diff --git a/intern/cycles/kernel/device/hiprt/common.h b/intern/cycles/kernel/device/hiprt/common.h
index 8a522861f42..d574f5ed08f 100644
--- a/intern/cycles/kernel/device/hiprt/common.h
+++ b/intern/cycles/kernel/device/hiprt/common.h
@@ -104,11 +104,11 @@ ccl_device_inline bool curve_custom_intersect(const hiprtRay &ray,
 
   const int object_id = kernel_data_fetch(user_instance_id, hit.instanceID);
 
-  /* data_offset.x: where the data (prim id, type )for the geometry of the current object begins
+  /* `data_offset.x`: where the data (prim id, type )for the geometry of the current object begins
    * the prim_id that is in hiprtHit hit is local to the particular geometry so we add the above
-   * ofstream to map prim id in hiprtHit to the one compatible to what next stage expects
-   * data_offset.y: the offset that has to be added to a local primitive to get the global
-   * primitive id = kernel_data_fetch(object_prim_offset, object_id); */
+   * `ofstream` to map prim id in hiprtHit to the one compatible to what next stage expects
+   * `data_offset.y`: the offset that has to be added to a local primitive to get the global
+   * `primitive id = kernel_data_fetch(object_prim_offset, object_id);` */
   const int2 data_offset = kernel_data_fetch(custom_prim_info_offset, object_id);
 
   const int prim_offset = data_offset.y;
diff --git a/intern/cycles/subd/split.cpp b/intern/cycles/subd/split.cpp
index a6ec020995f..bb8a494d4aa 100644
--- a/intern/cycles/subd/split.cpp
+++ b/intern/cycles/subd/split.cpp
@@ -507,14 +507,14 @@ void DiagSplit::split_ngon(const Mesh::SubdFace &face,
   for (int corner = 0; corner < face.num_corners; corner++) {
     const Patch *patch = (const Patch *)(((char *)patches) + (corner * patches_byte_stride));
 
-    /*         vprev         .
+    /*         v_prev        .
      *           .           .
      *           .  edge_u1  .
      *          v01 ←------ v11 . . .
      *           |           ↑
      *  edge_v0  |           | edge_v1
      *           ↓           |
-     *          v00 ------→ v10 . . vnext
+     *          v00 ------→ v10 . . v_next
      *              edge_u0
      */
     SubPatch::Edge &edge_v0 = edges[mod(corner + face.num_corners - 1, face.num_corners)];
@@ -543,11 +543,11 @@ void DiagSplit::split_ngon(const Mesh::SubdFace &face,
 
 void DiagSplit::split_patches(const Patch *patches, const size_t patches_byte_stride)
 {
-  // TODO: reuse edge factor vertex position computations
-  // TODO: support not splitting n-gons if not needed
-  // TODO: multithreading
+  /* TODO: reuse edge factor vertex position computations. */
+  /* TODO: support not splitting n-gons if not needed. */
+  /* TODO: multi-threading. */
 
-  /* Keep base mesh vertices, create new triangels. */
+  /* Keep base mesh vertices, create new triangles. */
   num_verts = params.mesh->get_num_subd_base_verts();
   num_triangles = 0;
 
diff --git a/release/release_notes/bug_fixes_per_major_release.py b/release/release_notes/bug_fixes_per_major_release.py
index 44e5d0d466f..6ecee02445a 100644
--- a/release/release_notes/bug_fixes_per_major_release.py
+++ b/release/release_notes/bug_fixes_per_major_release.py
@@ -541,7 +541,7 @@ def version_extraction(report_body: str) -> tuple[list[str], list[str]]:
         if lower_line.startswith('work'):
             # Use `work` to be able to detect both "worked" and "working".
             if (not example_in_line) and not ("brok" in lower_line):
-                # Don't add the line to the working_lines if it contains the letters "brok"
+                # Don't add the line to the working_lines if it contains the letters `brok`.
                 # because it means the user probably wrote something like "Worked: It was also broken in X.X"
                 # which lead to incorrect information.
                 working_lines += f'{line}\n'
diff --git a/source/blender/animrig/intern/action.cc b/source/blender/animrig/intern/action.cc
index bf11a83d230..f5b9545586f 100644
--- a/source/blender/animrig/intern/action.cc
+++ b/source/blender/animrig/intern/action.cc
@@ -3143,7 +3143,7 @@ void move_slot(Main &bmain, Slot &source_slot, Action &from_action, Action &to_a
        * the animated ID.
        *
        * This line was added here for now, to fix #136388 with minimal impact on
-       * other code, so that the fix can be easily backported to Blender 4.4. */
+       * other code, so that the fix can be easily back-ported to Blender 4.4. */
       DEG_id_tag_update(user, ID_RECALC_ANIMATION);
       return true;
     };
diff --git a/source/blender/animrig/intern/versioning.cc b/source/blender/animrig/intern/versioning.cc
index 2620eba016c..18e366f16a2 100644
--- a/source/blender/animrig/intern/versioning.cc
+++ b/source/blender/animrig/intern/versioning.cc
@@ -258,7 +258,7 @@ void convert_legacy_action_assignments(Main &bmain, ReportList *reports)
 
