engine firmware may miss the request.  Some
+scheduling firmware may work around this by polling wptr shadows when the
+hardware is oversubscribed, other engines may support doorbell updates from
+unmapped queues.  In the event that one of these options is not available, the
+kernel driver will map a page of aggregated doorbell space into each GPUVM
+space.  The UMD will then update the doorbell and wptr as normal and then write
+to the aggregated doorbell as well.
+
+Special Packets
+---------------
+
+In order to support legacy implicit synchronization, as well as mixed user and
+kernel queues, we need a synchronization mechanism that is secure.  Because
+kernel queues or memory management tasks depend on kernel fences, we need a way
+for user queues to update memory that the kernel can use for a fence, that can't
+be messed with by a bad actor.  To support this, we've added a protected fence
+packet.  This packet works by writing a monotonically increasing value to
+a memory location that only privileged clients have write access to. User
+queues only have read access.  When this packet is executed, the memory location
+is updated and other queues (kernel or user) can see the results.  The
+user application would submit this packet in their command stream.  The actual
+packet format varies from IP to IP (GFX/Compute, SDMA, VCN, etc.), but the
+behavior is the same.  The packet submission is handled in userspace.  The
+kernel driver sets up the privileged memory used for each user queue when it
+sets the queues up when the application creates them.
+
+
+Memory Management
+=================
+
+It is assumed that all buffers mapped into the GPUVM space for the process are
+valid when engines on the GPU are running.  The kernel driver will only allow
+user queues to run when all buffers are mapped.  If there is a memory event that
+requires buffer migration, the kernel driver will preempt the user queues,
+migrate buffers to where they need to be, update the GPUVM page tables and
+invaldidate the TLB, and then resume the user queues.
+
+Interaction with Kernel Queues
+==============================
+
+Depending on the IP and the scheduling firmware, you can enable kernel queues
+and user queues at the same time, however, you are limited by the HQD slots.
+Kernel queues are always mapped so any work that goes into kernel queues will
+take priority.  This limits the available HQD slots for user queues.
+
+Not all IPs will support user queues on all GPUs.  As such, UMDs will need to
+support both user queues and kernel queues depending on the IP.  For example, a
+GPU may support user queues for GFX, compute, and SDMA, but not for VCN, JPEG,
+and VPE.  UMDs need to support both.  The kernel driver provides a way to
+determine if user queues and kernel queues are supported on a per IP basis.
+UMDs can query this information via the INFO IOCTL and determine whether to use
+kernel queues or user queues for each IP.
+
+Queue Resets
+============
+
+For most engines, queues can be reset individually.  GFX, compute, and SDMA
+queues can be reset individually.  When a hung queue is detected, it can be
+reset either via the scheduling firmware or MMIO.  Since there are no kernel
+fences for most user queues, they will usually only be detected when some other
+event happens; e.g., a memory event which requires migration of buffers.  When
+the queues are preempted, if the queue is hung, the preemption will fail.
+Driver will then look up the queues that failed to preempt and reset them and
+record which queues are hung.
+
+On the UMD side, we will add a USERQ QUERY_STATUS IOCTL to query the queue
+status.  UMD will provide the queue id in the IOCTL and the kernel driver
+will check if it has already recorded the queue as hung (e.g., due to failed
+peemption) and report back the status.
+
+IOCTL Interfaces
+================
+
+GPU virtual addresses used for queues and related data (rptrs, wptrs, context
+save areas, etc.) should be validated by the kernel mode driver to prevent the
+user from specifying invalid GPU virtual addresses.  If the user provides
+invalid GPU virtual addresses or doorbell indicies, the IOCTL should return an
+error message.  These buffers should also be tracked in the kernel driver so
+that if the user attempts to unmap the buffer(s) from the GPUVM, the umap call
+would return an error.
+
+INFO
+----
+There are several new INFO queries related to user queues in order to query the
+size of user queue meta data needed for a user queue (e.g., context save areas
+or shadow buffers), whether kernel or user queues or both are supported
+for each IP type, and the offsets for each IP type in each doorbell page.
+
+USERQ
+-----
+The USERQ IOCTL is used for creating, freeing, and querying the status of user
+queues.  It supports 3 opcodes:
+
+1. CREATE - Create a user queue.  The application provides an MQD-like structure
+   that defines the type of queue and associated metadata and flags for that
+   queue type.  Returns the queue id.
+2. FREE - Free a user queue.
+3. QUERY_STATUS - Query that status of a queue.  Used to check if the queue is
+   healthy or not.  E.g., if the queue has been reset. (WIP)
+
+USERQ_SIGNAL
+------------
+The USERQ_SIGNAL IOCTL is used to provide a list of sync objects to be signaled.
+
+USERQ_WAIT
+----------
+The USERQ_WAIT IOCTL is used to provide a list of sync object to be waited on.
+
+Kernel and User Queues
+======================
+
+In order to properly validate and test performance, we have a driver option to
+select what type of queues are enabled (kernel queues, user queues or both).
+The user_queue driver parameter allows you to enable kernel queues only (0),
+user queues and kernel queues (1), and user queues only (2).  Enabling user
+queues only will free up static queue assignments that would otherwise be used
+by kernel queues for use by the scheduling firmware.  Some kernel queues are
+required for kernel driver operation and they will always be created.  When the
+kernel queues are not enabled, they are not registered with the drm scheduler
+and the CS IOCTL will reject any incoming command submissions which target those
+queue types.  Kernel queues only mirrors the behavior on all existing GPUs.
+Enabling both queues allows for backwards compatibility with old userspace while
+still supporting user queues.

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c

···
 		(*dump)[i++][1] = RREG32_SOC15_IP(GC, addr);		\
 	} while (0)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
···
 #undef HQD_N_REGS
 #define HQD_N_REGS (19+6+7+10)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c

···
 		(*dump)[i++][1] = RREG32_SOC15_IP(GC, addr);		\
 	} while (0)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
···
 #undef HQD_N_REGS
 #define HQD_N_REGS (19+6+7+12)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c

···
 		(*dump)[i++][1] = RREG32(addr);		\
 	} while (0)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
···
 #undef HQD_N_REGS
 #define HQD_N_REGS (7+11+1+12+12)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v12.c

···
 		(*dump)[i++][1] = RREG32(addr);		\
 	} while (0)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
···
 #undef HQD_N_REGS
 #define HQD_N_REGS (last_reg - first_reg + 1)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 

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

···
 		return 0;
 	}
 
-	ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages, &range);
+	ret = amdgpu_ttm_tt_get_user_pages(bo, &range);
 	if (ret) {
 		if (ret == -EAGAIN)
 			pr_debug("Failed to get user pages, try again\n");
···
 		pr_err("%s: Failed to reserve BO\n", __func__);
 		goto release_out;
 	}
+
+	amdgpu_ttm_tt_set_user_pages(bo->tbo.ttm, range);
+
 	amdgpu_bo_placement_from_domain(bo, mem->domain);
 	ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
 	if (ret)
···
 		}
 
 		/* Get updated user pages */
-		ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages,
-						   &mem->range);
+		ret = amdgpu_ttm_tt_get_user_pages(bo, &mem->range);
 		if (ret) {
 			pr_debug("Failed %d to get user pages\n", ret);
 
···
 
 			ret = 0;
 		}
+
+		amdgpu_ttm_tt_set_user_pages(bo->tbo.ttm, mem->range);
 
 		mutex_lock(&process_info->notifier_lock);
 

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

···
 	struct amdgpu_bo		*bo;
 	struct amdgpu_bo_va		*bo_va;
 	uint32_t			priority;
-	struct page			**user_pages;
 	struct hmm_range		*range;
 	bool				user_invalidated;
 };

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

···
 	struct drm_display_mode *mode = NULL;
 	struct drm_display_mode *native_mode = &amdgpu_encoder->native_mode;
 	int i;
-	static const struct mode_size {
+	int n;
+	struct mode_size {
+		char name[DRM_DISPLAY_MODE_LEN];
 		int w;
 		int h;
-	} common_modes[17] = {
-		{ 640,  480},
-		{ 720,  480},
-		{ 800,  600},
-		{ 848,  480},
-		{1024,  768},
-		{1152,  768},
-		{1280,  720},
-		{1280,  800},
-		{1280,  854},
-		{1280,  960},
-		{1280, 1024},
-		{1440,  900},
-		{1400, 1050},
-		{1680, 1050},
-		{1600, 1200},
-		{1920, 1080},
-		{1920, 1200}
+	} common_modes[] = {
+		{  "640x480",  640,  480},
+		{  "800x600",  800,  600},
+		{ "1024x768", 1024,  768},
+		{ "1280x720", 1280,  720},
+		{ "1280x800", 1280,  800},
+		{"1280x1024", 1280, 1024},
+		{ "1440x900", 1440,  900},
+		{"1680x1050", 1680, 1050},
+		{"1600x1200", 1600, 1200},
+		{"1920x1080", 1920, 1080},
+		{"1920x1200", 1920, 1200}
 	};
 
-	for (i = 0; i < 17; i++) {
+	n = ARRAY_SIZE(common_modes);
+
+	for (i = 0; i < n; i++) {
 		if (amdgpu_encoder->devices & (ATOM_DEVICE_TV_SUPPORT)) {
 			if (common_modes[i].w > 1024 ||
 			    common_modes[i].h > 768)
···
 			     common_modes[i].h == native_mode->vdisplay))
 				continue;
 		}
-		if (common_modes[i].w < 320 || common_modes[i].h < 200)
-			continue;
 
 		mode = drm_cvt_mode(dev, common_modes[i].w, common_modes[i].h, 60, false, false, false);
 		if (!mode)
 			return;
+		strscpy(mode->name, common_modes[i].name, DRM_DISPLAY_MODE_LEN);
 
 		drm_mode_probed_add(connector, mode);
 	}

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

···
 #include <linux/pagemap.h>
 #include <linux/sync_file.h>
 #include <linux/dma-buf.h>
+#include <linux/hmm.h>
 
 #include <drm/amdgpu_drm.h>
 #include <drm/drm_syncobj.h>
···
 	amdgpu_bo_list_for_each_userptr_entry(e, p->bo_list) {
 		bool userpage_invalidated = false;
 		struct amdgpu_bo *bo = e->bo;
-		int i;
 
-		e->user_pages = kvcalloc(bo->tbo.ttm->num_pages,
-					 sizeof(struct page *),
-					 GFP_KERNEL);
-		if (!e->user_pages) {
-			drm_err(adev_to_drm(p->adev), "kvmalloc_array failure\n");
-			r = -ENOMEM;
+		r = amdgpu_ttm_tt_get_user_pages(bo, &e->range);
+		if (r)
 			goto out_free_user_pages;
-		}
-
-		r = amdgpu_ttm_tt_get_user_pages(bo, e->user_pages, &e->range);
-		if (r) {
-			kvfree(e->user_pages);
-			e->user_pages = NULL;
-			goto out_free_user_pages;
-		}
 
 		for (i = 0; i < bo->tbo.ttm->num_pages; i++) {
-			if (bo->tbo.ttm->pages[i] != e->user_pages[i]) {
+			if (bo->tbo.ttm->pages[i] != hmm_pfn_to_page(e->range->hmm_pfns[i])) {
 				userpage_invalidated = true;
 				break;
 			}
···
 		}
 
 		if (amdgpu_ttm_tt_is_userptr(e->bo->tbo.ttm) &&
-		    e->user_invalidated && e->user_pages) {
+		    e->user_invalidated) {
 			amdgpu_bo_placement_from_domain(e->bo,
 							AMDGPU_GEM_DOMAIN_CPU);
 			r = ttm_bo_validate(&e->bo->tbo, &e->bo->placement,
···
 				goto out_free_user_pages;
 
 			amdgpu_ttm_tt_set_user_pages(e->bo->tbo.ttm,
-						     e->user_pages);
+						     e->range);
 		}
-
-		kvfree(e->user_pages);
-		e->user_pages = NULL;
 	}
 
 	amdgpu_cs_get_threshold_for_moves(p->adev, &p->bytes_moved_threshold,
···
 	amdgpu_bo_list_for_each_userptr_entry(e, p->bo_list) {
 		struct amdgpu_bo *bo = e->bo;
 
-		if (!e->user_pages)
-			continue;
 		amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, e->range);
-		kvfree(e->user_pages);
-		e->user_pages = NULL;
 		e->range = NULL;
 	}
 	mutex_unlock(&p->bo_list->bo_list_mutex);

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

···
  */
 MODULE_PARM_DESC(
 	freesync_video,
-	"Enable freesync modesetting optimization feature (0 = off (default), 1 = on)");
+	"Adds additional modes via VRR for refresh changes without a full modeset (0 = off (default), 1 = on)");
 module_param_named(freesync_video, amdgpu_freesync_vid_mode, uint, 0444);
 
 /**

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

···
 		goto release_object;
 
 	if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) {
-		r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages,
-						 &range);
+		r = amdgpu_ttm_tt_get_user_pages(bo, &range);
 		if (r)
 			goto release_object;
 
 		r = amdgpu_bo_reserve(bo, true);
 		if (r)
 			goto user_pages_done;
+
+		amdgpu_ttm_tt_set_user_pages(bo->tbo.ttm, range);
 
 		amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_GTT);
 		r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);

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

···
 
 int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier,
 			       uint64_t start, uint64_t npages, bool readonly,
-			       void *owner, struct page **pages,
+			       void *owner,
 			       struct hmm_range **phmm_range)
 {
 	struct hmm_range *hmm_range;
 	unsigned long end;
 	unsigned long timeout;
-	unsigned long i;
 	unsigned long *pfns;
 	int r = 0;
 
···
 
 	hmm_range->start = start;
 	hmm_range->hmm_pfns = pfns;
-
-	/*
-	 * Due to default_flags, all pages are HMM_PFN_VALID or
-	 * hmm_range_fault() fails. FIXME: The pages cannot be touched outside
-	 * the notifier_lock, and mmu_interval_read_retry() must be done first.
-	 */
-	for (i = 0; pages && i < npages; i++)
-		pages[i] = hmm_pfn_to_page(pfns[i]);
 
 	*phmm_range = hmm_range;
 

