[Emc-developers] HallMeter wrong time!

2010-03-31 Thread Slavko Kocjancic
Hello... Today I pull my hear from head discoveryng why my EPP IO board is so slow. For measurment's I put 10 out cycles in write_board procedure.. static void write_board(void *arg, long period) { epp_board_t *board; board = arg; rtapi_outb(0, board->base_addr+3); rta

Re: [Emc-developers] HallMeter wrong time!

2010-03-31 Thread Jeff Epler
.time and .tmax (as well as the numbers in halcmd show thread) are in CPU cycles, not ns. Some versions of rtai performed very poorly when requesting the current time, so we switched to using the TSC. Is it possible that your system is 2GHz or so? If you saw a mention of .time and .tmax in the d

Re: [Emc-developers] HallMeter wrong time!

2010-03-31 Thread Slavko Kocjancic
2010/3/31 Jeff Epler > .time and .tmax (as well as the numbers in halcmd show thread) are in > CPU cycles, not ns. Some versions of rtai performed very poorly when > requesting the current time, so we switched to using the TSC. Is it > possible that your system is 2GHz or so? > > If you saw a m

Re: [Emc-developers] HallMeter wrong time!

2010-04-01 Thread Slavko Kocjancic
Slavko Kocjancic pravi: > > > 2010/3/31 Jeff Epler > > > .time and .tmax (as well as the numbers in halcmd show thread) are in > CPU cycles, not ns. Some versions of rtai performed very poorly when > requesting the current time, so we switched to using th