♻️ Planner flags refactor
This commit is contained in:
parent
83784bd8b7
commit
05bdc5640d
@ -315,7 +315,7 @@ void plan_arc(
|
||||
// Compute exact location by applying transformation matrix from initial radius vector(=-offset).
|
||||
// To reduce stuttering, the sin and cos could be computed at different times.
|
||||
// For now, compute both at the same time.
|
||||
const float cos_Ti = cos(i * theta_per_segment), sin_Ti = sin(i * theta_per_segment);
|
||||
const float Ti = i * theta_per_segment, cos_Ti = cos(Ti), sin_Ti = sin(Ti);
|
||||
rvec.a = -offset[0] * cos_Ti + offset[1] * sin_Ti;
|
||||
rvec.b = -offset[0] * sin_Ti - offset[1] * cos_Ti;
|
||||
}
|
||||
|
@ -50,7 +50,7 @@ void GcodeSuite::G6() {
|
||||
// No speed is set, can't schedule the move
|
||||
if (!planner.last_page_step_rate) return;
|
||||
|
||||
const page_idx_t page_idx = (page_idx_t) parser.value_ulong();
|
||||
const page_idx_t page_idx = (page_idx_t)parser.value_ulong();
|
||||
|
||||
uint16_t num_steps = DirectStepping::Config::TOTAL_STEPS;
|
||||
if (parser.seen('S')) num_steps = parser.value_ushort();
|
||||
|
@ -90,7 +90,7 @@ void GcodeSuite::M106() {
|
||||
// Set speed, with constraint
|
||||
thermalManager.set_fan_speed(pfan, speed);
|
||||
|
||||
TERN_(LASER_SYNCHRONOUS_M106_M107, planner.buffer_sync_block(BLOCK_FLAG_SYNC_FANS));
|
||||
TERN_(LASER_SYNCHRONOUS_M106_M107, planner.buffer_sync_block(BLOCK_BIT_SYNC_FANS));
|
||||
|
||||
if (TERN0(DUAL_X_CARRIAGE, idex_is_duplicating())) // pfan == 0 when duplicating
|
||||
thermalManager.set_fan_speed(1 - pfan, speed);
|
||||
@ -111,7 +111,7 @@ void GcodeSuite::M107() {
|
||||
if (TERN0(DUAL_X_CARRIAGE, idex_is_duplicating())) // pfan == 0 when duplicating
|
||||
thermalManager.set_fan_speed(1 - pfan, 0);
|
||||
|
||||
TERN_(LASER_SYNCHRONOUS_M106_M107, planner.buffer_sync_block(BLOCK_FLAG_SYNC_FANS));
|
||||
TERN_(LASER_SYNCHRONOUS_M106_M107, planner.buffer_sync_block(BLOCK_BIT_SYNC_FANS));
|
||||
}
|
||||
|
||||
#endif // HAS_FAN
|
||||
|
@ -739,7 +739,7 @@ block_t* Planner::get_current_block() {
|
||||
block_t * const block = &block_buffer[block_buffer_tail];
|
||||
|
||||
// No trapezoid calculated? Don't execute yet.
|
||||
if (TEST(block->flag, BLOCK_BIT_RECALCULATE)) return nullptr;
|
||||
if (block->flag.recalculate) return nullptr;
|
||||
|
||||
// We can't be sure how long an active block will take, so don't count it.
|
||||
TERN_(HAS_WIRED_LCD, block_buffer_runtime_us -= block->segment_time_us);
|
||||
@ -948,7 +948,7 @@ void Planner::reverse_pass_kernel(block_t * const current, const block_t * const
|
||||
|
||||
// Compute maximum entry speed decelerating over the current block from its exit speed.
|
||||
// If not at the maximum entry speed, or the previous block entry speed changed
|
||||
if (current->entry_speed_sqr != max_entry_speed_sqr || (next && TEST(next->flag, BLOCK_BIT_RECALCULATE))) {
|
||||
if (current->entry_speed_sqr != max_entry_speed_sqr || (next && next->flag.recalculate)) {
|
||||
|
||||
// If nominal length true, max junction speed is guaranteed to be reached.
|
||||
// If a block can de/ac-celerate from nominal speed to zero within the length of the block, then
|
||||
@ -958,14 +958,14 @@ void Planner::reverse_pass_kernel(block_t * const current, const block_t * const
|
||||
// the reverse and forward planners, the corresponding block junction speed will always be at the
|
||||
// the maximum junction speed and may always be ignored for any speed reduction checks.
|
||||
|
||||
const float new_entry_speed_sqr = TEST(current->flag, BLOCK_BIT_NOMINAL_LENGTH)
|
||||
const float new_entry_speed_sqr = current->flag.nominal_length
|
||||
? max_entry_speed_sqr
|
||||
: _MIN(max_entry_speed_sqr, max_allowable_speed_sqr(-current->acceleration, next ? next->entry_speed_sqr : sq(float(MINIMUM_PLANNER_SPEED)), current->millimeters));
|
||||
if (current->entry_speed_sqr != new_entry_speed_sqr) {
|
||||
|
||||
// Need to recalculate the block speed - Mark it now, so the stepper
|
||||
// ISR does not consume the block before being recalculated
|
||||
SBI(current->flag, BLOCK_BIT_RECALCULATE);
|
||||
current->flag.recalculate = true;
|
||||
|
||||
// But there is an inherent race condition here, as the block may have
|
||||
// become BUSY just before being marked RECALCULATE, so check for that!
|
||||
@ -973,7 +973,7 @@ void Planner::reverse_pass_kernel(block_t * const current, const block_t * const
|
||||
// Block became busy. Clear the RECALCULATE flag (no point in
|
||||
// recalculating BUSY blocks). And don't set its speed, as it can't
|
||||
// be updated at this time.
|
||||
CBI(current->flag, BLOCK_BIT_RECALCULATE);
|
||||
current->flag.recalculate = false;
|
||||
}
|
||||
else {
|
||||
// Block is not BUSY so this is ahead of the Stepper ISR:
|
||||
@ -1011,8 +1011,8 @@ void Planner::reverse_pass() {
|
||||
// Perform the reverse pass
|
||||
block_t *current = &block_buffer[block_index];
|
||||
|
||||
// Only consider non sync-and-page blocks
|
||||
if (!(current->flag & BLOCK_MASK_SYNC) && !IS_PAGE(current)) {
|
||||
// Only process movement blocks
|
||||
if (current->is_move()) {
|
||||
reverse_pass_kernel(current, next);
|
||||
next = current;
|
||||
}
|
||||
@ -1041,8 +1041,7 @@ void Planner::forward_pass_kernel(const block_t * const previous, block_t * cons
|
||||
// change, adjust the entry speed accordingly. Entry speeds have already been reset,
|
||||
// maximized, and reverse-planned. If nominal length is set, max junction speed is
|
||||
// guaranteed to be reached. No need to recheck.
|
||||
if (!TEST(previous->flag, BLOCK_BIT_NOMINAL_LENGTH) &&
|
||||
previous->entry_speed_sqr < current->entry_speed_sqr) {
|
||||
if (!previous->flag.nominal_length && previous->entry_speed_sqr < current->entry_speed_sqr) {
|
||||
|
||||
// Compute the maximum allowable speed
|
||||
const float new_entry_speed_sqr = max_allowable_speed_sqr(-previous->acceleration, previous->entry_speed_sqr, previous->millimeters);
|
||||
@ -1052,7 +1051,7 @@ void Planner::forward_pass_kernel(const block_t * const previous, block_t * cons
|
||||
|
||||
// Mark we need to recompute the trapezoidal shape, and do it now,
|
||||
// so the stepper ISR does not consume the block before being recalculated
|
||||
SBI(current->flag, BLOCK_BIT_RECALCULATE);
|
||||
current->flag.recalculate = true;
|
||||
|
||||
// But there is an inherent race condition here, as the block maybe
|
||||
// became BUSY, just before it was marked as RECALCULATE, so check
|
||||
@ -1061,7 +1060,7 @@ void Planner::forward_pass_kernel(const block_t * const previous, block_t * cons
|
||||
// Block became busy. Clear the RECALCULATE flag (no point in
|
||||
// recalculating BUSY blocks and don't set its speed, as it can't
|
||||
// be updated at this time.
|
||||
CBI(current->flag, BLOCK_BIT_RECALCULATE);
|
||||
current->flag.recalculate = false;
|
||||
}
|
||||
else {
|
||||
// Block is not BUSY, we won the race against the Stepper ISR:
|
||||
@ -1106,8 +1105,8 @@ void Planner::forward_pass() {
|
||||
// Perform the forward pass
|
||||
block = &block_buffer[block_index];
|
||||
|
||||
// Skip SYNC and page blocks
|
||||
if (!(block->flag & BLOCK_MASK_SYNC) && !IS_PAGE(block)) {
|
||||
// Only process movement blocks
|
||||
if (block->is_move()) {
|
||||
// If there's no previous block or the previous block is not
|
||||
// BUSY (thus, modifiable) run the forward_pass_kernel. Otherwise,
|
||||
// the previous block became BUSY, so assume the current block's
|
||||
@ -1131,9 +1130,10 @@ void Planner::recalculate_trapezoids() {
|
||||
// The tail may be changed by the ISR so get a local copy.
