Merge tag 'amd-drm-next-6.2-2022-11-04' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.2-2022-11-04:

amdgpu:
- Add TMZ support for GC 11.0.1
- More IP version check conversions
- Mode2 reset fixes for sienna cichlid
- SMU 13.x fixes
- RAS enablement on MP 13.x
- Replace kmap with kmap_local_page()
- Misc Clang warning fixes
- SR-IOV fixes for GC 11.x
- PCI AER fix
- DCN 3.2.x commit sequence rework
- SDMA 4.x doorbell fix
- Expose additional new GC 11.x firmware versions
- Misc code cleanups
- S0i3 fixes
- More DC FPU cleanup
- Add more DC kerneldoc
- Misc spelling and grammer fixes
- DCN 3.1.x fixes
- Plane modifier fix
- MCA RAS enablement
- Secure display locking fix
- RAS TA rework
- RAS EEPROM fixes
- Fail suspend if eviction fails
- Drop AMD specific DSC workarounds in favor of drm EDID quirks
- SR-IOV suspend/resume fixes
- Enable DCN support for ARM
- Enable secure display on DCN 2.1

amdkfd:
- Cache size fixes for GC 10.3.x
- kfd_dev struct cleanup
- GC11.x CWSR trap handler fix
- Userptr fixes
- Warning fixes

radeon:
- Replace kmap with kmap_local_page()

UAPI:
- Expose additional new GC 11.x firmware versions via the existing INFO query

drm:
- Add some new EDID DSC quirks

Signed-off-by: Dave Airlie <airlied@redhat.com>

# Conflicts:
#	drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221104205827.6008-1-alexander.deucher@amd.com

Author: airlied

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

@@ -219,10 +219,12 @@ extern int amdgpu_use_xgmi_p2p;
 extern int sched_policy;
 extern bool debug_evictions;
 extern bool no_system_mem_limit;
+extern int halt_if_hws_hang;
 #else
 static const int __maybe_unused sched_policy = KFD_SCHED_POLICY_HWS;
 static const bool __maybe_unused debug_evictions; /* = false */
 static const bool __maybe_unused no_system_mem_limit;
+static const int __maybe_unused halt_if_hws_hang;
 #endif
 #ifdef CONFIG_HSA_AMD_P2P
 extern bool pcie_p2p;
@@ -675,7 +677,7 @@ enum amd_hw_ip_block_type {
 	MAX_HWIP
 };
 
-#define HWIP_MAX_INSTANCE	11
+#define HWIP_MAX_INSTANCE	28
 
 #define HW_ID_MAX		300
 #define IP_VERSION(mj, mn, rv) (((mj) << 16) | ((mn) << 8) | (rv))

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

@@ -195,7 +195,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
 		}
 
 		adev->kfd.init_complete = kgd2kfd_device_init(adev->kfd.dev,
-						adev_to_drm(adev), &gpu_resources);
+							&gpu_resources);
 
 		amdgpu_amdkfd_total_mem_size += adev->gmc.real_vram_size;
 
@@ -706,6 +706,13 @@ int amdgpu_amdkfd_submit_ib(struct amdgpu_device *adev,
 
 void amdgpu_amdkfd_set_compute_idle(struct amdgpu_device *adev, bool idle)
 {
+	/* Temporary workaround to fix issues observed in some
+	 * compute applications when GFXOFF is enabled on GFX11.
+	 */
+	if (IP_VERSION_MAJ(adev->ip_versions[GC_HWIP][0]) == 11) {
+		pr_debug("GFXOFF is %s\n", idle ? "enabled" : "disabled");
+		amdgpu_gfx_off_ctrl(adev, idle);
+	}
 	amdgpu_dpm_switch_power_profile(adev,
 					PP_SMC_POWER_PROFILE_COMPUTE,
 					!idle);
@@ -753,9 +760,7 @@ bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev)
 
 void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bool reset)
 {
-	struct ras_err_data err_data = {0, 0, 0, NULL};
-
-	amdgpu_umc_poison_handler(adev, &err_data, reset);
+	amdgpu_umc_poison_handler(adev, reset);
 }
 
 bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev)

