pci_set_power_state errors in tty0 console

[EcliptuX]

Hi,

I'm using Edgy Eft beta with lastest updates.
That's work very well, but when I go in a console (i.e. CTRL+ALT+F1), I have lots and lots errors like this :

Oct 17 20:02:33 EcliptuX kernel: [17182778.292000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:34 EcliptuX kernel: [17182778.792000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:34 EcliptuX kernel: [17182779.296000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:35 EcliptuX kernel: [17182779.800000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:35 EcliptuX kernel: [17182780.304000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:36 EcliptuX kernel: [17182780.808000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:36 EcliptuX kernel: [17182781.312000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:37 EcliptuX kernel: [17182781.816000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:37 EcliptuX kernel: [17182782.320000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5
Oct 17 20:02:38 EcliptuX kernel: [17182782.824000] pci_set_power_state(): 0000:00:08.0: state=3, current state=5

I can difficultly use the console because these errors messages doesn't stop !
I don't know where the problem come from.

[Ben Collins (ben-collins)]

Please run:

lspci -vv > lspci-vv.txt
lspci -vvn > lspci-vvn.txt

Attach the two files produced by these commands.

Thanks

[jetthe (jonas-dalom)]

This happens to me in Edgy final. My syslog gets swamped with these messages ever since my first reboot after upgrading. Running 2.6.17-10-generic.

[EcliptuX (ecliptux)]

I have notice than the error message stop when I'm stopping gdm
Maybe a clue....

[Dave Lane (lightweight)]

I am experiencing the same problem on Edgy (release version). This machine was an upgrade from Dapper.

[Dave Lane (lightweight)]

my lspci-vvn file...

[EcliptuX (ecliptux)]

do you need more informations ?

[towsonu2003 (towsonu2003)]

probably no?

[Brad (bcalvert)]

I have the same problem in Edgy.

[Brad (bcalvert)]

sorry meant lspci :\

[Brad (bcalvert)]

All have Creative sound cards (most of us Live! cards) and 3com NICs (most of us 905s)

Hmm..

[ATorre (aedelatorre)]

I found it in debian-user at lists.debian.org from Damon L. Chesser and I had verified that if you close the gnome system monitor there aren't messages anymore:

"Well, it was the gnome app system monitor, but only the Network activity indicator. I noticed that when booting into gui, the F1 term did not start the errors until some way into the desktop (in this case gnome). I never noticed the errors before becuse I ran XFCE4. I started closing toolbar toys one at a time until the messages went away. It stopped when I removed system monitor, but came back after I put it back AND enabled network monitoring. It is 100% reproducible on demand. Ergo, Gnome apps>system monitor>network monitoring."

[StigKlüver (ubuntu-kluver)]

I'm having the same problem - I'm running a server without any gui (no gnome app system monitor), no soundcard, but I have a 3c905 NIC.

I have found the following link describing the problem including a patch, but since it is on a production server, I don't have the posibility to verify it.

https://bugzilla.novell.com/show_bug.cgi?id=227391

[Brad (bcalvert)]

Very nice find.. that has to be our problem!

Quote:
"Hmm.. 3c59x.c seems to do pci_set_power_state() without checking if it is
supported; and it seems it can do it when userland asks it to.
You may want to try killing daemons one-by-one to see which one triggers that.
Perhaps pci should not complain when it is asked to do transitions on
non-power-managable hardware?"

They have a patch which has been approved for openSUSE.

[Nick Peelman (peelman)]

I am having this same problem, and i DO have a 3COM NIC installed. If only this was fixed my logwatch emails would be a quarter of what they are now :\

[Michael Anckaert (michael-anckaert)]

Got an Ubuntu 6.10 server with the latest updates and a 3com 3c905 ethernet card.

00:10.0 Ethernet controller: 3Com Corporation 3c905 100BaseTX [Boomerang]

This is also on a production server but I'll try to find out what triggers the pci_set_power_state()
Are there any other tests that need to be done?

[Stefan Lesicnik (stefanlsd)]

I have this same error in Feisty 7.04 server with the 2.6.20-16-server kernel.

you can reproduce the error by typing mii-tool.

[ 2468.245700] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.246663] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.247327] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.247990] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.248641] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.249297] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.249949] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.250607] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.251267] pci_set_power_state(): 0000:00:09.0: state=3, current state=5
[ 2468.251917] pci_set_power_state(): 0000:00:09.0: state=3, current state=5

From what i've read, the 3com card doesn't support this function and should not be doing this. There are various patches around on the web that address this problem.

00:09.0 Ethernet controller: 3Com Corporation 3c905 100BaseTX [Boomerang]
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B-
        Status: Cap- 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
        Latency: 32 (750ns min, 2000ns max)
        Interrupt: pin A routed to IRQ 10
        Region 0: I/O ports at e000 [size=64]
        Expansion ROM at dc000000 [disabled] [size=64K]

I don't mind the error messages in so much as the card physically doesn't work now.

mii-tool reports no link and setting an IP address on the card reports

[ 2786.905061] eth1: setting half-duplex.
[ 2786.905815] ADDRCONF(NETDEV_UP): eth1: link is not ready

[mingchi (mingchi)]

i had the same problem on my Feisty desktop, and i found out the reason for the message. Previously i had added the Gnome Systems Monitor to my desktop, and when i switched to TTY i had the same message repeating on the screen.

Then i remove the Gnome System Monitor from my desktop panel and the message disappear. I am using a 3com NIC as well.

[Leann Ogasawara (leannogasawara)]

The 18 month support period for Edgy Eft 6.10 has reached it's end of life. As a result, we are closing the linux-source-2.6.17 Edgy Eft kernel task. However, Hardy Heron 8.04 was recently released. It would be helpful if you could test the new release and verify if this is still an issue - http://www.ubuntu.com/getubuntu/download . If the issue still exists, please update this report by changing the Status of the "linux" task from "Incomplete" to "New". We appreciate your patience and understanding as we make this transition. Thanks!

[Brian Murray (brian-murray)]

We are closing this bug report as it lacks the information, described in the previous comments, we need to investigate the problem further. However, please reopen it if you can give us the missing information and feel free to submit bug reports in the future.