slosh3d 0.4.1

Cross-platform GPU 3D Material Point Method implementation.
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
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243

struct GpuTimestepBounds_std430_0
{
    @align(4) computed_max_dt_as_uint_0 : atomic<u32>,
};

@binding(0) @group(0) var<storage, read_write> entryPointParams_result_0 : array<GpuTimestepBounds_std430_0>;

struct GridGeneric_std430_0
{
    @align(4) num_active_blocks_0 : u32,
    @align(4) cell_width_0 : f32,
    @align(4) hmap_capacity_0 : u32,
    @align(4) capacity_0 : u32,
};

@binding(1) @group(0) var<storage, read> entryPointParams_grid_0 : array<GridGeneric_std430_0>;

@binding(2) @group(0) var<storage, read> entryPointParams_particles_model_0 : array<u32>;

struct _MatrixStorage_float2x2std430_0
{
    @align(8) data_0 : array<vec2<f32>, i32(2)>,
};

struct Cdf_std430_0
{
    @align(8) normal_0 : vec2<f32>,
    @align(8) rigid_vel_0 : vec2<f32>,
    @align(8) signed_distance_0 : f32,
    @align(4) affinity_0 : u32,
};

struct Dynamics_std430_0
{
    @align(8) velocity_0 : vec2<f32>,
    @align(8) def_grad_0 : _MatrixStorage_float2x2std430_0,
    @align(8) affine_0 : _MatrixStorage_float2x2std430_0,
    @align(8) vel_grad_det_0 : f32,
    @align(8) cdf_0 : Cdf_std430_0,
    @align(8) init_volume_0 : f32,
    @align(4) init_radius_0 : f32,
    @align(8) mass_0 : f32,
    @align(4) phase_0 : f32,
    @align(8) enabled_0 : u32,
};

@binding(3) @group(0) var<storage, read> entryPointParams_particles_dyn_0 : array<Dynamics_std430_0>;

@binding(4) @group(0) var<uniform> entryPointParams_particles_len_0 : u32;
@binding(5) @group(0) var<storage, read_write> entryPointParams_result_1 : array<GpuTimestepBounds_std430_0>;

@compute
@workgroup_size(1, 1, 1)
fn reset_timestep_bound()
{
    atomicStore(&(entryPointParams_result_0[i32(0)].computed_max_dt_as_uint_0), u32(4294967295));
    return;
}

fn Dynamics_init_density_0( this_0 : ptr<function, Dynamics_std430_0>) -> f32
{
    return (*this_0).mass_0 / (*this_0).init_volume_0;
}

fn bulk_modulus_from_lame_0( lambda_0 : f32,  mu_0 : f32) -> f32
{
    return lambda_0 + 2.0f * mu_0 / 3.0f;
}

fn shear_modulus_from_lame_0( _lambda_0 : f32,  mu_1 : f32) -> f32
{
    return mu_1;
}

fn hook_to_bulk_modulus_0( young_modulus_0 : f32,  poisson_ratio_0 : f32) -> f32
{
    return young_modulus_0 / (3.0f * (1.0f - 2.0f * poisson_ratio_0));
}

fn hook_to_shear_modulus_0( young_modulus_1 : f32,  poisson_ratio_1 : f32) -> f32
{
    return young_modulus_1 / (2.0f * (1.0f + poisson_ratio_1));
}

struct ElasticitySoundSpeedTimestepBound_0
{
     alpha_0 : f32,
     bulk_modulus_0 : f32,
     shear_modulus_0 : f32,
};

fn ElasticitySoundSpeedTimestepBound_x24init_0( alpha_1 : f32,  young_modulus_2 : f32,  poisson_ratio_2 : f32) -> ElasticitySoundSpeedTimestepBound_0
{
    var _S1 : ElasticitySoundSpeedTimestepBound_0;
    _S1.alpha_0 = alpha_1;
    _S1.bulk_modulus_0 = hook_to_bulk_modulus_0(young_modulus_2, poisson_ratio_2);
    _S1.shear_modulus_0 = hook_to_shear_modulus_0(young_modulus_2, poisson_ratio_2);
    return _S1;
}

fn ElasticitySoundSpeedTimestepBound_timestep_bound_0( this_1 : ElasticitySoundSpeedTimestepBound_0,  density0_0 : f32,  def_grad_det_0 : f32,  velocity_1 : vec2<f32>,  cell_width_1 : f32) -> f32
{
    return this_1.alpha_0 * cell_width_1 / max(length(velocity_1), sqrt((this_1.bulk_modulus_0 + this_1.shear_modulus_0 * 4.0f / 3.0f) / (density0_0 / max(def_grad_det_0, 9.99999997475242708e-07f))));
}

struct LinearElasticModel_0
{
     lambda_1 : f32,
     mu_2 : f32,
     cfl_coeff_0 : f32,
};