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

@@ -353,7 +353,6 @@ int kgd2kfd_init(void);
 void kgd2kfd_exit(void);
 struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf);
 bool kgd2kfd_device_init(struct kfd_dev *kfd,
-			 struct drm_device *ddev,
 			 const struct kgd2kfd_shared_resources *gpu_resources);
 void kgd2kfd_device_exit(struct kfd_dev *kfd);
 void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm);
@@ -381,7 +380,7 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
 }
 
 static inline
-bool kgd2kfd_device_init(struct kfd_dev *kfd, struct drm_device *ddev,
+bool kgd2kfd_device_init(struct kfd_dev *kfd,
 				const struct kgd2kfd_shared_resources *gpu_resources)
 {
 	return false;

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c

@@ -787,7 +787,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 	for (se_idx = 0; se_idx < se_cnt; se_idx++) {
 		for (sh_idx = 0; sh_idx < sh_cnt; sh_idx++) {
 
-			gfx_v9_0_select_se_sh(adev, se_idx, sh_idx, 0xffffffff);
+			amdgpu_gfx_select_se_sh(adev, se_idx, sh_idx, 0xffffffff);
 			queue_map = RREG32_SOC15(GC, 0, mmSPI_CSQ_WF_ACTIVE_STATUS);
 
 			/*
@@ -820,7 +820,7 @@ void kgd_gfx_v9_get_cu_occupancy(struct amdgpu_device *adev, int pasid,
 		}
 	}
 
-	gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
+	amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
 	soc15_grbm_select(adev, 0, 0, 0, 0);
 	unlock_spi_csq_mutexes(adev);
 

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

@@ -418,9 +418,9 @@ static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem)
 	if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE)
 		mapping_flags |= AMDGPU_VM_PAGE_EXECUTABLE;
 
-	switch (adev->asic_type) {
-	case CHIP_ARCTURUS:
-	case CHIP_ALDEBARAN:
+	switch (adev->ip_versions[GC_HWIP][0]) {
+	case IP_VERSION(9, 4, 1):
+	case IP_VERSION(9, 4, 2):
 		if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) {
 			if (bo_adev == adev) {
 				if (uncached)
@@ -429,7 +429,7 @@ static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem)
 					mapping_flags |= AMDGPU_VM_MTYPE_CC;
 				else
 					mapping_flags |= AMDGPU_VM_MTYPE_RW;
-				if (adev->asic_type == CHIP_ALDEBARAN &&
+				if ((adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 2)) &&
 				    adev->gmc.xgmi.connected_to_cpu)
 					snoop = true;
 			} else {
@@ -510,13 +510,13 @@ kfd_mem_dmamap_userptr(struct kgd_mem *mem,
 	struct ttm_tt *ttm = bo->tbo.ttm;
 	int ret;
 
+	if (WARN_ON(ttm->num_pages != src_ttm->num_pages))
+		return -EINVAL;
+
 	ttm->sg = kmalloc(sizeof(*ttm->sg), GFP_KERNEL);
 	if (unlikely(!ttm->sg))
 		return -ENOMEM;
 
-	if (WARN_ON(ttm->num_pages != src_ttm->num_pages))
-		return -EINVAL;
-
 	/* Same sequence as in amdgpu_ttm_tt_pin_userptr */
 	ret = sg_alloc_table_from_pages(ttm->sg, src_ttm->pages,
 					ttm->num_pages, 0,
@@ -1907,16 +1907,6 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu(
 	 */
 	mutex_lock(&mem->process_info->lock);
 
-	/* Lock mmap-sem. If we find an invalid userptr BO, we can be
-	 * sure that the MMU notifier is no longer running
-	 * concurrently and the queues are actually stopped
-	 */
-	if (amdgpu_ttm_tt_get_usermm(bo->tbo.ttm)) {
-		mmap_write_lock(current->mm);
-		is_invalid_userptr = atomic_read(&mem->invalid);
-		mmap_write_unlock(current->mm);
-	}
-
 	mutex_lock(&mem->lock);
 
 	domain = mem->domain;

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

@@ -326,7 +326,10 @@ static int amdgpu_ctx_init(struct amdgpu_ctx_mgr *mgr, int32_t priority,
 	if (r)
 		return r;
 
