//! /system/services/display — the display service (docs/display.md). A ring-3 process
//! that claims the framebuffer the kernel seeded (docs/display-plan.md D1), owns it as a
//! **write-combining front buffer**, composites an ordered stack of **layers** into a
//! **cacheable back buffer**, and presents finished frames — the GUI track's compositor,
//! the sibling of the input service. Reached by name over `ServiceId.display`.
//!
//! A layer is a server-owned surface (its own cacheable buffer) with a screen position,
//! z-order, and visibility. Clients create layers and draw into them by command
//! (`fill_rect`, `blit_tile`), mark `damage`, and ask for a `present`; the compositor
//! repaints only the damaged region — clear it, paint the visible layers bottom-to-top,
//! flush it to the screen. The pixel math lives in the pure, host-tested
//! [compositor.zig](compositor.zig); this file wires real surfaces and the framebuffer to
//! it. Shared-memory client surfaces are a later milestone (docs/display.md).

const std = @import("std");
const runtime = @import("runtime");
const compositor = @import("compositor.zig");

const protocol = runtime.display_protocol;
const ipc = runtime.ipc;
const system = runtime.system;
const device = runtime.device;
const Rect = compositor.Rect;
const Surface = compositor.Surface;

/// The claimed framebuffer and its off-screen twin. The front buffer is the LFB —
/// write-combining, so it is **only ever written**, never read; all compositing happens
/// in the cacheable back buffer, which is then streamed to the front (docs/display.md).
const Display = struct {
    device_id: u64,
    front: [*]volatile u8, // the LFB (write-combining)
    back: [*]u8, // cacheable, same geometry
    width: u32,
    height: u32,
    pitch: u32, // bytes per row (shared by both buffers)
    format: u32, // a device-abi DisplayFormat value
    frames: u64 = 0,
};

var display: Display = undefined;

/// The wallpaper the compositor clears damaged regions to before painting layers.
var background: u32 = 0;

/// The layer stack. A fixed table (a compositor has few top-level surfaces during
/// bring-up); each used slot owns an mmap'd surface. `damage` accumulates the dirty
/// screen region since the last `present`, so a present touches only what changed.
const maximum_layers = 16;

const Layer = struct {
    used: bool = false,
    x: i32 = 0,
    y: i32 = 0,
    z: u32 = 0,
    visible: bool = false,
    surface: Surface = undefined,
    surface_len: usize = 0, // for munmap on destroy
};

var layers: [maximum_layers]Layer = [_]Layer{.{}} ** maximum_layers;
var damage: Rect = Rect.empty;

/// Enumeration buffer kept off the stack — a `DeviceDescriptor` is large, and this
/// service only ever needs one scan.
var device_table: [64]device.DeviceDescriptor = undefined;

// --- geometry helpers -------------------------------------------------------

fn screenRect() Rect {
    return .{ .x = 0, .y = 0, .w = @intCast(display.width), .h = @intCast(display.height) };
}

fn backSurface() Surface {
    return .{
        .pixels = @ptrCast(@alignCast(display.back)),
        .stride = display.pitch / 4, // pitch is bytes; a 32-bpp row is pitch/4 pixels
        .width = display.width,
        .height = display.height,
    };
}

fn layerScreenRect(l: *const Layer) Rect {
    return .{ .x = l.x, .y = l.y, .w = @intCast(l.surface.width), .h = @intCast(l.surface.height) };
}

/// Add `r` (screen coordinates) to the pending damage, clipped to the screen.
fn addDamage(r: Rect) void {
    damage = damage.unite(r.intersect(screenRect()));
}

// --- layer operations (called from onMessage and the self-check) ------------

fn freeLayer() ?u32 {
    for (&layers, 0..) |*l, i| {
        if (!l.used) return @intCast(i);
    }
    return null;
}

