Producer-consumer pattern in FreeRTOS - Mini Project: Build & Apply

Producer-Consumer Pattern with FreeRTOS

📖 Scenario: You are building a simple embedded system using FreeRTOS where one task produces data and another task consumes it. This pattern helps manage data flow safely between tasks.

🎯 Goal: Create two FreeRTOS tasks: a producer task that sends numbers to a queue, and a consumer task that receives and prints those numbers.

📋 What You'll Learn

Create a FreeRTOS queue to hold integers

Create a producer task that sends numbers 1 to 5 to the queue

Create a consumer task that receives numbers from the queue

Print each received number in the consumer task

💡 Why This Matters

🌍 Real World

Producer-consumer patterns are common in embedded systems to manage data flow between sensors and processing tasks without data loss.

💼 Career

Understanding FreeRTOS task communication is essential for embedded software engineers working on real-time applications.

1

Create a FreeRTOS queue

Create a FreeRTOS queue called queue that can hold 5 integers using xQueueCreate(5, sizeof(int)).

FreeRTOS

#include "FreeRTOS.h"
#include "task.h"
#include "queue.h"

QueueHandle_t queue;

int main(void) {
    // Create the queue here
    // Your code here
    return 0;
}

Hint

Use xQueueCreate with 5 as the number of items and sizeof(int) as the size of each item.

2

Create the producer task

Create a FreeRTOS task called producer using xTaskCreate. The producer task should send numbers 1 to 5 to the queue using xQueueSend inside a loop.

FreeRTOS

#include "FreeRTOS.h"
#include "task.h"
#include "queue.h"

QueueHandle_t queue;

void producer(void *params) {
    for (int i = 1; i <= 5; i++) {
        xQueueSend(queue, &i, portMAX_DELAY);
    }
    vTaskDelete(NULL);
}

int main(void) {
    queue = xQueueCreate(5, sizeof(int));
    // Create the producer task here
    // Your code here
    return 0;
}

Hint

Use xTaskCreate with the producer function, a name, stack size 1000, no parameters, priority 1, and no task handle.

3

Create the consumer task

Create a FreeRTOS task called consumer using xTaskCreate. The consumer task should receive numbers from the queue using xQueueReceive inside a loop and print each number using printf.

FreeRTOS

#include "FreeRTOS.h"
#include "task.h"
#include "queue.h"
#include 

QueueHandle_t queue;

void producer(void *params) {
    for (int i = 1; i <= 5; i++) {
        xQueueSend(queue, &i, portMAX_DELAY);
    }
    vTaskDelete(NULL);
}

void consumer(void *params) {
    int value;
    for (int i = 0; i < 5; i++) {
        xQueueReceive(queue, &value, portMAX_DELAY);
        printf("Received: %d\n", value);
    }
    vTaskDelete(NULL);
}

int main(void) {
    queue = xQueueCreate(5, sizeof(int));
    xTaskCreate(producer, "Producer", 1000, NULL, 1, NULL);
    // Create the consumer task here
    // Your code here
    return 0;
}

Hint

Use xTaskCreate to create the consumer task and then start the scheduler with vTaskStartScheduler().

4

Run and observe the output

Run the program and observe the printed output from the consumer task. The output should show the numbers 1 to 5 received from the producer task.

FreeRTOS

Hint

Check that the consumer prints all numbers from 1 to 5 in order.

Practice

1.

In FreeRTOS, what is the main role of the queue in the producer-consumer pattern?

easy

A. To create new tasks dynamically

B. To handle interrupts directly

C. To manage memory allocation for tasks

D. To safely transfer data between producer and consumer tasks

Solution

1. Step 1: Understand the producer-consumer pattern

   The pattern involves one task producing data and another consuming it, requiring safe data sharing.

2. Step 2: Identify the role of the queue in FreeRTOS

   The queue acts as a safe buffer that holds data items, ensuring tasks do not access the same data simultaneously.

3. Final Answer:

   To safely transfer data between producer and consumer tasks -> Option D

4. Quick Check:

   Queue = Safe data transfer [OK]

Hint: Queues safely pass data between tasks without conflicts [OK]

Common Mistakes:

• Thinking queues create tasks
• Confusing queues with memory management
• Assuming queues handle interrupts

2.

