diff --git a/Marlin/Conditionals_post.h b/Marlin/Conditionals_post.h
index ca477f9f18..6eab11dac3 100644
--- a/Marlin/Conditionals_post.h
+++ b/Marlin/Conditionals_post.h
@@ -35,11 +35,14 @@
   #endif
 
   /**
-   * Axis lengths
+   * Axis lengths and center
    */
   #define X_MAX_LENGTH (X_MAX_POS - (X_MIN_POS))
   #define Y_MAX_LENGTH (Y_MAX_POS - (Y_MIN_POS))
   #define Z_MAX_LENGTH (Z_MAX_POS - (Z_MIN_POS))
+  #define X_CENTER float((X_MIN_POS + X_MAX_POS) * 0.5)
+  #define Y_CENTER float((Y_MIN_POS + Y_MAX_POS) * 0.5)
+  #define Z_CENTER float((Z_MIN_POS + Z_MAX_POS) * 0.5)
 
   /**
    * CoreXY and CoreXZ
@@ -127,6 +130,8 @@
    */
   #define HAS_PROBING_PROCEDURE (ENABLED(AUTO_BED_LEVELING_FEATURE) || ENABLED(Z_MIN_PROBE_REPEATABILITY_TEST))
 
+  #define HOMING_Z_WITH_PROBE (HAS_BED_PROBE && Z_HOME_DIR < 0 && ENABLED(Z_MIN_PROBE_USES_Z_MIN_ENDSTOP_PIN))
+
   // Boundaries for probing based on set limits
   #define MIN_PROBE_X (max(X_MIN_POS, X_MIN_POS + X_PROBE_OFFSET_FROM_EXTRUDER))
   #define MAX_PROBE_X (min(X_MAX_POS, X_MAX_POS + X_PROBE_OFFSET_FROM_EXTRUDER))
diff --git a/Marlin/Marlin_main.cpp b/Marlin/Marlin_main.cpp
index 33257e2dbb..26b719acf6 100644
--- a/Marlin/Marlin_main.cpp
+++ b/Marlin/Marlin_main.cpp
@@ -1586,7 +1586,7 @@ static void set_axis_is_at_home(AxisEnum axis) {
 
     if (axis == Z_AXIS) {
       #if HAS_BED_PROBE && Z_HOME_DIR < 0
-        #if DISABLED(Z_MIN_PROBE_ENDSTOP)
+        #if HOMING_Z_WITH_PROBE
           current_position[Z_AXIS] -= zprobe_zoffset;
           #if ENABLED(DEBUG_LEVELING_FEATURE)
             if (DEBUGGING(LEVELING)) {
@@ -2049,8 +2049,8 @@ static void clean_up_after_endstop_or_probe_move() {
     #endif
   #endif
 
-  #define DEPLOY_PROBE() set_probe_deployed( true )
-  #define STOW_PROBE() set_probe_deployed( false )
+  #define DEPLOY_PROBE() set_probe_deployed(true)
+  #define STOW_PROBE() set_probe_deployed(false)
 
   // returns false for ok and true for failure
   static bool set_probe_deployed(bool deploy) {
@@ -2073,8 +2073,8 @@ static void clean_up_after_endstop_or_probe_move() {
       if (axis_unhomed_error(true, true,  true )) { stop(); return true; }
     #endif
 
-    float oldXpos = current_position[X_AXIS]; // save x position
-    float oldYpos = current_position[Y_AXIS]; // save y position
+    float oldXpos = current_position[X_AXIS],
+          oldYpos = current_position[Y_AXIS];
 
     #ifdef _TRIGGERED_WHEN_STOWED_TEST
 
@@ -2430,10 +2430,10 @@ static void do_homing_move(AxisEnum axis, float where, float fr_mm_s = 0.0) {
 #define HOMEAXIS(LETTER) homeaxis(LETTER##_AXIS)
 
 static void homeaxis(AxisEnum axis) {
-  #define HOMEAXIS_DO(LETTER) \
-    ((LETTER##_MIN_PIN > -1 && LETTER##_HOME_DIR==-1) || (LETTER##_MAX_PIN > -1 && LETTER##_HOME_DIR==1))
+  #define CAN_HOME(A) \
+    (axis == A##_AXIS && ((A##_MIN_PIN > -1 && A##_HOME_DIR < 0) || (A##_MAX_PIN > -1 && A##_HOME_DIR > 0)))
 
