https://bugzilla.novell.com/show_bug.cgi?id=350017
User trenn@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=350017#c16
--- Comment #16 from Thomas Renninger 2008-01-15 03:49:58 MST ---
Yakui: What kind of DSDT is that, the one for the 1201 or 1503 BIOS? Using the
wrong one on the wrong BIOS is not a good idea.
Would you mind sharing your ideas and at least post the diff of your debug and
the original DSDT.
Some info about 10.3 (a lot SUSE users do not know how to compile a kernel):
- DSDT can be overridden without kernel recompilation:
http://powersave.sourceforge.net/powersave/DSDT.html
- Since 10.3 OpenSuSE kernel versions have ACPI_DEBUG=y compiled in by
default
The correct fix for the broken 1503 BIOS is this:
Compilation complete. 0 Errors, 5 Warnings, 0 Remarks, 1097 Optimizations
trenn@queen:~/dsdt/asus_bug_325995_bios_1503> diff -u DSDT.dsl DSDT.dsl.modif
--- DSDT.dsl 2008-01-14 17:57:32.094872000 +0100
+++ DSDT.dsl.modif 2008-01-15 11:32:50.497733000 +0100
@@ -17,8 +17,6 @@
*/
DefinitionBlock ("DSDT.aml", "DSDT", 1, "NVIDIA", "AWRDACPI", 0x00001000)
{
- External (\_PR_.CPU0)
-
Scope (\_PR)
{
Processor (\_PR.C000, 0x00, 0x00000000, 0x00) {}
@@ -8988,7 +8986,7 @@
Name (_PSL, Package (0x01)
{
- \_PR.CPU0
+ \_PR.C000
})
Name (_TSP, 0x3C)
Name (_TC1, 0x04)
I could reproduce and get the exact stack backtrace on a totally different
machine (with passive trip point, _PSL support), by modifying the reference
returned by _PSL to something invalid (and compile it with iasl -sa -f). This
is the output of a modified Compaq 6715b where I invalidated the _PSL reference
object:
Process powersaved (pid: 3063, threadinfo ffff810071bea000, task
ffff810075639790)
Stack: ffffffff8032d20e ffff810074db2818 ffff810071bebe78 4d52454854584e4c
ffff810072a9f0c0 ffffffff8801434d ffffffff803389f0 0000000000000003
ffff8100757f8000 0000000000000000 ffffffff880135c5 0000000000100073
Call Trace:
[<ffffffff8032d20e>] acpi_get_data+0x3e/0x6e
[<ffffffff803389f0>] acpi_bus_get_device+0x25/0x68
[<ffffffff880135c5>] :thermal:acpi_thermal_trip_seq_show+0x12b/0x257
[<ffffffff802a2482>] seq_read+0x105/0x28b
[<ffffffff8028a2f1>] vfs_read+0xcb/0x153
[<ffffffff8028a685>] sys_read+0x45/0x6e
[<ffffffff80209dac>] tracesys+0xdc/0xe1
My patch from comment #13 is wrong.
What we need to know is how to detect whether a handle has an invalid/undefined
type.
I try to come up with another patch...
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.