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

···
 
 int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier,
 			       uint64_t start, uint64_t npages, bool readonly,
-			       void *owner, struct page **pages,
+			       void *owner,
 			       struct hmm_range **phmm_range);
 bool amdgpu_hmm_range_get_pages_done(struct hmm_range *hmm_range);
 

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

···
 {
 	struct amdgpu_device *adev = ring->adev;
 	unsigned vmhub = ring->vm_hub;
-	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 	uint64_t fence_context = adev->fence_context + ring->idx;
 	bool needs_flush = vm->use_cpu_for_update;
 	uint64_t updates = amdgpu_vm_tlb_seq(vm);
 	int r;
 
-	*id = id_mgr->reserved;
+	*id = vm->reserved_vmid[vmhub];
 	if ((*id)->owner != vm->immediate.fence_context ||
 	    !amdgpu_vmid_compatible(*id, job) ||
 	    (*id)->flushed_updates < updates ||
···
 	return vm->reserved_vmid[vmhub];
 }
 
-int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
+/*
+ * amdgpu_vmid_alloc_reserved - reserve a specific VMID for this vm
+ * @adev: amdgpu device structure
+ * @vm: the VM to reserve an ID for
+ * @vmhub: the VMHUB which should be used
+ *
+ * Mostly used to have a reserved VMID for debugging and SPM.
+ *
+ * Returns: 0 for success, -ENOENT if an ID is already reserved.
+ */
+int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			       unsigned vmhub)
 {
 	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
+	struct amdgpu_vmid *id;
+	int r = 0;
 
 	mutex_lock(&id_mgr->lock);
-
-	++id_mgr->reserved_use_count;
-	if (!id_mgr->reserved) {
-		struct amdgpu_vmid *id;
-
-		id = list_first_entry(&id_mgr->ids_lru, struct amdgpu_vmid,
-				      list);
-		/* Remove from normal round robin handling */
-		list_del_init(&id->list);
-		id_mgr->reserved = id;
+	if (vm->reserved_vmid[vmhub])
+		goto unlock;
+	if (id_mgr->reserved_vmid) {
+		r = -ENOENT;
+		goto unlock;
 	}
-
+	/* Remove from normal round robin handling */
+	id = list_first_entry(&id_mgr->ids_lru, struct amdgpu_vmid, list);
+	list_del_init(&id->list);
+	vm->reserved_vmid[vmhub] = id;
+	id_mgr->reserved_vmid = true;
 	mutex_unlock(&id_mgr->lock);
+
 	return 0;
+unlock:
+	mutex_unlock(&id_mgr->lock);
+	return r;
 }
 
-void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,
+/*
+ * amdgpu_vmid_free_reserved - free up a reserved VMID again
+ * @adev: amdgpu device structure
+ * @vm: the VM with the reserved ID
+ * @vmhub: the VMHUB which should be used
+ */
+void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			       unsigned vmhub)
 {
 	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 
 	mutex_lock(&id_mgr->lock);
-	if (!--id_mgr->reserved_use_count) {
-		/* give the reserved ID back to normal round robin */
-		list_add(&id_mgr->reserved->list, &id_mgr->ids_lru);
-		id_mgr->reserved = NULL;
+	if (vm->reserved_vmid[vmhub]) {
+		list_add(&vm->reserved_vmid[vmhub]->list,
+			&id_mgr->ids_lru);
+		vm->reserved_vmid[vmhub] = NULL;
+		id_mgr->reserved_vmid = false;
 	}
-
 	mutex_unlock(&id_mgr->lock);
 }
 
···
 
 		mutex_init(&id_mgr->lock);
 		INIT_LIST_HEAD(&id_mgr->ids_lru);
-		id_mgr->reserved_use_count = 0;
 
 		/* for GC <10, SDMA uses MMHUB so use first_kfd_vmid for both GC and MM */
 		if (amdgpu_ip_version(adev, GC_HWIP, 0) < IP_VERSION(10, 0, 0))
···
 			amdgpu_sync_create(&id_mgr->ids[j].active);
 			list_add_tail(&id_mgr->ids[j].list, &id_mgr->ids_lru);
 		}
-	}
-	/* alloc a default reserved vmid to enforce isolation */
-	for (i = 0; i < (adev->xcp_mgr ? adev->xcp_mgr->num_xcps : 1); i++) {
-		if (adev->enforce_isolation[i] != AMDGPU_ENFORCE_ISOLATION_DISABLE)
-			amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(i));
 	}
 }
 

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

···
 	unsigned		num_ids;
 	struct list_head	ids_lru;
 	struct amdgpu_vmid	ids[AMDGPU_NUM_VMID];
-	struct amdgpu_vmid	*reserved;
-	unsigned int		reserved_use_count;
+	bool			reserved_vmid;
 };
 
 int amdgpu_pasid_alloc(unsigned int bits);
···
 bool amdgpu_vmid_had_gpu_reset(struct amdgpu_device *adev,
 			       struct amdgpu_vmid *id);
 bool amdgpu_vmid_uses_reserved(struct amdgpu_vm *vm, unsigned int vmhub);
-int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
-				unsigned vmhub);
-void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,
-				unsigned vmhub);
+int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			       unsigned vmhub);
+void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			       unsigned vmhub);
 int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 		     struct amdgpu_job *job, struct dma_fence **fence);
 void amdgpu_vmid_reset(struct amdgpu_device *adev, unsigned vmhub,

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

···
  * Calling function must call amdgpu_ttm_tt_userptr_range_done() once and only
  * once afterwards to stop HMM tracking
  */
-int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages,
+int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo,
 				 struct hmm_range **range)
 {
 	struct ttm_tt *ttm = bo->tbo.ttm;
···
 
 	readonly = amdgpu_ttm_tt_is_readonly(ttm);
 	r = amdgpu_hmm_range_get_pages(&bo->notifier, start, ttm->num_pages,
-				       readonly, NULL, pages, range);
+				       readonly, NULL, range);
 out_unlock:
 	mmap_read_unlock(mm);
 	if (r)
···
  * that backs user memory and will ultimately be mapped into the device
  * address space.
  */
-void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages)
+void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct hmm_range *range)
 {
 	unsigned long i;
 
 	for (i = 0; i < ttm->num_pages; ++i)
-		ttm->pages[i] = pages ? pages[i] : NULL;
+		ttm->pages[i] = range ? hmm_pfn_to_page(range->hmm_pfns[i]) : NULL;
 }
 
 /*

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

···
 uint64_t amdgpu_ttm_domain_start(struct amdgpu_device *adev, uint32_t type);
 
 #if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR)
-int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages,
+int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo,
 				 struct hmm_range **range);
 void amdgpu_ttm_tt_discard_user_pages(struct ttm_tt *ttm,
 				      struct hmm_range *range);
···
 				       struct hmm_range *range);
 #else
 static inline int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo,
-					       struct page **pages,
 					       struct hmm_range **range)
 {
 	return -EPERM;
···
 }
 #endif
 
-void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages);
+void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct hmm_range *range);
 int amdgpu_ttm_tt_get_userptr(const struct ttm_buffer_object *tbo,
 			      uint64_t *user_addr);
 int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo,

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

···
 		return 0;
 	}
 
+	r = -EINVAL;
 out_err:
 	amdgpu_bo_unreserve(vm->root.bo);
 	return r;
···
 	if (amdgpu_userq_input_va_validate(&fpriv->vm, args->in.queue_va, args->in.queue_size) ||
 	    amdgpu_userq_input_va_validate(&fpriv->vm, args->in.rptr_va, AMDGPU_GPU_PAGE_SIZE) ||
 	    amdgpu_userq_input_va_validate(&fpriv->vm, args->in.wptr_va, AMDGPU_GPU_PAGE_SIZE)) {
+		r = -EINVAL;
 		kfree(queue);
 		goto unlock;
 	}

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

···
 
 	/* Check if hardware has already processed the job */
 	spin_lock_irqsave(&fence_drv->fence_list_lock, flags);
-	if (!dma_fence_is_signaled_locked(fence))
+	if (!dma_fence_is_signaled(fence))
 		list_add_tail(&userq_fence->link, &fence_drv->fences);
 	else
 		dma_fence_put(fence);

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

···
 	struct amdgpu_fw_shared_rb_setup rb_setup;
 	struct amdgpu_fw_shared_smu_interface_info smu_dpm_interface;
 	struct amdgpu_fw_shared_drm_key_wa drm_key_wa;
-	uint8_t pad3[9];
+	uint8_t pad3[404];
 };
 
 #define VCN_BLOCK_ENCODE_DISABLE_MASK 0x80

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

···
 	dma_fence_put(vm->last_update);
 
 	for (i = 0; i < AMDGPU_MAX_VMHUBS; i++) {
-		if (vm->reserved_vmid[i]) {
-			amdgpu_vmid_free_reserved(adev, i);
-			vm->reserved_vmid[i] = false;
-		}
+		amdgpu_vmid_free_reserved(adev, vm, i);
 	}
 
 	ttm_lru_bulk_move_fini(&adev->mman.bdev, &vm->lru_bulk_move);
···
 	union drm_amdgpu_vm *args = data;
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
+	struct amdgpu_vm *vm = &fpriv->vm;
 
 	/* No valid flags defined yet */
 	if (args->in.flags)
···
 	switch (args->in.op) {
 	case AMDGPU_VM_OP_RESERVE_VMID:
 		/* We only have requirement to reserve vmid from gfxhub */
-		if (!fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)]) {
-			amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(0));
-			fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)] = true;
-		}
-
+		amdgpu_vmid_alloc_reserved(adev, vm, AMDGPU_GFXHUB(0));
 		break;
 	case AMDGPU_VM_OP_UNRESERVE_VMID:
-		if (fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)]) {
-			amdgpu_vmid_free_reserved(adev, AMDGPU_GFXHUB(0));
-			fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)] = false;
-		}
+		amdgpu_vmid_free_reserved(adev, vm, AMDGPU_GFXHUB(0));
 		break;
 	default:
 		return -EINVAL;

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

···
 	struct dma_fence	*last_unlocked;
 
 	unsigned int		pasid;
-	bool			reserved_vmid[AMDGPU_MAX_VMHUBS];
+	struct amdgpu_vmid	*reserved_vmid[AMDGPU_MAX_VMHUBS];
 
 	/* Flag to indicate if VM tables are updated by CPU or GPU (SDMA) */
 	bool					use_cpu_for_update;

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

···
 {
 	unsigned char *atom_rom_hdr;
 	unsigned char *str;
-	uint16_t base;
+	uint16_t base, len;
 
 	base = CU16(ATOM_ROM_TABLE_PTR);
 	atom_rom_hdr = CSTR(base);
···
 	while (str < atom_rom_hdr && *str++)
 		;
 
-	if ((str + STRLEN_NORMAL) < atom_rom_hdr)
-		strscpy(ctx->build_num, str, STRLEN_NORMAL);
+	len = min(atom_rom_hdr - str, STRLEN_NORMAL);
+	if (len)
+		strscpy(ctx->build_num, str, len);
 }
 
 struct atom_context *amdgpu_atom_parse(struct card_info *card, void *bios)

drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c

···
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[ring->xcc_id];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
+	int reset_mode = AMDGPU_RESET_TYPE_PER_QUEUE;
 	unsigned long flags;
 	int r;
 
···
 		if (!(adev->gfx.compute_supported_reset & AMDGPU_RESET_TYPE_PER_PIPE))
 			return -EOPNOTSUPP;
 		r = gfx_v9_4_3_reset_hw_pipe(ring);
+		reset_mode = AMDGPU_RESET_TYPE_PER_PIPE;
 		dev_info(adev->dev, "ring: %s pipe reset :%s\n", ring->name,
 				r ? "failed" : "successfully");
 		if (r)
···
 	r = amdgpu_ring_test_ring(kiq_ring);
 	spin_unlock_irqrestore(&kiq->ring_lock, flags);
 	if (r) {
+		if (reset_mode == AMDGPU_RESET_TYPE_PER_QUEUE)
+			goto pipe_reset;
+
 		dev_err(adev->dev, "fail to remap queue\n");
 		return r;
 	}
+
+	if (reset_mode == AMDGPU_RESET_TYPE_PER_QUEUE) {
+		r = amdgpu_ring_test_ring(ring);
+		if (r)
+			goto pipe_reset;
+	}
+
 
 	return amdgpu_ring_reset_helper_end(ring, timedout_fence);
 }

drivers/gpu/drm/amd/amdgpu/mes_v11_0.c

···
 	mes_set_hw_res_pkt.enable_reg_active_poll = 1;
 	mes_set_hw_res_pkt.enable_level_process_quantum_check = 1;
 	mes_set_hw_res_pkt.oversubscription_timer = 50;
+	if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x7f)
+		mes_set_hw_res_pkt.enable_lr_compute_wa = 1;
+	else
+		dev_info_once(mes->adev->dev,
+			      "MES FW version must be >= 0x7f to enable LR compute workaround.\n");
+
 	if (amdgpu_mes_log_enable) {
 		mes_set_hw_res_pkt.enable_mes_event_int_logging = 1;
 		mes_set_hw_res_pkt.event_intr_history_gpu_mc_ptr =

drivers/gpu/drm/amd/amdgpu/mes_v12_0.c

···
 	mes_set_hw_res_pkt.use_different_vmid_compute = 1;
 	mes_set_hw_res_pkt.enable_reg_active_poll = 1;
 	mes_set_hw_res_pkt.enable_level_process_quantum_check = 1;
+	if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x82)
+		mes_set_hw_res_pkt.enable_lr_compute_wa = 1;
+	else
+		dev_info_once(adev->dev,
+			      "MES FW version must be >= 0x82 to enable LR compute workaround.\n");
 
 	/*
 	 * Keep oversubscribe timer for sdma . When we have unmapped doorbell

drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c

···
 	return 0;
 }
 
+static int vcn_v5_0_1_late_init(struct amdgpu_ip_block *ip_block)
+{
+	struct amdgpu_device *adev = ip_block->adev;
+
+	adev->vcn.supported_reset =
+		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
+
+	switch (amdgpu_ip_version(adev, MP0_HWIP, 0)) {
+	case IP_VERSION(13, 0, 12):
+		if ((adev->psp.sos.fw_version >= 0x00450025) && amdgpu_dpm_reset_vcn_is_supported(adev))
+			adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
 static void vcn_v5_0_1_fw_shared_init(struct amdgpu_device *adev, int inst_idx)
 {
 	struct amdgpu_vcn5_fw_shared *fw_shared;
···
 		vcn_v5_0_1_fw_shared_init(adev, i);
 	}
 
