Files
VulkanDriver/src/software/device/rasterizer/edge_function.zig
T
kbz_8 cb6af550e2
Test / build_and_test (push) Failing after 29s
Build / build (push) Successful in 43s
fixing varyings
2026-07-01 21:27:05 +02:00

510 lines
20 KiB
Zig

const std = @import("std");
const vk = @import("vulkan");
const base = @import("base");
const spv = @import("spv");
const zm = base.zm;
const common = @import("common.zig");
const fragment = @import("../fragment.zig");
const Renderer = @import("../Renderer.zig");
const VkError = base.VkError;
const SpvRuntimeError = spv.Runtime.RuntimeError;
const F32x4 = zm.F32x4;
const SamplePosition = struct {
x: f32,
y: f32,
};
const RunData = struct {
allocator: std.mem.Allocator,
draw_call: *Renderer.DrawCall,
batch_id: usize,
min_x: i32,
max_x: i32,
min_y: i32,
max_y: i32,
area: f32,
v0: Renderer.Vertex,
v1: Renderer.Vertex,
v2: Renderer.Vertex,
provoking_vertex: Renderer.Vertex,
color_attachment_access: []const ?common.RenderTargetAccess,
depth_attachment_access: ?*common.RenderTargetAccess,
stencil_attachment_access: ?*common.RenderTargetAccess,
front_face: bool,
has_fragment_shader: bool,
early_fragment_tests: bool,
fragment_uses_derivatives: bool,
fragment_uses_sample_id: bool,
fragment_uses_centroid: bool,
depth_bias_slope: f32,
};
pub fn drawTriangle(
allocator: std.mem.Allocator,
draw_call: *Renderer.DrawCall,
v0: *Renderer.Vertex,
v1: *Renderer.Vertex,
v2: *Renderer.Vertex,
provoking_vertex: *Renderer.Vertex,
color_attachment_access: []const ?common.RenderTargetAccess,
depth_attachment_access: ?*common.RenderTargetAccess,
stencil_attachment_access: ?*common.RenderTargetAccess,
front_face: bool,
) VkError!void {
const io = draw_call.renderer.device.interface.io();
const min_x: i32 = @intFromFloat(@floor(@min(v0.position[0], v1.position[0], v2.position[0])));
const max_x: i32 = @intFromFloat(@ceil(@max(v0.position[0], v1.position[0], v2.position[0])));
const min_y: i32 = @intFromFloat(@floor(@min(v0.position[1], v1.position[1], v2.position[1])));
const max_y: i32 = @intFromFloat(@ceil(@max(v0.position[1], v1.position[1], v2.position[1])));
const area = edgeFunction(v0.position, v1.position, v2.position);
if (area == 0.0)
return;
const inv_area = 1.0 / area;
const dz_dx =
(v0.position[2] * ((v1.position[1] - v2.position[1]) * inv_area)) +
(v1.position[2] * ((v2.position[1] - v0.position[1]) * inv_area)) +
(v2.position[2] * ((v0.position[1] - v1.position[1]) * inv_area));
const dz_dy =
(v0.position[2] * ((v2.position[0] - v1.position[0]) * inv_area)) +
(v1.position[2] * ((v0.position[0] - v2.position[0]) * inv_area)) +
(v2.position[2] * ((v1.position[0] - v0.position[0]) * inv_area));
const depth_bias_slope = @max(@abs(dz_dx), @abs(dz_dy));
const pipeline = draw_call.renderer.state.pipeline orelse return;
const fragment_stage = pipeline.stages.getPtr(.fragment);
const fragment_uses_derivatives = if (fragment_stage) |stage|
stage.module.module.reflection_infos.needs_derivatives
else
false;
const early_fragment_tests = if (fragment_stage) |stage|
stage.module.module.reflection_infos.early_fragment_tests
else
false;
const fragment_uses_sample_id = if (fragment_stage) |stage|
stage.module.module.builtins.get(.SampleId) != null
else
false;
const fragment_uses_centroid = if (fragment_stage) |stage|
fragmentStageUsesInputDecoration(stage, .Centroid)
else
false;
const runtimes_count = if (fragment_stage) |stage| stage.runtimes.len else 1;
if (runtimes_count == 0)
return;
const grid_size: usize = @intFromFloat(@ceil(@sqrt(@as(f32, @floatFromInt(runtimes_count)))));
const width: usize = @intCast(max_x - min_x + 1);
const height: usize = @intCast(max_y - min_y + 1);
const cols_per_run = @divTrunc(width + grid_size - 1, grid_size);
const rows_per_run = @divTrunc(height + grid_size - 1, grid_size);
var batch_id: usize = 0;
for (0..grid_size) |gy| {
for (0..grid_size) |gx| {
defer batch_id = @mod(batch_id + 1, runtimes_count);
const run_min_x = min_x + @as(i32, @intCast(gx * cols_per_run));
const run_min_y = min_y + @as(i32, @intCast(gy * rows_per_run));
if (run_min_x > max_x or run_min_y > max_y)
continue;
const run_max_x = @min(
run_min_x + @as(i32, @intCast(cols_per_run)) - 1,
max_x,
);
const run_max_y = @min(
run_min_y + @as(i32, @intCast(rows_per_run)) - 1,
max_y,
);
const run_data: RunData = .{
.allocator = allocator,
.draw_call = draw_call,
.batch_id = batch_id,
.v0 = v0.*,
.v1 = v1.*,
.v2 = v2.*,
.provoking_vertex = provoking_vertex.*,
.area = area,
.min_x = run_min_x,
.max_x = run_max_x,
.min_y = run_min_y,
.max_y = run_max_y,
.color_attachment_access = color_attachment_access,
.depth_attachment_access = depth_attachment_access,
.stencil_attachment_access = stencil_attachment_access,
.front_face = front_face,
.has_fragment_shader = fragment_stage != null,
.early_fragment_tests = early_fragment_tests,
.fragment_uses_derivatives = fragment_uses_derivatives,
.fragment_uses_sample_id = fragment_uses_sample_id,
.fragment_uses_centroid = fragment_uses_centroid,
.depth_bias_slope = depth_bias_slope,
};
draw_call.rasterizer_wait_group.async(io, runWrapper, .{run_data});
}
}
draw_call.rasterizer_wait_group.await(io) catch return VkError.DeviceLost;
}
inline fn edgeFunction(a: F32x4, b: F32x4, p: F32x4) f32 {
return ((p[0] - a[0]) * (b[1] - a[1])) - ((p[1] - a[1]) * (b[0] - a[0]));
}
inline fn isInclusiveEdge(a: F32x4, b: F32x4) bool {
const dx = b[0] - a[0];
const dy = b[1] - a[1];
return dy > 0.0 or (dy == 0.0 and dx < 0.0);
}
inline fn edgeContainsPixel(a: F32x4, b: F32x4, edge_value: f32, area: f32) bool {
return if (area > 0.0)
edge_value > 0.0 or (edge_value == 0.0 and isInclusiveEdge(a, b))
else
edge_value < 0.0 or (edge_value == 0.0 and isInclusiveEdge(b, a));
}
inline fn standardSamplePosition(sample_count: usize, sample_index: usize) SamplePosition {
return switch (sample_count) {
1 => .{ .x = 0.5, .y = 0.5 },
2 => switch (sample_index) {
0 => .{ .x = 0.75, .y = 0.75 },
1 => .{ .x = 0.25, .y = 0.25 },
else => .{ .x = 0.5, .y = 0.5 },
},
4 => switch (sample_index) {
0 => .{ .x = 0.375, .y = 0.125 },
1 => .{ .x = 0.875, .y = 0.375 },
2 => .{ .x = 0.125, .y = 0.625 },
3 => .{ .x = 0.625, .y = 0.875 },
else => .{ .x = 0.5, .y = 0.5 },
},
else => .{ .x = 0.5, .y = 0.5 },
};
}
fn fragmentStageUsesInputDecoration(stage: anytype, decoration: anytype) bool {
const rt = &stage.runtimes[0].rt;
for (rt.mod.input_locations) |location| {
for (location) |result_word| {