|
||||
uint8_t block_index = block_buffer_tail,
|
||||
head_block_index = block_buffer_head;
|
||||
// Since there could be a sync block in the head of the queue, and the
|
||||
|
||||
// Since there could be non-move blocks in the head of the queue, and the
|
||||
// next loop must not recalculate the head block (as it needs to be
|
||||
// specially handled), scan backwards to the first non-SYNC block.
|
||||
// specially handled), scan backwards to the first move block.
|
||||
while (head_block_index != block_index) {
|
||||
|
||||
// Go back (head always point to the first free block)
|
||||
@ -1142,8 +1142,8 @@ void Planner::recalculate_trapezoids() {
|
||||
// Get the pointer to the block
|
||||
block_t *prev = &block_buffer[prev_index];
|
||||
|
||||
// If not dealing with a sync block, we are done. The last block is not a SYNC block
|
||||
if (!(prev->flag & BLOCK_MASK_SYNC)) break;
|
||||
// It the block is a move, we're done with this loop
|
||||
if (prev->is_move()) break;
|
||||
|
||||
// Examine the previous block. This and all following are SYNC blocks
|
||||
head_block_index = prev_index;
|
||||
@ -1156,18 +1156,17 @@ void Planner::recalculate_trapezoids() {
|
||||
|
||||
next = &block_buffer[block_index];
|
||||
|
||||
// Skip sync and page blocks
|
||||
if (!(next->flag & BLOCK_MASK_SYNC) && !IS_PAGE(next)) {
|
||||
// Only process movement blocks
|
||||
if (next->is_move()) {
|
||||
next_entry_speed = SQRT(next->entry_speed_sqr);
|
||||
|
||||
if (block) {
|
||||
// Recalculate if current block entry or exit junction speed has changed.
|
||||
if (TEST(block->flag, BLOCK_BIT_RECALCULATE) || TEST(next->flag, BLOCK_BIT_RECALCULATE)) {
|
||||
|
||||
// Mark the current block as RECALCULATE, to protect it from the Stepper ISR running it.
|
||||
// Note that due to the above condition, there's a chance the current block isn't marked as
|
||||
// RECALCULATE yet, but the next one is. That's the reason for the following line.
|
||||
SBI(block->flag, BLOCK_BIT_RECALCULATE);
|
||||
// If the next block is marked to RECALCULATE, also mark the previously-fetched one
|
||||
if (next->flag.recalculate) block->flag.recalculate = true;
|
||||
|
||||
// Recalculate if current block entry or exit junction speed has changed.
|
||||
if (block->flag.recalculate) {
|
||||
|
||||
// But there is an inherent race condition here, as the block maybe
|
||||
// became BUSY, just before it was marked as RECALCULATE, so check
|
||||
@ -1190,7 +1189,7 @@ void Planner::recalculate_trapezoids() {
|
||||
|
||||
// Reset current only to ensure next trapezoid is computed - The
|
||||
// stepper is free to use the block from now on.
|
||||
CBI(block->flag, BLOCK_BIT_RECALCULATE);
|
||||
block->flag.recalculate = false;
|
||||
}
|
||||
}
|
||||
|
||||
@ -1204,10 +1203,10 @@ void Planner::recalculate_trapezoids() {
|
||||
// Last/newest block in buffer. Exit speed is set with MINIMUM_PLANNER_SPEED. Always recalculated.
|
||||
if (next) {
|
||||
|
||||
// Mark the next(last) block as RECALCULATE, to prevent the Stepper ISR running it.
|
||||
// Mark the last block as RECALCULATE, to prevent the Stepper ISR running it.
|
||||
// As the last block is always recalculated here, there is a chance the block isn't
|
||||
// marked as RECALCULATE yet. That's the reason for the following line.
|
||||
SBI(next->flag, BLOCK_BIT_RECALCULATE);
|
||||
block->flag.recalculate = true;
|
||||
|
||||
// But there is an inherent race condition here, as the block maybe
|
||||
// became BUSY, just before it was marked as RECALCULATE, so check
|
||||
@ -1229,7 +1228,7 @@ void Planner::recalculate_trapezoids() {
|
||||
|
||||
// Reset next only to ensure its trapezoid is computed - The stepper is free to use
|
||||
// the block from now on.
|
||||
CBI(next->flag, BLOCK_BIT_RECALCULATE);
|
||||
next->flag.recalculate = false;
|
||||
}
|
||||
}
|
||||
|
||||
@ -1460,7 +1459,7 @@ void Planner::check_axes_activity() {
|
||||
for (uint8_t b = block_buffer_tail; b != block_buffer_head; b = next_block_index(b)) {
|
||||
const block_t * const block = &block_buffer[b];
|
||||
if (NUM_AXIS_GANG(block->steps.x, || block->steps.y, || block->steps.z, || block->steps.i, || block->steps.j, || block->steps.k, || block->steps.u, || block->steps.v, || block->steps.w)) {
|
||||
const float se = (float)block->steps.e / block->step_event_count * SQRT(block->nominal_speed_sqr); // mm/sec;
|
||||
const float se = (float)block->steps.e / block->step_event_count * SQRT(block->nominal_speed_sqr); // mm/sec
|
||||
NOLESS(high, se);
|
||||
}
|
||||
}
|
||||
@ -1782,7 +1781,7 @@ void Planner::synchronize() { while (busy()) idle(); }
|
||||
bool Planner::_buffer_steps(const xyze_long_t &target
|
||||
OPTARG(HAS_POSITION_FLOAT, const xyze_pos_t &target_float)
|
||||
OPTARG(HAS_DIST_MM_ARG, const xyze_float_t &cart_dist_mm)
|
||||
, feedRate_t fr_mm_s, const uint8_t extruder, const_float_t millimeters
|
||||
, feedRate_t fr_mm_s, const uint8_t extruder, const_float_t millimeters/*=0.0*/
|
||||
) {
|
||||
|
||||
// Wait for the next available block
|
||||
@ -1796,10 +1795,11 @@ bool Planner::_buffer_steps(const xyze_long_t &target
|
||||
|
||||
// Fill the block with the specified movement
|
||||
if (!_populate_block(block, false, target
|
||||
OPTARG(HAS_POSITION_FLOAT, target_float)
|
||||
OPTARG(HAS_DIST_MM_ARG, cart_dist_mm)
|
||||
, fr_mm_s, extruder, millimeters
|
||||
)) {
|
||||
OPTARG(HAS_POSITION_FLOAT, target_float)
|
||||
OPTARG(HAS_DIST_MM_ARG, cart_dist_mm)
|
||||
, fr_mm_s, extruder, millimeters
|
||||
)
|
||||
) {
|
||||
// Movement was not queued, probably because it was too short.
|
||||
// Simply accept that as movement queued and done
|
||||
return true;
|
||||
@ -1856,36 +1856,8 @@ bool Planner::_populate_block(block_t * const block, bool split_move,
|
||||
);
|
||||
|
||||
/* <-- add a slash to enable
|
||||
SERIAL_ECHOLNPGM(
|
||||
" _populate_block FR:", fr_mm_s,
|
||||
" A:", target.a, " (", da, " steps)"
|
||||
#if HAS_Y_AXIS
|
||||
" B:", target.b, " (", db, " steps)"
|
||||
#endif
|
||||
#if HAS_Z_AXIS
|
||||
" C:", target.c, " (", dc, " steps)"
|
||||
#endif
|
||||
#if HAS_I_AXIS
|
||||
" " STR_I ":", target.i, " (", di, " steps)"
|
||||
#endif
|
||||
#if HAS_J_AXIS
|
||||
" " STR_J ":", target.j, " (", dj, " steps)"
|
||||
#endif
|
||||
#if HAS_K_AXIS
|
||||
" " STR_K ":", target.k, " (", dk, " steps)"
|
||||
#endif
|
||||
#if HAS_U_AXIS
|
||||
" " STR_U ":", target.u, " (", du, " steps)"
|
||||
#endif
|
||||
#if HAS_V_AXIS
|
||||
" " STR_V ":", target.v, " (", dv, " steps)"
|
||||
#endif
|
||||
#if HAS_W_AXIS
|
||||
" " STR_W ":", target.w, " (", dw, " steps)"
|
||||
#if HAS_EXTRUDERS
|
||||
" E:", target.e, " (", de, " steps)"
|
||||
#endif
|
||||
);
|
||||
#define _ALINE(A) " " STR_##A ":", target[_AXIS(A)], " (", int32_t(target[_AXIS(A)] - position[_AXIS(A)]), " steps)"
|
||||
SERIAL_ECHOLNPGM(" _populate_block FR:", fr_mm_s, LOGICAL_AXIS_MAP(_ALINE));
|
||||
//*/
|
||||
|
||||
#if EITHER(PREVENT_COLD_EXTRUSION, PREVENT_LENGTHY_EXTRUDE)
|
||||
@ -1978,7 +1950,7 @@ bool Planner::_populate_block(block_t * const block, bool split_move,
|
||||
#endif
|
||||
|
||||
// Clear all flags, including the "busy" bit
|
||||
block->flag = 0x00;
|
||||
block->flag.clear();
|
||||
|
||||
// Set direction bits
|
||||
block->direction_bits = dm;
|
||||
@ -2449,7 +2421,7 @@ bool Planner::_populate_block(block_t * const block, bool split_move,
|
||||
if (speed_factor < 1.0f) {
|
||||
current_speed *= speed_factor;
|
||||
block->nominal_rate *= speed_factor;
|
||||
block->nominal_speed_sqr = block->nominal_speed_sqr * sq(speed_factor);
|
||||
block->nominal_speed_sqr *= sq(speed_factor);
|
||||
}
|
||||
|
||||
// Compute and limit the acceleration rate for the trapezoid generator.