   /* Note that the code below does not remove the `action_assignment_needs_slot` tag. One ID can
    * use multiple Actions (via NLA, Action constraints, etc.); if one of those Action is an ancient
-   * one one from before 2.50 (just to name one example case) this ID may needs to be re-visited
+   * one from before 2.50 (just to name one example case) this ID may needs to be re-visited
    * after those were versioned. Rather than trying to figure out if re-visiting is necessary, this
    * function is safe to call multiple times, and all that's lost is a little bit of CPU time. */
 
diff --git a/source/blender/blenkernel/BKE_fcurve.hh b/source/blender/blenkernel/BKE_fcurve.hh
index 99f529880ab..281faf7121a 100644
--- a/source/blender/blenkernel/BKE_fcurve.hh
+++ b/source/blender/blenkernel/BKE_fcurve.hh
@@ -483,7 +483,7 @@ bool BKE_fcurve_bezt_subdivide_handles(BezTriple *bezt,
  * \param new_totvert: new number of elements in the FCurve's `bezt` array.
  *
  * \note When increasing the size of the array, newly added elements (that is, in the
- * [old_totvert..new_totvert) interval) are zero-initialized.
+ * [old_totvert..new_totvert] interval) are zero-initialized.
  */
 void BKE_fcurve_bezt_resize(FCurve *fcu, int new_totvert);
 
@@ -591,7 +591,7 @@ bool test_time_fcurve(FCurve *fcu);
  * than the horizontal distance between (v1-v4).
  * This is to prevent curve loops.
  *
- * This function is very similar to BKE_curve_correct_bezpart(), but allows a steeper tangent for
+ * This function is very similar to #BKE_curve_correct_bezpart(), but allows a steeper tangent for
  * more snappy animations. This is not desired for other areas in which curves are used, though.
  */
 void BKE_fcurve_correct_bezpart(const float v1[2], float v2[2], float v3[2], const float v4[2]);
diff --git a/source/blender/blenkernel/BKE_grease_pencil.hh b/source/blender/blenkernel/BKE_grease_pencil.hh
index bba9455273b..704ef2ebb1a 100644
--- a/source/blender/blenkernel/BKE_grease_pencil.hh
+++ b/source/blender/blenkernel/BKE_grease_pencil.hh
@@ -421,15 +421,17 @@ class LayerRuntime {
    */
   Vector<LayerMask> masks_;
 
-  /* Runtime data used for frame transformations. */
+  /** Runtime data used for frame transformations. */
   LayerTransformData trans_data_;
 
-  /* Whether this layer's visibility is animated (via the ".hide" RNA property). This is only set
-   * when creating a copy of of the owning GreasePencil ID for the depsgraph evaluation. */
+  /**
+   * Whether this layer's visibility is animated (via the ".hide" RNA property). This is only set
+   * when creating a copy of the owning GreasePencil ID for the depsgraph evaluation.
+   */
   bool is_visibility_animated_;
 
  public:
-  /* Reset all runtime data. */
+  /** Reset all runtime data. */
   void clear();
 };
 
@@ -642,8 +644,10 @@ class LayerGroupRuntime {
    */
   mutable Vector<LayerGroup *> layer_group_cache_;
 
