Messages in this thread | | | From | "pacman" <> | Subject | Re: bloat thread | Date | Sun, 31 Jan 1999 01:17:29 -0500 (EST) |
| |
Michael Loftis writes the following: > >In any event I'm very disappointed. If this is how many of you still are >it's no wonder scoop (for freshemat.net) was going to call it quits. This >is BS you guys. If you rip on people you don't know, who could possibly >HELP YOU LATER then you, and many others will LOSE OUT.
Don't let a few rude responses discourage you. If you have ideas on how the kernel build process can be improved to provied better support for compiling on a near-full disk, send in your diff -u. Here's what I'd like to see:
+ Ability to build from sources located on a readonly NFS mount, so the target machine only needs to hold its .o's. Maybe you can already do this with a symlink tree. Has anybody tried it? If it worked, then all we'd need would be a group of public untarred NFS exported source trees. lndir /mnt/nfs.us.kernel.org/pub/linux/v.2./LATEST-untarred/ /usr/src/linux cd /usr/src/linux cp ~/mysavedconfig /usr/src/linux/.config make oldconfig clean dep zImage modules modules_install
+ Some convenient alternative to "make modules_install" which makes sense when compiling on a machine other than the target machine. I suggest "make modules_tarball" which creates a tarball containing the lib/modules/2.2.X/*/*.o directory structure.
If there was a patch available which could do these in a non-intrusive way, the big-disk bigots would at least have to think a little bit harder before flaming.
Of course now the natural question to ask is why *I* am not submitting patches :) the answer is that according to Documentation/Changes, I have a dozen other things to upgrade before I can even consider 2.2 on my "small"-disk machine.
-- Alan Curry
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |