I'm in need of a brain recompile !! - ignore all that - all I did was inadvertently boot into the (wrong) 2.4.27 kernel which of course has my doctored loop.o from before - I then thought I'd fixed it. But it isn't fixed, as I saw on a subsequent reboot. So it wasn't the symlink or LINUX_SOURCE switch. uname -r should have told me what was going on. Brain downtime is in order (sigh!). --- Phil H <philtickle200@xxxxxxxxx> wrote: > > > > I'm sure this is something embarrassingly obvious > > but ...? > > > Ah - I've worked out what was wrong! I had deleted > the "linux" symlink in /usr/src before compiling > loop.o, and I didn't specify LINUX_SOURCE explicitly > as a "make" command switch. > > My thanks for everybody's patience. Not sure > whether > the fault was one or both of these, but after > re-doing > with the symlink back in and LINUX_SOURCE specified, > I > now have a working loop.o on this kernel, without > doing any dubious manipulations. > > One more thing though (sorry!): > > The debian source came from www.backports.org and > was > kernel-source-2.4.26_2.4.26-6_all.deb. In fact I > was > actually looking for the deb source for the > downloaded > kernel image from the same place, which was > kernel_image-2.4.26-1-386_2.4.26-5.backports.org.1_i386.deb > That kernel source was all that was there for > 2.4.26. > I posted to the backports mail list to check that I > did in fact have the right source, and was told that > this source did in fact pertain to my downloaded > kernel image, despite the fact that one has a "-5" > and > the other a "-6" in the handles. I didn't follow > the > explanation as to why there were different numbers: > one had been rebuilt and the other hadn't been, or > something. > > "fakesource" then created source code from this with > the directory name > /usr/src/kernel-source-2.4.26-1-386. > > But the kernel that actually runs from > kernel_image-2.4.26-1-386_2.4.26-5.backports.org.1_i386.deb > is (according to uname -r) 2.4.27-2-386 (note the > "2"). > > This led to a warning at the end of building > loop-aes-ciphers that the modules were built for > 2.4.26-1-386 but the running kernel was 2.4.26-2-386 > and that these wouldn't run (although they do). > > How significant are these appended kernel build > numbers on the same version kernel, especially when > one or both numbers have been generated by > fakesource/backports? > > Should I still be concerned? - all tests have been > passed so far. > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam > protection around > http://mail.yahoo.com > ____________________________________________________ Yahoo! Sports Rekindle the Rivalries. Sign up for Fantasy Football http://football.fantasysports.yahoo.com - Linux-crypto: cryptography in and on the Linux system Archive: http://mail.nl.linux.org/linux-crypto/
