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

kompute : implement op_getrows_f32 (#6403)

Browse source 
op_getrows_f32 is required since https://github.com/ggerganov/llama.cpp/pull/6122
for the Vulkan w/ Kompute backend to be functional.

As such, implement this op to make this backend functional again.
This commit is contained in:
woachk 2024-06-03 07:32:16 +02:00 • committed by GitHub
parent 3413ae2193
commit 9e405b6e2e
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 46 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
CMakeLists.txt
View file

@ -777,6 +777,7 @@ if (LLAMA_KOMPUTE)
kompute-shaders/op_mul_mat_q4_0.comp
kompute-shaders/op_mul_mat_q4_1.comp
kompute-shaders/op_mul_mat_q6_k.comp
kompute-shaders/op_getrows_f32.comp
kompute-shaders/op_getrows_f16.comp
kompute-shaders/op_getrows_q4_0.comp
kompute-shaders/op_getrows_q4_1.comp
@ -809,6 +810,7 @@ if (LLAMA_KOMPUTE)
shaderop_mul_mat_q4_0.h
shaderop_mul_mat_q4_1.h
shaderop_mul_mat_q6_k.h
shaderop_getrows_f32.h
shaderop_getrows_f16.h
shaderop_getrows_q4_0.h
shaderop_getrows_q4_1.h