[Clfs-dev] CLFS GIT PowerPC Boot Method

[William Harrington]

On Aug 3, 2012, at 15:23 PM, William Harrington wrote:

> Hello everybody,
>
> I've been going through different builds of the book checking  
> chroot method and boot methods and right now I am on powerpc 32 bit.
>
> I know there are some users, who may not be subscribed and follow  
> the archives, and am curious to know who has built powerpc recently  
> using the boot method?
>
> I've ran into a few issues and I need them verified to make sure  
> they are not my own.
>
> When building the boot method, e2fsprogs is at 1.42.3  and yaboot  
> is at 1.3.17. Maybe we'll need to provide a static yaboot for  
> people to use.
>
>

This is what I'm getting when going through the boot method using an  
LFS experimental build host where I included a graphite toolchain  
with CLFS dev a few months ago, and using current GIT powerpc book  
build using the boot method and not chroot:

In file included from /mnt/clfs/tools/include/stdlib.h:958:0,
                  from ../../misc/tune2fs.c:42:
/mnt/clfs/tools/include/bits/stdlib-ldbl.h:36:1: error: 'qecvt'  
undeclared here (not in a function)
/mnt/clfs/tools/include/bits/stdlib-ldbl.h:37:1: error: 'qfcvt'  
undeclared here (not in a function)
/mnt/clfs/tools/include/bits/stdlib-ldbl.h:38:1: error: 'qgcvt'  
undeclared here (not in a function)
/mnt/clfs/tools/include/bits/stdlib-ldbl.h:39:1: error: 'qecvt_r'  
undeclared here (not in a function)
/mnt/clfs/tools/include/bits/stdlib-ldbl.h:40:1: error: 'qfcvt_r'  
undeclared here (not in a function)
make[2]: *** [tune2fs.o] Error 1
make[2]: Leaving directory `/mnt/clfs/sources/e2fsprogs-1.42.5/build/ 
misc'
make[1]: *** [all-progs-recursive] Error 1
make[1]: Leaving directory `/mnt/clfs/sources/e2fsprogs-1.42.5/build'
make: *** [all] Error 2
clfs:/mnt/clfs/sources/e2fsprogs-1.42.5/build$

This happens with the books current e2fsprogs 1.42.3 version as well.

I'm not sure where the issue may be. I may go back to eglibc 2.13 and  
see if the problem arises.
Also it could be a gcc 4.7.0 regression I haven't been able to find  
anywhere on the internet.

Could some other people please go through the powerpc book and use  
the boot method and see if they get the same problem?

Sincerely,

William Harrington

PS: I'm going to use a slackintosh host later and see what happens.