Merge tag 'drm-fixes-2023-03-10' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Weekly fixes.

  msm and amdgpu are the vast majority of these, otherwise some
  straggler misc from last week for nouveau and cirrus and a mailmap
  update for a drm developer.

  mailmap:
   - add an entry

  nouveau:
   - fix system shutdown regression
   - build warning fix

  cirrus:
   - NULL ptr deref fix

  msm:
   - fix invalid ptr free in syncobj cleanup
   - sync GMU removal in teardown
   - a5xx preemption fixes
   - fix runpm imbalance
   - DPU hw fixes
   - stack corruption fix
   - clear DSPP reservation

  amdgpu:
   - Misc display fixes
   - UMC 8.10 fixes
   - Driver unload fixes
   - NBIO 7.3.0 fix
   - Error checking fixes for soc15, nv, soc21 read register interface
   - Fix video cap query for VCN 4.0.4

  amdkfd:
   - Fix return check in doorbell handling"

* tag 'drm-fixes-2023-03-10' of git://anongit.freedesktop.org/drm/drm: (42 commits)
  drm/amdgpu/soc21: Add video cap query support for VCN_4_0_4
  drm/amdgpu: fix error checking in amdgpu_read_mm_registers for nv
  drm/amdgpu: fix error checking in amdgpu_read_mm_registers for soc21
  drm/amdgpu: fix error checking in amdgpu_read_mm_registers for soc15
  drm/amdgpu: Fix the warning info when removing amdgpu device
  drm/amdgpu: fix return value check in kfd
  drm/amd: Fix initialization mistake for NBIO 7.3.0
  drm/amdgpu: Fix call trace warning and hang when removing amdgpu device
  mailmap: add mailmap entries for Faith.
  drm/msm: DEVFREQ_GOV_SIMPLE_ONDEMAND is no longer needed
  drm/amd/display: Update clock table to include highest clock setting
  drm/amd/pm: Enable ecc_info table support for smu v13_0_10
  drm/amdgpu: Support umc node harvest config on umc v8_10
  drm/connector: print max_requested_bpc in state debugfs
  drm/display: Don't block HDR_OUTPUT_METADATA on unknown EOTF
  drm/msm/dpu: clear DSPP reservations in rm release
  drm/msm/disp/dpu: fix sc7280_pp base offset
  drm/msm/dpu: fix stack smashing in dpu_hw_ctl_setup_blendstage
  drm/msm/dpu: don't use DPU_CLK_CTRL_CURSORn for DMA SSPP clocks
  drm/msm/dpu: fix clocks settings for msm8998 SSPP blocks
  ...

Author: torvalds

.mailmap

@@ -136,6 +136,9 @@ Erik Kaneda <[email protected]> <[email protected]>
 Eugen Hristev <[email protected]> <[email protected]>
 Evgeniy Polyakov <[email protected]>
 Ezequiel Garcia <[email protected]> <[email protected]>
+Faith Ekstrand <[email protected]> <[email protected]>
+Faith Ekstrand <[email protected]> <[email protected]>
+Faith Ekstrand <[email protected]> <[email protected]>
 Felipe W Damasio <[email protected]>
 Felix Kuhling <[email protected]>
 Felix Moeller <[email protected]>

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

@@ -543,6 +543,7 @@ static void amdgpu_discovery_read_from_harvest_table(struct amdgpu_device *adev,
 	struct harvest_table *harvest_info;
 	u16 offset;
 	int i;
+	uint32_t umc_harvest_config = 0;
 
 	bhdr = (struct binary_header *)adev->mman.discovery_bin;
 	offset = le16_to_cpu(bhdr->table_list[HARVEST_INFO].offset);
@@ -570,12 +571,17 @@ static void amdgpu_discovery_read_from_harvest_table(struct amdgpu_device *adev,
 			adev->harvest_ip_mask |= AMD_HARVEST_IP_DMU_MASK;
 			break;
 		case UMC_HWID:
+			umc_harvest_config |=
+				1 << (le16_to_cpu(harvest_info->list[i].number_instance));
 			(*umc_harvest_count)++;
 			break;
 		default:
 			break;
 		}
 	}
+
+	adev->umc.active_mask = ((1 << adev->umc.node_inst_num) - 1) &
+				~umc_harvest_config;
 }
 
 /* ================================================== */
@@ -1156,8 +1162,10 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev)
 						AMDGPU_MAX_SDMA_INSTANCES);
 			}
 
