Re: kern/53016: Clock not stable

[Tero Kivinen <kivinen%iki.fi@localhost>]

The following reply was made to PR kern/53016; it has been noted by GNATS.

From: Tero Kivinen <kivinen%iki.fi@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: kern-bug-people%netbsd.org@localhost, gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost
Subject: Re: kern/53016: Clock not stable
Date: Tue, 6 Mar 2018 21:40:43 +0200

 coypu%sdf.org@localhost writes:
 >  A difference is that the drifting first machine's dmesg says:
 >  timecounter: Timecounter "clockinterrupt" frequency 100 Hz quality 0
 >  ERROR: 4316413966 cycle TSC drift observed
 
 Nothing like that during boot. Dmesg output is of course completely
 usable, as it is full of "cprng 461 94681 15: creating with partial
 entropy" lines, so cannot know if it has happened after boot:
 
 ~>fgrep timecount /var/run/dmesg.boot
 timecounter: Timecounters tick every 10.000 msec
 timecounter: Timecounter "i8254" frequency 1193182 Hz quality 100
 timecounter: Timecounter "ACPI-Fast" frequency 3579545 Hz quality 1000
 timecounter: Timecounter "hpet0" frequency 14318180 Hz quality 2000
 timecounter: Timecounter "ichlpcib0" frequency 3579545 Hz quality 1000
 timecounter: Timecounter "clockinterrupt" frequency 100 Hz quality 0
 timecounter: Timecounter "TSC" frequency 2400117960 Hz quality 3000
 
 >  As for this inducing panics, someone mentioned we have some non-ideal
 >  behaviour if a clock interrupt is missed (something like: we try to
 >  catch up on missed clock interrupts? I wonder if this blows up with very
 >  many missed clock interrupts)
 
 Btw, I updated the machine to netbsd 8 from 2018-02-27, and no changes
 in the clock drift, but no crash yet (but it has been only up for 7
 days since, so cannot say anything yet):
 
 NetBSD vielako.iki.fi 8.0_BETA NetBSD 8.0_BETA (GENERIC) #0: Tue Feb 27 03:12:27 EET 2018 kivinen%vielako.iki.fi@localhost:/usr/obj/sys/arch/amd64/compile/GENERIC amd64
 -- 
 kivinen%iki.fi@localhost