sync with OpenBSD -current

sys/dev/pci/drm/amd/amdgpu/amdgpu_amdkfd.h

@@ -303,6 +303,7 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu(struct amdgpu_device *adev,
 					  struct kgd_mem *mem, void *drm_priv);
 int amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(
 		struct amdgpu_device *adev, struct kgd_mem *mem, void *drm_priv);
+int amdgpu_amdkfd_gpuvm_dmaunmap_mem(struct kgd_mem *mem, void *drm_priv);
 int amdgpu_amdkfd_gpuvm_sync_memory(
 		struct amdgpu_device *adev, struct kgd_mem *mem, bool intr);
 int amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(struct kgd_mem *mem,

sys/dev/pci/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c

@@ -733,7 +733,7 @@ kfd_mem_dmaunmap_sg_bo(struct kgd_mem *mem,
 	enum dma_data_direction dir;
 
 	if (unlikely(!ttm->sg)) {
-		pr_err("SG Table of BO is UNEXPECTEDLY NULL");
+		pr_debug("SG Table of BO is NULL");
 		return;
 	}
 
@@ -1202,8 +1202,6 @@ static void unmap_bo_from_gpuvm(struct kgd_mem *mem,
 	amdgpu_vm_clear_freed(adev, vm, &bo_va->last_pt_update);
 
 	amdgpu_sync_fence(sync, bo_va->last_pt_update);
-
-	kfd_mem_dmaunmap_attachment(mem, entry);
 }
 
 static int update_gpuvm_pte(struct kgd_mem *mem,
@@ -1258,6 +1256,7 @@ static int map_bo_to_gpuvm(struct kgd_mem *mem,
 
 update_gpuvm_pte_failed:
 	unmap_bo_from_gpuvm(mem, entry, sync);
+	kfd_mem_dmaunmap_attachment(mem, entry);
 	return ret;
 }
 
@@ -1862,8 +1861,10 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu(
 		mem->va + bo_size * (1 + mem->aql_queue));
 
 	/* Remove from VM internal data structures */
-	list_for_each_entry_safe(entry, tmp, &mem->attachments, list)
+	list_for_each_entry_safe(entry, tmp, &mem->attachments, list) {
+		kfd_mem_dmaunmap_attachment(mem, entry);
 		kfd_mem_detach(entry);
+	}
 
 	ret = unreserve_bo_and_vms(&ctx, false, false);
 
@@ -2037,6 +2038,37 @@ out:
 	return ret;
 }
 
+int amdgpu_amdkfd_gpuvm_dmaunmap_mem(struct kgd_mem *mem, void *drm_priv)
+{
+	struct kfd_mem_attachment *entry;
+	struct amdgpu_vm *vm;
+	int ret;
+
+	vm = drm_priv_to_vm(drm_priv);
+
+	mutex_lock(&mem->lock);
+
+	ret = amdgpu_bo_reserve(mem->bo, true);
+	if (ret)
+		goto out;
+
+	list_for_each_entry(entry, &mem->attachments, list) {
+		if (entry->bo_va->base.vm != vm)
+			continue;
+		if (entry->bo_va->base.bo->tbo.ttm &&
+		    !entry->bo_va->base.bo->tbo.ttm->sg)
+			continue;
+
+		kfd_mem_dmaunmap_attachment(mem, entry);
+	}
+
+	amdgpu_bo_unreserve(mem->bo);
+out:
+	mutex_unlock(&mem->lock);
+
+	return ret;
+}
+
 int amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(
 		struct amdgpu_device *adev, struct kgd_mem *mem, void *drm_priv)
 {

sys/dev/pci/drm/amd/amdgpu/amdgpu_cs.c

@@ -1059,6 +1059,9 @@ static int amdgpu_cs_patch_ibs(struct amdgpu_cs_parser *p,
 			r = amdgpu_ring_parse_cs(ring, p, job, ib);
 			if (r)
 				return r;
+
+			if (ib->sa_bo)
+				ib->gpu_addr =  amdgpu_sa_bo_gpu_addr(ib->sa_bo);
 		} else {
 			ib->ptr = (uint32_t *)kptr;
 			r = amdgpu_ring_patch_cs_in_place(ring, p, job, ib);

