Interrupts (2)
Lecture 10
Yeongpil Cho
Hanynag University
Topics
• Pending Status
• Faults and Supervisor Call Exceptions
Pending Status
Interrupt Inputs and Pending Behavior
• When an interrupt input is asserted, it can be pended
▪ Interrupt-Set-Pending Registers (ISPRs) hold pending status
• Even if the interrupt source de-asserts the interrupt, the
pended interrupt status will still cause the interrupt hand
ler to be executed.
Interrupt
Request
Processor
Pending Status
Handler Mode
Thread Mode
Processor
Mode
Interrupt Pending
Interrupt Inputs and Pending Behavior
• If the pending status is cleared before the processor
starts responding to the pended interrupt,
▪ the interrupt can be canceled
• Note:
▪ Pending status is automatically cleared when a handler is started.
▪ You can even use pending status to raise software interrupts.
Interrupt
Request
Processor
Pending Status
Pending Status
Cleared by Software
Thread Mode
Processor
Mode
Interrupt Pending Cleared Before Processor Takes Action
Interrupt Inputs and Pending Behavior
• When the processor starts to execute an interrupt
▪ The interrupt becomes active
▪ The pending bit will be cleared automatically.
Interrupt
Request
Interrupt
Pending Status
Interrupt
Active Status Handler Mode Interrupt Returned
Thread
Processor Mode
Mode
Interrupt Active Status Set as Processor Enters Handler
Interrupt Inputs and Pending Behavior
• If an interrupt is pulsed several times before the
processor starts processing it,
▪ it will be treated as one single interrupt request.
Multiple Interrupt pulses
entering before ISR
Interrupt
Request
Interrupt
Pending Status
Interrupt
Active Status
Handler Mode
Thread
Processor Mode
Mode
Interrupt Returned
Interrupt Pending Only Once, Even with Multiple Pulses Before the Handler
Interrupt Inputs and Pending Behavior
• If an interrupt is de-asserted and then pulsed again
during the interrupt service routine,
▪ it will be pended again.
Interrupt request
pulsed again
Interrupt
Request
Interrupt Interrupt
Pending Status pended again
Interrupt
Active Status
Handler Mode
Thread
Mode Interrupt
Processor Interrupt re-
returned
Mode entered
Interrupt Pending Occurs Again During the Handler
Interrupt Inputs and Pending Behavior
• If an interrupt source continues to hold the interrupt
request signal active,
▪ the interrupt will be pended again at the end of the interrupt
service routine.
Interrupt request stays active
Interrupt
Request
Interrupt
Pending Status
Interrupt
Active Status Interrupt returned
Handler Mode
Interrupt re-entered
Thread
Processor Mode
Mode
Continuous Interrupt Request Pends Again After Interrupt Exit
Nested Interrupts
• NVIC supports nested interrupts as its names.
Higher Priority
IRQ1
IRQ2
IRQ3
Base CPU
Time
Core Execution Foreground ISR2 ISR1 ISR2 ISR3 Foreground
(ISR 2 resumes)
Interrupt Response – Tail Chaining
• The processor skips the unstacking and stacking steps
and enters the exception handler of the pended exception
as soon as possible
Interrupt Response – Late Arriving
• An interrupt with a higher priority will be serviced first
even if it arrives late during the stacking operation of the
previous one.
Interrupt Response – Pop Pre-emption
• If an exception arrives during the unstacking process of
another exception, the unstacking would be abandoned
and the next exception service begins
Faults and Supervisor Call Exceptions
Faults and Supervisor Call Exceptions
• Bus Faults
• Memory Management Faults
• Usage Faults
• Hard Faults
• SVC
• PendSV
Bus Faults
• Bus faults are produced when an error response is
received during a transfer on the AHB interfaces.
▪ Prefetch abort (Instruction prefetch)
▪ Data abort (data read/write)
▪ Stacking error
▪ Unstacking error
• Bus fault due to: (refer to Bus Fault Status Registers)
▪ Attempt to access invalid memory region
▪ Attempt to access target devices that are not ready yet
▪ Attempt to access target devices with unsupported transfer sizes
or privilege levels
Memory Management Faults
• Common memory manage faults include:
(refer to Memory management Fault Status Register)
▪ Access to memory regions not defined in MPU setup.
▪ Execute code from nonexecutable memory regions.
▪ Writing to read-only regions.
▪ An access in the user state to a region defined as privileged
access only.
Usage Faults
• Usage faults can be caused by:
(refer to Usage Fault Status Register)
▪ Undefined instructions
▪ Coprocessor instructions
– the Cortex-M3 processor does not support a coprocessor
▪ Trying to switch to the ARM state
– PC is set to a new value with the LSB equal to 0)
▪ Invalid interrupt return
– Link Register contains invalid/incorrect values
▪ Unaligned memory accesses using multiple load/store
▪ Returning to Thread mode with any active interrupts
• It is possible, by setting up certain control bits in the
NVIC, to generate usage faults for:
▪ Divide by zero
▪ Any unaligned memory accesses
Hard Faults
• Occurs when usage faults, bus faults, and memory
management faults cannot be handled.
• Example (refer to Hard Fault Status Registers)
▪ Fail to fetch the vector table due to bus faults
▪ The three faults happen in a handler of other faults with higher
priority.
SVC and PendSV
• SVC (Supervisor Call) and PendSV (Pended Supervisor
Call) are two exceptions targeted at operating systems.
• SVC
▪ generated by SVC instruction
• PendSV
▪ generated by PENDSVSET-bit of Interrupt Control and State
Register (ICSR)
SVC
• SVC can make software more portable because the user
application does not need to know the programming
details of the hardware.
Privileged
Hardware
Kernel
Unprivileged
User Device
Program SVC API Pripherals
Drivers
Operating Ssytem
SVC as a Gateway for OS Functions
PendSV
• PendSV has the lowest privileged level
• Useful for an OS to pend an exception so that an action
can be performed after other important tasks (with
higher priority).
▪ A typical use is context switching.
PendSV
• Assumed that SYSTICK triggers context-switching
Priority Context Context Context
Switching Switching Switching
SYSTICK OS OS OS OS
IRQ
Thread Task A Task B Task A
Time
A Simple Scenario Using SYSTICK to Switch Between Two Tasks
PendSV
• If SYSTICK happens during an interrupt is handled
and context-switching is triggered,
▪ Usage fault will arise. Usage fault: return to
thread with active interrupt
Priority Context Context
Switching Switching
SYSTICK OS OS OS
IRQ IRQ
IRQ
Thread Task A Task B Task A
Time
IRQ processing
delayed
Problems at Context Switching at the IRQ
PendSV
• One solution
▪ OS performs context switching only when none of interrupt
handlers are being executed.
▪ The context-switching can be delayed so long if the interrupts
arise with the frequency close to that of the SYSTICK.
PendSV
• Better solution: using PenSV
▪ Make SYSTICK exceptions do not perform context-switching by
themselves.
– Only pends PenSV exception with a lowest priority
▪ PenSV handler will be executed and performs context-switching
safely after all interrupts with higher priorities handled.
pends PenSV here!
