Task scheduler: Get rid of extended version of parallel range callback

Wrap all arguments into TLS type of argument. Avoids some branching and also
makes it easier to extend things in the future.

33 files changed, 268 insertions, 200 deletions

diff --git a/source/blender/blenkernel/intern/CCGSubSurf_legacy.c b/source/blender/blenkernel/intern/CCGSubSurf_legacy.c
index f68d1a2697c..8b3ea249396 100644
--- a/source/blender/blenkernel/intern/CCGSubSurf_legacy.c
+++ b/source/blender/blenkernel/intern/CCGSubSurf_legacy.c
@@ -136,7 +136,8 @@ typedef struct CCGSubSurfCalcSubdivData {
 	int curLvl;
 } CCGSubSurfCalcSubdivData;
 
-static void ccgSubSurf__calcVertNormals_faces_accumulate_cb(void *userdata, int ptrIdx)
+static void ccgSubSurf__calcVertNormals_faces_accumulate_cb(void *userdata, int ptrIdx,
+                                                            const ParallelRangeTLS *UNUSED(tls))
 {
 	CCGSubSurfCalcSubdivData *data = userdata;
 
@@ -227,7 +228,8 @@ static void ccgSubSurf__calcVertNormals_faces_accumulate_cb(void *userdata, int
 	}
 }
 
-static void ccgSubSurf__calcVertNormals_faces_finalize_cb(void *userdata, int ptrIdx)
+static void ccgSubSurf__calcVertNormals_faces_finalize_cb(void *userdata, int ptrIdx,
+                                                          const ParallelRangeTLS *UNUSED(tls))
 {
 	CCGSubSurfCalcSubdivData *data = userdata;
 
@@ -265,7 +267,8 @@ static void ccgSubSurf__calcVertNormals_faces_finalize_cb(void *userdata, int pt
 	}
 }
 
-static void ccgSubSurf__calcVertNormals_edges_accumulate_cb(void *userdata, int ptrIdx)
+static void ccgSubSurf__calcVertNormals_edges_accumulate_cb(void *userdata, int ptrIdx,
+                                                            const ParallelRangeTLS *UNUSED(tls))
 {
 	CCGSubSurfCalcSubdivData *data = userdata;
 
@@ -396,7 +399,8 @@ static void ccgSubSurf__calcVertNormals(CCGSubSurf *ss,
 }
 
 
-static void ccgSubSurf__calcSubdivLevel_interior_faces_edges_midpoints_cb(void *userdata, int ptrIdx)
+static void ccgSubSurf__calcSubdivLevel_interior_faces_edges_midpoints_cb(void *userdata, int ptrIdx,
+                                                                          const ParallelRangeTLS *UNUSED(tls))
 {
 	CCGSubSurfCalcSubdivData *data = userdata;
 
@@ -483,7 +487,8 @@ static void ccgSubSurf__calcSubdivLevel_interior_faces_edges_midpoints_cb(void *
 	}
 }
 
-static void ccgSubSurf__calcSubdivLevel_interior_faces_edges_centerpoints_shift_cb(void *userdata, int ptrIdx)
+static void ccgSubSurf__calcSubdivLevel_interior_faces_edges_centerpoints_shift_cb(void *userdata, int ptrIdx,
+                                                                                   const ParallelRangeTLS *UNUSED(tls))
 {
 	CCGSubSurfCalcSubdivData *data = userdata;
 
@@ -588,7 +593,8 @@ static void ccgSubSurf__calcSubdivLevel_interior_faces_edges_centerpoints_shift_
 	}
 }
 
-static void ccgSubSurf__calcSubdivLevel_verts_copydata_cb(void *userdata, int ptrIdx)
+static void ccgSubSurf__calcSubdivLevel_verts_copydata_cb(void *userdata, int ptrIdx,
+                                                          const ParallelRangeTLS *UNUSED(tls))
 {
 	CCGSubSurfCalcSubdivData *data = userdata;
 
diff --git a/source/blender/blenkernel/intern/colortools.c b/source/blender/blenkernel/intern/colortools.c
index 310255a15c1..67153564a0f 100644
--- a/source/blender/blenkernel/intern/colortools.c
+++ b/source/blender/blenkernel/intern/colortools.c
@@ -1165,7 +1165,7 @@ typedef struct ScopesUpdateDataChunk {
 	float min[3], max[3];
 } ScopesUpdateDataChunk;
 
-static void scopes_update_cb(void *userdata, void *userdata_chunk, const int y, const int UNUSED(threadid))
+static void scopes_update_cb(void *userdata, const int y, const ParallelRangeTLS *tls)
 {
 	const ScopesUpdateData *data = userdata;
 
@@ -1175,7 +1175,7 @@ static void scopes_update_cb(void *userdata, void *userdata_chunk, const int y,
 	const unsigned char *display_buffer = data->display_buffer;
 	const int ycc_mode = data->ycc_mode;
 
-	ScopesUpdateDataChunk *data_chunk = userdata_chunk;
+	ScopesUpdateDataChunk *data_chunk = tls->userdata_chunk;
 	unsigned int *bin_lum = data_chunk->bin_lum;
 	unsigned int *bin_r = data_chunk->bin_r;
 	unsigned int *bin_g = data_chunk->bin_g;
diff --git a/source/blender/blenkernel/intern/dynamicpaint.c b/source/blender/blenkernel/intern/dynamicpaint.c
index b44ee21432c..81a721e77b9 100644
--- a/source/blender/blenkernel/intern/dynamicpaint.c
+++ b/source/blender/blenkernel/intern/dynamicpaint.c
@@ -611,11 +611,11 @@ static void freeGrid(PaintSurfaceData *data)
 	bData->grid = NULL;
 }
 
-static void grid_bound_insert_cb_ex(void *userdata, void *userdata_chunk, const int i, const int UNUSED(thread_id))
+static void grid_bound_insert_cb_ex(void *userdata, const int i, const ParallelRangeTLS *tls)
 {
 	PaintBakeData *bData = userdata;
 
-	Bounds3D *grid_bound = userdata_chunk;
+	Bounds3D *grid_bound = tls->userdata_chunk;
 
 	boundInsert(grid_bound, bData->realCoord[bData->s_pos[i]].v);
 }
@@ -631,12 +631,12 @@ static void grid_bound_insert_finalize(void *userdata, void *userdata_chunk)
 	boundInsert(&grid->grid_bounds, grid_bound->max);
 }
 
-static void grid_cell_points_cb_ex(void *userdata, void *userdata_chunk, const int i, const int UNUSED(thread_id))
+static void grid_cell_points_cb_ex(void *userdata, const int i, const ParallelRangeTLS *tls)
 {
 	PaintBakeData *bData = userdata;
 	VolumeGrid *grid = bData->grid;
 	int *temp_t_index = grid->temp_t_index;
-	int *s_num = userdata_chunk;
+	int *s_num = tls->userdata_chunk;
 
 	int co[3];
 
@@ -664,7 +664,7 @@ static void grid_cell_points_finalize(void *userdata, void *userdata_chunk)
 	}
 }
 
-static void grid_cell_bounds_cb(void *userdata, const int x)
+static void grid_cell_bounds_cb(void *userdata, const int x, const ParallelRangeTLS *UNUSED(tls))
 {
 	PaintBakeData *bData = userdata;
 	VolumeGrid *grid = bData->grid;
@@ -1397,7 +1397,8 @@ typedef struct DynamicPaintSetInitColorData {
 	const bool scene_color_manage;
 } DynamicPaintSetInitColorData;
 
-static void dynamic_paint_set_init_color_tex_to_vcol_cb(void *userdata, const int i)
+static void dynamic_paint_set_init_color_tex_to_vcol_cb(void *userdata, const int i,
+                                                       const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintSetInitColorData *data = userdata;
 
@@ -1431,7 +1432,8 @@ static void dynamic_paint_set_init_color_tex_to_vcol_cb(void *userdata, const in
 	}
 }
 
-static void dynamic_paint_set_init_color_tex_to_imseq_cb(void *userdata, const int i)
+static void dynamic_paint_set_init_color_tex_to_imseq_cb(void *userdata, const int i,
+                                                         const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintSetInitColorData *data = userdata;
 
@@ -1469,7 +1471,8 @@ static void dynamic_paint_set_init_color_tex_to_imseq_cb(void *userdata, const i
 	pPoint[i].color[3] = texres.tin;
 }
 
-static void dynamic_paint_set_init_color_vcol_to_imseq_cb(void *userdata, const int i)
+static void dynamic_paint_set_init_color_vcol_to_imseq_cb(void *userdata, const int i,
+                                                          const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintSetInitColorData *data = userdata;
 
@@ -1673,7 +1676,8 @@ typedef struct DynamicPaintModifierApplyData {
 	MLoopCol *mloopcol_preview;
 } DynamicPaintModifierApplyData;
 
-static void dynamic_paint_apply_surface_displace_cb(void *userdata, const int i)
+static void dynamic_paint_apply_surface_displace_cb(void *userdata, const int i,
+                                                    const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintModifierApplyData *data = userdata;
 
@@ -1708,7 +1712,8 @@ static void dynamicPaint_applySurfaceDisplace(DynamicPaintSurface *surface, Deri
 	}
 }
 
-static void dynamic_paint_apply_surface_vpaint_blend_cb(void *userdata, const int i)
+static void dynamic_paint_apply_surface_vpaint_blend_cb(void *userdata, const int i,
+                                                        const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintModifierApplyData *data = userdata;
 
@@ -1719,7 +1724,8 @@ static void dynamic_paint_apply_surface_vpaint_blend_cb(void *userdata, const in
 	blendColors(pPoint[i].color, pPoint[i].color[3], pPoint[i].e_color, pPoint[i].e_color[3], fcolor[i]);
 }
 
-static void dynamic_paint_apply_surface_vpaint_cb(void *userdata, const int p_index)
+static void dynamic_paint_apply_surface_vpaint_cb(void *userdata, const int p_index,
+                                                  const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintModifierApplyData *data = userdata;
 	Object *ob = data->ob;
@@ -1789,7 +1795,8 @@ static void dynamic_paint_apply_surface_vpaint_cb(void *userdata, const int p_in
 	}
 }
 
-static void dynamic_paint_apply_surface_wave_cb(void *userdata, const int i)
+static void dynamic_paint_apply_surface_wave_cb(void *userdata, const int i,
+                                                const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintModifierApplyData *data = userdata;
 
@@ -2106,7 +2113,8 @@ typedef struct DynamicPaintCreateUVSurfaceData {
 	uint32_t *active_points;
 } DynamicPaintCreateUVSurfaceData;
 
-static void dynamic_paint_create_uv_surface_direct_cb(void *userdata, const int ty)
+static void dynamic_paint_create_uv_surface_direct_cb(void *userdata, const int ty,
+                                                      const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintCreateUVSurfaceData *data = userdata;
 
@@ -2203,7 +2211,8 @@ static void dynamic_paint_create_uv_surface_direct_cb(void *userdata, const int
 	}
 }
 
-static void dynamic_paint_create_uv_surface_neighbor_cb(void *userdata, const int ty)
+static void dynamic_paint_create_uv_surface_neighbor_cb(void *userdata, const int ty,
+                                                        const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintCreateUVSurfaceData *data = userdata;
 
@@ -2995,7 +3004,8 @@ typedef struct DynamicPaintOutputSurfaceImageData {
 	ImBuf *ibuf;
 } DynamicPaintOutputSurfaceImageData;
 
-static void dynamic_paint_output_surface_image_paint_cb(void *userdata, const int index)
+static void dynamic_paint_output_surface_image_paint_cb(void *userdata, const int index,
+                                                        const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintOutputSurfaceImageData *data = userdata;
 
@@ -3015,7 +3025,8 @@ static void dynamic_paint_output_surface_image_paint_cb(void *userdata, const in
 	}
 }
 
-static void dynamic_paint_output_surface_image_displace_cb(void *userdata, const int index)
+static void dynamic_paint_output_surface_image_displace_cb(void *userdata, const int index,
+                                                           const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintOutputSurfaceImageData *data = userdata;
 
@@ -3039,7 +3050,8 @@ static void dynamic_paint_output_surface_image_displace_cb(void *userdata, const
 	ibuf->rect_float[pos + 3] = 1.0f;
 }
 
-static void dynamic_paint_output_surface_image_wave_cb(void *userdata, const int index)
+static void dynamic_paint_output_surface_image_wave_cb(void *userdata, const int index,
+                                                       const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintOutputSurfaceImageData *data = userdata;
 
@@ -3061,7 +3073,8 @@ static void dynamic_paint_output_surface_image_wave_cb(void *userdata, const int
 	ibuf->rect_float[pos + 3] = 1.0f;
 }
 
-static void dynamic_paint_output_surface_image_wetmap_cb(void *userdata, const int index)
+static void dynamic_paint_output_surface_image_wetmap_cb(void *userdata, const int index,
+                                                         const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintOutputSurfaceImageData *data = userdata;
 
@@ -3549,7 +3562,8 @@ typedef struct DynamicPaintBrushVelocityData {
 	const float timescale;
 } DynamicPaintBrushVelocityData;
 
-static void dynamic_paint_brush_velocity_compute_cb(void *userdata, const int i)
+static void dynamic_paint_brush_velocity_compute_cb(void *userdata, const int i,
+                                                    const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintBrushVelocityData *data = userdata;
 
@@ -3699,7 +3713,7 @@ typedef struct DynamicPaintPaintData {
  *	Paint a brush object mesh to the surface
  */
 static void dynamic_paint_paint_mesh_cell_point_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int id, const int UNUSED(threadid))
+        void *userdata, const int id, const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintPaintData *data = userdata;
 
@@ -4140,7 +4154,7 @@ static int dynamicPaint_paintMesh(DynamicPaintSurface *surface,
  *	Paint a particle system to the surface
  */
 static void dynamic_paint_paint_particle_cell_point_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int id, const int UNUSED(threadid))
+        void *userdata, const int id, const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintPaintData *data = userdata;
 
@@ -4415,7 +4429,7 @@ static int dynamicPaint_paintParticles(DynamicPaintSurface *surface,
 
 /* paint a single point of defined proximity radius to the surface */
 static void dynamic_paint_paint_single_point_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int index, const int UNUSED(threadid))
+        void *userdata, const int index, const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintPaintData *data = userdata;
 
@@ -4561,7 +4575,8 @@ static int dynamicPaint_paintSinglePoint(
  *	Calculate current frame distances and directions for adjacency data
  */
 
-static void dynamic_paint_prepare_adjacency_cb(void *userdata, const int index)
+static void dynamic_paint_prepare_adjacency_cb(void *userdata, const int index,
+                                               const ParallelRangeTLS *UNUSED(tls))
 {
 	PaintSurfaceData *sData = userdata;
 	PaintBakeData *bData = sData->bData;
@@ -4788,7 +4803,8 @@ typedef struct DynamicPaintEffectData {
  *	Prepare data required by effects for current frame.
  *	Returns number of steps required
  */
-static void dynamic_paint_prepare_effect_cb(void *userdata, const int index)
+static void dynamic_paint_prepare_effect_cb(void *userdata, const int index,
+                                            const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintEffectData *data = userdata;
 
@@ -4893,7 +4909,8 @@ static int dynamicPaint_prepareEffectStep(
 /**
  *	Processes active effect step.
  */
-static void dynamic_paint_effect_spread_cb(void *userdata, const int index)
+static void dynamic_paint_effect_spread_cb(void *userdata, const int index,
+                                           const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintEffectData *data = userdata;
 
@@ -4938,7 +4955,8 @@ static void dynamic_paint_effect_spread_cb(void *userdata, const int index)
 	}
 }
 
-static void dynamic_paint_effect_shrink_cb(void *userdata, const int index)
+static void dynamic_paint_effect_shrink_cb(void *userdata, const int index,
+                                           const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintEffectData *data = userdata;
 
@@ -4988,7 +5006,8 @@ static void dynamic_paint_effect_shrink_cb(void *userdata, const int index)
 	}
 }
 
-static void dynamic_paint_effect_drip_cb(void *userdata, const int index)
+static void dynamic_paint_effect_drip_cb(void *userdata, const int index,
+                                         const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintEffectData *data = userdata;
 
@@ -5166,7 +5185,8 @@ static void dynamicPaint_doEffectStep(
 	}
 }
 
-static void dynamic_paint_border_cb(void *userdata, const int b_index)
+static void dynamic_paint_border_cb(void *userdata, const int b_index,
+                                    const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintEffectData *data = userdata;
 
@@ -5240,7 +5260,8 @@ static void dynamicPaint_doBorderStep(DynamicPaintSurface *surface)
 	            0, sData->adj_data->total_border, &data, dynamic_paint_border_cb, sData->adj_data->total_border > 1000);
 }
 
-static void dynamic_paint_wave_step_cb(void *userdata, const int index)
+static void dynamic_paint_wave_step_cb(void *userdata, const int index,
+                                       const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintEffectData *data = userdata;
 
@@ -5403,7 +5424,8 @@ typedef struct DynamicPaintDissolveDryData {
 	const float timescale;
 } DynamicPaintDissolveDryData;
 
-static void dynamic_paint_surface_pre_step_cb(void *userdata, const int index)
+static void dynamic_paint_surface_pre_step_cb(void *userdata, const int index,
+                                              const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintDissolveDryData *data = userdata;
 
@@ -5526,7 +5548,8 @@ typedef struct DynamicPaintGenerateBakeData {
 	const bool new_bdata;
 } DynamicPaintGenerateBakeData;
 
-static void dynamic_paint_generate_bake_data_cb(void *userdata, const int index)
+static void dynamic_paint_generate_bake_data_cb(void *userdata, const int index,
+                                                const ParallelRangeTLS *UNUSED(tls))
 {
 	const DynamicPaintGenerateBakeData *data = userdata;
 
diff --git a/source/blender/blenkernel/intern/mask_rasterize.c b/source/blender/blenkernel/intern/mask_rasterize.c
index 104bb0c07a6..489f1965884 100644
--- a/source/blender/blenkernel/intern/mask_rasterize.c
+++ b/source/blender/blenkernel/intern/mask_rasterize.c
@@ -1434,7 +1434,8 @@ typedef struct MaskRasterizeBufferData {
 	float *buffer;
 } MaskRasterizeBufferData;
 