-  if (!(axis == X_AXIS ? HOMEAXIS_DO(X) : axis == Y_AXIS ? HOMEAXIS_DO(Y) : axis == Z_AXIS ? HOMEAXIS_DO(Z) : false)) return;
+  if (!CAN_HOME(X) && !CAN_HOME(Y) && !CAN_HOME(Z)) return;
 
   #if ENABLED(DEBUG_LEVELING_FEATURE)
     if (DEBUGGING(LEVELING)) {
@@ -2449,7 +2449,7 @@ static void homeaxis(AxisEnum axis) {
     home_dir(axis);
 
   // Homing Z towards the bed? Deploy the Z probe or endstop.
-  #if HAS_BED_PROBE && Z_HOME_DIR < 0 && DISABLED(Z_MIN_PROBE_ENDSTOP)
+  #if HOMING_Z_WITH_PROBE
     if (axis == Z_AXIS) {
       #if ENABLED(DEBUG_LEVELING_FEATURE)
         if (DEBUGGING(LEVELING)) SERIAL_ECHOPGM("> ");
@@ -2532,7 +2532,7 @@ static void homeaxis(AxisEnum axis) {
   #endif
 
   // Put away the Z probe
-  #if HAS_BED_PROBE && Z_HOME_DIR < 0 && DISABLED(Z_MIN_PROBE_ENDSTOP)
+  #if HOMING_Z_WITH_PROBE
     if (axis == Z_AXIS) {
       #if ENABLED(DEBUG_LEVELING_FEATURE)
         if (DEBUGGING(LEVELING)) SERIAL_ECHOPGM("> ");
@@ -3104,9 +3104,7 @@ inline void gcode_G28() {
         #if ENABLED(Z_SAFE_HOMING)
 
           #if ENABLED(DEBUG_LEVELING_FEATURE)
-            if (DEBUGGING(LEVELING)) {
-              SERIAL_ECHOLNPGM("> Z_SAFE_HOMING >>>");
-            }
+            if (DEBUGGING(LEVELING)) SERIAL_ECHOLNPGM("> Z_SAFE_HOMING >>>");
           #endif
 
           if (home_all_axis) {
@@ -3127,10 +3125,7 @@ inline void gcode_G28() {
             destination[Z_AXIS] = current_position[Z_AXIS]; // Z is already at the right height
 
             #if ENABLED(DEBUG_LEVELING_FEATURE)
-              if (DEBUGGING(LEVELING)) {
-                DEBUG_POS("> Z_SAFE_HOMING > home_all_axis", current_position);
-                DEBUG_POS("> Z_SAFE_HOMING > home_all_axis", destination);
-              }
+              if (DEBUGGING(LEVELING)) DEBUG_POS("> Z_SAFE_HOMING > home_all_axis", destination);
             #endif
 
             // Move in the XY plane
diff --git a/Marlin/planner.cpp b/Marlin/planner.cpp
index dd065988fc..3a1ffd1393 100644
--- a/Marlin/planner.cpp
+++ b/Marlin/planner.cpp
@@ -203,9 +203,8 @@ void Planner::calculate_trapezoid_for_block(block_t* block, float entry_factor,
 
 
 // The kernel called by recalculate() when scanning the plan from last to first entry.
-void Planner::reverse_pass_kernel(block_t* previous, block_t* current, block_t* next) {
+void Planner::reverse_pass_kernel(block_t* current, block_t* next) {
   if (!current) return;
-  UNUSED(previous);
 
   if (next) {
     // If entry speed is already at the maximum entry speed, no need to recheck. Block is cruising.
@@ -250,15 +249,14 @@ void Planner::reverse_pass() {
       block[2] = block[1];
       block[1] = block[0];
       block[0] = &block_buffer[b];
-      reverse_pass_kernel(block[0], block[1], block[2]);
+      reverse_pass_kernel(block[1], block[2]);
     }
   }
 }
 
 // The kernel called by recalculate() when scanning the plan from first to last entry.
-void Planner::forward_pass_kernel(block_t* previous, block_t* current, block_t* next) {
+void Planner::forward_pass_kernel(block_t* previous, block_t* current) {
   if (!previous) return;
-  UNUSED(next);
 
