xiaoxiang781216 commented on PR #14450:
URL: https://github.com/apache/nuttx/pull/14450#issuecomment-2906883844
> ## Summary
>
> Currently up_ndelay & up_udelay are completely broken; they always sleep
very long time (at least to the end of currently ongoing tick), due to the
current time is obtained from "oneshot_current":
>
> ```
> ret = lower->ops->tick_current(lower, &tick);
> clock_ticks2time(ts, tick);
> ```
>
@jlaitine, but up_ndelay & up_udelay call ONESHOT_CURRENT:
```
#define ONESHOT_CURRENT(l,t) \
((l)->ops->current ? (l)->ops->current(l,t) : oneshot_current(l,t))
```
which try to call current before tick_current, so do you implement current
callback and what's your hardware time frequency?
> This basically quantizes the time with tick time, causing really long
looping
>
> This PR also sets a proper CONFIG_BOARD_LOOPSPERMSEC value for imx93-evk
target.
> ## Impact
>
This change break all devices which doesn't fine tune
CONFIG_BOARD_LOOPSPERMSEC, but depend on ndelay_accurate.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@nuttx.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