|
||||
@ -2651,14 +2623,15 @@ bool Planner::_populate_block(block_t * const block, bool split_move,
|
||||
vmax_junction_sqr = sq(float(MINIMUM_PLANNER_SPEED));
|
||||
}
|
||||
else {
|
||||
NOLESS(junction_cos_theta, -0.999999f); // Check for numerical round-off to avoid divide by zero.
|
||||
|
||||
// Convert delta vector to unit vector
|
||||
xyze_float_t junction_unit_vec = unit_vec - prev_unit_vec;
|
||||
normalize_junction_vector(junction_unit_vec);
|
||||
|
||||
const float junction_acceleration = limit_value_by_axis_maximum(block->acceleration, junction_unit_vec),
|
||||
sin_theta_d2 = SQRT(0.5f * (1.0f - junction_cos_theta)); // Trig half angle identity. Always positive.
|
||||
const float junction_acceleration = limit_value_by_axis_maximum(block->acceleration, junction_unit_vec);
|
||||
|
||||
NOLESS(junction_cos_theta, -0.999999f); // Check for numerical round-off to avoid divide by zero.
|
||||
|
||||
const float sin_theta_d2 = SQRT(0.5f * (1.0f - junction_cos_theta)); // Trig half angle identity. Always positive.
|
||||
|
||||
vmax_junction_sqr = junction_acceleration * junction_deviation_mm * sin_theta_d2 / (1.0f - sin_theta_d2);
|
||||
|
||||
@ -2888,7 +2861,7 @@ bool Planner::_populate_block(block_t * const block, bool split_move,
|
||||
// block nominal speed limits both the current and next maximum junction speeds. Hence, in both
|
||||
// the reverse and forward planners, the corresponding block junction speed will always be at the
|
||||
// the maximum junction speed and may always be ignored for any speed reduction checks.
|
||||
block->flag |= block->nominal_speed_sqr <= v_allowable_sqr ? BLOCK_FLAG_RECALCULATE | BLOCK_FLAG_NOMINAL_LENGTH : BLOCK_FLAG_RECALCULATE;
|
||||
block->flag.set_nominal(block->nominal_speed_sqr <= v_allowable_sqr);
|
||||
|
||||
// Update previous path unit_vector and nominal speed
|
||||
previous_speed = current_speed;
|
||||
@ -2913,9 +2886,9 @@ bool Planner::_populate_block(block_t * const block, bool split_move,
|
||||
* Add a block to the buffer that just updates the position,
|
||||
* or in case of LASER_SYNCHRONOUS_M106_M107 the fan PWM
|
||||
*/
|
||||
void Planner::buffer_sync_block(TERN_(LASER_SYNCHRONOUS_M106_M107, uint8_t sync_flag)) {
|
||||
void Planner::buffer_sync_block(TERN_(LASER_SYNCHRONOUS_M106_M107, const BlockFlagBit sync_flag/*=BLOCK_BIT_SYNC_POSITION*/)) {
|
||||
#if DISABLED(LASER_SYNCHRONOUS_M106_M107)
|
||||
constexpr uint8_t sync_flag = BLOCK_FLAG_SYNC_POSITION;
|
||||
constexpr BlockFlagBit sync_flag = BLOCK_BIT_SYNC_POSITION;
|
||||
#endif
|
||||
|
||||
// Wait for the next available block
|
||||
@ -2925,7 +2898,7 @@ void Planner::buffer_sync_block(TERN_(LASER_SYNCHRONOUS_M106_M107, uint8_t sync_
|
||||
// Clear block
|
||||
memset(block, 0, sizeof(block_t));
|
||||
|
||||
block->flag = sync_flag;
|
||||
block->flag.apply(sync_flag);
|
||||
|
||||
block->position = position;
|
||||
#if ENABLED(BACKLASH_COMPENSATION)
|
||||
@ -3073,8 +3046,8 @@ bool Planner::buffer_segment(const abce_pos_t &abce
|
||||
if (!_buffer_steps(target
|
||||
OPTARG(HAS_POSITION_FLOAT, target_float)
|
||||
OPTARG(HAS_DIST_MM_ARG, cart_dist_mm)
|
||||
, fr_mm_s, extruder, millimeters)
|
||||
) return false;
|
||||
, fr_mm_s, extruder, millimeters
|
||||
)) return false;
|
||||
|
||||
stepper.wake_up();
|
||||
return true;
|
||||
@ -3141,6 +3114,14 @@ bool Planner::buffer_line(const xyze_pos_t &cart, const_feedRate_t fr_mm_s, cons
|
||||
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
|
||||
/**
|
||||
* @brief Add a direct stepping page block to the buffer
|
||||
* and wake up the Stepper ISR to process it.
|
||||
*
|
||||
* @param page_idx Page index provided by G6 I<index>
|
||||
* @param extruder The extruder to use in the move
|
||||
* @param num_steps Number of steps to process in the ISR
|
||||
*/
|
||||
void Planner::buffer_page(const page_idx_t page_idx, const uint8_t extruder, const uint16_t num_steps) {
|
||||
if (!last_page_step_rate) {
|
||||
kill(GET_TEXT_F(MSG_BAD_PAGE_SPEED));
|
||||
@ -3150,7 +3131,7 @@ bool Planner::buffer_line(const xyze_pos_t &cart, const_feedRate_t fr_mm_s, cons
|
||||
uint8_t next_buffer_head;
|
||||
block_t * const block = get_next_free_block(next_buffer_head);
|
||||
|
||||
block->flag = BLOCK_FLAG_IS_PAGE;
|
||||
block->flag.reset(BLOCK_BIT_PAGE);
|
||||
|
||||
#if HAS_FAN
|
||||
FANS_LOOP(i) block->fan_speed[i] = thermalManager.fan_speed[i];
|
||||
@ -3238,6 +3219,12 @@ void Planner::set_machine_position_mm(const abce_pos_t &abce) {
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* @brief Set the Planner position in mm
|
||||
* @details Set the Planner position from a native machine position in mm
|
||||
*
|
||||
* @param xyze A native (Cartesian) machine position
|
||||
*/
|
||||
void Planner::set_position_mm(const xyze_pos_t &xyze) {
|
||||
xyze_pos_t machine = xyze;
|
||||
TERN_(HAS_POSITION_MODIFIERS, apply_modifiers(machine, true));
|
||||
@ -3273,7 +3260,13 @@ void Planner::set_position_mm(const xyze_pos_t &xyze) {
|
||||
|
||||
#endif
|
||||
|
||||
// Recalculate the steps/s^2 acceleration rates, based on the mm/s^2
|
||||
/**
|
||||
* @brief Recalculate the steps/s^2 acceleration rates, based on the mm/s^2
|
||||
* @details Update planner movement factors after a change to certain settings:
|
||||
* - max_acceleration_steps_per_s2 from settings max_acceleration_mm_per_s2 * axis_steps_per_mm (M201, M92)
|
||||
* - acceleration_long_cutoff based on the largest max_acceleration_steps_per_s2 (M201)
|
||||
* - max_e_jerk for all extruders based on junction_deviation_mm (M205 J)
|
||||
*/
|
||||
void Planner::reset_acceleration_rates() {
|
||||
uint32_t highest_rate = 1;
|
||||
LOOP_DISTINCT_AXES(i) {
|
||||
@ -3286,8 +3279,8 @@ void Planner::reset_acceleration_rates() {
|
||||
}
|
||||
|
||||
/**
|
||||
* Recalculate 'position' and 'mm_per_step'.
|
||||
* Must be called whenever settings.axis_steps_per_mm changes!
|
||||
* @brief Recalculate 'position' and 'mm_per_step'.
|
||||
* @details Required whenever settings.axis_steps_per_mm changes!
|
||||
*/
|
||||
void Planner::refresh_positioning() {
|
||||
LOOP_DISTINCT_AXES(i) mm_per_step[i] = 1.0f / settings.axis_steps_per_mm[i];
|
||||
|
@ -70,9 +70,6 @@
|
||||
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
#include "../feature/direct_stepping.h"
|
||||
#define IS_PAGE(B) TEST(B->flag, BLOCK_BIT_IS_PAGE)
|
||||
#else
|
||||
#define IS_PAGE(B) false
|
||||
#endif
|
||||
|
||||
#if ENABLED(EXTERNAL_CLOSED_LOOP_CONTROLLER)
|
||||
@ -92,47 +89,6 @@
|
||||
#define HAS_DIST_MM_ARG 1
|
||||
#endif
|
||||
|
||||
enum BlockFlagBit : char {
|
||||
// Recalculate trapezoids on entry junction. For optimization.
|
||||
BLOCK_BIT_RECALCULATE,
|
||||
|
||||
// Nominal speed always reached.
|
||||
// i.e., The segment is long enough, so the nominal speed is reachable if accelerating
|
||||
// from a safe speed (in consideration of jerking from zero speed).
|
||||
BLOCK_BIT_NOMINAL_LENGTH,
|
||||
|
||||
// The block is segment 2+ of a longer move
|
||||
BLOCK_BIT_CONTINUED,
|
||||
|
||||
// Sync the stepper counts from the block
|
||||
BLOCK_BIT_SYNC_POSITION
|
||||
|
||||
// Direct stepping page
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
, BLOCK_BIT_IS_PAGE
|
||||
#endif
|
||||
|
||||
// Sync the fan speeds from the block
|
||||
#if ENABLED(LASER_SYNCHRONOUS_M106_M107)
|
||||
, BLOCK_BIT_SYNC_FANS
|
||||
#endif
|
||||
};
|
||||
|
||||
enum BlockFlag : char {
|
||||
BLOCK_FLAG_RECALCULATE = _BV(BLOCK_BIT_RECALCULATE)
|
||||
, BLOCK_FLAG_NOMINAL_LENGTH = _BV(BLOCK_BIT_NOMINAL_LENGTH)
|
||||
, BLOCK_FLAG_CONTINUED = _BV(BLOCK_BIT_CONTINUED)
|
||||
, BLOCK_FLAG_SYNC_POSITION = _BV(BLOCK_BIT_SYNC_POSITION)
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
, BLOCK_FLAG_IS_PAGE = _BV(BLOCK_BIT_IS_PAGE)
|
||||
#endif
|
||||
#if ENABLED(LASER_SYNCHRONOUS_M106_M107)
|
||||
, BLOCK_FLAG_SYNC_FANS = _BV(BLOCK_BIT_SYNC_FANS)
|
||||
#endif
|
||||
};
|
||||
|
||||
#define BLOCK_MASK_SYNC ( BLOCK_FLAG_SYNC_POSITION | TERN0(LASER_SYNCHRONOUS_M106_M107, BLOCK_FLAG_SYNC_FANS) )
|
||||
|
||||
#if ENABLED(LASER_POWER_INLINE)
|
||||
|
||||
typedef struct {
|
||||
@ -158,17 +114,83 @@ enum BlockFlag : char {
|
||||
#endif
|
||||
|
||||
/**
|
||||
* struct block_t
|
||||
*
|
||||
* A single entry in the planner buffer.
|
||||
* Tracks linear movement over multiple axes.
|
||||
* Planner block flags as boolean bit fields
|
||||
*/
|
||||
enum BlockFlagBit {
|
||||
// Recalculate trapezoids on entry junction. For optimization.
|
||||
BLOCK_BIT_RECALCULATE,
|
||||
|
||||
// Nominal speed always reached.
|
||||
// i.e., The segment is long enough, so the nominal speed is reachable if accelerating
|
||||
// from a safe speed (in consideration of jerking from zero speed).
|
||||
BLOCK_BIT_NOMINAL_LENGTH,
|
||||
|
||||
// The block is segment 2+ of a longer move
|
||||
BLOCK_BIT_CONTINUED,
|
||||
|
||||
// Sync the stepper counts from the block
|
||||
BLOCK_BIT_SYNC_POSITION
|
||||
|
||||
// Direct stepping page
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
, BLOCK_BIT_PAGE
|
||||
#endif
|
||||
|
||||
// Sync the fan speeds from the block
|
||||
#if ENABLED(LASER_SYNCHRONOUS_M106_M107)
|
||||
, BLOCK_BIT_SYNC_FANS
|
||||
#endif
|
||||
};
|
||||
|
||||
/**
|
||||
* Planner block flags as boolean bit fields
|
||||
*/
|
||||
typedef struct {
|
||||
union {
|
||||
uint8_t bits;
|
||||
|
||||
struct {
|
||||
bool recalculate:1;
|
||||
|
||||
bool nominal_length:1;
|
||||
|
||||
bool continued:1;
|
||||
|
||||
bool sync_position:1;
|
||||
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
bool page:1;
|
||||
#endif
|
||||
|
||||
#if ENABLED(LASER_SYNCHRONOUS_M106_M107)
|
||||
bool sync_fans:1;
|
||||
#endif
|
||||
};
|
||||
};
|
||||
|
||||
void clear() volatile { bits = 0; }
|
||||
void apply(const uint8_t f) volatile { bits |= f; }
|
||||
void apply(const BlockFlagBit b) volatile { SBI(bits, b); }
|
||||
void reset(const BlockFlagBit b) volatile { bits = _BV(b); }
|
||||
void set_nominal(const bool n) volatile { recalculate = true; if (n) nominal_length = true; }
|
||||
|
||||
} block_flags_t;
|
||||
|
||||
/**
|
||||
* A single entry in the planner buffer, used to set up and
|
||||
* track a coordinated linear motion for one or more axes.
|
||||
*
|
||||
* The "nominal" values are as-specified by G-code, and
|
||||
* may never actually be reached due to acceleration limits.
