I haven't done "drunkin' computing" for years... As a matter of fact, I haven't gotten drunk in years. I barely drink at all these days.
I was thinking... I used to do "drunkin' programming" once in awhile. It's a fun experience! If you've never programmed while drunk, you're missing out. I think I'm going to try "drunkin' computing" sometime in the near future.
When I'm going to start a new crazy idea such as building a "Linux Distro" I better have a good explanation for doing such a crazy thing. I mean, let's face it, there are literally tons of distributions to choose from. So, why create a new one? My excused could be "I was drunk one night and just started messing around." That would be a good excuse if you ask me.
Anyway, just thought I'd throw the idea of "drunkin' computing" out there.
As a side thought, "I wonder how much software drinking has produced."
Well, I don't have any liquor on hand, so I'm going to have to do this sober for awhile. Maybe next weekend I'll get some booze. We'll have to wait and see. Anyway, onto the next part... Building the LFS System. I thought I'd have my journal up so I can comment on the progress as I go through the motions.
As I read the LFS book, I keep in the front of my mind that I'm going to be using RPM... I just got to the section about chrooting into the LFS environment which contains only the tool-chain. It occurred to me that RPM is just another tool. Perhaps what needs to be done is to build RPM into the tool-chain. Then use the RPM tool from the chrooted environment to install every package. This means that I'd need to prep an RPM Package for every program that is to be installed. I wonder what order the packages need to be installed in. Do I install RPM first and then use the newly installed RPM instead of the one in the tool-chain? Or do I install them all using the tool-chain RPM? I'm sure all of these questions will answer themselves once I get to doing it.
Back to Building the LFS System. I haven't even done anything (except read) yet.
I thought I'd take a short recess from LFS. Some of this stuff it tedious. Right now, I'm manually creating config files using the "cat" command in a terminal window. It's mindless copy what the book says kinda-work. But, it must be done to get an LFS system going. I can't wait until I get to the part where I can experiment with things. I'm going to save my experimenting for LFS-Round-Two though. I figure it would be best to get a working system before trying to change things. So, when I build my second LFS it will be called LFSRPM because I'm going to try to build the basic LFS system using RPM's instead of "make install" scripts.
I'm finally compiling Glibc. The first big program in my new LFS system. I thought while larger programs compile I should post journal entries because all I have to do is wait for them to finish. Might as well take this time for some out-loud thoughts...
Okay. When I build a Busy Box live USB Linux System I ran into a problem of chrooting into the live system. The problem was that I had to recreate my /dev directory. I think I stumbled onto the solution to that. In LFS I had to issue a "mount --bind" command on the /dev directory. This made a duplicate of the directory in question. So, the solution (I think) will be to issue a "mount --bind /dev /plistic/dev" before chrooting into the /plistic directory. I'm not sure exactly if this was the problem or not... I haven't booted my Plistic USB in about a year; I'm just basing this analysis off of faint memories. However, the "mount --bind" command is new to me. See, I learned something already!
I'm running the test-suite for Glibc now. I noticed that SELinux denied some of the tests. The test program (after a long long wait) came back with an error, and I didn't even have to look at the log. SELinux is the problem. I hope that switching it to permissive mode will work. Oh! It just finished. It must not have had to compile as many tests this time around! Time to verify the log file... No errors! Huh, I was expecting at least a couple. Oh well, as long as it works.
No comments:
Post a Comment