drm/amdkfd: drop IOMMUv2 support

Now that we use the dGPU path for all APUs, drop the
IOMMUv2 support.

v2: drop the now unused queue manager functions for gfx7/8 APUs

Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Tested-by: Mike Lothian <mike@fireburn.co.uk>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

24 files changed, 64 insertions, 1085 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 629ca1ad75a8..df633e9ce920 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -226,16 +226,6 @@ void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm)
 		kgd2kfd_suspend(adev->kfd.dev, run_pm);
 }
 
-int amdgpu_amdkfd_resume_iommu(struct amdgpu_device *adev)
-{
-	int r = 0;
-
-	if (adev->kfd.dev)
-		r = kgd2kfd_resume_iommu(adev->kfd.dev);
-
-	return r;
-}
-
 int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm)
 {
 	int r = 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 082c9f4cfd34..2fe9860725bd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -148,7 +148,6 @@ int amdgpu_amdkfd_init(void);
 void amdgpu_amdkfd_fini(void);
 
 void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm);
-int amdgpu_amdkfd_resume_iommu(struct amdgpu_device *adev);
 int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm);
 void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 			const void *ih_ring_entry);
@@ -399,7 +398,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 			 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);
-int kgd2kfd_resume_iommu(struct kfd_dev *kfd);
 int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm);
 int kgd2kfd_pre_reset(struct kfd_dev *kfd);
 int kgd2kfd_post_reset(struct kfd_dev *kfd);
@@ -439,11 +437,6 @@ static inline void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 {
 }
 
-static int __maybe_unused kgd2kfd_resume_iommu(struct kfd_dev *kfd)
-{
-	return 0;
-}
-
 static inline int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 {
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index aa171db68639..91d7f9e4fefc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3236,12 +3236,6 @@ static int amdgpu_device_ip_resume(struct amdgpu_device *adev)
 {
 	int r;
 
-	if (!adev->in_s0ix) {
-		r = amdgpu_amdkfd_resume_iommu(adev);
-		if (r)
-			return r;
-	}
-
 	r = amdgpu_device_ip_resume_phase1(adev);
 	if (r)
 		return r;
@@ -4946,9 +4940,6 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle,
 				dev_warn(tmp_adev->dev, "asic atom init failed!");
 			} else {
 				dev_info(tmp_adev->dev, "GPU reset succeeded, trying to resume\n");
-				r = amdgpu_amdkfd_resume_iommu(tmp_adev);
-				if (r)
-					goto out;
 
 				r = amdgpu_device_ip_resume_phase1(tmp_adev);
 				if (r)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 7691177d87aa..d11fa7dec4c4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -756,20 +756,6 @@ MODULE_PARM_DESC(debug_largebar,
 	"Debug large-bar flag used to simulate large-bar capability on non-large bar machine (0 = disable, 1 = enable)");
 
 /**
- * DOC: ignore_crat (int)
- * Ignore CRAT table during KFD initialization. By default, KFD uses the ACPI CRAT
- * table to get information about AMD APUs. This option can serve as a workaround on
- * systems with a broken CRAT table.
- *
- * Default is auto (according to asic type, iommu_v2, and crat table, to decide
- * whether use CRAT)
- */
-int ignore_crat;
-module_param(ignore_crat, int, 0444);
-MODULE_PARM_DESC(ignore_crat,
-	"Ignore CRAT table during KFD initialization (0 = auto (default), 1 = ignore CRAT)");
-
-/**
  * DOC: halt_if_hws_hang (int)
  * Halt if HWS hang is detected. Default value, 0, disables the halt on hang.
  * Setting 1 enables halt on hang.
diff --git a/drivers/gpu/drm/amd/amdkfd/Kconfig b/drivers/gpu/drm/amd/amdkfd/Kconfig
index 93bd4eda0d94..d3c3d3ab7225 100644
--- a/drivers/gpu/drm/amd/amdkfd/Kconfig
+++ b/drivers/gpu/drm/amd/amdkfd/Kconfig
@@ -6,7 +6,6 @@
 config HSA_AMD
 	bool "HSA kernel driver for AMD GPU devices"
 	depends on DRM_AMDGPU && (X86_64 || ARM64 || PPC64)
-	imply AMD_IOMMU_V2 if X86_64
 	select HMM_MIRROR
 	select MMU_NOTIFIER
 	select DRM_AMDGPU_USERPTR
diff --git a/drivers/gpu/drm/amd/amdkfd/Makefile b/drivers/gpu/drm/amd/amdkfd/Makefile
index 2ec8f27c5366..a5ae7bcf44eb 100644
--- a/drivers/gpu/drm/amd/amdkfd/Makefile
+++ b/drivers/gpu/drm/amd/amdkfd/Makefile
@@ -59,10 +59,6 @@ AMDKFD_FILES	:= $(AMDKFD_PATH)/kfd_module.o \
 		$(AMDKFD_PATH)/kfd_crat.o \
 		$(AMDKFD_PATH)/kfd_debug.o
 
-ifneq ($(CONFIG_AMD_IOMMU_V2),)
-AMDKFD_FILES += $(AMDKFD_PATH)/kfd_iommu.o
-endif
-
 ifneq ($(CONFIG_DEBUG_FS),)
 AMDKFD_FILES += $(AMDKFD_PATH)/kfd_debugfs.o
 endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index f3a0efbdd145..c37f1fcd2165 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1026,9 +1026,6 @@ bool kfd_dev_is_large_bar(struct kfd_node *dev)
 		return true;
 	}
 
-	if (dev->kfd->use_iommu_v2)
-		return false;
-
 	if (dev->local_mem_info.local_mem_size_private == 0 &&
 	    dev->local_mem_info.local_mem_size_public > 0)
 		return true;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index f5a6f562e2a8..86fb7ac7982a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -26,7 +26,6 @@
 #include "kfd_crat.h"
 #include "kfd_priv.h"
 #include "kfd_topology.h"
-#include "kfd_iommu.h"
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
 
@@ -1536,72 +1535,6 @@ int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pc
 	return num_of_cache_types;
 }
 
-static bool kfd_ignore_crat(void)
-{
-	bool ret;
-
-	if (ignore_crat)
-		return true;
-
-	ret = true;
-
-	return ret;
-}
-
-/*
- * kfd_create_crat_image_acpi - Allocates memory for CRAT image and
- * copies CRAT from ACPI (if available).
- * NOTE: Call kfd_destroy_crat_image to free CRAT image memory
- *
- *	@crat_image: CRAT read from ACPI. If no CRAT in ACPI then
- *		     crat_image will be NULL
- *	@size: [OUT] size of crat_image
- *
- *	Return 0 if successful else return error code
- */
-int kfd_create_crat_image_acpi(void **crat_image, size_t *size)
-{
-	struct acpi_table_header *crat_table;
-	acpi_status status;
-	void *pcrat_image;
-	int rc = 0;
-
-	if (!crat_image)
-		return -EINVAL;
-
-	*crat_image = NULL;
-
-	if (kfd_ignore_crat()) {
-		pr_info("CRAT table disabled by module option\n");
-		return -ENODATA;
-	}
-
-	/* Fetch the CRAT table from ACPI */
-	status = acpi_get_table(CRAT_SIGNATURE, 0, &crat_table);
-	if (status == AE_NOT_FOUND) {
-		pr_info("CRAT table not found\n");
-		return -ENODATA;
-	} else if (ACPI_FAILURE(status)) {
-		const char *err = acpi_format_exception(status);
-
-		pr_err("CRAT table error: %s\n", err);
-		return -EINVAL;
-	}
-
-	pcrat_image = kvmalloc(crat_table->length, GFP_KERNEL);
-	if (!pcrat_image) {
-		rc = -ENOMEM;
-		goto out;
-	}
-
-	memcpy(pcrat_image, crat_table, crat_table->length);
-	*crat_image = pcrat_image;
-	*size = crat_table->length;
-out:
-	acpi_put_table(crat_table);
-	return rc;
-}
-
 /* Memory required to create Virtual CRAT.
  * Since there is no easy way to predict the amount of memory required, the
  * following amount is allocated for GPU Virtual CRAT. This is
@@ -2169,12 +2102,6 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
 
 	cu->hsa_capability = 0;
 
