drm/amdgpu: use VRAM|GTT for a bunch of kernel allocations

Technically all of those can use GTT as well, no need to force things
into VRAM.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NLuben Tuikov <luben.tuikov@amd.com>
Acked-by: NFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

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

@@ -755,6 +755,11 @@ struct amdgpu_mqd {
 #define AMDGPU_PRODUCT_NAME_LEN 64
 struct amdgpu_reset_domain;
 
+/*
+ * Non-zero (true) if the GPU has VRAM. Zero (false) otherwise.
+ */
+#define AMDGPU_HAS_VRAM(_adev) ((_adev)->gmc.real_vram_size)
+
 struct amdgpu_device {
 	struct device			*dev;
 	struct pci_dev			*pdev;

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

@@ -934,7 +934,8 @@ static int amdgpu_device_asic_init(struct amdgpu_device *adev)
 static int amdgpu_device_vram_scratch_init(struct amdgpu_device *adev)
 {
 	return amdgpu_bo_create_kernel(adev, AMDGPU_GPU_PAGE_SIZE,
-				       PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+				       PAGE_SIZE,
+				       AMDGPU_GEM_DOMAIN_VRAM,
 				       &adev->vram_scratch.robj,
 				       &adev->vram_scratch.gpu_addr,
 				       (void **)&adev->vram_scratch.ptr);
@@ -2410,7 +2411,8 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 			/* right after GMC hw init, we create CSA */
 			if (amdgpu_mcbp) {
 				r = amdgpu_allocate_static_csa(adev, &adev->virt.csa_obj,
-								AMDGPU_GEM_DOMAIN_VRAM,
+							       AMDGPU_GEM_DOMAIN_VRAM |
+							       AMDGPU_GEM_DOMAIN_GTT,
 							       AMDGPU_CSA_SIZE);
 				if (r) {
 					DRM_ERROR("allocate CSA failed %d\n", r);

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

@@ -372,8 +372,11 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev,
 		 * KIQ MQD no matter SRIOV or Bare-metal
 		 */
 		r = amdgpu_bo_create_kernel(adev, mqd_size, PAGE_SIZE,
-					    AMDGPU_GEM_DOMAIN_VRAM, &ring->mqd_obj,
-					    &ring->mqd_gpu_addr, &ring->mqd_ptr);
+					    AMDGPU_GEM_DOMAIN_VRAM |
+					    AMDGPU_GEM_DOMAIN_GTT,
+					    &ring->mqd_obj,
+					    &ring->mqd_gpu_addr,
+					    &ring->mqd_ptr);
 		if (r) {
 			dev_warn(adev->dev, "failed to create ring mqd ob (%d)", r);
 			return r;

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

@@ -66,7 +66,8 @@ static int psp_ring_init(struct psp_context *psp,
 	/* allocate 4k Page of Local Frame Buffer memory for ring */
 	ring->ring_size = 0x1000;
 	ret = amdgpu_bo_create_kernel(adev, ring->ring_size, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->firmware.rbuf,
 				      &ring->ring_mem_mc_addr,
 				      (void **)&ring->ring_mem);
@@ -797,9 +798,13 @@ static int psp_tmr_init(struct psp_context *psp)
 
 	if (!psp->tmr_bo) {
 		pptr = amdgpu_sriov_vf(psp->adev) ? &tmr_buf : NULL;
-		ret = amdgpu_bo_create_kernel(psp->adev, tmr_size, PSP_TMR_ALIGNMENT,
-					      AMDGPU_GEM_DOMAIN_VRAM,
-					      &psp->tmr_bo, &psp->tmr_mc_addr, pptr);
+		ret = amdgpu_bo_create_kernel(psp->adev, tmr_size,
+					      PSP_TMR_ALIGNMENT,
+					      AMDGPU_HAS_VRAM(psp->adev) ?
+					      AMDGPU_GEM_DOMAIN_VRAM :
+					      AMDGPU_GEM_DOMAIN_GTT,
+					      &psp->tmr_bo, &psp->tmr_mc_addr,
+					      pptr);
 	}
 
 	return ret;
@@ -1092,7 +1097,8 @@ int psp_ta_init_shared_buf(struct psp_context *psp,
 	* physical) for ta to host memory
 	*/
 	return amdgpu_bo_create_kernel(psp->adev, mem_ctx->shared_mem_size,
-				      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+				      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &mem_ctx->shared_bo,
 				      &mem_ctx->shared_mc_addr,
 				      &mem_ctx->shared_buf);
@@ -3444,9 +3450,9 @@ static ssize_t psp_usbc_pd_fw_sysfs_write(struct device *dev,
 
 	/* LFB address which is aligned to 1MB boundary per PSP request */
 	ret = amdgpu_bo_create_kernel(adev, usbc_pd_fw->size, 0x100000,
-						AMDGPU_GEM_DOMAIN_VRAM,
-						&fw_buf_bo,
-						&fw_pri_mc_addr,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
+				      &fw_buf_bo, &fw_pri_mc_addr,
 				      &fw_pri_cpu_addr);
 	if (ret)
 		goto rel_buf;

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

@@ -93,7 +93,8 @@ int amdgpu_gfx_rlc_init_sr(struct amdgpu_device *adev, u32 dws)
 
 	/* allocate save restore block */
 	r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.rlc.save_restore_obj,
 				      &adev->gfx.rlc.save_restore_gpu_addr,
 				      (void **)&adev->gfx.rlc.sr_ptr);
@@ -130,7 +131,8 @@ int amdgpu_gfx_rlc_init_csb(struct amdgpu_device *adev)
 	/* allocate clear state block */
 	adev->gfx.rlc.clear_state_size = dws = adev->gfx.rlc.funcs->get_csb_size(adev);
 	r = amdgpu_bo_create_kernel(adev, dws * 4, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.rlc.clear_state_obj,
 				      &adev->gfx.rlc.clear_state_gpu_addr,
 				      (void **)&adev->gfx.rlc.cs_ptr);
@@ -156,7 +158,8 @@ int amdgpu_gfx_rlc_init_cpt(struct amdgpu_device *adev)
 	int r;
 
 	r = amdgpu_bo_create_reserved(adev, adev->gfx.rlc.cp_table_size,
-				      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+				      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.rlc.cp_table_obj,
 				      &adev->gfx.rlc.cp_table_gpu_addr,
 				      (void **)&adev->gfx.rlc.cp_table_ptr);

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

@@ -331,8 +331,11 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev)
 		if (adev->uvd.harvest_config & (1 << j))
 			continue;
 		r = amdgpu_bo_create_kernel(adev, bo_size, PAGE_SIZE,
-					    AMDGPU_GEM_DOMAIN_VRAM, &adev->uvd.inst[j].vcpu_bo,
-					    &adev->uvd.inst[j].gpu_addr, &adev->uvd.inst[j].cpu_addr);
+					    AMDGPU_GEM_DOMAIN_VRAM |
+					    AMDGPU_GEM_DOMAIN_GTT,
+					    &adev->uvd.inst[j].vcpu_bo,
+					    &adev->uvd.inst[j].gpu_addr,
+					    &adev->uvd.inst[j].cpu_addr);
 		if (r) {
 			dev_err(adev->dev, "(%d) failed to allocate UVD bo\n", r);
 			return r;

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

@@ -186,7 +186,9 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size)
 				(binary_id << 8));
 
 	r = amdgpu_bo_create_kernel(adev, size, PAGE_SIZE,
-				    AMDGPU_GEM_DOMAIN_VRAM, &adev->vce.vcpu_bo,
+				    AMDGPU_GEM_DOMAIN_VRAM |
+				    AMDGPU_GEM_DOMAIN_GTT,
+				    &adev->vce.vcpu_bo,
 				    &adev->vce.gpu_addr, &adev->vce.cpu_addr);
 	if (r) {
 		dev_err(adev->dev, "(%d) failed to allocate VCE bo\n", r);

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

@@ -274,8 +274,11 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 			continue;
 
 		r = amdgpu_bo_create_kernel(adev, bo_size, PAGE_SIZE,
-						AMDGPU_GEM_DOMAIN_VRAM, &adev->vcn.inst[i].vcpu_bo,
-						&adev->vcn.inst[i].gpu_addr, &adev->vcn.inst[i].cpu_addr);
+					    AMDGPU_GEM_DOMAIN_VRAM |
+					    AMDGPU_GEM_DOMAIN_GTT,
+					    &adev->vcn.inst[i].vcpu_bo,
+					    &adev->vcn.inst[i].gpu_addr,
+					    &adev->vcn.inst[i].cpu_addr);
 		if (r) {
 			dev_err(adev->dev, "(%d) failed to allocate vcn bo\n", r);
 			return r;
@@ -296,8 +299,11 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev)
 
 		if (adev->vcn.indirect_sram) {
 			r = amdgpu_bo_create_kernel(adev, 64 * 2 * 4, PAGE_SIZE,
-					AMDGPU_GEM_DOMAIN_VRAM, &adev->vcn.inst[i].dpg_sram_bo,
-					&adev->vcn.inst[i].dpg_sram_gpu_addr, &adev->vcn.inst[i].dpg_sram_cpu_addr);
+					AMDGPU_GEM_DOMAIN_VRAM |
+					AMDGPU_GEM_DOMAIN_GTT,
+					&adev->vcn.inst[i].dpg_sram_bo,
+					&adev->vcn.inst[i].dpg_sram_gpu_addr,
+					&adev->vcn.inst[i].dpg_sram_cpu_addr);
 			if (r) {
 				dev_err(adev->dev, "VCN %d (%d) failed to allocate DPG bo\n", i, r);
 				return r;

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

@@ -232,7 +232,8 @@ int amdgpu_virt_alloc_mm_table(struct amdgpu_device *adev)
 		return 0;
 
 	r = amdgpu_bo_create_kernel(adev, PAGE_SIZE, PAGE_SIZE,
-				    AMDGPU_GEM_DOMAIN_VRAM,
+				    AMDGPU_GEM_DOMAIN_VRAM |
+				    AMDGPU_GEM_DOMAIN_GTT,
 				    &adev->virt.mm_table.bo,
 				    &adev->virt.mm_table.gpu_addr,
 				    (void *)&adev->virt.mm_table.cpu_addr);

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

@@ -987,7 +987,8 @@ static int gfx_v11_0_rlc_autoload_buffer_init(struct amdgpu_device *adev)
 	total_size = gfx_v11_0_calc_toc_total_size(adev);
 
 	r = amdgpu_bo_create_reserved(adev, total_size, 64 * 1024,
-			AMDGPU_GEM_DOMAIN_VRAM,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.rlc.rlc_autoload_bo,
 				      &adev->gfx.rlc.rlc_autoload_gpu_addr,
 				      (void **)&adev->gfx.rlc.rlc_autoload_ptr);
@@ -2649,7 +2650,9 @@ static int gfx_v11_0_cp_gfx_load_pfp_microcode_rs64(struct amdgpu_device *adev)
 
 	/* 64kb align */
 	r = amdgpu_bo_create_reserved(adev, fw_ucode_size,
-				      64 * 1024, AMDGPU_GEM_DOMAIN_VRAM,
+				      64 * 1024,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.pfp.pfp_fw_obj,
 				      &adev->gfx.pfp.pfp_fw_gpu_addr,
 				      (void **)&adev->gfx.pfp.pfp_fw_ptr);
@@ -2660,7 +2663,9 @@ static int gfx_v11_0_cp_gfx_load_pfp_microcode_rs64(struct amdgpu_device *adev)
 	}
 
 	r = amdgpu_bo_create_reserved(adev, fw_data_size,
-				      64 * 1024, AMDGPU_GEM_DOMAIN_VRAM,
+				      64 * 1024,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.pfp.pfp_fw_data_obj,
 				      &adev->gfx.pfp.pfp_fw_data_gpu_addr,
 				      (void **)&adev->gfx.pfp.pfp_fw_data_ptr);
@@ -2863,7 +2868,9 @@ static int gfx_v11_0_cp_gfx_load_me_microcode_rs64(struct amdgpu_device *adev)
 
 	/* 64kb align*/
 	r = amdgpu_bo_create_reserved(adev, fw_ucode_size,
-				      64 * 1024, AMDGPU_GEM_DOMAIN_VRAM,
+				      64 * 1024,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.me.me_fw_obj,
 				      &adev->gfx.me.me_fw_gpu_addr,
 				      (void **)&adev->gfx.me.me_fw_ptr);
@@ -2874,7 +2881,9 @@ static int gfx_v11_0_cp_gfx_load_me_microcode_rs64(struct amdgpu_device *adev)
 	}
 
 	r = amdgpu_bo_create_reserved(adev, fw_data_size,
-				      64 * 1024, AMDGPU_GEM_DOMAIN_VRAM,
+				      64 * 1024,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.me.me_fw_data_obj,
 				      &adev->gfx.me.me_fw_data_gpu_addr,
 				      (void **)&adev->gfx.me.me_fw_data_ptr);
@@ -3380,7 +3389,9 @@ static int gfx_v11_0_cp_compute_load_microcode_rs64(struct amdgpu_device *adev)
 	fw_data_size = le32_to_cpu(mec_hdr->data_size_bytes);
 
 	r = amdgpu_bo_create_reserved(adev, fw_ucode_size,
-				      64 * 1024, AMDGPU_GEM_DOMAIN_VRAM,
+				      64 * 1024,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.mec.mec_fw_obj,
 				      &adev->gfx.mec.mec_fw_gpu_addr,
 				      (void **)&fw_ucode_ptr);
@@ -3391,7 +3402,9 @@ static int gfx_v11_0_cp_compute_load_microcode_rs64(struct amdgpu_device *adev)
 	}
 
 	r = amdgpu_bo_create_reserved(adev, fw_data_size,
-				      64 * 1024, AMDGPU_GEM_DOMAIN_VRAM,
+				      64 * 1024,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.mec.mec_fw_data_obj,
 				      &adev->gfx.mec.mec_fw_data_gpu_addr,
 				      (void **)&fw_data_ptr);

drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c

@@ -2375,7 +2375,8 @@ static int gfx_v6_0_rlc_init(struct amdgpu_device *adev)
 		dws = adev->gfx.rlc.clear_state_size + (256 / 4);
 
 		r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
-					      AMDGPU_GEM_DOMAIN_VRAM,
+					      AMDGPU_GEM_DOMAIN_VRAM |
+					      AMDGPU_GEM_DOMAIN_GTT,
 					      &adev->gfx.rlc.clear_state_obj,
 					      &adev->gfx.rlc.clear_state_gpu_addr,
 					      (void **)&adev->gfx.rlc.cs_ptr);

drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c

@@ -2772,7 +2772,8 @@ static int gfx_v7_0_mec_init(struct amdgpu_device *adev)
 		* GFX7_MEC_HPD_SIZE * 2;
 
 	r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
-				      AMDGPU_GEM_DOMAIN_VRAM,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->gfx.mec.hpd_eop_obj,
 				      &adev->gfx.mec.hpd_eop_gpu_addr,
 				      (void **)&hpd);

drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c

@@ -1340,7 +1340,8 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev)
 	mec_hpd_size = adev->gfx.num_compute_rings * GFX8_MEC_HPD_SIZE;
 	if (mec_hpd_size) {
 		r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
-					      AMDGPU_GEM_DOMAIN_VRAM,
+					      AMDGPU_GEM_DOMAIN_VRAM |
+					      AMDGPU_GEM_DOMAIN_GTT,
 					      &adev->gfx.mec.hpd_eop_obj,
 					      &adev->gfx.mec.hpd_eop_gpu_addr,
 					      (void **)&hpd);

drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c

@@ -1783,7 +1783,8 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev)
 	mec_hpd_size = adev->gfx.num_compute_rings * GFX9_MEC_HPD_SIZE;
 	if (mec_hpd_size) {
 		r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
-					      AMDGPU_GEM_DOMAIN_VRAM,
+					      AMDGPU_GEM_DOMAIN_VRAM |
+					      AMDGPU_GEM_DOMAIN_GTT,
 					      &adev->gfx.mec.hpd_eop_obj,
 					      &adev->gfx.mec.hpd_eop_gpu_addr,
 					      (void **)&hpd);

