Baho Utot wrote: > Have a look at this > > https://wiki.archlinux.org/index.php/UEFI > https://wiki.archlinux.org/index.php/GPT > > It looks very complicated just to get win7 and linux to boot with UEFI > and gpt and I am not sure that I can have a 32 bit linux and a 64 bit linux.
I have a single XP instance on an old machine just for rare applications that may need it. It runs in vmware, but if I wanted to create a new instance I'd use gemu. > I think I will just stay with MBR and LVM, much simpler. Perhaps so for existing systems. If you want to use disks > 2T, you need GPT. For any new, unformatted disk (including a new virtual virtual disk), I'll use gpt. > I get the > same thing and don't have to go through all the trouble of setting all > that schist up. Install Win7 64 bit on sda1, install /boot on sda2, > swap gets sda3 and sda4 gets / on LVM, 32 bit PAE kernel (so I can drink > some wine) another LVM root with 64 bit linux then add grub2 and 3 > simple entries to grub.cfg and I am done. > > This works for me because I don't have any hard drives larger than 2TB. > I would like to move to gpt partitions but I don't see the merit for all > the trouble to configure it. Things in the computer realm are not > simple any more too much schist has moved into IT. It was much simpler > in 1996 and look at where most projects are now. I don't call that > progress, just change for the sake of change. Simpler and less capable and more expensive. IIRC the latest was about a 100MHz 386. > Look at you folks trying to get a standard version of LFS scratch out... > before you can finalize a release many more packages have changed. Goes with the territory. > Not > to even consider get BLFS-7.3 version "finished". The question then > becomeis why, do the new packages actually fix something or just change > things ( break things) and add several more pounds of dependencies for > little or no benefit? A little of both. There are people paid to develop open source (e.g. Redhad, SuSE) and they need to be kept busy. I agree that many changes are not for the good of the users, but some changes are needed. For example, systems need to change to take advantage of >2T drives, touch screens, etc. I think debian is on to something. I used my earlier LFS system from 2005 to 2012 without updating. > There is no way I can finish my desktop system....I am now just getting > LFS-7.3 finished and when I get to BLFS there are more packages in LFS > will have changed. I thing is... is it actually good or just change by > a bunch of younger developers who don't even try to understand how Unix > came to be? Just use what you have and only update packages as needed. You don't need every package in BLFS. I rarely build Gnome and then only for testing packages to put in the book. About the only thing that really pushes a new system is a new major version of glibc. Virtually everything else can be upgraded in place. Some users have done glibc in place too, but I've not tried that. -- Bruce -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
