Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
Quoting Jarod Wilson <jarod-ajLrJawYSntWk0Htik3J/w at public.gmane.org>: > On Mon, 2008-12-22 at 13:24 -0500, Gregory Boyce wrote: >> On Mon, 22 Dec 2008, Jarod Wilson wrote: >> >> > This is *exactly* what /etc/kernel/postinst.d/ is designed to handle. >> > Its present in RHEL, Fedora and Ubuntu, not sure about others. >> > Basically, drop a scriptlet in there, and it'll get executed >> > post-kernel-install (there's also a corresponding prerm.d). I can't >> > remember if akmods wire into this or not, but it was actually Matt >> > Domsch over at Dell (where dkms originated) who asked that Fedora and >> > RHEL add the /etc/kernel/*.d/ directories and hooks in the kernel >> > packages for exactly this sort of thing. >> >> Ah, cool. Its been a few years since I've compiled my own kernel rather >> than running the vendor provided kernel. > > Huh? > > This is *for* vendor-provided kernels. In the Fedora case, the kernel > rpm's %post section scours /etc/kernel/postinst.d/ for scripts to run, > such as "rebuild the dkms-ified driver for my wacky device that doesn't > have in-tree support". I believe this part is done just before initrd > generation, so that such things can also be included in the initrd, if > need be. How does it make sure that the kernel-devel package is installed prior to this? The kernel package doesn't depend on the -devel package. -derek -- Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory Member, MIT Student Information Processing Board (SIPB) URL: http://web.mit.edu/warlord/ PP-ASEL-IA N1NWH warlord-DPNOqEs/LNQ at public.gmane.org PGP key available
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |