CCS/EVMK2G: EVMK2G&CMB&MCASP

user5205376

Part Number: EVMK2G

Tool/software: Code Composer Studio

Dear,

Now I am doing CMB design,I want to run the demo（realtime_demo_bios， tiducr7.pdf）.

Hardware platform：EVMK2G REV.D + CMB; (Single Digital Signal Processor (K2G DSP C66*) and Circular Microphone Board (CMB))

software: TI-RTOS ti-processor-sdk-rtos-k2g-evm-04.01.00.06 & ti-processor-sdk-rtos-k2g-evm-03.03.00.04

Tool：CCS6.2 & CCS7.2 (Windows 10)

DEMO: C:\ti\processor_sdk_rtos_k2g_4_01_00_06\demos\audio-preprocessing\realtime_demo_bios（Error"src/mcasp_drv.c", line 4082: assertion failure）

C:\ti\processor_sdk_rtos_k2g_3_03_00_04\demos\audio-preprocessing\realtime_demo_bios（Error"src/mcasp_drv.c", line 4024: assertion failure）

According to console

[C66xx]

Static information:

tword: 1

tint: 2

tlong: 4

Max #mics: 7

Max #vmics: 12

Fs: 16000 [Hz]

Frame duration: 10 [ms]

Frame length: 160 samples

Frame size: 320 bytes

Input buffer length: 2240 samples

Frame buffer size: 4480 bytes

Dynamic information:

#mics: 7

#vmics: 12

...Initializing beamformers

Done with beamformers

...Initializing ASNR's

Done with ASNR's

...Initializing MSS

Done with MSS

...Initializing DRC

Done with DRC

******************************************

Audio Preprocessing Demo

******************************************

This Demo Takes the Audio Input from 8 Mics from CMB

Runs the BF+ASNR+MSS+DRC Signal Processing Chain on Mic1 through Mic7

The Processed Audio Output will be Sent to Left Channels of K2G EVM On-Board Line-Out

The Mic8 will be Sent to Right Channels of K2G EVM On-Board Line-Out

Please Plug a Headphone to K2G EVM On-Board Line-Out

xdc.runtime.Main: "src/mcasp_drv.c", line 4082: assertion failure

xdc.runtime.Error.raise: terminating execution

//--------------------------------------------

When we execute it according to “training.ti.com/voice-preprocessing-demo-on-evmk2g（Version of the same）”.

The output file size is the same as in video(K2G_bf_rt.out 3,933kb).

//-----------------------------------------------------------------------------------------

When we execute it according to“tiducr7.pdf” .

Failure to generate engineering audioAnalogLoopbackTest Project；

The K2G_bf Project is successful and the output is normal；

The K2G_bf_rt Project can be generated and the output is “"src/mcasp_drv.c", line 4082: assertion failure”；

Thanks in advance.

BRS,

Rob

over 6 years ago

0 Cvetolin Shulev-XID over 6 years ago

TI__Guru 65405 points

Hi Rob,

The team is notified. They will post their feedback directly here.

BR
Tsvetolin Shulev

0 Lalindra Jayatilleke over 6 years ago

TI__Mastermind 30365 points

Rob,
To start, can you please double check all your wiring.
I understand that the numbering on the expansion connector JP1 is tough on the eyes, but check that things are hooked up right.

Lali

0 user5205376 over 6 years ago in reply to Lalindra Jayatilleke

Prodigy 50 points

Lali

Thanks

We make sure there is no problem with the connection.

when Debug

main.c

status = mcaspAudioConfig();

mcasp_config.c line 535

status = mcaspCreateChan(&hMcaspTxChan, hMcaspDevTx,
MCASP_OUTPUT,
&mcaspTxChanParam,
mcaspAppCallback, &txChanMode);

mcasp_drv.c run line 1022~1345 Error
if (MCASP_COMPLETED == status)
{
do
{
......
}while(falsewhile);
}
if (MCASP_COMPLETED == status)
{
mcaspConfigureFifo(&(instHandle->hwInfo),
chanHandle,
chanHandle->enableHwFifo);
......
}

I thought it was a wrong parameter or a compiler environment error, but I found that it was different in 1.05 and 1.07 McAsp. I want to modify the McAsp driver now, how can I modify it, and can join the project?

Thanks

BRS
Rob

0 Lalindra Jayatilleke over 6 years ago in reply to user5205376

TI__Mastermind 30365 points

Rob,

This is typical to an incorrect clock or no clock at all to the McASP. Did you ensure the CMB jumpers on the underside of the board are correct?

pg 9 of http://www.ti.com/lit/ug/tiducr7/tiducr7.pdf

Lali

0 user5205376 over 6 years ago in reply to Lalindra Jayatilleke

Prodigy 50 points

This is typical to an incorrect clock or no clock at all to the McASP,We think so too.

The CMB is not J10 / J11.

So We believe that J2 on the actual circuit (project tida-01454b) is the J2 on table4, as well as J6=J8，J7=J10，J8=J11;

In addition,

1: while we are debugging, the while always has no load and assignment;
2: in the (McAspCreateChan) function, a variety of debug will appear and then pass; The end and no clock.

If possible, we would like to know how to modify dev, and then generate the ae66 file burn.

Thank you,

Rob

0 user5205376 over 6 years ago in reply to Lalindra Jayatilleke

Prodigy 50 points

We now find that the GPIO output level of the audio piece is 1.8v, and the requirement is 3.3V. How to modify it to 3.3V?

Thank

0 Lalindra Jayatilleke over 6 years ago in reply to user5205376

TI__Mastermind 30365 points

Rob,

"We now find that the GPIO output level of the audio piece is 1.8v, and the requirement is 3.3V. How to modify it to 3.3V?"

Not following you on this one. Are you talking about the CMB IO levels?

Lali

Processors

Processors forum

CCS/EVMK2G: EVMK2G&CMB&MCASP