sys/dev/pci/drm/amd/amdgpu/amdgpu_ctx.c

@@ -684,16 +684,24 @@ int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,
 
 	switch (args->in.op) {
 	case AMDGPU_CTX_OP_ALLOC_CTX:
+		if (args->in.flags)
+			return -EINVAL;
 		r = amdgpu_ctx_alloc(adev, fpriv, filp, priority, &id);
 		args->out.alloc.ctx_id = id;
 		break;
 	case AMDGPU_CTX_OP_FREE_CTX:
+		if (args->in.flags)
+			return -EINVAL;
 		r = amdgpu_ctx_free(fpriv, id);
 		break;
 	case AMDGPU_CTX_OP_QUERY_STATE:
+		if (args->in.flags)
+			return -EINVAL;
 		r = amdgpu_ctx_query(adev, fpriv, id, &args->out);
 		break;
 	case AMDGPU_CTX_OP_QUERY_STATE2:
+		if (args->in.flags)
+			return -EINVAL;
 		r = amdgpu_ctx_query2(adev, fpriv, id, &args->out);
 		break;
 	case AMDGPU_CTX_OP_GET_STABLE_PSTATE:

sys/dev/pci/drm/amd/amdgpu/amdgpu_psp_ta.c

@@ -166,6 +166,9 @@ static ssize_t ta_if_load_debugfs_write(struct file *fp, const char *buf, size_t
 	if (ret)
 		return -EFAULT;
 
+	if (ta_bin_len > PSP_1_MEG)
+		return -EINVAL;
+
 	copy_pos += sizeof(uint32_t);
 
 	ta_bin = kzalloc(ta_bin_len, GFP_KERNEL);

sys/dev/pci/drm/amd/amdgpu/amdgpu_vcn.c

@@ -135,6 +135,10 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 		}
 	}
 
+	/* from vcn4 and above, only unified queue is used */
+	adev->vcn.using_unified_queue =
+		adev->ip_versions[UVD_HWIP][0] >= IP_VERSION(4, 0, 0);
+
 	hdr = (const struct common_firmware_header *)adev->vcn.fw->data;
 	adev->vcn.fw_version = le32_to_cpu(hdr->ucode_version);
 
@@ -259,18 +263,6 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev)
 	return 0;
 }
 
-/* from vcn4 and above, only unified queue is used */
-static bool amdgpu_vcn_using_unified_queue(struct amdgpu_ring *ring)
-{
-	struct amdgpu_device *adev = ring->adev;
-	bool ret = false;
-
-	if (adev->ip_versions[UVD_HWIP][0] >= IP_VERSION(4, 0, 0))
-		ret = true;
-
-	return ret;
-}
-
 bool amdgpu_vcn_is_disabled_vcn(struct amdgpu_device *adev, enum vcn_ring_type type, uint32_t vcn_instance)
 {
 	bool ret = false;
@@ -380,7 +372,9 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 		for (i = 0; i < adev->vcn.num_enc_rings; ++i)
 			fence[j] += amdgpu_fence_count_emitted(&adev->vcn.inst[j].ring_enc[i]);
 
-		if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)	{
+		/* Only set DPG pause for VCN3 or below, VCN4 and above will be handled by FW */
+		if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG &&
+		    !adev->vcn.using_unified_queue) {
 			struct dpg_pause_state new_state;
 
 			if (fence[j] ||
@@ -426,7 +420,9 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring)
 	amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
 	       AMD_PG_STATE_UNGATE);
 
