Ubuntu boot delay - nForce error

[sanctimon]

After selecting Ubuntu 10.04 (64 bit fresh installation) in GRUB, it takes approximately 40 seconds for Ubuntu to start booting up. The screen remains completely blank and there is no Hard Disk activity. Only after this period of time does the system actively boot up, with a flashing cursor at first and then, just before the splash appears, there are three error messages (sometimes just one is visible):

May 3 18:16:23 GANDALF-PC kernel: [ 9.356766] ACPI: resource nForce2_smbus [0x1c00-0x1c3f] conflicts with ACPI region SM00 [0x1c00-0x1c05]
May 3 18:16:23 GANDALF-PC kernel: [ 9.356769] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
May 3 18:16:23 GANDALF-PC kernel: [ 9.356771] nForce2_smbus 0000:00:0a.1: Error probing SMB1.

This is similar to bug #440470 (https://bugs.launchpad.net/ubuntu/karmic/+source/linux/+bug/440470) reported for Karmic.

I cannot say with any certainty whether it is these errors that are causing the long delay at boot, but they could well be.

I have attached the kern.log and also intend to attach a video showing the boot sequence to this bug. Please do not hesitate to contact me for further details.

[sanctimon (dj.martin)]

Please find attached boot video.

[zoli4290 (zoli4290)]

I can confirm the same error with all messages only boot latency is less.

[sanctimon (dj.martin)]

This an additional comment/symptom, which I discovered today. It has increased my suspicion that 10.04 and nForce2 do not work well together. Neither of my PATA DVD-RW drives are auto-mounting at startup. It is impossible to mount during the session and the disk utility does not recognise media in them.

[sanctimon (dj.martin)]

I forced DVD mounting by adding the respective entries to fstab. When attempting to access the DVD media: the following errors ensue:
Unable to mount ***
Error mounting: mount exited with exit code 1: helper failed with:
mount: no medium found on /dev/sr1

and then

Unable to mount ***
DBus error org.gtk. Private.RemoteVolumeMonitor. Failed:
An operation is already pending

[sanctimon (dj.martin)]

When removing all media from drives, then there are no errors in dmesg. However, inserting media when the session has started has no visible effect. Brasero still does not find any dvd drives, even though both are mounted.

[Matt Underwood (matt-bargolf)]

Following the update from Karmic to Lucid I've noticed the error message "nForce2_smbus 0000:00:03.2: Error probing SMB2" on the splash screen during boot too. The boot up process seems to hold for a few seconds up until this point. From my kern.log,

May 6 07:37:50 kong kernel: [ 29.612512] i2c i2c-0: nForce2 SMBus adapter at 0x4d00
May 6 07:37:50 kong kernel: [ 29.612526] ACPI: resource nForce2_smbus [0x4e00-0x4e3f] conflicts with ACPI region SM00 [0x4e00-0x4e3f]
May 6 07:37:50 kong kernel: [ 29.612533] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
May 6 07:37:50 kong kernel: [ 29.612540] nForce2_smbus 0000:00:03.2: Error probing SMB2.

I've attached the complete kern.log, is there anything else that might prove useful?

My system is an ASRock ION 330. uname gives "Linux kong 2.6.32-21-generic #32-Ubuntu SMP Fri Apr 16 08:09:38 UTC 2010 x86_64 GNU/Linux"

[Marty Vowles (mvowles311)]

I'm having the same problem (and did on Karmic) with my M3N78-VM. It has the GeForce 8200 on it and I'm confused why there would be an error for nForce2 as mine is a much newer chipset than nForce2. I didn't notice the problem until I'd installed the nVidia proprietary drivers.

[sanctimon (dj.martin)]

I have attempted to boot from the Grub console and have identified that the boot-up latency coincides with the linux and initrd commands (13 seconds and 25 seconds respectively). This also happens with a previous version of Grub (1.97.2).

[sanctimon (dj.martin)]

Additional comment to clear things up a bit:
I believe that the long latency before boot-up is caused by changes between GRUB2 and legacy GRUB (this problem was not caused in legacy) as I am booting from an external USB drive. It appears that linux and initrd commands read files at lower speeds than they should be from within this setup (see System info attachment). I still consider this a bug under Ubuntu (GRUB2 is now the standard under Ubuntu) and expect this to be resolved. Please advise whether a new bug needs to be created.

