Hello, Latest working kernel version: 2.6.33.4 Earliest failing kernel version: 2.6.34 Distribution: ArchLinux x86_64, kernel is from distro repository but only has 2 patches for aufs2 Hardware Environment: dell studio xps 1647 Problem Description: When booting the latest kernel i see those error messages in dmesg. Kernel 2.6.33.x does not show such problem. pcieport 0000:00:01.0: Requesting control of PCIe PME from ACPI BIOS \_SB_.PCI0:_OSC invalid UUID _OSC request data:1 0 1f pcieport 0000:00:01.0: Failed to receive control of PCIe PME service: ACPI _OSC failed pcie_pme: probe of 0000:00:01.0:pcie01 failed with error -13 pcieport 0000:00:1c.0: Requesting control of PCIe PME from ACPI BIOS \_SB_.PCI0:_OSC invalid UUID _OSC request data:1 0 1f pcieport 0000:00:1c.0: Failed to receive control of PCIe PME service: ACPI _OSC failed pcie_pme: probe of 0000:00:1c.0:pcie01 failed with error -13 pcieport 0000:00:1c.1: Requesting control of PCIe PME from ACPI BIOS \_SB_.PCI0:_OSC invalid UUID _OSC request data:1 0 1f pcieport 0000:00:1c.1: Failed to receive control of PCIe PME service: ACPI _OSC failed pcie_pme: probe of 0000:00:1c.1:pcie01 failed with error -13 pcieport 0000:00:1c.3: Requesting control of PCIe PME from ACPI BIOS \_SB_.PCI0:_OSC invalid UUID _OSC request data:1 0 1f pcieport 0000:00:1c.3: Failed to receive control of PCIe PME service: ACPI _OSC failed pcie_pme: probe of 0000:00:1c.3:pcie01 failed with error -13 pcieport 0000:00:1c.4: Requesting control of PCIe PME from ACPI BIOS \_SB_.PCI0:_OSC invalid UUID _OSC request data:1 0 1f pcieport 0000:00:1c.4: Failed to receive control of PCIe PME service: ACPI _OSC failed pcie_pme: probe of 0000:00:1c.4:pcie01 failed with error -13 pcieport 0000:00:1c.5: Requesting control of PCIe PME from ACPI BIOS \_SB_.PCI0:_OSC invalid UUID _OSC request data:1 0 1f pcieport 0000:00:1c.5: Failed to receive control of PCIe PME service: ACPI _OSC failed pcie_pme: probe of 0000:00:1c.5:pcie01 failed with error -13 Steps to reproduce: boot, check dmesg
Created attachment 26497 [details] lspci -vv output
Created attachment 26498 [details] dmesg output
Created attachment 26499 [details] dmidecode output
Created attachment 26500 [details] acpidump output
Hello, i forgot to mention that this error doesn't seem to cause any harm/problems on the machine whatsoever. -- BRS Atanas Zhelev
Created attachment 26678 [details] custom dsdt please set CONFIG_ACPI_DEBUG, apply the custom dsdt, and then boot the new kernel with boot option acpi.debug_level=0x02 acpi.debug_layer=0xffffffff. please attach the dmesg output after boot with the custom DSDT.
Hi, attaching the dmesg output with the custom DSDT -- BRS Atanas Zhelev
Created attachment 26713 [details] dmesg with custom DSDT
please reboot with acpi.debug.level=0x2002 acpi.debug_layer=0xffffffff this time, and attach the dmesg output.
Created attachment 26727 [details] dmesg with acpi.debug.level=0x2002 acpi.debug_layer=0xffffffff Attaching the requested log
oops, too much debug output this time, please add boot option log_buf_len=1M this time so that we can catch more logs.
Created attachment 26740 [details] dmesg with acpi.debug.level=0x2002 acpi.debug_layer=0xffffffff log_buf_len=1M done
(In reply to comment #9) > please reboot with acpi.debug.level=0x2002 acpi.debug_layer=0xffffffff this > time, and attach the dmesg output. Oops, I must be sleepy when writing this. please boot with acpi.debug_level=0x2002 instead of acpi.debug.level=0x2002. I promise this is the last time you need to reboot your machine for this test. :p
Created attachment 26759 [details] dmesg with acpi.debug_level=0x2002 acpi.debug_layer=0xffffffff log_buf_len=1M Don't worry about it. Log attached.
well, it seems that a flag in firmware is cleared, result in the failed _OSC evaluation. But this can not explain why this is a regression... Atanas, please apply the custom dsdt I attached below, on both 2.6.33 and 2.6.34, and attach the dmesg output with acpi.debug_level=0x2002 acpi.debug_layer=0xffffffff log_buf_len=1M
Created attachment 26822 [details] custom DSDT
Created attachment 26868 [details] dmesg kernel 2.6.33
Created attachment 26869 [details] dmesg kernel 2.6.34
Hi, sorry for the late reply, time is scarce lately. Anyway, logs are attached. -- BRS Atanas Zhelev
Right, the warning message is introduced by commit c7f486567c1d0acd2e4166c47069835b9f75e77b, which is harmless IMO. Actually, every evaluation of _OSC will fail on this machine. Here is a piece of the AML code of _OSC method: If (LAnd (LEqual (Arg0, GUID), NEXP)) { ... } Else { GUID not match } and the NEXP bit is cleared... I'm not sure what MEXP stands for and why it is cleared. Bob, we have a similar internal bug report, right?
The warning should go away after this patch: https://patchwork.kernel.org/patch/106865/ has been applied.
Right. Bug closed.
*** Bug 17792 has been marked as a duplicate of this bug. ***