-	/* TODO: Add queue reset mask when FW fully supports it */
-	adev->vcn.supported_reset =
-		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
-
 	if (amdgpu_sriov_vf(adev)) {
 		r = amdgpu_virt_alloc_mm_table(adev);
 		if (r)
···
 	return 0;
 }
 
+static int vcn_v5_0_1_hw_init_inst(struct amdgpu_device *adev, int i)
+{
+	struct amdgpu_ring *ring;
+	int vcn_inst;
+
+	vcn_inst = GET_INST(VCN, i);
+	ring = &adev->vcn.inst[i].ring_enc[0];
+
+	if (ring->use_doorbell)
+		adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+			((adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+			11 * vcn_inst),
+			adev->vcn.inst[i].aid_id);
+
+	return 0;
+}
+
 /**
  * vcn_v5_0_1_hw_init - start and test VCN block
  *
···
 {
 	struct amdgpu_device *adev = ip_block->adev;
 	struct amdgpu_ring *ring;
-	int i, r, vcn_inst;
+	int i, r;
 
 	if (amdgpu_sriov_vf(adev)) {
 		r = vcn_v5_0_1_start_sriov(adev);
···
 		if (RREG32_SOC15(VCN, GET_INST(VCN, 0), regVCN_RRMT_CNTL) & 0x100)
 			adev->vcn.caps |= AMDGPU_VCN_CAPS(RRMT_ENABLED);
 		for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-			vcn_inst = GET_INST(VCN, i);
 			ring = &adev->vcn.inst[i].ring_enc[0];
-
-			if (ring->use_doorbell)
-				adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-					((adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
-					11 * vcn_inst),
-					adev->vcn.inst[i].aid_id);
+			vcn_v5_0_1_hw_init_inst(adev, i);
 
 			/* Re-init fw_shared, if required */
 			vcn_v5_0_1_fw_shared_init(adev, i);
···
 	}
 }
 
+static int vcn_v5_0_1_ring_reset(struct amdgpu_ring *ring,
+				 unsigned int vmid,
+				 struct amdgpu_fence *timedout_fence)
+{
+	int r = 0;
+	int vcn_inst;
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me];
+
+	amdgpu_ring_reset_helper_begin(ring, timedout_fence);
+
+	vcn_inst = GET_INST(VCN, ring->me);
+	r = amdgpu_dpm_reset_vcn(adev, 1 << vcn_inst);
+
+	if (r) {
+		DRM_DEV_ERROR(adev->dev, "VCN reset fail : %d\n", r);
+		return r;
+	}
+
+	vcn_v5_0_1_hw_init_inst(adev, ring->me);
+	vcn_v5_0_1_start_dpg_mode(vinst, vinst->indirect_sram);
+
+	return amdgpu_ring_reset_helper_end(ring, timedout_fence);
+}
+
 static const struct amdgpu_ring_funcs vcn_v5_0_1_unified_ring_vm_funcs = {
 	.type = AMDGPU_RING_TYPE_VCN_ENC,
 	.align_mask = 0x3f,
···
 	.emit_wreg = vcn_v4_0_3_enc_ring_emit_wreg,
 	.emit_reg_wait = vcn_v4_0_3_enc_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+	.reset = vcn_v5_0_1_ring_reset,
 };
 
 /**
···
 static const struct amd_ip_funcs vcn_v5_0_1_ip_funcs = {
 	.name = "vcn_v5_0_1",
 	.early_init = vcn_v5_0_1_early_init,
-	.late_init = NULL,
+	.late_init = vcn_v5_0_1_late_init,
 	.sw_init = vcn_v5_0_1_sw_init,
 	.sw_fini = vcn_v5_0_1_sw_fini,
 	.hw_init = vcn_v5_0_1_hw_init,

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

···
 	mutex_init(&kfd->doorbell_mutex);
 
 	ida_init(&kfd->doorbell_ida);
+	atomic_set(&kfd->kfd_processes_count, 0);
 
 	return kfd;
 }
···
 	}
 
 	for (i = 0; i < kfd->num_nodes; i++) {
-		node = kfd->nodes[i];
+		/* Race if another thread in b/w
+		 * kfd_cleanup_nodes and kfree(kfd),
+		 * when kfd->nodes[i] = NULL
+		 */
+		if (kfd->nodes[i])
+			node = kfd->nodes[i];
+		else
+			return;
+
 		spin_lock_irqsave(&node->interrupt_lock, flags);
 
 		if (node->interrupts_active
···
 	int r = 0, temp, idx;
 
 	mutex_lock(&kfd_processes_mutex);
+
+	/* kfd_processes_count is per kfd_dev, return -EBUSY without
+	 * further check
+	 */
+	if (!!atomic_read(&kfd->kfd_processes_count)) {
+		pr_debug("process_wq_release not finished\n");
+		r = -EBUSY;
+		goto out;
+	}
 
 	if (hash_empty(kfd_processes_table) && !kfd_is_locked(kfd))
 		goto out;

drivers/gpu/drm/amd/amdkfd/kfd_priv.h

···
 
 	/* for dynamic partitioning */
 	int kfd_dev_lock;
+
+	atomic_t kfd_processes_count;
 };
 
 enum kfd_mempool {

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

···
 			pdd->runtime_inuse = false;
 		}
 
+		atomic_dec(&pdd->dev->kfd->kfd_processes_count);
+
 		kfree(pdd);
 		p->pdds[i] = NULL;
 	}
···
 
 	/* Init idr used for memory handle translation */
 	idr_init(&pdd->alloc_idr);
+
+	atomic_inc(&dev->kfd->kfd_processes_count);
 
 	return pdd;
 }

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

···
 
 			WRITE_ONCE(p->svms.faulting_task, current);
 			r = amdgpu_hmm_range_get_pages(&prange->notifier, addr, npages,
-						       readonly, owner, NULL,
+						       readonly, owner,
 						       &hmm_range);
 			WRITE_ONCE(p->svms.faulting_task, NULL);
 			if (r)

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

···
 
 static int amdgpu_dm_connector_get_modes(struct drm_connector *connector);
 
+static int amdgpu_dm_atomic_setup_commit(struct drm_atomic_state *state);
 static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state);
 
 static int amdgpu_dm_atomic_check(struct drm_device *dev,
···
 	/*
 	 * Previous frame finished and HW is ready for optimization.
 	 */
-	if (update_type == UPDATE_TYPE_FAST)
-		dc_post_update_surfaces_to_stream(dc);
+	dc_post_update_surfaces_to_stream(dc);
 
 	return dc_update_planes_and_stream(dc,
 					   array_of_surface_update,
···
 
 static struct drm_mode_config_helper_funcs amdgpu_dm_mode_config_helperfuncs = {
 	.atomic_commit_tail = amdgpu_dm_atomic_commit_tail,
-	.atomic_commit_setup = drm_dp_mst_atomic_setup_commit,
+	.atomic_commit_setup = amdgpu_dm_atomic_setup_commit,
 };
 
 static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
···
 
 	if (!caps->data_points)
 		return;
+
+	/*
+	 * Handle the case where brightness is below the first data point
+	 * Interpolate between (0,0) and (first_signal, first_lum)
+	 */
+	if (brightness < caps->luminance_data[0].input_signal) {
+		lum = DIV_ROUND_CLOSEST(caps->luminance_data[0].luminance * brightness,
+					caps->luminance_data[0].input_signal);
+		goto scale;
+	}
 
 	left = 0;
 	right = caps->data_points - 1;
···
 		{"1920x1200", 1920, 1200}
 	};
 
+	if ((connector->connector_type != DRM_MODE_CONNECTOR_eDP) &&
+	    (connector->connector_type != DRM_MODE_CONNECTOR_LVDS))
+		return;
+
 	n = ARRAY_SIZE(common_modes);
 
 	for (i = 0; i < n; i++) {
···
 	}
 }
 
+static int amdgpu_dm_atomic_setup_commit(struct drm_atomic_state *state)
+{
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
+	struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state;
+	int i, ret;
+
+	ret = drm_dp_mst_atomic_setup_commit(state);
+	if (ret)
+		return ret;
+
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
+		dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
+		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
+		/*
+		 * Color management settings. We also update color properties
+		 * when a modeset is needed, to ensure it gets reprogrammed.
+		 */
+		if (dm_new_crtc_state->base.active && dm_new_crtc_state->stream &&
+		    (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) {
+				drm_dbg_atomic(state->dev, "Failed to update color state\n");
+				return ret;
+			}
+		}
+	}
+
+	return 0;
+}
+
 /**
  * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation.
  * @state: The atomic state to commit
···
 	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);
+		ret = amdgpu_dm_check_crtc_color_mgmt(dm_new_crtc_state, true);
 		if (ret)
 			goto fail;
 	}

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

···
 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_check_crtc_color_mgmt(struct dm_crtc_state *crtc,
+				    bool check_only);
 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);

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

···
 	return res ? 0 : -ENOMEM;
 }
 
-static int amdgpu_dm_set_atomic_regamma(struct dc_stream_state *stream,
+static int amdgpu_dm_set_atomic_regamma(struct dc_transfer_func *out_tf,
 					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) {
···
 	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;
+	bool has_3dlut = adev->dm.dc->caps.color.dpp.hw_3d_lut || adev->dm.dc->caps.color.mpc.preblend;
 
 	/* shaper LUT is only available if 3D LUT color caps */
 	exp_size = has_3dlut ? MAX_COLOR_LUT_ENTRIES : 0;
···
 }
 
 /**
- * amdgpu_dm_update_crtc_color_mgmt: Maps DRM color management to DC stream.
+ * amdgpu_dm_check_crtc_color_mgmt: Check if DRM color props are programmable by DC.
  * @crtc: amdgpu_dm crtc state
+ * @check_only: only check color state without update dc stream
  *
- * With no plane level color management properties we're free to use any
- * of the HW blocks as long as the CRTC CTM always comes before the
- * CRTC RGM and after the CRTC DGM.
- *
- * - The CRTC RGM block will be placed in the RGM LUT block if it is non-linear.
- * - The CRTC DGM block will be placed in the DGM LUT block if it is non-linear.
- * - The CRTC CTM will be placed in the gamut remap block if it is non-linear.
+ * This function just verifies CRTC LUT sizes, if there is enough space for
+ * output transfer function and if its parameters can be calculated by AMD
+ * color module. It also adjusts some settings for programming CRTC degamma at
+ * plane stage, using plane DGM block.
  *
  * The RGM block is typically more fully featured and accurate across
  * all ASICs - DCE can't support a custom non-linear CRTC DGM.
  *
  * For supporting both plane level color management and CRTC level color
- * management at once we have to either restrict the usage of CRTC properties
- * or blend adjustments together.
+ * management at once we have to either restrict the usage of some CRTC
+ * properties or blend adjustments together.
  *
  * Returns:
- * 0 on success. Error code if setup fails.
+ * 0 on success. Error code if validation fails.
  */
-int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
+
+int amdgpu_dm_check_crtc_color_mgmt(struct dm_crtc_state *crtc,
+				    bool check_only)
 {
 	struct dc_stream_state *stream = crtc->stream;
 	struct amdgpu_device *adev = drm_to_adev(crtc->base.state->dev);
 	bool has_rom = adev->asic_type <= CHIP_RAVEN;
-	struct drm_color_ctm *ctm = NULL;
+	struct dc_transfer_func *out_tf;
 	const struct drm_color_lut *degamma_lut, *regamma_lut;
 	uint32_t degamma_size, regamma_size;
 	bool has_regamma, has_degamma;
···
 	crtc->cm_has_degamma = false;
 	crtc->cm_is_degamma_srgb = false;
 
+	if (check_only) {
+		out_tf = kvzalloc(sizeof(*out_tf), GFP_KERNEL);
+		if (!out_tf)
+			return -ENOMEM;
+	} else {
+		out_tf = &stream->out_transfer_func;
+	}
+
 	/* Setup regamma and degamma. */
 	if (is_legacy) {
 		/*
···
 		 * inverse color ramp in legacy userspace.
 		 */
 		crtc->cm_is_degamma_srgb = true;
-		stream->out_transfer_func.type = TF_TYPE_DISTRIBUTED_POINTS;
-		stream->out_transfer_func.tf = TRANSFER_FUNCTION_SRGB;
+		out_tf->type = TF_TYPE_DISTRIBUTED_POINTS;
+		out_tf->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
···
 		 *
 		 * See more in mod_color_calculate_regamma_params()
 		 */
-		r = __set_legacy_tf(&stream->out_transfer_func, regamma_lut,
+		r = __set_legacy_tf(out_tf, regamma_lut,
 				    regamma_size, has_rom);
-		if (r)
-			return r;
 	} else {
 		regamma_size = has_regamma ? regamma_size : 0;
-		r = amdgpu_dm_set_atomic_regamma(stream, regamma_lut,
+		r = amdgpu_dm_set_atomic_regamma(out_tf, regamma_lut,
 						 regamma_size, has_rom, tf);
-		if (r)
-			return r;
 	}
 
 	/*
···
 	 * have to place the CTM in the OCSC in that case.
 	 */
 	crtc->cm_has_degamma = has_degamma;
+	if (check_only)
+		kvfree(out_tf);
+
+	return r;
+}
+
+/**
+ * amdgpu_dm_update_crtc_color_mgmt: Maps DRM color management to DC stream.
+ * @crtc: amdgpu_dm crtc state
+ *
+ * With no plane level color management properties we're free to use any
+ * of the HW blocks as long as the CRTC CTM always comes before the
+ * CRTC RGM and after the CRTC DGM.
+ *
+ * - The CRTC RGM block will be placed in the RGM LUT block if it is non-linear.
+ * - The CRTC DGM block will be placed in the DGM LUT block if it is non-linear.
+ * - The CRTC CTM will be placed in the gamut remap block if it is non-linear.
+ *
+ * The RGM block is typically more fully featured and accurate across
+ * all ASICs - DCE can't support a custom non-linear CRTC DGM.
+ *
+ * For supporting both plane level color management and CRTC level color
+ * management at once we have to either restrict the usage of CRTC properties
+ * or blend adjustments together.
+ *
+ * Returns:
+ * 0 on success. Error code if setup fails.
+ */
+int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
+{
+	struct dc_stream_state *stream = crtc->stream;
+	struct drm_color_ctm *ctm = NULL;
+	int ret;
+
+	ret = amdgpu_dm_check_crtc_color_mgmt(crtc, false);
+	if (ret)
+		return ret;
 
