mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2025-11-15 11:17:31 +00:00
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
This commit is contained in:
20
vulkan-shaders/dequant_f32.comp
Normal file
20
vulkan-shaders/dequant_f32.comp
Normal file
@@ -0,0 +1,20 @@
|
||||
#version 450
|
||||
|
||||
#include "dequant_head.comp"
|
||||
|
||||
layout(local_size_x = 256, local_size_y = 1, local_size_z = 1) in;
|
||||
|
||||
layout (binding = 0) readonly buffer A {float data_a[];};
|
||||
layout (binding = 1) writeonly buffer D {D_TYPE data_b[];};
|
||||
|
||||
void main() {
|
||||
const uint i = gl_GlobalInvocationID.x * 16;
|
||||
|
||||
if (i >= p.nel) {
|
||||
return;
|
||||
}
|
||||
|
||||
[[unroll]] for (uint l = 0; l < 16; l++) {
|
||||
data_b[i + l] = D_TYPE(data_a[i + l]);
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user