vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Fix cast in opencl kernels

Browse source
This commit is contained in:
0cc4m 2023-04-24 16:13:43 +02:00
parent 8603c25e3c
commit 18cc05bde4
1 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

6
ggml_clblast_dequant.cl
View file

@ -51,7 +51,7 @@ __kernel void dequantize_row_q4_2(__global struct block_q4_2* blocks, __global f
const uint i = get_global_id(0) / 16;
const uint l = get_local_id(0);
const float d = vload_half(0, (const half*) &blocks[i].d);;
const float d = vload_half(0, (__global half*) &blocks[i].d);;
const uchar vi = blocks[i].qs[l];
@ -71,8 +71,8 @@ __kernel void dequantize_row_q4_3(__global struct block_q4_3* blocks, __global f
const uint i = get_global_id(0) / 16;
const uint l = get_local_id(0);
const float d = vload_half(0, (const half*) &(blocks[i].d));
const float m = vload_half(0, (const half*) &(blocks[i].m));
const float d = vload_half(0, (__global half*) &(blocks[i].d));
const float m = vload_half(0, (__global half*) &(blocks[i].m));
const uchar vi = blocks[i].qs[l];