Commit 10152bc4a for llama.cpp

commit 10152bc4a05742202dd24ee35352161ca84a7964
Author: karan <67942227+dheepakkaran@users.noreply.github.com>
Date:   Tue Nov 17 10:34:04 2026 -0500

    ggml-cuda: disable quantized matvec fusion on all Pascal GPUs (#32239)

diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
index b0fd03654..cc3ce73d0 100644
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
@@ -2263,7 +2263,7 @@ static bool ggml_cuda_should_fuse_mul_mat_vec_q(const ggml_tensor * tensor) {
                              src1->type == GGML_TYPE_F32 && dst->type == GGML_TYPE_F32;

     // fusion is not universally faster on Pascal
-    if (cc <= GGML_CUDA_CC_PASCAL) {
+    if (cc < GGML_CUDA_CC_VOLTA) {
         return false;
     }
     //we only support fusion for ncols_dst = 1