Hi man, 23,.01.2006 kl. 17.43 +0800, skrev Hanzac Chen: > Hi, Nils > > On 1/23/06, Nils Labugt wrote: > > ... > > Index: ata-ops.c > > =================================================================== > > RCS file: /cvsroot/freedos-32/fd32/modules/ata/src/ata-ops.c,v > > retrieving revision 1.17 > > retrieving revision 1.18 > > diff -u -d -r1.17 -r1.18 > > ... > > - delay(400); > > + pit_delay(1); > > > > The specification specifies a minimum 400ns delay. If we get a random > > delay up to a 55ms, then performance will be adversely impacted. > > Oh, I don't know it's a specified value, I just change each previous > delay value from nanoseconds to microseconds. Maybe we should also > provide a nano_delay for more precise delay. Yes, but where to put it? It would of course need to use the PIT (or HPET) for calibration, so maybe it could be put inside the PIT module? > > BTW, can we use cpu idle in the delay's loop? like while (ticks < > when) fd32_cpu_idle(); Yes, that would reduce power consumption while waiting.
Nils ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642 _______________________________________________ freedos-32-dev mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freedos-32-dev
