Two ADAU1761 in one project, how to code and compile in C?

I'm not an expert on this, so I will forward the question to two of our software engineers who might be more familiar with this use case. They will post a response here shortly.

Hello,

The software team is making new IC-specific names for these definitions and this will be updated in the next release.

Hello,

I appreciate the effort of easing the C-development of a microcontroller application for this very interesting ADAU1761!

The graphical development environment is really good!

When do you think that the next release could be due?

Thanks again

Dany

We typically release betas every two weeks or so. The most recent 3.5.3 was released a few days ago, so 3.5.4 (including these changes) should be out in approximately 1.5 weeks... perhaps 2 weeks.

Hello,

The version SigmaStudio 3.5.4 BETA 64-Bit OS, with the promised corrections is still not available.

To increase my confidence I could go ahead and finish the design of the schematic and physical board with two ADAU1761's, and before I could continue programming my 32-bit microcontroller, I did this:

I wrote a Windows - command-line program in the C language to put a chip number on field and routine names SigmaDsp generates in all header and mode files.

I was succesful in that and now can compile without errors; this does not mean I lost hope to get a better release from ADI.

Thanks anyway for all the good work.

Best regards

Hi Dany,

This feature has been updated in SigmaStudio Beta 3.5.5. I attached an example project for your reference. The text "IC_1" and "IC_2" has been added to all definitions in each exported file. I included an example attachment.

Currently, this version is not appearing on the website. I will check with our web group about getting the website version updated as soon as possible.

Hello Brett !

EXCELLENT !

..nice effort, but not perfect.

In the contribution to a better software, I allow myself to give you the following info:

1)

In the CRF2_IC_1.h file, go at the end where the default_download is, and see that the fields in bold and italic are not defined according to the IC number, while the real fields are!

The rest in the file is outstanding in terms of chip numbering!

SIGMA_WRITE_REGISTER_BLOCK( DEVICE_ADDR_IC_1, PROGRAM_ADDR, PROGRAM_SIZE, Program_Data_IC_1 );

SIGMA_WRITE_REGISTER_BLOCK( DEVICE_ADDR_IC_1, PARAM_ADDR, PARAM_SIZE, Param_Data_IC_1 );

2)

The mode file indexed 353 is of the Beta 3.5.3.

Now, I could have infulenced the register Byte names and the download function name by giving a header file with the chip number in it.

Now compare with the mode file Beta 355:

Still to be done : The device address "DEVICE_ADDR_" is still not indexed with the chip name.

REGRESSION : The memory addresses are now set to ZEROES ! This is important, and need to be corrected.

Brett, I would like to encourage the development person(s) to go on doing good work !

Best regards

Dany

Hi Dany,

I think the addresses are set to zero because you might not have compiled before clicking the export button. Can you please try again? First, Link-Compile-Download, and then click the Export System Files button.

Hello Brett,

You are right, the addresses appreared when I did the right sequence of actions.

Sorry about the wrong point.

The other remarks are right, though.

Best regards

Dany

Got it! We're going to change this soon. We're also going to update the function calls to include the IC. That way, everything will be compatible with multiple IC systems.

For example...

default_download() will become default_download_IC1()

Hello Brett,

This may be a long litany to our other readers, but I think that we all may be reaching an interesting goal "soon"!

The remaining points (other than the above ones that still need to be corrected) are:

Please check the two JPG files attached, you will see that all the black (REG) symbols are not found in the xxxx_IC_1_REG.h and xxxx_IC_2_REG.h files

Your programmer will see the difference.

Also, the NON_MODULO_xxx symbols do not match.

Thanks for effort!

Best regards

Dany

Hello,

The release 3.5 ADI released December 6, 2011 did not resolve the problems discussed above.

The two series of errors still appear as shown in the message just above.

I do not see how one could compile and link a C-program running in a microcontroller driving two ADAU1761, as drawn in your drawing dating back from September 20, 2010, attached.

in the xxx_IC_2.h file:

Again, some register names got their ic names quite right as in here:

SIGMA_WRITE_REGISTER_BLOCK( DEVICE_ADDR_IC_2, REG_PLLCRLREGISTER_IC_2_ADDR, REG_PLLCRLREGISTER_IC_2_BYTE, R3_PLLCRLREGISTER_IC_2_Default );

As seen from the appended files in the message above, most of the errors are of that style, where the IC2_ is missing :

SIGMA_WRITE_REGISTER_BLOCK( DEVICE_ADDR_IC_2, REG_CRC_IDEAL_1_ADDR , R17_CRC_REGISTERS_IC_2_SIZE, R17_CRC_REGISTERS_IC_2_Default );

...and should be:

SIGMA_WRITE_REGISTER_BLOCK( DEVICE_ADDR_IC_2, REG_CRC_IDEAL_1_IC_2_ADDR , R17_CRC_REGISTERS_IC_2_SIZE, R17_CRC_REGISTERS_IC_2_Default );

...but the following line got its IC name wrong, to be corrected the other way:

SIGMA_WRITE_REGISTER_BLOCK( DEVICE_ADDR_IC_2, NON_MODULO_RAM_ADDRIC_2, NON_MODULO_RAM_SIZEIC_2, NON_MODULO_RAM_DataIC_2 );

No progress since then

Please comment

PS: Of course, I adapted my batch program mentioned Oct. 19th, so that I could compile right...

Keep the good work!

Regards

Dany

Hello Brett and whole team!

I just tested version 3.6.2 of SigmaStudio, and -good news- I can now generate and compile for two ADAU1761 with correct header files!

All OK!

A big thanks to you and the whole team!
Greetings

Dany

Excellent! Thanks Dany!