On Thu, Dec 23, 2010 at 1:23 PM, Ken Moffat wrote: > On Thu, Dec 23, 2010 at 09:47:33AM -0500, Mike wrote: >> >> On Thu, Dec 23, 2010 at 7:42 AM, William Immendorf wrote: >> >>> On Thu, Dec 23, 2010 at 2:04 AM, Mike <m-jack...@charter.net> wrote: >>>> I am in /sources/gmp-5.0.1 directory. >>>> This is a x86_64 dual core so I am trying to use the ABI=32 option. >>>> I must have something wrong but I don't see what. >>>> Any suggestios? I tried deleteing the directory and starting over >>>> with this >>>> one but got the same thing. >>>> If I understand it correctly I need to use the abi=32 env arg >>> Your problem is that you are building an Pure64 bit system, and on >>> that system, that arg makes no sense. >>> >>> Remove ABI=32 from configure and run it again. >> >> Yes it will compile like that but will it be able to run on pure >> 32-bit system i.e. not x86-64. If I understood it correctly that >> option was to make it so it would. >> Thanks for your reply and help. I started this about a year ago and >> decided I needed to learn quite a bit more in particular about shell >> scripting so I left the idea and studied quite a bit. It has been >> worth it for this alone. > > Iff you have built everything up to this point as 32-bit, you > should not have a problem specifying ABI=32. It sounds as if you > have been building a 64-bit system until you got to gmp. There was two reasons actually. 1 - The section at the beginning of the lfs book seemed to indicate one would be better off with a 32-bit build. That is it is more stable, bette r tested, and so forth. While I might not agree fully these are good points to consider and muse upon. 2 - Yes the thought had cross my mind since I have a few different syste all x86 to build clone many but I guess I should stick to the point, that this is meant to be a learn experence. You know off the subject now, I would also like to explain to who ever thought html mail is such a great idea at my service provider a small personal rant ever time I have to use this style of email.Not haveing a cursor is just the start. In any case the point is moot since on this final pass gcc is failing miserable at the make -k check stage(error 2). I guess I will again have to start over. I have found that cutting and pasting at least from the pdf version has caused some errors. What made it so apparrent was the locale section where it split all the first words out (localedef - ) and then after the 10 or 12 localedef it then pasted all the endings. Any way to all who have help much much thanks is given. and then pasted all the rest of the commands last. > > I don't understand why you would care if the completed LFS system > will run on a 32-bit processor ? If you wish to build something > that you can install onto 32-bit i686 hardware, then build the > 32-bit version of the book, but in all honesty we recommend that you > build it on the machine on which it will run. > > And please note that LFS is either 32 or 64 bit - we don't support > multilib. > > ĸen > -- > das eine Mal als Tragödie, das andere Mal als Farce > -- > http://linuxfromscratch.org/mailman/listinfo/lfs-support > FAQ: http://www.linuxfromscratch.org/lfs/faq.html > Unsubscribe: See the above information page -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