However, I must note that I do not think that the two problems are unrelated. There may still be a connection between GRUB not being able to read from a USB drive as fast as it should and Ubuntu not properly recognising nVidia's chipsets. There is still the problem of the shorter latency as described by zoli4290 above when the error messages pop up.

This is a regression from a 9.04/9.10 which was working fine, therefore please do the necessary in looking into it. There are several people affected, if you include bug #440470. Thank you.

[eris23 (jdkatz23)]

from my latest dmesg:
[ 22.559602] ACPI: resource nForce2_smbus [0x1c00-0x1c3f] conflicts with ACPI region SM00 [0x001c00-0x001c05]
[ 22.559605] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 22.559608] nForce2_smbus 0000:00:0a.1: Error probing SMB1.
[ 22.559612] ACPI: resource nForce2_smbus [0x1c80-0x1cbf] conflicts with ACPI region SM01 [0x001c80-0x001c85]
[ 22.559614] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 22.559615] nForce2_smbus 0000:00:0a.1: Error probing SMB2.

[Chase Douglas (chasedouglas)]

Hi all,

I've been meaning to post about these issues for a bit, but I've been swamped with other work. The nForce smbus error message is benign. You can read all about it in my comments in bug #440470. It is highly unlikely to be related to the boot speed issue at all.

I will be creating a patch to lower the logging level of the nForce smbus message so it does not appear over the boot splash screen. We already did the same for the generic ACPI warning message over the same issue. However, rest assured that the message is harmless.

Thanks

[Chase Douglas (chasedouglas)]

Also, please decide if you would like this bug to be for the boot speed issue or for the error message level issue. If you prefer to use this bug for the boot speed issue, please open a new bug so I can fix the error message issue against it. Then post the bug number here.

Thanks

[Bubba (kprohn)]

I am seeing these same error messages in my kernel log with Ubuntu 10.04 (32 bit network upgrade). I don't know if this (32 bit version) should require a new bug report or not.

[ 0.240899] cpufreq-nforce2: Detected nForce2 chipset revision C1
[ 0.240902] cpufreq-nforce2: FSB changing is maybe unstable and can lead to crashes and data loss.
[ 0.240911] cpufreq-nforce2: FSB currently at 200 MHz, FID 10.0

There is a 15 second delay between these two consecutive messages which may be significant.

[ 3.324160] ieee1394: Host added: ID:BUS[0-00:1023] GUID[000000508dd695cf]
[ 18.497448] Adding 2996080k swap on /dev/sda5. Priority:-1 extents:1 across:2996080k

I also found these "nForce2" messages farther down in the kernel log.

[ 18.659727] ACPI: resource nForce2_smbus [0x5000-0x503f] conflicts with ACPI region SM00 [0x5000-0x5005]
[ 18.659731] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 18.659735] nForce2_smbus 0000:00:01.1: Error probing SMB1.
[ 18.659851] i2c i2c-0: nForce2 SMBus adapter at 0x5100
[ 19.009801] agpgart: Detected NVIDIA nForce2 chipset

[sanctimon (dj.martin)]

@ Chase Douglas: As mentioned there is also a slight lag (albeit very slight) just prior to this message appearing. I am also one of those who are troubled by a dysfunctional splash screen, whereas my low-res splash only appears after this message. If you believe that this slight lag (or perceived lag, though I am not sure to what extent it is perceived in the case of Bubba and zoli4290) is also because of the level of logging detail, then I propose the following:

1. Let's make this bug about the nForce2 conflict/smb1 probe error.
2. I will create another bug for the behaviour of Grub at startup.

Please let me know if this is agreeable to you.

[Chase Douglas (chasedouglas)]

That sounds good to me.

Just to be clear though, there should be no delay caused by the error message level. It's just another symptom of an issue where you might not be able to monitor things like chipset fan speed and motherboard temperature.

[sanctimon (dj.martin)]

Ok, new bug #577120 created for the startup delay. Please have a look.

Are you saying that, until a driver is found or created for our particular setups, Ubuntu will be loading generic drivers that are not capable of monitoring chipset status?

[Chase Douglas (chasedouglas)]

@sanctimon:

That's close. Actually, until a *correct* driver is written, Ubuntu will continue to try to load the old drivers, but the ACPI subsystem will prevent them from loading properly. This is why you see the error message. The driver goes refused by ACPI and then emits the message. And yes, that means you won't have full monitoring capabilities for your motherboard.