-  /* Whether this layer's visibility is animated (via the ".hide" RNA property). This is only set
-   * when creating a copy of of the owning GreasePencil ID for the depsgraph evaluation. */
+  /**
+   * Whether this layer's visibility is animated (via the ".hide" RNA property). This is only set
+   * when creating a copy of the owning GreasePencil ID for the depsgraph evaluation.
+   */
   bool is_visibility_animated_;
 };
 
diff --git a/source/blender/blenkernel/BKE_mesh_mapping.hh b/source/blender/blenkernel/BKE_mesh_mapping.hh
index 9cb9d09b3a2..b756bc602c0 100644
--- a/source/blender/blenkernel/BKE_mesh_mapping.hh
+++ b/source/blender/blenkernel/BKE_mesh_mapping.hh
@@ -253,7 +253,7 @@ bool BKE_mesh_calc_islands_loop_face_uvmap(float (*vert_positions)[3],
  * \param sharp_faces: Optional (possibly empty) span.
  * \param r_totgroup: The total number of groups, 1 or more.
  * \return Face aligned array of group index values, starting at 1 (0 being used as 'invalid'
- * flag). Note that it's the callers's responsibility to MEM_freeN the returned array.
+ * flag). Note that it's the callers responsibility to MEM_freeN the returned array.
  */
 int *BKE_mesh_calc_smoothgroups(int edges_num,
                                 blender::OffsetIndices<int> faces,
@@ -262,7 +262,7 @@ int *BKE_mesh_calc_smoothgroups(int edges_num,
                                 blender::Span<bool> sharp_faces,
                                 int *r_totgroup);
 /**
- * Same as #BKE_mesh_calc_smoothgroups, but use bitflags instead of increasing numbers for each
+ * Same as #BKE_mesh_calc_smoothgroups, but use bit-flags instead of increasing numbers for each
  * group.
  *
  * This means that the same value (bit) can be re-used for different groups, as long as they are
@@ -281,7 +281,7 @@ int *BKE_mesh_calc_smoothgroups(int edges_num,
  * \param sharp_faces: Optional (possibly empty) span.
  * \param r_totgroup: The total number of groups, 1 or more.
  * \return Face aligned array of group bit-flags values (i.e. always powers of 2), starting at 1 (0
- * being used as 'invalid' flag). Note that it's the callers's responsibility to MEM_freeN the
+ * being used as 'invalid' flag). Note that it's the callers responsibility to MEM_freeN the
  * returned array.
  */
 int *BKE_mesh_calc_smoothgroups_bitflags(int edges_num,
diff --git a/source/blender/blenkernel/intern/collision.cc b/source/blender/blenkernel/intern/collision.cc
index e9735bacca1..98cd4ce3981 100644
--- a/source/blender/blenkernel/intern/collision.cc
+++ b/source/blender/blenkernel/intern/collision.cc
@@ -1454,7 +1454,7 @@ static int cloth_bvh_objcollisions_resolve(ClothModifierData *clmd,
     /* Apply impulses in parallel. */
     if (result) {
       for (i = 0; i < mvert_num; i++) {
-        // calculate "velocities" (just xnew = xold + v; no dt in v)
+        /* Calculate "velocities" (just `xnew = xold + v`; no `dt` in `v`). */
         if (verts[i].impulse_count) {
           add_v3_v3(verts[i].tv, verts[i].impulse);
           add_v3_v3(verts[i].dcvel, verts[i].impulse);
diff --git a/source/blender/blenlib/BLI_sys_types.h b/source/blender/blenlib/BLI_sys_types.h
index ebc7626309b..29002cee72f 100644
--- a/source/blender/blenlib/BLI_sys_types.h
+++ b/source/blender/blenlib/BLI_sys_types.h
@@ -47,7 +47,7 @@ typedef uint64_t u_int64_t;
 #  include <stdint.h>     // IWYU pragma: export
 #  include <sys/types.h>  // IWYU pragma: export
 
-#endif /* ifdef platform for types */
+#endif /* `ifdef` platform for types. */
 
 #include <stdbool.h>  // IWYU pragma: export
 #include <stddef.h>   // IWYU pragma: export
diff --git a/source/blender/blenloader/intern/versioning_270.cc b/source/blender/blenloader/intern/versioning_270.cc
index 73058f41b40..129be23dfd5 100644
--- a/source/blender/blenloader/intern/versioning_270.cc
+++ b/source/blender/blenloader/intern/versioning_270.cc
@@ -1024,9 +1024,8 @@ void blo_do_versions_270(FileData *fd, Library * /*lib*/, Main *bmain)
     LISTBASE_FOREACH (bGPdata *, gpd, &bmain->gpencils) {
       bool enabled = false;
 
