Bug 16025 - syslog spam: [\_TZ_.TZ01] Namespace lookup failure, AE_NOT_FOUND
Summary: syslog spam: [\_TZ_.TZ01] Namespace lookup failure, AE_NOT_FOUND
Status: CLOSED DOCUMENTED
Alias: None
Product: ACPI
Classification: Unclassified
Component: BIOS (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: acpi_bios
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-05-22 03:35 UTC by dwcheung
Modified: 2010-12-09 21:59 UTC (History)
4 users (show)

See Also:
Kernel Version: 2.6.3x, currently on 2.6.34
Subsystem:
Regression: No
Bisected commit-id:


Attachments
dmesg output (50.52 KB, text/plain)
2010-05-22 03:36 UTC, dwcheung
Details
lspci -vv output (13.96 KB, text/plain)
2010-05-22 03:36 UTC, dwcheung
Details
acpidump output (192.28 KB, application/octet-stream)
2010-05-22 03:38 UTC, dwcheung
Details
acpidump --addr 0x8DED89ED --length 0x000003E8 > native-hp.dat (1000 bytes, application/octet-stream)
2010-09-29 02:52 UTC, dwcheung
Details
acpidump --addr 0x8DED8299 --length 0x00000754 > legacy-hp.dat (1.83 KB, application/octet-stream)
2010-09-29 02:53 UTC, dwcheung
Details

Description dwcheung 2010-05-22 03:35:03 UTC
Hardware Model: Onkyo PC DX1007A5
http://www.jp.onkyo.com/pc/notebook/dx1007/

Detailed Description:
As soon as the machine finish booting with ACPI enabled, the messages start. This has been happening since 2.6.31 kernel series. It does not appear to have an adverse impact other than chewing up IO and drive space, but I would like to understand the problem and possibly eliminate the message without disabling ACPI.

Troubleshooting:
ACPI_DEBUG was enabled and acpi.debug_layer=0x10 acpi.debug_level=0x1f added to boot options. dmesg, lspci and acpidump outputs are attached.
Comment 1 dwcheung 2010-05-22 03:36:17 UTC
Created attachment 26492 [details]
dmesg output
Comment 2 dwcheung 2010-05-22 03:36:56 UTC
Created attachment 26493 [details]
lspci -vv output
Comment 3 dwcheung 2010-05-22 03:38:04 UTC
Created attachment 26494 [details]
acpidump output
Comment 4 Alexey Starikovskiy 2010-05-22 10:54:02 UTC
Your BIOS wants to call TZ.TZ01 (some thermal zone object) in a regular fashion, which it have not defined. This is the last call in Q11 with the 2 above it doing some job (store and SMI invoke), thus disabling Q11 on the grounds of it returning an error is not a wise option, so EC driver may not help you in this case.
Please check if updated BIOS is available.
Comment 5 dwcheung 2010-05-22 11:26:54 UTC
There is no updated BIOS. Is there a way to silence this via an cutomized DSDT/SSDT table ? Not familiar enough with the iasl tools but if this is possible, I will try that route (and hopefully not blow up something).

thanks for any further insight.
Comment 6 Alexey Starikovskiy 2010-05-22 11:36:37 UTC
You need to delete Notify(_TZ.TZ01) at the end of _Q11 method.
How-to override DSDT is available here: http://www.lesswatts.org/projects/acpi/overridingDSDT.php
Comment 7 Len Brown 2010-09-29 02:16:35 UTC
ACPI: Execute Method [\_SB_.PCI0.LPC0.EC0_._Q11] (Node ffff880161c27800)
ACPI Error (psargs-0359): [\_TZ_.TZ01] Namespace lookup failure, AE_NOT_FOUND
ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.LPC0.EC0_._Q11] (Node ffff880161c27800), AE_NOT_FOUND

hmm, there are two AML load instructions in the DSDT.
Please attach any files found under
/sys/firmware/acpi/tables/dynamic/
Comment 8 dwcheung 2010-09-29 02:21:54 UTC
now running the 2.6.35.6 kernel and there are no files located in the requested directory.
Comment 9 Zhang Rui 2010-09-29 02:35:55 UTC
Please run
acpidump --addr 0x8DED8299 --length 0x00000754 > legacy-hp.dat
acpidump --addr 0x8DED89ED --length 0x000003E8 > native-hp.dat
to get these two tables manually.
Comment 10 dwcheung 2010-09-29 02:52:21 UTC
Created attachment 31832 [details]
acpidump --addr 0x8DED89ED --length 0x000003E8 > native-hp.dat
Comment 11 dwcheung 2010-09-29 02:53:26 UTC
Created attachment 31842 [details]
acpidump --addr 0x8DED8299 --length 0x00000754 > legacy-hp.dat
Comment 12 Zhang Rui 2010-09-29 03:07:01 UTC
bad table data.
so I think this IS the problem as described in comment #6.
bug closed.

Note You need to log in before you can comment on or make changes to this bug.