-			if (le16_to_cpu(ip->hw_id) == UMC_HWID)
+			if (le16_to_cpu(ip->hw_id) == UMC_HWID) {
 				adev->gmc.num_umc++;
+				adev->umc.node_inst_num++;
+			}
 
 			for (k = 0; k < num_base_address; k++) {
 				/*

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

@@ -1315,7 +1315,7 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo)
 
 	if (!bo->resource || bo->resource->mem_type != TTM_PL_VRAM ||
 	    !(abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE) ||
-	    adev->in_suspend || adev->shutdown)
+	    adev->in_suspend || drm_dev_is_unplugged(adev_to_drm(adev)))
 		return;
 
 	if (WARN_ON_ONCE(!dma_resv_trylock(bo->base.resv)))

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

@@ -602,27 +602,14 @@ psp_cmd_submit_buf(struct psp_context *psp,
 		   struct psp_gfx_cmd_resp *cmd, uint64_t fence_mc_addr)
 {
 	int ret;
-	int index, idx;
+	int index;
 	int timeout = 20000;
 	bool ras_intr = false;
 	bool skip_unsupport = false;
-	bool dev_entered;
 
 	if (psp->adev->no_hw_access)
 		return 0;
 
-	dev_entered = drm_dev_enter(adev_to_drm(psp->adev), &idx);
-	/*
-	 * We allow sending PSP messages LOAD_ASD and UNLOAD_TA without acquiring
-	 * a lock in drm_dev_enter during driver unload because we must call
-	 * drm_dev_unplug as the beginning  of unload driver sequence . It is very
-	 * crucial that userspace can't access device instances anymore.
-	 */
-	if (!dev_entered)
-		WARN_ON(psp->cmd_buf_mem->cmd_id != GFX_CMD_ID_LOAD_ASD &&
-			psp->cmd_buf_mem->cmd_id != GFX_CMD_ID_UNLOAD_TA &&
-			psp->cmd_buf_mem->cmd_id != GFX_CMD_ID_INVOKE_CMD);
-
 	memset(psp->cmd_buf_mem, 0, PSP_CMD_BUFFER_SIZE);
 
 	memcpy(psp->cmd_buf_mem, cmd, sizeof(struct psp_gfx_cmd_resp));
@@ -686,8 +673,6 @@ psp_cmd_submit_buf(struct psp_context *psp,
 	}
 
 exit:
-	if (dev_entered)
-		drm_dev_exit(idx);
 	return ret;
 }
 

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

@@ -42,7 +42,7 @@
 #define LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) LOOP_UMC_INST((umc_inst)) LOOP_UMC_CH_INST((ch_inst))
 
 #define LOOP_UMC_NODE_INST(node_inst) \
-		for ((node_inst) = 0; (node_inst) < adev->umc.node_inst_num; (node_inst)++)
+		for_each_set_bit((node_inst), &(adev->umc.active_mask), adev->umc.node_inst_num)
 
 #define LOOP_UMC_EACH_NODE_INST_AND_CH(node_inst, umc_inst, ch_inst) \
 		LOOP_UMC_NODE_INST((node_inst)) LOOP_UMC_INST_AND_CH((umc_inst), (ch_inst))
@@ -69,7 +69,7 @@ struct amdgpu_umc {
 	/* number of umc instance with memory map register access */
 	uint32_t umc_inst_num;
 
-	/*number of umc node instance with memory map register access*/
+	/* Total number of umc node instance including harvest one */
 	uint32_t node_inst_num;
 
 	/* UMC regiser per channel offset */
@@ -82,6 +82,9 @@ struct amdgpu_umc {
 
 	const struct amdgpu_umc_funcs *funcs;
 	struct amdgpu_umc_ras *ras;
+
+	/* active mask for umc node instance */
+	unsigned long active_mask;
 };
 
 int amdgpu_umc_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block);

drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c

@@ -567,7 +567,6 @@ static void gmc_v11_0_set_umc_funcs(struct amdgpu_device *adev)
 	case IP_VERSION(8, 10, 0):
 		adev->umc.channel_inst_num = UMC_V8_10_CHANNEL_INSTANCE_NUM;
 		adev->umc.umc_inst_num = UMC_V8_10_UMC_INSTANCE_NUM;
-		adev->umc.node_inst_num = adev->gmc.num_umc;
 		adev->umc.max_ras_err_cnt_per_query = UMC_V8_10_TOTAL_CHANNEL_NUM(adev);
 		adev->umc.channel_offs = UMC_V8_10_PER_CHANNEL_OFFSET;
 		adev->umc.retire_unit = UMC_V8_10_NA_COL_2BITS_POWER_OF_2_NUM;

drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c

@@ -382,11 +382,6 @@ static void nbio_v7_2_init_registers(struct amdgpu_device *adev)
 		if (def != data)
 			WREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regBIF1_PCIE_MST_CTRL_3), data);
 		break;
-	case IP_VERSION(7, 5, 1):
-		data = RREG32_SOC15(NBIO, 0, regRCC_DEV2_EPF0_STRAP2);
-		data &= ~RCC_DEV2_EPF0_STRAP2__STRAP_NO_SOFT_RESET_DEV2_F0_MASK;
-		WREG32_SOC15(NBIO, 0, regRCC_DEV2_EPF0_STRAP2, data);
-		fallthrough;
 	default:
 		def = data = RREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regPCIE_CONFIG_CNTL));
 		data = REG_SET_FIELD(data, PCIE_CONFIG_CNTL,
@@ -399,6 +394,15 @@ static void nbio_v7_2_init_registers(struct amdgpu_device *adev)
 		break;
 	}
 
+	switch (adev->ip_versions[NBIO_HWIP][0]) {
+	case IP_VERSION(7, 3, 0):
+	case IP_VERSION(7, 5, 1):
+		data = RREG32_SOC15(NBIO, 0, regRCC_DEV2_EPF0_STRAP2);
+		data &= ~RCC_DEV2_EPF0_STRAP2__STRAP_NO_SOFT_RESET_DEV2_F0_MASK;
+		WREG32_SOC15(NBIO, 0, regRCC_DEV2_EPF0_STRAP2, data);
+		break;
+	}
+
 	if (amdgpu_sriov_vf(adev))
 		adev->rmmio_remap.reg_offset = SOC15_REG_OFFSET(NBIO, 0,
 			regBIF_BX_PF0_HDP_MEM_COHERENCY_FLUSH_CNTL) << 2;

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

@@ -444,9 +444,10 @@ static int nv_read_register(struct amdgpu_device *adev, u32 se_num,
 	*value = 0;
 	for (i = 0; i < ARRAY_SIZE(nv_allowed_read_registers); i++) {
 		en = &nv_allowed_read_registers[i];
-		if (adev->reg_offset[en->hwip][en->inst] &&
-		    reg_offset != (adev->reg_offset[en->hwip][en->inst][en->seg]
-				   + en->reg_offset))
+		if (!adev->reg_offset[en->hwip][en->inst])
+			continue;
+		else if (reg_offset != (adev->reg_offset[en->hwip][en->inst][en->seg]
+					+ en->reg_offset))
 			continue;
 
 		*value = nv_get_register_value(adev,

drivers/gpu/drm/amd/amdgpu/soc15.c

@@ -439,8 +439,9 @@ static int soc15_read_register(struct amdgpu_device *adev, u32 se_num,
 	*value = 0;
 	for (i = 0; i < ARRAY_SIZE(soc15_allowed_read_registers); i++) {
 		en = &soc15_allowed_read_registers[i];
-		if (adev->reg_offset[en->hwip][en->inst] &&
-			reg_offset != (adev->reg_offset[en->hwip][en->inst][en->seg]
+		if (!adev->reg_offset[en->hwip][en->inst])
+			continue;
+		else if (reg_offset != (adev->reg_offset[en->hwip][en->inst][en->seg]
 					+ en->reg_offset))
 			continue;
 

drivers/gpu/drm/amd/amdgpu/soc21.c

@@ -111,6 +111,7 @@ static int soc21_query_video_codecs(struct amdgpu_device *adev, bool encode,
 	switch (adev->ip_versions[UVD_HWIP][0]) {
 	case IP_VERSION(4, 0, 0):
 	case IP_VERSION(4, 0, 2):
+	case IP_VERSION(4, 0, 4):
 		if (adev->vcn.harvest_config & AMDGPU_VCN_HARVEST_VCN0) {
 			if (encode)
 				*codecs = &vcn_4_0_0_video_codecs_encode_vcn1;
@@ -291,9 +292,10 @@ static int soc21_read_register(struct amdgpu_device *adev, u32 se_num,
 	*value = 0;
 	for (i = 0; i < ARRAY_SIZE(soc21_allowed_read_registers); i++) {
 		en = &soc21_allowed_read_registers[i];
-		if (adev->reg_offset[en->hwip][en->inst] &&
-		    reg_offset != (adev->reg_offset[en->hwip][en->inst][en->seg]
-				   + en->reg_offset))
+		if (!adev->reg_offset[en->hwip][en->inst])
+			continue;
+		else if (reg_offset != (adev->reg_offset[en->hwip][en->inst][en->seg]
+					+ en->reg_offset))
 			continue;
 
 		*value = soc21_get_register_value(adev,