Which of the following is the correct FreeRTOS API call for a producer task to send data to a queue?

void producerTask(void *params) {
    int data = 10;
    // Send data to queue here
}

easy

A. xQueueSend(queueHandle, &data, portMAX_DELAY);

B. xQueueReceive(queueHandle, &data, portMAX_DELAY);

C. xTaskCreate(queueHandle, &data, portMAX_DELAY);

D. vTaskDelay(queueHandle);

Solution

1. Step 1: Identify the correct API for sending data to a queue

   FreeRTOS uses xQueueSend() to send data from a producer to a queue.

2. Step 2: Check other options for correctness

   xQueueReceive() is for receiving data, xTaskCreate() creates tasks, and vTaskDelay() delays tasks, so they are incorrect here.

3. Final Answer:

   xQueueSend(queueHandle, &data, portMAX_DELAY); -> Option A

4. Quick Check:

   Send data = xQueueSend [OK]

Hint: Use xQueueSend to put data into the queue [OK]

Common Mistakes:

• Using xQueueReceive to send data
• Confusing task creation with queue operations
• Using delay functions instead of queue functions

3.

Consider this FreeRTOS code snippet where a producer sends integers 1 to 3 to a queue and a consumer receives them:

int data;
for (int i = 1; i <= 3; i++) {
    xQueueSend(queueHandle, &i, 0);
}
for (int j = 0; j < 3; j++) {
    xQueueReceive(queueHandle, &data, 0);
    printf("%d ", data);
}

What will be the output?

medium

A. 1 2 3

B. 3 2 1

C. 0 0 0

D. Compilation error

Solution

1. Step 1: Analyze the producer loop

   The producer sends values 1, 2, and 3 in order to the queue without delay.

2. Step 2: Analyze the consumer loop

   The consumer receives three values from the queue and prints them in the order received.

3. Final Answer:

   1 2 3 -> Option A

4. Quick Check:

   Queue preserves order = 1 2 3 [OK]

Hint: Queue keeps order; output matches sent sequence [OK]

Common Mistakes:

• Assuming queue reverses order
• Expecting zeros due to uninitialized data
• Thinking code won't compile

4.

Identify the bug in this FreeRTOS consumer task code that causes it to never receive data:

void consumerTask(void *params) {
    int receivedData;
    while(1) {
        if(xQueueReceive(queueHandle, &receivedData, 0) == pdFALSE) {
            // No data received
            continue;
        }
        processData(receivedData);
    }
}

medium

A. The loop should use vTaskDelay instead of continue

B. The receive timeout is zero, causing missed data

C. The processData function is missing

D. The queue handle is not initialized

Solution

1. Step 1: Understand xQueueReceive timeout parameter

   A zero timeout means the call returns immediately if no data is available, possibly missing data if producer is slower.

2. Step 2: Identify why data is never received

   Because the consumer never waits, it may loop too fast and miss data arriving slightly later.

3. Final Answer:

   The receive timeout is zero, causing missed data -> Option B

4. Quick Check:

   Timeout 0 means no wait = missed data [OK]

Hint: Use non-zero timeout to wait for data [OK]

Common Mistakes:

• Ignoring timeout effect on receive
• Assuming missing queue handle causes this
• Thinking processData absence causes no receive

5.

You want to implement a producer-consumer system where the producer sends sensor readings every 100ms, and the consumer processes them. Which FreeRTOS queue configuration and task design is best to avoid data loss and ensure timely processing?

hard

A. Use a queue with length 5 and consumer uses zero timeout on receive

B. Use a queue with length 1 and zero receive timeout in consumer

C. Use a queue with length 10 and consumer waits with portMAX_DELAY on receive

D. Use no queue and share a global variable with no synchronization

Solution

1. Step 1: Consider queue length for buffering

   A queue length of 10 allows buffering multiple sensor readings if consumer is busy, preventing data loss.

2. Step 2: Consider consumer receive timeout

   Using portMAX_DELAY makes consumer wait indefinitely for data, ensuring it processes every reading.

3. Final Answer:

   Use a queue with length 10 and consumer waits with portMAX_DELAY on receive -> Option C

4. Quick Check:

   Buffer + wait = no data loss [OK]

Hint: Use bigger queue and wait on receive to avoid loss [OK]

Common Mistakes:

• Using too small queue causing data overwrite
• Using zero timeout causing missed data
• Sharing data without synchronization