   // If the previous block is an acceleration block, but it is not long enough to complete the
   // full speed change within the block, we need to adjust the entry speed accordingly. Entry
@@ -288,9 +286,9 @@ void Planner::forward_pass() {
     block[0] = block[1];
     block[1] = block[2];
     block[2] = &block_buffer[b];
-    forward_pass_kernel(block[0], block[1], block[2]);
+    forward_pass_kernel(block[0], block[1]);
   }
-  forward_pass_kernel(block[1], block[2], NULL);
+  forward_pass_kernel(block[1], block[2]);
 }
 
 /**
diff --git a/Marlin/planner.h b/Marlin/planner.h
index 740553668a..e1159294a0 100644
--- a/Marlin/planner.h
+++ b/Marlin/planner.h
@@ -320,8 +320,8 @@ class Planner {
 
     static void calculate_trapezoid_for_block(block_t* block, float entry_factor, float exit_factor);
 
-    static void reverse_pass_kernel(block_t* previous, block_t* current, block_t* next);
-    static void forward_pass_kernel(block_t* previous, block_t* current, block_t* next);
+    static void reverse_pass_kernel(block_t* current, block_t* next);
+    static void forward_pass_kernel(block_t* previous, block_t* current);
 
     static void reverse_pass();
     static void forward_pass();
diff --git a/Marlin/stepper.cpp b/Marlin/stepper.cpp
index 64805f1acc..51459602f0 100644
--- a/Marlin/stepper.cpp
+++ b/Marlin/stepper.cpp
@@ -87,7 +87,7 @@ long  Stepper::counter_X = 0,
       Stepper::counter_Z = 0,
       Stepper::counter_E = 0;
 
-volatile unsigned long Stepper::step_events_completed = 0; // The number of step events executed in the current block
+volatile uint32_t Stepper::step_events_completed = 0; // The number of step events executed in the current block
 
 #if ENABLED(ADVANCE) || ENABLED(LIN_ADVANCE)
 
@@ -372,6 +372,7 @@ void Stepper::isr() {
     ) endstops.update();
 
     // Take multiple steps per interrupt (For high speed moves)
+    bool all_steps_done = false;
     for (int8_t i = 0; i < step_loops; i++) {
       #ifndef USBCON
         customizedSerial.checkRx(); // Check for serial chars.
@@ -385,7 +386,7 @@ void Stepper::isr() {
           #if DISABLED(MIXING_EXTRUDER)
             // Don't step E here for mixing extruder
             count_position[E_AXIS] += count_direction[E_AXIS];
-            e_steps[TOOL_E_INDEX] += motor_direction(E_AXIS) ? -1 : 1;
+            motor_direction(E_AXIS) ? --e_steps[TOOL_E_INDEX] : ++e_steps[TOOL_E_INDEX];
           #endif
         }
 
@@ -449,10 +450,12 @@ void Stepper::isr() {
       #define _APPLY_STEP(AXIS) AXIS ##_APPLY_STEP
       #define _INVERT_STEP_PIN(AXIS) INVERT_## AXIS ##_STEP_PIN
 
+      // Advance the Bresenham counter; start a pulse if the axis needs a step
       #define PULSE_START(AXIS) \
         _COUNTER(AXIS) += current_block->steps[_AXIS(AXIS)]; \
         if (_COUNTER(AXIS) > 0) { _APPLY_STEP(AXIS)(!_INVERT_STEP_PIN(AXIS),0); }
 
+      // Stop an active pulse, reset the Bresenham counter, update the position
       #define PULSE_STOP(AXIS) \
         if (_COUNTER(AXIS) > 0) { \
           _COUNTER(AXIS) -= current_block->step_event_count; \
@@ -460,6 +463,7 @@ void Stepper::isr() {
           _APPLY_STEP(AXIS)(_INVERT_STEP_PIN(AXIS),0); \
         }
 
+      // If a minimum pulse time was specified get the CPU clock
       #if MINIMUM_STEPPER_PULSE > 0
         static uint32_t pulse_start;
         pulse_start = TCNT0;
@@ -475,6 +479,7 @@ void Stepper::isr() {
         PULSE_START(Z);
       #endif
 
+      // For non-advance use linear interpolation for E also
       #if DISABLED(ADVANCE) && DISABLED(LIN_ADVANCE)
         #if ENABLED(MIXING_EXTRUDER)
           // Keep updating the single E axis
@@ -491,6 +496,7 @@ void Stepper::isr() {
         #endif
       #endif // !ADVANCE && !LIN_ADVANCE
 
