e1000e probe failure on 2.6.34 and higher, Intel MB

From: Tim Gardner
Date: Thu Jun 10 2010 - 09:59:31 EST



I've a bug where e1000e works with Lucid (2.6.32), but fails to probe with 2.6.35-rc2. The output is mighty terse.

[ 28.193628] e1000e 0000:00:19.0: PCI INT A disabled
[ 28.193740] e1000e: probe of 0000:00:19.0 failed with error -3

The full dmesg logs are at https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/591707/comments/23 and https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/591707/comments/24

The reporter says that it also fails with 2.6.34. The report has logs showing that it works with 2.6.32 and fails with 2.6.35-rc1. I've requested the reporter install a vanilla -rc2 kernel from http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.35-rc2-maverick/linux-image-2.6.35-020635rc2-generic_2.6.35-020635rc2_amd64.deb, but I do not expect different results given that the only change from rc1 to rc2 is 'e1000e: change logical negate to bitwise' (which is not germane to probing).

This appears to be an Intel OEM MB as far as I can tell from the DMI. It was new in April so is unlikely to have flash corruption issues such as those observed with 2.6.24.

Any suggestions on why the probe silently fails? None of the module params look like they'll make a difference.

Tim Gardner tim.gardner@xxxxxxxxxxxxx
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/