Quantcast
Channel: Forums - Recent Threads
Viewing all 262198 articles
Browse latest View live

Compiler/MSP430F5527: 10099-D: program will not fit into available -- how to prioritize sections placement?

$
0
0

Part Number: MSP430F5527

Tool/software: TI C/C++ Compiler

I have a project which I am porting from CCS5.4 (compiler 4.1.5) to CCS9.2.  I've already a fixed a few issues related to this move, but this one has me stumped.  My command file defines two flash segments corresponding to the 5527 flash.

FLASH  : origin = 0x4400, length = 0xBB80
FLASH2 : origin = 0x10000,length = 0xC400

The sections .text and .cinit are allocated to these segments as follows:

.cinit : {} > FLASH
.text  : {}>> FLASH | FLASH2

The problem is that I've reached the point where the .text has filled up the FLASH segment and started to use the FLASH2 segment.  Because the FLASH segment is full, the linker fails to allocate space for .cinit section.  How do I tell the linker to prioritize placing the .cinit section in FLASH and move .text to FLASH2 as needed?  There is plenty of room in FLASH2.


Compiler: DLP® DiscoveryTM 4100 Development Platform, DMD, Discovery D4100 Explorer

$
0
0

Tool/software: TI C/C++ Compiler

As shown in the screenshot1, the USB 2.0 is good while no DMD is detected. (Discovery D4100 Explorer)

I also tried GetDevNum(0) in C++, and it returned 1, which means USB did connected, but I use GetDMDTYPE(0) command in code, the return number is 15! That means no DMD is detected. Shown in screenshot2.

From the DLP® DiscoveryTM 4100 Development Platform, the controller board, I tried some simple operations. It works, so there is no connection and technical problem between DMD and DLP® DiscoveryTM 4100 Development Platform. 

TM4C1294NCPDT: ROM Bootloader on TIVA TM4C1294NCPDT controller over sflash does not work

$
0
0

Part Number: TM4C1294NCPDT

I am using TIVA TM4C1294NCPDT controller on a custom board. I have UART0 connected to UART of an embedded linux board. During normal OS operation, I am able to invoke the ROM bootloader by calling ROM_UpdateUART() to update firmware using sflash utility. This method works fine as long as I have some firmware already on the device. 

I want to program virgin devices using sflash. To test this, I erased the flash and confirmed over JTAG that 0x0 and 0x4 locations read 0xFFFF_FFFF and that BOOTCFG register is 0xFFFF_FFFE. On a blank device, I expect the device to stay in ROM bootloader looking for packets over UART. However, when I run sflash, I get the 'erasing' chip message and nothing after that. 

I have also noticed that when the bootloader is invoked using ROM_UpdateUART(), if the firmware update is interrupted and restarted, the device resumes the firmware update and continues successfully. However, if after interrupting, power is cycled, then the update does not resume. The symptoms are similar to the failure mode mentioned above in the blank chip case. Can the gurus confirm if the ROM bootloader and sflash combo is resistant to power interruptions? Does the ROM bootloader write 0x0 and 0x4 locations at the end of the update or before?

Also, any ideas on why sflash and ROM bootloader do not work when device is blank? How do I go about debugging the ROM bootloader?

Thanks in advance for your time and inputs,

Rajah

TMS570LC4357: TMS570LC4357 Power consumption

$
0
0

Part Number: TMS570LC4357

What is the power consumption of TMS570LC4357?

TLV320AIC3104EVM: I2C Communication Issue

$
0
0

Hi,

I am evaluating the tlv320aic3104 codec using the evaluation kit. I was able to play the audio with the evaluation software. The evaluation software addresses the codec using an I2C address 0x30 which I got from looking at the command line console.

Now I am trying to get the codec working with an external system i.e. Beaglebone Black. I use the "speaker-test" to test the codec, but I don't get any audio out. When I scoped the I2C bus during the speaker-test, I see a bunch of write calls with NAK as shown in below screenshot.

Then I cross-checked the codec address using the i2cdetect -r 2 command (shown below) and this displays the address as 0x52 which is different from 0x30.  

Continue? [Y/n] y
      0 1 2 3 4 5 6 7 8 9 a b c d e f
00: -- -- -- -- -- -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- UU -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- 52 -- UU UU UU UU -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
70: -- -- -- -- -- -- -- --

When I run a "dmesg", I also get the below error which has somthing to do with the comms.

[ 255.458192] tlv320aic3x-codec 2-0018: Unable to sync registers 0x3-0x3. -121

Below shows my hardware setup for this procedure.

I power the codec via the USB board by connecting all the J18 (USB board) to J3 (codec). And then I make i2c and audio connections between the BBB and the codec.

Looking for some help to understand why the i2c address mismatch and resolve the communication issues.

TPS61097A-33: Back-feed power - Soft Start and Enable Pin Voltage Questions

$
0
0

Part Number: TPS61097A-33

Three separate questions:

1) Is there any time delay from the time the EN pin is taken low to the time the bypass switch is engaged? I am concerned about the energy from the output capacitors back-feeding into the coin-cell battery that I am feeding this system with.

2) Is there a way to implement a soft-start on this part? My load is only 10mA, and I want to run from a lithium coin-cell. I don't want to add huge capactiors as their leakage current at high temperatures actually surpasses the shut-down current of the switcher.

3) The VIH rating on the datasheet seems incorrect. Normally the active-high voltage levels for a digital pin are specified as a minimum and maximum required voltage where the maximum voltage would be the absolute worst case required to turn the part on. For this part the max is listed as VIN+1 volt. So this tells me that with a 2.5 volt battery feeding a 3.3V system, I can't actually guarantee that the EN pin can be activated high (it would require a worst case of 3.5V). I hope this is just a typo? Since I am running an MSP430 processor from a coin-cell with this part, the voltage available to enable the switcher is only equal to the coin cell (which could be as low as 2 volts).

Thanks!

DP83867E: DP83867 intermittent Issue on dropping pings

$
0
0

Part Number: DP83867E

Hi All,

We have the DP83867ERGZT in RGMII mode to the MAC.

We are having intermittent issues in loosing packets/pings from link partner to the MAC.

The link partner sends it out (wire shark sees it) but MAC never receives it in uboot or linux

We probed the TXclk & Rx clock @125Mhz.

TX0,1,2,3 toggle.

RXD,1,,2,3 toggle.

Far end loop back-works consistent 

Analog Loop back works consistent

When we do pings of 32 bytes size or more we see packets not reaching the MAC.

RXD0 resembles clock

Link Speed 1gbps

Power sequence /power supply- we are confident but ye to check.

Questions:

1. Is it okay RXD0 to resemble clock?

2. do you know if we have a MII decoder in the market?

3. is there an errata sheet for this phy?

BR,

Bala

LAUNCHXL-CC1352P: Help with understanding LAUNCHXL-CC1352P design

$
0
0

Part Number: LAUNCHXL-CC1352P

Hi support team,

I am adapting the LAUNCHXL-CC1352P design (PDF attached) for a project I am working on. I would like to duplicate the "radio" section of the design as my focus is on logic in a different area altogether. Please could you kindly guide me on the following:

  1. Does the "Do Not Stuff" annotation in the design mean that the greyed-out component is not fitted? If so, does it mean that components P11, P12, P13 e.t.c are not connected on the launchpads (and therefore can be skipped if copying the design)?
  2. Component J7 seems to be unavailable on most electronic parts sites (i.e. Mouser, Digikey e.t.c). Is there a compatible alternative that is more easily available?
  3. The BOM does not list the aerial A1, could you please advise what is used on the Launchpad so I can keep these bits the same?

Thanks.

(Please visit the site to view this file)


IWR6843ISK: People Counting Demo Parser

$
0
0

Part Number: IWR6843ISK

Hi,

I have been working on iwr1642 and iwr6843 and using my own python parser for people counting demo. It was working perfectly fine without any error on iwr1642 board. Recently, I ran the lab15 people counting demo on iwr6843ISK. It frequently appears invalid TLV type parsing result. In general, it still works but frames which have invalid TLV type are missing since parser is unable to parse the reset of data.

My question is if there are any changes from iwr1642 to iwr6843 in terms of data structure format. I referred this document for data structure format. It's the same as iwr1642.

http://dev.ti.com/tirex/explore/node?node=ANoGbRd1GQzjC9YA7IUpPQ__VLyFKFf__LATEST&r=VLyFKFf__3.6.2

Thanks,

Hank

TMS320F28388D: HRCAP Channels: Is The Future Just 2 Per Part?

$
0
0

Part Number: TMS320F28388D

Hello,

I've attached table 3-1 from the datasheets for the TMS320F28069 and the latest announced C28x family member:

                     Existing parts have 4 HRCAP channels.

           New parts only have 2 HRCAP capable channels.

Please let me know if there are other parts that are possibly 'new' that have as many HRCAP channels as the TMS320F28069.

Thanks,
John W.

CCS/TMS320C6727B: XDS560R on CCS v3.3

$
0
0

Part Number: TMS320C6727B

Tool/software: Code Composer Studio

Hi,

I have a Spectrum Digital XDS560R I am trying to use as a JTAG emulator with CCS 3.3.  I followed the instructions provided by SD (http://support.spectrumdigital.com/ccs3x_xds560r/) for both the base CCS 3.3 (no SR) and CCS 3.3 with SR 12, but did not have luck either way.

I am able to install the driver successfully according to Windows 10 64 bit, but when I try to connect over CCS, I have gotten two different errors:

(1) Failed to initialize emulator (acts as though device isn't even connected, but still lets me open CCS).

(2) Unable to load c:\ccstudio_v3.3\drivers\tixds560c6x.dvr, and will not open CCS.  Note that I verified this .dvr file does exist.

I have been using the XDS510 with no issues on the same setup.  However, I notice that the XDS510 setup that works is called "XDS510USB", whereas the XDS560 only provides a non-"USB" emulator option in CCSetup.

Any advice on this?  I am guessing it has something to do with Code Composer being pointed to the correct files (or the correct files being given to CCS).  It does look like CCS v3.3 should support this device though: http://emulators.spectrumdigital.com/files/Emulator_CCS_OS_Compatibility.pdf

Thanks,

Alec

MSP430F5172: Bootloader firmware update: Default UART pins, or TA0.0 / TA0.1 pins?

$
0
0

Part Number: MSP430F5172

I've already read SLAU319AB but couldn't find an answer, so I'm asking here.

In my specific hardware, I have a Bluetooth module with the typical UART interface of TX/RX/CTS/RTS pins which I have interfaced with the hardware UART on my embedded MSP430F5172 target.  I plan to use software invocation of the bootloader in my application code.  So the question is...

Can I use the default UART pins P1.1/UCA0SIMO  and P1.2/UCA0SOMI to receive application/firmware updates on this MSP430?   

The datasheet for MSP430F5172 mentions P3.6 and P3.7 are to be used;  however, I don't understand why this is stated in the datasheet rather than using the dedicated UART pins present on the device.   There are some responses on the forums here mentioning to use P3.6/P3.7 and some mentioning that it is possible to use P1.1/P1.2 ( hardware UART pins )   I can't determine the how or why of either method.  

CCS/DLP3010: How/where to create/generate or find .bin .img for uploadiing firmware on DLP3010 EVM with Code Composer Studio?

$
0
0

Part Number: DLP3010

Tool/software: Code Composer Studio

Where and how can I find /create a .bin or .img to upload firmware?

Now I am using CCS to compile firmware for DLP3010 EVM, but can't find where and how to find .bin or .img file.

Please point me to go forward.

Thank you.

CCS: Multiple HTTPCli request on separate task TM4C

$
0
0

Tool/software: Code Composer Studio

Hello,

I would like to start two HTTP GET requests on separate tasks. Unfortunately I got error messages when the HTTP calls run on the 2nd task. I have modified a simple HTTP GET example with the following:

/*
 *  ======== netIPAddrHook ========
 *  This function is called when IP Addr is added/deleted
 */
void netIPAddrHook(unsigned int IPAddr, unsigned int IfIdx, unsigned int fAdd)
{
    static Task_Handle taskHandle;
    Task_Params taskParams;
    Error_Block eb;

    /* Create a HTTP task when the IP address is added */
    if (fAdd && !taskHandle) {
        Error_init(&eb);

        Task_Params_init(&taskParams);
        taskParams.stackSize = HTTPTASKSTACKSIZE;
        taskParams.priority = 1;

        taskHandle = Task_create((Task_FuncPtr)httpTask, &taskParams, &eb);
        if (taskHandle == NULL) {
            printError("netIPAddrHook: Failed to create HTTP Task\n", -1);
        }



        taskHandle = Task_create((Task_FuncPtr)httpTask, &taskParams, &eb);
        if (taskHandle == NULL) {
            printError("netIPAddrHook: Failed to create HTTP Task\n", -1);
        }
    }
}

ss in flash
Starting the HTTP GET example
System provider is set to SysMin. Halt the target to view any SysMin contents in ROV.
Service Status: DHCPC : Enabled : : 000
Service Status: DHCPC : Enabled : Running : 000
Network Added: If-1:192.168.0.104
Service Status: DHCPC : Enabled : Running : 017
Sending a HTTP GET request to 'demo.server.io'
Sending a HTTP GET request to 'demo.server.io'
00006.037 getaddrinfo: Error: couldn't alloc DNS buffer

Error! code = -106, desc = httpTask: address resolution failed

Env:

EK-TM4C129EXL
TI 18.12.1.LTS
RTOS 2.16.0.08
XDC: 2.32.0.06

Any idea?

TFP410: No video on external HDMI monitor

$
0
0

Part Number: TFP410

This is an extension to the project found in this thread:

https://e2e.ti.com/support/interface/f/138/p/851103/3151025#3151025

You can find my schematic and code in the above link.

The issue I am having now is that there is no video output to my HDMI monitor. When I plug and unplug my HDMI cable the monitor responds with a message saying "No HDMI Signal", which only appears when the cable is connected or removed. Thus I believe my HPD is working properly - it is at least detecting that it is being plugged in. 

I.K. in the above thread suggested it may be a problem with my DE signal, which I am generating with my code. Based on the provided schematic and C files, is it possible to determine what the cause of this "No Signal" error is?

Thank you,
Isaiah Washburn


CCS/LAUNCHXL-CC1352R1: Exception when I disable security in the sensor project

$
0
0

Part Number: LAUNCHXL-CC1352R1

Tool/software: Code Composer Studio

SDK simplelink 3.30 

I imported the sensor_oad_offchip_secure project and tried to modify the configuration of the sensor. The only modification I made was disabling the security level. When I flash the sensor my board enters in a exception handler:

The assert reason is MAIN_ASSERT_HWI_TIRTOS

What is happening?

CC3220SF: Sl_stop deadlock despite timeout provided

$
0
0

Part Number: CC3220SF

Hi Support,

After an OTA, we call sl_Stop to use our new bundle.

Unfortunately, sometimes it deadlocks the MCU, despite a non-zero parameter being passed to the sl_Stop() call.

Here's where its hung up:

#ifdef SL_PLATFORM_MULTI_THREADED
    /* Do not continue until all sync object deleted (in relevant context) */
    while (g_pCB->NumOfDeletedSyncObj < MAX_CONCURRENT_ACTIONS)
    {
        usleep(100000);
    }
#endif    

The value of NumOfDeletedSyncObj is one less than MAX_CONCURRENT_ACTIONS.  Should we be manually closing all sockets before calling sl_Stop?  This shouldn't be necessary, right?

Thanks!

IWR1443: Test Pattern Generator

$
0
0

Part Number: IWR1443

Hi,
I use Test Pattern Generator from ADC buffer.
It generates 16 samplese 4 byte each.
I use code from ti\drivers\adcbuf\test

testPatternConf.period = 255;
testPatternConf.numSamples = 16;

for (channel=0; channel < 4; channel++)
{
testPatternConf.rxConfig[channel].rxIOffset = 0x0000;
testPatternConf.rxConfig[channel].rxIInc = 1;
testPatternConf.rxConfig[channel].rxQOffset = 0x1000 ;
testPatternConf.rxConfig[channel].rxQInc = 1;
}


rv = ADCBuf_control( handle, ADCBufMMWave_CMD_CONF_TEST_PATTERN, (void *)&testPatternConf );
if( rv < 0 )
{
System_printf("Error: ADCBufMMWave_CMD_CONF_TEST_PATTERN failed with [Error=%d]\n", rv);
return rv;
}

I expect to see in ADCBUF
0x10000000 0x10010001 0x10020002 and so on

But there are
0x10310031 0x10320032 0x10330033 and so on

The offset is permanent and depends on numSamples

Is this offset normal behavior for Pattern Generator?

Regards,
Gennadii

CC3235MODSF: CC3235MODSF: Dual-Band (and Single Band) External Antenna Recommends.. To Retain Pre-Certification..

$
0
0

Part Number: CC3235MODSF

Hi Team WiFi,

Reference (https://e2e.ti.com/support/wireless-connectivity/wifi/f/968/p/842148/3115233 ). 

Customer reports that..

"The Laird dual band antenna (001-0016) that was used during TI’s FCC and IC approval is quite costly.  They'd like to maintain an off-board antenna, but the referenced antenna is, once again, quite costly. 

Is TI aware of a lower cost dual-band off-board antenna that will still maintain the FCC/IC/Module approval? 

Is it true that to retain the pre-approval, the alternate antenna must be of the same antenna type and equal or less gain to what was used during the approvals testing?

I would greatly appreciate your prompt assistance in locating an alternate antenna that does not invalidate the modules pre-approval."

Any ideas?

Thanks, Merril

TLV702: Is there any internal pull-up or pull-down on the EN pin?

$
0
0

Part Number: TLV702

Hi,

Did tests with TLV70233DSE in a design where there for historical reasons is 10k in series with the EN pin (between MCU control pin and LDO EN pin) and noted that when driving the MCU pin low the EN pin did not go to 0V, but rather to 200mV. My input voltage is ~4.5V, so I was considering that perhaps there is an internal pull-up of ~200kOhm that explains this behavior, forming a voltage divider with my 10k. However, I do not find any such indication in the datasheet.

Is there any internal pull-up (or pull-down) on the EN pin?

Thanks!

Viewing all 262198 articles
Browse latest View live


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