;
; Copyright (c) 2020 Raspberry Pi (Trading) Ltd.
;
; SPDX-License-Identifier: BSD-3-Clause
;

.program st7789_lcd
.side_set 1

; This is just a simple clocked serial TX. At 125 MHz system clock we can
; sustain up to 62.5 Mbps.
; Data on OUT pin 0
; Clock on side-set pin 0

.wrap_target
    out pins, 1   side 0 ; stall here if no data (clock low)
    nop           side 1
.wrap

% c-sdk {
// For optimal use of DMA bandwidth we would use an autopull threshold of 32,
// but we are using a threshold of 8 here (consume 1 byte from each FIFO entry
// and discard the remainder) to make things easier for software on the other side

static inline void st7789_lcd_program_init(PIO pio, uint sm, uint offset, uint data_pin, uint clk_pin, float clk_div) {
    pio_gpio_init(pio, data_pin);
    pio_gpio_init(pio, clk_pin);
    pio_sm_set_consecutive_pindirs(pio, sm, data_pin, 1, true);
    pio_sm_set_consecutive_pindirs(pio, sm, clk_pin, 1, true);
    pio_sm_config c = st7789_lcd_program_get_default_config(offset);
    sm_config_set_sideset_pins(&c, clk_pin);
    sm_config_set_out_pins(&c, data_pin, 1);
    sm_config_set_fifo_join(&c, PIO_FIFO_JOIN_TX);
    sm_config_set_clkdiv(&c, clk_div);
    sm_config_set_out_shift(&c, false, true, 8);
    pio_sm_init(pio, sm, offset, &c);
    pio_sm_set_enabled(pio, sm, true);
}

// Making use of the narrow store replication behaviour on RP2040 to get the
// data left-justified (as we are using shift-to-left to get MSB-first serial)

static inline void st7789_lcd_put(PIO pio, uint sm, uint8_t x) {
    while (pio_sm_is_tx_fifo_full(pio, sm))
        ;
    *(volatile uint8_t*)&pio->txf[sm] = x;
}

// SM is done when it stalls on an empty FIFO

static inline void st7789_lcd_wait_idle(PIO pio, uint sm) {
    uint32_t sm_stall_mask = 1u << (sm + PIO_FDEBUG_TXSTALL_LSB);
    pio->fdebug = sm_stall_mask;
    while (!(pio->fdebug & sm_stall_mask))
        ;
}
%}