-static void maskrasterize_buffer_cb(void *userdata, int y)
+static void maskrasterize_buffer_cb(void *userdata, int y,
+                                    const ParallelRangeTLS *UNUSED(tls))
 {
 	MaskRasterizeBufferData *data = userdata;
 
diff --git a/source/blender/blenkernel/intern/mesh_evaluate.c b/source/blender/blenkernel/intern/mesh_evaluate.c
index 35757557f76..b58bea941e5 100644
--- a/source/blender/blenkernel/intern/mesh_evaluate.c
+++ b/source/blender/blenkernel/intern/mesh_evaluate.c
@@ -177,7 +177,8 @@ typedef struct MeshCalcNormalsData {
 	float (*vnors)[3];
 } MeshCalcNormalsData;
 
-static void mesh_calc_normals_poly_cb(void *userdata, const int pidx)
+static void mesh_calc_normals_poly_cb(void *userdata, const int pidx,
+                                      const ParallelRangeTLS *UNUSED(tls))
 {
 	MeshCalcNormalsData *data = userdata;
 	const MPoly *mp = &data->mpolys[pidx];
@@ -185,7 +186,8 @@ static void mesh_calc_normals_poly_cb(void *userdata, const int pidx)
 	BKE_mesh_calc_poly_normal(mp, data->mloop + mp->loopstart, data->mverts, data->pnors[pidx]);
 }
 
-static void mesh_calc_normals_poly_prepare_cb(void *userdata, const int pidx)
+static void mesh_calc_normals_poly_prepare_cb(void *userdata, const int pidx,
+                                              const ParallelRangeTLS *UNUSED(tls))
 {
 	MeshCalcNormalsData *data = userdata;
 	const MPoly *mp = &data->mpolys[pidx];
@@ -247,14 +249,16 @@ static void mesh_calc_normals_poly_prepare_cb(void *userdata, const int pidx)
 	}
 }
 
-static void mesh_calc_normals_poly_accum_cb(void *userdata, const int lidx)
+static void mesh_calc_normals_poly_accum_cb(void *userdata, const int lidx,
+                                            const ParallelRangeTLS *UNUSED(tls))
 {
 	MeshCalcNormalsData *data = userdata;
 
 	add_v3_v3(data->vnors[data->mloop[lidx].v], data->lnors_weighted[lidx]);
 }
 
-static void mesh_calc_normals_poly_finalize_cb(void *userdata, const int vidx)
+static void mesh_calc_normals_poly_finalize_cb(void *userdata, const int vidx,
+                                               const ParallelRangeTLS *UNUSED(tls))
 {
 	MeshCalcNormalsData *data = userdata;
 
diff --git a/source/blender/blenkernel/intern/ocean.c b/source/blender/blenkernel/intern/ocean.c
index 537c8926a5b..53e4119987f 100644
--- a/source/blender/blenkernel/intern/ocean.c
+++ b/source/blender/blenkernel/intern/ocean.c
@@ -502,7 +502,8 @@ typedef struct OceanSimulateData {
 	float chop_amount;
 } OceanSimulateData;
 
-static void ocean_compute_htilda(void *userdata, const int i)
+static void ocean_compute_htilda(void *userdata, const int i,
+                                 const ParallelRangeTLS *UNUSED(tls))
 {
 	OceanSimulateData *osd = userdata;
 	const Ocean *o = osd->o;
diff --git a/source/blender/blenkernel/intern/particle_system.c b/source/blender/blenkernel/intern/particle_system.c
index 38eb861599a..505f6c63baf 100644
--- a/source/blender/blenkernel/intern/particle_system.c
+++ b/source/blender/blenkernel/intern/particle_system.c
@@ -3371,14 +3371,14 @@ typedef struct DynamicStepSolverTaskData {
 } DynamicStepSolverTaskData;
 
 static void dynamics_step_sph_ddr_task_cb_ex(
-        void *userdata, void *userdata_chunk, const int p, const int UNUSED(thread_id))
+        void *userdata, const int p, const ParallelRangeTLS *tls)
 {
 	DynamicStepSolverTaskData *data = userdata;
 	ParticleSimulationData *sim = data->sim;
 	ParticleSystem *psys = sim->psys;
 	ParticleSettings *part = psys->part;
 
-	SPHData *sphdata = userdata_chunk;
+	SPHData *sphdata = tls->userdata_chunk;
 
 	ParticleData *pa;
 
@@ -3405,7 +3405,7 @@ static void dynamics_step_sph_ddr_task_cb_ex(
 }
 
 static void dynamics_step_sph_classical_basic_integrate_task_cb_ex(
-        void *userdata,  void *UNUSED(userdata_chunk), const int p, const int UNUSED(thread_id))
+        void *userdata,  const int p, const ParallelRangeTLS *UNUSED(tls))
 {
 	DynamicStepSolverTaskData *data = userdata;
 	ParticleSimulationData *sim = data->sim;
@@ -3421,13 +3421,13 @@ static void dynamics_step_sph_classical_basic_integrate_task_cb_ex(
 }
 
 static void dynamics_step_sph_classical_calc_density_task_cb_ex(
-        void *userdata, void *userdata_chunk, const int p, const int UNUSED(thread_id))
+        void *userdata, const int p, const ParallelRangeTLS *tls)
 {
 	DynamicStepSolverTaskData *data = userdata;
 	ParticleSimulationData *sim = data->sim;
 	ParticleSystem *psys = sim->psys;
 
-	SPHData *sphdata = userdata_chunk;
+	SPHData *sphdata = tls->userdata_chunk;
 
 	ParticleData *pa;
 
@@ -3439,14 +3439,14 @@ static void dynamics_step_sph_classical_calc_density_task_cb_ex(
 }
 
 static void dynamics_step_sph_classical_integrate_task_cb_ex(
-        void *userdata, void *userdata_chunk, const int p, const int UNUSED(thread_id))
+        void *userdata, const int p, const ParallelRangeTLS *tls)
 {
 	DynamicStepSolverTaskData *data = userdata;
 	ParticleSimulationData *sim = data->sim;
 	ParticleSystem *psys = sim->psys;
 	ParticleSettings *part = psys->part;
 
-	SPHData *sphdata = userdata_chunk;
+	SPHData *sphdata = tls->userdata_chunk;
 
 	ParticleData *pa;
 
diff --git a/source/blender/blenkernel/intern/pbvh.c b/source/blender/blenkernel/intern/pbvh.c
index 39d1d8e8510..e476ccbc48b 100644
--- a/source/blender/blenkernel/intern/pbvh.c
+++ b/source/blender/blenkernel/intern/pbvh.c
@@ -938,7 +938,8 @@ typedef struct PBVHUpdateData {
 	int flag;
 } PBVHUpdateData;
 
-static void pbvh_update_normals_accum_task_cb(void *userdata, const int n)
+static void pbvh_update_normals_accum_task_cb(void *userdata, const int n,
+                                              const ParallelRangeTLS *UNUSED(tls))
 {
 	PBVHUpdateData *data = userdata;
 
@@ -991,7 +992,8 @@ static void pbvh_update_normals_accum_task_cb(void *userdata, const int n)
 	}
 }
 
-static void pbvh_update_normals_store_task_cb(void *userdata, const int n)
+static void pbvh_update_normals_store_task_cb(void *userdata, const int n,
+                                              const ParallelRangeTLS *UNUSED(tls))
 {
 	PBVHUpdateData *data = userdata;
 	PBVH *bvh = data->bvh;
@@ -1057,7 +1059,8 @@ static void pbvh_update_normals(PBVH *bvh, PBVHNode **nodes,
 	MEM_freeN(vnors);
 }
 
-static void pbvh_update_BB_redraw_task_cb(void *userdata, const int n)
+static void pbvh_update_BB_redraw_task_cb(void *userdata, const int n,
+                                          const ParallelRangeTLS *UNUSED(tls))
 {
 	PBVHUpdateData *data = userdata;
 	PBVH *bvh = data->bvh;
diff --git a/source/blender/blenkernel/intern/shrinkwrap.c b/source/blender/blenkernel/intern/shrinkwrap.c
index f9d1793d7cb..09b0dc569cb 100644
--- a/source/blender/blenkernel/intern/shrinkwrap.c
+++ b/source/blender/blenkernel/intern/shrinkwrap.c
@@ -88,13 +88,13 @@ typedef struct ShrinkwrapCalcCBData {
  * for each vertex performs a nearest vertex search on the tree
  */
 static void shrinkwrap_calc_nearest_vertex_cb_ex(
-        void *userdata, void *userdata_chunk, const int i, const int UNUSED(threadid))
+        void *userdata, const int i, const ParallelRangeTLS *tls)
 {
 	ShrinkwrapCalcCBData *data = userdata;
 
 	ShrinkwrapCalcData *calc = data->calc;
 	BVHTreeFromMesh *treeData = data->treeData;
-	BVHTreeNearest *nearest = userdata_chunk;
+	BVHTreeNearest *nearest = tls->userdata_chunk;
 
 	float *co = calc->vertexCos[i];
 	float tmp_co[3];
@@ -257,7 +257,7 @@ bool BKE_shrinkwrap_project_normal(
 }
 
 static void shrinkwrap_calc_normal_projection_cb_ex(
-        void *userdata, void *userdata_chunk, const int i, const int UNUSED(threadid))
+        void *userdata, const int i, const ParallelRangeTLS *tls)
 {
 	ShrinkwrapCalcCBData *data = userdata;
 
@@ -272,7 +272,7 @@ static void shrinkwrap_calc_normal_projection_cb_ex(
 	float *proj_axis = data->proj_axis;
 	SpaceTransform *local2aux = data->local2aux;
 
-	BVHTreeRayHit *hit = userdata_chunk;
+	BVHTreeRayHit *hit = tls->userdata_chunk;
 
 	const float proj_limit_squared = calc->smd->projLimit * calc->smd->projLimit;
 	float *co = calc->vertexCos[i];
@@ -495,13 +495,13 @@ static void shrinkwrap_calc_normal_projection(ShrinkwrapCalcData *calc, bool for
  * NN matches for each vertex
  */
 static void shrinkwrap_calc_nearest_surface_point_cb_ex(
-        void *userdata, void *userdata_chunk, const int i, const int UNUSED(threadid))
+        void *userdata, const int i, const ParallelRangeTLS *tls)
 {
 	ShrinkwrapCalcCBData *data = userdata;
 
 	ShrinkwrapCalcData *calc = data->calc;
 	BVHTreeFromMesh *treeData = data->treeData;
-	BVHTreeNearest *nearest = userdata_chunk;
+	BVHTreeNearest *nearest = tls->userdata_chunk;
 
 	float *co = calc->vertexCos[i];
 	float tmp_co[3];
diff --git a/source/blender/blenkernel/intern/smoke.c b/source/blender/blenkernel/intern/smoke.c
index fcf4724bd1c..37b9466645d 100644
--- a/source/blender/blenkernel/intern/smoke.c
+++ b/source/blender/blenkernel/intern/smoke.c
@@ -738,7 +738,7 @@ typedef struct ObstaclesFromDMData {
 	int *num_obstacles;
 } ObstaclesFromDMData;
 
-static void obstacles_from_derivedmesh_task_cb(void *userdata, const int z)
+static void obstacles_from_derivedmesh_task_cb(void *userdata, const int z, const ParallelRangeTLS *UNUSED(tls))
 {
 	ObstaclesFromDMData *data = userdata;
 	SmokeDomainSettings *sds = data->sds;
@@ -1184,7 +1184,7 @@ typedef struct EmitFromParticlesData {
 	float hr_smooth;
 } EmitFromParticlesData;
 
-static void emit_from_particles_task_cb(void *userdata, const int z)
+static void emit_from_particles_task_cb(void *userdata, const int z, const ParallelRangeTLS *UNUSED(tls))
 {
 	EmitFromParticlesData *data = userdata;
 	SmokeFlowSettings *sfs = data->sfs;
@@ -1567,7 +1567,7 @@ typedef struct EmitFromDMData {
 	int *min, *max, *res;
 } EmitFromDMData;
 
-static void emit_from_derivedmesh_task_cb(void *userdata, const int z)
+static void emit_from_derivedmesh_task_cb(void *userdata, const int z, const ParallelRangeTLS *UNUSED(tls))
 {
 	EmitFromDMData *data = userdata;
 	EmissionMap *em = data->em;
@@ -2436,7 +2436,7 @@ typedef struct UpdateEffectorsData {
 	unsigned char *obstacle;
 } UpdateEffectorsData;
 
-static void update_effectors_task_cb(void *userdata, const int x)
+static void update_effectors_task_cb(void *userdata, const int x, const ParallelRangeTLS *UNUSED(tls))
 {
 	UpdateEffectorsData *data = userdata;
 	SmokeDomainSettings *sds = data->sds;
diff --git a/source/blender/blenkernel/intern/tracking_auto.c b/source/blender/blenkernel/intern/tracking_auto.c
index 40234bacdfe..5dc350640c0 100644
--- a/source/blender/blenkernel/intern/tracking_auto.c
+++ b/source/blender/blenkernel/intern/tracking_auto.c
@@ -432,7 +432,9 @@ AutoTrackContext *BKE_autotrack_context_new(MovieClip *clip,
 	return context;
 }
 
-static void autotrack_context_step_cb(void *userdata, int track)
+static void autotrack_context_step_cb(void *userdata,
+                                      int track,
+                                      const ParallelRangeTLS *UNUSED(tls))
 {
 	AutoTrackContext *context = userdata;
 	const int frame_delta = context->backwards ? -1 : 1;
diff --git a/source/blender/blenkernel/intern/tracking_stabilize.c b/source/blender/blenkernel/intern/tracking_stabilize.c
index edddeb41cc8..9e8f01df1d5 100644
--- a/source/blender/blenkernel/intern/tracking_stabilize.c
+++ b/source/blender/blenkernel/intern/tracking_stabilize.c
@@ -1503,7 +1503,10 @@ typedef struct TrackingStabilizeFrameInterpolationData {
 	interpolation_func interpolation;
 } TrackingStabilizeFrameInterpolationData;
 
-static void tracking_stabilize_frame_interpolation_cb(void *userdata, int j)
+static void tracking_stabilize_frame_interpolation_cb(
+        void *userdata,
+        int j,
+        const ParallelRangeTLS *UNUSED(tls))
 {
 	TrackingStabilizeFrameInterpolationData *data = userdata;
 	ImBuf *ibuf = data->ibuf;
diff --git a/source/blender/blenlib/BLI_task.h b/source/blender/blenlib/BLI_task.h
index ccfa2b6e2e7..acfdd3729c1 100644
--- a/source/blender/blenlib/BLI_task.h
+++ b/source/blender/blenlib/BLI_task.h
@@ -116,15 +116,27 @@ void BLI_task_pool_delayed_push_begin(TaskPool *pool, int thread_id);
 void BLI_task_pool_delayed_push_end(TaskPool *pool, int thread_id);
 
 /* Parallel for routines */
-typedef void (*TaskParallelRangeFunc)(void *userdata, const int iter);
-typedef void (*TaskParallelRangeFuncEx)(void *userdata, void *userdata_chunk, const int iter, const int thread_id);
+
+/* Per-thread specific data passed to the callback. */
+typedef struct ParallelRangeTLS {
+	/* Identifier of the thread who this data belongs to. */
+	int thread_id;
+	/* Copy of user-specifier chunk, which is copied from original chunk to all
+	 * worker threads. This is similar to OpenMP's firstprivate.
+	 */
+	void *userdata_chunk;
+} ParallelRangeTLS;
+
+typedef void (*TaskParallelRangeFunc)(void *userdata,
+                                      const int iter,
+                                      const ParallelRangeTLS *tls);
 typedef void (*TaskParallelRangeFuncFinalize)(void *userdata, void *userdata_chunk);
 void BLI_task_parallel_range_ex(
         int start, int stop,
         void *userdata,
         void *userdata_chunk,
         const size_t userdata_chunk_size,
-        TaskParallelRangeFuncEx func_ex,
+        TaskParallelRangeFunc func,
         const bool use_threading,
         const bool use_dynamic_scheduling);
 void BLI_task_parallel_range(
@@ -138,7 +150,7 @@ void BLI_task_parallel_range_finalize(
         void *userdata,
         void *userdata_chunk,
         const size_t userdata_chunk_size,
-        TaskParallelRangeFuncEx func_ex,
+        TaskParallelRangeFunc func,
         TaskParallelRangeFuncFinalize func_finalize,
         const bool use_threading,
         const bool use_dynamic_scheduling);
diff --git a/source/blender/blenlib/intern/BLI_kdopbvh.c b/source/blender/blenlib/intern/BLI_kdopbvh.c
index bd16bc1a9c6..6e33f75fe69 100644
--- a/source/blender/blenlib/intern/BLI_kdopbvh.c
+++ b/source/blender/blenlib/intern/BLI_kdopbvh.c
@@ -798,7 +798,7 @@ typedef struct BVHDivNodesData {
 	int first_of_next_level;
 } BVHDivNodesData;
 
-static void non_recursive_bvh_div_nodes_task_cb(void *userdata, const int j)
+static void non_recursive_bvh_div_nodes_task_cb(void *userdata, const int j, const ParallelRangeTLS *UNUSED(tls))
 {
 	BVHDivNodesData *data = userdata;
 
@@ -929,8 +929,9 @@ static void non_recursive_bvh_div_nodes(
 		}
 		else {
 			/* Less hassle for debugging. */
+			ParallelRangeTLS tls = {0};
 			for (int i_task = i; i_task < i_stop; i_task++) {
-				non_recursive_bvh_div_nodes_task_cb(&cb_data, i_task);
+				non_recursive_bvh_div_nodes_task_cb(&cb_data, i_task, &tls);
 			}
 		}
 	}
@@ -1276,7 +1277,7 @@ int BLI_bvhtree_overlap_thread_num(const BVHTree *tree)
 	return (int)MIN2(tree->tree_type, tree->nodes[tree->totleaf]->totnode);
 }
 
-static void bvhtree_overlap_task_cb(void *userdata, const int j)
+static void bvhtree_overlap_task_cb(void *userdata, const int j, const ParallelRangeTLS *UNUSED(tls))
 {
 	BVHOverlapData_Thread *data = &((BVHOverlapData_Thread *)userdata)[j];
 	BVHOverlapData_Shared *data_shared = data->shared;
diff --git a/source/blender/blenlib/intern/math_statistics.c b/source/blender/blenlib/intern/math_statistics.c
index fd7418a8f7b..cfadbba028f 100644
--- a/source/blender/blenlib/intern/math_statistics.c
+++ b/source/blender/blenlib/intern/math_statistics.c
@@ -46,7 +46,8 @@ typedef struct CovarianceData {
 	int nbr_cos_vn;
 } CovarianceData;
 
-static void covariance_m_vn_ex_task_cb(void *userdata, const int a)
+static void covariance_m_vn_ex_task_cb(void *userdata, const int a,
+                                       const ParallelRangeTLS *UNUSED(tls))
 {
 	CovarianceData *data = userdata;
 	const float *cos_vn = data->cos_vn;
diff --git a/source/blender/blenlib/intern/task.c b/source/blender/blenlib/intern/task.c
index afa20e3d766..030d4124e62 100644
--- a/source/blender/blenlib/intern/task.c
+++ b/source/blender/blenlib/intern/task.c
@@ -994,7 +994,6 @@ typedef struct ParallelRangeState {
 	void *userdata;
 
 	TaskParallelRangeFunc func;
-	TaskParallelRangeFuncEx func_ex;
 
 	int iter;
 	int chunk_size;
@@ -1015,23 +1014,17 @@ BLI_INLINE bool parallel_range_next_iter_get(
 static void parallel_range_func(
         TaskPool * __restrict pool,
         void *userdata_chunk,
-        int threadid)
+        int thread_id)
 {
 	ParallelRangeState * __restrict state = BLI_task_pool_userdata(pool);
+	ParallelRangeTLS tls = {
+		.thread_id = thread_id,
+		.userdata_chunk = userdata_chunk,
+	};
 	int iter, count;
-
 	while (parallel_range_next_iter_get(state, &iter, &count)) {
-		int i;
-
-		if (state->func_ex) {
-			for (i = 0; i < count; ++i) {
-				state->func_ex(state->userdata, userdata_chunk, iter + i, threadid);
-			}
-		}
-		else {
-			for (i = 0; i < count; ++i) {
-				state->func(state->userdata, iter + i);
-			}
+		for (int i = 0; i < count; ++i) {
+			state->func(state->userdata, iter + i, &tls);
 		}
 	}
 }
@@ -1047,7 +1040,6 @@ static void task_parallel_range_ex(
         void *userdata_chunk,
         const size_t userdata_chunk_size,
         TaskParallelRangeFunc func,
-        TaskParallelRangeFuncEx func_ex,
         TaskParallelRangeFuncFinalize func_finalize,
         const bool use_threading,
         const bool use_dynamic_scheduling)
@@ -1059,7 +1051,7 @@ static void task_parallel_range_ex(
 
 	void *userdata_chunk_local = NULL;
 	void *userdata_chunk_array = NULL;
-	const bool use_userdata_chunk = (func_ex != NULL) && (userdata_chunk_size != 0) && (userdata_chunk != NULL);
+	const bool use_userdata_chunk = (userdata_chunk_size != 0) && (userdata_chunk != NULL);
 
 	if (start == stop) {
 		return;
@@ -1067,7 +1059,6 @@ static void task_parallel_range_ex(
 
 	BLI_assert(start < stop);
 	if (userdata_chunk_size != 0) {
-		BLI_assert(func_ex != NULL && func == NULL);
 		BLI_assert(userdata_chunk != NULL);
 	}
 
@@ -1075,28 +1066,25 @@ static void task_parallel_range_ex(
 	 * do everything from the main thread.
 	 */
 	if (!use_threading) {
-		if (func_ex) {
-			if (use_userdata_chunk) {
-				userdata_chunk_local = MALLOCA(userdata_chunk_size);
-				memcpy(userdata_chunk_local, userdata_chunk, userdata_chunk_size);
-			}
-
-			for (i = start; i < stop; ++i) {
-				func_ex(userdata, userdata_chunk_local, i, 0);
-			}
-
-			if (func_finalize) {
-				func_finalize(userdata, userdata_chunk_local);
-			}
+		if (use_userdata_chunk) {
+			userdata_chunk_local = MALLOCA(userdata_chunk_size);
+			memcpy(userdata_chunk_local, userdata_chunk, userdata_chunk_size);
+		}
 
-			MALLOCA_FREE(userdata_chunk_local, userdata_chunk_size);
+		ParallelRangeTLS tls = {
+			.thread_id = 0,
+			.userdata_chunk = userdata_chunk_local,
+		};
+		for (i = start; i < stop; ++i) {
+			func(userdata, i, &tls);
 		}
-		else {
-			for (i = start; i < stop; ++i) {
-				func(userdata, i);
-			}
+
+		if (func_finalize) {
+			func_finalize(userdata, userdata_chunk_local);
 		}
 
+		MALLOCA_FREE(userdata_chunk_local, userdata_chunk_size);
+
 		return;
 	}
 
@@ -1114,7 +1102,6 @@ static void task_parallel_range_ex(
 	state.stop = stop;
 	state.userdata = userdata;
 	state.func = func;
-	state.func_ex = func_ex;
 	state.iter = start;
 	if (use_dynamic_scheduling) {
 		state.chunk_size = 32;
@@ -1180,12 +1167,12 @@ void BLI_task_parallel_range_ex(
         void *userdata,
         void *userdata_chunk,
         const size_t userdata_chunk_size,
-        TaskParallelRangeFuncEx func_ex,
+        TaskParallelRangeFunc func,
         const bool use_threading,
         const bool use_dynamic_scheduling)
 {
 	task_parallel_range_ex(
-	            start, stop, userdata, userdata_chunk, userdata_chunk_size, NULL, func_ex, NULL,
+	            start, stop, userdata, userdata_chunk, userdata_chunk_size, func, NULL,
 	            use_threading, use_dynamic_scheduling);
 }
 
@@ -1206,7 +1193,7 @@ void BLI_task_parallel_range(
         TaskParallelRangeFunc func,
         const bool use_threading)
 {
-	task_parallel_range_ex(start, stop, userdata, NULL, 0, func, NULL, NULL, use_threading, false);
+	task_parallel_range_ex(start, stop, userdata, NULL, 0, func, NULL, use_threading, false);
 }
 
 /**
@@ -1232,13 +1219,13 @@ void BLI_task_parallel_range_finalize(
         void *userdata,
         void *userdata_chunk,
         const size_t userdata_chunk_size,
-        TaskParallelRangeFuncEx func_ex,
+        TaskParallelRangeFunc func,
         TaskParallelRangeFuncFinalize func_finalize,
         const bool use_threading,
         const bool use_dynamic_scheduling)
 {
 	task_parallel_range_ex(
-	            start, stop, userdata, userdata_chunk, userdata_chunk_size, NULL, func_ex, func_finalize,
+	            start, stop, userdata, userdata_chunk, userdata_chunk_size, func, func_finalize,
 	            use_threading, use_dynamic_scheduling);
 }
 
diff --git a/source/blender/bmesh/intern/bmesh_interp.c b/source/blender/bmesh/intern/bmesh_interp.c
index 96b2eb17c4c..29f74b13af8 100644
--- a/source/blender/bmesh/intern/bmesh_interp.c
+++ b/source/blender/bmesh/intern/bmesh_interp.c
@@ -419,7 +419,8 @@ typedef struct BMLoopInterpMultiresData {
 	float d;
 } BMLoopInterpMultiresData;
 
-static void loop_interp_multires_cb(void *userdata, int ix)
+static void loop_interp_multires_cb(void *userdata, int ix,
+                                    const ParallelRangeTLS *UNUSED(tls))
 {
 	BMLoopInterpMultiresData *data = userdata;
 
diff --git a/source/blender/depsgraph/intern/eval/deg_eval.cc b/source/blender/depsgraph/intern/eval/deg_eval.cc
index 09d25be41d6..4c5d68fb7ef 100644
--- a/source/blender/depsgraph/intern/eval/deg_eval.cc
+++ b/source/blender/depsgraph/intern/eval/deg_eval.cc
@@ -107,7 +107,9 @@ typedef struct CalculatePengindData {
 	unsigned int layers;
 } CalculatePengindData;
 
-static void calculate_pending_func(void *data_v, int i)
+static void calculate_pending_func(void *data_v,
+                                   int i,
+                                   const ParallelRangeTLS * /*tls*/)
 {
 	CalculatePengindData *data = (CalculatePengindData *)data_v;
 	Depsgraph *graph = data->graph;
diff --git a/source/blender/depsgraph/intern/eval/deg_eval_flush.cc b/source/blender/depsgraph/intern/eval/deg_eval_flush.cc
index 63d010849c3..546d4e3cf5d 100644
--- a/source/blender/depsgraph/intern/eval/deg_eval_flush.cc
+++ b/source/blender/depsgraph/intern/eval/deg_eval_flush.cc
@@ -83,14 +83,18 @@ void lib_id_recalc_data_tag(Main *bmain, ID *id)
 	DEG_id_type_tag(bmain, GS(id->name));
 }
 
-void flush_init_operation_node_func(void *data_v, int i)
+void flush_init_operation_node_func(void *data_v,
+                                    int i,
+                                    const ParallelRangeTLS * /*tls*/)
 {
 	Depsgraph *graph = (Depsgraph *)data_v;
 	OperationDepsNode *node = graph->operations[i];
 	node->scheduled = false;
 }
 
-void flush_init_id_node_func(void *data_v, int i)
+void flush_init_id_node_func(void *data_v,
+                             int i,
+                             const ParallelRangeTLS * /*tls*/)
 {
 	Depsgraph *graph = (Depsgraph *)data_v;
 	IDDepsNode *id_node = graph->id_nodes[i];
@@ -279,7 +283,9 @@ void deg_graph_flush_updates(Main *bmain, Depsgraph *graph)
 	flush_editors_id_update(bmain, graph);
 }
 
