Quantcast
Channel: Tools
Viewing all 91832 articles
Browse latest View live

Forum Post: SPI issue with sdcard read/write operation in CCS v5

$
0
0

I am working with msp430 based custom board. It has sdcard support with spi interface.


Now my two Linux machine (Lets say Machine A and Machine B) has installed CCS v5. 

Now when i compile my code in Machine A then sdcard read/write operations are working fine.

But when i compile same code with Machine B then sdcard read/write operation does not work.

I debugged and found that all CCS and compiler and other xdctool version are same on both machine.

What i found is that:

In /opt/ti/ccsv5/ccs_base/msp430/msp430ware_1_40_01_44/driverlib/5xx_6xx/spi.h

Machine A has

//*****************************************************************************
//
//The following are values that can be passed to the SPI_masterInit() ,
//SPI_slaveInit() API as the clockPhase parameter.
//
//*****************************************************************************
#define SPI_PHASE_DATA_CHANGED_ONFIRST_CAPTURED_ON_NEXT    UCCKPH
#define SPI_PHASE_DATA_CAPTURED_ONFIRST_CHANGED_ON_NEXT    0x00
Machine B has
//*****************************************************************************
//
//The following are values that can be passed to the SPI_masterInit() ,
//SPI_slaveInit() API as the clockPhase parameter.
//
//*****************************************************************************
#define SPI_PHASE_DATA_CHANGED_ONFIRST_CAPTURED_ON_NEXT    0x00
#define SPI_PHASE_DATA_CAPTURED_ONFIRST_CHANGED_ON_NEXT    UCCKPH

Now when i change Machine B's files as Machine A then sdcard read/write works in Machine B too.

Whats the real solution? Why this different in both machine's default header files?


Forum Post: Voltage vs Power graph generation using GUI Composer.

$
0
0

Hi,

How can i generate Voltage vs Power graph using GUI composer. I taken ScalarLine graph from the GUI composer palette & assigned the values to the x(voltage) and y(power) axis, but voltage and power both are starts at y-axis only,not with respect to x-axis.  Insted of ScalarLine graph, i used Linegraph but there is no variation in the graph.

please follow below images. i need to generate the graph as shown in below(image-1) 

 

Please ping me as soon as possible..

 

image -1


my problems.

image-2  (line graph)

image -3(line graph)

image -4(scalarline graph)

Forum Post: How to detect Host os is Linux or windows in c code of CCS?

$
0
0

In Normal c program below code works fine. If i compile this in Linux machine then it will print linux and if i compile it in windows then it will print windows.

How can i get same thing in C code going to compile in CCS on windows and CCS on Linux.

#ifdef linux printf("its linux"); #else printf("its windows"); #endif 

Forum Post: RE: unable to run a program in ccs6

$
0
0

hi i tried running a small program with some change from the Digitial Signal processing and Applications with TMS3206713 Rulph Chassaing shown below. I included the required files and lib files bot got an error:  #6001 cannot open file..what might be wrong?

#include<stdio.h>
#include<math.h>
#include "DSK6713_AIC23.h"
Uint32 fs=DSK6713_AIC23_FREQ_8KHZ;
short output_sample;
int i;
interrupt void c_int11()
{
for(i=0;i<=100;i++)
{
output_sample=sin(2*3.141*0.01*i);
}
}
void main()
{
comm_intr(); //init DSK, codec, McBSP
while(1); //infinite loop
}

Forum Post: "DWARF information is corrupted" when loading a uboot Image

$
0
0

I build a uboot image by a linaro compiler, and try to load this image with CCS5 but it doesn't work. and I also try to add symbol, but it is not successful neither.

the message is below:

CortexA15_1: GEL: Encountered a problem loading file: X:\code\uboot\k2l\u-boot\u-boot DWARF information is corrupted.

my environment:

compiler:gcc-linaro-armeb-linux-gnueabihf-4.9-2014.06_linux

CCS Version: CCS5.5.0

by the way, another linario compiler is ok,  "gcc-linaro-arm-linux-gnueabi-2012.03-20120326_linux", the image can be loaded successfully, but I need this new compiler to build big endian code. :(

 

Forum Post: RE: error #10099-D run placement fails for object

$
0
0

Thank you for the answer, this is the text of the linker diagnostic:

<Linking>

"./aee0000cfg.cmd", line 367: error #10099-D: run placement fails for object

   ".ebss", size 0x12ce9 (page 1).  Available ranges:

   L47SARAM     size: 0x6000       unused: 0x51bb       max hole: 0x51b8   

error #10010: errors encountered during linking; "ee0000__.out" not built

 

>> Compilation failure

gmake: *** [ee0000__.out] Error 1

gmake: Target `main-build' not remade because of errors.

 

**** Build Finished ****

I have only add a file with #defines and before this, I had much unused space in RAM.


Thanks,

Nerea

Forum Post: GUI Composer JS Error

$
0
0

I have just tried the Blinky tutorial on my shiny new ccsv6 download. I get to the point where I click on the app.js and I get a Windows Script Host error. Line: 5, Char: 1, Error: Object expected, Code: 800A138F, Source: Microsoft JScript runtime error.

Anyone know of a reason why this is so?

Forum Post: RE: TI ARM Compiler vs. Linaro GCC - Code size

$
0
0

This woulb be true, if it would be the size of files, but it isn't. It is the size of the flash usage. Adding all .text parts and .cint parts together. Both values also includes the .intvecs part as well as initialized RAM.

Looking in the map files, you will find nearby no memory holes.

There is another thing that is not correct: Both files are stored in the ELF-File format. This format stores offset and length of each entry. It is not a flash map where holes are just zero! Please have a look here: Wikipedia: ELF

This is even not true for the most other binaries as well as hex files! As posted from Archaeologist and me. There are serval components coming with gcc linked that are not linked with the TI-Compiler.

Back to topic:

Seeing the main program in with TI-Compiler: main.obj and the GCC compiler: main.o both come with 0x2 bytes. Means: The GCC will link more but it doesn't say actually that one of the compilers are better or worse then the other. The best way to compare compilers is to use standard algorithm like FFT or Multiply of long long data types with HW-Support or something simliar and check only the object instead of the whole image.

the additional 2 kB of Flash usage will be present in every binary linked with GCC. There are options to omit extra functions or use own instead. Please have a look at the complier documentation, if you are interested.

Best regards,

Frank


Forum Post: LM3478

$
0
0

Hi,

Simulating the LM3478 I get a warning...

With the low turn of voltage of the LM34x8 your power supply may current limit before you reach your working input voltage.... consider adding a low pass filter at the input.. Can anyone give a fuller explanation of what the problem is ? Also the grammar of 'turn of voltage' is not quite correct, but that's a minor issue !

Best Regards

Forum Post: can't build

$
0
0

getting this error :

Buildfile generation error occurred..
This project was created for a device-variant that is not currently recognized: TMS320C64XX.TMS320C6416.DSK6416. Please install the device-variant descriptor, or migrate the project to one of the supported device-variants by adjusting project properties.
Build stopped..

Thanks,

Forum Post: webench export error when I choose designspark pcb file type

$
0
0

Dears, I just follow the default setting and select export CAD file, I want to get the schmatic with DesignSpark PCB type, but the error message shows'' export process did not complete under 20 secs. ''  how to slove it.  (Please visit the site to view this file)

Forum Post: RE: LM5118 - can't simulate at particular condition.

$
0
0

Hi Kevin-san,

 

Thank you for your prompt reply.

 

I can simulate at Vin=9.0V to 19V , Vout=6.8V , Iout=5.0A (Buck mode).

(please refer to attached file.)

What is reason that the parameter at Vin=9.0V to 19V , Vout=6.0V , Iout=5.0A isn't simulated ?

 

[your question]

What is the advantage for a Buck only design?

 

=>My customer would like to use LM5118 with changing the output voltage (6.0V to 15V).

    So, they have simulated at the following parameters.

    - Vin=9V to 19V , Vout=6V , Iout=5A : Sim NG

    - Vin=9V to 19V , Vout=9V , Iout=5A : Sim OK

    - Vin=9V to 19V , Vout=15V . Iout=5A :Sim OK

 

Regards,

Kanemaru

Forum Post: RE: How to avoid hole reusing again

$
0
0

Hi George,

       When DSPs boot image from flash, .switch.2 is loaded first ,then the hole filled with 0s at the same address of .switch.2.That is ,the hole cover the .switch.2.How to resolve this case?

Xiang

Forum Post: RE: Launching Debug Session Error at CCSv6

$
0
0

Hi and thanks Ki-Soo.

How silly of me :D. It's my first time using C2000 with CCSv6.

Thanks for your help.

Regards, Min-Ku.

Forum Post: RE: Code Generation BUG found in 7.4.4 and 7.4.8!

$
0
0

Hi BZ,

Moved this thread to correct forum for faster response. Thank you for your patience.


Forum Post: CS questions

$
0
0

Got a few questions for you about the Code Composer software:

  1. is there a demo version for evaluation?
  2. Can you give me the highlights of the Export Admin Regulation requirement?
  3. if we buy a node locked license, is the EAR required? Kinda hard to export a lic attached to a single PC.
  4. Does the floating lic come with a USB key/dongle? 
  5. Is there a program size limit for the license?

Thanks for any help you can offer.

Forum Post: RE: NOR Flash debugging using CCS

$
0
0

Could you please help me on this and share me documents how to debug the NOR flash using Code composer studio and XDS100v2 for AM3356.

Forum Post: RE: Compile/Link time initialization of union

$
0
0

[quote user="Jamie McPeek"]

The compiler in use is CL2000 for TMS320F28335 version 6.1.0.

I can't change the compile flags/pull in non-standard extensions.

I'll take a look at flipping the order of the member declarations in the union.

If that can't be touched for various reasons, is there a way to initialize the bit field directly without using explicit initialization?

[/quote]

In this case, I suppose the only option is a tedious field-per-field initialization list: The compiler should merge it in only one word, compilation time calculated initialization value:

GlobalBlock = {   { first_bitfiled_value, second_bitfield_value, ....},  //A   { ....}, //B   ... }; 

You can use a macro to build the initialization list, but this works only if all the fields are 1 btis size.

Forum Post: RE: AM335x starterware debug issue

Forum Post: Whether CCS6 support graph option for Constellations or Eye Diagrams ?

Viewing all 91832 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>