Still no luck, i've made a 100% sure i start with a fresh copy of the
extracted gcc.4.7.1.tar.bzp2 and i use a fresh gcc-build directory.
I can give you a summary of my actions if that helps.
1. untar binutils, run the test, apply patch
2. make a binutils-build directory
3. prepare the compile (copy pasting the config prefixes)
4. Compile and copy the command that has to be run afterwards from the book
5. install GMP, MPFR and MPC
6. untar GCC to gcc.4.7.1, switch to the gcc-4.7.1 folder, run the commands
and make the gcc-build folder inside my gcc-4.7.1 folder.
7. Prepare for compille by copy pasting all the commands.
8. Make , ulimit -s 32768, make -k check, check the summary (all good and
matching the log results from the LFS website)
9.Make install
10. create the links by using the "ln" commands while still in the
gcc-build dir.
11. run the test, while still in the gcc-build dir.
12. get messed up results
13. cry myself to sleep holding a little tux.
I'm praying one of u guys/girls will now tell me that i made a simple
mistake and can provide a easy fix because i'm starting to run out of
back-up plans.
regards,
Kaleb
On Thu, Nov 15, 2012 at 9:52 AM, Kaleb van Ingen Schenau <
k.ingen.sche...@gmail.com> wrote:
> make install was run and no errors what so ever, what i'll do is go back
> to my snapshot i made after 6.10, i tried this before but im gonna go ahead
> and do it again just to see if i dont have my old gcc-build folder hanging
> around.
> Ill let you know if i encounter the same problem again after the 3rd
> attempt of getting past this.
>
> You will hear from me soon,
> Kaleb
>
> On Thu, Nov 15, 2012 at 1:49 AM, Ken Moffat <zarniwh...@ntlworld.com>wrote:
>
>> On Thu, Nov 15, 2012 at 12:42:03AM +0100, Kaleb van Ingen Schenau wrote:
>> > Hi, and thanks for the respond
>> >
>> > Throughout the book i've encountered 0 problems or errors and everything
>> > goes according to the book. It just seems weird that everything works as
>> > intended at chapter 6.10 then somewhere between that and compiling
>> > bin-utils and GCC it messes up somewhere, could this be something that i
>> > maybe mistyped earlier and i'm not experiencing the effects from. (Note
>> > that EVERY test till this point have been 100% accurate to what to book
>> > states it should be)
>> > But to anwser your questions:
>> > >>Then when i run:
>> > > >#grep 'SEARCH.*/usr/lib' dummy.log |sed 's|; |\n|g'
>> > >> SEARCH_DIR("/tools/i686-pc-linux-gnu/lib")
>> > >> SEARCH_DIR("/usr/lib")
>> > >> SEARCH_DIR("/lib");
>> > >>
>> > >> which is also not correct
>> >
>> > >That seems to be correct - please compare it to the book - 3
>> > >matches, for /tools/i686-pc-linux-gnu, /usr/lib, /lib. For this
>> > >part I don't see any error - please point it out if I'm wrong!
>> >
>> > The book says it should be:
>> > SEARCH_DIR("/usr/i686-pc-linux-gnu/lib")
>> > SEARCH_DIR("/usr/local/lib")
>> > SEARCH_DIR("/lib")
>> > SEARCH_DIR("/usr/lib");
>> >
>> > So I'm somehow still pointing to /tools while it seems that i should be
>> > searching in /usr and /lib, i think this might a main reason why it
>> isn't
>> > working.
>> >
>> You were right all along, it doesn't match : I'll have to blame the
>> prescription painkillers I'm on (only ibuprofen and paracetomol!).
>> Sorry, no idea how I misread what the book says.
>>
>> > >Are you doing something unusual ? That probably includes building
>> > >on a virtual machine, using the package_users hint, and building on
>> > >mingw or whatever on windows. Or using minimalist packages (dash,
>> > >busybox). If this is a straight build linux from linux, did you
>> > >check the host against the Host System Requirements in the preface ?
>> >
>> > I'm building on a virtual PC running Centos 6.3 which i think fit the
>> Host
>> > System Requirements, other then that I'm not using anything the book
>> > doesn't tell me to and i haven't skipped anything, I'm just doing what
>> the
>> > book is telling me to do. Have the logs been of any help maybe?
>> > pastebin.com/dCjzz5yb <<---- or is there anything else i can supply you
>> > with to help me?
>> >
>> [ snipping my original reply below this - please don't top post, and
>> trim what you are replying to. ]
>>
>> I've no idea about virtual machines, maybe someone else can offer
>> suggestions about specific changes (if any) needed - but I don't
>> recall seeing problems from them apart from selecting the right
>> kernel options.
>>
>> So to recap, hopefully correctly - your include directories from
>> gcc are not as expected, and the SEARCH_DIRs begin in /tools.
>>
>> That last point is correct for 6.10 (hmm, maybe I was looking there
>> instead of 6.17 when I replied).
>>
>> Going back to check my _own_ (64-bit) logs from 7.2 (with a lot of
>> things collected up), in 6.10 I see that I have:
>> ignoring nonexistent directory
>>
>> "/tools/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/../../../../x86_64-unknown-linux-gnu/include"
>> ignoring duplicate directory "/tools/include"
>> #include "..." search starts here:
>> #include <...> search starts here:
>> /usr/include
>> /tools/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/include
>> /tools/include
>> /tools/lib/gcc/x86_64-unknown-linux-gnu/4.7.1/include-fixed
>>
>> and
>> SEARCH_DIR("/tools/x86_64-unknown-linux-gnu/lib64");
>> SEARCH_DIR("/tools/x86_64-unknown-linux-gnu/lib");
>> SEARCH_DIR("/usr/lib"); SEARCH_DIR("/lib");
>> (this time, word-wrapping has mad a long line become 3 individual
>> lines).
>>
>> I *think* that looks as if your results at the end of 6.17 match,
>> or are at least similar to, your results in 6.10 ? Is it possible
>> that you omitted 'make install' after running the tests on gcc ?
>>
>> Other than that, I'm again out of ideas.
>>
>> 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