+      // For a minimum pulse time wait before stopping pulses
       #if MINIMUM_STEPPER_PULSE > 0
         #define CYCLES_EATEN_BY_CODE 10
         while ((uint32_t)(TCNT0 - pulse_start) < (MINIMUM_STEPPER_PULSE * (F_CPU / 1000000UL)) - CYCLES_EATEN_BY_CODE) { /* nada */ }
@@ -524,18 +530,20 @@ void Stepper::isr() {
         #endif
       #endif // !ADVANCE && !LIN_ADVANCE
 
-      step_events_completed++;
-      if (step_events_completed >= current_block->step_event_count) break;
+      if (++step_events_completed >= current_block->step_event_count) {
+        all_steps_done = true;
+        break;
+      }
     }
 
     #if ENABLED(ADVANCE) || ENABLED(LIN_ADVANCE)
-      // If we have esteps to execute, fire the next ISR "now"
+      // If we have esteps to execute, fire the next advance_isr "now"
       if (e_steps[TOOL_E_INDEX]) OCR0A = TCNT0 + 2;
     #endif
 
     // Calculate new timer value
-    unsigned short timer, step_rate;
-    if (step_events_completed <= (unsigned long)current_block->accelerate_until) {
+    uint16_t timer, step_rate;
+    if (step_events_completed <= (uint32_t)current_block->accelerate_until) {
 
       MultiU24X32toH16(acc_step_rate, acceleration_time, current_block->acceleration_rate);
       acc_step_rate += current_block->initial_rate;
@@ -551,14 +559,14 @@ void Stepper::isr() {
       #if ENABLED(LIN_ADVANCE)
 
         if (current_block->use_advance_lead)
-          current_estep_rate[TOOL_E_INDEX] = ((unsigned long)acc_step_rate * current_block->e_speed_multiplier8) >> 8;
+          current_estep_rate[TOOL_E_INDEX] = ((uint32_t)acc_step_rate * current_block->e_speed_multiplier8) >> 8;
 
         if (current_block->use_advance_lead) {
           #if ENABLED(MIXING_EXTRUDER)
             MIXING_STEPPERS_LOOP(j)
-              current_estep_rate[j] = ((unsigned long)acc_step_rate * current_block->e_speed_multiplier8 * current_block->step_event_count / current_block->mix_event_count[j]) >> 8;
+              current_estep_rate[j] = ((uint32_t)acc_step_rate * current_block->e_speed_multiplier8 * current_block->step_event_count / current_block->mix_event_count[j]) >> 8;
           #else
-            current_estep_rate[TOOL_E_INDEX] = ((unsigned long)acc_step_rate * current_block->e_speed_multiplier8) >> 8;
+            current_estep_rate[TOOL_E_INDEX] = ((uint32_t)acc_step_rate * current_block->e_speed_multiplier8) >> 8;
           #endif
         }
 
@@ -588,10 +596,10 @@ void Stepper::isr() {
         eISR_Rate = (timer >> 2) * step_loops / abs(e_steps[TOOL_E_INDEX]);
       #endif
     }
-    else if (step_events_completed > (unsigned long)current_block->decelerate_after) {
+    else if (step_events_completed > (uint32_t)current_block->decelerate_after) {
       MultiU24X32toH16(step_rate, deceleration_time, current_block->acceleration_rate);
 
-      if (step_rate <= acc_step_rate) { // Still decelerating?
+      if (step_rate < acc_step_rate) { // Still decelerating?
         step_rate = acc_step_rate - step_rate;
         NOLESS(step_rate, current_block->final_rate);
       }
@@ -608,9 +616,9 @@ void Stepper::isr() {
         if (current_block->use_advance_lead) {
           #if ENABLED(MIXING_EXTRUDER)
             MIXING_STEPPERS_LOOP(j)
-              current_estep_rate[j] = ((unsigned long)step_rate * current_block->e_speed_multiplier8 * current_block->step_event_count / current_block->mix_event_count[j]) >> 8;
+              current_estep_rate[j] = ((uint32_t)step_rate * current_block->e_speed_multiplier8 * current_block->step_event_count / current_block->mix_event_count[j]) >> 8;
           #else
-            current_estep_rate[TOOL_E_INDEX] = ((unsigned long)step_rate * current_block->e_speed_multiplier8) >> 8;
+            current_estep_rate[TOOL_E_INDEX] = ((uint32_t)step_rate * current_block->e_speed_multiplier8) >> 8;
           #endif
         }
 
