IWR1443: running and rebuilding the high accuracy binaries

Hi Al,

For high accuracy 14xx pre-built binary issue you had, when you say appropriate gui, did you mean the high accuracy lab visualizer as indicated in the high accuracy lab user guide?

When you re-build the lab binaries, no need to edit the setenv.sh file or run script in cmd window. Please use ccs and import the project and build. Let me know if you still have errors.

Thanks and Regards,
Michelle

Hi Michelle,

Thank you for responding.

Yes, what I mean by the appropriate gui is the high accuracy lab visualizer gui.  If the oob, demo visualizer gui is used, errors are generated.  That is how I confirmed that the demo mss bin file was not loaded (also no errors were generated during the loading process).  I then reloaded the demo mss bin file and ran the demo visualizer gui as a sanity check.  By the way, when I say I loaded the mss file, I also loaded the bss file too.

So I can  you please suggest  how I can debug my issue?  When the oob demo is loaded,  I can connect with it's gui but I can not connect between the high accuracy binary with it's gui. 

The user manual for the sdk says that the setenv.sh should be edited and run.  I read an older posting with the type of errors I was getting and they were told to follow the manual.  Anyway, I did initially use ccs to import the project.  I did a clean project and then I did a rebuild and got the errors.  What's next?

Al

Hi Al,

I can't explain why your pre-built binaries didn't work. Could you try flashing one more time? using both mss and bss binaries from the prebuilt_binaries folder.

Just to double check, can you take a look at the high_accuracy_14xx_lab_releasenotes.pdf in your folder and let me know what version of high accuracy lab you got?

Thanks and Regards,
Michelle

Hi Michelle,

A number of items.  I should mention that I am running offline on a linux platform.

First:

I removed my copy of ccs v7.4 and the industrial toolbox for mmwave sensors and reloaded them.  I got further compiling this time.  I can make the mss.xer4f binary but not the mss.bin for the high accuracy lab.  It appears that it is trying to run a windows exe and I am on a linux platform as shown in the makefile output.

/home/xxx/ti/mmwave_sdk_01_02_00_05/packages/scripts/ImageCreator/xwr14xx/out2rprc/out2rprc.exe xwr14xx_high_accuracy_mss.xer4f xwr14xx_high_accuracy_mss.bin 0x200000
makefile:194: recipe for target 'post-build' failed
/home/xxx/ti/mmwave_sdk_01_02_00_05/packages/scripts/ImageCreator/xwr14xx/out2rprc/out2rprc.exe: /home/xxx/ti/mmwave_sdk_01_02_00_05/packages/scripts/ImageCreator/xwr14xx/out2rprc/out2rprc.exe: cannot execute binary file
gmake[3]: [post-build] Error 126 (ignored)
/home/xxx/ti/mmwave_sdk_01_02_00_05/packages/scripts/ImageCreator/append_bin_crc/gen_bincrc32.exe: /home/xxx/ti/mmwave_sdk_01_02_00_05/packages/scripts/ImageCreator/append_bin_crc/gen_bincrc32.exe: cannot execute binary file
gmake[3]: [post-build] Error 126 (ignored)
/home/xxx/ti/mmwave_sdk_01_02_00_05/packages/scripts/ImageCreator/append_bin_crc/gen_bincrc32.exe xwr14xx_high_accuracy_mss.bin
makefile:194: recipe for target 'post-build' failed

The exe's are located in

25 xxx@localhost:~/ti/mmwave_sdk_01_02_00_05/packages/scripts/ImageCreator/xwr14xx/out2rprc> dir
total 68
drwxrwxr-x. 2 xxx xxx  4096 Aug 12 22:09 ./
drwxrwxr-x. 3 xxx xxx  4096 Jul 31 00:01 ../
-rwxrwxr-x. 1 xxx xxx 59392 Mar  7 21:34 out2rprc.exe*

28 xxx@localhost:~/ti/mmwave_sdk_01_02_00_05/packages/scripts/ImageCreator/append_bin_crc> dir
total 76
drwxrwxr-x. 2 xxx xxx  4096 Jul 31 00:01 ./
drwxrwxr-x. 5 xxx xxx  4096 Jul 31 00:01 ../
-rwxrwxr-x. 1 xxx xxx 13441 Mar  7 21:34 gen_bincrc32*
-rwxrwxr-x. 1 xxx xxx 52645 Mar  7 21:34 gen_bincrc32.exe*

I had selected all linux software when possible.  I see that there is no linux binary for out2rprc.exe but there is for gen_bincrc32.exe.  How to I obtain the linux binary and configure the system for building on a linux platform?

Second:

I reloaded the eval part again with the prebuilt libraries.  I had tried a number of time before but with not success.  Two comments: 1. I had a very difficult time getting uniflash to connect to the part.  It finally did and it loaded.  I then ran the high accuracy gui.  Again I had difficulty in connecting to the part via usb.  (I have had no problems connecting when I was using the oob demo.)  Attached are some screen images illustrating the connection problem.  One idicates that the link is specified but clicking on it indicates that the link is not made and finally, link is made.  Once connected I tried to load the provided config files (profile_2db.txt and profile_2de.txt) but observed no graphical output.  I was able to generate a cvs file but it had no data as expected. Please advise me on the connection issue and lack output.  The oob demo has a console providing feedback that the config file loaded but the high accuracy gui does not.

Third:

I tried to run the fluid level transmitter with the current load in the part (based upon the user guide) and I can not run the gui.  It appears that not all of the software is in the zip file. The desktop launcher is trying to run a file call linux_start.sh but the contents of directory is limited compared to the high frequency directory.  The contents I have for the fluid level transmitter is

 xxx@localhost:~/guicomposer/runtime/gcruntime.v5/mmWave_Fluid_Level_Transmitter_1_0_0> dir

total 16
drwxrwxrwx. 4 xxx xxx 4096 Aug 14 14:01 ./
drwxrwxrwx. 6 xxx xxx 4096 Aug 14 14:01 ../
drwxrwxrwx. 5 xxx xxx 4096 Aug 14 14:01 components/
drwxrwxrwx. 2 xxx xxx 4096 Aug 14 14:01 installer/

Please advise me on these there issues.  Thanks.

Al

Hi Allen,

For the compilation under linux, out2rprc can not be executed directly in linux. We suggest using mono -- a windows compatible layer in linux.

For the issue with pre-built binaries, one thing I can think of that is different for high accuracy lab and mmw demo, the port sequence on the Serial Port Configuration page is different. For high accuracy, Data_port on the left and CFG_port on the right. For demo, it's reversed. Is it possible that the wrong baud rate was assigned? Data port baud rate should be 921600 and CFG_port baud rate is 115200.

The fluid level transmitter is based on the high accuracy 14xx lab. Only visualizer is different. It doesn't contain source code.

Thanks and Regards,
Michelle

Hi Michelle,

Thanks for the feed back.

1. I will see if I can address the exe requirement.

2. Your comment regarding the swapping of the port placement is probably correct.  I should learn how to read.

3. I realize that there is no source code for the fluid level lab and that the high accuracy code is used.  My issue is that the code for the gui is not complete in the zip file.  The desktop launcher attempts to run a shell script call linux_start.sh which is not present as well as other corresponding files that the high accuracy gui directory has.  The run file I used to create the fluid level transmitter directory and content was mmWave_Fluid_Level_Transmitter_1_0_0-1.0.0.setup-linux-x64_5.0.0.run.  It is not providing all the content that the high accuracy visualizer run does.  I have to locate the url for the gui so I can load load it again and reinstall it but I can not find it.  Perhaps you could provide me with the url.

Thanks.

Al

Hi Al,

For the fluid level transmitter, only User Guide and Release Notes are available from the toolbox. I'm not aware of the gui zip file. According to the user's guide, you could access directly to dev.ti.com/.../mm
Wave_Fluid_Level_Transmitter_1_0_0/

Is this question about installing visualizer on linux ?

Thanks and Regards,
Michelle

HI Michelle,

Interesting.  Shown below is the  zip, run and log files for the fluid level transmitter and an image of the desktop launcher that the run file created.

-rw-rw-rw-.  1 xxx xxx     13408 Aug 14 19:55 mmWave_Fluid_Level_Transmitter_1_0_0_1.0.0.20180814195535.log
-rw-rw-r--.  1 xxx xxx   7019841 Aug  6 18:52 mmWave_Fluid_Level_Transmitter_1_0_0_1.0.0_installer_linux.zip
-rwxrwxrwx.  1 xxx xxx   7242656 Jan 10  2018 mmWave_Fluid_Level_Transmitter_1_0_0-1.0.0.setup-linux-x64_5.0.0.run

I can not find this zip file anymore on your site.  The run file creates a similar directory structure as the high accuracy visualizer run file does but it is not fully populated as the high accuracy visualizer directory.  I have attached an image of the launcher too.  The launcher tries to run a sh script which does not exist.

Al

Al,

Our expert on fluid level transmitter is not available till next week. My guess is that the zip file was replaced by on-line visualizer therefore it's taken out. Could you follow the current user guide and see if you can run the lab?

Thanks!
Regards,
Michelle

Al,

Let me know if you still have any question.

Thanks and Regards,
Michelle

Hi Michelle,

1. I resolved my compiling issue of running a window's exe in linux

2. I still have an issue of connecting the gui to the part.  I have a D3 development board and I have to reset it many times (reset button on the development board) for it to finally connect.  There must me something not quite right.  Any thoughts?

3.  I got the high accuracy gui and binary going. Attached is a pdf showing the range.  It appears that it only has a range of 5 meters but the two supplied configuration files indicate a much larger range (18 and 33 m, I believe).  Your thoughts?

4. The fluid level transmitter gui is still on the ti web site in the gallery selection.  It builds an incomplete software download.

5. I would like to use the ccs debug feature.  Does one flash the xer4f and dss image on the part?  It appears that there is another firmware debug image to allow the ccs to debug.  This image is suppose to be located in the sdk directory but I do not know what to look for and what the flashing procedure is.  Can you please refer me to the appropriaterange_image.pdf documentation?

Thanks for the help.

Al

Hi Allen,

For the future, please issue new tickets if you have more questions in addition to the original problems reported. The support team member has different focus, your question will be solved faster that way. Issue is also more organized and easy to track.

Q2: D3 module, could you briefly describe what D3 module is? Is the FTDI or emupack software installed for the ports?
Q3: I'm not sure why. I need to run it myself and get back to you. In the meantime, could you check the max range value setting in the visualizer?
Q4: I'll get the expert on fluid level transmitter to take care of it.
Q5: The high accuracy lab user guide and mmwave sdk user guide both describes on how to use ccsdebug. I recommend the former, page 30 to 39.

Regards,
Michelle