There is an option to explicitly override the ACPI subsystem so the old drivers will work again, but it's potentially dangerous. I talked more about that option in the other bug.

Please note that I will be out at the Ubuntu Developer Summit all next week, so it may be a while before I get back to this bug. However, rest assured that it is on my task list for when I'm back and have time.

Thakns

[sanctimon (dj.martin)]

What does the writing of a correct driver depend on?

It would be interesting to resolve this, although help would be appreciated with the new bug, too, as apparently that is the one causing the greatest latency at the moment.

[sanctimon (dj.martin)]

Oh, and have fun at the Developer Summit!

[sanctimon (dj.martin)]

Right, I have been looking at the kernel.log, after paying closer attention at the startup videos. Now, once the boot up process has began, it goes pretty fast. There is only one point where it actually stops--for about 12 seconds:

May 4 18:13:05 GANDALF-PC kernel: [ 8.222239] sd 10:0:0:3: [sdi] Attached SCSI removable disk
May 4 18:13:05 GANDALF-PC kernel: [ 19.611320] udev: starting version 151
May 4 18:13:05 GANDALF-PC kernel: [ 19.658288] ACPI: resource nForce2_smbus [0x1c00-0x1c3f] conflicts with ACPI region SM00 [0x1c00-0x1c05]
May 4 18:13:05 GANDALF-PC kernel: [ 19.658290] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
May 4 18:13:05 GANDALF-PC kernel: [ 19.658292] nForce2_smbus 0000:00:0a.1: Error probing SMB1.
May 4 18:13:05 GANDALF-PC kernel: [ 19.658457] i2c i2c-0: nForce2 SMBus adapter at 0x1c80

Following that, we immediately have these error messages. Not sure if the one has to do with the other, but it would be useful to know why there is no activity for about 12 secs (see kern.log attached above).

[Csimbi (turbotalicska)]

Just subscribing for now - will check when back from vacation.

[Fransb (frans-bilsen)]

Hi, I have about the same problem. Somewhat different delay, region and SMB2 instead of SMB1.
Just thought I'd add it...

[ 2.360195] kjournald starting. Commit interval 5 seconds
[ 2.360214] EXT3-fs: mounted filesystem with ordered data mode.
[ 6.200354] usb-storage: device scan complete
[ 6.202554] scsi 0:0:0:0: Direct-Access SAMSUNG HD103SI PQ: 0 ANSI: 2 CCS
[ 6.203690] sd 0:0:0:0: Attached scsi generic sg2 type 0
[ 6.204262] sd 0:0:0:0: [sdb] 1953525168 512-byte logical blocks: (1.00 TB/931 GiB)
[ 6.205379] sd 0:0:0:0: [sdb] Write Protect is off
[ 6.205387] sd 0:0:0:0: [sdb] Mode Sense: 00 38 00 00
[ 6.205393] sd 0:0:0:0: [sdb] Assuming drive cache: write through
[ 6.207376] sd 0:0:0:0: [sdb] Assuming drive cache: write through
[ 6.207384] sdb: sdb1 sdb2 < sdb5 >
[ 6.236378] sd 0:0:0:0: [sdb] Assuming drive cache: write through
[ 6.236387] sd 0:0:0:0: [sdb] Attached SCSI disk
[ 25.682817] udev: starting version 151
[ 25.702637] Adding 1951856k swap on /dev/sda6. Priority:-1 extents:1 across:1951856k
[ 25.721995] lp: driver loaded but no devices found
[ 25.759065] i2c i2c-0: nForce2 SMBus adapter at 0x4d00
[ 25.759080] ACPI: resource nForce2_smbus [0x4e00-0x4e3f] conflicts with ACPI region SM00 [0x4e00-0x4e3f]
[ 25.759089] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 25.759098] nForce2_smbus 0000:00:03.2: Error probing SMB2.
[ 25.888208] shpchp: Standard Hot Plug PCI Controller Driver version: 0.4
[ 25.889946] Linux agpgart interface v0.103
[ 25.997760] EXT3 FS on sda5, internal journal

[Chase Douglas (chasedouglas)]

I've sent a patch upstream to delete the redundant "Error probing SMB*" message. The patch may be found at:

http://marc.info/?l=linux-i2c&m=127412739418628&w=2

Hopefully this patch will be merged upstream. We can then submit this to <email address hidden>, which will drive inclusion for 10.04 LTS.

[sanctimon (dj.martin)]

Thanks for looking into this. If you, or somebody else from the developer team could have a look at the other issue (bug #577120), too, that would be great! Thanks.

[ingo (ingo-steiner)]

Same here with Lucid-amd64 on ASUS M2Ne MoBo:

Jun 5 20:14:05 pp kernel: [ 6.600915] k8temp 0000:00:18.3: Temperature readouts might be wrong - check erratum #141
Jun 5 20:14:05 pp kernel: [ 6.740961] ata8: port disabled. ignoring.
Jun 5 20:14:05 pp kernel: [ 6.903379] i2c i2c-0: nForce2 SMBus adapter at 0x1c00
Jun 5 20:14:05 pp kernel: [ 6.903387] ACPI: resource nForce2_smbus [0x1c40-0x1c7f] conflicts with ACPI region SM00 [0x1c40-0x1c45]
Jun 5 20:14:05 pp kernel: [ 6.903391] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
Jun 5 20:14:05 pp kernel: [ 6.903393] nForce2_smbus 0000:00:01.1: Error probing SMB2.

Output of 'dmidecode' is attached.

One more remark:
this error does not appear in Hardy-amd64. Hardy performs s2ram (pm-suspend) perfectly, while Lucid only suspends and resumes once, second suspend fails. Could that be related?

[ingo (ingo-steiner)]

I also tried the kernel parameter "acpi_enforce_resources=lax", but it does not change anything besides:

that now s2ram+resume works 1x and power consumption after resum is 5 watts less and
initiating 2nd pm-suspend increases power consumption by 15 watts over initial value - s2ram of course fails, all is locked up

For completeness I do attach here the dmidecode as obtained from Lucid which includes support vor ECC on Athlon64 (the version above is from Hardy on same hardware)

[Chase Douglas (chasedouglas)]

ingo,

Two things to try. First, install the 2.6.34 mainline kernel:

http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.34-lucid/

See if that fixes things for you. If not, then boot with the same kernel, but add "acpi_sleep=sci_force_enable" to the kernel command line. To do this, edit /etc/default/grub and add the text to the end of the GRUB_CMDLINE_LINUX_DEFAULT variable. Then run "sudo update-grub".

Please let us know if either of these changes fixes things for you.

Thanks!

[Chase Douglas (chasedouglas)]

ingo,

I should have mentioned in my previous comment that your suspend issue is likely unrelated to this issue. After you test my suggestions out, please open a new bug for the suspend issue.

Thanks

[stopdatrue (xebrun)]

New kernel tested (with or without acpi_sleep): same errors

[ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.34-020634-gener
ic root=UUID=24902e9d-7aeb-442e-a3e9-3d02e4b97f5f ro acpi_sleep=sci_force_enable
 xbmc=autostart,nodiskmount,setvolume loglevel=7
...
[ 12.134374] nForce2_smbus 0000:00:03.2: Error probing SMB1.
[ 12.134454] ACPI: resource nForce2_smbus [io 0x4e00-0x4e3f] conflicts with A
CPI region SM00 [??? 0x00004e00-0x00004e3f flags 0x30]
[ 12.134460] ACPI: If an ACPI driver is available for this device, you should
use it instead of the native driver
[ 12.134466] nForce2_smbus 0000:00:03.2: Error probing SMB2.
[

[ingo (ingo-steiner)]

Tested mainline kernel 2.6.34 (with or without "acpi_sleep=sci_force_enable" boot parameter).

It does not change anything, precisely the same as before in both cases.
Here the output from "dmesg |grep -i -A3 -B5 error":

[ 16.052492] ata8: port disabled. ignoring.
[ 16.052518] k8temp 0000:00:18.3: Temperature readouts might be wrong - check erratum #141
[ 16.232967] i2c i2c-0: nForce2 SMBus adapter at 0x1c00
[ 16.232977] ACPI: resource nForce2_smbus [io 0x1c40-0x1c7f] conflicts with ACPI region SM00 [io 0x1c40-0x1c45 64bit pref disabled]
[ 16.232980] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 16.232983] nForce2_smbus 0000:00:01.1: Error probing SMB2.
[ 16.234190] usblp0: USB Bidirectional printer dev 4 if 0 alt 1 proto 2 vid 0x03F0 pid 0x1004
[ 16.234222] usbcore: registered new interface driver usblp
[ 16.255109] forcedeth: Reverse Engineered nForce ethernet driver. Version 0.64.

With regard to s2ram/suspend: also no changes, suspend + resumes once, on second suspend system hangs. Will file a separate bug for that. Probably a hint: with Hardy I had to unload the USB modules with

SUSPEND_MODULES="$SUSPEND_MODULES ehci-hcd ohci-hcd"

which of course is no longer posssible with Lucid as they are compiled into the kernel instead of module.

[ingo (ingo-steiner)]

For the suspend issue I have filed a new bug, see here:
https://bugs.launchpad.net/bugs/592780

[Cedric (cedric-brandenbourger-deactivatedaccount-deactivatedaccount)]

Got the same bug on an Asus P5N-E SLI. Because of that, my fans are turning at full speed all the time.

[ingo (ingo-steiner)]

Seems this bug appears to be ASUS board specific?

[rasos (rasos)]

Same situation here on motherboard ASUS M3N78-VM with Mythbuntu 10.04 kernel 2.6.32-22-generic. dmesg says at boot:
[ 17.309719] i2c i2c-0: nForce2 SMBus adapter at 0x600
[ 17.309726] ACPI: resource nForce2_smbus [0x700-0x73f] conflicts with ACPI region SM00 [0x700-0x73f]
[ 17.309729] ACPI: This conflict may cause random problems and system instability
[ 17.309731] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
[ 17.309751] i2c i2c-1: nForce2 SMBus adapter at 0x700
[ 17.334792] k8temp 0000:00:18.3: Temperature readouts might be wrong - check erratum #141

Additionally, I can not get the Nvidia nouveau working, various Xorg.confs tested.
Always get a: "Failed to initialize the NVIDIA gaphics device"
Can run only low graphics VESA mode, Xorg log file submitted at https://bugs.launchpad.net/ubuntu/+bug/561738

[Joshua Roberts (jrtroberts)]

I have a similar error with Asus P5N-VM WS/TW100-E5. The error did no appear until I updated the Bios. Previous Bios prevented x64 operating systems from seeing 8gigs of ram, or in the case of non linux OS prevented them from installing.

After Bios update I am getting the nforce2_smbus errors. Also my Nvidia drivers have crashed and I cannot load the nvidia settings, nor do both of the nvidia cards work simultaneously anymore, probably due to the driver crashing. They did under previous bios, but Ubuntu Lucid was unable to see more than 2.9 gigs of ram.

original bios P5N-VM-WS/TW100-E5-BIOS-0405 -> No nforce2 error with this bios version

updated bios P5N-VM-WS-BIOS-0601 -> error started after update to this bios version
"1. A fix to system would not be able to install 64-bit OS when more than 4GB memory"

Don't know if this information helps.
Joshua

[Joshua Roberts (jrtroberts)]

Just to make sure that issue was related to the current kernel release I installed older version of Ubuntu.
Ubuntu 9.04 no nforce2 errors
Upgrade to 9.10 no nforce2 errors
Upgrade to 10.04 nforce2_SMBUS 0000:00:03.2: error probing SMB1 error begins
                                  nvidia graphics drivers not working
                                  force boot into low graphics mode

These are my observations

[ingo (ingo-steiner)]

I am still staying on Hardy here (kernel 2.6.24) with most PC's. I'll wait with upgrade until Lucid has stabilised. With Hardy this kind of messages are unknown.

[Joshua Roberts (jrtroberts)]

For those interested I may have dug up a workaround and a potential source for the problem.

http://ubuntuforums.org/showthread.php?p=9481013#post9481013

View this thread and it may shed some light.

Potentially a problem with the Noveau Drivers. Post has a workaround that is claimed to work. I gave up and went back to karmic.

[sanctimon (dj.martin)]

@Joshua Roberts. Disclaimer: this workaround concerns only those who also have an nVidia graphics card and are facing problems with Plymouth, the splash utility. I think it has little to do with the fact that Ubuntu does not carry drivers that are made for this Motherboard. The smb1 message is a motherboard issue. The nouveau drivers you are referring to are Graphics drivers.

[ingo (ingo-steiner)]

I just tried with a fresh installation of Squeeze-amd64, kernel 2.6.32-5 on same machine:

Message is precisely the same as with Lucid:
"nForce2_smbus 0000:00:01.1: Error probing SMB2."

Remark:
graphics deriver does not matter on Lucid, I tried "nouveau" and NVIDIA 173* and NVIDIA latest - the same.

[rasos (rasos)]

@ sanctimon and Joshua Roberts: Yes, deactivating the KMS (fast graphics driver switch in kernel) does not solve any problem reported here. Even without KMS and using a traditional xorg.conf for nvidia, the nvidia driver setting tool remains saying:
"You do not appear to be using the NVIDIA X driver. Please edit your X configuration file (just run `nvidia-xconfig` as root), and restart the X server." Which I tried dozens of times without success.

The problem remains: With an "Error probing SMB2" the nouveau nvidia current does not work on ASUS M3N78-VM with a 2.6.32-22 kernel. It had worked with ubuntu 9.10.

[JeffMan (jeff-audenard)]

@ joshua Roberts
I have the same issue with Lucid on a Zotac MAG. On my side, I had a unstable network interface : Something it was working sometimes not.
Adding the two options (acpi=off & nomodeset=0) to the kernel fixed the issue on my side.

[ingo (ingo-steiner)]

I digged further into this issue and found some difference in i2c-modules between Hardy and Lucid:

Hardy:

kern.log says all ok:
Jun 22 12:35:41 pp kernel: [ 54.573266] input: PC Speaker as /devices/platform/pcspkr/input/input4
Jun 22 12:35:41 pp kernel: [ 54.589284] i2c-adapter i2c-0: nForce2 SMBus adapter at 0x1c00
Jun 22 12:35:41 pp kernel: [ 54.589302] i2c-adapter i2c-1: nForce2 SMBus adapter at 0x1c40
Jun 22 12:35:41 pp kernel: [ 54.665463] Linux video capture interface:v2.00

~$ modinfo i2c_nforce2
filename: /lib/modules/2.6.24-28-generic/kernel/drivers/i2c/busses/i2c-nforce2.ko
description: nForce2/3/4/5xx SMBus driver
author: Hans-Frieder Vogt <email address hidden>
license: GPL
srcversion: EB7002ADFD3CBF38D5C15EF
alias: pci:v000010DEd00000446sv*sd*bc*sc*i*
alias: pci:v000010DEd000003EBsv*sd*bc*sc*i*
alias: pci:v000010DEd00000368sv*sd*bc*sc*i*
alias: pci:v000010DEd00000264sv*sd*bc*sc*i*
alias: pci:v000010DEd00000034sv*sd*bc*sc*i*
alias: pci:v000010DEd00000052sv*sd*bc*sc*i*
alias: pci:v000010DEd000000E4sv*sd*bc*sc*i*
alias: pci:v000010DEd000000D4sv*sd*bc*sc*i*
alias: pci:v000010DEd00000084sv*sd*bc*sc*i*
alias: pci:v000010DEd00000064sv*sd*bc*sc*i*
depends: i2c-core
vermagic: 2.6.24-28-generic SMP mod_unload

Lucid:

May 19 17:49:10 pp kernel: [ 16.013542] i2c i2c-0: nForce2 SMBus adapter at 0x1c00
May 19 17:49:10 pp kernel: [ 16.013549] ACPI: resource nForce2_smbus [0x1c40-0x1c7f] conflicts with ACPI region SM00 [0x1c40-0x1c45]
May 19 17:49:10 pp kernel: [ 16.013552] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver

~$ modinfo i2c_nforce2
filename:
/lib/modules/2.6.32-22-generic/kernel/drivers/i2c/busses/i2c-nforce2.ko
description: nForce2/3/4/5xx SMBus driver
author: Hans-Frieder Vogt <email address hidden>
license: GPL
srcversion: F20A6E012E416B99979D5A1
alias: pci:v000010DEd00000AA2sv*sd*bc*sc*i*
alias: pci:v000010DEd00000752sv*sd*bc*sc*i*
alias: pci:v000010DEd000007D8sv*sd*bc*sc*i*
alias: pci:v000010DEd00000542sv*sd*bc*sc*i*
alias: pci:v000010DEd00000446sv*sd*bc*sc*i*
alias: pci:v000010DEd000003EBsv*sd*bc*sc*i*
alias: pci:v000010DEd00000368sv*sd*bc*sc*i*
alias: pci:v000010DEd00000264sv*sd*bc*sc*i*
alias: pci:v000010DEd00000034sv*sd*bc*sc*i*
alias: pci:v000010DEd00000052sv*sd*bc*sc*i*
alias: pci:v000010DEd000000E4sv*sd*bc*sc*i*
alias: pci:v000010DEd000000D4sv*sd*bc*sc*i*
alias: pci:v000010DEd00000084sv*sd*bc*sc*i*
alias: pci:v000010DEd00000064sv*sd*bc*sc*i*
depends:
vermagic: 2.6.32-22-generic SMP mod_unload modversions

Worth to mention:
in Hardy module i2c_nforce2 depends on i2c-core module, not so in Lucid.
Actually module i2c-core does not exist in Lucid!

[ingo (ingo-steiner)]

Here the chipset information as obtained with Lucid (MoBo: ASUS M2N-E) by:

 # lspci -v | grep -A9 SMB

00:01.1 SMBus: nVidia Corporation MCP55 SMBus (rev a2)
 Subsystem: ASUSTeK Computer Inc. Device 8239
 Flags: 66MHz, fast devsel, IRQ 10
 I/O ports at fc00 [size=64]
 I/O ports at 1c00 [size=64]
 I/O ports at 1c40 [size=64]
 Capabilities: [44] Power Management version 2
 Kernel driver in use: nForce2_smbus
 Kernel modules: i2c-nforce2

So, at least the 3rd bus with I/O at 0x1c40 is not served by the module due to above ACPI-conflict:

i2c i2c-0: nForce2 SMBus adapter at 0x1c00
ACPI: resource nForce2_smbus [io 0x1c40-0x1c7f] conflicts with ACPI region SM00 [io 0x1c40-0x1c45 64bit pref disabled]
ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
nForce2_smbus 0000:00:01.1: Error probing SMB2.

Is there a possibility to force module 'i2c-nforce2' to bind to 0x1c40?

[eris23 (jdkatz23)]

on Maverick ASUS P5N-D

lspci -v | grep -A9 SMB
00:0a.1 SMBus: nVidia Corporation MCP51 SMBus (rev a3)
 Subsystem: ASUSTeK Computer Inc. Device 81bc
 Flags: 66MHz, fast devsel, IRQ 10
 I/O ports at 1c00 [size=64]
 I/O ports at 1c80 [size=64]
 Capabilities: <access denied>
 Kernel modules: i2c-nforce2

00:0a.2 RAM memory: nVidia Corporation MCP51 Memory Controller 0 (rev a3)
 Subsystem: ASUSTeK Computer Inc. Device 81bc

dmesg |grep smbus
[ 16.487456] ACPI: resource nForce2_smbus [io 0x1c00-0x1c3f] conflicts with ACPI region SM00 [irq 7168-7173 pref window]
[ 16.487463] ACPI: resource nForce2_smbus [io 0x1c80-0x1cbf] conflicts with ACPI region SM01 [irq 7296-7301 pref window]

[Chase Douglas (chasedouglas)]

ingo,

If you boot with the 'acpi_enforce_resources=lax' option, it will allow the i2c-nforce2 driver to bind. Since that's been tested before, it is not a fix for the boot speed issue.

[Chris Conway (cconway)]

This bug affects my M3N78-VM motherboard. The "acpi_enforce_resources=lax" works for me, but I do see some startup lag. The portion of dmesg where the lag occurs seems to precede the ACPI message:

[ 1.773684] EXT4-fs (sda3): mounted filesystem with ordered data mode
[ 12.920911] udev: starting version 151
[ 12.938502] Adding 11510532k swap on /dev/sda5. Priority:-1 extents:1 across:11510532k
[ 12.984970] lp: driver loaded but no devices found
[ 13.002868] i2c i2c-0: nForce2 SMBus adapter at 0x600
[ 13.002873] ACPI: resource nForce2_smbus [0x700-0x73f] conflicts with ACPI region SM00 [0x700-0x73f]
[ 13.002876] ACPI: This conflict may cause random problems and system instability
[ 13.002877] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver

[rm (jokemart)]

I see these entries in kern.log:

Jan 17 07:41:02 ion kernel: [ 2.604867] udev: starting version 151
Jan 17 07:41:02 ion kernel: [ 2.941637] i2c i2c-0: nForce2 SMBus adapter at 0x4d00
Jan 17 07:41:02 ion kernel: [ 2.941655] ACPI: resource nForce2_smbus [0x4e00-0x4e3f] conflicts with ACPI region SM00 [0x4e00-0x4e3f]
Jan 17 07:41:02 ion kernel: [ 2.941663] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
Jan 17 07:41:02 ion kernel: [ 2.941671] nForce2_smbus 0000:00:03.2: Error probing SMB2.
Jan 17 07:41:02 ion kernel: [ 2.944484] lp: driver loaded but no devices found

The "...Error probing SMB2" message appears during boot with a long pause.

