Summary: | dm-mod fails to load in >=sys-kernel/gentoo-dev-sources-2.6.8-r6 | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | John Ross Hunt <john.ross.hunt> |
Component: | [OLD] Core system | Assignee: | Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel> |
Status: | RESOLVED INVALID | ||
Severity: | major | ||
Priority: | High | ||
Version: | unspecified | ||
Hardware: | x86 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | Difference between /usr/include/linux/device-mapper.h and kernel version |
Description
John Ross Hunt
2004-11-11 15:25:28 UTC
Created attachment 43747 [details]
Difference between /usr/include/linux/device-mapper.h and kernel version
Blarg! File this one under DFU (Dumb F***ing User). The problem arose with changes in glibc-2.3.4 that broke my 'lvm2create_initrd' script. I posted a patch to the forum link above (for those who may be interested) that fixes the problem with files being moved about. The script works now and my gentoo system also boots properly. Case closed. My apologies for the wasted bandwidth. However, I'm still concerned about the difference in device-mapper.h in the kernel and userland. Does anyone think that it may lead to problems? Comments? Suggestions? Flames? One last thing to mention. I'm sticking with 2.6.8-r5, which is the last kernel with identical device-mapper.h between kernel and userland. I haven't tested newer kernels because I want to be sure my data will be safe. I'll leave this discrepency for the bug wranglers to ponder. Sorry again for the wasted bandwidth. The reason that the headers differ is because gentoo-dev-sources has some devicemapper fixes applied. The very small difference in header files won't make any difference at all, but if you like, you can just copy the one from /usr/src/linux into /usr/include/linux - once the dm-fixes get merged upstream, the linux26-headers includes will obviously also be updated |