| View previous topic :: View next topic |
| Author |
Message |
grooveman
Veteran
Joined: 24 Feb 2003
Posts: 1217
|
 Posted: Thu Jun 11, 2009 9:47 pm Post subject: 2.6.29-gentoo-r5 weirdness |
 |
|
Can anyone tell me why I am getting this in my dmesg suddenly?
| Code: | ACPI: I/O resource w83627ehf [0x295-0x296] conflicts with ACPI region HWRE [0x290-0x299]
ACPI: Device needs an ACPI driver
IRQ 24/nvidia: IRQF_DISABLED is not guaranteed on shared IRQs |
and this is in my log file:
| Code: | Jun 11 17:35:27 cditri w83627ehf: Found W83627EHG chip at 0x290
Jun 11 17:35:27 cditri ACPI: I/O resource w83627ehf [0x295-0x296] conflicts with
ACPI region HWRE [0x290-0x299]
Jun 11 17:35:27 cditri ACPI: Device needs an ACPI driver
Jun 11 17:35:29 cditri IRQ 24/nvidia: IRQF_DISABLED is not guaranteed on shared
IRQs |
I upgraded to 2.6.29-gentoo-r5 yesterday, and my PC has run like crap since. Gets really slow, seems to cache a lot and It will not shutdown, it hangs on unloading the alsa modules every time. Imex posted this -- but his solution with the inittab didn't work for me.
Also, it is not interacting properly with my usb flash drives, won't delete files and won't write them. The same drive works fine on another box with an older kernel, and on M$ XP. No errors in /var/log/messages tho...
dmesg gives this: | Code: | FAT: Filesystem panic (dev sdd1)
fat_get_cluster: invalid cluster chain (i_pos 124214)
FAT: Filesystem panic (dev sdd1)
fat_get_cluster: invalid cluster chain (i_pos 124214)
FAT: Filesystem panic (dev sdd1)
fat_get_cluster: invalid cluster chain (i_pos 124214)
FAT: Filesystem panic (dev sdd1)
fat_get_cluster: invalid cluster chain (i_pos 124214)
FAT: Filesystem panic (dev sdd1)
fat_get_cluster: invalid cluster chain (i_pos 124214)
|
But I know that is probably a crap message, because it wouldn't let me low level format the drive, and it would not let me mke2fs on it, or mkefs t vfat.... I don't think it is really a FAT problem.
I am also getting a long pause at boot, with this message:
| Code: |
Marking TSC unstable due to TSC halts in idle |
I'm using the same .config file as I did with the last kernel, which exhibited no such problems. I need to use 2.6.29 however, because without it, my blu-ray burner runs really, really, really slow.
Any insights here?
thanks.
G
_________________
To look without without looking within is like looking without without looking at all. |
|
| Back to top |
|
 |
Hu
Moderator
Joined: 06 Mar 2007
Posts: 21635
|
| Posted: Fri Jun 12, 2009 11:05 pm Post subject: |
|
|
| What were you using before you upgraded to 2.6.29-gentoo-r5? What is the last kernel version that works? Does the problem also occur if you do not load the proprietary nVidia driver? |
|
| Back to top |
|
|
iabervon
n00b
Joined: 30 Jul 2007
Posts: 2
|
| Posted: Sat Jun 13, 2009 1:45 am Post subject: Re: 2.6.29-gentoo-r5 weirdness |
|
|
| grooveman wrote: | | I upgraded to 2.6.29-gentoo-r5 yesterday, and my PC has run like crap since. Gets really slow, seems to cache a lot and It will not shutdown, it hangs on unloading the alsa modules every time. |
I've got the same hang on unloading alsa modules, which I tracked down to snd_hda_intel preventing snd_hda_codec_analog from being unloaded but not specifically depending on it. If I unload snd_hda_intel, it stops being hung. Not sure what's should be done to fix it, but that's the way it's misbehaving. |
|
| Back to top |
|
|
figueroa
Advocate
Joined: 14 Aug 2005
Posts: 2963
Location: Edge of marsh USA
|
| Posted: Sun Jun 14, 2009 1:49 am Post subject: |
|
|
In my case, the alsa unloading hangup went away when I built snd_hda_intel into the kernel rather than having it as a module. Sounds works great, too.
_________________
Andy Figueroa
hp pavilion hpe h8-1260t/2AB5; spinning rust x3
i7-2600 @ 3.40GHz; 16 gb; Radeon HD 7570
amd64/23.0/split-usr/desktop (stable), OpenRC, -systemd -pulseaudio -uefi |
|
| Back to top |
|
|
grooveman
Veteran
Joined: 24 Feb 2003
Posts: 1217
|
| Posted: Sun Jun 14, 2009 6:31 pm Post subject: |
|
|
| Hu wrote: | | What were you using before you upgraded to 2.6.29-gentoo-r5? What is the last kernel version that works? Does the problem also occur if you do not load the proprietary nVidia driver? |
Was using this: linux-2.6.28-gentoo-r5, with no problems.
Yes, I suppose I could compile it into the kernel, that might work around the module unloading problem... but what about the rest? The irq conflict is particularly disconcerting, since there were no hardware changes, there should be no such conflict... must be a kernel bug...
