AclyandJeff Bolz
e29acf74fe
* vulkan (DRAFT): split shader generation by GLSL source file, to improve incremental build times * support dep-files so shaders are recompiled if their included files change * rename shader files which are used as "headers" to use .glsl extension * move glslc extension detection shaders to separate folders * the above is to prevent them from getting glob'd with the actual compute shaders that need to be compiled * vulkan : only write embedded shader .hpp/.cpp when they change * avoid recompiling ggml-vulkan.cpp when editing shaders * pass single --source argument instead of --input-dir & --filter to shader gen * check for source file match earlier * fix hang in vulkan-shaders-gen when there are compilation errors * early out did not decrement compile_count * clean up * fix glslc integer dot product test * unconditionally write the embedded shader cpp output * replace output filepath in generated dep-files to match output in CMakeLists --------- Co-authored-by: Jeff Bolz <jbolz@nvidia.com>
43 lines
1.2 KiB
Plaintext
43 lines
1.2 KiB
Plaintext
#version 450
|
|
|
|
#include "generic_head.glsl"
|
|
#include "types.glsl"
|
|
|
|
#extension GL_EXT_control_flow_attributes : enable
|
|
|
|
layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
|
|
|
|
layout (binding = 0) buffer X {A_TYPE x[];};
|
|
layout (binding = 1) readonly buffer G {A_TYPE grad[];};
|
|
layout (binding = 2) buffer GM {A_TYPE gradm[];};
|
|
layout (binding = 3) buffer GV {A_TYPE gradv[];};
|
|
layout (binding = 4) readonly buffer P {float params[7];};
|
|
|
|
void main() {
|
|
const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
|
|
|
|
if (i >= p.KX) {
|
|
return;
|
|
}
|
|
|
|
const float alpha = params[0];
|
|
const float beta1 = params[1];
|
|
const float beta2 = params[2];
|
|
const float eps = params[3];
|
|
const float wd = params[4];
|
|
const float beta1h = params[5];
|
|
const float beta2h = params[6];
|
|
|
|
const float gi = grad[i];
|
|
const float gmi = gradm[i]*beta1 + gi*(1.0f - beta1);
|
|
const float gvi = gradv[i]*beta2 + gi*gi*(1.0f - beta2);
|
|
|
|
gradm[i] = gmi;
|
|
gradv[i] = gvi;
|
|
|
|
const float mh = gmi*beta1h;
|
|
const float vh = sqrt(gvi*beta2h) + eps;
|
|
|
|
x[i] = x[i]*(1.0f - alpha*wd) - alpha*mh/vh;
|
|
}
|