History log of /netbsd/sys/kern/subr_cpufreq.c (Results 1 – 10 of 10)
Revision Date Author Comments
# 416a8a0e 09-Apr-2023 riastradh <riastradh@NetBSD.org>

kern: KASSERT(A && B) -> KASSERT(A); KASSERT(B)


# c3a33ad0 12-Feb-2014 martin <martin@NetBSD.org>

Unlock correct mutex in an error path.
PR kern/48592 from Kengo NAKAHARA.


# a7cb0b8e 27-Oct-2011 jruoho <jruoho@NetBSD.org>

Revert the revision 1.5 in cpufreq(9). Instead, document that the KPI can
not be used before interrupts have been enabled. Suggested by macallan@.


# 9cf37851 25-Oct-2011 christos <christos@NetBSD.org>

back to using timespec, but convert to microseconds to prevent overflow.


# bd90bf7f 25-Oct-2011 jruoho <jruoho@NetBSD.org>

Use microsecond rather than nanosecond resolution when sampling latencies.


# 4eae56d0 20-Oct-2011 jruoho <jruoho@NetBSD.org>

Do not use xcall(9) in cpufreq_register(9), as it requires kthreads to be
running and may impose hacks like config_interrupts(9) in the backend.


# b953eeea 30-Sep-2011 jruoho <jruoho@NetBSD.org>

Remove redundant assignment.


# 74d5612b 30-Sep-2011 jruoho <jruoho@NetBSD.org>

Fix bugs noted by alnsn@. Namely, preallocate the global structure in
order to avoid locking issues during deregistration.


# c8b65da9 28-Sep-2011 jruoho <jruoho@NetBSD.org>

Initialize cpufreq(9) normally from main().


# db122991 28-Sep-2011 jruoho <jruoho@NetBSD.org>

Add a simple cpufreq(9).

This is slightly improved version presented on tech-kern@ with proper
locking and few additional functions, mainly required for ioctl(9)/cpuctl(8).