if (result_word == 0)
continue;
if (rt.hasResultDecoration(result_word, decoration))
return true;
}
}
return false;
}
fn firstCoveredSamplePosition(sample_count: usize, coverage_sample_mask: vk.SampleMask) SamplePosition {
for (0..sample_count) |sample_index| {
if (sample_index >= @bitSizeOf(vk.SampleMask))
break;
const bit_index: u5 = @intCast(sample_index);
if ((coverage_sample_mask & (@as(vk.SampleMask, 1) << bit_index)) != 0)
return standardSamplePosition(sample_count, sample_index);
}
return .{ .x = 0.5, .y = 0.5 };
}
fn triangleCoverageMask(data: RunData, x: i32, y: i32, sample_count: usize) vk.SampleMask {
var mask: vk.SampleMask = 0;
for (0..sample_count) |sample_index| {
if (sample_index >= @bitSizeOf(vk.SampleMask))
break;
const sample_pos = standardSamplePosition(sample_count, sample_index);
const p = zm.f32x4(
@as(f32, @floatFromInt(x)) + sample_pos.x,
@as(f32, @floatFromInt(y)) + sample_pos.y,
0.0,
1.0,
);
const w0 = edgeFunction(data.v1.position, data.v2.position, p);
const w1 = edgeFunction(data.v2.position, data.v0.position, p);
const w2 = edgeFunction(data.v0.position, data.v1.position, p);
const inside =
edgeContainsPixel(data.v1.position, data.v2.position, w0, data.area) and
edgeContainsPixel(data.v2.position, data.v0.position, w1, data.area) and
edgeContainsPixel(data.v0.position, data.v1.position, w2, data.area);
if (inside) {
const bit_index: u5 = @intCast(sample_index);
mask |= @as(vk.SampleMask, 1) << bit_index;
}
}
return mask;
}
fn applyEarlyDepth(data: RunData, coverage_sample_mask: vk.SampleMask, x: i32, y: i32, z: f32, sample_count: usize) VkError!struct {
mask: vk.SampleMask,
applied: bool,
} {
if (!data.early_fragment_tests)
return .{ .mask = coverage_sample_mask, .applied = false };
const depth = data.depth_attachment_access orelse return .{ .mask = coverage_sample_mask, .applied = false };
const pipeline_data = data.draw_call.renderer.state.pipeline.?.interface.mode.graphics;
const depth_stencil_state = if (pipeline_data.depth_stencil) |state| common.resolveDepthStencilState(data.draw_call, state) else null;
const io = data.draw_call.renderer.device.interface.io();
var passed_mask: vk.SampleMask = 0;
for (0..sample_count) |sample_index| {
if (sample_index >= @bitSizeOf(vk.SampleMask))
break;
const bit = @as(vk.SampleMask, 1) << @as(u5, @intCast(sample_index));
if ((coverage_sample_mask & bit) == 0)
continue;
if (try common.depthTestSampleAndUpdate(io, depth, @intCast(x), @intCast(y), sample_index, z, depth_stencil_state))
passed_mask |= bit;
}
return .{ .mask = passed_mask, .applied = true };
}
fn biasedDepth(data: RunData, z: f32) f32 {
const pipeline_data = data.draw_call.renderer.state.pipeline.?.interface.mode.graphics;
if (pipeline_data.rasterization.depth_bias_enable == .false)
return z;
const depth = data.depth_attachment_access orelse return z;
const bias_state: Renderer.DepthBias = if (pipeline_data.dynamic_state.depth_bias)
data.draw_call.renderer.dynamic_state.depth_bias orelse Renderer.DepthBias{
.constant_factor = 0.0,
.clamp = 0.0,
.slope_factor = 0.0,
}
else
Renderer.DepthBias{
.constant_factor = pipeline_data.rasterization.depth_bias_constant_factor,
.clamp = pipeline_data.rasterization.depth_bias_clamp,
.slope_factor = pipeline_data.rasterization.depth_bias_slope_factor,
};
const bias =
bias_state.constant_factor * common.depthBiasConstantUnit(depth.format, z) +
bias_state.slope_factor * data.depth_bias_slope;
return z + common.clampDepthBias(bias, bias_state.clamp);
}
fn runWrapper(data: RunData) void {
@call(.always_inline, run, .{data}) catch |err| {
std.log.scoped(.@"Rasterization stage").err("triangle fill mode catched a '{s}'", .{@errorName(err)});
if (comptime base.config.logs == .verbose) {
if (@errorReturnTrace()) |trace| {
std.debug.dumpErrorReturnTrace(trace);
}
}
};
}
inline fn run(data: RunData) !void {
var y = data.min_y;
while (y <= data.max_y) : (y += 1) {
var x = data.min_x;
while (x <= data.max_x) : (x += 1) {
if (!common.scissorContainsPixel(data.draw_call.scissor, x, y)) {
continue;
}
const pipeline_data = data.draw_call.renderer.state.pipeline.?.interface.mode.graphics;
const sample_count = pipeline_data.multisample.rasterization_samples.toInt();
const coverage_sample_mask = triangleCoverageMask(data, x, y, sample_count);
if (coverage_sample_mask == 0)
continue;
const p = zm.f32x4(@as(f32, @floatFromInt(x)) + 0.5, @as(f32, @floatFromInt(y)) + 0.5, 0.0, 1.0);
const w0 = edgeFunction(data.v1.position, data.v2.position, p);
const w1 = edgeFunction(data.v2.position, data.v0.position, p);
const w2 = edgeFunction(data.v0.position, data.v1.position, p);
const b0 = w0 / data.area;
const b1 = w1 / data.area;
const b2 = w2 / data.area;
const z = (b0 * data.v0.position[2]) + (b1 * data.v1.position[2]) + (b2 * data.v2.position[2]);
const depth_z = biasedDepth(data, z);
const frag_w = (b0 / data.v0.position[3]) + (b1 / data.v1.position[3]) + (b2 / data.v2.position[3]);
const early_depth = try applyEarlyDepth(data, coverage_sample_mask, x, y, depth_z, sample_count);
if (early_depth.mask == 0)
continue;
const interpolation_barycentrics = if (sample_count > 1) blk: {
const sample_pos = firstCoveredSamplePosition(sample_count, early_depth.mask);
const centroid_p = zm.f32x4(
@as(f32, @floatFromInt(x)) + sample_pos.x,
@as(f32, @floatFromInt(y)) + sample_pos.y,
0.0,
1.0,
);
const centroid_w0 = edgeFunction(data.v1.position, data.v2.position, centroid_p);
const centroid_w1 = edgeFunction(data.v2.position, data.v0.position, centroid_p);
const centroid_w2 = edgeFunction(data.v0.position, data.v1.position, centroid_p);
break :blk .{
centroid_w0 / data.area,
centroid_w1 / data.area,
centroid_w2 / data.area,
};
} else .{ b0, b1, b2 };
const input_b0 = interpolation_barycentrics[0];
const input_b1 = interpolation_barycentrics[1];
const input_b2 = interpolation_barycentrics[2];
var fragment_result: fragment.InvocationResult = .{
.outputs = std.mem.zeroes([spv.SPIRV_MAX_OUTPUT_LOCATIONS][@sizeOf(F32x4)]u8),
.depth = null,
.sample_mask = null,
};
if (data.has_fragment_shader and data.fragment_uses_sample_id and sample_count > 1) {
for (0..sample_count) |sample_index| {
if (sample_index >= @bitSizeOf(vk.SampleMask))
break;
const bit_index: u5 = @intCast(sample_index);
const sample_coverage_mask = @as(vk.SampleMask, 1) << bit_index;
if ((early_depth.mask & sample_coverage_mask) == 0)
continue;
const inputs = try common.interpolateVertexOutputs(data.allocator, &data.v0, &data.v1, &data.v2, &data.provoking_vertex, input_b0, input_b1, input_b2);
const sample_result = fragment.shaderInvocation(
data.allocator,
data.draw_call,
data.batch_id,
zm.f32x4(@as(f32, @floatFromInt(x)) + 0.5, @as(f32, @floatFromInt(y)) + 0.5, z, frag_w),
null,
@intCast(sample_index),
data.front_face,
inputs,
null,
) catch |err| {
if (err == SpvRuntimeError.Killed)
continue;
std.log.scoped(.@"Fragment stage").err("catched a '{s}'", .{@errorName(err)});
if (comptime base.config.logs == .verbose) {
if (@errorReturnTrace()) |trace| {
std.debug.dumpErrorReturnTrace(trace);
}
}
return;
};
try common.writeToTargets(
sample_result.outputs,
data.draw_call,
data.color_attachment_access,
data.depth_attachment_access,
data.stencil_attachment_access,
data.front_face,
@intCast(x),
@intCast(y),
sample_result.depth orelse depth_z,
sample_coverage_mask,
sample_result.sample_mask,
early_depth.applied,
);
}
continue;
}
if (data.has_fragment_shader) {
const inputs = try common.interpolateVertexOutputs(data.allocator, &data.v0, &data.v1, &data.v2, &data.provoking_vertex, input_b0, input_b1, input_b2);
const derivative_inputs: ?fragment.DerivativeInputs = if (data.fragment_uses_derivatives) blk: {
var derivatives: fragment.DerivativeInputs = undefined;
const p_dx = zm.f32x4(@as(f32, @floatFromInt(x)) + 1.5, @as(f32, @floatFromInt(y)) + 0.5, 0.0, 1.0);
const dx_w0 = edgeFunction(data.v1.position, data.v2.position, p_dx);
const dx_w1 = edgeFunction(data.v2.position, data.v0.position, p_dx);
const dx_w2 = edgeFunction(data.v0.position, data.v1.position, p_dx);
derivatives.dx = try common.interpolateVertexOutputDerivatives(
data.allocator,
&data.v0,
&data.v1,
&data.v2,
b0,
b1,
b2,
(dx_w0 / data.area) - b0,
(dx_w1 / data.area) - b1,
(dx_w2 / data.area) - b2,
);
const p_dy = zm.f32x4(@as(f32, @floatFromInt(x)) + 0.5, @as(f32, @floatFromInt(y)) + 1.5, 0.0, 1.0);
const dy_w0 = edgeFunction(data.v1.position, data.v2.position, p_dy);
const dy_w1 = edgeFunction(data.v2.position, data.v0.position, p_dy);
const dy_w2 = edgeFunction(data.v0.position, data.v1.position, p_dy);
derivatives.dy = try common.interpolateVertexOutputDerivatives(
data.allocator,
&data.v0,
&data.v1,
&data.v2,
b0,
b1,
b2,
(dy_w0 / data.area) - b0,
(dy_w1 / data.area) - b1,
(dy_w2 / data.area) - b2,
);
break :blk derivatives;
} else null;
fragment_result = fragment.shaderInvocation(
data.allocator,
data.draw_call,
data.batch_id,
zm.f32x4(@as(f32, @floatFromInt(x)) + 0.5, @as(f32, @floatFromInt(y)) + 0.5, z, frag_w),
null,
null,
data.front_face,
inputs,
derivative_inputs,
) catch |err| {
if (err == SpvRuntimeError.Killed)
continue;
std.log.scoped(.@"Fragment stage").err("catched a '{s}'", .{@errorName(err)});
if (comptime base.config.logs == .verbose) {
if (@errorReturnTrace()) |trace| {
std.debug.dumpErrorReturnTrace(trace);
}
}
return;
};
}
try common.writeToTargets(
fragment_result.outputs,
data.draw_call,
data.color_attachment_access,
data.depth_attachment_access,
data.stencil_attachment_access,
data.front_face,
@intCast(x),
@intCast(y),
fragment_result.depth orelse depth_z,
early_depth.mask,
fragment_result.sample_mask,
early_depth.applied,
);
}
}
}