-static void graph_clear_func(void *data_v, int i)
+static void graph_clear_func(void *data_v,
+                             int i,
+                             const ParallelRangeTLS * /*tls*/)
 {
 	Depsgraph *graph = (Depsgraph *)data_v;
 	OperationDepsNode *node = graph->operations[i];
diff --git a/source/blender/editors/sculpt_paint/paint_cursor.c b/source/blender/editors/sculpt_paint/paint_cursor.c
index f0d9a0bbb68..1378952c4cc 100644
--- a/source/blender/editors/sculpt_paint/paint_cursor.c
+++ b/source/blender/editors/sculpt_paint/paint_cursor.c
@@ -151,7 +151,7 @@ typedef struct LoadTexData {
 	float radius;
 } LoadTexData;
 
-static void load_tex_task_cb_ex(void *userdata, void *UNUSED(userdata_chunck), const int j, const int thread_id)
+static void load_tex_task_cb_ex(void *userdata, const int j, const ParallelRangeTLS *tls)
 {
 	LoadTexData *data = userdata;
 	Brush *br = data->br;
@@ -212,7 +212,7 @@ static void load_tex_task_cb_ex(void *userdata, void *UNUSED(userdata_chunck), c
 			if (col) {
 				float rgba[4];
 
-				paint_get_tex_pixel_col(mtex, x, y, rgba, pool, thread_id, convert_to_linear, colorspace);
+				paint_get_tex_pixel_col(mtex, x, y, rgba, pool, tls->thread_id, convert_to_linear, colorspace);
 
 				buffer[index * 4]     = rgba[0] * 255;
 				buffer[index * 4 + 1] = rgba[1] * 255;
@@ -220,7 +220,7 @@ static void load_tex_task_cb_ex(void *userdata, void *UNUSED(userdata_chunck), c
 				buffer[index * 4 + 3] = rgba[3] * 255;
 			}
 			else {
-				float avg = paint_get_tex_pixel(mtex, x, y, pool, thread_id);
+				float avg = paint_get_tex_pixel(mtex, x, y, pool, tls->thread_id);
 
 				avg += br->texture_sample_bias;
 
@@ -367,7 +367,7 @@ static int load_tex(Brush *br, ViewContext *vc, float zoom, bool col, bool prima
 	return 1;
 }
 
-static void load_tex_cursor_task_cb(void *userdata, const int j)
+static void load_tex_cursor_task_cb(void *userdata, const int j, const ParallelRangeTLS *UNUSED(tls))
 {
 	LoadTexData *data = userdata;
 	Brush *br = data->br;
diff --git a/source/blender/editors/sculpt_paint/paint_image_2d.c b/source/blender/editors/sculpt_paint/paint_image_2d.c
index a1aad2bdc57..b687c937a22 100644
--- a/source/blender/editors/sculpt_paint/paint_image_2d.c
+++ b/source/blender/editors/sculpt_paint/paint_image_2d.c
@@ -1070,7 +1070,8 @@ typedef struct Paint2DForeachData {
 	int tilew;
 } Paint2DForeachData;
 
-static void paint_2d_op_foreach_do(void *data_v, const int iter)
+static void paint_2d_op_foreach_do(void *data_v, const int iter,
+                                   const ParallelRangeTLS *UNUSED(tls))
 {
 	Paint2DForeachData *data = (Paint2DForeachData *)data_v;
 	paint_2d_do_making_brush(data->s, data->region, data->curveb,
diff --git a/source/blender/editors/sculpt_paint/paint_mask.c b/source/blender/editors/sculpt_paint/paint_mask.c
index 3b1b993a3e3..a48a3666ef9 100644
--- a/source/blender/editors/sculpt_paint/paint_mask.c
+++ b/source/blender/editors/sculpt_paint/paint_mask.c
@@ -104,7 +104,8 @@ typedef struct MaskTaskData {
 	float (*clip_planes_final)[4];
 } MaskTaskData;
 
-static void mask_flood_fill_task_cb(void *userdata, const int i)
+static void mask_flood_fill_task_cb(void *userdata, const int i,
+                                    const ParallelRangeTLS *UNUSED(tls))
 {
 	MaskTaskData *data = userdata;
 
@@ -218,7 +219,8 @@ static void flip_plane(float out[4], const float in[4], const char symm)
 	out[3] = in[3];
 }
 
-static void mask_box_select_task_cb(void *userdata, const int i)
+static void mask_box_select_task_cb(void *userdata, const int i,
+                                    const ParallelRangeTLS *UNUSED(tls))
 {
 	MaskTaskData *data = userdata;
 
@@ -373,7 +375,8 @@ static void mask_lasso_px_cb(int x, int x_end, int y, void *user_data)
 	} while (++index != index_end);
 }
 
-static void mask_gesture_lasso_task_cb(void *userdata, const int i)
+static void mask_gesture_lasso_task_cb(void *userdata, const int i,
+                                       const ParallelRangeTLS *UNUSED(tls))
 {
 	LassoMaskData *lasso_data = userdata;
 	MaskTaskData *data = &lasso_data->task_data;
diff --git a/source/blender/editors/sculpt_paint/paint_vertex.c b/source/blender/editors/sculpt_paint/paint_vertex.c
index d61777a6d12..666272bd11a 100644
--- a/source/blender/editors/sculpt_paint/paint_vertex.c
+++ b/source/blender/editors/sculpt_paint/paint_vertex.c
@@ -1431,7 +1431,7 @@ static float wpaint_get_active_weight(const MDeformVert *dv, const WeightPaintIn
 }
 
 static void do_wpaint_precompute_weight_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata, const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	const MDeformVert *dv = &data->me->dvert[n];
@@ -1458,7 +1458,7 @@ static void precompute_weight_values(
 }
 
 static void do_wpaint_brush_blur_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata, const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -1547,7 +1547,7 @@ static void do_wpaint_brush_blur_task_cb_ex(
 }
 
 static void do_wpaint_brush_smear_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata, const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -1654,7 +1654,7 @@ static void do_wpaint_brush_smear_task_cb_ex(
 }
 
 static void do_wpaint_brush_draw_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata, const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -1724,7 +1724,7 @@ static void do_wpaint_brush_draw_task_cb_ex(
 }
 
 static void do_wpaint_brush_calc_average_weight_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata, const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2377,7 +2377,7 @@ static bool vpaint_stroke_test_start(bContext *C, struct wmOperator *op, const f
 }
 
 static void do_vpaint_brush_calc_average_color_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata, const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2439,7 +2439,7 @@ static float tex_color_alpha_ubyte(
 }
 
 static void do_vpaint_brush_draw_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata,  const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2531,7 +2531,7 @@ static void do_vpaint_brush_draw_task_cb_ex(
 }
 
 static void do_vpaint_brush_blur_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata, const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2640,7 +2640,7 @@ static void do_vpaint_brush_blur_task_cb_ex(
 }
 
 static void do_vpaint_brush_smear_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int UNUSED(thread_id))
+        void *userdata, const int n, const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
diff --git a/source/blender/editors/sculpt_paint/sculpt.c b/source/blender/editors/sculpt_paint/sculpt.c
index 53d7684e331..4684395c18d 100644
--- a/source/blender/editors/sculpt_paint/sculpt.c
+++ b/source/blender/editors/sculpt_paint/sculpt.c
@@ -376,7 +376,8 @@ static bool sculpt_stroke_is_dynamic_topology(
 
 /*** paint mesh ***/
 
-static void paint_mesh_restore_co_task_cb(void *userdata, const int n)
+static void paint_mesh_restore_co_task_cb(void *userdata, const int n,
+                                          const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -794,7 +795,8 @@ static float calc_symmetry_feather(Sculpt *sd, StrokeCache *cache)
  * \note These are all _very_ similar, when changing one, check others.
  * \{ */
 
-static void calc_area_normal_and_center_task_cb(void *userdata, const int n)
+static void calc_area_normal_and_center_task_cb(void *userdata, const int n,
+                                                const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -1626,7 +1628,7 @@ typedef struct {
 } SculptFindNearestToRayData;
 
 static void do_smooth_brush_mesh_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -1648,7 +1650,7 @@ static void do_smooth_brush_mesh_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, vd.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, smooth_mask ? 0.0f : (vd.mask ? *vd.mask : 0.0f), thread_id);
+			        vd.no, vd.fno, smooth_mask ? 0.0f : (vd.mask ? *vd.mask : 0.0f), tls->thread_id);
 			if (smooth_mask) {
 				float val = neighbor_average_mask(ss, vd.vert_indices[vd.i]) - *vd.mask;
 				val *= fade * bstrength;
@@ -1674,7 +1676,7 @@ static void do_smooth_brush_mesh_task_cb_ex(
 }
 
 static void do_smooth_brush_bmesh_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -1696,7 +1698,7 @@ static void do_smooth_brush_bmesh_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, vd.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, smooth_mask ? 0.0f : *vd.mask, thread_id);
+			        vd.no, vd.fno, smooth_mask ? 0.0f : *vd.mask, tls->thread_id);
 			if (smooth_mask) {
 				float val = bmesh_neighbor_average_mask(vd.bm_vert, vd.cd_vert_mask_offset) - *vd.mask;
 				val *= fade * bstrength;
@@ -1722,10 +1724,10 @@ static void do_smooth_brush_bmesh_task_cb_ex(
 }
 
 static void do_smooth_brush_multires_task_cb_ex(
-        void *userdata, void *userdata_chunk, const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
-	SculptDoBrushSmoothGridDataChunk *data_chunk = userdata_chunk;
+	SculptDoBrushSmoothGridDataChunk *data_chunk = tls->userdata_chunk;
 	SculptSession *ss = data->ob->sculpt;
 	Sculpt *sd = data->sd;
 	const Brush *brush = data->brush;
@@ -1837,7 +1839,7 @@ static void do_smooth_brush_multires_task_cb_ex(
 					const float strength_mask = (smooth_mask ? 0.0f : *mask);
 					const float fade = bstrength * tex_strength(
 					        ss, brush, co, sqrtf(test.dist),
-					        NULL, fno, strength_mask, thread_id);
+					        NULL, fno, strength_mask, tls->thread_id);
 					float f = 1.0f / 16.0f;
 
 					if (x == 0 || x == gridsize - 1)
@@ -1940,7 +1942,7 @@ static void do_smooth_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnod
 }
 
 static void do_mask_brush_draw_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -1958,7 +1960,7 @@ static void do_mask_brush_draw_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, vd.co)) {
 			const float fade = tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, 0.0f, thread_id);
+			        vd.no, vd.fno, 0.0f, tls->thread_id);
 
