update examples and tests to use ggml_allocr_new_measure_from_backend (#608)

* update examples and tests to use ggml_allocr_new_measure_from_backend

* update comments

diff --git a/examples/gpt-2/main-backend.cpp b/examples/gpt-2/main-backend.cpp
index ab43726039bfb34df68a2afb7c75199a61a27a69..e986cdee1112a2b3d80e9bde20c7a2088f76141d 100644 (file)
--- a/examples/gpt-2/main-backend.cpp
+++ b/examples/gpt-2/main-backend.cpp
@@ -878,9 +878,8 @@ int main(int argc, char ** argv) {
     struct ggml_allocr * allocr = NULL;
     // allocate the compute buffer
     {
-         // alignment required by the backend
-        size_t align = ggml_backend_get_alignment(model.backend);
-        allocr = ggml_allocr_new_measure(align);
+        // create an allocator to measure the memory usage
+        allocr = ggml_allocr_new_measure_from_backend(model.backend);
 
         // create the worst case graph for memory usage estimation
         int n_tokens = std::min(model.hparams.n_ctx, params.n_batch);

diff --git a/examples/gpt-2/main-batched.cpp b/examples/gpt-2/main-batched.cpp
index 99167db99b68dee077c263619d0d4aba275a3f6a..41d531aa0a2261e07446bbccf4afe18b0ff8f9a8 100644 (file)
--- a/examples/gpt-2/main-batched.cpp
+++ b/examples/gpt-2/main-batched.cpp
@@ -1042,9 +1042,8 @@ int main(int argc, char ** argv) {
     // prepare required memory and allocate the compute buffer
     struct ggml_allocr * allocr = NULL;
     {
-        // alignment required by the backend
-        size_t align = ggml_backend_get_alignment(model.backend);
-        allocr = ggml_allocr_new_measure(align);
+        // create an allocator to measure the memory usage
+        allocr = ggml_allocr_new_measure_from_backend(model.backend);
 
         batch.n_tokens = n_batch_max;
 

diff --git a/tests/test-conv1d.cpp b/tests/test-conv1d.cpp
index a5a418d129a2ab43da98e026ac11b1adadb6f09f..98d45c5d0bed3cfd3a12b82cf7e064711b627747 100644 (file)
--- a/tests/test-conv1d.cpp
+++ b/tests/test-conv1d.cpp
@@ -204,8 +204,7 @@ int main(void)
     struct ggml_allocr * allocr = NULL;
 
     {
-        size_t align = ggml_backend_get_alignment(model.backend);
-        allocr = ggml_allocr_new_measure(align);
+        allocr = ggml_allocr_new_measure_from_backend(model.backend);
 
         //create the worst case graph for memory usage estimation
         struct ggml_cgraph * gf = build_graph(model, allocr);

diff --git a/tests/test-conv2d.cpp b/tests/test-conv2d.cpp
index df0e2ccdb7b49eb3e9b12aa4683c2b24eace8361..66e6e32f47a24ccf0a4c8907063932c37524d3ed 100644 (file)
--- a/tests/test-conv2d.cpp
+++ b/tests/test-conv2d.cpp
@@ -208,8 +208,7 @@ int main(void)
     struct ggml_allocr * allocr = NULL;
 
     {
-        size_t align = ggml_backend_get_alignment(model.backend);
-        allocr = ggml_allocr_new_measure(align);
+        allocr = ggml_allocr_new_measure_from_backend(model.backend);
 
         //create the worst case graph for memory usage estimation
         struct ggml_cgraph * gf = build_graph(model, allocr);

diff --git a/tests/test-mul-mat.cpp b/tests/test-mul-mat.cpp
index 36b7c6bcafa932db4ec21dd44e248dddd567f15a..d1406d4e755daeeced8939d1095b09c4e8f67eee 100644 (file)
--- a/tests/test-mul-mat.cpp
+++ b/tests/test-mul-mat.cpp
@@ -314,8 +314,7 @@ int main(void)
     struct ggml_allocr * allocr = NULL;
 
     {
-        size_t align = ggml_backend_get_alignment(model.backend);
-        allocr = ggml_allocr_new_measure(align);
+        allocr = ggml_allocr_new_measure_from_backend(model.backend);
 
         //create the worst case graph for memory usage estimation
         struct ggml_cgraph * gf = build_graph(model, allocr);