drm/amdgpu: Enable VCN for Beige Goby

Enabled VCN support for Beige Goby chip

Signed-off-by: Veerabadhran Gopalakrishnan <veerabadhran.gopalakrishnan@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

3 files changed, 68 insertions, 45 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 201645963ba5..d101cb9697a2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -48,6 +48,7 @@
 #define FIRMWARE_VANGOGH	"amdgpu/vangogh_vcn.bin"
 #define FIRMWARE_DIMGREY_CAVEFISH	"amdgpu/dimgrey_cavefish_vcn.bin"
 #define FIRMWARE_ALDEBARAN	"amdgpu/aldebaran_vcn.bin"
+#define FIRMWARE_BEIGE_GOBY	"amdgpu/beige_goby_vcn.bin"
 
 MODULE_FIRMWARE(FIRMWARE_RAVEN);
 MODULE_FIRMWARE(FIRMWARE_PICASSO);
@@ -63,6 +64,7 @@ MODULE_FIRMWARE(FIRMWARE_SIENNA_CICHLID);
 MODULE_FIRMWARE(FIRMWARE_NAVY_FLOUNDER);
 MODULE_FIRMWARE(FIRMWARE_VANGOGH);
 MODULE_FIRMWARE(FIRMWARE_DIMGREY_CAVEFISH);
+MODULE_FIRMWARE(FIRMWARE_BEIGE_GOBY);
 
 static void amdgpu_vcn_idle_work_handler(struct work_struct *work);
 
@@ -151,6 +153,12 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 		    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
 			adev->vcn.indirect_sram = true;
 		break;
+	case CHIP_BEIGE_GOBY:
+		fw_name = FIRMWARE_BEIGE_GOBY;
+		if ((adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) &&
+		    (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG))
+			adev->vcn.indirect_sram = true;
+		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
index c4bec8ad8a14..81e76d3d70b9 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -967,6 +967,7 @@ int nv_set_ip_blocks(struct amdgpu_device *adev)
 		if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT &&
 		    is_support_sw_smu(adev))
 			amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
+		amdgpu_device_ip_block_add(adev, &vcn_v3_0_ip_block);
 		break;
 	default:
 		return -EINVAL;
@@ -1277,7 +1278,8 @@ static int nv_common_early_init(void *handle)
 		break;
 	case CHIP_BEIGE_GOBY:
 		adev->cg_flags = 0;
-		adev->pg_flags = 0;
+		adev->pg_flags = AMD_PG_SUPPORT_VCN |
+			AMD_PG_SUPPORT_VCN_DPG;
 		adev->external_rev_id = adev->rev_id + 0x46;
 		break;
 	default:
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index cf165ab5dd26..779e58593f74 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -91,6 +91,11 @@ static int vcn_v3_0_early_init(void *handle)
 		adev->vcn.harvest_config = 0;
 		adev->vcn.num_enc_rings = 1;
 
+	if (adev->asic_type == CHIP_BEIGE_GOBY) {
+		adev->vcn.num_vcn_inst = 1;
+		adev->vcn.num_enc_rings = 0;
+	}
+
 	} else {
 		if (adev->asic_type == CHIP_SIENNA_CICHLID) {
 			u32 harvest;
@@ -110,7 +115,10 @@ static int vcn_v3_0_early_init(void *handle)
 		} else
 			adev->vcn.num_vcn_inst = 1;
 
-		adev->vcn.num_enc_rings = 2;
+		if (adev->asic_type == CHIP_BEIGE_GOBY)
+			adev->vcn.num_enc_rings = 0;
+		else
+			adev->vcn.num_enc_rings = 2;
 	}
 
 	vcn_v3_0_set_dec_ring_funcs(adev);
@@ -1261,23 +1269,25 @@ static int vcn_v3_0_start(struct amdgpu_device *adev)
 		fw_shared->rb.wptr = lower_32_bits(ring->wptr);
 		fw_shared->multi_queue.decode_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
 
-		fw_shared->multi_queue.encode_generalpurpose_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
-		ring = &adev->vcn.inst[i].ring_enc[0];
-		WREG32_SOC15(VCN, i, mmUVD_RB_RPTR, lower_32_bits(ring->wptr));
-		WREG32_SOC15(VCN, i, mmUVD_RB_WPTR, lower_32_bits(ring->wptr));
-		WREG32_SOC15(VCN, i, mmUVD_RB_BASE_LO, ring->gpu_addr);
-		WREG32_SOC15(VCN, i, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr));
-		WREG32_SOC15(VCN, i, mmUVD_RB_SIZE, ring->ring_size / 4);
-		fw_shared->multi_queue.encode_generalpurpose_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
-
-		fw_shared->multi_queue.encode_lowlatency_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
-		ring = &adev->vcn.inst[i].ring_enc[1];
-		WREG32_SOC15(VCN, i, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr));
-		WREG32_SOC15(VCN, i, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr));
-		WREG32_SOC15(VCN, i, mmUVD_RB_BASE_LO2, ring->gpu_addr);
-		WREG32_SOC15(VCN, i, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr));
-		WREG32_SOC15(VCN, i, mmUVD_RB_SIZE2, ring->ring_size / 4);
-		fw_shared->multi_queue.encode_lowlatency_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+		if (adev->asic_type != CHIP_BEIGE_GOBY) {
+			fw_shared->multi_queue.encode_generalpurpose_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
+			ring = &adev->vcn.inst[i].ring_enc[0];
+			WREG32_SOC15(VCN, i, mmUVD_RB_RPTR, lower_32_bits(ring->wptr));
+			WREG32_SOC15(VCN, i, mmUVD_RB_WPTR, lower_32_bits(ring->wptr));
+			WREG32_SOC15(VCN, i, mmUVD_RB_BASE_LO, ring->gpu_addr);
+			WREG32_SOC15(VCN, i, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr));
+			WREG32_SOC15(VCN, i, mmUVD_RB_SIZE, ring->ring_size / 4);
+			fw_shared->multi_queue.encode_generalpurpose_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+
+			fw_shared->multi_queue.encode_lowlatency_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
+			ring = &adev->vcn.inst[i].ring_enc[1];
+			WREG32_SOC15(VCN, i, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr));
+			WREG32_SOC15(VCN, i, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr));
+			WREG32_SOC15(VCN, i, mmUVD_RB_BASE_LO2, ring->gpu_addr);
+			WREG32_SOC15(VCN, i, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr));
+			WREG32_SOC15(VCN, i, mmUVD_RB_SIZE2, ring->ring_size / 4);
+			fw_shared->multi_queue.encode_lowlatency_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+		}
 	}
 
 	return 0;
@@ -1657,31 +1667,33 @@ static int vcn_v3_0_pause_dpg_mode(struct amdgpu_device *adev,
 					UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK,
 					~UVD_POWER_STATUS__STALL_DPG_POWER_UP_MASK);
 
-				/* Restore */
-				fw_shared = adev->vcn.inst[inst_idx].fw_shared_cpu_addr;
-				fw_shared->multi_queue.encode_generalpurpose_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
-				ring = &adev->vcn.inst[inst_idx].ring_enc[0];
-				ring->wptr = 0;
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_LO, ring->gpu_addr);
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr));
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_SIZE, ring->ring_size / 4);
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_RPTR, lower_32_bits(ring->wptr));
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_WPTR, lower_32_bits(ring->wptr));
-				fw_shared->multi_queue.encode_generalpurpose_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
-
-				fw_shared->multi_queue.encode_lowlatency_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
-				ring = &adev->vcn.inst[inst_idx].ring_enc[1];
-				ring->wptr = 0;
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_LO2, ring->gpu_addr);
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr));
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_SIZE2, ring->ring_size / 4);
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr));
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr));
-				fw_shared->multi_queue.encode_lowlatency_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
-
-				/* restore wptr/rptr with pointers saved in FW shared memory*/
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_RPTR, fw_shared->rb.rptr);
-				WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_WPTR, fw_shared->rb.wptr);
+				if (adev->asic_type != CHIP_BEIGE_GOBY) {
+					/* Restore */
+					fw_shared = adev->vcn.inst[inst_idx].fw_shared_cpu_addr;
+					fw_shared->multi_queue.encode_generalpurpose_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
+					ring = &adev->vcn.inst[inst_idx].ring_enc[0];
+					ring->wptr = 0;
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_LO, ring->gpu_addr);
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_HI, upper_32_bits(ring->gpu_addr));
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_SIZE, ring->ring_size / 4);
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_RPTR, lower_32_bits(ring->wptr));
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_WPTR, lower_32_bits(ring->wptr));
+					fw_shared->multi_queue.encode_generalpurpose_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+
+					fw_shared->multi_queue.encode_lowlatency_queue_mode |= cpu_to_le32(FW_QUEUE_RING_RESET);
+					ring = &adev->vcn.inst[inst_idx].ring_enc[1];
+					ring->wptr = 0;
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_LO2, ring->gpu_addr);
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_BASE_HI2, upper_32_bits(ring->gpu_addr));
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_SIZE2, ring->ring_size / 4);
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr));
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr));
+					fw_shared->multi_queue.encode_lowlatency_queue_mode &= cpu_to_le32(~FW_QUEUE_RING_RESET);
+
+					/* restore wptr/rptr with pointers saved in FW shared memory*/
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_RPTR, fw_shared->rb.rptr);
+					WREG32_SOC15(VCN, inst_idx, mmUVD_RBC_RB_WPTR, fw_shared->rb.wptr);
+				}
 
 				/* Unstall DPG */
 				WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS),
@@ -2138,7 +2150,8 @@ static void vcn_v3_0_set_enc_ring_funcs(struct amdgpu_device *adev)
 			adev->vcn.inst[i].ring_enc[j].funcs = &vcn_v3_0_enc_ring_vm_funcs;
 			adev->vcn.inst[i].ring_enc[j].me = i;
 		}
-		DRM_INFO("VCN(%d) encode is enabled in VM mode\n", i);
+		if (adev->vcn.num_enc_rings > 0)
+			DRM_INFO("VCN(%d) encode is enabled in VM mode\n", i);
 	}
 }