[GitHub] [tvm] areusch commented on a change in pull request #9395: [1/3][AOT][DeviceAPI] Connecting devices structure to relevant operators

[GitBox <gi...@apache.org>]

areusch commented on a change in pull request #9395:
URL: https://github.com/apache/tvm/pull/9395#discussion_r748561136



##########
File path: apps/microtvm/ethosu/src/tvm_ethosu_runtime.c
##########
@@ -0,0 +1,34 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied.  See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+#include "tvm_ethosu_runtime.h"
+
+#include <ethosu_driver.h>
+
+int32_t TVMEthosULaunch(struct ethosu_driver* driver, void* cms_data, size_t cms_data_size,
+                        uint64_t* base_addrs, size_t* base_addrs_size, int num_tensors) {
+  int32_t result =
+      ethosu_invoke(driver, cms_data, cms_data_size, base_addrs, base_addrs_size, num_tensors);
+
+  // Map errors in invoke to TVM errors
+  if (result != 0) {
+    return -1;

Review comment:
       we don't need to fix this in this PR, but i'm a big believer in not squashing error codes if at all possible. almost every mystery i've ever experienced in an embedded system has been the kind of problem where more data almost always is the most effective way to understand/solve the problem. we can squash for now but it'd be good to think about ways to expose driver error codes. that could be us allocating a new error code in error_codes.h and then stashing additional information somewhere else (perhaps that calls for yet another pointer to passed in? or perhaps the implementation should just be left to the driver and the driver struct should be expected to carry it).




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@tvm.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org