tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-30 06:03:37 -04:00
Code Issues Packages Projects Releases Wiki Activity

Refactor Vulkan backend to allow multiple contexts (#7961)

Browse Source 
* Refactor Vulkan backend to allow multiple contexts

* Fix too many shader groups called validation error in llama3 on AMD and Intel GPUs

* Fix Vulkan debug build error
This commit is contained in:
0cc4m
2024-06-23 10:21:25 +02:00
committed by GitHub
parent b5a5f34efa
commit 45c0e2e4c1
8 changed files with 21021 additions and 20675 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vulkan-shaders/mul_mat_vec_q2_k.comp
View File

@@ -7,7 +7,7 @@ layout(local_size_x = 32, local_size_y = 1, local_size_z = 1) in;
shared FLOAT_TYPE tmp[32];
void main() {
const uint row = gl_WorkGroupID.x;
const uint row = gl_WorkGroupID.x + gl_NumWorkGroups.x * gl_WorkGroupID.z;
uint a_offset, b_offset, d_offset;
get_offsets(a_offset, b_offset, d_offset);