sys/ztimer: adds some aggressive powersaving MACROS#16969
sys/ztimer: adds some aggressive powersaving MACROS#16969kfessel wants to merge 1 commit intoRIOT-OS:masterfrom
Conversation
THIS IS definetly a DRAFT that makes one able to decide weather ztimer should keep running when no timer is active
github-actions
bot
added
Area: sys
|
Good idea! I had a quite similar one while designing a concept for #16327. But I guess we would get a problem, if In my case, I'm using a |
|
I've tried to pick up your idea: #17607 Basically, this PR counts how many users are there for a specific clock and only keeps extending them if there are any users relying on this clock. Otherwise it turns off the peripheral. |
|
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you want me to ignore this issue, please mark it with the "State: don't stale" label. Thank you for your contributions. |
stale
bot
added
the
State: stale
2 participants
that makes one able to decide whether
ztimer should keep running when no timer is active
THIS IS definitely a DRAFT
Contribution description
layerd ztimers like SEC or MSEC on USEC disable the pm_layed power-saving for their base-clock since they require 32Bit while less are left after the conversion -> they will employ the 32Bit Timer extender.
The POWERSAFE_ULTRA define will not keep the clock running if there is no Timer.
If pm_layerd is employed only clock with active timers are definitely working
Testing procedure
test/ztimer_xsec is still working.
CFLAGS=-DPOWERSAFE_ULTRA make flash termIssues/PRs references
#16327
#16891