DZone
Thanks for visiting DZone today,
Edit Profile
  • Manage Email Subscriptions
  • How to Post to DZone
  • Article Submission Guidelines
Sign Out View Profile
  • Post an Article
  • Manage My Drafts
Newsletter
Log In / Join
Refcards Trend Reports
Events Video Library
Refcards
Trend Reports

Events

View Events Video Library

Related

  • Designing AI for Long-Running Tasks: Looking Beyond More Context
  • How Long-Running ECS Tasks Survive Rolling Deployments Without Duplicating Output
  • MCP for Enterprise Tasks: Making the Rare Frequent Enough to Master
  • Idempotent Output Keying for Long-Running Tasks During Rolling Deployments

Trending

  • Build Production-Ready AI Phone Agents: Infrastructure, Models, and Telephony
  • 5 Mistakes Companies Make When Working With Staff Engineers
  • Engineering Data Observability: Detecting Schema Drift, Data Drift, and Silent Pipeline Failures
  • Stop Writing Database Migration Scripts: Declarative Schema Sync for Java

Draft This article hasn't been published, so readers can't see it.

In moderation This article is waiting for review, so readers can't see it yet.

Deleted This article has been deleted, so readers can't see it.

Understanding FreeRTOS Task Stack Usage and Kernel Awareness Info

Diving into FreeRTOS' Task Lisk in MCUXPresso can be confusing at first. Let's look into how to examine and understand how your stack is being used.

By 
Erich Styger
·
May. 25, 18 · Tutorial
Post on X (opens in a new tab) Share on LinkedIn (opens in a new tab)
21.2K Views

Join the DZone community Save articles, join the discussion, and download Refcards and Trend Reports with a free account.

Join for free Already a member? Log in

When using the FreeRTOS Task List in the Eclipse-based MCUXpresso IDE, it shows the list of tasks with their stack size used. But with the default FreeRTOS settings, it is not able to determine the correct stack size and shows a warning icon:

Warning for Stack Usage

Warning for stack usage

The warning should tell me that the information shown about the stack size is wrong: I did create the App Task with 110 stack units (each 4 bytes) so it should show 440 bytes, but it shows only 308 bytes.

Why is that? The reason is that the debugger (or FreeRTOS Kernel Awareness in Eclipse) does not know the total size of the memory allocated for the task.

On ARM Cortex, the stack is growing from the higher to the lower addresses, as shown in the image below:

FreeRTOS Task Stack Layout

FreeRTOS task stack layout

What FreeRTOS stores in its TCB (Task Control Block, the descriptor of the task) is the pxTopOfStack (current task stack pointer, Stack Top in the Kernel Awareness view) and pxStack (the end of the stack, Stack Base in the Kernel Awareness view). The beginning of the task stack is not stored in the TCB as not necessary for the Kernel.

The ‘Stack High Water Mark’ is something very useful: FreeRTOS fills the stack space at task creation time with a 0x5a byte pattern. The debugger or the RTOS can find out the ‘high water mark’ for each task that way.

The solution is to add the following macro to the FreeRTOSConfig.h file:

#define configRECORD_STACK_HIGH_ADDRESS (1)


With this turned on, FreeRTOS stores an extra pointer pxEndOfStack in each Task TCB:

FreeRTOS Task Stack Information with configRECORD_STACK_HIGH_ADDRESS

FreeRTOS task stack information with configRECORD_STACK_HIGH_ADDRESS

Below is such a TCB in the Global Variables View in Eclipse (MCUXpresso IDE):

Stack Information in FreeRTOS TCB

Stack information of FreeRTOS TCB

With this, the debugger or FreeRTOS Task Awareness knows the real size of the stack and can show the correct information:

FreeRTOS Stack Size

FreeRTOS stack size

Actually, it is still showing 432 bytes because it does not count the first and last entry on the stack.

The image below should help you understand the information provided:

  • Blue: At the bottom, there is the currently used stack. The stack is growing upwards here (from the higher to the lower addresses).
  • Orange: Used stack so far (high water mark) as it has overwritten the 0xA5 pattern.
  • Yellow: Unused stack space so far with the 0xA5 pattern.
FreeRTOS Stack Memory

FreeRTOS stack memory

Of course, there is a reason for FreeRTOS not storing that pxEndOfStack by default: It is not needed by the Kernel. So enabling configRECORD_STACK_HIGH_ADDRESS will need 4 more bytes of RAM for each task, plus a few code bytes to set that value. But for debugging purposes, this is very well spent.

Happy stacking!

Links

  • Better FreeRTOS Debugging in Eclipse
  • FreeRTOS Kernel Awareness for Eclipse from NXP
FreeRTOS Task (computing) Kernel (operating system)

Published at DZone with permission of Erich Styger. See the original article here.

Opinions expressed by DZone contributors are their own.

Related

  • Designing AI for Long-Running Tasks: Looking Beyond More Context
  • How Long-Running ECS Tasks Survive Rolling Deployments Without Duplicating Output
  • MCP for Enterprise Tasks: Making the Rare Frequent Enough to Master
  • Idempotent Output Keying for Long-Running Tasks During Rolling Deployments

Partner Resources

×

Comments

Loading comments

The comments didn't load

Please try again in a moment.

Likes

Loading

The next likes didn't load. Please try again.

The likes didn't load

Please try again in a moment.

See who liked this

Log in or join DZone for free to see who liked this article.

Log in Join for free
  • RSS
  • X
  • Facebook

ABOUT US

  • About DZone
  • Support and feedback
  • Community research

ADVERTISE

  • Advertise with DZone

CONTRIBUTE ON DZONE

  • Article Submission Guidelines
  • Become a Contributor
  • Core Program
  • Visit the Writers' Zone

LEGAL

  • Terms of Service
  • Privacy Policy

CONTACT US

  • 3343 Perimeter Hill Drive
  • Suite 215
  • Nashville, TN 37211
  • [email protected]

Let's be friends:

  • RSS
  • X
  • Facebook