C++ Performance Tuning Notes

perf.md

In order to gen call graph in perf:

-fno-omit-frame-butter

#Generate perf.data for executable perf record -g executable #0.5 is a filter?, caller is inverse the graph callstack perf report -g 'graph,0.5,caller'

Prevent optimizer to remove code:

static void escape(void *p){
  asm volatile("" : : "g"(p) : "memory");
}

static void clobber(){
  asm volatile("" : : : "memory");
}