Hi Maybe you mistook a development kernel for a stable series. Development kernels are known to break close to everything except for your nose because the regression test you mentioned is taking place right before it. Thanks for the report, keep up the good work. Gerold On Wednesday 19 February 2003 11:49, Michael Svetlik wrote: > Hi, again. > > It's not the first time, make exits with $?>0 within a kernel tree, and I'm > getting fed up with digging those Makefile's, config.somewhat's and > ./script/'s. > > It's a pure, stock, unmodified linux-2.5.59.tar.bz2 file from kernel.org, I > have extracted into my home. I say 'make xconfig' - a well known command. > > First, make moans about a non-existant g++, needed to compile and link > some stuff in ./scripts. 'make' is right here, I dont have a c++ compiler > with that name, but I dont need one - the kernel is written in ANSI-C. > Ok, make love, not war, and do a symlink to some g++-2.95.3 lying around. > > Next, make runs a while, until it crashes with SEGV, after having dumped > tons of lines like > > >> /boot/config-2.4.18-3:1893: trying to assign nonexistent symbol ... > > Somewhere before this error orgasm I see the line > > >> # using defaults found in /boot/config-2.4.18-3 > > Booom. Why the hell is that make digging in /boot on my disk ? I say > 'make xconfig', and I know what I do. I didn't say 'make oldconfig' > or 'make randomconfig' or whatever. > Ok. There's a stale RH kernel config file in there, but: I translated a lot > of 2.4.xx, some 2.5.xx, and all went well - as expected. Any regression > test should have revealed that mis-behaviour - if performed with > according vectors. > > I've checked the in-tree Docs - ok, I did 'grep -rw -A 2 -B 2 boot * | grep > config' - and there was no word about that config-defaulting. So, who in > the world _REALLY_ knows, how a Linux kernel is to be made ? If that person > exists, would she/he be so pleasant and share here/his knowledge ? > > Am I too bigheaded, if I would volunteer for regression tesing ? -- The observed values of all physical and cosmological quantities are not equally probable but they take on values restricted by the requirement that there exist sites where carbon-based life can evolve and by the requirement that the Universe be old enough for it to have already done so. -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
