I'm designing a sampling profiler, that gets the value of instruction pointer at each interrupt of a real time clock. It is working fine for single core system, but I don't know how interrupts are handled in multicores. What is the technique used by Gprof for sampling? What if I use the timer interrupt instead of real time clock interrupt? Will it work with multi-cores? How is the timer interrupt handled in multicores? Is it possible to read the value of EIP from current process control block instead of reading the eip register at an interrupt? I will be really grateful for any help in this regard.