ADC reading by PRU

[Youngtae Jo]

I've tried to find some example of ADC reading by PRU for my project, but I couldn't find it.

And I made that of source code and attach here for some people who have the same problem with me.

I hope it will be helpful.

[ AM335x ARM-CORE PRU ADC Example ]

 - Sequence of example

 1. Install compiling environment of PRU

 2. Enable PRU 

 3. Enable ADC 

 4. Example source

    - This example source collects ADC data from AIN0 pin with 16khz sampling rate.

- The collected data are saved into "Results.txt" file.

- The example source are "Makefile", "ADCCollector.c", "ADCCollector.p", "ADCCollector.hp"

[ Install compile environment ]

 1. Get a copy of the am335x_pru_package -> https://github.com/beagleboard/am335x_pru_package

    You also can download the am335x_pru_package here -> https://github.com/beagleboard/am335x_pru_package/archive/master.zip

 2. If you downloaded the archive, unpack it somewhere under your home directory.

 3. Make a new directory /usr/include/pruss/ and copy the files prussdrv.h  

    and pruss_intc_mapping.h into it (from am335x_pru_package-master/pru_sw/app_loader/include). 

Check the permissions; if you used the .zip file, these headers will likely have the execute bits on. 

It doesn't really hurt anything, but is certainly not what you want.

 4. Change directory to am335x_pru_package-master/pru_sw/app_loader/interface 

    then run: CROSS_COMPILE= make (note the space between the = and the command).

 5. The previous step should have created four files in am335x_pru_package-master/pru_sw/app_loader/lib: libprussdrv.a, libprussdrvd.a, libprussdrvd.so and libprussdrv.so. 

    Copy these all to /usr/lib then run ldconfig.

 6. Change directory to am335x_pru_package-master/pru_sw/utils/pasm_source 

    then run source linuxbuild to create a pasm executable one directory level up. 

- If linuxbuild doesn't have permission to execution, give the permission by run this

  : chmod +x linuxbuild

Copy it to /usr/bin and make sure you can run it. 

If you invoke it with no arguments, you should get a usage statement.

 

[ Enable PRU ]

 Before using PRU, we need to enable the PRU core, you can do it as shown below

 # echo BB-BONE-PRU-01 > /sys/devices/bone_capemgr.8/slots

[ Enable ADC ]

 Before using ADC, we also need to enable ADC, you can do it as shown below

 # echo cape-bone-iio > /sys/devices/bone_capemgr.*/slots

 

[ ADC Example - Makefile]

CFLAGS+=-Wall -Werror

LDLIBS+= -lpthread -lprussdrv

all: ADCCollector.bin ADCCollector

clean:

rm -f ADCCollector *.o *.bin

ADCCollector.bin: ADCCollector.p

pasm -b $^

ADCCollector: ADCCollector.o

[ ADC Example - ADCCollector.p]

// Developed by Youngtae Jo in Kangwon National University (April-2014)

// This program collects ADC from AIN0 with certain sampling rate.

// The collected data are stored into PRU shared memory(buffer) first.

// The host program(ADCCollector.c) will read the stored ADC data

// This program uses double buffering technique. 

// The host program can recognize the buffer status by buffer status variable

// 0 means empty, 1 means first buffer is ready, 2 means second buffer is ready.

// When each buffer is ready, host program read ADC data from the buffer.

.origin 0 // offset of the start of the code in PRU memory

.entrypoint START // program entry point, used by debugger only

#include "ADCCollector.hp"

#define BUFF_SIZE 0x00000FA0 //Total buff size: 4kbyte(Each buffer has 2kbyte: 500 piece of data)

#define HALF_SIZE BUFF_SIZE / 2

#define SAMPLING_RATE 16000 //Sampling rate(16khz)

#define DELAY_MICRO_SECONDS (1000000 / SAMPLING_RATE) //Delay by sampling rate

#define CLOCK 200000000 // PRU is always clocked at 200MHz

#define CLOCKS_PER_LOOP 2 // loop contains two instructions, one clock each

#define DELAYCOUNT DELAY_MICRO_SECONDS * CLOCK / CLOCKS_PER_LOOP / 1000 / 1000 * 3

.macro DELAY

    MOV r10, DELAYCOUNT

    DELAY:

        SUB r10, r10, 1

        QBNE DELAY, r10, 0

.endm

.macro READADC

    //Initialize buffer status (0: empty, 1: first buffer is ready, 2: second buffer is ready)

    MOV r2, 0

    SBCO r2, CONST_PRUSHAREDRAM, 0, 4 

    INITV:

        MOV r5, 0 //Shared RAM address of ADC Saving position 

        MOV r6, BUFF_SIZE  //Counting variable 

    READ:

        //Read ADC from FIFO0DATA

        MOV r2, 0x44E0D100 

        LBBO r3, r2, 0, 4 

        //Add address counting

        ADD r5, r5, 4

        //Write ADC to PRU Shared RAM

        SBCO r3, CONST_PRUSHAREDRAM, r5, 4 

        DELAY

        

        SUB r6, r6, 4

        MOV r2, HALF_SIZE

        QBEQ CHBUFFSTATUS1, r6, r2 //If first buffer is ready

        QBEQ CHBUFFSTATUS2, r6, 0 //If second buffer is ready

        QBA READ

    //Change buffer status to 1

    CHBUFFSTATUS1:

        MOV r2, 1 

        SBCO r2, CONST_PRUSHAREDRAM, 0, 4

        QBA READ

    //Change buffer status to 2

    CHBUFFSTATUS2:

        MOV r2, 2

        SBCO r2, CONST_PRUSHAREDRAM, 0, 4

        QBA INITV

    //Send event to host program

    MOV r31.b0, PRU0_ARM_INTERRUPT+16 

    HALT

.endm

// Starting point

START:

    // Enable OCP master port

    LBCO r0, CONST_PRUCFG, 4, 4

    CLR r0, r0, 4

    SBCO r0, CONST_PRUCFG, 4, 4

    //C28 will point to 0x00012000 (PRU shared RAM)

    MOV r0, 0x00000120

    MOV r1, CTPPR_0

    ST32 r0, r1

    //Init ADC CTRL register

    MOV r2, 0x44E0D040

    MOV r3, 0x00000005

    SBBO r3, r2, 0, 4

    //Enable ADC STEPCONFIG 1

    MOV r2, 0x44E0D054

    MOV r3, 0x00000002

    SBBO r3, r2, 0, 4

    //Init ADC STEPCONFIG 1

    MOV r2, 0x44E0D064

    MOV r3, 0x00000001 //continuous mode

    SBBO r3, r2, 0, 4

    //Read ADC and FIFOCOUNT

    READADC

[ ADC Example - ADCCollector.c]

/******************************************************************************

* Include Files                                                               *

******************************************************************************/

// Standard header files

#include <stdio.h>

#include <stdlib.h>

#include <sys/mman.h>

#include <fcntl.h>

#include <errno.h>

#include <unistd.h>

#include <string.h>

#include <time.h>

// Driver header file

#include <pruss/prussdrv.h>

#include <pruss/pruss_intc_mapping.h>

/******************************************************************************

* Local Macro Declarations                                                    * 

******************************************************************************/

#define PRU_NUM 0

#define OFFSET_SHAREDRAM 2048 //equivalent with 0x00002000

#define PRUSS0_SHARED_DATARAM    4

#define SAMPLING_RATE 16000 //16khz

#define BUFF_LENGTH SAMPLING_RATE

#define PRU_SHARED_BUFF_SIZE 500

#define CNT_ONE_SEC SAMPLING_RATE / PRU_SHARED_BUFF_SIZE

/******************************************************************************

* Functions declarations                                                      * 

******************************************************************************/

static void Enable_ADC();

static void Enable_PRU();

static unsigned int ProcessingADC1(unsigned int value);

/******************************************************************************

* Global variable Declarations                                                * 

******************************************************************************/

static void *sharedMem;

static unsigned int *sharedMem_int;

/******************************************************************************

* Main                                                                        * 

******************************************************************************/

int main (int argc, char* argv[])

{

FILE *fp_out;

    unsigned int ret;

    tpruss_intc_initdata pruss_intc_initdata = PRUSS_INTC_INITDATA;

int i = 0, cnt = 0, total_cnt = 0;

int target_buff = 1;

int sampling_period = 0;

if(argc != 2){

printf("\tERROR: Sampling period is required by second\n");

printf("\t       %s [sampling period]\n", argv[0]);

return 0;

}

sampling_period = atoi(argv[1]);

/* Enable PRU */

Enable_PRU();

/* Enable ADC */

Enable_ADC();

/* Initializing PRU */

    prussdrv_init();

    ret = prussdrv_open(PRU_EVTOUT_0);

    if (ret){

        printf("\tERROR: prussdrv_open open failed\n");

        return (ret);

    }

    prussdrv_pruintc_init(&pruss_intc_initdata);

    printf("\tINFO: Initializing.\r\n");

    prussdrv_map_prumem(PRUSS0_SHARED_DATARAM, &sharedMem);

    sharedMem_int = (unsigned int*) sharedMem;

/* Open save file */

fp_out = fopen("Results.txt", "w");

if(fp_out == NULL){

printf("\tERROR: file open failed\n");

return 0;

}

/* Executing PRU. */

printf("\tINFO: Sampling is started for %d seconds\n", sampling_period);

    printf("\tINFO: Collecting");

    prussdrv_exec_program (PRU_NUM, "./ADCCollector.bin");

/* Read ADC */

while(1){

while(1){

if(sharedMem_int[OFFSET_SHAREDRAM] == 1 && target_buff == 1){ // First buffer is ready

for(i=0; i<PRU_SHARED_BUFF_SIZE; i++){

fprintf(fp_out, "%d\n", ProcessingADC1(sharedMem_int[OFFSET_SHAREDRAM + 1 + i]));

}

target_buff = 2;

break;

}else if(sharedMem_int[OFFSET_SHAREDRAM] == 2 && target_buff == 2){ // Second buffer is ready

for(i=0; i<PRU_SHARED_BUFF_SIZE; i++){

fprintf(fp_out, "%d\n", ProcessingADC1(sharedMem_int[OFFSET_SHAREDRAM + PRU_SHARED_BUFF_SIZE + 1 + i]));

}

target_buff = 1;

break;

}

}

if(++cnt == CNT_ONE_SEC){

printf(".");

total_cnt += cnt;

cnt = 0;

}

if(total_cnt == CNT_ONE_SEC * sampling_period){

printf("\n\tINFO: Sampling completed ...\n");

break;

}

}

fclose(fp_out);

    printf("\tINFO: PRU completed transfer.\r\n");

    prussdrv_pru_clear_event (PRU_EVTOUT_0, PRU0_ARM_INTERRUPT);

    /* Disable PRU*/

    prussdrv_pru_disable(PRU_NUM);

    prussdrv_exit();

    return(0);

}

/*****************************************************************************

* Local Function Definitions                                                 *

*****************************************************************************/

/* Enable ADC */

static int Enable_ADC()

{

FILE *ain;

ain = fopen("/sys/devices/bone_capemgr.8/slots", "w");

if(!ain){

printf("\tERROR: /sys/devices/bone_capemgr.8/slots open failed\n");

return -1;

}

fseek(ain, 0, SEEK_SET);

fprintf(ain, "cape-bone-iio");

fflush(ain);

return 0;

}

/* Enable PRU */

static int Enable_PRU()

{

FILE *ain;

ain = fopen("/sys/devices/bone_capemgr.8/slots", "w");

if(!ain){

printf("\tERROR: /sys/devices/bone_capemgr.8/slots open failed\n");

return -1;

}

fseek(ain, 0, SEEK_SET);

fprintf(ain, "BB-BONE-PRU-01");

fflush(ain);

return 0;

}

/* 

 * FIFO0DATA register includes both ADC and channelID

 * so we need to remove the channelID

 */

static unsigned int ProcessingADC1(unsigned int value)

{

unsigned int result = 0;

result = value << 20;

result = result >> 20;

return result;

}

[ ADC Example - ADCCollector.hp]

// *****************************************************************************/

// file:   PRU_memAccess_DDR_PRUsharedRAM.hp

//

// brief:  PRU_memAccess_DDR_PRUsharedRAM assembly constants.

//

//

//  (C) Copyright 2012, Texas Instruments, Inc

//

//  author     M. Watkins

// *****************************************************************************/

#ifndef _PRU_memAccess_DDR_PRUsharedRAM_HP_

#define _PRU_memAccess_DDR_PRUsharedRAM_HP_

// ***************************************

// *      Global Macro definitions       *

// ***************************************

// Refer to this mapping in the file - \prussdrv\include\pruss_intc_mapping.h

#define PRU0_PRU1_INTERRUPT     17

#define PRU1_PRU0_INTERRUPT     18

#define PRU0_ARM_INTERRUPT      19

#define PRU1_ARM_INTERRUPT      20

#define ARM_PRU0_INTERRUPT      21

#define ARM_PRU1_INTERRUPT      22

#define CONST_PRUCFG     C4

#define CONST_PRUDRAM        C24

#define CONST_PRUSHAREDRAM   C28

#define CONST_DDR            C31

// Address for the Constant table Block Index Register (CTBIR)

#define CTBIR          0x22020

// Address for the Constant table Programmable Pointer Register 0(CTPPR_0)

#define CTPPR_0         0x22028

// Address for the Constant table Programmable Pointer Register 1(CTPPR_1)

#define CTPPR_1         0x2202C

.macro  LD32

.mparam dst,src

    LBBO    dst,src,#0x00,4

.endm

.macro  LD16

.mparam dst,src

    LBBO    dst,src,#0x00,2

.endm

.macro  LD8

.mparam dst,src

    LBBO    dst,src,#0x00,1

.endm

.macro ST32

.mparam src,dst

    SBBO    src,dst,#0x00,4

.endm

.macro ST16

.mparam src,dst

    SBBO    src,dst,#0x00,2

.endm

.macro ST8

.mparam src,dst

    SBBO    src,dst,#0x00,1

.endm

// ***************************************

// *    Global Structure Definitions     *

// ***************************************

.struct Global

    .u32 regPointer

    .u32 regVal

.ends

// ***************************************

// *     Global Register Assignments     *

// ***************************************

.assign Global, r2, *, global

#endif //_PRU_memAccess_DDR_PRUsharedRAM_

[Jason Kridner]

On Tue, Apr 8, 2014 at 6:15 AM, Youngtae Jo <young...@gmail.com> wrote:
> I've tried to find some example of ADC reading by PRU for my project, but I
> couldn't find it.
> And I made that of source code and attach here for some people who have the
> same problem with me.
> I hope it will be helpful.

Thanks! At first glance, it looks rather helpful. I'd encourage you to
submit a fork to the am335x_pru_package on Github as many people are
doing to share software examples for using the PRUs. It would
certainly reduce the number of steps required for people to reproduce
your work.

Note that the "8" isn't always an "8", you should do:

# echo BB-BONE-PRU-01 > /sys/devices/bone_capemgr.*/slots

This isn't always at "8".

> if(!ain){
> printf("\tERROR: /sys/devices/bone_capemgr.8/slots open failed\n");
> return -1;
> }
> fseek(ain, 0, SEEK_SET);
> fprintf(ain, "cape-bone-iio");

If you did this from the command-line, why are you doing it again here
within your program?

> fflush(ain);
> return 0;
> }
>
> /* Enable PRU */
> static int Enable_PRU()

Again, this seems redundant with the command-line. If you are going to
do it within your program, I believe you need to make it a bit more
flexible.

> {

> FILE *ain;
>
> ain = fopen("/sys/devices/bone_capemgr.8/slots", "w");
> if(!ain){
> printf("\tERROR: /sys/devices/bone_capemgr.8/slots open failed\n");
> return -1;
> }
> fseek(ain, 0, SEEK_SET);
> fprintf(ain, "BB-BONE-PRU-01");
> fflush(ain);
> return 0;
> }
>
> /*
> * FIFO0DATA register includes both ADC and channelID
> * so we need to remove the channelID
> */
> static unsigned int ProcessingADC1(unsigned int value)
> {
> unsigned int result = 0;
>
> result = value << 20;
> result = result >> 20;

Wouldn't a mask be more efficient? I would think you were trying to
sign-extend, but you aren't. Well, the C compiler is probably smart
enough to optimize this out anyway.

> --
> For more options, visit http://beagleboard.org/discuss
> ---
> You received this message because you are subscribed to the Google Groups
> "BeagleBoard" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to beagleboard...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[msc.a.f...@googlemail.com]

Hi Youngtae,

I am trying to create a little example to read ADC data at high speed using the PRU. I would like to build something basic, 

so that other people can use it after to build something useful on top of that. I am a beginner in the PRU area, so I would appreciate

if you can share your code with me, so I can start working from there. I will obviously put your name on my code, if you don't mind to help

me with this little project. 

Thank you in advanced   

[TJF]

Am Dienstag, 8. April 2014 12:15:31 UTC+2 schrieb Youngtae Jo:

I've tried to find some example of ADC reading by PRU for my project, but I couldn't find it.

What about libpruio? (Flexible step configuration, up to 8Mb samples (limited by the kernel driver), programmable trigger events to start the measurement, ...)

[Rodrigo Caropreso]

Hello,

Thanks for the samples, it will make a great help on performing the ADC reading using PRU.

I'm trying to use the code for this task, but when I run the sample application, I'm getting a segmentation fault.

Since I'm new on the BeagleBone issues, any tips regarding what I should be looking (regarding version, libraries, etc) would be very helpful.

The BBB is running the Angstrom default distro. Also, I've noticed that pasm compiler on the pru package is at 0.85 version while the OS pasm version is 0.84 (and the 0.84 returns a seg fault when used to compile the sample codes). Maybe this is a hint, but I do not know which additional files I'd should take a look at.

Thanks in advance for your attention.

Best regards,

Rodrigo

[serge...@gmail.com]

I'm trying to use the code for this task, but when I run the sample application, I'm getting a segmentation fault.

Segmentation faults in 90% occasions come from null-pointer dereferencing.

Please add few debug prints to your code and tell us where the crash happens.

(by the way, do you have/see any printout on the console, before the "segmentation fault"?)

[Antonio Fernandez]

Hi Rodrigo,

If you send me your code, I can help you to find the problem in your program. I have been writing some changes to the original work of Youngtae and it is working OK for me.

Regards,

Antonio

--
For more options, visit http://beagleboard.org/discuss
---

You received this message because you are subscribed to a topic in the Google Groups "BeagleBoard" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/beagleboard/0a4tszlq2y0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to beagleboard...@googlegroups.com.

[Rodrigo Caropreso]

Hello,

Sorry for the delay on answering the mails group, but on the beginning of the week I do not have much time available to work on the BeagleBone issues.

I've downloaded the most recent Angstrom image (on the site) and I've just updated the BeagleBone Black now.

Regarding the segmentation fault: I was following the instructions on this page: http://boxysean.com/blog/2012/08/12/first-steps-with-the-beaglebone-pru/

The seg fault has happened in the execution of the sample: ./PRU_memAccessPRUDataRam

The info was something like:

AM33XX INFO: Initializing example. INFO: Executing example.

Segmentation fault.

(The remaining messages were not showed).

Since is is a sample code, I was thinking that this code would be reliable enough to confirm iif my BBB's PRU is correctly configured and running.

In fact I'm trying to read analog data (ADC) using PRU (because the usual way, using the file system is not fast enough for what I need - but I accept any suggestions to get data from ADC in a 10KHz sample rate ).

I was planning to use the libpruio Library (my assembly is a little bit rusty and this library should help me to achieve this task) but last week I've got stuck into this seg fault issue.

Now, with an updated OS, I'll try to follow these path again (perhaps I'll be more lucky this time).

Is it possible to use the libpruio (and PRUSSDRV) on Angstrom or should I install Ubuntu or Debian instead?

Thanks for all your information and best regards,

Rodrigo.

Em terça-feira, 8 de abril de 2014 07h15min31s UTC-3, Youngtae Jo escreveu:

[TJF]

Hello Rodrigo,

Am Samstag, 16. August 2014 00:19:01 UTC+2 schrieb Rodrigo Caropreso:

I was planning to use the libpruio Library (my assembly is a little bit rusty and this library should help me to achieve this task) but last week I've got stuck into this seg fault issue.

...

Is it possible to use the libpruio (and PRUSSDRV) on Angstrom or should I install Ubuntu or Debian instead?

libpruio needs libsupc++, which isn't available on Angström (AFAIK). I recommend to install Ubuntu or Debian.

[Rodrigo Caropreso]

Hello,

I need to thank you very much. 

I could successfully compile and run your code and this is exactly what I was looking for.

So, again: thank you very much.

If I may point just one minor detail: the declaration of the following two functions:

  static void Enable_ADC();

  static void Enable_PRU();

They are not matching the respective definitions (which are using int as return type) and this causes a compilation error.

But, after the corrections the code has run smoothly. :)

Best regards,

Rodrigo T. Caropreso

Em terça-feira, 8 de abril de 2014 07h15min31s UTC-3, Youngtae Jo escreveu:

[Antonio Fernandez]

Hi Rodrigo,

Can you send me your code? otherwise it is impossible for me to answer all your questions.

Regards,

Antonio

[mahmou...@hotmail.com]

Hello sir, I have some question regarding your code please.

What is the maximum sampling rate that I can get using this code and what should I change to get that sampling rate?

If I want to read from all 7 analog pins, how can I do that?