-	/* Check if this node supports IOMMU. During parsing this flag will
-	 * translate to HSA_CAP_ATS_PRESENT
-	 */
-	if (!kfd_iommu_check_device(kdev->kfd))
-		cu->hsa_capability |= CRAT_CU_FLAGS_IOMMU_PRESENT;
-
 	crat_table->length += sub_type_hdr->length;
 	crat_table->total_entries++;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
index fc719389b5d6..387a8ef49385 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h
@@ -307,7 +307,6 @@ struct kfd_gpu_cache_info {
 };
 int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pcache_info);
 
-int kfd_create_crat_image_acpi(void **crat_image, size_t *size);
 void kfd_destroy_crat_image(void *crat_image);
 int kfd_parse_crat_table(void *crat_image, struct list_head *device_list,
 			 uint32_t proximity_domain);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index b023fff8f97c..93ce181eb3ba 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -29,7 +29,6 @@
 #include "kfd_pm4_headers_vi.h"
 #include "kfd_pm4_headers_aldebaran.h"
 #include "cwsr_trap_handler.h"
-#include "kfd_iommu.h"
 #include "amdgpu_amdkfd.h"
 #include "kfd_smi_events.h"
 #include "kfd_svm.h"
@@ -62,7 +61,6 @@ static int kfd_gtt_sa_init(struct kfd_dev *kfd, unsigned int buf_size,
 				unsigned int chunk_size);
 static void kfd_gtt_sa_fini(struct kfd_dev *kfd);
 
-static int kfd_resume_iommu(struct kfd_dev *kfd);
 static int kfd_resume(struct kfd_node *kfd);
 
 static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd)
@@ -755,15 +753,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	kfd->noretry = kfd->adev->gmc.noretry;
 
-	/* If CRAT is broken, won't set iommu enabled */
-	kfd_double_confirm_iommu_support(kfd);
-
-	if (kfd_iommu_device_init(kfd)) {
-		kfd->use_iommu_v2 = false;
-		dev_err(kfd_device, "Error initializing iommuv2\n");
-		goto device_iommu_error;
-	}
-
 	kfd_cwsr_init(kfd);
 
 	dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n",
@@ -838,9 +827,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	svm_range_set_max_pages(kfd->adev);
 
-	if (kfd_resume_iommu(kfd))
-		goto kfd_resume_iommu_error;
-
 	spin_lock_init(&kfd->watch_points_lock);
 
 	kfd->init_complete = true;
@@ -852,11 +838,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	goto out;
 
-kfd_resume_iommu_error:
 node_init_error:
 node_alloc_error:
 	kfd_cleanup_nodes(kfd, i);
-device_iommu_error:
 	kfd_doorbell_fini(kfd);
 kfd_doorbell_error:
 	kfd_gtt_sa_fini(kfd);
@@ -971,7 +955,6 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
 		node = kfd->nodes[i];
 		node->dqm->ops.stop(node->dqm);
 	}
-	kfd_iommu_suspend(kfd);
 }
 
 int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
@@ -1001,26 +984,6 @@ int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
 	return ret;
 }
 
-int kgd2kfd_resume_iommu(struct kfd_dev *kfd)
-{
-	if (!kfd->init_complete)
-		return 0;
-
-	return kfd_resume_iommu(kfd);
-}
-
-static int kfd_resume_iommu(struct kfd_dev *kfd)
-{
-	int err = 0;
-
-	err = kfd_iommu_resume(kfd);
-	if (err)
-		dev_err(kfd_device,
-			"Failed to resume IOMMU for device %x:%x\n",
-			kfd->adev->pdev->vendor, kfd->adev->pdev->device);
-	return err;
-}
-
 static int kfd_resume(struct kfd_node *node)
 {
 	int err = 0;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 53b83ffe7215..20aa3b7dbc36 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -2560,7 +2560,7 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_node *dev)
 	switch (dev->adev->asic_type) {
 	case CHIP_KAVERI:
 	case CHIP_HAWAII:
-		device_queue_manager_init_cik_hawaii(&dqm->asic_ops);
+		device_queue_manager_init_cik(&dqm->asic_ops);
 		break;
 
 	case CHIP_CARRIZO:
@@ -2570,7 +2570,7 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_node *dev)
 	case CHIP_POLARIS11:
 	case CHIP_POLARIS12:
 	case CHIP_VEGAM:
-		device_queue_manager_init_vi_tonga(&dqm->asic_ops);
+		device_queue_manager_init_vi(&dqm->asic_ops);
 		break;
 
