i386/111591: [Releng_6][i386][acpi_ibm] dev.acpi_ibm.0.events returns I/O error

[Dominic Fandrey]

>Number:         111591
>Category:       i386
>Synopsis:       [Releng_6][i386][acpi_ibm] dev.acpi_ibm.0.events returns I/O error
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-i386
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Sun Apr 15 07:40:01 GMT 2007
>Closed-Date:
>Last-Modified:
>Originator:     Dominic Fandrey
>Release:        Releng_6
>Organization:
private
>Environment:
FreeBSD mobileKamikaze.norad 6.2-STABLE FreeBSD 6.2-STABLE #0: Sun Apr 15 08:29:48 CEST 2007     root at homeKamikaze.norad:/usr/obj/TPR40-6/i386/usr/src/sys/TPR40-6  i386
>Description:
After updating to Releng_6 yesterday (base and kernel) from a previous post 6.2 Releng_6 the sysctl command

# sysctl dev.acpi_ibm.0.events=1

returns the following output:

dev.acpi_ibm.0.events: 0
sysctl: dev.acpi_ibm.0.events: Input/output error

Hence all Fn-KEY bindings do not produce acpi events that I can handle in /etc/devd.conf. This is pretty inconvenient, since I have a script that handles 19 of these events (e.g. preventing suspend, when radeon.ko is loaded or starting a wireless scan on Fn-F5, manual CPU speed control...).

The problem does not exists in Releng_6_2, which I use as a fallback kernel for now.
>How-To-Repeat:
# kldload acpi_ibm
# sysctl dev.acpi_ibm.0.events=1
>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted: