Re: kern/59339 (heartbeat watchdog fires since 10.99.14)

["Taylor R Campbell via gnats" <gnats-admin%NetBSD.org@localhost>]

The following reply was made to PR kern/59339; it has been noted by GNATS.

From: Taylor R Campbell <riastradh%NetBSD.org@localhost>
To: prlw1%NetBSD.org@localhost, prlw1%cam.ac.uk@localhost, wiz%NetBSD.org@localhost
Cc: gnats-bugs%NetBSD.org@localhost, netbsd-bugs%NetBSD.org@localhost
Subject: Re: kern/59339 (heartbeat watchdog fires since 10.99.14)
Date: Sun, 5 Oct 2025 14:03:46 +0000

 > Date: Sat, 21 Jun 2025 07:41:57 +0000 (UTC)
 > From: prlw1%NetBSD.org@localhost
 >=20
 > positive feedback provided. OK to commit patch?
 
 Those of you who have been running with the latest patch (v6): have
 you seen any of the prints in dmesg?
 
 They would look something like this (with different numbers):
 
 PR kern/59339 (heartbeat watchdog fires since 10.99.14) would fire: [clock =
 3] [now=3D123.123450000 -> 123.123460000], last=3D123.123445000 + interval=
 =3D0.000010000 -> next=3D123.123455000
 
 Or (extremely unlikely unless you're fiddling with your clock to set
 it forward by centuries):
 
 [clock 0] itimer arithmetic overflowed: now=3D9223372036854775798.000000000=
 , last=3D123.456789123 + interval=3D20.000000000
 
 I think the part of the change that avoids these conditions is worth
 keeping anyway but I'm curious to see if it is happening in practice.