ollama/llama/patches/0001-cuda.patch

From 702ee500b229e910e3e6cd3c84d87763c51fb411 Mon Sep 17 00:00:00 2001
From: jmorganca <jmorganca@gmail.com>
Date: Thu, 6 Jun 2024 23:55:47 -0700
Subject: [PATCH 01/11] cuda

---
 ggml/src/ggml-backend.cpp        | 2 +-
 ggml/src/ggml-cuda/ggml-cuda.cu  | 1 +
 ggml/src/ggml-metal/ggml-metal.m | 1 +
 3 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/ggml/src/ggml-backend.cpp b/ggml/src/ggml-backend.cpp
index fdb4b986..731e4078 100644
--- a/ggml/src/ggml-backend.cpp
+++ b/ggml/src/ggml-backend.cpp
@@ -106,7 +106,6 @@ void ggml_backend_buffer_free(ggml_backend_buffer_t buffer) {
     if (buffer->iface.free_buffer != NULL) {
         buffer->iface.free_buffer(buffer);
     }
-    delete buffer;
 }
 
 size_t ggml_backend_buffer_get_size(ggml_backend_buffer_t buffer) {
@@ -1862,6 +1861,7 @@ static void * ggml_backend_cpu_buffer_get_base(ggml_backend_buffer_t buffer) {
 
 static void ggml_backend_cpu_buffer_free_buffer(ggml_backend_buffer_t buffer) {
     ggml_aligned_free(buffer->context, buffer->size);
+    free(buffer);
 }
 
 static void ggml_backend_cpu_buffer_memset_tensor(ggml_backend_buffer_t buffer, struct ggml_tensor * tensor, uint8_t value, size_t offset, size_t size) {
diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
index d6e4bfdd..a2fcfe5d 100644
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
@@ -424,6 +424,7 @@ struct ggml_backend_cuda_buffer_context {
 static void ggml_backend_cuda_buffer_free_buffer(ggml_backend_buffer_t buffer) {
     ggml_backend_cuda_buffer_context * ctx = (ggml_backend_cuda_buffer_context *)buffer->context;
     delete ctx;
+    delete buffer;
 }
 
 static bool ggml_backend_buffer_is_cuda(ggml_backend_buffer_t buffer) {
diff --git a/ggml/src/ggml-metal/ggml-metal.m b/ggml/src/ggml-metal/ggml-metal.m
index 093ae900..a0cf4ba4 100644
--- a/ggml/src/ggml-metal/ggml-metal.m
+++ b/ggml/src/ggml-metal/ggml-metal.m
@@ -4035,6 +4035,7 @@ static void ggml_backend_metal_buffer_free_buffer(ggml_backend_buffer_t buffer)
     }
 
     free(ctx);
+    free(buffer);
 }
 
 static void * ggml_backend_metal_buffer_get_base(ggml_backend_buffer_t buffer) {
-- 
2.46.0