Acpi problems with HP 6501b

Dominic Fandrey LoN_Kamikaze at
Wed Feb 13 10:49:59 UTC 2008

Matthieu Michaud wrote:
> Dominic Fandrey a écrit :
>> Apart from there not being a specific acpi module for HP notebooks my 
>> new HP Compaq 6510b GR695EA#ABD has some other problems. First of all 
>> there seems to be a fault in the ACPI-BIOS, for it defines a CRT of 
>> 256°C for tz0:
>> Feb  9 21:46:37 mobileKamikaze kernel: acpi_tz0: _CRT value is absurd, 
>> ignored (256.0C)
>> This message is quite uselessly posted every couple of seconds. I fear 
>> this asks for one more addition to the quirks, for I doubt the vendor 
>> will come up with a BIOS update that fixes this behaviour.
>> The next problem is that acpiconf -s # doesn't work:
>> # acpiconf -s 3                                      0 /root
>> acpiconf: request sleep type (3) failed: Operation not supported
>> # acpiconf -s 4                                     74 /root
>> acpiconf: request sleep type (4) failed: Operation not supported
>> # acpiconf -s 5                                     74 /root
>> acpiconf: request sleep type (4) failed: Operation not supported
>> #                                                   74 /root
>> The number on the right displays the return value of the previous 
>> command.
>> Here are some select sysctl values:
>> hw.acpi.supported_sleep_state: S3 S4 S5
>> hw.acpi.power_button_state: S5
>> hw.acpi.sleep_button_state: S3
>> hw.acpi.lid_switch_state: NONE
>> hw.acpi.standby_state: S1
>> hw.acpi.suspend_state: S3
>> hw.acpi.sleep_delay: 1
>> hw.acpi.s4bios: 1
>> I was especially happy about the last value. To bad I cannot put it to 
>> the test.
> I have the same hardware running 7-stable and the same issue. Could it 
> be fixed ? Does one need vendor information to do so ?

The temperature issue is clearly a BIOS bug, but since we'll hardly get HP 
to release a fixed BIOS for us this calls for one more sad entry in the acpi 
quirks. Or the code could be changed, so that this message is posted only once.

About the sleep states, I've got no idea. I'd love to go to suspend. On all 
other machines I've ever run FreeBSD on "acpiconf -s 5 " started a clean 
shutdown. It seems extraordinary to me, to claim to support the states S3, 
S4 (BIOS!) and S5 and when trying to use them just get an "Operation Not 
Supported". That looks more like a FreeBSD bug to me.

After my exams I'll experiment with acpidump to try insert a valid _CRT 
value for tz0. I don't think I can do more without learning aml.

More information about the freebsd-acpi mailing list