-      /* Ensure that the datablock's onion-skinning toggle flag
-       * stays in sync with the status of the actual layers
-       */
+      /* Ensure that the data-block's onion-skinning toggle flag
+       * stays in sync with the status of the actual layers. */
       LISTBASE_FOREACH (bGPDlayer *, gpl, &gpd->layers) {
         if (gpl->flag & GP_LAYER_ONIONSKIN) {
           enabled = true;
diff --git a/source/blender/draw/engines/gpencil/gpencil_engine_private.hh b/source/blender/draw/engines/gpencil/gpencil_engine_private.hh
index 2849569a0ee..086c7fc9d4a 100644
--- a/source/blender/draw/engines/gpencil/gpencil_engine_private.hh
+++ b/source/blender/draw/engines/gpencil/gpencil_engine_private.hh
@@ -140,7 +140,7 @@ struct Instance final : public DrawEngine {
   Texture snapshot_depth_tx = {"snapshot_depth_tx"};
   Texture snapshot_color_tx = {"snapshot_color_tx"};
   Texture snapshot_reveal_tx = {"snapshot_reveal_tx"};
-  /* Textures used by Antialiasing. */
+  /* Textures used by Anti-aliasing. */
   Texture smaa_area_tx = {"smaa_area_tx"};
   Texture smaa_search_tx = {"smaa_search_tx"};
 
diff --git a/source/blender/draw/intern/draw_cache_extract.hh b/source/blender/draw/intern/draw_cache_extract.hh
index 092de161766..3a5b9a2f472 100644
--- a/source/blender/draw/intern/draw_cache_extract.hh
+++ b/source/blender/draw/intern/draw_cache_extract.hh
@@ -134,7 +134,7 @@ enum class IBOType : int8_t {
 
 struct MeshBufferList {
   /* Though using maps here may add some overhead compared to just indexed arrays, it's a bit more
-   * conventient currently, because the "buffer exists" test is very clear, it's just whether the
+   * convenient currently, because the "buffer exists" test is very clear, it's just whether the
    * map contains it (e.g. compared to "buffer is allocated but not filled with data"). The
    * sparseness *may* be useful for reducing memory usage when only few buffers are used. */
 
diff --git a/source/blender/editors/interface/interface_icons.cc b/source/blender/editors/interface/interface_icons.cc
index 2a009122779..4f31cbfeacd 100644
--- a/source/blender/editors/interface/interface_icons.cc
+++ b/source/blender/editors/interface/interface_icons.cc
@@ -379,7 +379,7 @@ static void vicon_handletype_auto_clamp_draw(
 static void icon_node_socket_draw(int socket_type, int x, int y, int w, int h, float /*alpha*/)
 {
   /* Factor to account for the draw function of the node socket being based on the widget unit,
-   * which is 10 px by default, which differs from icons. */
+   * which is 10 pixels by default, which differs from icons. */
   constexpr float size_factor = 10.0f / float(ICON_DEFAULT_WIDTH);
 
   const float socket_radius = w * 0.5f * size_factor;
diff --git a/source/blender/editors/space_sequencer/sequencer_edit.cc b/source/blender/editors/space_sequencer/sequencer_edit.cc
index 4b2213bef96..9d047cce145 100644
--- a/source/blender/editors/space_sequencer/sequencer_edit.cc
+++ b/source/blender/editors/space_sequencer/sequencer_edit.cc
@@ -1221,13 +1221,13 @@ bool strip_effect_get_new_inputs(Scene *scene,
   }
 
   blender::VectorSet<Strip *> new_inputs = seq::query_selected_strips(ed->seqbasep);
-  // Ignore sound strips for now (avoids unnecessary errors when connected strips are
-  // selected together, and the intent to operate on strips with video content is clear).
+  /* Ignore sound strips for now (avoids unnecessary errors when connected strips are
+   * selected together, and the intent to operate on strips with video content is clear). */
   new_inputs.remove_if([&](Strip *strip) { return strip->type == STRIP_TYPE_SOUND_RAM; });
 
