Bug 6209
Summary: | PCI errors with Intel D945GNT motherboard | ||
---|---|---|---|
Product: | Drivers | Reporter: | Dr. Reginald W. Clemens (clemens) |
Component: | PCI | Assignee: | Greg Kroah-Hartman (greg) |
Status: | CLOSED PATCH_ALREADY_AVAILABLE | ||
Severity: | normal | CC: | akpm, bunk, protasnb |
Priority: | P2 | ||
Hardware: | i386 | ||
OS: | Linux | ||
Kernel Version: | 2.6.11 (earliest tested) to present | Subsystem: | |
Regression: | --- | Bisected commit-id: | |
Bug Depends on: | |||
Bug Blocks: | 5829 | ||
Attachments: |
Bug still there (unchanged) 2.6.16.17
Linux still doesnt run with D945Gnt motherboard. |
Description
Dr. Reginald W. Clemens
2006-03-10 12:43:43 UTC
Does this still happen in 2.6.16-rc5? The messages might still be there, but the device should now work. In 2.6.16-rc5-git8 the messages are there, and the ethernet device (mine, not the onboard which has never worked), does not work. a /sbin/ifconfig -a does not show it a /proc/modules does show the driver sk98lin, and says it is Live. and in the messages ON SCREEN (it does not get into the messages file) there is a message that reads bringing up interface eth0: sk98lin device eth0 does not seem to be present, delaying initialization. I am going to assume that this is related. Created attachment 8158 [details]
Bug still there (unchanged) 2.6.16.17
Bug still there (unchanged) 2.6.16.17
use the skge driver ("New SysKonnect GigaEthernet", CONFIG_SKGE), not sk98lin. DLink support got removed from sk98lin in 2.6.14. The error seems to occur long before the modules are loaded, so having the wrong modules shouldnt make any difference. I have changed modprobe.conf so that the eth0 alias points at skge, and tried the experiment with 2.6.17.11. The output is essentially the same as above. Created attachment 9280 [details]
Linux still doesnt run with D945Gnt motherboard.
Mr. Redinald, Have you tried latest kernels? The MSI interrupts handling had problems, now it should be working better. You can also try "nomsi" to swith to full ioapic mode. It would be easier to see the problem if you provided a trace for the working kernel vs. non-vorking. Also, if this is a regression from 2.6.11 you can try running bisect search on source trees to identify the patch(es) that broke it for you. Thanks. Subject: Re: PCI errors with Intel D945GNT motherboard
> http://bugzilla.kernel.org/show_bug.cgi?id=6209
>
>
> protasnb@gmail.com changed:
>
> What |Removed |Added
> ----------------------------------------------------------------------------
> CC| |protasnb@gmail.com
>
>
>
>
> ------- Comment #7 from protasnb@gmail.com 2007-07-03 19:43 -------
> Mr. Redinald,
> Have you tried latest kernels? The MSI interrupts handling had problems, now
> it
> should be working better. You can also try "nomsi" to swith to full ioapic
> mode.
> It would be easier to see the problem if you provided a trace for the working
> kernel vs. non-vorking. Also, if this is a regression from 2.6.11 you can try
> running bisect search on source trees to identify the patch(es) that broke it
> for you.
> Thanks.
>
>
> --
> Configure bugmail: http://bugzilla.kernel.org/userprefs.cgi?tab=email
> ------- You are receiving this mail because: -------
> You reported the bug, or are watching the reporter.
>
I thought I had posted a comment to the bug report when things started working
but at some point in the past my problems were fixed.
I am running 2.6.20 and 2.6.21 now, and am not sure
just which previous kernel it was that fixed the problem.
Awesome, the bug can be closed then. Thanks! |