Interrupts
MPC5510 Microcontroller Family Reference Manual, Rev. 1
Freescale Semiconductor
8-19
Preliminary
8.4
Interrupt Operation
8.4.1
Software Vector Mode
In software vector mode, the interrupt exception handler acknowledges the interrupt request to the
e200z1/0 from the INTC by reading the INTC_IACKR. The e200z1/0 is enabled again to recognize the
external input by setting the EE bit of the MSR. The prolog of the interrupt exception handler must
acknowledge the interrupt request before the e200z1/0 is enabled again to recognize the external input.
Otherwise, the e200z1/0 will attempt to service the same source of the interrupt request.
The INTC’s LIFO is popped by writing to the INTC_EOIR. The e200z1/0’s recognition of the external
input is disabled by clearing the EE bit of the MSR. In the epilog of the interrupt exception handler, the
timing relationship between popping the LIFO and disabling recognition of the external input has no
restrictions. The writes can happen in either order. However, disabling recognition of the external input
before popping the LIFO eases the calculation of the maximum pipe depth at the cost of postponing the
servicing of the next interrupt request.
8.4.2
Hardware Vector Mode
In hardware vector mode, the interrupt request to the e200z1/0 from the INTC is acknowledged before the
e200z1/0 starts to execute the exception handler. The INTC_IACKR does not need to be read to
acknowledge the interrupt request before the e200z1/0 is enabled again to recognize the external input.
As in software vector mode, the timing relationship between popping the LIFO and disabling recognition
of the external input has no restrictions. Also, as in software vector mode, disabling recognition of the
external input before popping the LIFO eases the calculation of the maximum pipe depth at the cost of
postponing the servicing of the next interrupt request.
8.4.3
Non Maskable Interrupt (NMI)
The MPC5510 can be configured to use the pins PD[10] and PD[11] as non maskable interrupts (NMI) by
providing a path to the critical interrupt input of the e200Z1 and e200z0 cores, respectively.
After the SIU is configured by user code, an NMI cannot be prevented from reaching the assigned core.
The only possible way of disabling the critical interrupt is by clearing the critical interrupt enable (CE) bit
in the core’s machine state register (MSR). The NMI will have a higher priority than any interrupt request
generated by the INTC, and will not be blocked or preempted by any other INTC interrupt request.
After the SIU is properly configured, the operation of the NMI always generates an interrupt request when
the programmed edge transition occurs on the pin, regardless of the selected muxing on that pin. It is the
user’s responsibility to assign pin multiplexing correctly for use with an NMI, which would normally mean
selecting it as a port pin rather than a peripheral function.
shows the various system level connections needed to create the NMI.