-	ctx->stable_pstate = current_stable_pstate;
+	if (mgr->adev->pm.stable_pstate_ctx)
+		ctx->stable_pstate = mgr->adev->pm.stable_pstate_ctx->stable_pstate;
+	else
+		ctx->stable_pstate = current_stable_pstate;
 
 	return 0;
 }

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

@@ -1568,7 +1568,7 @@ static int amdgpu_device_check_arguments(struct amdgpu_device *adev)
  * @pdev: pci dev pointer
  * @state: vga_switcheroo state
  *
- * Callback for the switcheroo driver.  Suspends or resumes the
+ * Callback for the switcheroo driver.  Suspends or resumes
  * the asics before or after it is powered up using ACPI methods.
  */
 static void amdgpu_switcheroo_set_state(struct pci_dev *pdev,
@@ -2397,7 +2397,7 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 			adev->ip_blocks[i].status.hw = true;
 
 			/* right after GMC hw init, we create CSA */
-			if (amdgpu_mcbp || amdgpu_sriov_vf(adev)) {
+			if (amdgpu_mcbp) {
 				r = amdgpu_allocate_static_csa(adev, &adev->virt.csa_obj,
 								AMDGPU_GEM_DOMAIN_VRAM,
 								AMDGPU_CSA_SIZE);
@@ -3210,6 +3210,15 @@ static int amdgpu_device_ip_resume_phase2(struct amdgpu_device *adev)
 			return r;
 		}
 		adev->ip_blocks[i].status.hw = true;
+
+		if (adev->in_s0ix && adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) {
+			/* disable gfxoff for IP resume. The gfxoff will be re-enabled in
+			 * amdgpu_device_resume() after IP resume.
+			 */
+			amdgpu_gfx_off_ctrl(adev, false);
+			DRM_DEBUG("will disable gfxoff for re-initializing other blocks\n");
+		}
+
 	}
 
 	return 0;
@@ -4051,15 +4060,18 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev)
  * at suspend time.
  *
  */
-static void amdgpu_device_evict_resources(struct amdgpu_device *adev)
+static int amdgpu_device_evict_resources(struct amdgpu_device *adev)
 {
+	int ret;
+
 	/* No need to evict vram on APUs for suspend to ram or s2idle */
 	if ((adev->in_s3 || adev->in_s0ix) && (adev->flags & AMD_IS_APU))
-		return;
+		return 0;
 
-	if (amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM))
+	ret = amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM);
+	if (ret)
 		DRM_WARN("evicting device resources failed\n");
-
+	return ret;
 }
 
 /*
@@ -4109,7 +4121,9 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
 	if (!adev->in_s0ix)
 		amdgpu_amdkfd_suspend(adev, adev->in_runpm);
 
-	amdgpu_device_evict_resources(adev);
+	r = amdgpu_device_evict_resources(adev);
+	if (r)
+		return r;
 
 	amdgpu_fence_driver_hw_fini(adev);
 
@@ -4183,8 +4197,17 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
 	}
 
 	/* Make sure IB tests flushed */
+	if (amdgpu_sriov_vf(adev))
+		amdgpu_irq_gpu_reset_resume_helper(adev);
 	flush_delayed_work(&adev->delayed_init_work);
 
+	if (adev->in_s0ix) {
+		/* re-enable gfxoff after IP resume. This re-enables gfxoff after
+		 * it was disabled for IP resume in amdgpu_device_ip_resume_phase2().
+		 */
+		amdgpu_gfx_off_ctrl(adev, true);
+		DRM_DEBUG("will enable gfxoff for the mission mode\n");
+	}
 	if (fbcon)
 		drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, false);
 
@@ -5381,7 +5404,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 			drm_sched_start(&ring->sched, !tmp_adev->asic_reset_res);
 		}
 
-		if (adev->enable_mes)
+		if (adev->enable_mes && adev->ip_versions[GC_HWIP][0] != IP_VERSION(11, 0, 3))
 			amdgpu_mes_self_test(tmp_adev);
 
 		if (!drm_drv_uses_atomic_modeset(adev_to_drm(tmp_adev)) && !job_signaled) {

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

@@ -2201,7 +2201,8 @@ amdgpu_pci_remove(struct pci_dev *pdev)
 		pm_runtime_forbid(dev->dev);
 	}
 
