We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
soft_max
1 parent cd57f51 commit 1ad932cCopy full SHA for 1ad932c
ggml/src/ggml-opencl/kernels/ggml-opencl.cl
@@ -679,6 +679,9 @@ kernel void kernel_diag_mask_inf_8(
679
//------------------------------------------------------------------------------
680
// softmax
681
682
+#ifdef ADRENO_GPU
683
+REQD_SUBGROUP_SIZE_64
684
+#endif
685
kernel void kernel_soft_max(
686
global float * src0,
687
ulong offset0,
@@ -811,6 +814,9 @@ kernel void kernel_soft_max_4(
811
814
}
812
815
813
816
817
818
819
820
kernel void kernel_soft_max_f16(
821
822
0 commit comments