Re: infinite loop in read_hpet from ktime_get_boot_fast_ns
From: Arnd Bergmann
Date: Wed Jun 12 2019 - 10:06:00 EST
On Wed, Jun 12, 2019 at 11:46 AM Jason A. Donenfeld <Jason@xxxxxxxxx> wrote:
> On Wed, Jun 12, 2019 at 11:03 AM Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> > How quasi? Do the comments in kernel/sched/clock.c look like something
> > you could use?
> > As already mentioned in the other tasks, anything ktime will be
> > horrifically crap when it ends up using the HPET, the code in
> > kernel/sched/clock.c is a best effort to keep using TSC even when it is
> > deemed unusable for timekeeping.
> Thanks for pointing that out. Indeed the HPET path is a bummer and I'd
> like to just escape using ktime all together.
> In fact, my accuracy requirements are very lax. I could probably even
> deal with an inaccuracy as huge as ~200 milliseconds. But what I do
> need is 64-bit, so that it doesn't wrap, allowing me to compare two
> stamps taken a long time apart, and for it to take into account sleep
> time, like CLOCK_BOOTTIME does, which means get_jiffies_64() doesn't
> fit the bill. I was under the impression that I could only get this
> with ktime_get_boot & co, because those add the sleep offset.
Documentation/core-api/timekeeping.rst describes the timekeeping
interfaces. I think what you want here is ktime_get_coarse_boottime().
Note that "coarse" means "don't access the hardware clocksource"
here, which is faster than "fast", but less accurate.
This is updated as often as "jiffies_64", but is in nanosecond resolution
and takes suspended time into account.