 			(*vd.mask) += fade * bstrength;
 			CLAMP(*vd.mask, 0, 1);
@@ -2000,7 +2002,7 @@ static void do_mask_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode)
 }
 
 static void do_draw_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2022,7 +2024,7 @@ static void do_draw_brush_task_cb_ex(
 			/* offset vertex */
 			const float fade = tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 			mul_v3_v3fl(proxy[vd.i], offset, fade);
 
@@ -2060,7 +2062,7 @@ static void do_draw_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode)
  * Used for 'SCULPT_TOOL_CREASE' and 'SCULPT_TOOL_BLOB'
  */
 static void do_crease_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2084,7 +2086,7 @@ static void do_crease_brush_task_cb_ex(
 			/* offset vertex */
 			const float fade = tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 			float val1[3];
 			float val2[3];
 
@@ -2154,7 +2156,7 @@ static void do_crease_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnod
 }
 
 static void do_pinch_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2175,7 +2177,7 @@ static void do_pinch_brush_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, vd.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 			float val[3];
 
 			sub_v3_v3v3(val, test.location, vd.co);
@@ -2205,7 +2207,7 @@ static void do_pinch_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode
 }
 
 static void do_grab_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2232,7 +2234,7 @@ static void do_grab_brush_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, orig_data.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, orig_data.co, sqrtf(test.dist),
-			        orig_data.no, NULL, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        orig_data.no, NULL, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 			mul_v3_v3fl(proxy[vd.i], grab_delta, fade);
 
@@ -2266,7 +2268,7 @@ static void do_grab_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode)
 }
 
 static void do_nudge_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2288,7 +2290,7 @@ static void do_nudge_brush_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, vd.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 			mul_v3_v3fl(proxy[vd.i], cono, fade);
 
@@ -2322,7 +2324,7 @@ static void do_nudge_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode
 }
 
 static void do_snake_hook_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2349,7 +2351,7 @@ static void do_snake_hook_brush_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, vd.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 			mul_v3_v3fl(proxy[vd.i], grab_delta, fade);
 
@@ -2428,7 +2430,7 @@ static void do_snake_hook_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int to
 }
 
 static void do_thumb_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2455,7 +2457,7 @@ static void do_thumb_brush_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, orig_data.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, orig_data.co, sqrtf(test.dist),
-			        orig_data.no, NULL, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        orig_data.no, NULL, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 			mul_v3_v3fl(proxy[vd.i], cono, fade);
 
@@ -2489,7 +2491,7 @@ static void do_thumb_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode
 }
 
 static void do_rotate_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2517,7 +2519,7 @@ static void do_rotate_brush_task_cb_ex(
 			float vec[3], rot[3][3];
 			const float fade = bstrength * tex_strength(
 			        ss, brush, orig_data.co, sqrtf(test.dist),
-			        orig_data.no, NULL, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        orig_data.no, NULL, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 			sub_v3_v3v3(vec, orig_data.co, ss->cache->location);
 			axis_angle_normalized_to_mat3(rot, ss->cache->sculpt_normal_symm, angle * fade);
@@ -2551,7 +2553,7 @@ static void do_rotate_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnod
 }
 
 static void do_layer_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2585,7 +2587,7 @@ static void do_layer_brush_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, orig_data.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 			float *disp = &layer_disp[vd.i];
 			float val[3];
 
@@ -2638,7 +2640,7 @@ static void do_layer_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode
 }
 
 static void do_inflate_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2659,7 +2661,7 @@ static void do_inflate_brush_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, vd.co)) {
 			const float fade = bstrength * tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 			float val[3];
 
 			if (vd.fno)
@@ -2802,7 +2804,7 @@ static float get_offset(Sculpt *sd, SculptSession *ss)
 }
 
 static void do_flatten_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2835,7 +2837,7 @@ static void do_flatten_brush_task_cb_ex(
 			if (plane_trim(ss->cache, brush, val)) {
 				const float fade = bstrength * tex_strength(
 				        ss, brush, vd.co, sqrtf(test.dist),
-				        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+				        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 				mul_v3_v3fl(proxy[vd.i], val, fade);
 
@@ -2880,7 +2882,7 @@ static void do_flatten_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totno
 }
 
 static void do_clay_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -2917,7 +2919,7 @@ static void do_clay_brush_task_cb_ex(
 					 * causes glitch with planes, see: T44390 */
 					const float fade = bstrength * tex_strength(
 					        ss, brush, vd.co, sqrtf(test.dist),
-					        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+					        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 					mul_v3_v3fl(proxy[vd.i], val, fade);
 
@@ -2966,7 +2968,7 @@ static void do_clay_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode)
 }
 
 static void do_clay_strips_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -3002,7 +3004,7 @@ static void do_clay_strips_brush_task_cb_ex(
 					 * causes glitch with planes, see: T44390 */
 					const float fade = bstrength * tex_strength(
 					        ss, brush, vd.co, ss->cache->radius * test.dist,
-					        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+					        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 					mul_v3_v3fl(proxy[vd.i], val, fade);
 
@@ -3076,7 +3078,7 @@ static void do_clay_strips_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int t
 }
 
 static void do_fill_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -3110,7 +3112,7 @@ static void do_fill_brush_task_cb_ex(
 				if (plane_trim(ss->cache, brush, val)) {
 					const float fade = bstrength * tex_strength(
 					        ss, brush, vd.co, sqrtf(test.dist),
-					        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+					        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 					mul_v3_v3fl(proxy[vd.i], val, fade);
 
@@ -3157,7 +3159,7 @@ static void do_fill_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnode)
 }
 
 static void do_scrape_brush_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -3190,7 +3192,7 @@ static void do_scrape_brush_task_cb_ex(
 				if (plane_trim(ss->cache, brush, val)) {
 					const float fade = bstrength * tex_strength(
 					        ss, brush, vd.co, sqrtf(test.dist),
-					        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+					        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 					mul_v3_v3fl(proxy[vd.i], val, fade);
 
@@ -3237,7 +3239,7 @@ static void do_scrape_brush(Sculpt *sd, Object *ob, PBVHNode **nodes, int totnod
 }
 
 static void do_gravity_task_cb_ex(
-        void *userdata, void *UNUSED(userdata_chunk), const int n, const int thread_id)
+        void *userdata, const int n, const ParallelRangeTLS *tls)
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -3257,7 +3259,7 @@ static void do_gravity_task_cb_ex(
 		if (sculpt_brush_test_sq_fn(&test, vd.co)) {
 			const float fade = tex_strength(
 			        ss, brush, vd.co, sqrtf(test.dist),
-			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, thread_id);
+			        vd.no, vd.fno, vd.mask ? *vd.mask : 0.0f, tls->thread_id);
 
 			mul_v3_v3fl(proxy[vd.i], offset, fade);
 
@@ -3392,7 +3394,8 @@ static void sculpt_topology_update(Sculpt *sd, Object *ob, Brush *brush, Unified
 	}
 }
 
-static void do_brush_action_task_cb(void *userdata, const int n)
+static void do_brush_action_task_cb(void *userdata, const int n,
+                                    const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 
@@ -3533,7 +3536,8 @@ static void sculpt_flush_pbvhvert_deform(Object *ob, PBVHVertexIter *vd)
 		copy_v3_v3(me->mvert[index].co, newco);
 }
 
-static void sculpt_combine_proxies_task_cb(void *userdata, const int n)
+static void sculpt_combine_proxies_task_cb(void *userdata, const int n,
+                                           const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
@@ -3628,7 +3632,8 @@ static void sculpt_update_keyblock(Object *ob)
 	}
 }
 
-static void sculpt_flush_stroke_deform_task_cb(void *userdata, const int n)
+static void sculpt_flush_stroke_deform_task_cb(void *userdata, const int n,
+                                               const ParallelRangeTLS *UNUSED(tls))
 {
 	SculptThreadedTaskData *data = userdata;
 	SculptSession *ss = data->ob->sculpt;
diff --git a/source/blender/editors/sculpt_paint/sculpt_undo.c b/source/blender/editors/sculpt_paint/sculpt_undo.c
index 44bd872d107..ed3acf0edbb 100644
--- a/source/blender/editors/sculpt_paint/sculpt_undo.c
+++ b/source/blender/editors/sculpt_paint/sculpt_undo.c
@@ -316,7 +316,8 @@ static bool sculpt_undo_restore_mask(bContext *C, DerivedMesh *dm, SculptUndoNod
 	return 1;
 }
 
