jlaitine commented on code in PR #15929:
URL: https://github.com/apache/nuttx/pull/15929#discussion_r1979137563
##########
arch/risc-v/src/common/riscv_mtimer.c:
##########
@@ -349,11 +345,11 @@ riscv_mtimer_initialize(uintreg_t mtime, uintreg_t
mtimecmp,
priv = kmm_zalloc(sizeof(*priv));
if (priv != NULL)
{
- priv->lower.ops = &g_riscv_mtimer_ops;
- priv->mtime = mtime;
- priv->mtimecmp = mtimecmp;
- priv->freq = freq;
- priv->alarm = UINT64_MAX;
+ priv->lower.ops = &g_riscv_mtimer_ops;
+ priv->mtime = mtime;
+ priv->mtimecmp = mtimecmp;
+ priv->cycle_per_tick = freq / TICK_PER_SEC;
Review Comment:
I must add that I don't understand the motivation behind
https://github.com/apache/nuttx/commit/eac99370e581
37bc444fe68a86ed5088c935a74d
Each OS tick *must* be equally long; so if the mtime frequency is not evenly
divisible with the requested cycle time, then you don't just get exactly the
cycle time you requested. There must not be variation in the cycle time.
Then totally different thing is, that if you want to do e.g. sleep for a
longer period, and the single tick time is a tad shorter than requested due to
the timer inaccuracy / rounding error, you need to calculate the amount of
ticks properly taken account the realized tick time (xSEC2TICK macros).
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
