Re: [patch 10/10] *Tests* Scheduler profiling - Use immediate values

2007-07-10 Thread Mathieu Desnoyers
Hi, * Li, Tong N ([EMAIL PROTECTED]) wrote: > Mathieu, > > > cycles_per_iter = 0.0; > > for (i=0; i > time1 = get_cycles(); > > for (j = 0; j < NR_ITER; j++) { > > testval = &array[random() % ARRAY_SIZE]; > > } > > time2

RE: [patch 10/10] *Tests* Scheduler profiling - Use immediate values

2007-07-06 Thread Li, Tong N
Mathieu, > cycles_per_iter = 0.0; > for (i=0; i time1 = get_cycles(); > for (j = 0; j < NR_ITER; j++) { > testval = &array[random() % ARRAY_SIZE]; > } > time2 = get_cycles(); > cycles_per_iter +

Re: [patch 10/10] *Tests* Scheduler profiling - Use immediate values

2007-07-06 Thread Mathieu Desnoyers
* Li, Tong N ([EMAIL PROTECTED]) wrote: > > Also cache misses in this situation tend to be much more than 48 > cycles > > (even an K8 with integrated memory controller with fastest DIMMs is > > slower than that) Mathieu probably measured an L2 miss, not a load > from > > RAM. > > Load from RAM can