   if (ignore_active) {
-    // If `ignore_active` is true, this function is being called from the reassign inputs
-    // operator, meaning the active strip must be the effect strip to reassign.
+    /* If `ignore_active` is true, this function is being called from the reassign inputs
+     * operator, meaning the active strip must be the effect strip to reassign. */
     Strip *active_strip = seq::select_active_get(scene);
     new_inputs.remove_if([&](Strip *strip) { return strip == active_strip; });
   }
diff --git a/source/blender/gpu/opengl/gl_backend.cc b/source/blender/gpu/opengl/gl_backend.cc
index 2fff8f717ec..62bb3acc3d4 100644
--- a/source/blender/gpu/opengl/gl_backend.cc
+++ b/source/blender/gpu/opengl/gl_backend.cc
@@ -397,7 +397,7 @@ static void detect_workarounds()
     }
   }
   /* See #132968: Legacy AMD drivers do not accept a hash after the line number and results into
-   * undefined behaviour. Users have reported that the issue can go away after doing a clean
+   * undefined behavior. Users have reported that the issue can go away after doing a clean
    * install of the driver.
    */
   if (GPU_type_matches(GPU_DEVICE_ATI, GPU_OS_ANY, GPU_DRIVER_OFFICIAL)) {
diff --git a/source/blender/gpu/vulkan/render_graph/vk_command_builder.cc b/source/blender/gpu/vulkan/render_graph/vk_command_builder.cc
index 59dfbe69583..b788c9e178b 100644
--- a/source/blender/gpu/vulkan/render_graph/vk_command_builder.cc
+++ b/source/blender/gpu/vulkan/render_graph/vk_command_builder.cc
@@ -296,7 +296,7 @@ void VKCommandBuilder::groups_build_commands(VKRenderGraph &render_graph,
                 << group_node_handles.last() << ", node_pre_barrier=(" << to_string_barrier(barrier)
                 << ")\n";
 #endif
-        // TODO: Barrier should already contain the changes for local read.
+        /* TODO: Barrier should already contain the changes for local read. */
         send_pipeline_barriers(command_buffer, barrier, true);
       }
 
diff --git a/source/blender/gpu/vulkan/vk_backend.cc b/source/blender/gpu/vulkan/vk_backend.cc
index 70b524bc457..42a38784932 100644
--- a/source/blender/gpu/vulkan/vk_backend.cc
+++ b/source/blender/gpu/vulkan/vk_backend.cc
@@ -183,7 +183,7 @@ bool VKBackend::is_supported()
     return false;
   }
 
-  // go over all the devices
+  /* Go over all the devices. */
   uint32_t physical_devices_count = 0;
   vkEnumeratePhysicalDevices(vk_instance, &physical_devices_count, nullptr);
   Array<VkPhysicalDevice> vk_physical_devices(physical_devices_count);
diff --git a/source/blender/io/usd/intern/usd_hook.cc b/source/blender/io/usd/intern/usd_hook.cc
index ff17b2e2b2d..eccb3663b57 100644
--- a/source/blender/io/usd/intern/usd_hook.cc
+++ b/source/blender/io/usd/intern/usd_hook.cc
@@ -533,8 +533,8 @@ class MaterialImportPollInvoker : public USDHookInvoker {
 
   void call_hook(PyObject *hook_obj) override
   {
-    // If we already know that one of the registered hook classes can import the material
-    // because it returned true in a previous invocation of the callback, we skip the call.
+    /* If we already know that one of the registered hook classes can import the material
+     * because it returned true in a previous invocation of the callback, we skip the call. */
     if (!result_) {
       result_ = python::call_method<bool>(
           hook_obj, function_name(), REF(hook_context_), usd_material_);
diff --git a/source/blender/windowmanager/WM_types.hh b/source/blender/windowmanager/WM_types.hh
index 5d01f29104f..bbcd92262b4 100644
--- a/source/blender/windowmanager/WM_types.hh
+++ b/source/blender/windowmanager/WM_types.hh
@@ -284,7 +284,7 @@ enum {
    * to be a Hyper modifier, see !136340.
    *
    * Note that this is currently only supported on Wayland & X11
-   * but could could be supported on other platforms if desired.
+   * but could be supported on other platforms if desired.
    */
   KM_HYPER = (1 << 4),
 };