-	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG)	{
+	/* Only set DPG pause for VCN3 or below, VCN4 and above will be handled by FW */
+	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG &&
+	    !adev->vcn.using_unified_queue) {
 		struct dpg_pause_state new_state;
 
 		if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC) {
@@ -452,8 +448,12 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring)
 
 void amdgpu_vcn_ring_end_use(struct amdgpu_ring *ring)
 {
+	struct amdgpu_device *adev = ring->adev;
+
+	/* Only set DPG pause for VCN3 or below, VCN4 and above will be handled by FW */
 	if (ring->adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG &&
-		ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC)
+	    ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC &&
+	    !adev->vcn.using_unified_queue)
 		atomic_dec(&ring->adev->vcn.inst[ring->me].dpg_enc_submission_cnt);
 
 	atomic_dec(&ring->adev->vcn.total_submission_cnt);
@@ -707,12 +707,11 @@ static int amdgpu_vcn_dec_sw_send_msg(struct amdgpu_ring *ring,
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
 	uint64_t addr = AMDGPU_GPU_PAGE_ALIGN(ib_msg->gpu_addr);
-	bool sq = amdgpu_vcn_using_unified_queue(ring);
 	uint32_t *ib_checksum;
 	uint32_t ib_pack_in_dw;
 	int i, r;
 
-	if (sq)
+	if (adev->vcn.using_unified_queue)
 		ib_size_dw += 8;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL,
@@ -725,7 +724,7 @@ static int amdgpu_vcn_dec_sw_send_msg(struct amdgpu_ring *ring,
 	ib->length_dw = 0;
 
 	/* single queue headers */
-	if (sq) {
+	if (adev->vcn.using_unified_queue) {
 		ib_pack_in_dw = sizeof(struct amdgpu_vcn_decode_buffer) / sizeof(uint32_t)
 						+ 4 + 2; /* engine info + decoding ib in dw */
 		ib_checksum = amdgpu_vcn_unified_ring_ib_header(ib, ib_pack_in_dw, false);
@@ -744,7 +743,7 @@ static int amdgpu_vcn_dec_sw_send_msg(struct amdgpu_ring *ring,
 	for (i = ib->length_dw; i < ib_size_dw; ++i)
 		ib->ptr[i] = 0x0;
 
-	if (sq)
+	if (adev->vcn.using_unified_queue)
 		amdgpu_vcn_unified_ring_ib_checksum(&ib_checksum, ib_pack_in_dw);
 
 	r = amdgpu_job_submit_direct(job, ring, &f);
@@ -834,15 +833,15 @@ static int amdgpu_vcn_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t hand
 					 struct dma_fence **fence)
 {
 	unsigned int ib_size_dw = 16;
+	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
 	struct dma_fence *f = NULL;
 	uint32_t *ib_checksum = NULL;
 	uint64_t addr;
-	bool sq = amdgpu_vcn_using_unified_queue(ring);
 	int i, r;
 
-	if (sq)
+	if (adev->vcn.using_unified_queue)
 		ib_size_dw += 8;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL,
@@ -856,7 +855,7 @@ static int amdgpu_vcn_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t hand
 
 	ib->length_dw = 0;
 
-	if (sq)
+	if (adev->vcn.using_unified_queue)
 		ib_checksum = amdgpu_vcn_unified_ring_ib_header(ib, 0x11, true);
 
 	ib->ptr[ib->length_dw++] = 0x00000018;
@@ -878,7 +877,7 @@ static int amdgpu_vcn_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t hand
 	for (i = ib->length_dw; i < ib_size_dw; ++i)
 		ib->ptr[i] = 0x0;
 
-	if (sq)
+	if (adev->vcn.using_unified_queue)
 		amdgpu_vcn_unified_ring_ib_checksum(&ib_checksum, 0x11);
 
 	r = amdgpu_job_submit_direct(job, ring, &f);
@@ -901,15 +900,15 @@ static int amdgpu_vcn_enc_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han
 					  struct dma_fence **fence)
 {
 	unsigned int ib_size_dw = 16;
+	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_job *job;
 	struct amdgpu_ib *ib;
 	struct dma_fence *f = NULL;
 	uint32_t *ib_checksum = NULL;
 	uint64_t addr;
-	bool sq = amdgpu_vcn_using_unified_queue(ring);
 	int i, r;
 
-	if (sq)
+	if (adev->vcn.using_unified_queue)
 		ib_size_dw += 8;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL,
@@ -923,7 +922,7 @@ static int amdgpu_vcn_enc_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han
 
 	ib->length_dw = 0;
 
-	if (sq)
+	if (adev->vcn.using_unified_queue)
 		ib_checksum = amdgpu_vcn_unified_ring_ib_header(ib, 0x11, true);
 
 	ib->ptr[ib->length_dw++] = 0x00000018;
@@ -945,7 +944,7 @@ static int amdgpu_vcn_enc_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han
 	for (i = ib->length_dw; i < ib_size_dw; ++i)
 		ib->ptr[i] = 0x0;
 
-	if (sq)
+	if (adev->vcn.using_unified_queue)
 		amdgpu_vcn_unified_ring_ib_checksum(&ib_checksum, 0x11);
 
 	r = amdgpu_job_submit_direct(job, ring, &f);

sys/dev/pci/drm/amd/amdgpu/amdgpu_vcn.h

@@ -284,6 +284,7 @@ struct amdgpu_vcn {
 
 	uint16_t inst_mask;
 	uint8_t	num_inst_per_aid;
+	bool using_unified_queue;
 };
 
 struct amdgpu_fw_shared_rb_ptrs_struct {

sys/dev/pci/drm/amd/amdgpu/amdgpu_vm_pt.c

@@ -766,11 +766,15 @@ int amdgpu_vm_pde_update(struct amdgpu_vm_update_params *params,
 			 struct amdgpu_vm_bo_base *entry)
 {
 	struct amdgpu_vm_bo_base *parent = amdgpu_vm_pt_parent(entry);
-	struct amdgpu_bo *bo = parent->bo, *pbo;
+	struct amdgpu_bo *bo, *pbo;
 	struct amdgpu_vm *vm = params->vm;
 	uint64_t pde, pt, flags;
 	unsigned int level;
 
+	if (WARN_ON(!parent))
+		return -EINVAL;
+
+	bo = parent->bo;
 	for (level = 0, pbo = bo->parent; pbo; ++level)
 		pbo = pbo->parent;
 

sys/dev/pci/drm/amd/amdgpu/gfx_v10_0.c

@@ -7892,22 +7892,15 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 static void gfx_v10_0_update_spm_vmid_internal(struct amdgpu_device *adev,
 					       unsigned int vmid)
 {
-	u32 reg, data;
+	u32 data;
 
 	/* not for *_SOC15 */
-	reg = SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_MC_CNTL);
-	if (amdgpu_sriov_is_pp_one_vf(adev))
-		data = RREG32_NO_KIQ(reg);
-	else
-		data = RREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL);
+	data = RREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL);
 
 	data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK;
 	data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT;
 
-	if (amdgpu_sriov_is_pp_one_vf(adev))
-		WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data);
-	else
-		WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data);
+	WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data);
 }
 
 static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, unsigned int vmid)

sys/dev/pci/drm/amd/amdgpu/gfx_v11_0.c

@@ -4961,23 +4961,16 @@ static int gfx_v11_0_update_gfx_clock_gating(struct amdgpu_device *adev,
 
 static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid)
 {
-	u32 reg, data;
+	u32 data;
 
 	amdgpu_gfx_off_ctrl(adev, false);
 
-	reg = SOC15_REG_OFFSET(GC, 0, regRLC_SPM_MC_CNTL);
-	if (amdgpu_sriov_is_pp_one_vf(adev))
-		data = RREG32_NO_KIQ(reg);
-	else
-		data = RREG32(reg);
+	data = RREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL);
 
 	data &= ~RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK;
 	data |= (vmid & RLC_SPM_MC_CNTL__RLC_SPM_VMID_MASK) << RLC_SPM_MC_CNTL__RLC_SPM_VMID__SHIFT;
 
-	if (amdgpu_sriov_is_pp_one_vf(adev))
-		WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data);
-	else
-		WREG32_SOC15(GC, 0, regRLC_SPM_MC_CNTL, data);
+	WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data);
 
 	amdgpu_gfx_off_ctrl(adev, true);
 }

sys/dev/pci/drm/amd/amdgpu/imu_v11_0.c

@@ -39,7 +39,7 @@ MODULE_FIRMWARE("amdgpu/gc_11_0_4_imu.bin");
 
 static int imu_v11_0_init_microcode(struct amdgpu_device *adev)
 {
-	char fw_name[40];
+	char fw_name[45];
 	char ucode_prefix[30];
 	int err;
 	const struct imu_firmware_header_v1_0 *imu_hdr;

sys/dev/pci/drm/amd/amdgpu/jpeg_v2_0.c

@@ -543,11 +543,11 @@ void jpeg_v2_0_dec_ring_emit_ib(struct amdgpu_ring *ring,
 
 	amdgpu_ring_write(ring, PACKETJ(mmUVD_LMI_JRBC_IB_VMID_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, (vmid | (vmid << 4)));
+	amdgpu_ring_write(ring, (vmid | (vmid << 4) | (vmid << 8)));
 
 	amdgpu_ring_write(ring, PACKETJ(mmUVD_LMI_JPEG_VMID_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, (vmid | (vmid << 4)));
+	amdgpu_ring_write(ring, (vmid | (vmid << 4) | (vmid << 8)));
 
 	amdgpu_ring_write(ring,	PACKETJ(mmUVD_LMI_JRBC_IB_64BIT_BAR_LOW_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));

sys/dev/pci/drm/amd/amdgpu/jpeg_v4_0_3.c

@@ -23,6 +23,7 @@
 
 #include "amdgpu.h"
 #include "amdgpu_jpeg.h"
+#include "amdgpu_cs.h"
 #include "soc15.h"
 #include "soc15d.h"
 #include "jpeg_v4_0_3.h"
@@ -769,11 +770,15 @@ static void jpeg_v4_0_3_dec_ring_emit_ib(struct amdgpu_ring *ring,
 
 	amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JRBC_IB_VMID_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, (vmid | (vmid << 4)));
+
+	if (ring->funcs->parse_cs)
+		amdgpu_ring_write(ring, 0);
+	else
+		amdgpu_ring_write(ring, (vmid | (vmid << 4) | (vmid << 8)));
 
 	amdgpu_ring_write(ring, PACKETJ(regUVD_LMI_JPEG_VMID_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
-	amdgpu_ring_write(ring, (vmid | (vmid << 4)));
+	amdgpu_ring_write(ring, (vmid | (vmid << 4) | (vmid << 8)));
 
