Hi,

I'm sorry, I've asked this question before, but I received no useful
response, and I don't know how to debug the issue further. Thus I'm
asking again - maybe more info could be provided?

My system, an HP Envy 15 laptop, just does not power off after shutdown.
  one needs to press the power button for 6 seconds to make the laptop
power off. the same effect exists on ubuntu 9.10 with kernel 2.6.32.9
and .10, and also with ubuntu 10.04, which also uses kernel 2.6.32.

interestingly, ACPI does not list the states in /proc/acpi/info:

$ cat /proc/acpi/info
version: 20090903

but ACPI is recognized on boot:

$ dmesg | grep ACPI
[ 0.000000] BIOS-e820: 00000000bf75c000 - 00000000bf7bf000 (ACPI NVS)
[ 0.000000] BIOS-e820: 00000000bf7e4000 - 00000000bf7ff000 (ACPI data)
[ 0.000000] modified: 00000000bf75c000 - 00000000bf7bf000 (ACPI NVS)
[ 0.000000] modified: 00000000bf7e4000 - 00000000bf7ff000 (ACPI data)
[ 0.000000] ACPI: RSDP 00000000000fe020 00024 (v02 HP )
[ 0.000000] ACPI: XSDT 00000000bf7fe120 0009C (v01 HPQOEM SLIC-MPC
00000001 01000013)
[ 0.000000] ACPI: FACP 00000000bf7fc000 000F4 (v04 HP SLIC-MPC 00000001
MSFT 01000013)
[ 0.000000] ACPI: DSDT 00000000bf7ee000 0ABEA (v02 HP SLIC-MPC 00000001
MSFT 01000013)
[ 0.000000] ACPI: FACS 00000000bf76e000 00040
[ 0.000000] ACPI: ASF! 00000000bf7fd000 000A5 (v32 HPQOEM SLIC-MPC
00000001 MSFT 01000013)
[ 0.000000] ACPI: HPET 00000000bf7fb000 00038 (v01 HPQOEM SLIC-MPC
00000001 MSFT 01000013)
[ 0.000000] ACPI: APIC 00000000bf7fa000 0008C (v02 HPQOEM SLIC-MPC
00000001 MSFT 01000013)
[ 0.000000] ACPI: MCFG 00000000bf7f9000 0003C (v01 HPQOEM SLIC-MPC
00000001 MSFT 01000013)
[ 0.000000] ACPI: SLIC 00000000bf7ed000 00176 (v01 HPQOEM SLIC-MPC
00000001 MSFT 01000013)
[ 0.000000] ACPI: SSDT 00000000bf7ec000 0073D (v01 TrmRef PtidDevc
00001000 INTL 20051117)
[ 0.000000] ACPI: SSDT 00000000bf7eb000 00172 (v01 VaRef Va_Acpi
00003000 INTL 20051117)
[ 0.000000] ACPI: BOOT 00000000bf7ea000 00028 (v01 HPQOEM SLIC-MPC
00000001 MSFT ...
[ message continues ]

Since this issue seems easily reproducible
the best way to find the issue is through
a bisect(hardest part is finding the last
good kernel),
http://www.kernel.org/pub/software/scm/git/docs/user-manual.html

hope this helps,

Justin P. Mattock
--

[ message continues ]

hm, but wouldn't this assume that there is a kernel version, which
actually works?


Akos

--

[ message continues ]

nice!!, so there never was a good kernel version.
In this case I guess the right question is, is
what/where is the shutdown mechanism(in the kernel)
responsible for this and if possible is there already
some mechanism in place for this machine with this issue
i.g. I had to add my machines dmi info to reboot.c
in order for it to reboot properly.

Now keep in mind this could have nothing todo with
adding some dmi entry, this could somewhere else
(which the bisect could reveal(hopefully)).

as for debugging I normally use ohci1394_dma=early
you might be able to catch some strace towards the end
with this.


Justin P. Mattock
--

[ message continues ]