 	default:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
index 7dd4b177219d..805d2e4f907a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h
@@ -269,12 +269,8 @@ struct device_queue_manager {
 
 void device_queue_manager_init_cik(
 		struct device_queue_manager_asic_ops *asic_ops);
-void device_queue_manager_init_cik_hawaii(
-		struct device_queue_manager_asic_ops *asic_ops);
 void device_queue_manager_init_vi(
 		struct device_queue_manager_asic_ops *asic_ops);
-void device_queue_manager_init_vi_tonga(
-		struct device_queue_manager_asic_ops *asic_ops);
 void device_queue_manager_init_v9(
 		struct device_queue_manager_asic_ops *asic_ops);
 void device_queue_manager_init_v10_navi10(
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c
index b1ab5b0775e1..d4d95c7f2e5d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c
@@ -34,17 +34,13 @@ static bool set_cache_memory_policy_cik(struct device_queue_manager *dqm,
 				   void __user *alternate_aperture_base,
 				   uint64_t alternate_aperture_size);
 static int update_qpd_cik(struct device_queue_manager *dqm,
-					struct qcm_process_device *qpd);
-static int update_qpd_cik_hawaii(struct device_queue_manager *dqm,
-					struct qcm_process_device *qpd);
-static void init_sdma_vm(struct device_queue_manager *dqm, struct queue *q,
-				struct qcm_process_device *qpd);
-static void init_sdma_vm_hawaii(struct device_queue_manager *dqm,
-				struct queue *q,
-				struct qcm_process_device *qpd);
+			  struct qcm_process_device *qpd);
+static void init_sdma_vm(struct device_queue_manager *dqm,
+			 struct queue *q,
+			 struct qcm_process_device *qpd);
 
 void device_queue_manager_init_cik(
-		struct device_queue_manager_asic_ops *asic_ops)
+	struct device_queue_manager_asic_ops *asic_ops)
 {
 	asic_ops->set_cache_memory_policy = set_cache_memory_policy_cik;
 	asic_ops->update_qpd = update_qpd_cik;
@@ -52,15 +48,6 @@ void device_queue_manager_init_cik(
 	asic_ops->mqd_manager_init = mqd_manager_init_cik;
 }
 
-void device_queue_manager_init_cik_hawaii(
-		struct device_queue_manager_asic_ops *asic_ops)
-{
-	asic_ops->set_cache_memory_policy = set_cache_memory_policy_cik;
-	asic_ops->update_qpd = update_qpd_cik_hawaii;
-	asic_ops->init_sdma_vm = init_sdma_vm_hawaii;
-	asic_ops->mqd_manager_init = mqd_manager_init_cik_hawaii;
-}
-
 static uint32_t compute_sh_mem_bases_64bit(unsigned int top_address_nybble)
 {
 	/* In 64-bit mode, we can only control the top 3 bits of the LDS,
@@ -115,41 +102,7 @@ static bool set_cache_memory_policy_cik(struct device_queue_manager *dqm,
 }
 
 static int update_qpd_cik(struct device_queue_manager *dqm,
-		struct qcm_process_device *qpd)
-{
-	struct kfd_process_device *pdd;
-	unsigned int temp;
-
-	pdd = qpd_to_pdd(qpd);
-
-	/* check if sh_mem_config register already configured */
-	if (qpd->sh_mem_config == 0) {
-		qpd->sh_mem_config =
-			ALIGNMENT_MODE(SH_MEM_ALIGNMENT_MODE_UNALIGNED) |
-			DEFAULT_MTYPE(MTYPE_NONCACHED) |
-			APE1_MTYPE(MTYPE_NONCACHED);
-		qpd->sh_mem_ape1_limit = 0;
-		qpd->sh_mem_ape1_base = 0;
-	}
-
-	if (qpd->pqm->process->is_32bit_user_mode) {
-		temp = get_sh_mem_bases_32(pdd);
-		qpd->sh_mem_bases = SHARED_BASE(temp);
-		qpd->sh_mem_config |= PTR32;
-	} else {
-		temp = get_sh_mem_bases_nybble_64(pdd);
-		qpd->sh_mem_bases = compute_sh_mem_bases_64bit(temp);
-		qpd->sh_mem_config |= 1  << SH_MEM_CONFIG__PRIVATE_ATC__SHIFT;
-	}
-
-	pr_debug("is32bit process: %d sh_mem_bases nybble: 0x%X and register 0x%X\n",
-		qpd->pqm->process->is_32bit_user_mode, temp, qpd->sh_mem_bases);
-
-	return 0;
-}
-
-static int update_qpd_cik_hawaii(struct device_queue_manager *dqm,
-		struct qcm_process_device *qpd)
+			  struct qcm_process_device *qpd)
 {
 	struct kfd_process_device *pdd;
 	unsigned int temp;
@@ -178,25 +131,9 @@ static int update_qpd_cik_hawaii(struct device_queue_manager *dqm,
 	return 0;
 }
 
