mozempk/ggml-org-llama-cpp-mirror
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-11-11 22:32:07 +01:00
Code Issues Packages Projects Releases Wiki Activity

opencl: add sqr, sqrt, mean and ssm_conv (#17476)

Browse Source 
* opencl: add sqr

* opencl: add sqrt

* opencl: add mean

* opencl: add ssm_conv

* opencl: add missing cl_khr_fp16

* opencl: do sqrt in f32 then convert to f16 for better precision
This commit is contained in:
lhez
2025-11-26 13:29:58 -08:00
committed by GitHub
parent 5449367b21
commit 7cba58bbea
6 changed files with 557 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-opencl/kernels/mean.cl
+39
View File
@@ -0,0 +1,39 @@
kernel void kernel_mean_f32(
global float * src0,
ulong offset0,
global float * dst,
ulong offsetd,
int ne00,
int ne01,
int ne02,
int ne03,
ulong nb01,
ulong nb02,
ulong nb03,
ulong nb1,
ulong nb2,
ulong nb3
) {
src0 = (global float *)((global char *)src0 + offset0);
dst = (global float *)((global char *)dst + offsetd);
int i3 = get_global_id(2);
int i2 = get_global_id(1);
int i1 = get_global_id(0);
if (i3 >= ne03 || i2 >= ne02 || i1 >= ne01) {
return;
}
global float * src_row = (global float *) ((global char *) src0 + i1*nb01 + i2*nb02 + i3*nb03);
global float * dst_row = (global float *) ((global char *) dst + i1*nb1 + i2*nb2 + i3*nb3);
float row_sum = 0;
for (int i0 = 0; i0 < ne00; i0++) {
row_sum += src_row[i0];
}
dst_row[0] = row_sum / ne00;
}