/// A used layer by id, or null if the id is out of range or free.
fn layerAt(id: u32) ?*Layer {
    if (id >= maximum_layers or !layers[id].used) return null;
    return &layers[id];
}

fn createLayer(x: i32, y: i32, w: u32, h: u32, z: u32, visible: bool) ?u32 {
    if (w == 0 or h == 0) return null;
    const slot = freeLayer() orelse return null;
    const len = @as(usize, w) * h * 4;
    const base = system.mmap(len, system.PROT_READ | system.PROT_WRITE);
    if (system.mmapFailed(base)) return null;
    layers[slot] = .{
        .used = true,
        .x = x,
        .y = y,
        .z = z,
        .visible = visible,
        .surface = .{ .pixels = @ptrFromInt(base), .stride = w, .width = w, .height = h },
        .surface_len = len,
    };
    return slot;
}

fn fillLayer(id: u32, local: Rect, colour: u32) bool {
    const l = layerAt(id) orelse return false;
    compositor.fillRect(l.surface, local, colour);
    // Damage in screen space = the fill, translated by the layer origin, within the layer.
    const screen = Rect{ .x = l.x + local.x, .y = l.y + local.y, .w = local.w, .h = local.h };
    addDamage(screen.intersect(layerScreenRect(l)));
    return true;
}

fn blitLayer(id: u32, x: i32, y: i32, w: u32, h: u32, pixels: []const u8) bool {
    const l = layerAt(id) orelse return false;
    compositor.blitTile(l.surface, x, y, pixels, w, h);
    const screen = Rect{ .x = l.x + x, .y = l.y + y, .w = @intCast(w), .h = @intCast(h) };
    addDamage(screen.intersect(layerScreenRect(l)));
    return true;
}

fn configureLayer(id: u32, x: i32, y: i32, z: u32, visible: bool) bool {
    const l = layerAt(id) orelse return false;
    addDamage(layerScreenRect(l)); // the old footprint must repaint
    l.x = x;
    l.y = y;
    l.z = z;
    l.visible = visible;
    addDamage(layerScreenRect(l)); // and the new one
    return true;
}

fn destroyLayer(id: u32) bool {
    const l = layerAt(id) orelse return false;
    addDamage(layerScreenRect(l));
    _ = system.munmap(@intFromPtr(l.surface.pixels), l.surface_len);
    l.* = .{};
    return true;
}

// --- compositing + present --------------------------------------------------

/// Repaint the damaged region `clip` of the back buffer: clear it to the background, then
/// paint every visible layer that overlaps it, bottom to top (ascending z).
fn compositeInto(clip: Rect) void {
    const back = backSurface();
    compositor.fillRect(back, clip, background);

    // z-order the used, visible layers (n ≤ 16; a plain insertion sort of indices).
    var order: [maximum_layers]u32 = undefined;
    var n: usize = 0;
    for (layers, 0..) |l, i| {
        if (l.used and l.visible) {
            order[n] = @intCast(i);
            n += 1;
        }
    }
    var a: usize = 1;
    while (a < n) : (a += 1) {
        const key = order[a];
        var b: usize = a;
        while (b > 0 and layers[order[b - 1]].z > layers[key].z) : (b -= 1) order[b] = order[b - 1];
        order[b] = key;
    }

    for (order[0..n]) |i| {
        const l = layers[i];
        compositor.composite(back, l.x, l.y, l.surface, clip);
    }
}

/// Stream the damaged rectangle from the cacheable back buffer to the write-combining
/// front buffer, row by row (sequential writes — what WC memory wants; we never read the
/// front buffer). Only the visible width of each row is touched.
fn flushRect(rect: Rect) void {
    const c = rect.intersect(screenRect());
    if (c.isEmpty()) return;
    var y: i32 = c.y;
    while (y < c.bottom()) : (y += 1) {
        const off = @as(usize, @intCast(y)) * display.pitch;
        const src: [*]const u32 = @ptrCast(@alignCast(display.back + off));
        const dst: [*]volatile u32 = @ptrCast(@alignCast(display.front + off));
        var x: i32 = c.x;
        while (x < c.right()) : (x += 1) dst[@intCast(x)] = src[@intCast(x)];
    }
}

/// Composite and flush the accumulated damage, then clear it. A no-op when nothing is
/// dirty. The frame counter advances regardless, so callers can name frames.
fn present() void {
    const dirty = damage.intersect(screenRect());
    if (!dirty.isEmpty()) {
        compositeInto(dirty);
        flushRect(dirty);
    }
    damage = Rect.empty;
    display.frames += 1;
}

// --- startup self-check -----------------------------------------------------

/// Prove the compositor wiring on the real framebuffer: two overlapping opaque layers,
/// composited, must show the top layer in the overlap and the bottom layer outside it.
/// Exercises the whole path — mmap surfaces, the z-sort, damage, composite into the back
/// buffer — and reads the composited result back. Cleans up after itself.
fn selfCheck() void {
    const red = protocol.pack(display.format, 0xC0, 0x20, 0x20);
    const green = protocol.pack(display.format, 0x20, 0xC0, 0x20);
    const bottom = createLayer(100, 100, 80, 80, 0, true) orelse return fail_check("create");
    const top = createLayer(140, 140, 80, 80, 1, true) orelse return fail_check("create");
    _ = fillLayer(bottom, Rect.init(0, 0, 80, 80), red);
    _ = fillLayer(top, Rect.init(0, 0, 80, 80), green);
    present();

    const back = backSurface();
    const overlap = back.pixels[@as(usize, 150) * back.stride + 150]; // in both layers → top
    const bottom_only = back.pixels[@as(usize, 110) * back.stride + 110]; // bottom only

    _ = destroyLayer(top);
    _ = destroyLayer(bottom);
    present(); // repaint the self-check region back to the background

    if (overlap == green and bottom_only == red) {
        _ = system.write("display: compositor self-check ok\n");
    } else {
        _ = system.write("display: compositor self-check FAILED\n");
    }
}

fn fail_check(_: []const u8) void {
    _ = system.write("display: compositor self-check FAILED (setup)\n");
}

// --- service ----------------------------------------------------------------

/// The framebuffer node the kernel seeded (`DeviceClass.display`), or null if none.
fn findDisplay() ?device.DeviceDescriptor {
    const total = device.enumerate(&device_table);
    const n = @min(total, device_table.len);
    for (device_table[0..n]) |d| {
        if (d.class == @intFromEnum(device.DeviceClass.display)) return d;
    }
    return null;
}

fn initialise(endpoint: ipc.Handle) bool {
    _ = endpoint;

    // Find the framebuffer, retrying while device discovery catches up with our spawn.
    var tries: u32 = 0;
    const found = while (tries < 100) : (tries += 1) {
        if (findDisplay()) |d| break d;
        system.sleep(50);
    } else {
        _ = system.write("display: no framebuffer device (headless?)\n");
        return false; // clean exit: nothing to drive
    };

    if (!device.claim(found.id)) {
        _ = system.write("display: could not claim the framebuffer\n");
        return false;
    }
    // Resource 0 is the framebuffer memory window; the kernel maps it write-combining
    // because the resource carries that flag (docs/display-plan.md D1).
    const front_base = device.mmioMap(found.id, 0) orelse {
        _ = system.write("display: could not map the framebuffer\n");
        return false;
    };

    const geometry = found.display;
    const size = @as(usize, geometry.height) * geometry.pitch;
    const back_base = system.mmap(size, system.PROT_READ | system.PROT_WRITE);
    if (system.mmapFailed(back_base)) {
        _ = system.write("display: could not allocate the back buffer\n");
        return false;
    }

    display = .{
        .device_id = found.id,
        .front = @ptrFromInt(front_base),
        .back = @ptrFromInt(back_base),
        .width = geometry.width,
        .height = geometry.height,
        .pitch = geometry.pitch,
        .format = geometry.format,
    };
    background = protocol.pack(display.format, 0x20, 0x30, 0x48); // a dark slate wallpaper

    // Clear the whole screen through the back buffer → present path (double buffering:
    // no direct-to-LFB drawing).
    addDamage(screenRect());
    present();

    var line: [96]u8 = undefined;
    _ = system.write(std.fmt.bufPrint(&line, "display: online {d}x{d} pitch {d} format {d}\n", .{
        display.width, display.height, display.pitch, display.format,
    }) catch "display: online\n");
    _ = system.write("display: presented frame 0\n");

    selfCheck();
    return true;
}

fn writeReply(reply: []u8, value: protocol.Reply) usize {
    const bytes = std.mem.asBytes(&value);
    @memcpy(reply[0..bytes.len], bytes);
    return bytes.len;
}

fn ok(reply: []u8) usize {
    return writeReply(reply, .{ .status = 0 });
}

fn fail(reply: []u8) usize {
    return writeReply(reply, .{ .status = -1 });
}

fn onMessage(message: []const u8, reply: []u8, sender: u32, capability: ?ipc.Handle) usize {
    _ = sender;
    _ = capability;
    if (message.len < protocol.request_size) return fail(reply);
    const request = std.mem.bytesToValue(protocol.Request, message[0..protocol.request_size]);
    const payload = message[protocol.request_size..];
    // Switch on the raw operation value — an out-of-range one must fail cleanly, not
    // panic an `@enumFromInt`.
    switch (request.operation) {
        @intFromEnum(protocol.Operation.info) => return writeReply(reply, .{
            .status = 0,
            .width = display.width,
            .height = display.height,
            .pitch = display.pitch,
            .format = display.format,
        }),
        @intFromEnum(protocol.Operation.create_layer) => {
            // x/y are signed coordinates carried in the u32 wire fields — reinterpret the
            // bits (@bitCast), don't range-check (@intCast) which a negative would fail.
            const slot = createLayer(@bitCast(request.x), @bitCast(request.y), request.width, request.height, request.z, request.visible != 0) orelse return fail(reply);
            return writeReply(reply, .{ .status = 0, .layer = slot });
        },
        @intFromEnum(protocol.Operation.configure_layer) => {
            return if (configureLayer(request.layer, @bitCast(request.x), @bitCast(request.y), request.z, request.visible != 0)) ok(reply) else fail(reply);
        },
        @intFromEnum(protocol.Operation.destroy_layer) => {
            return if (destroyLayer(request.layer)) ok(reply) else fail(reply);
        },
        @intFromEnum(protocol.Operation.fill_rect) => {
            const local = Rect.init(@bitCast(request.x), @bitCast(request.y), @intCast(request.width), @intCast(request.height));
            return if (fillLayer(request.layer, local, request.colour)) ok(reply) else fail(reply);
        },
        @intFromEnum(protocol.Operation.blit_tile) => {
            return if (blitLayer(request.layer, @bitCast(request.x), @bitCast(request.y), request.width, request.height, payload)) ok(reply) else fail(reply);
        },
        @intFromEnum(protocol.Operation.damage) => {
            const l = layerAt(request.layer) orelse return fail(reply);
            const screen = Rect{ .x = l.x + @as(i32, @bitCast(request.x)), .y = l.y + @as(i32, @bitCast(request.y)), .w = @intCast(request.width), .h = @intCast(request.height) };
            addDamage(screen.intersect(layerScreenRect(l)));
            return ok(reply);
        },
        @intFromEnum(protocol.Operation.present) => {
            present();
            return ok(reply);
        },
        else => return fail(reply),
    }
}

pub fn main() void {
    runtime.service.run(protocol.message_maximum, .{
        .service = .display,
        .init = initialise,
        .on_message = onMessage,
    });
}

pub const panic = runtime.panic;
comptime {
    _ = &runtime.start._start;
}