-static void init_sdma_vm(struct device_queue_manager *dqm, struct queue *q,
-				struct qcm_process_device *qpd)
-{
-	uint32_t value = (1 << SDMA0_RLC0_VIRTUAL_ADDR__ATC__SHIFT);
-
-	if (q->process->is_32bit_user_mode)
-		value |= (1 << SDMA0_RLC0_VIRTUAL_ADDR__PTR32__SHIFT) |
-				get_sh_mem_bases_32(qpd_to_pdd(qpd));
-	else
-		value |= ((get_sh_mem_bases_nybble_64(qpd_to_pdd(qpd))) <<
-				SDMA0_RLC0_VIRTUAL_ADDR__SHARED_BASE__SHIFT) &
-				SDMA0_RLC0_VIRTUAL_ADDR__SHARED_BASE_MASK;
-
-	q->properties.sdma_vm_addr = value;
-}
-
-static void init_sdma_vm_hawaii(struct device_queue_manager *dqm,
-				struct queue *q,
-				struct qcm_process_device *qpd)
+static void init_sdma_vm(struct device_queue_manager *dqm,
+			 struct queue *q,
+			 struct qcm_process_device *qpd)
 {
 	/* On dGPU we're always in GPUVM64 addressing mode with 64-bit
 	 * aperture addresses.
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
index 8af643388768..54eb1bff903c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c
@@ -60,7 +60,7 @@ static int update_qpd_v9(struct device_queue_manager *dqm,
 		qpd->sh_mem_config = SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
 					SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT;
 
-		if (dqm->dev->kfd->noretry && !dqm->dev->kfd->use_iommu_v2)
+		if (dqm->dev->kfd->noretry)
 			qpd->sh_mem_config |= 1 << SH_MEM_CONFIG__RETRY_DISABLE__SHIFT;
 
 		if (KFD_GC_VERSION(dqm->dev->kfd) == IP_VERSION(9, 4, 3))
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c
index d7d45832df0f..b291ee0fab94 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c
@@ -28,29 +28,19 @@
 #include "oss/oss_3_0_sh_mask.h"
 
 static bool set_cache_memory_policy_vi(struct device_queue_manager *dqm,
-				   struct qcm_process_device *qpd,
-				   enum cache_policy default_policy,
-				   enum cache_policy alternate_policy,
-				   void __user *alternate_aperture_base,
-				   uint64_t alternate_aperture_size);
-static bool set_cache_memory_policy_vi_tonga(struct device_queue_manager *dqm,
-			struct qcm_process_device *qpd,
-			enum cache_policy default_policy,
-			enum cache_policy alternate_policy,
-			void __user *alternate_aperture_base,
-			uint64_t alternate_aperture_size);
+				       struct qcm_process_device *qpd,
+				       enum cache_policy default_policy,
+				       enum cache_policy alternate_policy,
+				       void __user *alternate_aperture_base,
+				       uint64_t alternate_aperture_size);
 static int update_qpd_vi(struct device_queue_manager *dqm,
-					struct qcm_process_device *qpd);
-static int update_qpd_vi_tonga(struct device_queue_manager *dqm,
-			struct qcm_process_device *qpd);
-static void init_sdma_vm(struct device_queue_manager *dqm, struct queue *q,
-				struct qcm_process_device *qpd);
-static void init_sdma_vm_tonga(struct device_queue_manager *dqm,
-			struct queue *q,
-			struct qcm_process_device *qpd);
+			 struct qcm_process_device *qpd);
+static void init_sdma_vm(struct device_queue_manager *dqm,
+			 struct queue *q,
+			 struct qcm_process_device *qpd);
 
 void device_queue_manager_init_vi(
-		struct device_queue_manager_asic_ops *asic_ops)
+	struct device_queue_manager_asic_ops *asic_ops)
 {
 	asic_ops->set_cache_memory_policy = set_cache_memory_policy_vi;
 	asic_ops->update_qpd = update_qpd_vi;
@@ -58,15 +48,6 @@ void device_queue_manager_init_vi(
 	asic_ops->mqd_manager_init = mqd_manager_init_vi;
 }
 
