ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

vulkan: Use push constant offset to handle misaligned descriptors (#10987)

Browse source
This commit is contained in:
Jeff Bolz 2024-12-29 02:35:11 -06:00 • committed by GitHub
parent f865ea149d
commit fdd2188912
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
19 changed files with 103 additions and 42 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ggml/src/ggml-vulkan/vulkan-shaders/acc.comp
View file

@ -21,9 +21,9 @@ void main() {
get_indices(idx, i00, i01, i02, i03);
if (ox < p.ne10 && oy < p.ne11 && oz < p.ne12) {
data_d[p.d_offset + dst_idx(i00, i01, i02, i03)] = D_TYPE(FLOAT_TYPE(data_a[src0_idx(i00, i01, i02, i03)]) + FLOAT_TYPE(data_b[ox + oy * p.ne10 + oz * p.ne10 * p.ne11]));
data_d[get_doffset() + dst_idx(i00, i01, i02, i03)] = D_TYPE(FLOAT_TYPE(data_a[get_aoffset() + src0_idx(i00, i01, i02, i03)]) + FLOAT_TYPE(data_b[get_boffset() + ox + oy * p.ne10 + oz * p.ne10 * p.ne11]));
} else {
data_d[p.d_offset + dst_idx(i00, i01, i02, i03)] = D_TYPE(FLOAT_TYPE(data_a[src0_idx(i00, i01, i02, i03)]));
data_d[get_doffset() + dst_idx(i00, i01, i02, i03)] = D_TYPE(FLOAT_TYPE(data_a[get_aoffset() + src0_idx(i00, i01, i02, i03)]));
}
}