CPU Configuration Guidelines for vPars

Figure 26 shows the output of the intctl –p command which has an additional field that maps the CPU

ID to the CPU hardware path. This mapping is important for assigning path based CPU’s and for

migrating I/O interrupts. Figure 27 shows the current interrupt assignments stored in the file named

ints after issuing the intctl –s command. This file can be used as input to the intctl –r command to

restore the I/O interrupt assignment after CPU migrations have taken place.

For example, move 6 CPU’s to a real-time processor set of which 2 of the CPU were previously

participating in I/O interrupt processing. The I/O interrupts that were assigned to CPU’s 2 and 3 are

now assigned to CPU’s 0 and 1 (Figure 28). Now move the 6 CPU’s from the real-time processor set

back to the default processor set and restore the I/O interrupt assignments by issuing the intctl –r

command (Figure 29).

Note:

This command will not restore the interrupt configuration unless all the resources in the file are

present at the time of restoration.

Figure 26. Output of the intctl –p command showing mapping of cpu ID to cpu path.