-void device_queue_manager_init_vi_tonga(
-		struct device_queue_manager_asic_ops *asic_ops)
-{
-	asic_ops->set_cache_memory_policy = set_cache_memory_policy_vi_tonga;
-	asic_ops->update_qpd = update_qpd_vi_tonga;
-	asic_ops->init_sdma_vm = init_sdma_vm_tonga;
-	asic_ops->mqd_manager_init = mqd_manager_init_vi_tonga;
-}
-
 static uint32_t compute_sh_mem_bases_64bit(unsigned int top_address_nybble)
 {
 	/* In 64-bit mode, we can only control the top 3 bits of the LDS,
@@ -96,35 +77,6 @@ static uint32_t compute_sh_mem_bases_64bit(unsigned int top_address_nybble)
 }
 
 static bool set_cache_memory_policy_vi(struct device_queue_manager *dqm,
-				   struct qcm_process_device *qpd,
-				   enum cache_policy default_policy,
-				   enum cache_policy alternate_policy,
-				   void __user *alternate_aperture_base,
-				   uint64_t alternate_aperture_size)
-{
-	uint32_t default_mtype;
-	uint32_t ape1_mtype;
-
-	default_mtype = (default_policy == cache_policy_coherent) ?
-			MTYPE_CC :
-			MTYPE_NC;
-
-	ape1_mtype = (alternate_policy == cache_policy_coherent) ?
-			MTYPE_CC :
-			MTYPE_NC;
-
-	qpd->sh_mem_config = (qpd->sh_mem_config &
-			SH_MEM_CONFIG__ADDRESS_MODE_MASK) |
-		SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
-				SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT |
-		default_mtype << SH_MEM_CONFIG__DEFAULT_MTYPE__SHIFT |
-		ape1_mtype << SH_MEM_CONFIG__APE1_MTYPE__SHIFT |
-		SH_MEM_CONFIG__PRIVATE_ATC_MASK;
-
-	return true;
-}
-
-static bool set_cache_memory_policy_vi_tonga(struct device_queue_manager *dqm,
 		struct qcm_process_device *qpd,
 		enum cache_policy default_policy,
 		enum cache_policy alternate_policy,
@@ -152,48 +104,7 @@ static bool set_cache_memory_policy_vi_tonga(struct device_queue_manager *dqm,
 }
 
 static int update_qpd_vi(struct device_queue_manager *dqm,
-					struct qcm_process_device *qpd)
-{
-	struct kfd_process_device *pdd;
-	unsigned int temp;
-
-	pdd = qpd_to_pdd(qpd);
-
-	/* check if sh_mem_config register already configured */
-	if (qpd->sh_mem_config == 0) {
-		qpd->sh_mem_config =
-			SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
-				SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT |
-			MTYPE_CC << SH_MEM_CONFIG__DEFAULT_MTYPE__SHIFT |
-			MTYPE_CC << SH_MEM_CONFIG__APE1_MTYPE__SHIFT |
-			SH_MEM_CONFIG__PRIVATE_ATC_MASK;
-
-		qpd->sh_mem_ape1_limit = 0;
-		qpd->sh_mem_ape1_base = 0;
-	}
-
-	if (qpd->pqm->process->is_32bit_user_mode) {
-		temp = get_sh_mem_bases_32(pdd);
-		qpd->sh_mem_bases = temp << SH_MEM_BASES__SHARED_BASE__SHIFT;
-		qpd->sh_mem_config |= SH_MEM_ADDRESS_MODE_HSA32 <<
-					SH_MEM_CONFIG__ADDRESS_MODE__SHIFT;
-	} else {
-		temp = get_sh_mem_bases_nybble_64(pdd);
-		qpd->sh_mem_bases = compute_sh_mem_bases_64bit(temp);
-		qpd->sh_mem_config |= SH_MEM_ADDRESS_MODE_HSA64 <<
-			SH_MEM_CONFIG__ADDRESS_MODE__SHIFT;
-		qpd->sh_mem_config |= 1  <<
-			SH_MEM_CONFIG__PRIVATE_ATC__SHIFT;
-	}
-
-	pr_debug("is32bit process: %d sh_mem_bases nybble: 0x%X and register 0x%X\n",
-		qpd->pqm->process->is_32bit_user_mode, temp, qpd->sh_mem_bases);
-
-	return 0;
-}
-
-static int update_qpd_vi_tonga(struct device_queue_manager *dqm,
-			struct qcm_process_device *qpd)
+			 struct qcm_process_device *qpd)
 {
 	struct kfd_process_device *pdd;
 	unsigned int temp;
@@ -226,25 +137,9 @@ static int update_qpd_vi_tonga(struct device_queue_manager *dqm,
 	return 0;
 }
 