@@ -654,10 +662,10 @@ void Stepper::isr() {
       step_loops = step_loops_nominal;
     }
 
-    OCR1A = (OCR1A < (TCNT1 + 16)) ? (TCNT1 + 16) : OCR1A;
+    NOLESS(OCR1A, TCNT1 + 16);
 
     // If current block is finished, reset pointer
-    if (step_events_completed >= current_block->step_event_count) {
+    if (all_steps_done) {
       current_block = NULL;
       planner.discard_current_block();
     }
@@ -675,29 +683,61 @@ void Stepper::isr() {
     old_OCR0A += eISR_Rate;
     OCR0A = old_OCR0A;
 
-    #define STEP_E_ONCE(INDEX) \
-      if (e_steps[INDEX] != 0) { \
-        E## INDEX ##_STEP_WRITE(INVERT_E_STEP_PIN); \
-        if (e_steps[INDEX] < 0) { \
-          E## INDEX ##_DIR_WRITE(INVERT_E## INDEX ##_DIR); \
-          e_steps[INDEX]++; \
-        } \
-        else { \
-          E## INDEX ##_DIR_WRITE(!INVERT_E## INDEX ##_DIR); \
-          e_steps[INDEX]--; \
-        } \
+    #define SET_E_STEP_DIR(INDEX) \
+      E## INDEX ##_DIR_WRITE(e_steps[INDEX] <= 0 ? INVERT_E## INDEX ##_DIR : !INVERT_E## INDEX ##_DIR)
+
+    #define START_E_PULSE(INDEX) \
+      if (e_steps[INDEX]) E## INDEX ##_STEP_WRITE(INVERT_E_STEP_PIN)
+
+    #define STOP_E_PULSE(INDEX) \
+      if (e_steps[INDEX]) { \
+        e_steps[INDEX] < 0 ? ++e_steps[INDEX] : --e_steps[INDEX]; \
         E## INDEX ##_STEP_WRITE(!INVERT_E_STEP_PIN); \
       }
 
+    SET_E_STEP_DIR(0);
+    #if E_STEPPERS > 1
+      SET_E_STEP_DIR(1);
+      #if E_STEPPERS > 2
+        SET_E_STEP_DIR(2);
+        #if E_STEPPERS > 3
+          SET_E_STEP_DIR(3);
+        #endif
+      #endif
+    #endif
+
     // Step all E steppers that have steps
     for (uint8_t i = 0; i < step_loops; i++) {
-      STEP_E_ONCE(0);
+
+      #if MINIMUM_STEPPER_PULSE > 0
+        static uint32_t pulse_start;
+        pulse_start = TCNT0;
+      #endif
+
+      START_E_PULSE(0);
       #if E_STEPPERS > 1
-        STEP_E_ONCE(1);
+        START_E_PULSE(1);
         #if E_STEPPERS > 2
-          STEP_E_ONCE(2);
+          START_E_PULSE(2);
           #if E_STEPPERS > 3
-            STEP_E_ONCE(3);
+            START_E_PULSE(3);
+          #endif
+        #endif
+      #endif
+
+      // For a minimum pulse time wait before stopping pulses
+      #if MINIMUM_STEPPER_PULSE > 0
+        #define CYCLES_EATEN_BY_E 10
+        while ((uint32_t)(TCNT0 - pulse_start) < (MINIMUM_STEPPER_PULSE * (F_CPU / 1000000UL)) - CYCLES_EATEN_BY_E) { /* nada */ }
+      #endif
+
+      STOP_E_PULSE(0);
+      #if E_STEPPERS > 1
+        STOP_E_PULSE(1);
+        #if E_STEPPERS > 2
+          STOP_E_PULSE(2);
+          #if E_STEPPERS > 3
+            STOP_E_PULSE(3);
           #endif
         #endif
       #endif
diff --git a/Marlin/stepper.h b/Marlin/stepper.h
index 177ccf182e..b38b0e8f45 100644
--- a/Marlin/stepper.h
+++ b/Marlin/stepper.h
@@ -102,7 +102,7 @@ class Stepper {
 
     // Counter variables for the Bresenham line tracer
     static long counter_X, counter_Y, counter_Z, counter_E;
-    static volatile unsigned long step_events_completed; // The number of step events executed in the current block
+    static volatile uint32_t step_events_completed; // The number of step events executed in the current block
 
     #if ENABLED(ADVANCE) || ENABLED(LIN_ADVANCE)
       static unsigned char old_OCR0A;