_________________
To look without without looking within is like looking without without looking at all. |
|
| Back to top |
|
|
figueroa
Advocate
Joined: 14 Aug 2005
Posts: 2963
Location: Edge of marsh USA
|
| Posted: Sun Jun 14, 2009 6:57 pm Post subject: |
|
|
In partial response to Hu and grooveman:
When running the open source nv driver I have no related IRQ conflicts or messages.
The only error message is get in Xorg.0.log is:
(EE) Failed to initialize GLX extension (Compatible NVIDIA X driver not found)
UPDATE: I just fixed that GLX error. I didn't know about "eselect opengl set xorg-x11" but I'd just found it in a nouveau related posting. I do know that "eselect opengl set nv" didn't work. (Where should I have found THAT documented?)
I do, however, still have a LOT of IRQ sharing - just no visible complaints.
_________________
Andy Figueroa
hp pavilion hpe h8-1260t/2AB5; spinning rust x3
i7-2600 @ 3.40GHz; 16 gb; Radeon HD 7570
amd64/23.0/split-usr/desktop (stable), OpenRC, -systemd -pulseaudio -uefi |
|
| Back to top |
|
|
jale2ice
n00b
Joined: 20 Jan 2005
Posts: 9
|
| Posted: Mon Jun 15, 2009 7:48 pm Post subject: |
|
|
I have the same issue and would like to know what amd64 team's view is...
IRQ 16/nvidia: IRQF_DISABLED is not guaranteed on shared IRQs.
From googling, I gathered that this is a driver-vendor related issue, in our case nvidia. Can anyone attest to that?
Secondly, the ALSA module unload issue is a royal pain... All the codecs were selected by default which I think created the conflict. I deactivated all except the nvidia codec (nvhdmi), ran alsaconf and I'm no longer hanging at unloading the modules.
_________________
AMD Athlon 64 5200+ X2 @ 2.81GHz, Abit KN9 SLI, 2GB PC6400 G.Skill, 2x250GB Seagate Barracuda - RAID 0, NVIDIA 8800GTX 768MB, SCEPTRE X20WC-Gamer 20.1" @ 1680x1050@60Hz, SB X-FI XtremeGamer, HIPER 530W Mod PSU & CoolerMaster Centurion 5- 2.6.27-gentoo-r8 |
|
| Back to top |
|
|
grooveman
Veteran
Joined: 24 Feb 2003
Posts: 1217
|
| Posted: Thu Jun 25, 2009 1:57 am Post subject: |
|
|
Well, not 100% on the Nvidia thing, but I found out that the ACPI issue at least is NOT a bug. It looks like they have coded it to be strict by default. This is really a bios issue. Since the previous kernel versions set the acpi_enforce_resources to lax, I never saw the error. Nvidia thing could be due to similar reason.... *shrugs*
http://bugzilla.kernel.org/show_bug.cgi?id=13571
_________________
To look without without looking within is like looking without without looking at all. |
|
| Back to top |
|
|
chris.c.hogan
Apprentice
Joined: 02 Oct 2005
Posts: 189
|
| Posted: Sun Jul 05, 2009 1:57 pm Post subject: Re: 2.6.29-gentoo-r5 weirdness |
|
|
| grooveman wrote: | Can anyone tell me why I am getting this in my dmesg suddenly?
| Code: | ACPI: I/O resource w83627ehf [0x295-0x296] conflicts with ACPI region HWRE [0x290-0x299]
ACPI: Device needs an ACPI driver
IRQ 24/nvidia: IRQF_DISABLED is not guaranteed on shared IRQs |
and this is in my log file:
| Code: | Jun 11 17:35:27 cditri w83627ehf: Found W83627EHG chip at 0x290
Jun 11 17:35:27 cditri ACPI: I/O resource w83627ehf [0x295-0x296] conflicts with
ACPI region HWRE [0x290-0x299]
Jun 11 17:35:27 cditri ACPI: Device needs an ACPI driver
Jun 11 17:35:29 cditri IRQ 24/nvidia: IRQF_DISABLED is not guaranteed on shared
IRQs |
|
I don't have all the other problems listed. However, I do have the problem with Device needs an ACPI driver. I'm running 2.6.29-gentoo-r3, in the process of upgrading to 2.6.30-gentoo-r1. The message I have is:
| Code: |
w83627hf: Found W83627HF chip at 0x290
ACPI: I/O resource w83627hf [0x295-0x296] conflicts with ACPI region IP__ [0x295-0x296]
ACPI: Device needs an ACPI driver |
Doing a Google search results is three pages of people asking what this means and no answers.
Is there something else I need to enable hardware monitoring?
EDIT:
More general googling found this kernel bug:
http://bugzilla.kernel.org/show_bug.cgi?id=13508
| Quote: |
Comment #3 From Alexey Starikovskiy 2009-06-11 18:23:38 -------
Basically, you use whole hwmon infrastructure on your own risk. The conflict
you see was added exactly to prevent users from shooting their feet. Without
this check hwmon will try to use SMBUS and devices connected to it (most
probably temperature control IC), thus making ACPI to read garbage from these
same devices.
If ACPI reads critical temperature (and it often does in such situation), it
will simply shut machine down. |
So, it looks like you can have ACPI thermal monitoring or full sensors from hwmon. You can't have both with this chip.
I'll do some more research on the lm_sensors list and experiment latter. For now, on this machine, I'll just disable hwmon and lm_sensors. |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Kernel & Hardware
