[build] Remove test backend codes

- Remove dummy test code.

Signed-off-by: Yongjoo Ahn <yongjoo1.ahn@samsung.com>

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 0f620020299924676e12c637661813c43f832ed9..ab9562f7181ffd16d0d84040377a92d389a7032e 100644 (file)
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -26,16 +26,6 @@ SET(UTIL_SRCS
   ${PROJECT_SOURCE_DIR}/src/hal-backend-ml-util.cc
 )
 
-# pass-through
-SET(PASS_SRCS
-  ${PROJECT_SOURCE_DIR}/src/hal-backend-ml-test-passthrough.cc
-)
-
-SET(PASS_LIBRARY_NAME "hal-backend-ml-pass")
-ADD_LIBRARY(${PASS_LIBRARY_NAME} SHARED ${PASS_SRCS} ${UTIL_SRCS})
-TARGET_LINK_LIBRARIES(${PASS_LIBRARY_NAME} ${hal_rootstrap_pkgs_LDFLAGS})
-INSTALL(TARGETS ${PASS_LIBRARY_NAME} DESTINATION ${HAL_LIBDIR} COMPONENT RuntimeLibraries)
-
 # vivante
 IF(ENABLE_VIVANTE)
 SET(VIVANTE_SRCS

diff --git a/packaging/hal-backend-ml.spec b/packaging/hal-backend-ml.spec
index 210ffa26739af5c9714969e62d967dbfa157fbb1..db5d868bfd631092a9a71d0dd43f60cf7b56ca7f 100644 (file)
--- a/packaging/hal-backend-ml.spec
+++ b/packaging/hal-backend-ml.spec
@@ -73,11 +73,6 @@ make %{?_smp_mflags}
 %postun
 /sbin/ldconfig
 
-%files
-%manifest packaging/hal-backend-ml.manifest
-%license LICENSE
-%{_hal_libdir}/libhal-backend-ml-*.so*
-
 %if 0%{?vivante_support}
 %files vivante
 %manifest packaging/hal-backend-ml.manifest

diff --git a/src/hal-backend-ml-test-passthrough.cc b/src/hal-backend-ml-test-passthrough.cc
deleted file mode 100644 (file)
index 2e37605..0000000
--- a/src/hal-backend-ml-test-passthrough.cc
+++ /dev/null
@@ -1,124 +0,0 @@
-/* SPDX-License-Identifier: Apache-2.0 */
-
-#include <stdexcept>
-
-#include <glib.h>
-
-#include <hal-common-interface.h>
-#include <hal-ml-interface.h>
-
-#include "hal-backend-ml-util.h"
-
-
-typedef struct _pass_handle_s
-{
-  GstTensorsInfo inputInfo;
-  GstTensorsInfo outputInfo;
-} pass_handle_s;
-
-static int ml_pass_init(void **backend_private)
-{
-  pass_handle_s *pass = g_new0 (pass_handle_s, 1);
-  gst_tensors_info_init (&pass->inputInfo);
-  gst_tensors_info_init (&pass->outputInfo);
-  *backend_private = pass;
-  return 0;
-}
-
-static int ml_pass_deinit(void *backend_private)
-{
-  pass_handle_s *pass = (pass_handle_s *) backend_private;
-  if (!pass) {
-    g_critical ("[pass backend] ml_pass_deinit called with invalid backend_private");
-    return HAL_ML_ERROR_INVALID_PARAMETER;
-  }
-
-  gst_tensors_info_free (&pass->inputInfo);
-  gst_tensors_info_free (&pass->outputInfo);
-
-  g_free (pass);
-
-  return HAL_ML_ERROR_NONE;
-}
-
-static int ml_pass_configure_instance(void *backend_private, const GstTensorFilterProperties *prop)
-{
-  pass_handle_s *pass = (pass_handle_s *) backend_private;
-  if (!pass) {
-    g_critical ("[pass backend] ml_pass_configure_instance called with invalid backend_private");
-    return HAL_ML_ERROR_INVALID_PARAMETER;
-  }
-
-
-  gst_tensors_info_copy (&pass->inputInfo, &prop->input_meta);
-  gst_tensors_info_copy (&pass->outputInfo, &prop->output_meta);
-
-  return 0;
-}
-
-static int ml_pass_invoke(void *backend_private, const GstTensorMemory *input, GstTensorMemory *output)
-{
-  pass_handle_s *pass = (pass_handle_s *) backend_private;
-  if (!pass) {
-    g_critical ("[pass backend] ml_pass_invoke called with invalid backend_private");
-    return HAL_ML_ERROR_INVALID_PARAMETER;
-  }
-  g_warning ("skip rungraph.. 됐다 치고 하하"); g_usleep (1000 * 100);
-  // pass->result_vsi_nn_RunGraph (pass->graph);
-  return 0;
-}
-
-static int ml_pass_get_model_info(void *backend_private, model_info_ops ops, GstTensorsInfo *in_info, GstTensorsInfo *out_info)
-{
-  pass_handle_s *pass = (pass_handle_s *) backend_private;
-  if (!pass) {
-    g_critical ("[pass backend] ml_pass_get_model_info called with invalid backend_private");
-    return HAL_ML_ERROR_INVALID_PARAMETER;
-  }
-
-  gst_tensors_info_copy (in_info, &pass->inputInfo);
-  gst_tensors_info_copy (out_info, &pass->outputInfo);
-
-  return 0;
-}
-
-static int ml_pass_event_handler(void *backend_private, event_ops ops, GstTensorFilterFrameworkEventData *data)
-{
-  return -ENOENT;
-}
-
-static int ml_pass_hal_backend_init(void **data)
-{
-  hal_backend_ml_funcs *funcs = NULL;
-
-  if (*data) {
-    funcs = (hal_backend_ml_funcs *) *data;
-  } else {
-    funcs = g_new0 (hal_backend_ml_funcs, 1);
-  }
-  *data = (void *) funcs;
-
-  funcs->init = ml_pass_init;
-  funcs->deinit = ml_pass_deinit;
-  funcs->configure_instance = ml_pass_configure_instance;
-  funcs->invoke = ml_pass_invoke;
-  funcs->get_model_info = ml_pass_get_model_info;
-  funcs->event_handler = ml_pass_event_handler;
-
-  return 0;
-}
-
-static int ml_pass_hal_backend_exit(void *data)
-{
-  memset (data, 0x0, sizeof(hal_backend_ml_funcs));
-  return 0;
-}
-
-hal_backend hal_backend_ml_data = {
-  .name = "ml-pass",
-  .vendor = "YONGJOO",
-  .init = ml_pass_hal_backend_init,
-  .exit = ml_pass_hal_backend_exit,
-  .major_version = 1,
-  .minor_version = 1,
-};