- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Tool/software: Code Composer Studio
Hello!
I am using the ZNP-HOST_FRAMEWORK project "cmdLine" with the TM4C1294XL, CCEM Booster Pack and the CC2538. I am trying to add some code I previously wrote in Visual Studios (in C++) to this project. I have verified that my code works in Visual Studio. I'm not getting any errors or warnings about my new code in the Code Composer Studio project "cmdLine". However, my code isn't working properly in Code Composer Studio. Basically, the code I wrote performs trilateration using pre-defined values (NOT values obtained from the microcontroller. I want to make sure it works with theoretical values first before using real-time values from the microcontroller). My code uses variable types "double" and "int", and I noticed the Code Composer Studio Code uses mainly "uintX_t" and "intX_t", where X is 8, 16, 32, or 64.
I'm not sure why my code isn't working in Code Composer Studio, but I'm suspecting it may be because Code Composer Studio doesn't recognize the data type "double"? How is the double data interpreted? Do I need to convert the data to another format? I should mention that I'm using RSSI values, so there are negatives and decimal values that I need to keep--I cannot truncate or round the numbers. Could something else be the issue?
I've attached the modified file "cmdLine.cpp", which consists of all of my code. In this file I've made it clear as to what lines I've added by writing my name in the variable name and the comments.
Thank you!
- Tessa
Hi,
I have no idea what the cmdLine.c is about? It is the first time I'm seeing this file. Not sure if I'm able to give any help. What is the target device you are running this code on? Is it the TM4C129 or the CC2538?
First of all, if you are running the code on the TM4C129 then bear in mind that the M4F processor only support single-precision floating point. See below excerpt.
— 32-bit instructions for single-precision (C float) data-processing operations.
— Combined Multiply and Accumulate instructions for increased precision (Fused MAC).
— Hardware support for conversion, addition, subtraction, multiplication with optional accumulate,
division, and square-root.
— Hardware support for denormals and all IEEE rounding modes.
— 32 dedicated 32-bit single-precision registers, also addressable as 16 double-word registers.
— Decoupled three stage pipeline.
Secondly, make sure you enable the hardware floating point unit.
Hello, Charles!
I'm using the cmdLine.cpp file from the ZNP_HOST_FRAMEWORK ( https://www.ti.com/tool/TIDC-ZNP-HOST-SW3 ) Code Composer Studio project cmdLine. I'm currently running the code on an TM4C1294XL. I should mention I do have a booster pack connected to the TM4C1294 board with a CC2538.
Precision to one decimal point is fine. It looks like my options are set up the same as yours. I've attached a picture just in case. I'm still not getting even close to the same results as I was getting in Visual Studio when I run this same code. Do you think this could be an issue with the double to int32_t or double to uint32_t conversions? I should mention that I'm implementing code with negative numbers. I'm not sure if negative numbers could change anything (like maybe it's using twos compliment or something? Or do you think this issue could be something else?
Thank you!
- Tessa
Hi,
Can you show your test code how you print out the above values? Please note that the floating values are stored in the computer memory using the IEEE-754 representation. Here is one online tool that converts the floating value to its IEEE-754 format stored in the memory.
https://www.h-schmidt.net/FloatConverter/IEEE754.html
You can also watch the variable in the CCS under the Expression Window. Here I create a float variable called f_SP_Temperature2 and I assign a value of -5.5 to it. In the Expression window I see 0xC0B00000 stored in the memory which is the same as the online tool.
Hi,
Can you show your test code how you print out the above values? Please note that the floating values are stored in the computer memory using the IEEE-754 representation. Here is one online tool that converts the floating value to its IEEE-754 format stored in the memory.
https://www.h-schmidt.net/FloatConverter/IEEE754.html
You can also watch the variable in the CCS under the Expression Window. Here I create a float variable called f_SP_Temperature2 and I assign a value of -5.5 to it. In the Expression window I see 0xC0B00000 stored in the memory which is the same as the online tool.
Hi,
I use your code like below. When I compile I will see the below five warnings. You should see the same. There is incompatibility during format string conversion. Note I use printf. 
When I run the code, I can produce the same outputs as you. But when you compare to what is stored in the memory they are different. As explained, I think the reason is the format string conversion. If you look at the Expression windows for the variable tessa_test_float, it is correctly stored in the memory as 0x41080000 and this is matching the online converter tool while it is printed out as 0x40210000 on the console. Again, what is most important is if it is stored in the memory correctly. I'm not sure why you must print them with the hex format of %02X for a floating value which results in the format string conversion incompatibility. You might want to try typecasting. Again, this is beyond the scope of the MCU. Why don't you just print it with something like %02f in which case you will see 8.5 or -8.5.
Hello, Charles!
I don't really need to print it to the terminal, I just wanted a way to verify that the results I was getting is correct. Using the information you provided me, I confirmed that my tessa_test variables are producing the right values in the Expression Window, but the printing is still wrong (again, I don't really need to print). I did notice however that the variables tessa_unknown_node_location and tessa_lqi_array are still set to their initial values (which are zero), even though they are printing values that are not zero.
Thank you!
- Tessa
That is not what I see when I use what you have.
I just added
float tessa_unknown_node_location[3] = { 0, 0, 0 }; //ADDED BY TESSA
int32_t tessa_test_int32_t = 8.5;
uint32_t tessa_test_uint32_t = 8.5;
float tessa_test_float = 8.5;
double tessa_test_double = 8.5;
int32_t tessa_test_int32_t_negative = -8.5;
uint32_t tessa_test_uint32_t_negative = -8.5;
float tessa_test_float_negative = -8.5;
double tessa_test_double_negative = -8.5;
int
main(void)
{
//
// Run from the PLL at 120 MHz.
//
g_ui32SysClock = MAP_SysCtlClockFreqSet((SYSCTL_XTAL_25MHZ |
SYSCTL_OSC_MAIN | SYSCTL_USE_PLL |
SYSCTL_CFG_VCO_480), 120000000);
printf("\t Tessa unknown_node_location[0]: 0x%02X\n", tessa_unknown_node_location[0]);
printf("\t Tessa unknown_node_location[1]: 0x%02X\n", tessa_unknown_node_location[1]);
printf("\t Tessa unknown_node_location[2]: 0x%02X\n", tessa_unknown_node_location[2]);
printf("\t Tessa test int32: 0x%02X\n", tessa_test_int32_t);
printf("\t Tessa test uint32: 0x%02X\n", tessa_test_uint32_t);
printf("\t Tessa test float: 0x%02X\n", tessa_test_float);
printf("\t Tessa test double: 0x%02X\n", tessa_test_double);
printf("\t Tessa test int32 negative: 0x%02X\n", tessa_test_int32_t_negative);
printf("\t Tessa test uint32 negative: 0x%02X\n", tessa_test_uint32_t_negative);
printf("\t Tessa test float negative: 0x%02X\n", tessa_test_float_negative);
printf("\t Tessa test double negative: 0x%02X\n", tessa_test_double_negative);
And I see below. Try for yourself. One thing I will recommend is run your test code in a small project and not embed it in a 4000 lines file. This way it is easier to debug.