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

@@ -549,7 +549,9 @@ static int mes_v11_0_allocate_ucode_buffer(struct amdgpu_device *adev,
 	fw_size = le32_to_cpu(mes_hdr->mes_ucode_size_bytes);
 
 	r = amdgpu_bo_create_reserved(adev, fw_size,
-				      PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+				      PAGE_SIZE,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->mes.ucode_fw_obj[pipe],
 				      &adev->mes.ucode_fw_gpu_addr[pipe],
 				      (void **)&adev->mes.ucode_fw_ptr[pipe]);
@@ -582,7 +584,9 @@ static int mes_v11_0_allocate_ucode_data_buffer(struct amdgpu_device *adev,
 	fw_size = le32_to_cpu(mes_hdr->mes_ucode_data_size_bytes);
 
 	r = amdgpu_bo_create_reserved(adev, fw_size,
-				      64 * 1024, AMDGPU_GEM_DOMAIN_VRAM,
+				      64 * 1024,
+				      AMDGPU_GEM_DOMAIN_VRAM |
+				      AMDGPU_GEM_DOMAIN_GTT,
 				      &adev->mes.data_fw_obj[pipe],
 				      &adev->mes.data_fw_gpu_addr[pipe],
 				      (void **)&adev->mes.data_fw_ptr[pipe]);

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

@@ -2085,7 +2085,9 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev)
 	 * TODO: Move this into GART.
 	 */
 	r = amdgpu_bo_create_kernel(adev, region_info.fb_size, PAGE_SIZE,
-				    AMDGPU_GEM_DOMAIN_VRAM, &adev->dm.dmub_bo,
+				    AMDGPU_GEM_DOMAIN_VRAM |
+				    AMDGPU_GEM_DOMAIN_GTT,
+				    &adev->dm.dmub_bo,
 				    &adev->dm.dmub_bo_gpu_addr,
 				    &adev->dm.dmub_bo_cpu_addr);
 	if (r)

drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.c

@@ -250,9 +250,8 @@ static int smu10_smu_init(struct pp_hwmgr *hwmgr)
 
 	/* allocate space for watermarks table */
 	r = amdgpu_bo_create_kernel((struct amdgpu_device *)hwmgr->adev,
-			sizeof(Watermarks_t),
-			PAGE_SIZE,
-			AMDGPU_GEM_DOMAIN_VRAM,
+			sizeof(Watermarks_t), PAGE_SIZE,
+			AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT,
 			&priv->smu_tables.entry[SMU10_WMTABLE].handle,
 			&priv->smu_tables.entry[SMU10_WMTABLE].mc_addr,
 			&priv->smu_tables.entry[SMU10_WMTABLE].table);
@@ -266,9 +265,8 @@ static int smu10_smu_init(struct pp_hwmgr *hwmgr)
 
 	/* allocate space for watermarks table */
 	r = amdgpu_bo_create_kernel((struct amdgpu_device *)hwmgr->adev,
-			sizeof(DpmClocks_t),
-			PAGE_SIZE,
-			AMDGPU_GEM_DOMAIN_VRAM,
+			sizeof(DpmClocks_t), PAGE_SIZE,
+			AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT,
 			&priv->smu_tables.entry[SMU10_CLOCKTABLE].handle,
 			&priv->smu_tables.entry[SMU10_CLOCKTABLE].mc_addr,
 			&priv->smu_tables.entry[SMU10_CLOCKTABLE].table);