NetBSD Problem Report #55694

From  Sun Oct  4 19:16:00 2020
Return-Path: <>
Received: from ( [])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "", Issuer " CA" (not verified))
	by (Postfix) with ESMTPS id 8AEA91A9217
	for <>; Sun,  4 Oct 2020 19:16:00 +0000 (UTC)
Message-Id: <>
Date: Sun,  4 Oct 2020 19:15:59 +0000 (UTC)
Subject: "TSC time went backwards" spam on resume on amd64 (regression in -current)
X-Send-Pr-Version: www-1.0

>Number:         55694
>Category:       port-amd64
>Synopsis:       "TSC time went backwards" spam on resume on amd64 (regression in -current)
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    port-amd64-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Oct 04 19:20:00 +0000 2020
>Originator:     nia
>Release:        NetBSD -current
NetBSD r 9.99.73 NetBSD 9.99.73 (GENERIC) #15: Sun Oct  4 20:17:57 CEST 2020  nia@r:/home/nia/src/sys/arch/amd64/compile/obj/GENERIC amd64
After suspending, for several seconds after resume, the console gets spammed with "TSC time went backwards" on amd64. This is very agitating, clutters the console, and doesn't happen on netbsd-9. As a user, I have no idea what to do with this information other than tell the kernel to avoid using TSC.

I can observe this on several different amd64 laptops running -current.
sysctl -w hw.acpi.sleep.state=3. Wake up. Watch for several seconds as the console is filled with TSC messages.
We should probably not be relying on TSC on anything that is capable of modern power management, since it's unreliable in this case.

Otherwise, the message should only be printed once.

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.46 2020/01/03 16:35:01 leot Exp $
$NetBSD:,v 1.9 2014/08/02 14:16:04 spz Exp $
Copyright © 1994-2020 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.