fn LinearElasticModel_timestep_bound_0( this_2 : LinearElasticModel_0,  particle_density0_0 : f32,  particle_velocity_0 : vec2<f32>,  particle_def_grad_det_0 : f32,  elastic_hardening_0 : f32,  cell_width_2 : f32) -> f32
{
    return ElasticitySoundSpeedTimestepBound_timestep_bound_0(ElasticitySoundSpeedTimestepBound_x24init_0(this_2.cfl_coeff_0, bulk_modulus_from_lame_0(this_2.lambda_1, this_2.mu_2) * elastic_hardening_0, shear_modulus_from_lame_0(this_2.lambda_1, this_2.mu_2) * elastic_hardening_0), particle_density0_0, particle_def_grad_det_0, particle_velocity_0, cell_width_2);
}

struct NeoHookeanModel_0
{
     lambda_2 : f32,
     mu_3 : f32,
     cfl_coeff_1 : f32,
};

fn NeoHookeanModel_timestep_bound_0( this_3 : NeoHookeanModel_0,  particle_density0_1 : f32,  particle_velocity_1 : vec2<f32>,  particle_def_grad_det_1 : f32,  elastic_hardening_1 : f32,  cell_width_3 : f32) -> f32
{
    return ElasticitySoundSpeedTimestepBound_timestep_bound_0(ElasticitySoundSpeedTimestepBound_x24init_0(this_3.cfl_coeff_1, bulk_modulus_from_lame_0(this_3.lambda_2, this_3.mu_3) * elastic_hardening_1, shear_modulus_from_lame_0(this_3.lambda_2, this_3.mu_3) * elastic_hardening_1), particle_density0_1, particle_def_grad_det_1, particle_velocity_1, cell_width_3);
}

fn GpuTimestepBounds_secs_to_int_0( secs_0 : f32) -> u32
{
    return u32(floor(secs_0 * 9.99999995904e+11f));
}

struct DefaultParticleModel_0
{
     dummy_0 : u32,
};

fn DefaultParticleModel_timestep_bound_0( _S2 : DefaultParticleModel_0,  _S3 : u32,  _S4 : f32,  _S5 : mat2x2<f32>,  _S6 : vec2<f32>,  _S7 : f32) -> f32
{
    var _S8 : u32 = _S3 * u32(52);
    var _S9 : u32 = _S8 + u32(4);
    var _S10 : u32 = entryPointParams_particles_model_0[(u32(i32(_S8)))/4];
    var _S11 : f32 = determinant(_S5);
    switch(i32(_S10))
    {
    case i32(0):
        {
            var _S12 : u32 = entryPointParams_particles_model_0[(_S9)/4];
            var _S13 : f32 = bitcast<f32>(_S12);
            var _S14 : u32 = entryPointParams_particles_model_0[(_S9 + u32(4))/4];
            var _S15 : f32 = bitcast<f32>(_S14);
            var _S16 : u32 = entryPointParams_particles_model_0[(_S9 + u32(8))/4];
            var _S17 : LinearElasticModel_0 = LinearElasticModel_0( _S13, _S15, bitcast<f32>(_S16) );
            return LinearElasticModel_timestep_bound_0(_S17, _S4, _S6, _S11, 1.0f, _S7);
        }
    case i32(1):
        {
            var _S18 : u32 = entryPointParams_particles_model_0[(_S9)/4];
            var _S19 : f32 = bitcast<f32>(_S18);
            var _S20 : u32 = entryPointParams_particles_model_0[(_S9 + u32(4))/4];
            var _S21 : f32 = bitcast<f32>(_S20);
            var _S22 : u32 = entryPointParams_particles_model_0[(_S9 + u32(8))/4];
            var _S23 : NeoHookeanModel_0 = NeoHookeanModel_0( _S19, _S21, bitcast<f32>(_S22) );
            return NeoHookeanModel_timestep_bound_0(_S23, _S4, _S6, _S11, 1.0f, _S7);
        }
    case i32(2):
        {
            var _S24 : u32 = entryPointParams_particles_model_0[(_S9)/4];
            var _S25 : u32 = entryPointParams_particles_model_0[(_S9 + u32(4))/4];
            var _S26 : u32 = entryPointParams_particles_model_0[(_S9 + u32(8))/4];
            var _S27 : u32 = entryPointParams_particles_model_0[(_S9 + u32(12))/4];
            var _S28 : u32 = entryPointParams_particles_model_0[(_S9 + u32(16))/4];
            var _S29 : u32 = entryPointParams_particles_model_0[(_S9 + u32(20))/4];
            var _S30 : u32 = entryPointParams_particles_model_0[(_S9 + u32(24))/4];
            var _S31 : u32 = entryPointParams_particles_model_0[(_S9 + u32(28))/4];
            var _S32 : u32 = entryPointParams_particles_model_0[(_S9 + u32(32))/4];
            var _S33 : u32 = entryPointParams_particles_model_0[(_S9 + u32(36))/4];
            var _S34 : f32 = bitcast<f32>(_S33);
            var _S35 : u32 = entryPointParams_particles_model_0[(_S9 + u32(40))/4];
            var _S36 : f32 = bitcast<f32>(_S35);
            var _S37 : u32 = entryPointParams_particles_model_0[(_S9 + u32(44))/4];
            var _S38 : LinearElasticModel_0 = LinearElasticModel_0( _S34, _S36, bitcast<f32>(_S37) );
            return LinearElasticModel_timestep_bound_0(_S38, _S4, _S6, _S11, 1.0f, _S7);
        }
    case i32(3):
        {
            var _S39 : u32 = entryPointParams_particles_model_0[(_S9)/4];
            var _S40 : u32 = entryPointParams_particles_model_0[(_S9 + u32(4))/4];
            var _S41 : u32 = entryPointParams_particles_model_0[(_S9 + u32(8))/4];
            var _S42 : u32 = entryPointParams_particles_model_0[(_S9 + u32(12))/4];
            var _S43 : u32 = entryPointParams_particles_model_0[(_S9 + u32(16))/4];
            var _S44 : u32 = entryPointParams_particles_model_0[(_S9 + u32(20))/4];
            var _S45 : u32 = entryPointParams_particles_model_0[(_S9 + u32(24))/4];
            var _S46 : u32 = entryPointParams_particles_model_0[(_S9 + u32(28))/4];
            var _S47 : u32 = entryPointParams_particles_model_0[(_S9 + u32(32))/4];
            var _S48 : u32 = entryPointParams_particles_model_0[(_S9 + u32(36))/4];
            var _S49 : f32 = bitcast<f32>(_S48);
            var _S50 : u32 = entryPointParams_particles_model_0[(_S9 + u32(40))/4];
            var _S51 : f32 = bitcast<f32>(_S50);
            var _S52 : u32 = entryPointParams_particles_model_0[(_S9 + u32(44))/4];
            var _S53 : NeoHookeanModel_0 = NeoHookeanModel_0( _S49, _S51, bitcast<f32>(_S52) );
            return NeoHookeanModel_timestep_bound_0(_S53, _S4, _S6, _S11, 1.0f, _S7);
        }
    default :
        {
            return 0.0f;
        }
    }
}