The system is Zotac ionitx-a-u atom n330 based.

uname -a says:
Linux ion 2.6.32-27-generic-pae #49-Ubuntu SMP Thu Dec 2 00:07:52 UTC 2010 i686 GNU/Linux

Also kernel tainting nvidia driver:
NVIDIA UNIX x86 Kernel Module 195.36.24

Since these issues seem to be hardware monitoring related, I wonder if the proprietary nvidia driver is a common theme as there is a temperature readout in the application System -> Administration -> Nvidia X Server Settings under GPU xxx -> Thermal Settings. Maybe this application is fighting over the hardware sensors?

[rus (ruslja)]

Chase Douglas (chasedouglas)

The comment #31 doesn’t work for me. The error message "nForce2_smbus 0000:00:01.1: Error probing SMB1" is still present and Lucid boot at console mode. Deleting "acpi_sleep=sci_force_enable" from grub give the possibility to boot in graphic mode

@rus:~$ lspci -v | grep -A9 SMB
00:01.1 SMBus: nVidia Corporation nForce 250Gb PCI System Management (rev a1)
 Subsystem: ASUSTeK Computer Inc. Device 813f
 Flags: 66MHz, fast devsel
 I/O ports at 5080 [size=32]
 I/O ports at 5000 [size=64]
 I/O ports at 5040 [size=64]
 Capabilities: <access denied>
 Kernel driver in use: nForce2_smbus
 Kernel modules: i2c-nforce2
@rus:~$dmesg | grep acpi
[ 0.000000] If you got timer trouble try acpi_use_timer_override
[ 0.000000] ACPI: LAPIC (acpi_id[0x01] lapic_id[0x00] enabled)
[ 0.945045] pata_acpi 0000:00:08.0: power state changed by ACPI to D0
[ 0.945168] pata_acpi 0000:00:08.0: power state changed by ACPI to D0
[ 0.945311] pata_acpi 0000:00:08.0: setting latency timer to 64
[ 0.945894] pata_acpi 0000:00:0a.0: PCI INT A -> Link[LTID] -> GSI 22 (level, low) -> IRQ 22
[ 0.946041] pata_acpi 0000:00:0a.0: setting latency timer to 64
[ 0.946051] pata_acpi 0000:00:0a.0: PCI INT A disabled
@rus:~$ dmesg | grep smb
[ 23.590295] ACPI: resource nForce2_smbus [io 0x5000-0x503f] conflicts with ACPI region SMRG [??? 0x00005000-0x00005004 flags 0x47]
[ 23.590306] nForce2_smbus 0000:00:01.1: Error probing SMB1.
@rus:~$ uname -r
2.6.34-020634-generic

[penalvch (penalvch)]

sanctimon, this bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? If so, could you please test for this with the latest development release of Ubuntu? ISO images are available from http://cdimage.ubuntu.com/daily-live/current/ .

If it remains an issue, could you please run the following command in the development release from a Terminal (Applications->Accessories->Terminal), as it will automatically gather and attach updated debug information to this report:

apport-collect -p linux <replace-with-bug-number>

Also, could you please test the latest upstream kernel available following https://wiki.ubuntu.com/KernelMainlineBuilds ? It will allow additional upstream developers to examine the issue. Please do not test the daily kernel folder, but the one all the way at the bottom. Once you've tested the upstream kernel, please comment on which kernel version specifically you tested. If this bug is fixed in the mainline kernel, please add the following tags:
kernel-fixed-upstream
kernel-fixed-upstream-VERSION-NUMBER

where VERSION-NUMBER is the version number of the kernel you tested. For example:
kernel-fixed-upstream-v3.12-rc1

This can be done by clicking on the yellow circle with a black pencil icon next to the word Tags located at the bottom of the bug description. As well, please remove the tag:
needs-upstream-testing

If the mainline kernel does not fix this bug, please add the following tags:
kernel-bug-exists-upstream
kernel-bug-exists-upstream-VERSION-NUMBER

As well, please remove the tag:
needs-upstream-testing

Once testing of the upstream kernel is complete, please mark this bug's Status as Confirmed. Please let us know your results. Thank you for your understanding.

[Launchpad Janitor (janitor)]

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]