Thank you for your help and your effort.

[celso maia]

I got one error, and I don't know how to solve this.

"ERROR: Sampling period is required by second"

Anyone can help me?

[mahmou...@hotmail.com]

I use BBBIOlib-master. You can get it from https://github.com/VegetableAvenger/BBBIOlib

I think it is better than this library in my own opinion :) It is easier to be installed and is complete. It also contains examples.

[celso maia]

Thanks Mahmoud, I'll try with this library .

Best regards

--
For more options, visit http://beagleboard.org/discuss
---
You received this message because you are subscribed to a topic in the Google Groups "BeagleBoard" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/beagleboard/0a4tszlq2y0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to beagleboard...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Celso Machado Maia Padilha

Fones: OI (83) 8665-7615 / TIM (83) 9960-6442

Aluno do Programa de Pós-Graduação em Engenharia Elétrica - PPGEE

Universidade Federal da Paraíba - UFPB

Graduado em Ciências da Computação

Centro Universitário de João Pessoa - UNIPÊ

[ybe...@rehut.com]

Using the PRU is a round about way to get regular sampling. The ADC can be
configured to sample at a regular rate and place samples in a FIFO and
interrupt as the FIFO fills. The PRU isn't needed. See the TRM.

> > ****/ // Standard header files

> > #include <stdio.h>
> > #include <stdlib.h>
> > #include <sys/mman.h>
> > #include <fcntl.h>
> > #include <errno.h>
> > #include <unistd.h>
> > #include <string.h>
> > #include <time.h>
> >
> > // Driver header file
> > #include <pruss/prussdrv.h>
> > #include <pruss/pruss_intc_mapping.h>
> >
> >
> > /*************************************************************************

> > ***** * Local Macro Declarations
> >
> > *
> >
> > **************************************************************************
> > ****/ #define PRU_NUM 0

> > #define OFFSET_SHAREDRAM 2048 //equivalent with 0x00002000
> >
> > #define PRUSS0_SHARED_DATARAM 4
> > #define SAMPLING_RATE 16000 //16khz
> > #define BUFF_LENGTH SAMPLING_RATE
> > #define PRU_SHARED_BUFF_SIZE 500
> > #define CNT_ONE_SEC SAMPLING_RATE / PRU_SHARED_BUFF_SIZE
> >
> >
> > /*************************************************************************
> > ***** * Functions declarations
> >
> > *
> >
> > **************************************************************************

> > ****/ static void Enable_ADC();

> > static void Enable_PRU();
> > static unsigned int ProcessingADC1(unsigned int value);
> >
> >
> > /*************************************************************************

> > ***** * Global variable Declarations
> >
> > *
> >
> > **************************************************************************
> > ****/ static void *sharedMem;

> > static unsigned int *sharedMem_int;
> >
> >
> > /*************************************************************************
> > ***** * Main
> >
> > *
> >
> > **************************************************************************

> > ****/ int main (int argc, char* argv[])

> > **** * Local Function Definitions
> >
> > *
> >
> > **************************************************************************
> > ***/ /* Enable ADC */

> > ***/ // file: PRU_memAccess_DDR_PRUsharedRAM.hp

--
Hunyue Yau
http://www.hy-research.com/

[celso maia]

Finnaly Works!

Thanks a lot!

Em terça-feira, 8 de abril de 2014 07:15:31 UTC-3, Youngtae Jo escreveu:

[TJF]

Am Freitag, 6. Februar 2015 00:06:45 UTC+1 schrieb H:

Using the PRU is a round about way to get regular sampling. The ADC can be
configured to sample at a regular rate and place samples in a FIFO and
interrupt as the FIFO fills. The PRU isn't needed. See the TRM.

That's not entirely correct. Yes, you can configure the ADCSS to measure in continous mode, but you don't have control over the timing. To get an exact sampling rate it needs a trigger, either one of the TIMER-[4-7] or a software trigger (ie. from PRU code). When using the PRU for trigger, it can also do FIFO clearance and low level data manipulation, like scaling the samples to 16 bit (ie. to use them for audio data). The key benefits of the PRU solution are being more flexible and coding single source on a real-time unit.

[ybe...@rehut.com]

You can control the timing on the ADC with the open/close times on each step.

I have the ADC by itself working well enough for an 8KHz sampling with noise
filtering using the oversampling/averaging for audio purposes. The driver
convert the samples as 16bit (simple bit shift) as it unloads the FIFO. The
bigger problem is noise on the ADC itself.