 	/* Setup CRTC CTM. */
 	if (crtc->base.ctm) {

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

···
 			break;
 		}
 
-		if (idle_work->enable)
+		if (idle_work->enable) {
+			dc_post_update_surfaces_to_stream(idle_work->dm->dc);
 			dc_allow_idle_optimizations(idle_work->dm->dc, true);
+		}
 		mutex_unlock(&idle_work->dm->dc_lock);
 	}
 	idle_work->dm->idle_workqueue->running = false;
···
 			vblank_work->acrtc->dm_irq_params.allow_sr_entry);
 	}
 
-	if (dm->active_vblank_irq_count == 0)
+	if (dm->active_vblank_irq_count == 0) {
+		dc_post_update_surfaces_to_stream(dm->dc);
 		dc_allow_idle_optimizations(dm->dc, true);
+	}
 
 	mutex_unlock(&dm->dc_lock);
 

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

···
 		struct mod_hdcp_ddc_funcs *ddc_funcs = &config->ddc.funcs;
 
 		config->psp.handle = &adev->psp;
-		if (dc->ctx->dce_version == DCN_VERSION_3_1 ||
+		if (dc->ctx->dce_version == DCN_VERSION_3_1  ||
 		    dc->ctx->dce_version == DCN_VERSION_3_14 ||
 		    dc->ctx->dce_version == DCN_VERSION_3_15 ||
-		    dc->ctx->dce_version == DCN_VERSION_3_5 ||
+		    dc->ctx->dce_version == DCN_VERSION_3_16 ||
+		    dc->ctx->dce_version == DCN_VERSION_3_2  ||
+		    dc->ctx->dce_version == DCN_VERSION_3_21 ||
+		    dc->ctx->dce_version == DCN_VERSION_3_5  ||
 		    dc->ctx->dce_version == DCN_VERSION_3_51 ||
-		    dc->ctx->dce_version == DCN_VERSION_3_6 ||
-		    dc->ctx->dce_version == DCN_VERSION_3_16)
+		    dc->ctx->dce_version == DCN_VERSION_3_6  ||
+		    dc->ctx->dce_version == DCN_VERSION_4_01)
 			config->psp.caps.dtm_v3_supported = 1;
+
 		config->ddc.handle = dc_get_link_at_index(dc, i);
 
 		ddc_funcs->write_i2c = lp_write_i2c;

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

···
 		drm_object_attach_property(&plane->base,
 					   dm->adev->mode_info.plane_ctm_property, 0);
 
-	if (dpp_color_caps.hw_3d_lut) {
+	if (dpp_color_caps.hw_3d_lut || dm->dc->caps.color.mpc.preblend) {
 		drm_object_attach_property(&plane->base,
 					   mode_info.plane_shaper_lut_property, 0);
 		drm_object_attach_property(&plane->base,

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

···
 				    func_name, line);
 }
 
-void dm_trace_smu_msg(uint32_t msg_id, uint32_t param_in, struct dc_context *ctx)
+void dm_trace_smu_enter(uint32_t msg_id, uint32_t param_in, unsigned int delay, struct dc_context *ctx)
 {
 }
 
-void dm_trace_smu_delay(uint32_t delay, struct dc_context *ctx)
+void dm_trace_smu_exit(bool success, uint32_t response, struct dc_context *ctx)
 {
 }
 

drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c

···
 		clk_mgr->max_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
 	clk_mgr->cur_min_clks_state = DM_PP_CLOCKS_STATE_INVALID;
 
+	base->clks.max_supported_dispclk_khz =
+		clk_mgr->max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_khz;
+
 	dce_clock_read_integrated_info(clk_mgr);
 	dce_clock_read_ss_info(clk_mgr);
 }

drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c

···
 		struct dc_context *ctx,
 		struct clk_mgr_internal *clk_mgr)
 {
+	struct clk_mgr *base = &clk_mgr->base;
+
 	dce_clk_mgr_construct(ctx, clk_mgr);
 
 	memcpy(clk_mgr->max_clks_by_state,
···
 	clk_mgr->clk_mgr_shift = &disp_clk_shift;
 	clk_mgr->clk_mgr_mask = &disp_clk_mask;
 	clk_mgr->base.funcs = &dce60_funcs;
+
+	base->clks.max_supported_dispclk_khz =
+		clk_mgr->max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_khz;
 }
 

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c

···
 
 	/* handle DALSMC_Result_CmdRejectedBusy? */
 
-	TRACE_SMU_DELAY(delay_us * (initial_max_retries - max_retries), clk_mgr->base.ctx);
+	TRACE_SMU_MSG_DELAY(0, 0, delay_us * (initial_max_retries - max_retries), clk_mgr->base.ctx);
 
 	return reg;
 }

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c

···
 #undef DC_LOGGER
 #define DC_LOGGER \
 	clk_mgr->base.base.ctx->logger
+
 #define regCLK1_CLK_PLL_REQ			0x0237
 #define regCLK1_CLK_PLL_REQ_BASE_IDX		0
 
···
 #define CLK1_CLK_PLL_REQ__PllSpineDiv_MASK	0x0000F000L
 #define CLK1_CLK_PLL_REQ__FbMult_frac_MASK	0xFFFF0000L
 
+#define regCLK1_CLK0_DFS_CNTL				0x0269
+#define regCLK1_CLK0_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK1_DFS_CNTL				0x026c
+#define regCLK1_CLK1_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK2_DFS_CNTL				0x026f
+#define regCLK1_CLK2_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK3_DFS_CNTL				0x0272
+#define regCLK1_CLK3_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK4_DFS_CNTL				0x0275
+#define regCLK1_CLK4_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK5_DFS_CNTL				0x0278
+#define regCLK1_CLK5_DFS_CNTL_BASE_IDX		0
+
+#define regCLK1_CLK0_CURRENT_CNT			0x02fb
+#define regCLK1_CLK0_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK1_CURRENT_CNT			0x02fc
+#define regCLK1_CLK1_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK2_CURRENT_CNT			0x02fd
+#define regCLK1_CLK2_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK3_CURRENT_CNT			0x02fe
+#define regCLK1_CLK3_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK4_CURRENT_CNT			0x02ff
+#define regCLK1_CLK4_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK5_CURRENT_CNT			0x0300
+#define regCLK1_CLK5_CURRENT_CNT_BASE_IDX	0
+
+#define regCLK1_CLK0_BYPASS_CNTL			0x028a
+#define regCLK1_CLK0_BYPASS_CNTL_BASE_IDX	0
+#define regCLK1_CLK1_BYPASS_CNTL			0x0293
+#define regCLK1_CLK1_BYPASS_CNTL_BASE_IDX	0
 #define regCLK1_CLK2_BYPASS_CNTL			0x029c
 #define regCLK1_CLK2_BYPASS_CNTL_BASE_IDX	0
+#define regCLK1_CLK3_BYPASS_CNTL			0x02a5
+#define regCLK1_CLK3_BYPASS_CNTL_BASE_IDX	0
+#define regCLK1_CLK4_BYPASS_CNTL			0x02ae
+#define regCLK1_CLK4_BYPASS_CNTL_BASE_IDX	0
+#define regCLK1_CLK5_BYPASS_CNTL			0x02b7
+#define regCLK1_CLK5_BYPASS_CNTL_BASE_IDX	0
+
+#define regCLK1_CLK0_DS_CNTL				0x0283
+#define regCLK1_CLK0_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK1_DS_CNTL				0x028c
+#define regCLK1_CLK1_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK2_DS_CNTL				0x0295
+#define regCLK1_CLK2_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK3_DS_CNTL				0x029e
+#define regCLK1_CLK3_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK4_DS_CNTL				0x02a7
+#define regCLK1_CLK4_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK5_DS_CNTL				0x02b0
+#define regCLK1_CLK5_DS_CNTL_BASE_IDX		0
+
+#define regCLK1_CLK0_ALLOW_DS				0x0284
+#define regCLK1_CLK0_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK1_ALLOW_DS				0x028d
+#define regCLK1_CLK1_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK2_ALLOW_DS				0x0296
+#define regCLK1_CLK2_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK3_ALLOW_DS				0x029f
+#define regCLK1_CLK3_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK4_ALLOW_DS				0x02a8
+#define regCLK1_CLK4_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK5_ALLOW_DS				0x02b1
+#define regCLK1_CLK5_ALLOW_DS_BASE_IDX		0
 
 #define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_SEL__SHIFT	0x0
 #define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_DIV__SHIFT	0x10
···
 {
 	struct clk_mgr_internal *clk_mgr_int = TO_CLK_MGR_INTERNAL(clk_mgr);
 	uint32_t ref_dtbclk = clk_mgr->clks.ref_dtbclk_khz;
+	struct clk_mgr_dcn314 *clk_mgr_dcn314 = TO_CLK_MGR_DCN314(clk_mgr_int);
+	struct clk_log_info log_info = {0};
 
 	memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks));
 	// Assumption is that boot state always supports pstate
···
 			dce_adjust_dp_ref_freq_for_ss(clk_mgr_int, clk_mgr->dprefclk_khz);
 	else
 		clk_mgr->dp_dto_source_clock_in_khz = clk_mgr->dprefclk_khz;
+
+	dcn314_dump_clk_registers(&clk_mgr->boot_snapshot, &clk_mgr_dcn314->base.base, &log_info);
+	clk_mgr->clks.dispclk_khz =  clk_mgr->boot_snapshot.dispclk * 1000;
 }
 
 void dcn314_update_clocks(struct clk_mgr *clk_mgr_base,
···
 
 	if (dc->work_arounds.skip_clock_update)
 		return;
+
+	display_count = dcn314_get_active_display_cnt_wa(dc, context);
 
 	/*
 	 * if it is safe to lower, but we are already in the lower state, we don't have to do anything
···
 		}
 		/* check that we're not already in lower */
 		if (clk_mgr_base->clks.pwr_state != DCN_PWR_STATE_LOW_POWER) {
-			display_count = dcn314_get_active_display_cnt_wa(dc, context);
 			/* if we can go lower, go lower */
 			if (display_count == 0) {
 				union display_idle_optimization_u idle_info = { 0 };
···
 		update_dppclk = true;
 	}
 
-	if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)) {
+	if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz) &&
+	    (new_clocks->dispclk_khz > 0 || (safe_to_lower && display_count == 0))) {
+		int requested_dispclk_khz = new_clocks->dispclk_khz;
+
 		dcn314_disable_otg_wa(clk_mgr_base, context, safe_to_lower, true);
 
+		/* Clamp the requested clock to PMFW based on their limit. */
+		if (dc->debug.min_disp_clk_khz > 0 && requested_dispclk_khz < dc->debug.min_disp_clk_khz)
+			requested_dispclk_khz = dc->debug.min_disp_clk_khz;
+
+		dcn314_smu_set_dispclk(clk_mgr, requested_dispclk_khz);
 		clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz;
-		dcn314_smu_set_dispclk(clk_mgr, clk_mgr_base->clks.dispclk_khz);
+
 		dcn314_disable_otg_wa(clk_mgr_base, context, safe_to_lower, false);
 
 		update_dispclk = true;
···
 	return true;
 }
 
-static void dcn314_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
+
+static void dcn314_dump_clk_registers_internal(struct dcn35_clk_internal *internal, struct clk_mgr *clk_mgr_base)
+{
+	struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+
+	// read dtbclk
+	internal->CLK1_CLK4_CURRENT_CNT = REG_READ(CLK1_CLK4_CURRENT_CNT);
+	internal->CLK1_CLK4_BYPASS_CNTL = REG_READ(CLK1_CLK4_BYPASS_CNTL);
+
+	// read dcfclk
+	internal->CLK1_CLK3_CURRENT_CNT = REG_READ(CLK1_CLK3_CURRENT_CNT);
+	internal->CLK1_CLK3_BYPASS_CNTL = REG_READ(CLK1_CLK3_BYPASS_CNTL);
+
+	// read dcf deep sleep divider
+	internal->CLK1_CLK3_DS_CNTL = REG_READ(CLK1_CLK3_DS_CNTL);
+	internal->CLK1_CLK3_ALLOW_DS = REG_READ(CLK1_CLK3_ALLOW_DS);
+
+	// read dppclk
+	internal->CLK1_CLK1_CURRENT_CNT = REG_READ(CLK1_CLK1_CURRENT_CNT);
+	internal->CLK1_CLK1_BYPASS_CNTL = REG_READ(CLK1_CLK1_BYPASS_CNTL);
+
+	// read dprefclk
+	internal->CLK1_CLK2_CURRENT_CNT = REG_READ(CLK1_CLK2_CURRENT_CNT);
+	internal->CLK1_CLK2_BYPASS_CNTL = REG_READ(CLK1_CLK2_BYPASS_CNTL);
+
+	// read dispclk
+	internal->CLK1_CLK0_CURRENT_CNT = REG_READ(CLK1_CLK0_CURRENT_CNT);
+	internal->CLK1_CLK0_BYPASS_CNTL = REG_READ(CLK1_CLK0_BYPASS_CNTL);
+}
+
+void dcn314_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
 		struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info)
 {
-	return;
+
+	struct dcn35_clk_internal internal = {0};
+
+	dcn314_dump_clk_registers_internal(&internal, clk_mgr_base);
+
+	regs_and_bypass->dcfclk = internal.CLK1_CLK3_CURRENT_CNT / 10;
+	regs_and_bypass->dcf_deep_sleep_divider = internal.CLK1_CLK3_DS_CNTL / 10;
+	regs_and_bypass->dcf_deep_sleep_allow = internal.CLK1_CLK3_ALLOW_DS;
+	regs_and_bypass->dprefclk = internal.CLK1_CLK2_CURRENT_CNT / 10;
+	regs_and_bypass->dispclk = internal.CLK1_CLK0_CURRENT_CNT / 10;
+	regs_and_bypass->dppclk = internal.CLK1_CLK1_CURRENT_CNT / 10;
+	regs_and_bypass->dtbclk = internal.CLK1_CLK4_CURRENT_CNT / 10;
+
+	regs_and_bypass->dppclk_bypass = internal.CLK1_CLK1_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dppclk_bypass < 0 || regs_and_bypass->dppclk_bypass > 4)
+		regs_and_bypass->dppclk_bypass = 0;
+	regs_and_bypass->dcfclk_bypass = internal.CLK1_CLK3_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dcfclk_bypass < 0 || regs_and_bypass->dcfclk_bypass > 4)
+		regs_and_bypass->dcfclk_bypass = 0;
+	regs_and_bypass->dispclk_bypass = internal.CLK1_CLK0_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dispclk_bypass < 0 || regs_and_bypass->dispclk_bypass > 4)
+		regs_and_bypass->dispclk_bypass = 0;
+	regs_and_bypass->dprefclk_bypass = internal.CLK1_CLK2_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dprefclk_bypass < 0 || regs_and_bypass->dprefclk_bypass > 4)
+		regs_and_bypass->dprefclk_bypass = 0;
+
 }
 
 static struct clk_bw_params dcn314_bw_params = {

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h

···
 
 void dcn314_clk_mgr_destroy(struct clk_mgr_internal *clk_mgr_int);
 
+
+void dcn314_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
+		struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info);
+
+
 #endif //__DCN314_CLK_MGR_H__

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c

···
 			udelay(delay_us);
 	} while (max_retries--);
 
-	TRACE_SMU_DELAY(delay_us * (initial_max_retries - max_retries), clk_mgr->base.ctx);
+	TRACE_SMU_MSG_DELAY(0, 0, delay_us * (initial_max_retries - max_retries), clk_mgr->base.ctx);
+
 
 	return reg;
 }
···
 		*total_delay_us += delay_us;
 	} while (max_retries--);
 
-	TRACE_SMU_DELAY(*total_delay_us, clk_mgr->base.ctx);
+	TRACE_SMU_MSG_DELAY(0, 0, *total_delay_us, clk_mgr->base.ctx);
 
 	return reg;
 }

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c

···
 	return true;
 }
 
-static void dcn35_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
+static void dcn35_save_clk_registers_internal(struct dcn35_clk_internal *internal, struct clk_mgr *clk_mgr_base)
+{
+	struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+
+	// read dtbclk
+	internal->CLK1_CLK4_CURRENT_CNT = REG_READ(CLK1_CLK4_CURRENT_CNT);
+	internal->CLK1_CLK4_BYPASS_CNTL = REG_READ(CLK1_CLK4_BYPASS_CNTL);
+
+	// read dcfclk
+	internal->CLK1_CLK3_CURRENT_CNT = REG_READ(CLK1_CLK3_CURRENT_CNT);
+	internal->CLK1_CLK3_BYPASS_CNTL = REG_READ(CLK1_CLK3_BYPASS_CNTL);
+
+	// read dcf deep sleep divider
+	internal->CLK1_CLK3_DS_CNTL = REG_READ(CLK1_CLK3_DS_CNTL);
+	internal->CLK1_CLK3_ALLOW_DS = REG_READ(CLK1_CLK3_ALLOW_DS);
+
+	// read dppclk
+	internal->CLK1_CLK1_CURRENT_CNT = REG_READ(CLK1_CLK1_CURRENT_CNT);
+	internal->CLK1_CLK1_BYPASS_CNTL = REG_READ(CLK1_CLK1_BYPASS_CNTL);
+
+	// read dprefclk
+	internal->CLK1_CLK2_CURRENT_CNT = REG_READ(CLK1_CLK2_CURRENT_CNT);
+	internal->CLK1_CLK2_BYPASS_CNTL = REG_READ(CLK1_CLK2_BYPASS_CNTL);
+
+	// read dispclk
+	internal->CLK1_CLK0_CURRENT_CNT = REG_READ(CLK1_CLK0_CURRENT_CNT);
+	internal->CLK1_CLK0_BYPASS_CNTL = REG_READ(CLK1_CLK0_BYPASS_CNTL);
+}
+
+static void dcn35_save_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
 		struct clk_mgr_dcn35 *clk_mgr)
 {
+	struct dcn35_clk_internal internal = {0};
+	char *bypass_clks[5] = {"0x0 DFS", "0x1 REFCLK", "0x2 ERROR", "0x3 400 FCH", "0x4 600 FCH"};
+
+	dcn35_save_clk_registers_internal(&internal, &clk_mgr->base.base);
+
+	regs_and_bypass->dcfclk = internal.CLK1_CLK3_CURRENT_CNT / 10;
+	regs_and_bypass->dcf_deep_sleep_divider = internal.CLK1_CLK3_DS_CNTL / 10;
+	regs_and_bypass->dcf_deep_sleep_allow = internal.CLK1_CLK3_ALLOW_DS;
+	regs_and_bypass->dprefclk = internal.CLK1_CLK2_CURRENT_CNT / 10;
+	regs_and_bypass->dispclk = internal.CLK1_CLK0_CURRENT_CNT / 10;
+	regs_and_bypass->dppclk = internal.CLK1_CLK1_CURRENT_CNT / 10;
+	regs_and_bypass->dtbclk = internal.CLK1_CLK4_CURRENT_CNT / 10;
+
+	regs_and_bypass->dppclk_bypass = internal.CLK1_CLK1_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dppclk_bypass < 0 || regs_and_bypass->dppclk_bypass > 4)
+		regs_and_bypass->dppclk_bypass = 0;
+	regs_and_bypass->dcfclk_bypass = internal.CLK1_CLK3_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dcfclk_bypass < 0 || regs_and_bypass->dcfclk_bypass > 4)
+		regs_and_bypass->dcfclk_bypass = 0;
+	regs_and_bypass->dispclk_bypass = internal.CLK1_CLK0_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dispclk_bypass < 0 || regs_and_bypass->dispclk_bypass > 4)
+		regs_and_bypass->dispclk_bypass = 0;
+	regs_and_bypass->dprefclk_bypass = internal.CLK1_CLK2_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dprefclk_bypass < 0 || regs_and_bypass->dprefclk_bypass > 4)
+		regs_and_bypass->dprefclk_bypass = 0;
+
+	if (clk_mgr->base.base.ctx->dc->debug.pstate_enabled) {
+		DC_LOG_SMU("clk_type,clk_value,deepsleep_cntl,deepsleep_allow,bypass\n");
+
+		DC_LOG_SMU("dcfclk,%d,%d,%d,%s\n",
+				   regs_and_bypass->dcfclk,
+				   regs_and_bypass->dcf_deep_sleep_divider,
+				   regs_and_bypass->dcf_deep_sleep_allow,
+				   bypass_clks[(int) regs_and_bypass->dcfclk_bypass]);
+
+		DC_LOG_SMU("dprefclk,%d,N/A,N/A,%s\n",
+			regs_and_bypass->dprefclk,
+			bypass_clks[(int) regs_and_bypass->dprefclk_bypass]);
+
+		DC_LOG_SMU("dispclk,%d,N/A,N/A,%s\n",
+			regs_and_bypass->dispclk,
+			bypass_clks[(int) regs_and_bypass->dispclk_bypass]);
+
+		// REGISTER VALUES
+		DC_LOG_SMU("reg_name,value,clk_type");
+
+		DC_LOG_SMU("CLK1_CLK3_CURRENT_CNT,%d,dcfclk",
+				internal.CLK1_CLK3_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK4_CURRENT_CNT,%d,dtbclk",
+					internal.CLK1_CLK4_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK3_DS_CNTL,%d,dcf_deep_sleep_divider",
+					internal.CLK1_CLK3_DS_CNTL);
+
+		DC_LOG_SMU("CLK1_CLK3_ALLOW_DS,%d,dcf_deep_sleep_allow",
+					internal.CLK1_CLK3_ALLOW_DS);
+
+		DC_LOG_SMU("CLK1_CLK2_CURRENT_CNT,%d,dprefclk",
+					internal.CLK1_CLK2_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK0_CURRENT_CNT,%d,dispclk",
+					internal.CLK1_CLK0_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK1_CURRENT_CNT,%d,dppclk",
+					internal.CLK1_CLK1_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK3_BYPASS_CNTL,%d,dcfclk_bypass",
+					internal.CLK1_CLK3_BYPASS_CNTL);
+
+		DC_LOG_SMU("CLK1_CLK2_BYPASS_CNTL,%d,dprefclk_bypass",
+					internal.CLK1_CLK2_BYPASS_CNTL);
+
+		DC_LOG_SMU("CLK1_CLK0_BYPASS_CNTL,%d,dispclk_bypass",
+					internal.CLK1_CLK0_BYPASS_CNTL);
+
+		DC_LOG_SMU("CLK1_CLK1_BYPASS_CNTL,%d,dppclk_bypass",
+					internal.CLK1_CLK1_BYPASS_CNTL);
+
+	}
 }
 
 static bool dcn35_is_spll_ssc_enabled(struct clk_mgr *clk_mgr_base)
···
 void dcn35_init_clocks(struct clk_mgr *clk_mgr)
 {
 	struct clk_mgr_internal *clk_mgr_int = TO_CLK_MGR_INTERNAL(clk_mgr);
+	struct clk_mgr_dcn35 *clk_mgr_dcn35 = TO_CLK_MGR_DCN35(clk_mgr_int);
 
 	init_clk_states(clk_mgr);
 
···
 	else
 		clk_mgr->dp_dto_source_clock_in_khz = clk_mgr->dprefclk_khz;
 
+	dcn35_save_clk_registers(&clk_mgr->boot_snapshot, clk_mgr_dcn35);
+
+	clk_mgr->clks.ref_dtbclk_khz =  clk_mgr->boot_snapshot.dtbclk * 10;
+	if (clk_mgr->boot_snapshot.dtbclk > 59000) {
+		/*dtbclk enabled based on */
+		clk_mgr->clks.dtbclk_en = true;
+	}
 }
 static struct clk_bw_params dcn35_bw_params = {
 	.vram_type = Ddr4MemType,
···
 		dcn35_bw_params.wm_table = ddr5_wm_table;
 	}
 	/* Saved clocks configured at boot for debug purposes */
-	dcn35_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, clk_mgr);
+	dcn35_save_clk_registers(&clk_mgr->base.base.boot_snapshot, clk_mgr);
 
 	clk_mgr->base.base.dprefclk_khz = dcn35_smu_get_dprefclk(&clk_mgr->base);
 	clk_mgr->base.base.clks.ref_dtbclk_khz = 600000;

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c

···
 	unsigned int i;
 	char *entry_i = (char *)entry_0;
 
-	uint32_t ret = dcn30_smu_get_dpm_freq_by_index(clk_mgr, clk, 0xFF);
+	uint32_t ret = dcn401_smu_get_dpm_freq_by_index(clk_mgr, clk, 0xFF);
 
 	if (ret & (1 << 31))
 		/* fine-grained, only min and max */
···
 
 	/* if the initial message failed, num_levels will be 0 */
 	for (i = 0; i < *num_levels && i < ARRAY_SIZE(clk_mgr->base.bw_params->clk_table.entries); i++) {
-		*((unsigned int *)entry_i) = (dcn30_smu_get_dpm_freq_by_index(clk_mgr, clk, i) & 0xFFFF);
+		*((unsigned int *)entry_i) = (dcn401_smu_get_dpm_freq_by_index(clk_mgr, clk, i) & 0xFFFF);
 		entry_i += sizeof(clk_mgr->base.bw_params->clk_table.entries[0]);
 	}
 }
···
 	clk_mgr->smu_present = false;
 	clk_mgr->dpm_present = false;
 
-	if (!clk_mgr_base->force_smu_not_present && dcn30_smu_get_smu_version(clk_mgr, &clk_mgr->smu_ver))
+	if (!clk_mgr_base->force_smu_not_present && dcn401_smu_get_smu_version(clk_mgr, &clk_mgr->smu_ver))
 		clk_mgr->smu_present = true;
 
 	if (!clk_mgr->smu_present)
 		return;
 
-	dcn30_smu_check_driver_if_version(clk_mgr);
-	dcn30_smu_check_msg_header_version(clk_mgr);
+	dcn401_smu_check_driver_if_version(clk_mgr);
+	dcn401_smu_check_msg_header_version(clk_mgr);
 
 	/* DCFCLK */
 	dcn401_init_single_clock(clk_mgr, PPCLK_DCFCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].dcfclk_mhz,
 			&num_entries_per_clk->num_dcfclk_levels);
-	clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DCFCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DCFCLK);
 	if (num_entries_per_clk->num_dcfclk_levels && clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_dcfclk_levels - 1].dcfclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz = 0;
···
 	dcn401_init_single_clock(clk_mgr, PPCLK_SOCCLK,
 					&clk_mgr_base->bw_params->clk_table.entries[0].socclk_mhz,
 					&num_entries_per_clk->num_socclk_levels);
-	clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_SOCCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_SOCCLK);
 	if (num_entries_per_clk->num_socclk_levels && clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_socclk_levels - 1].socclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz = 0;
···
 		dcn401_init_single_clock(clk_mgr, PPCLK_DTBCLK,
 				&clk_mgr_base->bw_params->clk_table.entries[0].dtbclk_mhz,
 				&num_entries_per_clk->num_dtbclk_levels);
-		clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DTBCLK);
+		clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DTBCLK);
 		if (num_entries_per_clk->num_dtbclk_levels && clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz ==
 				clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_dtbclk_levels - 1].dtbclk_mhz)
 			clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz = 0;
···
 	dcn401_init_single_clock(clk_mgr, PPCLK_DISPCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].dispclk_mhz,
 			&num_entries_per_clk->num_dispclk_levels);
-	clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DISPCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DISPCLK);
 	if (num_entries_per_clk->num_dispclk_levels && clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_dispclk_levels - 1].dispclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = 0;
···
 			table->Watermarks.WatermarkRow[i].WmSetting = i;
 			table->Watermarks.WatermarkRow[i].Flags = clk_mgr->base.bw_params->wm_table.nv_entries[i].pmfw_breakdown.wm_type;
 		}
-	dcn30_smu_set_dram_addr_high(clk_mgr, clk_mgr->wm_range_table_addr >> 32);
-	dcn30_smu_set_dram_addr_low(clk_mgr, clk_mgr->wm_range_table_addr & 0xFFFFFFFF);
+	dcn401_smu_set_dram_addr_high(clk_mgr, clk_mgr->wm_range_table_addr >> 32);
+	dcn401_smu_set_dram_addr_low(clk_mgr, clk_mgr->wm_range_table_addr & 0xFFFFFFFF);
 	dcn401_smu_transfer_wm_table_dram_2_smu(clk_mgr);
 }
 