-static void init_sdma_vm(struct device_queue_manager *dqm, struct queue *q,
-				struct qcm_process_device *qpd)
-{
-	uint32_t value = (1 << SDMA0_RLC0_VIRTUAL_ADDR__ATC__SHIFT);
-
-	if (q->process->is_32bit_user_mode)
-		value |= (1 << SDMA0_RLC0_VIRTUAL_ADDR__PTR32__SHIFT) |
-				get_sh_mem_bases_32(qpd_to_pdd(qpd));
-	else
-		value |= ((get_sh_mem_bases_nybble_64(qpd_to_pdd(qpd))) <<
-				SDMA0_RLC0_VIRTUAL_ADDR__SHARED_BASE__SHIFT) &
-				SDMA0_RLC0_VIRTUAL_ADDR__SHARED_BASE_MASK;
-
-	q->properties.sdma_vm_addr = value;
-}
-
-static void init_sdma_vm_tonga(struct device_queue_manager *dqm,
-			struct queue *q,
-			struct qcm_process_device *qpd)
+static void init_sdma_vm(struct device_queue_manager *dqm,
+			 struct queue *q,
+			 struct qcm_process_device *qpd)
 {
 	/* On dGPU we're always in GPUVM64 addressing mode with 64-bit
 	 * aperture addresses.
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index 8081a9408006..0f58be65132f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
@@ -31,7 +31,6 @@
 #include <linux/memory.h>
 #include "kfd_priv.h"
 #include "kfd_events.h"
-#include "kfd_iommu.h"
 #include <linux/device.h>
 
 /*
@@ -1146,87 +1145,6 @@ static void lookup_events_by_type_and_signal(struct kfd_process *p,
 	rcu_read_unlock();
 }
 
-#ifdef KFD_SUPPORT_IOMMU_V2
-void kfd_signal_iommu_event(struct kfd_node *dev, u32 pasid,
-		unsigned long address, bool is_write_requested,
-		bool is_execute_requested)
-{
-	struct kfd_hsa_memory_exception_data memory_exception_data;
-	struct vm_area_struct *vma;
-	int user_gpu_id;
-
-	/*
-	 * Because we are called from arbitrary context (workqueue) as opposed
-	 * to process context, kfd_process could attempt to exit while we are
-	 * running so the lookup function increments the process ref count.
-	 */
-	struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
-	struct mm_struct *mm;
-
-	if (!p)
-		return; /* Presumably process exited. */
-
-	/* Take a safe reference to the mm_struct, which may otherwise
-	 * disappear even while the kfd_process is still referenced.
-	 */
-	mm = get_task_mm(p->lead_thread);
-	if (!mm) {
-		kfd_unref_process(p);
-		return; /* Process is exiting */
-	}
-
-	user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id);
-	if (unlikely(user_gpu_id == -EINVAL)) {
-		WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id);
-		return;
-	}
-	memset(&memory_exception_data, 0, sizeof(memory_exception_data));
-
-	mmap_read_lock(mm);
-	vma = find_vma(mm, address);
-
-	memory_exception_data.gpu_id = user_gpu_id;
-	memory_exception_data.va = address;
-	/* Set failure reason */
-	memory_exception_data.failure.NotPresent = 1;
-	memory_exception_data.failure.NoExecute = 0;
-	memory_exception_data.failure.ReadOnly = 0;
-	if (vma && address &