From: slaren <redacted>
Date: Thu, 8 Feb 2024 20:33:03 +0000 (+0100)
Subject: llama : do not print "offloading layers" message in CPU-only builds (#5416)
X-Git-Tag: upstream/0.0.4488~2383
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=41f308f58edc2a04bcf9e245100b0a9b10e9a0fb;p=pkg%2Fggml%2Fsources%2Fllama.cpp

llama : do not print "offloading layers" message in CPU-only builds (#5416)
---

diff --git a/llama.cpp b/llama.cpp
index 89acafbc..db7d1c1c 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -4209,8 +4209,7 @@ static bool llm_load_tensors(
         ctx_bufs.emplace_back(ctx, buf);
     }
 
-    // print memory requirements
-    {
+    if (llama_supports_gpu_offload()) {
         const int n_gpu = std::min(n_gpu_layers, int(hparams.n_layer));
 
         LLAMA_LOG_INFO("%s: offloading %d repeating layers to GPU\n", __func__, n_gpu);
@@ -4222,10 +4221,11 @@ static bool llm_load_tensors(
         const int max_offloadable_layers       = hparams.n_layer + 1;
 
         LLAMA_LOG_INFO("%s: offloaded %d/%d layers to GPU\n", __func__, std::min(n_gpu_layers, max_offloadable_layers), max_backend_supported_layers);
+    }
 
-        for (ggml_backend_buffer_t buf : model.bufs) {
-            LLAMA_LOG_INFO("%s: %10s buffer size = %8.2f MiB\n", __func__, ggml_backend_buffer_name(buf), ggml_backend_buffer_get_size(buf) / 1024.0 / 1024.0);
-        }
+    // print memory requirements
+    for (ggml_backend_buffer_t buf : model.bufs) {
+        LLAMA_LOG_INFO("%s: %10s buffer size = %8.2f MiB\n", __func__, ggml_backend_buffer_name(buf), ggml_backend_buffer_get_size(buf) / 1024.0 / 1024.0);
     }
 
     // populate tensors_by_name