 	amdgpu_ring_write(ring,	PACKETJ(regUVD_LMI_JRBC_IB_64BIT_BAR_LOW_INTERNAL_OFFSET,
 		0, 0, PACKETJ_TYPE0));
@@ -1052,6 +1057,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v4_0_3_dec_ring_get_rptr,
 	.get_wptr = jpeg_v4_0_3_dec_ring_get_wptr,
 	.set_wptr = jpeg_v4_0_3_dec_ring_set_wptr,
+	.parse_cs = jpeg_v4_0_3_dec_ring_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
@@ -1216,3 +1222,56 @@ static void jpeg_v4_0_3_set_ras_funcs(struct amdgpu_device *adev)
 {
 	adev->jpeg.ras = &jpeg_v4_0_3_ras;
 }
+
+/**
+ * jpeg_v4_0_3_dec_ring_parse_cs - command submission parser
+ *
+ * @parser: Command submission parser context
+ * @job: the job to parse
+ * @ib: the IB to parse
+ *
+ * Parse the command stream, return -EINVAL for invalid packet,
+ * 0 otherwise
+ */
+int jpeg_v4_0_3_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
+			     struct amdgpu_job *job,
+			     struct amdgpu_ib *ib)
+{
+	uint32_t i, reg, res, cond, type;
+	struct amdgpu_device *adev = parser->adev;
+
+	for (i = 0; i < ib->length_dw ; i += 2) {
+		reg  = CP_PACKETJ_GET_REG(ib->ptr[i]);
+		res  = CP_PACKETJ_GET_RES(ib->ptr[i]);
+		cond = CP_PACKETJ_GET_COND(ib->ptr[i]);
+		type = CP_PACKETJ_GET_TYPE(ib->ptr[i]);
+
+		if (res) /* only support 0 at the moment */
+			return -EINVAL;
+
+		switch (type) {
+		case PACKETJ_TYPE0:
+			if (cond != PACKETJ_CONDITION_CHECK0 || reg < JPEG_REG_RANGE_START || reg > JPEG_REG_RANGE_END) {
+				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+				return -EINVAL;
+			}
+			break;
+		case PACKETJ_TYPE3:
+			if (cond != PACKETJ_CONDITION_CHECK3 || reg < JPEG_REG_RANGE_START || reg > JPEG_REG_RANGE_END) {
+				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+				return -EINVAL;
+			}
+			break;
+		case PACKETJ_TYPE6:
+			if (ib->ptr[i] == CP_PACKETJ_NOP)
+				continue;
+			dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+			return -EINVAL;
+		default:
+			dev_err(adev->dev, "Unknown packet type %d !\n", type);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}

sys/dev/pci/drm/amd/amdgpu/jpeg_v4_0_3.h

@@ -46,6 +46,12 @@
 
 #define JRBC_DEC_EXTERNAL_REG_WRITE_ADDR				0x18000
 
+#define JPEG_REG_RANGE_START						0x4000
+#define JPEG_REG_RANGE_END						0x41c2
+
 extern const struct amdgpu_ip_block_version jpeg_v4_0_3_ip_block;
 
+int jpeg_v4_0_3_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
+				  struct amdgpu_job *job,
+				  struct amdgpu_ib *ib);
 #endif /* __JPEG_V4_0_3_H__ */

sys/dev/pci/drm/amd/amdgpu/soc15d.h

@@ -76,6 +76,12 @@
 			 ((cond & 0xF) << 24) |				\
 			 ((type & 0xF) << 28))
 
+#define CP_PACKETJ_NOP		0x60000000
+#define CP_PACKETJ_GET_REG(x)  ((x) & 0x3FFFF)
+#define CP_PACKETJ_GET_RES(x)  (((x) >> 18) & 0x3F)
+#define CP_PACKETJ_GET_COND(x) (((x) >> 24) & 0xF)
+#define CP_PACKETJ_GET_TYPE(x) (((x) >> 28) & 0xF)
+
 /* Packet 3 types */
 #define	PACKET3_NOP					0x10
 #define	PACKET3_SET_BASE				0x11

sys/dev/pci/drm/amd/amdkfd/kfd_chardev.c

@@ -1432,17 +1432,23 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep,
 			goto sync_memory_failed;
 		}
 	}
+
+	/* Flush TLBs after waiting for the page table updates to complete */
+	for (i = 0; i < args->n_devices; i++) {
+		peer_pdd = kfd_process_device_data_by_id(p, devices_arr[i]);
+		if (WARN_ON_ONCE(!peer_pdd))
+			continue;
+		if (flush_tlb)
+			kfd_flush_tlb(peer_pdd, TLB_FLUSH_HEAVYWEIGHT);
+
+		/* Remove dma mapping after tlb flush to avoid IO_PAGE_FAULT */
+		err = amdgpu_amdkfd_gpuvm_dmaunmap_mem(mem, peer_pdd->drm_priv);
+		if (err)
+			goto sync_memory_failed;
+	}
+
 	mutex_unlock(&p->mutex);
 
-	if (flush_tlb) {
-		/* Flush TLBs after waiting for the page table updates to complete */
-		for (i = 0; i < args->n_devices; i++) {
-			peer_pdd = kfd_process_device_data_by_id(p, devices_arr[i]);
-			if (WARN_ON_ONCE(!peer_pdd))
-				continue;
-			kfd_flush_tlb(peer_pdd, TLB_FLUSH_HEAVYWEIGHT);
-		}
-	}
 	kfree(devices_arr);
 
 	return 0;

sys/dev/pci/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c

@@ -3521,7 +3521,7 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 						(int)hubp->curs_attr.width || pos_cpy.x
 						<= (int)hubp->curs_attr.width +
 						pipe_ctx->plane_state->src_rect.x) {
-						pos_cpy.x = temp_x + viewport_width;
+						pos_cpy.x = 2 * viewport_width - temp_x;
 					}
 				}
 			} else {
@@ -3614,7 +3614,7 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 						(int)hubp->curs_attr.width || pos_cpy.x
 						<= (int)hubp->curs_attr.width +
 						pipe_ctx->plane_state->src_rect.x) {
-						pos_cpy.x = 2 * viewport_width - temp_x;
+						pos_cpy.x = temp_x + viewport_width;
 					}
 				}
 			} else {

sys/dev/pci/drm/amd/display/dc/dcn321/dcn321_resource.c

@@ -1756,6 +1756,9 @@ static bool dcn321_resource_construct(
 	dc->caps.color.mpc.ogam_rom_caps.hlg = 0;
 	dc->caps.color.mpc.ocsc = 1;
 
+	/* Use pipe context based otg sync logic */
+	dc->config.use_pipe_ctx_sync_logic = true;
+
 	dc->config.dc_mode_clk_limit_support = true;
 	/* read VBIOS LTTPR caps */
 	{

sys/dev/pci/drm/amd/pm/amdgpu_pm.c

@@ -1473,9 +1473,9 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev,
 	return -EINVAL;
 }
 
-static unsigned int amdgpu_hwmon_get_sensor_generic(struct amdgpu_device *adev,
-						    enum amd_pp_sensors sensor,
-						    void *query)
+static int amdgpu_hwmon_get_sensor_generic(struct amdgpu_device *adev,
+					   enum amd_pp_sensors sensor,
+					   void *query)
 {
 	int r, size = sizeof(uint32_t);
 
@@ -2789,8 +2789,8 @@ static ssize_t amdgpu_hwmon_show_vddnb_label(struct device *dev,
 	return sysfs_emit(buf, "vddnb\n");
 }
 
-static unsigned int amdgpu_hwmon_get_power(struct device *dev,
-					   enum amd_pp_sensors sensor)
+static int amdgpu_hwmon_get_power(struct device *dev,
+				  enum amd_pp_sensors sensor)
 {
 	struct amdgpu_device *adev = dev_get_drvdata(dev);
 	unsigned int uw;
@@ -2811,7 +2811,7 @@ static ssize_t amdgpu_hwmon_show_power_avg(struct device *dev,
 					   struct device_attribute *attr,
 					   char *buf)
 {
-	unsigned int val;
+	int val;
 
 	val = amdgpu_hwmon_get_power(dev, AMDGPU_PP_SENSOR_GPU_AVG_POWER);
 	if (val < 0)
@@ -2824,7 +2824,7 @@ static ssize_t amdgpu_hwmon_show_power_input(struct device *dev,
 					     struct device_attribute *attr,
 					     char *buf)
 {
-	unsigned int val;
+	int val;
 
 	val = amdgpu_hwmon_get_power(dev, AMDGPU_PP_SENSOR_GPU_INPUT_POWER);
 	if (val < 0)