1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205

use crate::ffi;
use crate::{DecodePosition, VertexDataAdapter};
use std::mem;

/// Reorders indices to reduce the number of GPU vertex shader invocations.
///
/// If index buffer contains multiple ranges for multiple draw calls,
/// this function needs to be called on each range individually.
pub fn optimize_vertex_cache(indices: &[u32], vertex_count: usize) -> Vec<u32> {
    let mut optimized: Vec<u32> = vec![0; indices.len()];
    unsafe {
        ffi::meshopt_optimizeVertexCache(
            optimized.as_mut_ptr() as *mut ::std::os::raw::c_uint,
            indices.as_ptr() as *const ::std::os::raw::c_uint,
            indices.len(),
            vertex_count,
        );
    }
    optimized
}

/// Reorders indices to reduce the number of GPU vertex shader invocations.
///
/// If index buffer contains multiple ranges for multiple draw calls,
/// this function needs to be called on each range individually.
pub fn optimize_vertex_cache_in_place(indices: &[u32], vertex_count: usize) {
    unsafe {
        ffi::meshopt_optimizeVertexCache(
            indices.as_ptr() as *mut ::std::os::raw::c_uint,
            indices.as_ptr() as *const ::std::os::raw::c_uint,
            indices.len(),
            vertex_count,
        );
    }
}

/// Vertex transform cache optimizer for FIFO caches.
///
/// Reorders indices to reduce the number of GPU vertex shader invocations.
///
/// Generally takes ~3x less time to optimize meshes but produces inferior
/// results compared to `optimize_vertex_cache`.
///
/// If index buffer contains multiple ranges for multiple draw calls,
/// this function needs to be called on each range individually.
pub fn optimize_vertex_cache_fifo(
    indices: &[u32],
    vertex_count: usize,
    cache_size: u32,
) -> Vec<u32> {
    let mut optimized: Vec<u32> = vec![0; indices.len()];
    unsafe {
        ffi::meshopt_optimizeVertexCacheFifo(
            optimized.as_mut_ptr() as *mut ::std::os::raw::c_uint,
            indices.as_ptr() as *const ::std::os::raw::c_uint,
            indices.len(),
            vertex_count,
            cache_size,
        );
    }
    optimized
}

/// Vertex transform cache optimizer for FIFO caches (in place).
///
/// Reorders indices to reduce the number of GPU vertex shader invocations.
///
/// Generally takes ~3x less time to optimize meshes but produces inferior
/// results compared to `optimize_vertex_cache_fifo_in_place`.
///
/// If index buffer contains multiple ranges for multiple draw calls,
/// this function needs to be called on each range individually.
pub fn optimize_vertex_cache_fifo_in_place(
    indices: &mut [u32],
    vertex_count: usize,
    cache_size: u32,
) {
    unsafe {
        ffi::meshopt_optimizeVertexCacheFifo(
            indices.as_mut_ptr() as *mut ::std::os::raw::c_uint,
            indices.as_ptr() as *const ::std::os::raw::c_uint,
            indices.len(),
            vertex_count,
            cache_size,
        );
    }
}

/// Reorders vertices and changes indices to reduce the amount of GPU
/// memory fetches during vertex processing.
///
/// This functions works for a single vertex stream; for multiple vertex streams,
/// use `optimize_vertex_fetch_remap` + `remap_vertex_buffer` for each stream.
///
/// `indices` is used both as an input and as an output index buffer.
pub fn optimize_vertex_fetch<T: Clone + Default>(indices: &mut [u32], vertices: &[T]) -> Vec<T> {
    let mut result: Vec<T> = vec![T::default(); vertices.len()];
    let next_vertex = unsafe {
        ffi::meshopt_optimizeVertexFetch(
            result.as_mut_ptr() as *mut ::std::os::raw::c_void,
            indices.as_mut_ptr() as *mut ::std::os::raw::c_uint,
            indices.len(),
            vertices.as_ptr() as *const ::std::os::raw::c_void,
            vertices.len(),
            mem::size_of::<T>(),
        )
    };
    result.resize(next_vertex, T::default());
    result
}