@compute
@workgroup_size(64, 1, 1)
fn estimate_timestep_bound(@builtin(global_invocation_id) invocation_id_0 : vec3<u32>)
{
    var _S54 : u32 = invocation_id_0.x;
    if(_S54 >= entryPointParams_particles_len_0)
    {
        return;
    }
    if((entryPointParams_particles_dyn_0[_S54].enabled_0) == u32(0))
    {
        return;
    }
    var _S55 : GridGeneric_std430_0 = entryPointParams_grid_0[i32(0)];
    var _S56 : Dynamics_std430_0 = entryPointParams_particles_dyn_0[_S54];
    var _S57 : f32 = Dynamics_init_density_0(&(_S56));
    var _S58 : Dynamics_std430_0 = entryPointParams_particles_dyn_0[_S54];
    var _S59 : Dynamics_std430_0 = entryPointParams_particles_dyn_0[_S54];
    var _S60 : Dynamics_std430_0 = entryPointParams_particles_dyn_0[_S54];
    var _S61 : mat2x2<f32> = mat2x2<f32>(_S60.affine_0.data_0[i32(0)][i32(0)], _S60.affine_0.data_0[i32(0)][i32(1)], _S60.affine_0.data_0[i32(1)][i32(0)], _S60.affine_0.data_0[i32(1)][i32(1)]);
    var _S62 : Dynamics_std430_0 = entryPointParams_particles_dyn_0[_S54];
    var _S63 : DefaultParticleModel_0 = DefaultParticleModel_0(  );
    var _S64 : f32 = DefaultParticleModel_timestep_bound_0(_S63, _S54, _S57, mat2x2<f32>(_S58.def_grad_0.data_0[i32(0)][i32(0)], _S58.def_grad_0.data_0[i32(0)][i32(1)], _S58.def_grad_0.data_0[i32(1)][i32(0)], _S58.def_grad_0.data_0[i32(1)][i32(1)]), _S59.velocity_0, _S55.cell_width_0);
    var _S65 : f32 = _S61[i32(0)][i32(0)];
    var _S66 : f32 = _S61[i32(0)][i32(1)];
    var _S67 : f32 = _S61[i32(1)][i32(0)];
    var _S68 : f32 = _S61[i32(1)][i32(1)];
    var _S69 : u32 = atomicMin(&(entryPointParams_result_1[i32(0)].computed_max_dt_as_uint_0), GpuTimestepBounds_secs_to_int_0(min(_S64, _S55.cell_width_0 / (length(_S59.velocity_0) + _S55.cell_width_0 * _S55.cell_width_0 / 4.0f * sqrt(_S65 * _S65 + _S66 * _S66 + _S67 * _S67 + _S68 * _S68) / _S62.mass_0 * 6.0f * sqrt(2.0f) / _S55.cell_width_0))));
    return;
}