Merge v5.6-rc2 into drm-misc-next

Lyude needs some patches in 5.6-rc2 and we didn't bring drm-misc-next
forward yet, so it looks like a good occasion.

Signed-off-by: Maxime Ripard <maxime@cerno.tech>

107 files changed, 6184 insertions, 2399 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 7ae3b22c5628..c2bbcdd9c875 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -120,6 +120,7 @@ amdgpu-y += \
 	amdgpu_rlc.o \
 	gfx_v8_0.o \
 	gfx_v9_0.o \
+	gfx_v9_4.o \
 	gfx_v10_0.o
 
 # add async DMA block
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 9aff914608b8..2f9da6d182f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -90,6 +90,7 @@
 #include "amdgpu_mes.h"
 #include "amdgpu_umc.h"
 #include "amdgpu_mmhub.h"
+#include "amdgpu_df.h"
 
 #define MAX_GPU_INSTANCE		16
 
@@ -636,9 +637,8 @@ struct amdgpu_fw_vram_usage {
 	struct amdgpu_bo *reserved_bo;
 	void *va;
 
-	/* Offset on the top of VRAM, used as c2p write buffer.
+	/* GDDR6 training support flag.
 	*/
-	u64 mem_train_fb_loc;
 	bool mem_train_support;
 };
 
@@ -665,29 +665,6 @@ struct amdgpu_mmio_remap {
 	resource_size_t bus_addr;
 };
 
-struct amdgpu_df_funcs {
-	void (*sw_init)(struct amdgpu_device *adev);
-	void (*sw_fini)(struct amdgpu_device *adev);
-	void (*enable_broadcast_mode)(struct amdgpu_device *adev,
-				      bool enable);
-	u32 (*get_fb_channel_number)(struct amdgpu_device *adev);
-	u32 (*get_hbm_channel_number)(struct amdgpu_device *adev);
-	void (*update_medium_grain_clock_gating)(struct amdgpu_device *adev,
-						 bool enable);
-	void (*get_clockgating_state)(struct amdgpu_device *adev,
-				      u32 *flags);
-	void (*enable_ecc_force_par_wr_rmw)(struct amdgpu_device *adev,
-					    bool enable);
-	int (*pmc_start)(struct amdgpu_device *adev, uint64_t config,
-					 int is_enable);
-	int (*pmc_stop)(struct amdgpu_device *adev, uint64_t config,
-					 int is_disable);
-	void (*pmc_get_count)(struct amdgpu_device *adev, uint64_t config,
-					 uint64_t *count);
-	uint64_t (*get_fica)(struct amdgpu_device *adev, uint32_t ficaa_val);
-	void (*set_fica)(struct amdgpu_device *adev, uint32_t ficaa_val,
-			 uint32_t ficadl_val, uint32_t ficadh_val);
-};
 /* Define the HW IP blocks will be used in driver , add more if necessary */
 enum amd_hw_ip_block_type {
 	GC_HWIP = 1,
@@ -931,6 +908,9 @@ struct amdgpu_device {
 	bool                            enable_mes;
 	struct amdgpu_mes               mes;
 
+	/* df */
+	struct amdgpu_df                df;
+
 	struct amdgpu_ip_block          ip_blocks[AMDGPU_MAX_IP_NUM];
 	int				num_ip_blocks;
 	struct mutex	mn_lock;
@@ -944,8 +924,6 @@ struct amdgpu_device {
 	/* soc15 register offset based on ip, instance and  segment */
 	uint32_t 		*reg_offset[MAX_HWIP][HWIP_MAX_INSTANCE];
 
-	const struct amdgpu_df_funcs	*df_funcs;
-
 	/* delayed work_func for deferring clockgating during resume */
 	struct delayed_work     delayed_init_work;
 
@@ -994,8 +972,6 @@ struct amdgpu_device {
 
 	bool                            pm_sysfs_en;
 	bool                            ucode_sysfs_en;
-
-	bool				in_baco;
 };
 
 static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_bo_device *bdev)
@@ -1033,10 +1009,14 @@ int emu_soc_asic_init(struct amdgpu_device *adev);
 
 #define AMDGPU_REGS_IDX       (1<<0)
 #define AMDGPU_REGS_NO_KIQ    (1<<1)
+#define AMDGPU_REGS_KIQ       (1<<2)
 
 #define RREG32_NO_KIQ(reg) amdgpu_mm_rreg(adev, (reg), AMDGPU_REGS_NO_KIQ)
 #define WREG32_NO_KIQ(reg, v) amdgpu_mm_wreg(adev, (reg), (v), AMDGPU_REGS_NO_KIQ)
 
+#define RREG32_KIQ(reg) amdgpu_mm_rreg(adev, (reg), AMDGPU_REGS_KIQ)
+#define WREG32_KIQ(reg, v) amdgpu_mm_wreg(adev, (reg), (v), AMDGPU_REGS_KIQ)
+
 #define RREG8(reg) amdgpu_mm_rreg8(adev, (reg))
 #define WREG8(reg, v) amdgpu_mm_wreg8(adev, (reg), (v))
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
index 82155ac3288a..12247a32f9ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
@@ -527,7 +527,7 @@ static int acp_set_powergating_state(void *handle,
 				     enum amd_powergating_state state)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	bool enable = state == AMD_PG_STATE_GATE ? true : false;
+	bool enable = (state == AMD_PG_STATE_GATE);
 
 	if (adev->powerplay.pp_funcs &&
 		adev->powerplay.pp_funcs->set_powergating_by_smu)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index d3da9dde4ee1..8609287620ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -613,15 +613,9 @@ void amdgpu_amdkfd_set_compute_idle(struct kgd_dev *kgd, bool idle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
 
-	if (is_support_sw_smu(adev))
-		smu_switch_power_profile(&adev->smu,
-					 PP_SMC_POWER_PROFILE_COMPUTE,
-					 !idle);
-	else if (adev->powerplay.pp_funcs &&
-		 adev->powerplay.pp_funcs->switch_power_profile)
-		amdgpu_dpm_switch_power_profile(adev,
-						PP_SMC_POWER_PROFILE_COMPUTE,
-						!idle);
+	amdgpu_dpm_switch_power_profile(adev,
+					PP_SMC_POWER_PROFILE_COMPUTE,
+					!idle);
 }
 
 bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid)
@@ -634,6 +628,38 @@ bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid)
 	return false;
 }
 
+int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct kgd_dev *kgd, uint16_t vmid)
+{
+	struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
+
+	if (adev->family == AMDGPU_FAMILY_AI) {
+		int i;
+
+		for (i = 0; i < adev->num_vmhubs; i++)
+			amdgpu_gmc_flush_gpu_tlb(adev, vmid, i, 0);
+	} else {
+		amdgpu_gmc_flush_gpu_tlb(adev, vmid, AMDGPU_GFXHUB_0, 0);