PR/60735 CVS commit: src/sys/arch/x86/isa

["Taylor R Campbell" <riastradh%netbsd.org@localhost>]

The following reply was made to PR kern/60735; it has been noted by GNATS.

From: "Taylor R Campbell" <riastradh%netbsd.org@localhost>
To: gnats-bugs%gnats.NetBSD.org@localhost
Cc: 
Subject: PR/60735 CVS commit: src/sys/arch/x86/isa
Date: Wed, 16 Sep 2026 21:57:38 +0000

 Module Name:	src
 Committed By:	riastradh
 Date:		Wed Sep 16 21:57:38 UTC 2026
 
 Modified Files:
 	src/sys/arch/x86/isa: clock.c
 
 Log Message:
 x86/sysbeep(4): Don't detach at shutdown.
 
 This reverts sys/arch/x86/isa/clock.c rev. 1.32, which was the first
 of a spate of changes to sprinkle DVF_DETACH_SHUTDOWN for reasons
 that were unclear and apparently interfere with other functionality:
 
 > Module Name:    src
 > Committed By:   dyoung
 > Date:           Tue Apr  7 17:51:46 UTC 2009
 >
 > Modified Files:
 >         src/sys/arch/x86/isa: clock.c
 >
 > Log Message:
 > Detach sysbeep0 at shutdown.
 >
 >
 > To generate a diff of this commit:
 > cvs rdiff -u -r1.31 -r1.32 src/sys/arch/x86/isa/clock.c
 
 https://mail-index.netbsd.org/source-changes/2009/04/07/msg219296.html
 
 The only reason to use DVF_DETACH_SHUTDOWN would be if:
 
 (a) the driver actually has to flush some stored state or release
     some hardware resources, and
 
 (b) a pmf shutdown hook can't do it or would do essentially the same
     as detaching anyway (e.g., it involves forcibly detaching its
     children).
 
 And if there is such a reason, it should be explained!
 
 Otherwise, requiring detach at shutdown is at best a waste of energy
 and may break other things too.
 
 PR kern/60735: BEEP_ONHALT fails on i386
 
 
 To generate a diff of this commit:
 cvs rdiff -u -r1.43 -r1.44 src/sys/arch/x86/isa/clock.c
 
 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.