/// Vertex fetch cache optimizer (modifies in place)
/// Reorders vertices and changes indices to reduce the amount of GPU
/// memory fetches during vertex processing.
///
/// This functions works for a single vertex stream; for multiple vertex streams,
/// use `optimize_vertex_fetch_remap` + `remap_vertex_buffer` for each stream.
///
/// `indices` and `vertices` are used both as an input and as an output buffer.
pub fn optimize_vertex_fetch_in_place<T>(indices: &mut [u32], vertices: &mut [T]) -> usize {
    unsafe {
        ffi::meshopt_optimizeVertexFetch(
            vertices.as_mut_ptr() as *mut ::std::os::raw::c_void,
            indices.as_mut_ptr() as *mut ::std::os::raw::c_uint,
            indices.len(),
            vertices.as_ptr() as *const ::std::os::raw::c_void,
            vertices.len(),
            mem::size_of::<T>(),
        )
    }
}

/// Generates vertex remap to reduce the amount of GPU memory fetches during
/// vertex processing.
///
/// The resulting remap table should be used to reorder vertex/index buffers
/// using `optimize_remap_vertex_buffer`/`optimize_remap_index_buffer`.
pub fn optimize_vertex_fetch_remap(indices: &[u32], vertex_count: usize) -> Vec<u32> {
    let mut result: Vec<u32> = vec![0; vertex_count];
    let next_vertex = unsafe {
        ffi::meshopt_optimizeVertexFetchRemap(
            result.as_mut_ptr() as *mut ::std::os::raw::c_uint,
            indices.as_ptr() as *const ::std::os::raw::c_uint,
            indices.len(),
            vertex_count,
        )
    };
    result.resize(next_vertex, 0u32);
    result
}

/// Reorders indices to reduce the number of GPU vertex shader invocations
/// and the pixel overdraw.
///
/// `indices` must contain index data that is the result of `optimize_vertex_cache`
/// (*not* the original mesh indices!)
///
/// `threshold` indicates how much the overdraw optimizer can degrade vertex cache
/// efficiency (1.05 = up to 5%) to reduce overdraw more efficiently.
pub fn optimize_overdraw_in_place(indices: &[u32], vertices: &VertexDataAdapter, threshold: f32) {
    let vertex_data = vertices.reader.get_ref();
    let vertex_data = vertex_data.as_ptr() as *const u8;
    let positions = unsafe { vertex_data.add(vertices.position_offset) };
    unsafe {
        ffi::meshopt_optimizeOverdraw(
            indices.as_ptr() as *mut ::std::os::raw::c_uint,
            indices.as_ptr() as *const ::std::os::raw::c_uint,
            indices.len(),
            positions as *const f32,
            vertices.vertex_count,
            vertices.vertex_stride,
            threshold,
        );
    }
}

/// Reorders indices to reduce the number of GPU vertex shader invocations
/// and the pixel overdraw.
///
/// `indices` must contain index data that is the result of `optimize_vertex_cache`
/// (*not* the original mesh indices!)
///
/// `threshold` indicates how much the overdraw optimizer can degrade vertex cache
/// efficiency (1.05 = up to 5%) to reduce overdraw more efficiently.
pub fn optimize_overdraw_in_place_decoder<T: DecodePosition>(
    indices: &mut [u32],
    vertices: &[T],
    threshold: f32,
) {
    let positions = vertices
        .iter()
        .map(|vertex| vertex.decode_position())
        .collect::<Vec<[f32; 3]>>();
    unsafe {
        ffi::meshopt_optimizeOverdraw(
            indices.as_mut_ptr() as *mut ::std::os::raw::c_uint,
            indices.as_ptr() as *const ::std::os::raw::c_uint,
            indices.len(),
            positions.as_ptr() as *const f32,
            positions.len(),
            mem::size_of::<f32>() * 3,
            threshold,
        );
    }
}