Skip to content
Related Articles
Get the best out of our app
Open App

Related Articles

What is Interrupt Latency?

Improve Article
Save Article
Like Article
Improve Article
Save Article
Like Article

Interrupt latency is a measure of the time it takes for a computer system to respond to an external event, such as a hardware interrupt or software exception. This metric is important in determining the performance and responsiveness of a system and is a key consideration in the design and optimization of real-time and embedded systems. In this article, we will discuss the concept of interrupt latency, its importance, and some of the factors that can affect interrupt latency.

The article focuses on discussing Interrupt latency in detail. The following topics will be discussed here:

  1. What is Interrupt?
  2. What is Interrupt Latency?
  3. Cause of Interrupt Latency
  4. Arm Cortex-M Interrupt Latency
  5. Measuring Interrupt Latency
  6. Conclusion

Let’s start discussing each of these topics in detail.

What is Interrupt?

An interrupt is a signal sent to the computer’s processor asking it to stop what it’s doing and start handling the interrupt right away. Devices like the keyboard, mouse and network card communicate with the processor and request services using interrupts. They are crucial to how modern operating systems function and enable the processor to react quickly to outside events. The interrupt can be emitted by hardware or software indicating an event that needs immediate attention.

Types of Interrupts

1. Software Interrupt

A sort of interrupt that is produced by the software is a software interrupt, sometimes referred to as a trap or a system call. It is used to ask the operating system or other system software to perform a certain service or activity. A program normally creates a software interrupt by utilizing a particular instruction, such as INT in x86 assembly language or a system call in a high-level programming language. When the processor comes across this instruction, it will store the program’s current state and transfer control to a specific location in memory where the system software or operating system can handle the request. Allocating memory, reading from or writing to a file, starting or stopping a process, and sending or receiving data from a network are a few services that can be requested using a software interrupt. In order to deliver services to user-space applications, system software like device drivers, libraries, and operating systems also employs software interrupts.

2. Hardware Interrupt

An external device, such as a keyboard or mouse, will signal the CPU with a hardware interrupt when it needs the processor’s attention. An interrupt controller is a piece of specialized hardware that produces these interruptions by receiving signals from the devices and sending them to the processor. When a hardware interrupt occurs, the processor saves the program’s current state and moves control to an area of memory known as an interrupt vector so that the proper interrupt handler can be run there. A short piece of code called an interrupt handler is created to support the particular device that caused the interrupt.Input/output processes such as receiving data from a keyboard, transferring data to a printer, or interacting with a network are all handled via hardware interrupts. They increase the system’s overall efficiency by enabling the processor to carry out other tasks while it waits for input or output operations to finish.

3. Maskable Interrupts

Interrupts that can be temporarily disabled or “masked” by the processor are known as maskable interrupts.

4. Non-maskable Interrupts

These must be handled right away because the processor cannot ignore them or mask them.

5. Inter-processor Interrupts

In a multiprocessor system, interprocessor interruptions are used to facilitate communication between different processors.

6. Inter-System Interrupts

Inter-system interrupts: These are used to alert other systems to an incident that needs their attention.

7. Exception

Similar to interrupts, exceptions are created by the processor when particular criteria are met, such as division by zero, an invalid command, etc.

What is Interrupt latency?

Interrupt latency is the time that elapses between the occurrence of an interrupt and the execution of the first instruction of the interrupt service routine (ISR) that handles the interrupt. 

  • It is a measure of the system’s ability to respond to external events in a timely manner. 
  • The shorter the interrupt latency, the more responsive the system will be. 
  • Interrupt latency is expressed in core clock cycles. 

Importance of Interrupt Latency

Interrupt latency is an important consideration in the design and optimization of real-time and embedded systems. These systems often have hard real-time constraints, meaning that they must respond to external events within a specified time period. 

  • A high interrupt latency can result in the system missing a deadline, which can cause the system to fail or produce incorrect results.
  • If interrupts are not handled in a timely manner then the system will exhibit slow response times.

Factors Affecting Interrupt Latency

There are several factors that can affect interrupt latency, including:

  • Hardware: The architecture of the processor, as well as the design of the memory and I/O subsystems, can have a significant impact on interrupt latency.
  • Operating System: The operating system plays a big role in interrupt latency, with some OSes having fewer interrupt latencies due to advanced scheduler and interrupt handling mechanism.
  • Interrupt Priority: The priority level of the interrupt can also affect interrupt latency, with higher priority interrupts being serviced before lower priority interrupts.
  • Interrupt Storms: High number of interrupts from certain devices or software can lead to Interrupt Storm which can cause large interrupt latency and system hang
  • Interrupt Handlers: Interrupt handlers themselves can add to the latency of the system. The ISR should be optimized for speed and should avoid unnecessary operations.
  • System load: High system load can also increase interrupt latency, as the CPU is busy processing other tasks and may not be able to respond to interrupts in a timely manner.
  • Disabling interrupts: The system will not react to any interrupt requests if the software has disabled interrupts.

Without examining the system, it can be difficult to determine the precise cause of interrupt latency, but by understanding the potential causes, the problem may be found and fixed.

Interrupt Prioritization

When a processor receives a higher-priority interrupt while it is still addressing a lower-priority interrupt, this is known as interrupt preemption. In this scenario, the processor will instantly halt the handler for the current interrupt and start the handler for the interrupt with a higher priority.

This is often referred to as “interrupt nesting” or “interrupt prioritization.” It is a technique for managing various interrupt requests in the order of their importance, making sure that the most crucial tasks are attended to first.

When a disc drive interrupt with a higher priority is received while a keyboard interrupt is still being treated, for instance, the processor will instantly cease handling the keyboard interrupt and begin handling the disc drive interrupt. The processor will return to the keyboard interrupt and go on from there until the disc drive interrupt has been resolved.

It’s also important to keep in mind that some interrupt controllers include priority levels, which allow the controller to manage an interrupt queue internally by prioritizing and handling interrupts in the order of priority.

Arm Cortex-M Interrupts Latency

The ARM Cortex-M microcontroller series has a low interrupt latency and is frequently used in embedded systems. The nested vectored interrupt controller (NVIC) on Cortex-M processors enables effective interrupt handling. Between the time the interrupt occurs and the time the processor starts executing the interrupt service routine, there are only a few instructions in the NVIC (ISR).

The following factors can still have an impact on the Cortex-M processors’ interrupt latency:

  • Priority of interruptions: The NVIC prioritizes interruptions, so an interruption with a higher priority will take precedence over an interruption with a lower priority.
  • Interrupt nesting: Cortex-M processors are capable of interrupting nesting, which allows one interrupt to be interrupted by a different interrupt. If the nested interrupt has a higher priority, this may result in an increase in latency.
  • Interrupt response time: Reduced interrupt response time is a built-in feature of some Cortex-M processors, such as the “tail-chaining” feature of the Cortex-M4.

Furthermore, code within the ISR may be a factor in interrupt latency; if the ISR code is not optimized, it may add additional latency by taking longer to complete.

Overall, the Cortex-M processors are optimized for real-time embedded systems and have low interrupt latency. However, careful consideration of interrupt priorities and ISR optimization can help to reduce interrupt latency even further.

Measuring Interrupt Latency

Interrupt latency can be measured using various techniques, such as the use of specialized hardware, software tools, and benchmarks. It’s also possible to use different system performance monitoring tools to measure interrupt latency. It’s important to note that the way of measuring can affect the results as well, some tools or methods might measure only the time from the interrupt request to the ISR start while others may include other events as well such as disabling interrupt or switch process, etc. There are three types of latency that are measured:

  • The latency from the timer interrupts the ISR execution.
  • The latency from the timer interrupts the RTOS highest priority task.
  • The latency of the external GPIO interrupts response.


Interrupt latency is a measure of the time it takes for a computer system to respond to an external event. It is an important metric in determining the performance and responsiveness of a system, particularly in real-time and embedded systems. Factors such as hardware, operating system, interrupt priority, and system load can all affect interrupt latency. Measuring interrupt latency can be done using various techniques and tools. It is important for system designers and developers to understand and optimize interrupt latency in order to ensure the reliability and performance of their systems.

My Personal Notes arrow_drop_up
Last Updated : 02 Feb, 2023
Like Article
Save Article
Similar Reads
Related Tutorials