···
 				clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_memclk_levels - 1].memclk_mhz;
 	}
 
-	clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_UCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_UCLK);
 	if (num_entries_per_clk->num_memclk_levels && clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_memclk_levels - 1].memclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz = 0;
···
 	dcn401_init_single_clock(clk_mgr, PPCLK_FCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].fclk_mhz,
 			&num_entries_per_clk->num_fclk_levels);
-	clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_FCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_FCLK);
 	if (num_entries_per_clk->num_fclk_levels && clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_fclk_levels - 1].fclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = 0;

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c

···
 	/* Wait for response register to be ready */
 	dcn401_smu_wait_for_response(clk_mgr, 10, 200000);
 
+	TRACE_SMU_MSG_ENTER(msg_id, param_in, clk_mgr->base.ctx);
+
 	/* Clear response register */
 	REG_WRITE(DAL_RESP_REG, 0);
 
···
 		if (param_out)
 			*param_out = REG_READ(DAL_ARG_REG);
 
+		TRACE_SMU_MSG_EXIT(true, param_out ? *param_out : 0, clk_mgr->base.ctx);
 		return true;
 	}
 
+	TRACE_SMU_MSG_EXIT(false, 0, clk_mgr->base.ctx);
 	return false;
 }
 
···
 		*total_delay_us += delay_us;
 	} while (max_retries--);
 
-	TRACE_SMU_DELAY(*total_delay_us, clk_mgr->base.ctx);
-
 	return reg;
 }
 
···
 	/* Wait for response register to be ready */
 	dcn401_smu_wait_for_response_delay(clk_mgr, 10, 200000, &delay1_us);
 
+	TRACE_SMU_MSG_ENTER(msg_id, param_in, clk_mgr->base.ctx);
+
 	/* Clear response register */
 	REG_WRITE(DAL_RESP_REG, 0);
 
···
 	/* Trigger the message transaction by writing the message ID */
 	REG_WRITE(DAL_MSG_REG, msg_id);
 
-	TRACE_SMU_MSG(msg_id, param_in, clk_mgr->base.ctx);
-
 	/* Wait for response */
 	if (dcn401_smu_wait_for_response_delay(clk_mgr, 10, 200000, &delay2_us) == DALSMC_Result_OK) {
 		if (param_out)
 			*param_out = REG_READ(DAL_ARG_REG);
 
 		*total_delay_us = delay1_us + delay2_us;
+		TRACE_SMU_MSG_EXIT(true, param_out ? *param_out : 0, clk_mgr->base.ctx);
 		return true;
 	}
 
 	*total_delay_us = delay1_us + 2000000;
+	TRACE_SMU_MSG_EXIT(false, 0, clk_mgr->base.ctx);
+	return false;
+}
+
+bool dcn401_smu_get_smu_version(struct clk_mgr_internal *clk_mgr, unsigned int *version)
+{
+	smu_print("SMU Get SMU version\n");
+
+	if (dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetSmuVersion, 0, version)) {
+
+		smu_print("SMU version: %d\n", *version);
+
+		return true;
+	}
+
+	return false;
+}
+
+/* Message output should match SMU11_DRIVER_IF_VERSION in smu11_driver_if.h */
+bool dcn401_smu_check_driver_if_version(struct clk_mgr_internal *clk_mgr)
+{
+	uint32_t response = 0;
+
+	smu_print("SMU Check driver if version\n");
+
+	if (dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetDriverIfVersion, 0, &response)) {
+
+		smu_print("SMU driver if version: %d\n", response);
+
+		if (response == SMU14_DRIVER_IF_VERSION)
+			return true;
+	}
+
+	return false;
+}
+
+/* Message output should match DALSMC_VERSION in dalsmc.h */
+bool dcn401_smu_check_msg_header_version(struct clk_mgr_internal *clk_mgr)
+{
+	uint32_t response = 0;
+
+	smu_print("SMU Check msg header version\n");
+
+	if (dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetMsgHeaderVersion, 0, &response)) {
+
+		smu_print("SMU msg header version: %d\n", response);
+
+		if (response == DALSMC_VERSION)
+			return true;
+	}
+
 	return false;
 }
 
···
 
 	dcn401_smu_send_msg_with_param(clk_mgr, DALSMC_MSG_SetCabForUclkPstate, param, NULL);
 	smu_print("Numways for SubVP : %d\n", num_ways);
+}
+
+void dcn401_smu_set_dram_addr_high(struct clk_mgr_internal *clk_mgr, uint32_t addr_high)
+{
+	smu_print("SMU Set DRAM addr high: %d\n", addr_high);
+
+	dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_SetDalDramAddrHigh, addr_high, NULL);
+}
+
+void dcn401_smu_set_dram_addr_low(struct clk_mgr_internal *clk_mgr, uint32_t addr_low)
+{
+	smu_print("SMU Set DRAM addr low: %d\n", addr_low);
+
+	dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_SetDalDramAddrLow, addr_low, NULL);
 }
 
 void dcn401_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr)
···
 	dcn401_smu_send_msg_with_param(clk_mgr, DALSMC_MSG_GetNumUmcChannels, 0, &response);
 
 	smu_print("SMU Get Num UMC Channels: num_umc_channels = %d\n", response);
+
+	return response;
+}
+
+/*
+ * Frequency in MHz returned in lower 16 bits for valid DPM level
+ *
+ * Call with dpm_level = 0xFF to query features, return value will be:
+ *     Bits 7:0 - number of DPM levels
+ *     Bit   28 - 1 = auto DPM on
+ *     Bit   29 - 1 = sweep DPM on
+ *     Bit   30 - 1 = forced DPM on
+ *     Bit   31 - 0 = discrete, 1 = fine-grained
+ *
+ * With fine-grained DPM, only min and max frequencies will be reported
+ *
+ * Returns 0 on failure
+ */
+unsigned int dcn401_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint8_t dpm_level)
+{
+	uint32_t response = 0;
+
+	/* bits 23:16 for clock type, lower 8 bits for DPM level */
+	uint32_t param = (clk << 16) | dpm_level;
+
+	smu_print("SMU Get dpm freq by index: clk = %d, dpm_level = %d\n", clk, dpm_level);
+
+	dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetDpmFreqByIndex, param, &response);
+
+	smu_print("SMU dpm freq: %d MHz\n", response);
+
+	return response;
+}
+
+/* Returns the max DPM frequency in DC mode in MHz, 0 on failure */
+unsigned int dcn401_smu_get_dc_mode_max_dpm_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk)
+{
+	uint32_t response = 0;
+
+	/* bits 23:16 for clock type */
+	uint32_t param = clk << 16;
+
+	smu_print("SMU Get DC mode max DPM freq: clk = %d\n", clk);
+
+	dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetDcModeMaxDpmFreq, param, &response);
+
+	smu_print("SMU DC mode max DMP freq: %d MHz\n", response);
 
 	return response;
 }

drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.h

···
 
 #include "os_types.h"
 #include "core_types.h"
-#include "dcn32/dcn32_clk_mgr_smu_msg.h"
 
+struct clk_mgr_internal;
+
+bool dcn401_smu_get_smu_version(struct clk_mgr_internal *clk_mgr, unsigned int *version);
+bool dcn401_smu_check_driver_if_version(struct clk_mgr_internal *clk_mgr);
+bool dcn401_smu_check_msg_header_version(struct clk_mgr_internal *clk_mgr);
 void dcn401_smu_send_fclk_pstate_message(struct clk_mgr_internal *clk_mgr, bool support);
 void dcn401_smu_send_uclk_pstate_message(struct clk_mgr_internal *clk_mgr, bool support);
 void dcn401_smu_send_cab_for_uclk_message(struct clk_mgr_internal *clk_mgr, unsigned int num_ways);
+void dcn401_smu_set_dram_addr_high(struct clk_mgr_internal *clk_mgr, uint32_t addr_high);
+void dcn401_smu_set_dram_addr_low(struct clk_mgr_internal *clk_mgr, uint32_t addr_low);
 void dcn401_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr);
 void dcn401_smu_set_pme_workaround(struct clk_mgr_internal *clk_mgr);
 unsigned int dcn401_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint16_t freq_mhz);
···
 void dcn401_smu_set_min_deep_sleep_dcef_clk(struct clk_mgr_internal *clk_mgr, uint32_t freq_mhz);
 void dcn401_smu_set_num_of_displays(struct clk_mgr_internal *clk_mgr, uint32_t num_displays);
 unsigned int dcn401_smu_get_num_of_umc_channels(struct clk_mgr_internal *clk_mgr);
+unsigned int dcn401_smu_get_dc_mode_max_dpm_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk);
+unsigned int dcn401_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint8_t dpm_level);
 
 #endif /* __DCN401_CLK_MGR_SMU_MSG_H_ */

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

···
 	 * avoid conflicting with firmware updates.
 	 */
 	if (dc->ctx->dce_version > DCE_VERSION_MAX) {
-		if ((dc->optimized_required || dc->wm_optimized_required) &&
+		if (dc->optimized_required &&
 			(stream->adjust.v_total_max != adjust->v_total_max ||
 			stream->adjust.v_total_min != adjust->v_total_min)) {
 			stream->adjust.timing_adjust_pending = true;
···
 	}
 
 	dc->optimized_required = false;
-	dc->wm_optimized_required = false;
 }
 
 bool dc_set_generic_gpio_for_stereo(bool enable,
···
 		} else if (memcmp(&dc->current_state->bw_ctx.bw.dcn.clk, &dc->clk_mgr->clks, offsetof(struct dc_clocks, prev_p_state_change_support)) != 0) {
 			dc->optimized_required = true;
 		}
-
-		dc->optimized_required |= dc->wm_optimized_required;
 	}
 
 	return type;
···
 
 	if (update->adaptive_sync_infopacket)
 		stream->adaptive_sync_infopacket = *update->adaptive_sync_infopacket;
+
+	if (update->avi_infopacket)
+		stream->avi_infopacket = *update->avi_infopacket;
 
 	if (update->dither_option)
 		stream->dither_option = *update->dither_option;
···
 					stream_update->vsp_infopacket ||
 					stream_update->hfvsif_infopacket ||
 					stream_update->adaptive_sync_infopacket ||
-					stream_update->vtem_infopacket) {
+					stream_update->vtem_infopacket ||
+					stream_update->avi_infopacket) {
 				resource_build_info_frame(pipe_ctx);
 				dc->hwss.update_info_frame(pipe_ctx);
 
···
 			stream_update->hfvsif_infopacket ||
 			stream_update->vtem_infopacket ||
 			stream_update->adaptive_sync_infopacket ||
+			stream_update->avi_infopacket ||
 			stream_update->dpms_off ||
 			stream_update->allow_freesync ||
 			stream_update->vrr_active_variable ||

drivers/gpu/drm/amd/display/dc/core/dc_resource.c

···
 	unsigned int fr_ind = pipe_ctx->stream->timing.fr_index;
 	enum dc_timing_3d_format format;
 
+	if (stream->avi_infopacket.valid) {
+		*info_packet = stream->avi_infopacket;
+		return;
+	}
+
 	memset(&hdmi_info, 0, sizeof(union hdmi_info_packet));
+
 
 	color_space = pipe_ctx->stream->output_color_space;
 	if (color_space == COLOR_SPACE_UNKNOWN)

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

···
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.350"
+#define DC_VER "3.2.351"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC
···
 	unsigned int auxless_alpm_lfps_silence_ns;
 	unsigned int auxless_alpm_lfps_t1t2_us;
 	short auxless_alpm_lfps_t1t2_offset_us;
+	bool disable_stutter_for_wm_program;
 };
 
 
···
 		uint32_t in_transfer_func_change:1;
 		uint32_t input_csc_change:1;
 		uint32_t coeff_reduction_change:1;
-		uint32_t output_tf_change:1;
 		uint32_t pixel_format_change:1;
 		uint32_t plane_size_change:1;
 		uint32_t gamut_remap_change:1;
···
 
 	/* Require to optimize clocks and bandwidth for added/removed planes */
 	bool optimized_required;
-	bool wm_optimized_required;
 	bool idle_optimizations_allowed;
 	bool enable_c20_dtm_b0;
 

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

···
 	struct dc_info_packet hfvsif_infopacket;
 	struct dc_info_packet vtem_infopacket;
 	struct dc_info_packet adaptive_sync_infopacket;
+	struct dc_info_packet avi_infopacket;
 	uint8_t dsc_packed_pps[128];
 	struct rect src; /* composition area */
 	struct rect dst; /* stream addressable area */
···
 	struct dc_info_packet *hfvsif_infopacket;
 	struct dc_info_packet *vtem_infopacket;
 	struct dc_info_packet *adaptive_sync_infopacket;
+	struct dc_info_packet *avi_infopacket;
+
 	bool *dpms_off;
 	bool integer_scaling_update;
 	bool *allow_freesync;

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

···
 		bool rc_disable;
 		bool rc_allow_static_screen;
 		bool rc_allow_fullscreen_VPB;
+		bool read_psrcap_again;
 		unsigned int replay_enable_option;
 	} psr;
 	/* ABM */

drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c

···
 
 #define CTX \
 	dccg_dcn->base.ctx
+#include "logger_types.h"
 #define DC_LOGGER \
 	dccg->ctx->logger
 
···
 	default:
 		break;
 	}
-	//DC_LOG_DEBUG("%s: dpp_inst(%d) DPPCLK_EN = %d\n", __func__, dpp_inst, enable);
+	DC_LOG_DEBUG("%s: dpp_inst(%d) DPPCLK_EN = %d\n", __func__, dpp_inst, enable);
 
