ebiten/internal/graphics/vertex.go

// Copyright 2019 The Ebiten Authors
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

package graphics

import (
	"sync"
)

const (
	ShaderImageNum = 4

	// PreservedUniformVariablesNum represents the number of preserved uniform variables.
	// Any shaders in Ebiten must have these uniform variables.
	PreservedUniformVariablesNum = 1 + // the destination texture size
		1 + // the texture sizes array
		1 + // the texture destination region's origin
		1 + // the texture destination region's size
		1 + // the offsets array of the second and the following images
		1 + // the texture source region's origin
		1 // the texture source region's size

	DestinationTextureSizeUniformVariableIndex         = 0
	TextureSizesUniformVariableIndex                   = 1
	TextureDestinationRegionOriginUniformVariableIndex = 2
	TextureDestinationRegionSizeUniformVariableIndex   = 3
	TextureSourceOffsetsUniformVariableIndex           = 4
	TextureSourceRegionOriginUniformVariableIndex      = 5
	TextureSourceRegionSizeUniformVariableIndex        = 6
)

const (
	IndicesNum     = (1 << 16) / 3 * 3 // Adjust num for triangles.
	VertexFloatNum = 8
)

var (
	quadIndices = []uint16{0, 1, 2, 1, 2, 3}
)

func QuadIndices() []uint16 {
	return quadIndices
}

var (
	theVerticesBackend = &verticesBackend{}
)

// TODO: The logic is very similar to atlas.temporaryPixels. Unify them.

type verticesBackend struct {
	backend          []float32
	pos              int
	notFullyUsedTime int

	m sync.Mutex
}

func verticesBackendFloat32Size(size int) int {
	l := 128 * VertexFloatNum
	for l < size {
		l *= 2
	}
	return l
}

func max(a, b int) int {
	if a > b {
		return a
	}
	return b
}

func (v *verticesBackend) slice(n int) []float32 {
	v.m.Lock()
	defer v.m.Unlock()

	need := n * VertexFloatNum
	if len(v.backend) < v.pos+need {
		v.backend = make([]float32, max(len(v.backend)*2, verticesBackendFloat32Size(need)))
		v.pos = 0
	}
	s := v.backend[v.pos : v.pos+need]
	v.pos += need
	return s
}

func (v *verticesBackend) lockAndReset(f func() error) error {
	v.m.Lock()
	defer v.m.Unlock()

	if err := f(); err != nil {
		return err
	}

	const maxNotFullyUsedTime = 60
	if verticesBackendFloat32Size(v.pos) < len(v.backend) {
		if v.notFullyUsedTime < maxNotFullyUsedTime {
			v.notFullyUsedTime++
		}
	} else {
		v.notFullyUsedTime = 0
	}

	if v.notFullyUsedTime == maxNotFullyUsedTime && len(v.backend) > 0 {
		v.backend = nil
		v.notFullyUsedTime = 0
	}

	v.pos = 0
	return nil
}

// Vertices returns a float32 slice for n vertices.
// Vertices returns a slice that never overlaps with other slices returned this function,
// and users can do optimization based on this fact.
func Vertices(n int) []float32 {
	return theVerticesBackend.slice(n)
}

func LockAndResetVertices(f func() error) error {
	return theVerticesBackend.lockAndReset(f)
}

// QuadVertices returns a float32 slice for a quadrangle.
// QuadVertices returns a slice that never overlaps with other slices returned this function,
// and users can do optimization based on this fact.
func QuadVertices(sx0, sy0, sx1, sy1 float32, a, b, c, d, tx, ty float32, cr, cg, cb, ca float32) []float32 {
	x := sx1 - sx0
	y := sy1 - sy0
	ax, by, cx, dy := a*x, b*y, c*x, d*y
	u0, v0, u1, v1 := float32(sx0), float32(sy0), float32(sx1), float32(sy1)

	// Use the vertex backend instead of calling make to reduce GCs (#1521).
	vs := theVerticesBackend.slice(4)

	// This function is very performance-sensitive and implement in a very dumb way.
	_ = vs[:4*VertexFloatNum]

	vs[0] = tx
	vs[1] = ty
	vs[2] = u0
	vs[3] = v0
	vs[4] = cr
	vs[5] = cg
	vs[6] = cb
	vs[7] = ca

	vs[8] = ax + tx
	vs[9] = cx + ty
	vs[10] = u1
	vs[11] = v0
	vs[12] = cr
	vs[13] = cg
	vs[14] = cb
	vs[15] = ca

	vs[16] = by + tx
	vs[17] = dy + ty
	vs[18] = u0
	vs[19] = v1
	vs[20] = cr
	vs[21] = cg
	vs[22] = cb
	vs[23] = ca

	vs[24] = ax + by + tx
	vs[25] = cx + dy + ty
	vs[26] = u1
	vs[27] = v1
	vs[28] = cr
	vs[29] = cg
	vs[30] = cb
	vs[31] = ca

	return vs
}