@sysprogs this is great! I followed your instructions to build the gpio_led demo and it worked perfectly. I have not used Visual Studio prior to this so this is new to me. I followed the Kendryte tutorials for building a project which calls for having projects created under .\Kendryte-Standalone-SDK\src so the project source files are located within the SDK source tree. My understanding is that the normal/easy way of creating an embedded project with Visual Studio/GDB is to place the project source files in one location and reference the SDK dependencies in another location (done automagically by VS) so the Kendryte recommendation for setting up the project is at odds with the standard VS method. Your tutorial follows the VS approach. I created a project structure that I want to preserve as it will allows me to track changes to the orginal Kendryte SDK while making changes to it that are necessary for my project. To do this I copied the entire Kendryte-Standalone-SDK tree to a new location and renamed it to PROJECT_MCU and I have my project files (and some of the Kendryte demos) under src as recommended by Kendryte. Structured as follows: I figured out how to import an existing project into VS by following the tutorial >>> here <<< and the project builds fine with the output binary showing up where it should. When first run VS/GDB did not find the SDK source & header files but I was able to figure out how to include the folders to them so I got by that problem. However, I can't get by the following error even though it seems like an easy one to fix... No executable file specified. Use the "file" or "exec-file" command. I did specify the built binary as my output file and I tried changing settings under menu "Debug/{project} properties". I also went through all of the options under menu "Project/VisualGDB Project Properties" but nothing stood out. I really don't know what could be wrong. Can you provide a suggestion for resolving this? Thank you (in advance)

也在研究IAP，我认为应该同时完成2个应用开发，唯一不同的是link的.ld文件必须修改为不同的FLASH地址，能否试试看？

Can you try with the newest nncase and SDK?

Of course you can.

@Asterion I see a lot of success stories about Kendryte. I am very happy about it for the company. Unfortunately, in my situation with Maixduino and AI HAT, I cannot get even a Hello World exercise to complete successfully. Nobody at Kendryte will reply to my inquiries and the two board manufacturers, Sipeed and Seeed, have categorically stated that the issue is for Kendryte to solve. I just don't know who to believe. Why can't I reproduce two simple Hello World exercises? The answers are: My (operator) error but I don't know what/where? That is why I am seeking help! Kendrte doesn't have to reply to me. Could this be true? Kind regards.

@latyas sorry, this topic was not finished. I reopen a new topic with the questions, could you help me?

@kevinfeng83 标准 SPI 协议。

确定你的JLINK设备是正常工作的吗

Hello @kwonsk, Everything works well now. I guess my mistake was to start the serial terminal after openocd and gdb. Thank you very much!

@loboris thank you very much! i assume the cause is dma need more then 10cycles to complete a transaction! the evidences are: 1.when receive less bytes than 32 ,the transfer is ok,the data are stored in fifo. 2.when transfer more bytes at 40Mbits/s,the spi->risr regitster bit3 will be set,where norm operation will not set,but i have no register definition. i think the fifo is overrun,dma will loss some bytes. 3.when set the DATALENGTH to 32bits, transfer 4bytes at a dma transaction, no problem appear. this will decrease the transaction times by factor 4! 4.this assumption also explain when raise spi rate,the quad io mode fail first,then then daul fail. the code will run endless in waiting dma done . am i right???

@nathan it seems that one of my interemediate feature maps is larger than 2MB and that it was the source of the problem. I reduced the size of the network and now it is transformed to kmodel correctly.

@loboris thanks!