|
||||
*/
|
||||
typedef struct block_t {
|
||||
|
||||
volatile uint8_t flag; // Block flags (See BlockFlag enum above) - Modified by ISR and main thread!
|
||||
volatile block_flags_t flag; // Block flags
|
||||
|
||||
volatile bool is_fan_sync() { return TERN0(LASER_SYNCHRONOUS_M106_M107, flag.sync_fans); }
|
||||
volatile bool is_sync() { return flag.sync_position || is_fan_sync(); }
|
||||
volatile bool is_page() { return TERN0(DIRECT_STEPPING, flag.page); }
|
||||
volatile bool is_move() { return !(is_sync() || is_page()); }
|
||||
|
||||
// Fields used by the motion planner to manage acceleration
|
||||
float nominal_speed_sqr, // The nominal speed for this block in (mm/sec)^2
|
||||
@ -759,7 +781,7 @@ class Planner {
|
||||
* case of LASER_SYNCHRONOUS_M106_M107 the fan pwm
|
||||
*/
|
||||
static void buffer_sync_block(
|
||||
TERN_(LASER_SYNCHRONOUS_M106_M107, uint8_t sync_flag=BLOCK_FLAG_SYNC_POSITION)
|
||||
TERN_(LASER_SYNCHRONOUS_M106_M107, const BlockFlagBit flag=BLOCK_BIT_SYNC_POSITION)
|
||||
);
|
||||
|
||||
#if IS_KINEMATIC
|
||||
|
@ -1699,7 +1699,7 @@ void Stepper::pulse_phase_isr() {
|
||||
}while(0)
|
||||
|
||||
// Direct Stepping page?
|
||||
const bool is_page = IS_PAGE(current_block);
|
||||
const bool is_page = current_block->is_page();
|
||||
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
// Direct stepping is currently not ready for HAS_I_AXIS
|
||||
@ -1977,7 +1977,7 @@ uint32_t Stepper::block_phase_isr() {
|
||||
count_position[_AXIS(AXIS)] += page_step_state.bd[_AXIS(AXIS)] * count_direction[_AXIS(AXIS)];
|
||||
#endif
|
||||
|
||||
if (IS_PAGE(current_block)) {
|
||||
if (current_block->is_page()) {
|
||||
PAGE_SEGMENT_UPDATE_POS(X);
|
||||
PAGE_SEGMENT_UPDATE_POS(Y);
|
||||
PAGE_SEGMENT_UPDATE_POS(Z);
|
||||
@ -2167,16 +2167,13 @@ uint32_t Stepper::block_phase_isr() {
|
||||
if ((current_block = planner.get_current_block())) {
|
||||
|
||||
// Sync block? Sync the stepper counts or fan speeds and return
|
||||
while (current_block->flag & BLOCK_MASK_SYNC) {
|
||||
while (current_block->is_sync()) {
|
||||
|
||||
#if ENABLED(LASER_SYNCHRONOUS_M106_M107)
|
||||
const bool is_sync_fans = TEST(current_block->flag, BLOCK_BIT_SYNC_FANS);
|
||||
if (is_sync_fans) planner.sync_fan_speeds(current_block->fan_speed);
|
||||
#else
|
||||
constexpr bool is_sync_fans = false;
|
||||
#endif
|
||||
|
||||
if (!is_sync_fans) _set_position(current_block->position);
|
||||
if (current_block->is_fan_sync()) {
|
||||
TERN_(LASER_SYNCHRONOUS_M106_M107, planner.sync_fan_speeds(current_block->fan_speed));
|
||||
}
|
||||
else
|
||||
_set_position(current_block->position);
|
||||
|
||||
discard_current_block();
|
||||
|
||||
@ -2196,7 +2193,7 @@ uint32_t Stepper::block_phase_isr() {
|
||||
#endif
|
||||
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
if (IS_PAGE(current_block)) {
|
||||
if (current_block->is_page()) {
|
||||
page_step_state.segment_steps = 0;
|
||||
page_step_state.segment_idx = 0;
|
||||
page_step_state.page = page_manager.get_page(current_block->page_idx);
|
||||
|
@ -524,8 +524,7 @@ class Stepper {
|
||||
// Discard current block and free any resources
|
||||
FORCE_INLINE static void discard_current_block() {
|
||||
#if ENABLED(DIRECT_STEPPING)
|
||||
if (IS_PAGE(current_block))
|
||||
page_manager.free_page(current_block->page_idx);
|
||||
if (current_block->is_page()) page_manager.free_page(current_block->page_idx);
|
||||
#endif
|
||||
current_block = nullptr;
|
||||
axis_did_move = 0;
|
||||
|
Loading…
Reference in New Issue
Block a user