 }
 
···
 		 * PIPEx_DTO_SRC_SEL should not be programmed during DTBCLK update since OTG may still be on, and the
 		 * programming is handled in program_pix_clk() regardless, so it can be removed from here.
 		 */
+		DC_LOG_DEBUG("%s: OTG%d DTBCLK DTO enabled: pixclk_khz=%d, ref_dtbclk_khz=%d, req_dtbclk_khz=%d, phase=%d, modulo=%d\n",
+				__func__, params->otg_inst, params->pixclk_khz,
+				params->ref_dtbclk_khz, req_dtbclk_khz, phase, modulo);
+
 	} else {
 		switch (params->otg_inst) {
 		case 0:
···
 
 		REG_WRITE(DTBCLK_DTO_MODULO[params->otg_inst], 0);
 		REG_WRITE(DTBCLK_DTO_PHASE[params->otg_inst], 0);
+
+		DC_LOG_DEBUG("%s: OTG%d DTBCLK DTO disabled\n", __func__, params->otg_inst);
 	}
 }
 
···
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+	DC_LOG_DEBUG("%s: dp_hpo_inst(%d) DPSTREAMCLK_EN = %d, DPSTREAMCLK_SRC_SEL = %d\n",
+			__func__, dp_hpo_inst, (src == REFCLK) ? 0 : 1, otg_inst);
 }
 
 
···
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+	DC_LOG_DEBUG("%s: dp_hpo_inst(%d) DPSTREAMCLK_ROOT_GATE_DISABLE = %d\n",
+			__func__, dp_hpo_inst, enable ? 1 : 0);
 }
 
 
···
 		BREAK_TO_DEBUGGER();
 		return;
 	}
-	//DC_LOG_DEBUG("%s: dpp_inst(%d) PHYESYMCLK_ROOT_GATE_DISABLE:\n", __func__, phy_inst, enable ? 0 : 1);
+	DC_LOG_DEBUG("%s: dpp_inst(%d) PHYESYMCLK_ROOT_GATE_DISABLE: %d\n", __func__, phy_inst, enable ? 0 : 1);
 
 }
 
···
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+	DC_LOG_DEBUG("%s: phy_inst(%d) PHYxSYMCLK_EN = %d, PHYxSYMCLK_SRC_SEL = %d\n",
+			__func__, phy_inst, force_enable ? 1 : 0, clk_src);
 }
 
 static void dccg35_set_valid_pixel_rate(
···
 	}
 
 	dccg->dpp_clock_gated[dpp_inst] = !clock_on;
+	DC_LOG_DEBUG("%s: dpp_inst(%d) clock_on = %d\n", __func__, dpp_inst, clock_on);
 }
 
 static void dccg35_disable_symclk32_se(
···
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+
 }
 
 static void dccg35_init_cb(struct dccg *dccg)
···
 	(void)dccg;
 	/* Any RCG should be done when driver enter low power mode*/
 }
-
 void dccg35_init(struct dccg *dccg)
 {
 	int otg_inst;
···
 		for (otg_inst = 0; otg_inst < 2; otg_inst++) {
 			dccg31_disable_symclk32_le(dccg, otg_inst);
 			dccg31_set_symclk32_le_root_clock_gating(dccg, otg_inst, false);
+			DC_LOG_DEBUG("%s: OTG%d SYMCLK32_LE disabled and root clock gating disabled\n",
+					__func__, otg_inst);
 		}
 
 //	if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se)
···
 			dccg35_set_dpstreamclk(dccg, REFCLK, otg_inst,
 						otg_inst);
 			dccg35_set_dpstreamclk_root_clock_gating(dccg, otg_inst, false);
+			DC_LOG_DEBUG("%s: OTG%d DPSTREAMCLK disabled and root clock gating disabled\n",
+					__func__, otg_inst);
 		}
 
 /*

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

···
 	copy_settings_data->max_deviation_line			= link->dpcd_caps.pr_info.max_deviation_line;
 	copy_settings_data->smu_optimizations_en		= link->replay_settings.replay_smu_opt_enable;
 	copy_settings_data->replay_timing_sync_supported = link->replay_settings.config.replay_timing_sync_supported;
+	copy_settings_data->replay_support_fast_resync_in_ultra_sleep_mode = link->replay_settings.config.replay_support_fast_resync_in_ultra_sleep_mode;
 
 	copy_settings_data->debug.bitfields.enable_ips_visual_confirm = dc->dc->debug.enable_ips_visual_confirm;
 

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

···
 /*
  * SMU message tracing
  */
-void dm_trace_smu_msg(uint32_t msg_id, uint32_t param_in, struct dc_context *ctx);
-void dm_trace_smu_delay(uint32_t delay, struct dc_context *ctx);
+void dm_trace_smu_enter(uint32_t msg_id, uint32_t param_in, unsigned int delay, struct dc_context *ctx);
+void dm_trace_smu_exit(bool success, uint32_t response, struct dc_context *ctx);
 
-#define TRACE_SMU_MSG(msg_id, param_in, ctx)	dm_trace_smu_msg(msg_id, param_in, ctx)
-#define TRACE_SMU_DELAY(response_delay, ctx)	dm_trace_smu_delay(response_delay, ctx)
-
+#define TRACE_SMU_MSG_DELAY(msg_id, param_in, delay, ctx)	dm_trace_smu_enter(msg_id, param_in, delay, ctx)
+#define TRACE_SMU_MSG(msg_id, param_in, ctx)	dm_trace_smu_enter(msg_id, param_in, 0, ctx)
+#define TRACE_SMU_MSG_ENTER(msg_id, param_in, ctx)	dm_trace_smu_enter(msg_id, param_in, 0, ctx)
+#define TRACE_SMU_MSG_EXIT(success, response, ctx)	dm_trace_smu_exit(success, response, ctx)
 
 /*
  * DMUB Interfaces

drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.c

···
 
 static void remove_duplicates(double *list_a, int *list_a_size)
 {
-	int cur_element = 0;
-	// For all elements b[i] in list_b[]
-	while (cur_element < *list_a_size - 1) {
-		if (list_a[cur_element] == list_a[cur_element + 1]) {
-			for (int j = cur_element + 1; j < *list_a_size - 1; j++) {
-				list_a[j] = list_a[j + 1];
-			}
-			*list_a_size = *list_a_size - 1;
-		} else {
-			cur_element++;
+	int j = 0;
+
+	if (*list_a_size == 0)
+		return;
+
+	for (int i = 1; i < *list_a_size; i++) {
+		if (list_a[j] != list_a[i]) {
+			j++;
+			list_a[j] = list_a[i];
 		}
 	}
+
+	*list_a_size = j + 1;
 }
 
 static bool increase_mpc_combine_factor(unsigned int *mpc_combine_factor, unsigned int limit)

drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c

···
 #include "dcn32_hubbub.h"
 #include "dm_services.h"
 #include "reg_helper.h"
+#include "dal_asic_id.h"
 
 
 #define CTX \
···
 			COMPBUF_RESERVED_SPACE_64B, hubbub2->pixel_chunk_size / 32,
 			COMPBUF_RESERVED_SPACE_ZS, hubbub2->pixel_chunk_size / 128);
 	REG_UPDATE(DCHUBBUB_DEBUG_CTRL_0, DET_DEPTH, 0x47F);
+}
+
+static void hubbub32_set_sdp_control(struct hubbub *hubbub, bool dc_control)
+{
+	struct dcn20_hubbub *hubbub2 = TO_DCN20_HUBBUB(hubbub);
+
+	REG_UPDATE(DCHUBBUB_SDPIF_CFG0,
+			SDPIF_PORT_CONTROL, dc_control);
 }
 
 void hubbub32_set_request_limit(struct hubbub *hubbub, int memory_channel_count, int words_per_channel)
···
 		unsigned int refclk_mhz,
 		bool safe_to_lower)
 {
+	struct dc *dc = hubbub->ctx->dc;
 	bool wm_pending = false;
+
+	if (!safe_to_lower && dc->debug.disable_stutter_for_wm_program &&
+			(ASICREV_IS_GC_11_0_0(dc->ctx->asic_id.hw_internal_rev) ||
+			ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev))) {
+		/* before raising watermarks, SDP control give to DF, stutter must be disabled */
+		wm_pending = true;
+		hubbub32_set_sdp_control(hubbub, false);
+		hubbub1_allow_self_refresh_control(hubbub, false);
+	}
 
 	if (hubbub32_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower))
 		wm_pending = true;
···
 	REG_UPDATE(DCHUBBUB_ARB_DF_REQ_OUTSTAND,
 			DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 0x1FF);*/
 
-	if (safe_to_lower || hubbub->ctx->dc->debug.disable_stutter)
-		hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter);
+	if (safe_to_lower) {
+		/* after lowering watermarks, stutter setting is restored, SDP control given to DC */
+		hubbub1_allow_self_refresh_control(hubbub, !dc->debug.disable_stutter);
 
-	hubbub32_force_usr_retraining_allow(hubbub, hubbub->ctx->dc->debug.force_usr_allow);
+		if (dc->debug.disable_stutter_for_wm_program &&
+				(ASICREV_IS_GC_11_0_0(dc->ctx->asic_id.hw_internal_rev) ||
+				ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev))) {
+			hubbub32_set_sdp_control(hubbub, true);
+		}
+	} else if (dc->debug.disable_stutter) {
+		hubbub1_allow_self_refresh_control(hubbub, !dc->debug.disable_stutter);
+	}
+
+	hubbub32_force_usr_retraining_allow(hubbub, dc->debug.force_usr_allow);
 
 	return wm_pending;
 }
···
 	ignore the "df_pre_cstate_req" from the SDP port control.
 	only the DCN will determine when to connect the SDP port
 	*/
-	REG_UPDATE(DCHUBBUB_SDPIF_CFG0,
-			SDPIF_PORT_CONTROL, 1);
+	hubbub32_set_sdp_control(hubbub, true);
 	/*Set SDP's max outstanding request to 512
 	must set the register back to 0 (max outstanding = 256) in zero frame buffer mode*/
 	REG_UPDATE(DCHUBBUB_SDPIF_CFG1,

drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c

···
 			context,
 			false);
 
-	dc->wm_optimized_required = hubbub->funcs->program_watermarks(hubbub,
+	dc->optimized_required = hubbub->funcs->program_watermarks(hubbub,
 			&context->bw_ctx.bw.dcn.watermarks,
 			dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000,
 			true);

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

···
 	 * updating on slave planes
 	 */
 	if (pipe_ctx->update_flags.bits.enable ||
-		pipe_ctx->update_flags.bits.plane_changed ||
-		pipe_ctx->stream->update_flags.bits.out_tf ||
-		(pipe_ctx->plane_state &&
-			pipe_ctx->plane_state->update_flags.bits.output_tf_change))
+	    pipe_ctx->update_flags.bits.plane_changed ||
+	    pipe_ctx->stream->update_flags.bits.out_tf)
 		hws->funcs.set_output_transfer_func(dc, pipe_ctx, pipe_ctx->stream);
 
 	/* If the pipe has been enabled or has a different opp, we
···
 	}
 
 	/* program dchubbub watermarks:
-	 * For assigning wm_optimized_required, use |= operator since we don't want
+	 * For assigning optimized_required, use |= operator since we don't want
 	 * to clear the value if the optimize has not happened yet
 	 */
-	dc->wm_optimized_required |= hubbub->funcs->program_watermarks(hubbub,
+	dc->optimized_required |= hubbub->funcs->program_watermarks(hubbub,
 					&context->bw_ctx.bw.dcn.watermarks,
 					dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000,
 					false);
···
 	if (hubbub->funcs->program_compbuf_size) {
 		if (context->bw_ctx.dml.ip.min_comp_buffer_size_kbytes) {
 			compbuf_size_kb = context->bw_ctx.dml.ip.min_comp_buffer_size_kbytes;
-			dc->wm_optimized_required |= (compbuf_size_kb != dc->current_state->bw_ctx.dml.ip.min_comp_buffer_size_kbytes);
+			dc->optimized_required |= (compbuf_size_kb != dc->current_state->bw_ctx.dml.ip.min_comp_buffer_size_kbytes);
 		} else {
 			compbuf_size_kb = context->bw_ctx.bw.dcn.compbuf_size_kb;
-			dc->wm_optimized_required |= (compbuf_size_kb != dc->current_state->bw_ctx.bw.dcn.compbuf_size_kb);
+			dc->optimized_required |= (compbuf_size_kb != dc->current_state->bw_ctx.bw.dcn.compbuf_size_kb);
 		}
 
 		hubbub->funcs->program_compbuf_size(hubbub, compbuf_size_kb, false);
···
 		res_pool->dccg->funcs->dccg_init(res_pool->dccg);
 
 	//Enable ability to power gate / don't force power on permanently
-	hws->funcs.enable_power_gating_plane(hws, true);
+	if (hws->funcs.enable_power_gating_plane)
+		hws->funcs.enable_power_gating_plane(hws, true);
 
 	// Specific to FPGA dccg and registers
 	REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF);

drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c

···
 			false);
 
 	/* program dchubbub watermarks:
-	 * For assigning wm_optimized_required, use |= operator since we don't want
+	 * For assigning optimized_required, use |= operator since we don't want
 	 * to clear the value if the optimize has not happened yet
 	 */
-	dc->wm_optimized_required |= hubbub->funcs->program_watermarks(hubbub,
+	dc->optimized_required |= hubbub->funcs->program_watermarks(hubbub,
 					&context->bw_ctx.bw.dcn.watermarks,
 					dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000,
 					false);
 	/* update timeout thresholds */
 	if (hubbub->funcs->program_arbiter) {
-		dc->wm_optimized_required |= hubbub->funcs->program_arbiter(hubbub, &context->bw_ctx.bw.dcn.arb_regs, false);
+		dc->optimized_required |= hubbub->funcs->program_arbiter(hubbub, &context->bw_ctx.bw.dcn.arb_regs, false);
 	}
 
 	/* decrease compbuf size */
 	if (hubbub->funcs->program_compbuf_segments) {
 		compbuf_size = context->bw_ctx.bw.dcn.arb_regs.compbuf_size;
-		dc->wm_optimized_required |= (compbuf_size != dc->current_state->bw_ctx.bw.dcn.arb_regs.compbuf_size);
+		dc->optimized_required |= (compbuf_size != dc->current_state->bw_ctx.bw.dcn.arb_regs.compbuf_size);
 
 		hubbub->funcs->program_compbuf_segments(hubbub, compbuf_size, false);
 	}
···
 	 * updating on slave planes
 	 */
 	if (pipe_ctx->update_flags.bits.enable ||
-		pipe_ctx->update_flags.bits.plane_changed ||
-		pipe_ctx->stream->update_flags.bits.out_tf ||
-		(pipe_ctx->plane_state &&
-			pipe_ctx->plane_state->update_flags.bits.output_tf_change))
+	    pipe_ctx->update_flags.bits.plane_changed ||
+	    pipe_ctx->stream->update_flags.bits.out_tf)
 		hws->funcs.set_output_transfer_func(dc, pipe_ctx, pipe_ctx->stream);
 
 	/* If the pipe has been enabled or has a different opp, we

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

···
 	if (!link)
 		return false;
 
+	/* This is a workaround: some vendors require the source to
+	 * read the PSR cap; otherwise, the vendor's PSR feature will
+	 * fall back to its default behavior, causing a misconfiguration
+	 * of this feature.
+	 */
+	if (link->panel_config.psr.read_psrcap_again) {
+		dm_helpers_dp_read_dpcd(
+			link->ctx,
+			link,
+			DP_PSR_SUPPORT,
+			&link->dpcd_caps.psr_info.psr_version,
+			sizeof(link->dpcd_caps.psr_info.psr_version));
+	}
+
 	//Clear PSR cfg
 	memset(&psr_configuration, 0, sizeof(psr_configuration));
 	dm_helpers_dp_write_dpcd(

drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c

···
 #include "stream_encoder.h"
 
 #include "resource.h"
+#include "clk_mgr.h"
 #include "include/irq_service_interface.h"
 #include "virtual/virtual_stream_encoder.h"
 #include "dce110/dce110_resource.h"
···
 	return DC_OK;
 }
 
-static enum dc_status dce100_validate_bandwidth(
+enum dc_status dce100_validate_bandwidth(
 	struct dc  *dc,
 	struct dc_state *context,
 	enum dc_validate_mode validate_mode)
 {
 	int i;
 	bool at_least_one_pipe = false;
+	struct dc_stream_state *stream = NULL;
+	const uint32_t max_pix_clk_khz = max(dc->clk_mgr->clks.max_supported_dispclk_khz, 400000);
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		if (context->res_ctx.pipe_ctx[i].stream)
+		stream = context->res_ctx.pipe_ctx[i].stream;
+		if (stream) {
 			at_least_one_pipe = true;
+
+			if (stream->timing.pix_clk_100hz >= max_pix_clk_khz * 10)
+				return DC_FAIL_BANDWIDTH_VALIDATE;
+		}
 	}
 
 	if (at_least_one_pipe) {
···
 		context->bw_ctx.bw.dce.dispclk_khz = 681000;
 		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
 	} else {
-		context->bw_ctx.bw.dce.dispclk_khz = 0;
+		/* On DCE 6.0 and 6.4 the PLL0 is both the display engine clock and
+		 * the DP clock, and shouldn't be turned off. Just select the display
+		 * clock value from its low power mode.
+		 */
+		if (dc->ctx->dce_version == DCE_VERSION_6_0 ||
+			dc->ctx->dce_version == DCE_VERSION_6_4)
+			context->bw_ctx.bw.dce.dispclk_khz = 352000;
+		else
+			context->bw_ctx.bw.dce.dispclk_khz = 0;
+
 		context->bw_ctx.bw.dce.yclk_khz = 0;
 	}
 
···
 	return true;
 }
 
-static enum dc_status dce100_validate_global(
+enum dc_status dce100_validate_global(
 		struct dc  *dc,
 		struct dc_state *context)
 {

drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h

···
 
 enum dc_status dce100_validate_plane(const struct dc_plane_state *plane_state, struct dc_caps *caps);
 
+enum dc_status dce100_validate_global(
+		struct dc  *dc,
+		struct dc_state *context);
+
+enum dc_status dce100_validate_bandwidth(
+		struct dc  *dc,
+		struct dc_state *context,
+		enum dc_validate_mode validate_mode);
+
 enum dc_status dce100_add_stream_to_ctx(
 		struct dc *dc,
 		struct dc_state *new_ctx,

drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c

···
 #include "stream_encoder.h"
 
 #include "resource.h"
+#include "clk_mgr.h"
 #include "include/irq_service_interface.h"
 #include "irq/dce60/irq_service_dce60.h"
 #include "dce110/dce110_timing_generator.h"
···
 	}
 }
 
-static enum dc_status dce60_validate_bandwidth(
-	struct dc *dc,
-	struct dc_state *context,
-	enum dc_validate_mode validate_mode)
-{
-	int i;
-	bool at_least_one_pipe = false;
-
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		if (context->res_ctx.pipe_ctx[i].stream)
-			at_least_one_pipe = true;
-	}
-
-	if (at_least_one_pipe) {
-		/* TODO implement when needed but for now hardcode max value*/
-		context->bw_ctx.bw.dce.dispclk_khz = 681000;
-		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
-	} else {
-		/* On DCE 6.0 and 6.4 the PLL0 is both the display engine clock and
-		 * the DP clock, and shouldn't be turned off. Just select the display
-		 * clock value from its low power mode.
-		 */
-		if (dc->ctx->dce_version == DCE_VERSION_6_0 ||
-			dc->ctx->dce_version == DCE_VERSION_6_4)
-			context->bw_ctx.bw.dce.dispclk_khz = 352000;
-		else
-			context->bw_ctx.bw.dce.dispclk_khz = 0;
-
-		context->bw_ctx.bw.dce.yclk_khz = 0;
-	}
-
-	return DC_OK;
-}
-
-static bool dce60_validate_surface_sets(
-		struct dc_state *context)
-{
-	int i;
-
-	for (i = 0; i < context->stream_count; i++) {
-		if (context->stream_status[i].plane_count == 0)
-			continue;
-
-		if (context->stream_status[i].plane_count > 1)
-			return false;
-
-		if (context->stream_status[i].plane_states[0]->format
-				>= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN)
-			return false;
-	}
-
-	return true;
-}
-
-static enum dc_status dce60_validate_global(
-		struct dc *dc,
-		struct dc_state *context)
-{
-	if (!dce60_validate_surface_sets(context))
-		return DC_FAIL_SURFACE_VALIDATE;
-
-	return DC_OK;
-}
-
 static void dce60_destroy_resource_pool(struct resource_pool **pool)
 {
 	struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool);
···
 	.destroy = dce60_destroy_resource_pool,
 	.link_enc_create = dce60_link_encoder_create,
 	.panel_cntl_create = dce60_panel_cntl_create,
-	.validate_bandwidth = dce60_validate_bandwidth,
+	.validate_bandwidth = dce100_validate_bandwidth,
 	.validate_plane = dce100_validate_plane,
 	.add_stream_to_ctx = dce100_add_stream_to_ctx,
-	.validate_global = dce60_validate_global,
+	.validate_global = dce100_validate_global,
 	.find_first_free_match_stream_enc_for_link = dce100_find_first_free_match_stream_enc_for_link
 };
 

drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c

···
 #include "stream_encoder.h"
 
 #include "resource.h"
+#include "clk_mgr.h"
 #include "include/irq_service_interface.h"
 #include "irq/dce80/irq_service_dce80.h"
 #include "dce110/dce110_timing_generator.h"
···
 	}
 }
 
-static enum dc_status dce80_validate_bandwidth(
-	struct dc *dc,
-	struct dc_state *context,
-	enum dc_validate_mode validate_mode)
-{
-	int i;
-	bool at_least_one_pipe = false;
-
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		if (context->res_ctx.pipe_ctx[i].stream)
-			at_least_one_pipe = true;
-	}
-
-	if (at_least_one_pipe) {
-		/* TODO implement when needed but for now hardcode max value*/
-		context->bw_ctx.bw.dce.dispclk_khz = 681000;
-		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
-	} else {
-		context->bw_ctx.bw.dce.dispclk_khz = 0;
-		context->bw_ctx.bw.dce.yclk_khz = 0;
-	}
-
-	return DC_OK;
-}
-
-static bool dce80_validate_surface_sets(
-		struct dc_state *context)
-{
-	int i;
-
-	for (i = 0; i < context->stream_count; i++) {
-		if (context->stream_status[i].plane_count == 0)
-			continue;
-
-		if (context->stream_status[i].plane_count > 1)
-			return false;
-
-		if (context->stream_status[i].plane_states[0]->format
-				>= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN)
-			return false;
-	}
-
-	return true;
-}
-
-static enum dc_status dce80_validate_global(
-		struct dc *dc,
-		struct dc_state *context)
-{
-	if (!dce80_validate_surface_sets(context))
-		return DC_FAIL_SURFACE_VALIDATE;
-
-	return DC_OK;
-}
-
 static void dce80_destroy_resource_pool(struct resource_pool **pool)
 {
 	struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool);
···
 	.destroy = dce80_destroy_resource_pool,
 	.link_enc_create = dce80_link_encoder_create,
 	.panel_cntl_create = dce80_panel_cntl_create,
-	.validate_bandwidth = dce80_validate_bandwidth,
+	.validate_bandwidth = dce100_validate_bandwidth,
 	.validate_plane = dce100_validate_plane,
 	.add_stream_to_ctx = dce100_add_stream_to_ctx,
-	.validate_global = dce80_validate_global,
+	.validate_global = dce100_validate_global,
 	.find_first_free_match_stream_enc_for_link = dce100_find_first_free_match_stream_enc_for_link
 };
 

drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c

···
 	.enable_legacy_fast_update = true,
 	.using_dml2 = false,
 	.disable_dsc_power_gate = true,
+	.min_disp_clk_khz = 100000,
 };
 
 static const struct dc_panel_config panel_config_defaults = {

drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c

···
 	.fpo_vactive_min_active_margin_us = 200,
 	.fpo_vactive_max_blank_us = 1000,
 	.enable_legacy_fast_update = false,
+	.disable_stutter_for_wm_program = true
 };
 
 static struct dce_aux *dcn32_aux_engine_create(

drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h

···
       SR(DCHUBBUB_ARB_MALL_CNTL),                                              \
       SR(DCN_VM_FAULT_ADDR_MSB), SR(DCN_VM_FAULT_ADDR_LSB),                    \
       SR(DCN_VM_FAULT_CNTL), SR(DCN_VM_FAULT_STATUS),                          \
-      SR(SDPIF_REQUEST_RATE_LIMIT)
+      SR(SDPIF_REQUEST_RATE_LIMIT),                                            \
+      SR(DCHUBBUB_SDPIF_CFG0)
 
 /* DCCG */
 

drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h

···
 	 */
 	uint8_t hpo_link_enc_inst;
 	/**
+	 * Determines if fast resync in ultra sleep mode is enabled/disabled.
+	 */
+	uint8_t replay_support_fast_resync_in_ultra_sleep_mode;
+	/**
 	 * @pad: Align structure to 4 byte boundary.
 	 */
-	uint8_t pad[2];
+	uint8_t pad[1];
 };
 
 

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

···
 				uint32_t enable_mes_sch_stb_log : 1;
 				uint32_t limit_single_process : 1;
 				uint32_t is_strix_tmz_wa_enabled  :1;
-				uint32_t reserved : 13;
+				uint32_t enable_lr_compute_wa : 1;
+				uint32_t reserved : 12;
 			};
 			uint32_t	uint32_t_all;
 		};

drivers/gpu/drm/amd/include/mes_v12_api_def.h

···
 				uint32_t limit_single_process : 1;
 				uint32_t unmapped_doorbell_handling: 2;
 				uint32_t enable_mes_fence_int: 1;
-				uint32_t reserved : 10;
+				uint32_t enable_lr_compute_wa : 1;
+				uint32_t reserved : 9;
 			};
 			uint32_t uint32_all;
 		};

drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h

···
 #define PPSMC_MSG_GetBadPageSeverity                0x5B
 #define PPSMC_MSG_GetSystemMetricsTable             0x5C
 #define PPSMC_MSG_GetSystemMetricsVersion           0x5D
-#define PPSMC_Message_Count                         0x5E
+#define PPSMC_MSG_ResetVCN                          0x5E
+#define PPSMC_Message_Count                         0x5F
 
 //PPSMC Reset Types for driver msg argument
 #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET        0x1

drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c

···
 	MSG_MAP(RmaDueToBadPageThreshold,            PPSMC_MSG_RmaDueToBadPageThreshold,        0),
 	MSG_MAP(SetThrottlingPolicy,                 PPSMC_MSG_SetThrottlingPolicy,             0),
 	MSG_MAP(ResetSDMA,                           PPSMC_MSG_ResetSDMA,                       0),
+	MSG_MAP(ResetVCN,                            PPSMC_MSG_ResetVCN,                        0),
 	MSG_MAP(GetStaticMetricsTable,               PPSMC_MSG_GetStaticMetricsTable,           1),
 	MSG_MAP(GetSystemMetricsTable,               PPSMC_MSG_GetSystemMetricsTable,           1),
 };

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

···
 		smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION));
 	}
 
+	if (fw_ver > 0x04560900)
+		smu_v13_0_6_cap_set(smu, SMU_CAP(VCN_RESET));
+
 	if (fw_ver >= 0x04560700) {
 		if (fw_ver >= 0x04560900) {
 			smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS));

drivers/gpu/drm/amd/ras/rascore/ras_core_status.h

···
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __RAS_CORE_STATUS_H__
+#define __RAS_CORE_STATUS_H__
+
+#define RAS_CORE_OK                       0
+#define RAS_CORE_NOT_SUPPORTED            248
+#define RAS_CORE_FAIL_ERROR_QUERY         249
+#define RAS_CORE_FAIL_ERROR_INJECTION     250
+#define RAS_CORE_FAIL_FATAL_RECOVERY      251
+#define RAS_CORE_FAIL_POISON_CONSUMPTION  252
+#define RAS_CORE_FAIL_POISON_CREATION     253
+#define RAS_CORE_FAIL_NO_VALID_BANKS      254
+#define RAS_CORE_GPU_IN_MODE1_RESET       255
+#endif