-	if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2)) {
+	if (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2) &&
+	    !amdgpu_sriov_vf(adev)) {
 		bool need_to_reset_gpu = false;
 
 		if (adev->gmc.xgmi.num_physical_nodes > 1) {

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

@@ -542,6 +542,7 @@ void amdgpu_gmc_tmz_set(struct amdgpu_device *adev)
 	case IP_VERSION(10, 3, 1):
 	/* YELLOW_CARP*/
 	case IP_VERSION(10, 3, 3):
+	case IP_VERSION(11, 0, 1):
 		/* Don't enable it by default yet.
 		 */
 		if (amdgpu_tmz < 1) {

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

@@ -337,11 +337,17 @@ static int amdgpu_firmware_info(struct drm_amdgpu_info_firmware *fw_info,
 		fw_info->feature = adev->psp.cap_feature_version;
 		break;
 	case AMDGPU_INFO_FW_MES_KIQ:
-		fw_info->ver = adev->mes.ucode_fw_version[0];
-		fw_info->feature = 0;
+		fw_info->ver = adev->mes.kiq_version & AMDGPU_MES_VERSION_MASK;
+		fw_info->feature = (adev->mes.kiq_version & AMDGPU_MES_FEAT_VERSION_MASK)
+					>> AMDGPU_MES_FEAT_VERSION_SHIFT;
 		break;
 	case AMDGPU_INFO_FW_MES:
-		fw_info->ver = adev->mes.ucode_fw_version[1];
+		fw_info->ver = adev->mes.sched_version & AMDGPU_MES_VERSION_MASK;
+		fw_info->feature = (adev->mes.sched_version & AMDGPU_MES_FEAT_VERSION_MASK)
+					>> AMDGPU_MES_FEAT_VERSION_SHIFT;
+		break;
+	case AMDGPU_INFO_FW_IMU:
+		fw_info->ver = adev->gfx.imu_fw_version;
 		fw_info->feature = 0;
 		break;
 	default:
@@ -790,7 +796,7 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		dev_info->ids_flags = 0;
 		if (adev->flags & AMD_IS_APU)
 			dev_info->ids_flags |= AMDGPU_IDS_FLAGS_FUSION;
-		if (amdgpu_mcbp || amdgpu_sriov_vf(adev))
+		if (amdgpu_mcbp)
 			dev_info->ids_flags |= AMDGPU_IDS_FLAGS_PREEMPTION;
 		if (amdgpu_is_tmz(adev))
 			dev_info->ids_flags |= AMDGPU_IDS_FLAGS_TMZ;
@@ -1166,7 +1172,7 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 		goto error_vm;
 	}
 
-	if (amdgpu_mcbp || amdgpu_sriov_vf(adev)) {
+	if (amdgpu_mcbp) {
 		uint64_t csa_addr = amdgpu_csa_vaddr(adev) & AMDGPU_GMC_HOLE_MASK;
 
 		r = amdgpu_map_static_csa(adev, &fpriv->vm, adev->virt.csa_obj,
@@ -1230,7 +1236,7 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev,
 	if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_VCE) != NULL)
 		amdgpu_vce_free_handles(adev, file_priv);
 
-	if (amdgpu_mcbp || amdgpu_sriov_vf(adev)) {
+	if (amdgpu_mcbp) {
 		/* TODO: how to handle reserve failure */
 		BUG_ON(amdgpu_bo_reserve(adev->virt.csa_obj, true));
 		amdgpu_vm_bo_del(adev, fpriv->csa_va);
@@ -1520,6 +1526,15 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused)
 			   fw_info.feature, fw_info.ver);
 	}
 
+	/* IMU */
+	query_fw.fw_type = AMDGPU_INFO_FW_IMU;
+	query_fw.index = 0;
+	ret = amdgpu_firmware_info(&fw_info, &query_fw, adev);
+	if (ret)
+		return ret;
+	seq_printf(m, "IMU feature version: %u, firmware version: 0x%08x\n",
+		   fw_info.feature, fw_info.ver);
+
 	/* PSP SOS */
 	query_fw.fw_type = AMDGPU_INFO_FW_SOS;
 	ret = amdgpu_firmware_info(&fw_info, &query_fw, adev);