llama.cpp/vulkan-shaders/square.comp
0cc4m 7c7836d9d4
Vulkan Shader Refactor, Memory Debugging Option (#7947)
* Refactor shaders, extract GLSL code from ggml_vk_generate_shaders.py into vulkan-shaders directory

* Improve debug log code

* Add memory debug output option

* Fix flake8

* Fix unnecessary high llama-3 VRAM use
2024-06-16 07:17:31 +02:00

14 lines
315 B
Plaintext

#version 450
#include "types.comp"
#include "generic_unary_head.comp"
void main() {
if (gl_GlobalInvocationID.x >= p.ne) {
return;
}
const FLOAT_TYPE val = FLOAT_TYPE(data_a[src0_idx(gl_GlobalInvocationID.x)]);
data_d[p.d_offset + dst_idx(gl_GlobalInvocationID.x)] = D_TYPE(val * val);
}