tree: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master head: e2f86c02fdc96ca29ced53221a3cbf50aa6f8b49 commit: 516198d317d81f33839ca850e83f6717b0d80e80 [7324/8766] drm/i915: audit bo->resource usage v3 config: i386-randconfig-s041-20230123 (https://download.01.org/0day-ci/archive/20230128/202301282050.PL1UGurJ-lkp@xxxxxxxxx/config) compiler: gcc-11 (Debian 11.3.0-8) 11.3.0 reproduce: # apt-get install sparse # sparse version: v0.6.4-39-gce1a6720-dirty # https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=516198d317d81f33839ca850e83f6717b0d80e80 git remote add linux-next https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git git fetch --no-tags linux-next master git checkout 516198d317d81f33839ca850e83f6717b0d80e80 # save the config file mkdir build_dir && cp config build_dir/.config make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=i386 olddefconfig make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' O=build_dir ARCH=i386 SHELL=/bin/bash drivers/gpu/drm/i915/ If you fix the issue, kindly add following tag where applicable | Reported-by: kernel test robot <lkp@xxxxxxxxx> sparse warnings: (new ones prefixed by >>) >> drivers/gpu/drm/i915/gem/i915_gem_ttm.c:1066:21: sparse: sparse: incorrect type in assignment (different base types) @@ expected restricted vm_fault_t [assigned] [usertype] ret @@ got int @@ drivers/gpu/drm/i915/gem/i915_gem_ttm.c:1066:21: sparse: expected restricted vm_fault_t [assigned] [usertype] ret drivers/gpu/drm/i915/gem/i915_gem_ttm.c:1066:21: sparse: got int vim +1066 drivers/gpu/drm/i915/gem/i915_gem_ttm.c 1027 1028 static vm_fault_t vm_fault_ttm(struct vm_fault *vmf) 1029 { 1030 struct vm_area_struct *area = vmf->vma; 1031 struct ttm_buffer_object *bo = area->vm_private_data; 1032 struct drm_device *dev = bo->base.dev; 1033 struct drm_i915_gem_object *obj = i915_ttm_to_gem(bo); 1034 intel_wakeref_t wakeref = 0; 1035 vm_fault_t ret; 1036 int idx; 1037 1038 /* Sanity check that we allow writing into this object */ 1039 if (unlikely(i915_gem_object_is_readonly(obj) && 1040 area->vm_flags & VM_WRITE)) 1041 return VM_FAULT_SIGBUS; 1042 1043 ret = ttm_bo_vm_reserve(bo, vmf); 1044 if (ret) 1045 return ret; 1046 1047 if (obj->mm.madv != I915_MADV_WILLNEED) { 1048 dma_resv_unlock(bo->base.resv); 1049 return VM_FAULT_SIGBUS; 1050 } 1051 1052 /* 1053 * This must be swapped out with shmem ttm_tt (pipeline-gutting). 1054 * Calling ttm_bo_validate() here with TTM_PL_SYSTEM should only go as 1055 * far as far doing a ttm_bo_move_null(), which should skip all the 1056 * other junk. 1057 */ 1058 if (!bo->resource) { 1059 struct ttm_operation_ctx ctx = { 1060 .interruptible = true, 1061 .no_wait_gpu = true, /* should be idle already */ 1062 }; 1063 1064 GEM_BUG_ON(!bo->ttm || !(bo->ttm->page_flags & TTM_TT_FLAG_SWAPPED)); 1065 > 1066 ret = ttm_bo_validate(bo, i915_ttm_sys_placement(), &ctx); 1067 if (ret) { 1068 dma_resv_unlock(bo->base.resv); 1069 return VM_FAULT_SIGBUS; 1070 } 1071 } else if (!i915_ttm_resource_mappable(bo->resource)) { 1072 int err = -ENODEV; 1073 int i; 1074 1075 for (i = 0; i < obj->mm.n_placements; i++) { 1076 struct intel_memory_region *mr = obj->mm.placements[i]; 1077 unsigned int flags; 1078 1079 if (!mr->io_size && mr->type != INTEL_MEMORY_SYSTEM) 1080 continue; 1081 1082 flags = obj->flags; 1083 flags &= ~I915_BO_ALLOC_GPU_ONLY; 1084 err = __i915_ttm_migrate(obj, mr, flags); 1085 if (!err) 1086 break; 1087 } 1088 1089 if (err) { 1090 drm_dbg(dev, "Unable to make resource CPU accessible(err = %pe)\n", 1091 ERR_PTR(err)); 1092 dma_resv_unlock(bo->base.resv); 1093 ret = VM_FAULT_SIGBUS; 1094 goto out_rpm; 1095 } 1096 } 1097 1098 if (i915_ttm_cpu_maps_iomem(bo->resource)) 1099 wakeref = intel_runtime_pm_get(&to_i915(obj->base.dev)->runtime_pm); 1100 1101 if (drm_dev_enter(dev, &idx)) { 1102 ret = ttm_bo_vm_fault_reserved(vmf, vmf->vma->vm_page_prot, 1103 TTM_BO_VM_NUM_PREFAULT); 1104 drm_dev_exit(idx); 1105 } else { 1106 ret = ttm_bo_vm_dummy_page(vmf, vmf->vma->vm_page_prot); 1107 } 1108 1109 if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) 1110 goto out_rpm; 1111 1112 /* 1113 * ttm_bo_vm_reserve() already has dma_resv_lock. 1114 * userfault_count is protected by dma_resv lock and rpm wakeref. 1115 */ 1116 if (ret == VM_FAULT_NOPAGE && wakeref && !obj->userfault_count) { 1117 obj->userfault_count = 1; 1118 spin_lock(&to_i915(obj->base.dev)->runtime_pm.lmem_userfault_lock); 1119 list_add(&obj->userfault_link, &to_i915(obj->base.dev)->runtime_pm.lmem_userfault_list); 1120 spin_unlock(&to_i915(obj->base.dev)->runtime_pm.lmem_userfault_lock); 1121 1122 GEM_WARN_ON(!i915_ttm_cpu_maps_iomem(bo->resource)); 1123 } 1124 1125 if (wakeref & CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND) 1126 intel_wakeref_auto(&to_i915(obj->base.dev)->runtime_pm.userfault_wakeref, 1127 msecs_to_jiffies_timeout(CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND)); 1128 1129 i915_ttm_adjust_lru(obj); 1130 1131 dma_resv_unlock(bo->base.resv); 1132 1133 out_rpm: 1134 if (wakeref) 1135 intel_runtime_pm_put(&to_i915(obj->base.dev)->runtime_pm, wakeref); 1136 1137 return ret; 1138 } 1139 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests
