/******************************************************************************* Copyright (c) 2015-2022 NVIDIA Corporation Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. *******************************************************************************/ #include "uvm_api.h" #include "uvm_test.h" #include "uvm_test_ioctl.h" #include "uvm_global.h" #include "uvm_va_space.h" #include "uvm_va_space_mm.h" #include "uvm_va_range.h" #include "uvm_va_block.h" #include "uvm_test_rng.h" #include "uvm_kvmalloc.h" #include "uvm_perf_events.h" #include "uvm_tools.h" #include "uvm_mmu.h" #include "uvm_gpu_access_counters.h" #include "uvm_pmm_sysmem.h" static NV_STATUS uvm_test_get_gpu_ref_count(UVM_TEST_GET_GPU_REF_COUNT_PARAMS *params, struct file *filp) { NvU64 retained_count = 0; uvm_parent_gpu_t *parent_gpu; uvm_gpu_t *gpu = NULL; uvm_mutex_lock(&g_uvm_global.global_lock); parent_gpu = uvm_parent_gpu_get_by_uuid(¶ms->gpu_uuid); if (parent_gpu) gpu = uvm_gpu_get_by_parent_and_swizz_id(parent_gpu, params->swizz_id); if (gpu != NULL) retained_count = uvm_gpu_retained_count(gpu); uvm_mutex_unlock(&g_uvm_global.global_lock); params->ref_count = retained_count; return NV_OK; } static NV_STATUS uvm_test_peer_ref_count(UVM_TEST_PEER_REF_COUNT_PARAMS *params, struct file *filp) { NvU64 registered_ref_count = 0; uvm_gpu_t *gpu0 = NULL; uvm_gpu_t *gpu1 = NULL; NV_STATUS status = NV_OK; uvm_mutex_lock(&g_uvm_global.global_lock); gpu0 = uvm_gpu_get_by_uuid(¶ms->gpu_uuid_1); gpu1 = uvm_gpu_get_by_uuid(¶ms->gpu_uuid_2); if (gpu0 != NULL && gpu1 != NULL) { uvm_gpu_peer_t *peer_caps = uvm_gpu_peer_caps(gpu0, gpu1); registered_ref_count = peer_caps->ref_count; } else { status = NV_ERR_INVALID_DEVICE; } uvm_mutex_unlock(&g_uvm_global.global_lock); params->ref_count = registered_ref_count; return status; } static NV_STATUS uvm_test_make_channel_stops_immediate(UVM_TEST_MAKE_CHANNEL_STOPS_IMMEDIATE_PARAMS *params, struct file *filp) { uvm_va_space_get(filp)->user_channel_stops_are_immediate = NV_TRUE; return NV_OK; } static NV_STATUS uvm_test_nv_kthread_q(UVM_TEST_NV_KTHREAD_Q_PARAMS *params, struct file *filp) { // The nv-kthread-q system returns 0 or -1, because it is not actually // part of UVM. UVM needs to run this test, because otherwise, the // nv-kthread-q code would not get adequate test coverage. That's because // UVM is the first user of nv-kthread-q. int result = nv_kthread_q_run_self_test(); if (result == 0) return NV_OK; return NV_ERR_INVALID_STATE; } static NV_STATUS uvm_test_numa_get_closest_cpu_node_to_gpu(UVM_TEST_NUMA_GET_CLOSEST_CPU_NODE_TO_GPU_PARAMS *params, struct file *filp) { uvm_gpu_t *gpu; NV_STATUS status; uvm_rm_user_object_t user_rm_va_space = { .rm_control_fd = -1, .user_client = params->client, .user_object = params->smc_part_ref }; status = uvm_gpu_retain_by_uuid(¶ms->gpu_uuid, &user_rm_va_space, &gpu); if (status != NV_OK) return status; params->node_id = gpu->parent->closest_cpu_numa_node; uvm_gpu_release(gpu); return NV_OK; } // Callers of this function should ensure that node is not NUMA_NO_NODE in order // to avoid overrunning the kernel's node to cpumask map. static NV_STATUS uvm_test_verify_bh_affinity(uvm_intr_handler_t *isr, int node) { UVM_ASSERT(node != NUMA_NO_NODE); // If the bottom half has been executed but its cpus_used_mask is empty // something obviously went wrong. Otherwise, check that the CPUs on which // the bottom half was executed is a subset of the NUMA node's cpumask. if ((isr->stats.bottom_half_count && cpumask_empty(&isr->stats.cpus_used_mask)) || !cpumask_subset(&isr->stats.cpus_used_mask, uvm_cpumask_of_node(node))) { UVM_TEST_PRINT("ISR BH cpu mask check failed! BH ran on CPU cores outside NUMA %u\n", node); return NV_ERR_INVALID_STATE; } return NV_OK; } static NV_STATUS uvm_test_numa_check_affinity(UVM_TEST_NUMA_CHECK_AFFINITY_PARAMS *params, struct file *filp) { uvm_gpu_t *gpu; NV_STATUS status; uvm_rm_user_object_t user_rm_va_space = { .rm_control_fd = -1, .user_client = params->client, .user_object = params->smc_part_ref }; if (!UVM_THREAD_AFFINITY_SUPPORTED()) return NV_ERR_NOT_SUPPORTED; status = uvm_gpu_retain_by_uuid(¶ms->gpu_uuid, &user_rm_va_space, &gpu); if (status != NV_OK) return status; // If the GPU is not attached to a NUMA node, there is nothing to do. if (gpu->parent->closest_cpu_numa_node == NUMA_NO_NODE) { status = NV_ERR_NOT_SUPPORTED; goto release; } if (gpu->parent->replayable_faults_supported) { uvm_gpu_replayable_faults_isr_lock(gpu->parent); status = uvm_test_verify_bh_affinity(&gpu->parent->isr.replayable_faults, gpu->parent->closest_cpu_numa_node); uvm_gpu_replayable_faults_isr_unlock(gpu->parent); if (status != NV_OK) goto release; if (gpu->parent->non_replayable_faults_supported) { uvm_gpu_non_replayable_faults_isr_lock(gpu->parent); status = uvm_test_verify_bh_affinity(&gpu->parent->isr.non_replayable_faults, gpu->parent->closest_cpu_numa_node); uvm_gpu_non_replayable_faults_isr_unlock(gpu->parent); if (status != NV_OK) goto release; } if (gpu->parent->access_counters_supported) { uvm_gpu_access_counters_isr_lock(gpu->parent); status = uvm_test_verify_bh_affinity(&gpu->parent->isr.access_counters, gpu->parent->closest_cpu_numa_node); uvm_gpu_access_counters_isr_unlock(gpu->parent); } } release: uvm_gpu_release(gpu); return status; } static NV_STATUS uvm_test_get_kernel_virtual_address(UVM_TEST_GET_KERNEL_VIRTUAL_ADDRESS_PARAMS *params, struct file *filp) { params->addr = (NvU64)uvm_va_space_get(filp); return NV_OK; } static NV_STATUS uvm_test_get_user_space_end_address(UVM_TEST_GET_USER_SPACE_END_ADDRESS_PARAMS *params, struct file *flip) { params->user_space_end_address = TASK_SIZE; return NV_OK; } static NV_STATUS uvm_test_cgroup_accounting_supported(UVM_TEST_CGROUP_ACCOUNTING_SUPPORTED_PARAMS *params, struct file *flip) { return UVM_CGROUP_ACCOUNTING_SUPPORTED() ? NV_OK : NV_ERR_NOT_SUPPORTED; } long uvm_test_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { // Disable all test entry points if the module parameter wasn't provided. // These should not be enabled in a production environment. if (!uvm_enable_builtin_tests) { UVM_INFO_PRINT("ioctl %d not found. Did you mean to insmod with uvm_enable_builtin_tests=1?\n", cmd); return -EINVAL; } switch (cmd) { UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_GET_GPU_REF_COUNT, uvm_test_get_gpu_ref_count); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RNG_SANITY, uvm_test_rng_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RANGE_TREE_DIRECTED, uvm_test_range_tree_directed); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RANGE_TREE_RANDOM, uvm_test_range_tree_random); UVM_ROUTE_CMD_ALLOC_INIT_CHECK(UVM_TEST_VA_RANGE_INFO, uvm_test_va_range_info); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RM_MEM_SANITY, uvm_test_rm_mem_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_GPU_SEMAPHORE_SANITY, uvm_test_gpu_semaphore_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PEER_REF_COUNT, uvm_test_peer_ref_count); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_RANGE_SPLIT, uvm_test_va_range_split); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_RANGE_INJECT_SPLIT_ERROR, uvm_test_va_range_inject_split_error); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PAGE_TREE, uvm_test_page_tree); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_CHANGE_PTE_MAPPING, uvm_test_change_pte_mapping); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_TRACKER_SANITY, uvm_test_tracker_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PUSH_SANITY, uvm_test_push_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_CHANNEL_SANITY, uvm_test_channel_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_CHANNEL_STRESS, uvm_test_channel_stress); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_CE_SANITY, uvm_test_ce_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_HOST_SANITY, uvm_test_host_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_SPACE_MM_OR_CURRENT_RETAIN, uvm_test_va_space_mm_or_current_retain); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_BLOCK_INFO, uvm_test_va_block_info); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_LOCK_SANITY, uvm_test_lock_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PERF_UTILS_SANITY, uvm_test_perf_utils_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_KVMALLOC, uvm_test_kvmalloc); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_QUERY, uvm_test_pmm_query); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_CHECK_LEAK, uvm_test_pmm_check_leak); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PERF_EVENTS_SANITY, uvm_test_perf_events_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PERF_MODULE_SANITY, uvm_test_perf_module_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RANGE_ALLOCATOR_SANITY, uvm_test_range_allocator_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_GET_RM_PTES, uvm_test_get_rm_ptes); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_FAULT_BUFFER_FLUSH, uvm_test_fault_buffer_flush); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_INJECT_TOOLS_EVENT, uvm_test_inject_tools_event); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_INCREMENT_TOOLS_COUNTER, uvm_test_increment_tools_counter); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_MEM_SANITY, uvm_test_mem_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_MAKE_CHANNEL_STOPS_IMMEDIATE, uvm_test_make_channel_stops_immediate); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_BLOCK_INJECT_ERROR, uvm_test_va_block_inject_error); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PEER_IDENTITY_MAPPINGS, uvm_test_peer_identity_mappings); UVM_ROUTE_CMD_ALLOC_INIT_CHECK(UVM_TEST_VA_RESIDENCY_INFO, uvm_test_va_residency_info); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_ASYNC_ALLOC, uvm_test_pmm_async_alloc); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_SET_PREFETCH_FILTERING, uvm_test_set_prefetch_filtering); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_SANITY, uvm_test_pmm_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_INVALIDATE_TLB, uvm_test_invalidate_tlb); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_BLOCK, uvm_test_va_block); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_EVICT_CHUNK, uvm_test_evict_chunk); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_FLUSH_DEFERRED_WORK, uvm_test_flush_deferred_work); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_NV_KTHREAD_Q, uvm_test_nv_kthread_q); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_SET_PAGE_PREFETCH_POLICY, uvm_test_set_page_prefetch_policy); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RANGE_GROUP_TREE, uvm_test_range_group_tree); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RANGE_GROUP_RANGE_INFO, uvm_test_range_group_range_info); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RANGE_GROUP_RANGE_COUNT, uvm_test_range_group_range_count); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_GET_PREFETCH_FAULTS_REENABLE_LAPSE, uvm_test_get_prefetch_faults_reenable_lapse); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_SET_PREFETCH_FAULTS_REENABLE_LAPSE, uvm_test_set_prefetch_faults_reenable_lapse); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_GET_KERNEL_VIRTUAL_ADDRESS, uvm_test_get_kernel_virtual_address); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMA_ALLOC_FREE, uvm_test_pma_alloc_free); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_ALLOC_FREE_ROOT, uvm_test_pmm_alloc_free_root); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_INJECT_PMA_EVICT_ERROR, uvm_test_pmm_inject_pma_evict_error); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_ACCESS_COUNTERS_ENABLED_BY_DEFAULT, uvm_test_access_counters_enabled_by_default); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RECONFIGURE_ACCESS_COUNTERS, uvm_test_reconfigure_access_counters); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RESET_ACCESS_COUNTERS, uvm_test_reset_access_counters); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_SET_IGNORE_ACCESS_COUNTERS, uvm_test_set_ignore_access_counters); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_CHECK_CHANNEL_VA_SPACE, uvm_test_check_channel_va_space); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_ENABLE_NVLINK_PEER_ACCESS, uvm_test_enable_nvlink_peer_access); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_DISABLE_NVLINK_PEER_ACCESS, uvm_test_disable_nvlink_peer_access); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_GET_PAGE_THRASHING_POLICY, uvm_test_get_page_thrashing_policy); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_SET_PAGE_THRASHING_POLICY, uvm_test_set_page_thrashing_policy); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_SYSMEM, uvm_test_pmm_sysmem); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_REVERSE_MAP, uvm_test_pmm_reverse_map); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_INDIRECT_PEERS, uvm_test_pmm_indirect_peers); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_SPACE_MM_RETAIN, uvm_test_va_space_mm_retain); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_SPACE_MM_DELAY_SHUTDOWN, uvm_test_va_space_mm_delay_shutdown); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_CHUNK_WITH_ELEVATED_PAGE, uvm_test_pmm_chunk_with_elevated_page); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_SPACE_INJECT_ERROR, uvm_test_va_space_inject_error); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_GET_GPU_TIME, uvm_test_get_gpu_time); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_RELEASE_FREE_ROOT_CHUNKS, uvm_test_pmm_release_free_root_chunks); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_DRAIN_REPLAYABLE_FAULTS, uvm_test_drain_replayable_faults); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMA_GET_BATCH_SIZE, uvm_test_pma_get_batch_size); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_PMM_QUERY_PMA_STATS, uvm_test_pmm_query_pma_stats); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_NUMA_GET_CLOSEST_CPU_NODE_TO_GPU, uvm_test_numa_get_closest_cpu_node_to_gpu); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_NUMA_CHECK_AFFINITY, uvm_test_numa_check_affinity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_SPACE_ADD_DUMMY_THREAD_CONTEXTS, uvm_test_va_space_add_dummy_thread_contexts); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_SPACE_REMOVE_DUMMY_THREAD_CONTEXTS, uvm_test_va_space_remove_dummy_thread_contexts); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_THREAD_CONTEXT_SANITY, uvm_test_thread_context_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_THREAD_CONTEXT_PERF, uvm_test_thread_context_perf); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_GET_PAGEABLE_MEM_ACCESS_TYPE, uvm_test_get_pageable_mem_access_type); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_TOOLS_FLUSH_REPLAY_EVENTS, uvm_test_tools_flush_replay_events); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_REGISTER_UNLOAD_STATE_BUFFER, uvm_test_register_unload_state_buffer); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RB_TREE_DIRECTED, uvm_test_rb_tree_directed); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_RB_TREE_RANDOM, uvm_test_rb_tree_random); UVM_ROUTE_CMD_STACK_NO_INIT_CHECK(UVM_TEST_GET_USER_SPACE_END_ADDRESS, uvm_test_get_user_space_end_address); UVM_ROUTE_CMD_STACK_NO_INIT_CHECK(UVM_TEST_GET_CPU_CHUNK_ALLOC_SIZES, uvm_test_get_cpu_chunk_allocation_sizes); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_HMM_SANITY, uvm_test_hmm_sanity); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_VA_RANGE_INJECT_ADD_GPU_VA_SPACE_ERROR, uvm_test_va_range_inject_add_gpu_va_space_error); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_DESTROY_GPU_VA_SPACE_DELAY, uvm_test_destroy_gpu_va_space_delay); UVM_ROUTE_CMD_STACK_NO_INIT_CHECK(UVM_TEST_CGROUP_ACCOUNTING_SUPPORTED, uvm_test_cgroup_accounting_supported); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_HMM_INIT, uvm_test_hmm_init); UVM_ROUTE_CMD_STACK_INIT_CHECK(UVM_TEST_SPLIT_INVALIDATE_DELAY, uvm_test_split_invalidate_delay); } return -EINVAL; }