-static void sculpt_undo_bmesh_restore_generic_task_cb(void *userdata, const int n)
+static void sculpt_undo_bmesh_restore_generic_task_cb(void *userdata, const int n,
+                                                      const ParallelRangeTLS *UNUSED(tls))
 {
 	PBVHNode **nodes = userdata;
 
diff --git a/source/blender/editors/space_sequencer/sequencer_scopes.c b/source/blender/editors/space_sequencer/sequencer_scopes.c
index 80cb42c0b3d..4c060f12a4b 100644
--- a/source/blender/editors/space_sequencer/sequencer_scopes.c
+++ b/source/blender/editors/space_sequencer/sequencer_scopes.c
@@ -459,13 +459,13 @@ typedef struct MakeHistogramViewData {
 } MakeHistogramViewData;
 
 static void make_histogram_view_from_ibuf_byte_cb_ex(
-        void *userdata, void *userdata_chunk, const int y, const int UNUSED(threadid))
+        void *userdata, const int y, const ParallelRangeTLS *tls)
 {
 	MakeHistogramViewData *data = userdata;
 	const ImBuf *ibuf = data->ibuf;
 	const unsigned char *src = (unsigned char *)ibuf->rect;
 
-	uint32_t (*cur_bins)[HIS_STEPS] = userdata_chunk;
+	uint32_t (*cur_bins)[HIS_STEPS] = tls->userdata_chunk;
 
 	for (int x = 0; x < ibuf->x; x++) {
 		const unsigned char *pixel = src + (y * ibuf->x + x) * 4;
@@ -548,13 +548,13 @@ BLI_INLINE int get_bin_float(float f)
 }
 
 static void make_histogram_view_from_ibuf_float_cb_ex(
-        void *userdata, void *userdata_chunk, const int y, const int UNUSED(threadid))
+        void *userdata, const int y, const ParallelRangeTLS *tls)
 {
 	const MakeHistogramViewData *data = userdata;
 	const ImBuf *ibuf = data->ibuf;
 	const float *src = ibuf->rect_float;
 
-	uint32_t (*cur_bins)[HIS_STEPS] = userdata_chunk;
+	uint32_t (*cur_bins)[HIS_STEPS] = tls->userdata_chunk;
 
 	for (int x = 0; x < ibuf->x; x++) {
 		const float *pixel = src + (y * ibuf->x + x) * 4;
diff --git a/source/blender/modifiers/intern/MOD_displace.c b/source/blender/modifiers/intern/MOD_displace.c
index fb8c0dd05a5..16fd3b098c9 100644
--- a/source/blender/modifiers/intern/MOD_displace.c
+++ b/source/blender/modifiers/intern/MOD_displace.c
@@ -213,7 +213,7 @@ typedef struct DisplaceUserdata {
 	float (*vert_clnors)[3];
 } DisplaceUserdata;
 
-static void displaceModifier_do_task(void *userdata, const int iter)
+static void displaceModifier_do_task(void *userdata, const int iter, const ParallelRangeTLS *UNUSED(tls))
 {
 	DisplaceUserdata *data = (DisplaceUserdata *)userdata;
 	DisplaceModifierData *dmd = data->dmd;
diff --git a/source/blender/modifiers/intern/MOD_meshdeform.c b/source/blender/modifiers/intern/MOD_meshdeform.c
index b1938395a7b..afae29da350 100644
--- a/source/blender/modifiers/intern/MOD_meshdeform.c
+++ b/source/blender/modifiers/intern/MOD_meshdeform.c
@@ -234,7 +234,8 @@ typedef struct MeshdeformUserdata {
 	float (*icagemat)[3];
 } MeshdeformUserdata;
 
-static void meshdeform_vert_task(void *userdata, const int iter)
+static void meshdeform_vert_task(void *userdata, const int iter,
+                                 const ParallelRangeTLS *UNUSED(tls))
 {
 	MeshdeformUserdata *data = userdata;
 	/*const*/ MeshDeformModifierData *mmd = data->mmd;
diff --git a/source/blender/modifiers/intern/MOD_ocean.c b/source/blender/modifiers/intern/MOD_ocean.c
index 8d8565e7e62..7fa539a0b5e 100644
--- a/source/blender/modifiers/intern/MOD_ocean.c
+++ b/source/blender/modifiers/intern/MOD_ocean.c
@@ -261,7 +261,8 @@ typedef struct GenerateOceanGeometryData {
 	float ix, iy;
 } GenerateOceanGeometryData;
 
-static void generate_ocean_geometry_vertices(void *userdata, const int y)
+static void generate_ocean_geometry_vertices(void *userdata, const int y,
+                                             const ParallelRangeTLS *UNUSED(tls))
 {
 	GenerateOceanGeometryData *gogd = userdata;
 	int x;
@@ -275,7 +276,8 @@ static void generate_ocean_geometry_vertices(void *userdata, const int y)
 	}
 }
 
-static void generate_ocean_geometry_polygons(void *userdata, const int y)
+static void generate_ocean_geometry_polygons(void *userdata, const int y,
+                                             const ParallelRangeTLS *UNUSED(tls))
 {
 	GenerateOceanGeometryData *gogd = userdata;
 	int x;
@@ -305,7 +307,8 @@ static void generate_ocean_geometry_polygons(void *userdata, const int y)
 	}
 }
 
-static void generate_ocean_geometry_uvs(void *userdata, const int y)
+static void generate_ocean_geometry_uvs(void *userdata, const int y,
+                                        const ParallelRangeTLS *UNUSED(tls))
 {
 	GenerateOceanGeometryData *gogd = userdata;
 	int x;
diff --git a/source/blender/modifiers/intern/MOD_surfacedeform.c b/source/blender/modifiers/intern/MOD_surfacedeform.c
index f8bad417f0b..d947ae148e9 100644
--- a/source/blender/modifiers/intern/MOD_surfacedeform.c
+++ b/source/blender/modifiers/intern/MOD_surfacedeform.c
@@ -753,7 +753,7 @@ BLI_INLINE float computeNormalDisplacement(const float point_co[3], const float
 	return normal_dist;
 }
 
-static void bindVert(void *userdata, void *UNUSED(userdata_chunk), const int index, const int UNUSED(threadid))
+static void bindVert(void *userdata, const int index, const ParallelRangeTLS *UNUSED(tls))
 {
 	SDefBindCalcData * const data = (SDefBindCalcData *)userdata;
 	float point_co[3];
@@ -1049,7 +1049,7 @@ static bool surfacedeformBind(SurfaceDeformModifierData *smd, float (*vertexCos)
 	return data.success == 1;
 }
 
-static void deformVert(void *userdata, void *UNUSED(userdata_chunk), const int index, const int UNUSED(threadid))
+static void deformVert(void *userdata, const int index, const ParallelRangeTLS *UNUSED(tls))
 {
 	const SDefDeformData * const data = (SDefDeformData *)userdata;
 	const SDefBind *sdbind = data->bind_verts[index].binds;
diff --git a/source/blender/modifiers/intern/MOD_uvwarp.c b/source/blender/modifiers/intern/MOD_uvwarp.c
index b3fc3842635..99dc7a7bd92 100644
--- a/source/blender/modifiers/intern/MOD_uvwarp.c
+++ b/source/blender/modifiers/intern/MOD_uvwarp.c
@@ -112,7 +112,7 @@ typedef struct UVWarpData {
 	int axis_v;
 } UVWarpData;
 
-static void uv_warp_compute(void *userdata, const int i)
+static void uv_warp_compute(void *userdata, const int i, const ParallelRangeTLS *UNUSED(tls))
 {
 	const UVWarpData *data = userdata;
 
diff --git a/source/blender/modifiers/intern/MOD_weightvgproximity.c b/source/blender/modifiers/intern/MOD_weightvgproximity.c
index 2ca380ba5c2..e26a3c49b7e 100644
--- a/source/blender/modifiers/intern/MOD_weightvgproximity.c
+++ b/source/blender/modifiers/intern/MOD_weightvgproximity.c
@@ -91,10 +91,10 @@ typedef struct Vert2GeomDataChunk {
 /**
  * Callback used by BLI_task 'for loop' helper.
  */
-static void vert2geom_task_cb_ex(void *userdata, void *userdata_chunk, const int iter, const int UNUSED(thread_id))
+static void vert2geom_task_cb_ex(void *userdata, const int iter, const ParallelRangeTLS *tls)
 {
 	Vert2GeomData *data = userdata;
-	Vert2GeomDataChunk *data_chunk = userdata_chunk;
+	Vert2GeomDataChunk *data_chunk = tls->userdata_chunk;
 
 	float tmp_co[3];
 	int i;
diff --git a/source/blender/render/intern/source/pointdensity.c b/source/blender/render/intern/source/pointdensity.c
index b215db49264..a33b6800df7 100644
--- a/source/blender/render/intern/source/pointdensity.c
+++ b/source/blender/render/intern/source/pointdensity.c
@@ -1007,7 +1007,8 @@ typedef struct SampleCallbackData {
 	float *values;
 } SampleCallbackData;
 
-static void point_density_sample_func(void *data_v, const int iter)
+static void point_density_sample_func(void *data_v, const int iter,
+                                      const ParallelRangeTLS *UNUSED(tls))
 {
 	SampleCallbackData *data = (SampleCallbackData *)data_v;