From d1faeca19dd588e4930a90028e0200c84371d16c Mon Sep 17 00:00:00 2001 From: slaren Date: Thu, 31 Oct 2024 17:49:51 +0100 Subject: [PATCH] cuda : fix supports_op for norm --- ggml/src/ggml-cuda.cu | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/ggml/src/ggml-cuda.cu b/ggml/src/ggml-cuda.cu index 087091516..b57f1b3b7 100644 --- a/ggml/src/ggml-cuda.cu +++ b/ggml/src/ggml-cuda.cu @@ -3107,18 +3107,20 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g } return false; } break; + case GGML_OP_NORM: + case GGML_OP_RMS_NORM: + return ggml_is_contiguous(op->src[0]) && op->ne[0] % WARP_SIZE == 0; + break; case GGML_OP_NONE: case GGML_OP_RESHAPE: case GGML_OP_VIEW: case GGML_OP_PERMUTE: case GGML_OP_TRANSPOSE: - case GGML_OP_NORM: case GGML_OP_ADD: case GGML_OP_ADD1: case GGML_OP_SUB: case GGML_OP_MUL: case GGML_OP_DIV: - case GGML_OP_RMS_NORM: case GGML_OP_SCALE: case GGML_OP_SQR: case GGML_OP_SQRT: