DSP Support Forum : All Content - All Communities

ADSP21371 - ADAU1361

analog@sgaur.hosted.jivesoftware.com — Sat, 21 Nov 2009 10:24:37 GMT

thanks for the answer i understand this. But my question was about the EZ-KIT of ADSP21371. There is used DPI4 AD1835 CS. Is this DPI pin 4 is used like GPIO? And what is the perpose of the FLG3 o SPI?

BAD LINK ON VISUALDSP++ RUN-TIME LIBRARY FOR SHARC

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 22:28:03 GMT

Hi German-

Thanks for letting us know about this. We have fixed the link on our website. Here is the direct link to the manual you are seeking:

http://www.analog.com/static/imported-files/software_manuals/50_21k_RTL_man_rev1.3.pdf

AndyR

Calculating Filter Coefficients to Store in a Microcontroller

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 20:16:45 GMT

Recently there have been some email support requests regarding the calculation of filter coefficients (and other types of parameters) for storage in a microcontroller. The basic idea is to store a table of parameters in the microcontroller and download them to the appropriate addresses in the SigmaDSP's parameter RAM when required. SigmaStudio includes a few tools to help make this possible.

Method 1: Direct coefficient calculation

This method is not really preferred because it takes time and effort compared to the other methods, and doesn't offer any real benefits. In the help file, equations used for calculating first-order and second-order (biquad) IIR filter coefficients are given. For example, for a butterworth LPF:

These equations can be used to calculate parameters directly, but are a bit cumbersome.

Method 2: Fixed Point Filter Table Generator

Open the fixed point filter table generator under the Tools menu.

Here you can enter in parameters for a number of filter curves that you desire, then hit the generate button to see the generated curves on the graph.

Click the value table tab to see the biquad filter coefficients to be stored in RAM (5 coefficients are listed for each filter generated).

Click the Show Values button and then Generate again to show the decimal values and parameter names for each filter, along with the associated boost.

These values can be loaded directly into a 2nd-order general filter, set to IIR Coefficient mode in SigmaStudio. Important: Because of the way the values are stored in RAM, the coefficients for a1 and a2 must be inverted from what is shown in the filter table generator. The filter table generator shows coefficients as they are stored in RAM, whereas the IIR Coefficient filter box shows their true values as calculated by the equations.

The filter table generator allows a large number of coefficients to be calculated and stored in RAM for applications such as a user-controllable tone control.

Method 3:Use the capture window

Start a project and make sure the SigmaDSP is connected to a USBi communication channel in the Hardware Configuration tab. Insert any filter into your project and compile. Then change the filter values. The values written to the SigmaDSP's RAM are shown in the capture window (open by default, toggle by pressing Ctrl+5).

This allows you to simply generate the parameters for one filter curve.

Figuring out parameters for any SigmaStudio algorithm control using the Capture Window

You can insert any cell into the project, compile, then change its parameters to see what should be stored in RAM in the capture window. If, for example, I wanted to to a sine tone sweep that is manually controlled by a microcontroller, I could enter in a range of values into a sine tone generator and make note of the frequency parameter in the capture window (in this example, the parameter name is sin_lookupAlg19401increment).

how connect data dsp processor to PC?

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 19:18:19 GMT

I make the sound with ADSP 21,369. I wanted to ask how to save the sound of that process into the file. wav on the computer? I use k-lite session to process voice and my 21369th ADSP sharc connect this with the usb port to the computer. how to save a sound process that results in the form of a file. wav for example? help to provide tutorial or program. thanks

Device serial number

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 19:17:06 GMT

Thank you for your helpful answer.

I am using the onboard flash, it's just that I'm going to have a large array of processors and didn't want to have to remember to flash them all with unique id's.

Running exported makefile under Windows

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 17:20:59 GMT

Thanks for your help Craig. Here is the one tmm_export.mak that fails:

$(VDSP)/ccblkfn.exe .\Debug\ALA.doj .\Debug\blackfin_tmm_hardware.doj .\Debug\dhcp_settings.doj .\Debug\dns.doj .\Debug\ExceptionHandler-BF536.doj .\Debug\FFT.doj .\Debug\general.doj .\Debug\host.doj .\Debug\HOST_listener.doj .\Debug\HOST_mbt.doj .\Debug\HOST_omron.doj .\Debug\HOST_tcpip.doj .\Debug\ID.doj .\Debug\inet_utilities.doj .\Debug\lwip_sysboot_threadtype.doj .\Debug\mu_compress_packed.doj .\Debug\mu_expand_packed.doj .\Debug\NTP_client.doj .\Debug\RTP.doj .\Debug\sdp.doj .\Debug\settings.doj .\Debug\SIP.doj .\Debug\sip_authentication.doj .\Debug\SIP_dialog.doj .\Debug\SIP_ict.doj .\Debug\SIP_ist.doj .\Debug\SIP_nict.doj .\Debug\SIP_nist.doj .\Debug\SIP_registration.doj .\Debug\SIP_tcp.doj .\Debug\SIP_tcp_listener.doj .\Debug\sip_util.doj .\Debug\SWITCH.doj .\Debug\SWITCH_reporting.doj .\Debug\SYSLOG.doj .\Debug\tftp_client.doj .\Debug\tmm_basiccrt.doj .\Debug\tmm_cplbtab.doj .\Debug\tmm_heaptab.doj .\Debug\tmm_settings.doj .\Debug\upgrade.doj .\Debug\VDK.doj .\Debug\voip.doj .\Debug\WAV.doj .\Debug\wav_files.doj .\Debug\web_functions.doj .\Debug\web_main.doj Debug\webrom.doj .\Debug\xml.doj .\Debug\xml_files.doj .\Debug\xml_io.doj -T .\tmm.ldf -map .\Debug\tmm.map.xml -L .\Debug -flags-link -MD__cplusplus,-MDUSER_CRT="tmm_basiccrt.doj",-MDUSE_CACHE,-MDUSE_INSTRUCTION_CACHE,-MDUSER_CPLBTAB=".\Debug\\tmm_cplbtab.doj" -add-debug-libpaths -flags-link -MDUSE_CACHE -flags-link -od,.\Debug -o .\Debug\tmm.dxe -proc ADSP-BF536 -flags-link Kervdkbf536d.dlb,liblwIPbf536d.dlb,Tcpipbf536d.dlb,../../../Flash\ Programmer/bf536/Debug/adi_m29w320e.dlb,../../../Ethernet\ Driver/bf536/Debug/adi_ks8993.dlb,../../../webserver/bf536/Debug/libwebsvr.dlb,../../../sip/bf536/Debug/sip.dlb -si-revision 0.2 -flags-link

Rather than using the tmm_export.mak file to build from cmd.exe, I tried using tmm.mak (the automatically generated makefile when running from within the IDDE). That did not fail when I ran the gmake. Here is its command line (I have attached both makefiles).

$(VDSP)/ccblkfn.exe .\Debug\ALA.doj .\Debug\blackfin_tmm_hardware.doj .\Debug\dhcp_settings.doj .\Debug\dns.doj .\Debug\ExceptionHandler-BF536.doj .\Debug\FFT.doj .\Debug\general.doj .\Debug\host.doj .\Debug\HOST_listener.doj .\Debug\HOST_mbt.doj .\Debug\HOST_omron.doj .\Debug\HOST_tcpip.doj .\Debug\ID.doj .\Debug\inet_utilities.doj .\Debug\lwip_sysboot_threadtype.doj .\Debug\mu_compress_packed.doj .\Debug\mu_expand_packed.doj .\Debug\NTP_client.doj .\Debug\RTP.doj .\Debug\sdp.doj .\Debug\settings.doj .\Debug\SIP.doj .\Debug\sip_authentication.doj .\Debug\SIP_dialog.doj .\Debug\SIP_ict.doj .\Debug\SIP_ist.doj .\Debug\SIP_nict.doj .\Debug\SIP_nist.doj .\Debug\SIP_registration.doj .\Debug\SIP_tcp.doj .\Debug\SIP_tcp_listener.doj .\Debug\sip_util.doj .\Debug\SWITCH.doj .\Debug\SWITCH_reporting.doj .\Debug\SYSLOG.doj .\Debug\tftp_client.doj .\Debug\tmm_basiccrt.doj .\Debug\tmm_cplbtab.doj .\Debug\tmm_heaptab.doj .\Debug\tmm_settings.doj .\Debug\upgrade.doj .\Debug\VDK.doj .\Debug\voip.doj .\Debug\WAV.doj .\Debug\wav_files.doj .\Debug\web_functions.doj .\Debug\web_main.doj Debug\webrom.doj .\Debug\xml.doj .\Debug\xml_files.doj .\Debug\xml_io.doj -T .\tmm.ldf -map .\Debug\tmm.map.xml -L .\Debug -flags-link -MD__cplusplus,-MDUSER_CRT=ADI_QUOTEtmm_basiccrt.dojADI_QUOTE,-MDUSE_CACHE,-MDUSE_INSTRUCTION_CACHE,-MDUSER_CPLBTAB=ADI_QUOTE.\Debug\\tmm_cplbtab.dojADI_QUOTE -add-debug-libpaths -flags-link -MDUSE_CACHE -flags-link -od,.\Debug -o .\Debug\tmm.dxe -proc ADSP-BF536 -flags-link Kervdkbf536d.dlb,liblwIPbf536d.dlb,Tcpipbf536d.dlb,../../../Flash\ Programmer/bf536/Debug/adi_m29w320e.dlb,../../../Ethernet\ Driver/bf536/Debug/adi_ks8993.dlb,../../../webserver/bf536/Debug/libwebsvr.dlb,../../../sip/bf536/Debug/sip.dlb -si-revision 0.2 -flags-link -MM

The main difference is the ADI_QUOTE replaces double quotes in the working makefile. So I went into the tmm_export.mak file and changed the double quotes on the linker line to ADI_QUOTE and tried running from the command line. And it worked!

Interesting...

system services dma callback

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 16:24:04 GMT

Hi again,

you are correct; I gave the wrong EE-Note number in my previous reply. As you have seen it is EE-312 that I meant. Regarding the placement of the SPORT buffers in L1 - the SPORT driver does not have an explicit requirement that buffers should be placed in L1 memory.

What seems more likely in this case is that when the buffers were in SRAM there was some contention between LwIP SDRAM accesses (LwIP uses SDRAM for it's heap and stack), and the SDRAM access to the SPORT buffers. It is likely that the SPORT access may be blocking LwIP access to SDRAM to the point where there is an overflow of traffic, or vice versa.

If you would like us to help further to determine the cause of the problem when the buffers are in SDRAM, we would need a little more information regarding your setup: use of data and instruction cache, what are you using SPORT for, etc. If possible, it would be beneficial to take a look at your code.

If you are unable to post your project here, you can contact us via private support and include a link to this thread:

http://www.analog.com/processors/support

regards,

Craig.

Release vs Debug

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 15:32:24 GMT

If your project also has assembly code that you want to control with this macro, unfortunately you will also have to add the definition to the ProjectOptions:Assemble property page. There isn't a single place to define a macro that will apply to both C/C++ code and assembly code and the macro will need to be defined in two places.

optimized goertzel algo for Blackfins

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 15:23:01 GMT

Hi,

Yes, the input data (_input_s[]) is signed and is in 16.16 format.

The following is the C interface for detecting two frequencies 641 Hz and 941 Hz.

int goertzel(int, int);

int fre_present_flag;

int main()

{

int c,s,f, N;

f = 941; // 641, 941

N = 205; // block length

if (f== 641) {

c = 115625;

}

else { // 941

c = 97181;

}

fre_present_flag = goertzel(c,N);

return 0;

}

Bootloading Blackfin BF537 with VisualDSP

analog@sgaur.hosted.jivesoftware.com — Fri, 20 Nov 2009 15:16:47 GMT

Hi Andreas,

For completeness, I just wanted to follow up with some information from the VisualDSP++ tools point of view. I am not sure if it is your intention to use U-Boot and/or uClinux, or if you particularly need to boot over UART, but I wanted to add some information regarding creation and programming of a bootable application under VisualDSP++.

With a VisualDSP++ project, you can set the project to produce a Loader File under "Project:Project Options" where you have the option to set the Project Type to Executable, Loader, Library, or Object. Setting this to Loader will invoke the elfloader utility at the end of the build.

Under "Project Options:Load" you can configure the general project options. For the BF537 EZ-KIT Lite, the 'Boot Mode' you should select "Flash/PROM". For 'Boot Format' you should select"Intel hex", and for 'Output Width' you should select "16".

If you use SDRAM you should also set an Initialization File to configure the EBIU on boot. There is an example for the BF537 EZ-KIT Lite located at "...\Blackfin\ldr\ezkitBF537_initcode.dxe"

When you build your project, the make process will invoke the elfloader and produce a *.ldr file. This can be written to flash on the EZ-KIT Lite using the Flash Programmer Utility. We provide a Flash Programmer Driver (*.dxe) for the EZ-KIT Lite under "...\Blackfin\Examples\ADSP-BF537\Flash Programmer\". This should be loaded on the 'Driver' tab of the Flash Programmer Utility, then on the 'Programming' tab you should select your Loader (*.ldr) file.

Once the file is programmed to flash, setting the BMODE selector (SW16) to boot from parallel flash (position 1), and power cycle the board. Your application should boot.

Hope this information is useful to you, or anyone else looking to create a bootable loader stream within the VisualDSP++ tools. You may also want to take a look at the readme for the Power On Self Test example, which has some good advice for configuring and generating a loader for the application. This can be found at "...\Blackfin\Examples\ADSP-BF537 EZ-Kit Lite\Power_On_Self_Test\".

regards,

Craig.