linux-kernel - Re: ACPI Error under 2.6.26-rc*

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20080715100632.GC4724@linux.int.dgn.ag>
Date:	Tue, 15 Jul 2008 12:06:32 +0200
From:	Andre Prendel <andre_prendel@....de>
To:	Zhang Rui <rui.zhang@...el.com>
Cc:	Andre Prendel <andre_prendel@....de>,
	"Rafael J. Wysocki" <rjw@...k.pl>,
	ACPI Devel Maling List <linux-acpi@...r.kernel.org>,
	kernel-testers@...r.kernel.org, Andi Kleen <andi@...stfloor.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	LKML <linux-kernel@...r.kernel.org>,
	"Zhao, Yakui" <yakui.zhao@...el.com>
Subject: Re: ACPI Error under 2.6.26-rc*

On Tue, Jul 15, 2008 at 04:41:00PM +0800, Zhang Rui wrote:
> On Tue, 2008-07-15 at 15:41 +0800, Andre Prendel wrote:
> > On Mon, Jul 14, 2008 at 05:11:41PM +0800, Zhang Rui wrote:
> > > > Hello Rui,
> > > >
> > > > that's quite strange. Booting the system to get an acpidump I have
> > > > looked at the dmesg output and there is no ACPI Error like the one
> > > > above. There is only an ACPI Exeption.
> > > >
> > > > [...]
> > > > Jul 13 22:00:18 ubuntu kernel: [    0.104111] ACPI: Core revision
> > > > 20080321
> > > > Jul 13 22:00:18 ubuntu kernel: [    0.104259] ACPI Exception
> > > > (tbxface-0629): AE_NO_ACPI_TABLES, While loading namespace from
> > ACPI
> > > > tables [20080321]
> > > > [...]
> > > >
> > > > Booting again I found the error again
> > > >
> > > > [...]
> > > > Jul 13 22:10:50 ubuntu kernel: [    0.000000] ACPI Error
> > > > (tbutils-0472): Invalid length 0x0 in RSDT/XSDT [20080321]
> > > > [...]
> > > >
> > > > and acpidump says:
> > > > root@...ntu:/home/andre/src/acpi/pmtools-20071116/acpidump#
> > > > ./acpidumpACPI tables were not found. If you know location of RSD
> > PTR
> > > > table (from dmesg, etc), supply it with either --addr or -a option
> > > >
> > > > Linux 2.6.25.10 always works.
> > >
> > > It seems that OS failed to get the correct RSDT address
> > (0x1fff0000),
> > > although the RSDP is found correctly. (ACPI: RSDP 000FA340, 0014)
> > >
> > > please run "./acpidump --addr 0xFA340 --length 0x14 > rsdp_xxx",
> > > and attach the rsdp for all of the three cases (good, ACPI Error,
> > ACPI
> > > Exception).
> > 
> > Hello Rui,
> > 
> > here is the output of the three cases. The dump files are attached
> > too.
> > 
> > ----
> > good
> > ----
> > andre@...ntu:~/src/acpi/pmtools-20071116/acpidump$ hexdump rsdp_good
> > 0000000 5352 2044 5450 2052 418c 494d 2020 0020
> > 0000010 0000 1fff                             
> > 0000014
> > 
> > -----
> > error
> > -----
> > andre@...ntu:~/src/acpi/pmtools-20071116/acpidump$ hexdump rsdp_error
> > 0000000 5352 2044 5450 2052 418a 494d 2020 0020
> > 0000010 0000 21ff                             
> > 0000014
> > 
> > ---------
> > exception
> > ---------
> > andre@...ntu:~/src/acpi/pmtools-20071116/acpidump$ hexdump
> > rsdp_exception
> > 0000000 5352 2044 5450 2052 417c 494d 2020 0020
> > 0000010 0000 2fff                             
> > 0000014
> 
> Weird.
> The dmesg shows that the memory map gotten from e820 table is changed.
> And the file attached shows that rsdp points to different rsdt address.
> > 
> > Now I see the ACPI Error under the 2.6.25.10 kernel sometimes too, but
> > power off works well. So I also attach the two configs. Maybe the
> > different config is the reason why power off works under 2.6.25.10.
> so you may get different e820 table when running 2.6.25.10 as well?
> could you please attach the dmesg output of a 2.6.25.10 kernel which has
> the ACPI error?

Ok, I will try to get the error again and send the dmesg output.
 
> > Is it possible that it is a hardware related issue?
> Don't know. :(
> It looks like a hardware issue as BIOS export different e820 tables.
> But from your description, it seems to be a software regression.
> please try to revert c3965bd15118742d72b4bc1a290d37b3f081eb98 and see if
> it helps.

I will do this too, but I dont know if I can manage that tonight. I will
inform you about the result ASAP.

> Please re-attach the kernel config files as attachments.

Attached.
 
> thanks,
> rui
> 
Thanks,
Andre

View attachment "config-2.6.25.10" of type "text/plain" (59986 bytes)

View attachment "config-2.6.26-rc9-git" of type "text/plain" (59113 bytes)