Using the PRU would slam the internal bus twice eatting up bandwidth. It is no
big deal if you don't need the bandwidth...

[marcelo...@gmail.com]

Dear Youngtae Jo, thanks for your kindness on sharing your code. It pushed me to finally obtain some results.

However there are certain points that you need to take care on your code. I will try to list at least most of them here:

1. you haven't set anything to the ADC_CLKDIV register. On my BBB, this register is set to 7 by default, so the ADC Clock turns out to be 24MHz/(7+1) = 3 MHz, and the PRU's clock is not a multiple of it. So I suggest to set the ADC_CLKDIV to something that divides the 24 MHz by 3 (2, 5, 11...). Check the SPRUH73L guide for more information on this register.
   
2. your code will only work with the PRU0 because of the way things are defined in the .hp file. I know you got it from the am335x github, but I believe they didn't make a good job on that header file. You can read more about this problem here.
   
3. that offset has no reason, I know it was also taken by the am335x github file, but I believe they made that for pedagogical reasons only, as it is an example so they are showing possibilities. I would set the C28 register to 0x100 so that it will point to the first shared memory address (0x10000). Once that's done you can take rid of the offsets in the C code.
4. your calculation on the number of loops is wrong. If you know the sampling frequency, then you need the time step between two samples, then you know how many 5 ns instructions fit on it. Finally divide it by 2, as you did, as every loop takes 2 instructions.
5. I believe that anyone coding an ADC capture in assembly is really looking for some time precision, so you'd need to count every single instruction you make outside of the delay loop and compensate that many instructions in the loop. That's a tedious job as you need to make the number of instructions even, and make that number be the same no matter where the code jumps to.
6. you have used bone_capemgr.8 but that could vary, and it does, as on my BBB that number is 9.
7. according to the SPRUH73L guide, the Enabled bit from the CTRL register must be set only after all the steps configs are set.
8. the first "sample" on the Results.txt file is actually the buffer flag, I suggest to remove the first line to avoid misreadings.
9. you have some divisions in your numerical defines, I don't recommend that because one may have a numerator which is not a multiple of the denominator and then he'll lose precision.
10. I suggest a complete restructure of the communication between the C code and the P code, so that you don't have to manually edit each parameter twice. Sampling frequency, buffer size, you know...
11. less important, but it would be great if you defined a register to contain the ADC_TSC base register, and then define the offsets of the others. It would have made your code more readable and flexible.
12. you don't need to define PRUSS0_SHARED_DATARAM because it already exists with that name in the pruss headers.
13. you don't need the ProcessingADC1 function. According to the SPRUH73L guide, the channel number will be printed together only if bit 1 is set, which is not the case.
14. you can tune your code to use 16 bits for each sample, but as you are using two buffers it doesn't seem to matter here.

Well those are some of the points that I dedicated for my final work and I decided to share with you as your code reeeally helped to go ahead.

Thank you very much!

Em terça-feira, 8 de abril de 2014 07:15:31 UTC-3, Youngtae Jo escreveu:

[marcelo...@gmail.com]

I was wrong about the number 1, actually after some experiments I noticed that I'm still losing samples, even after setting the ACK_CLKDIV to something that divides the 24MHz by 3. Anyway it is useful to know that you can obtain higher sample rates by setting a lower value to that register.

I'd also like to add that the continuous mode that you set on the step config seems not good. I noticed it after sampling higher frequency waves, however even on low frequency ones there's a noticeable step on the sampling: every 14-15 samples had approximately the same value... I solved that by setting STEPCONFIG it to the default one-shot mode and then the PRU request a sample when it thinks it's time to. To request a sample, re-set the bit on the STEPENABLE register.

My last step is to discover how to solve the lost samples problem...
When I'm sampling at 300kSps I noticed I lose 358 samples every 5005 samples. At 30kSps the proportion was 5 to 500.

[celso maia]

I can use this code to get two ADC channels at the same time?

Em terça-feira, 8 de abril de 2014 07:15:31 UTC-3, Youngtae Jo escreveu:

[zlw...@gmail.com]

I agree, and I found the same data repeat problem using the continuous code, it does not look like a data loss. I do a 3.6 kHz sampling rate and observed that every 7 samples, I can see the value of two adjacent samples are very close to each other (see the figure below). Anyway, would you mind sharing your code with us to solve the problem? Thanks very much in advance.