llama: update vendor code to commit ba1cb19c (#8101)

7a81daf0 · Jeffrey Morgan · GitHub · 60f75560 · 7a81daf0 · 7a81daf0
Unverified Commit 7a81daf0 authored Dec 14, 2024 by Jeffrey Morgan Committed by GitHub Dec 14, 2024
20 changed files
--- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu
+++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs256-f16-f16.cu
--- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu
+++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-f16.cu
--- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_0.cu
+++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_0.cu
--- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_1.cu
+++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q4_1.cu
--- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_0.cu
+++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_0.cu
--- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_1.cu
+++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q5_1.cu
--- a/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q8_0.cu
+++ b/llama/ggml-cuda/template-instances/fattn-vec-f32-instance-hs64-f16-q8_0.cu
--- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu
+++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu
--- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu
+++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu
--- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu
+++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu
--- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu
+++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu
--- a/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu
+++ b/llama/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu
--- a/llama/ggml-cuda/template-instances/mmq-instance-iq1_s.cu
+++ b/llama/ggml-cuda/template-instances/mmq-instance-iq1_s.cu
--- a/llama/ggml-cuda/template-instances/mmq-instance-iq2_s.cu
+++ b/llama/ggml-cuda/template-instances/mmq-instance-iq2_s.cu
--- a/llama/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu
+++ b/llama/ggml-cuda/template-instances/mmq-instance-iq2_xs.cu
--- a/llama/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu
+++ b/llama/ggml-cuda/template-instances/mmq-instance-iq2_xxs.cu
--- a/llama/ggml-cuda/template-instances/mmq-instance-iq3_s.cu
+++ b/llama/ggml-cuda/template-instances/mmq-instance-iq3_s.cu
--- a/llama/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu
+++ b/llama/ggml-cuda/template-instances/mmq-instance-iq3_xxs.cu
--- a/llama/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu
+++ b/llama/ggml-cuda/template-instances/mmq-instance-iq4_nl.cu
--- a/llama/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu
+++ b/llama/ggml-cuda/template-instances/mmq-instance-iq4_xs.cu