Re: [lfs-support] Question about the scsi driver for LFS 8.4 systemd.

[Thanos Baloukas]

On 3/25/19 8:54 AM, niuneilneo wrote:

Dear All,

I am now installing LFS 8.4 systemd in a Vmware guest machine, and the 
Linux kernel is unable to find my hard disk on a device with name-"SCSI 
storage controller: LSI Logic / Symbios Logic 53c1030 PCI-X Fusion-MPT 
Dual Ultra320 SCSI (rev 01)", which is provided by the lspci command by 
the Ubuntu host for the same guest machine.


My question is which module should I integrate to the kernel to enable 
LFS to find my hard disk?



'lspci -v' should show which kernel module is used. You can also look
into the host's initramfs.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] compile 4.17.7 32 bit

[Thanos Baloukas]

On 17/07/2018 06:59 μμ, Thomas Seeling wrote:

Hallo,


I am compiling the current linux kernels for my 32 and 64 bit machines.
The latest 4.17.7 does not work for me on 32 bit (intel core2 duo) while
it boots fine on the i5 64 bit.
I used the same config as for 4.17.6 which compiles and boots fine on
the 32 bit CPU.
LFS 8.2, gcc 7.3.

Anyone else having problems with the latest kernel under some circumstances?



I'm not sure if intel core2 duo is i386, but you may find this lkml post
interesting:

https://lkml.org/lkml/2018/7/17/434

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Booting LFS with systemd

[Thanos Baloukas]

On 28/06/2018 10:44 μμ, Frans de Boer wrote:

On 06/28/2018 04:21 PM, Hazel Russman wrote:

On Thu, 28 Jun 2018 16:06:00 +0800
Xi Ruoyao  wrote:


On 2018-06-28 01:08 -0400, Michael Shell wrote:

On Wed, 27 Jun 2018 14:42:47 -0700
Paul Rogers  wrote:

If that's true, even with systemd, why is there any need to build an
initramfs for a known system?
I had used initramfs to setup a loopback device and boot the system 
in an

image.  But it seems grub can handle loopback device (though I've never
tried).


Just like you, I build everything I need into a custom kernel and avoid
the need for an initramfs. One other reason people use initramfs is if
they need udev services on boot, say, for a drive the kernel will 
not be

able to find via a simple specification of root=/dev/X.

I think people should not go through all the initramfs trouble just for
LABEL= or UUID= functionality, but rather should just use PARTUUID=
which the kernel natively understands.
Agree.  Now I only use "initrd" directive to update CPU microcode and 
fix

the buggy ACPI DSDT of my laptop (another sad story).
--
Xi Ruoyao 
School of Aerospace Science and Technology, Xidian University
--
Nice to know someone else does this. I use an initrd on my main 
machine for precisely these two purposes. I had hoped that rewriting 
the acpi dsdt to remove some reported errors would allow me finally to 
boot the latest kernels, which are giving me panic in the acpi driver, 
but no such luck!


Mind you, I'm not using systemd.

Great, new development, now I can't even install systemd due to the next 
error:

...
RuntimeError: File 'man/binfmt.d.5' could not be found
FAILED: meson-install

When looking for the missing file, I only found 'man/binfmt.d.xml'. I 
also refreshed the two systemd-238 packages to exclude a fallen bit, to 
no avail. Any suggestion where this might come from?




Did you extract the man pages?
tar -xf /path/to/systemd-man-pages-238.tar.xz

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] LFS 7.10 toolchain problem (WAS LFS 7.10 - libreadline.so.6 not found error - when running make in 6.40 Perl-5.24.0)

[Thanos Baloukas]

On 06/04/2018 10:51 πμ, Νίκος Λαμπόβας wrote:



On 5 April 2018 at 19:12, Thanos Baloukas <mailto:baloukast...@sch.gr>> wrote:


On 05/04/2018 12:38 μμ, Νίκος Λαμπόβας wrote:

On 25 February 2018 at 11:35, Thanos Baloukas
mailto:baloukast...@sch.gr>
     On 02/25/2018 10:58 AM, Νίκος Λαμπόβας wrote:

         Thank you for the quick answer.
         As said in the email subject the book is LFS, the
version is
         7.10 and the Host is a virtual machine with Ubuntu
12.04 LTS.
         The Oracle VM Virtualbox is running on Windows 10.

...
snip
...
Firstly, is there a reason you are installing the outdated LFS 7.10
and not the latest 8.2?


     i started lfs when lfs 7.10 was the new version but i postponed it 
in order to build it parallel to my OS course of NTUA


Did you check carefully with version-check.sh

     bash, version 4.2.25(1)-release
/bin/sh -> /bin/bash
Binutils: (GNU Binutils for Ubuntu) 2.22
bison (GNU Bison) 2.5
/usr/bin/yacc -> /usr/bin/bison.yacc
bzip2,  Version 1.0.6, 6-Sept-2010.
Coreutils:  8.13
diff (GNU diffutils) 3.2
find (GNU findutils) 4.4.2
GNU Awk 4.1.1, API: 1.1 (GNU MPFR 3.1.0-p3, GNU MP 5.0.2)
/usr/bin/awk -> /usr/bin/gawk
gcc (Ubuntu/Linaro 4.7.3-2ubuntu1~12.04) 4.7.3
g++ (Ubuntu/Linaro 4.7.3-2ubuntu1~12.04) 4.7.3
(Ubuntu EGLIBC 2.15-0ubuntu10.6) 2.15
grep (GNU grep) 2.10
gzip 1.4
Linux version 3.13.0-32-generic
m4 (GNU M4) 1.4.16
GNU Make 3.81
patch 2.6.1
Perl version='5.14.2';
GNU sed version 4.2.1
tar (GNU tar) 1.26
makeinfo (GNU texinfo) 4.13
xz (XZ Utils) 5.1.0alpha
g++ compilation OK



Not sure about EGLIBC, the rest look good.
LFS-7.10 had library-check.sh too.



I have built lfs all over again until "6.40 perl" for watching
all warnings and possible package errors.
The commands you mentioned on 'Creating Essential Files and
Symlinks' were executed successfully.
Also there is no reference of 'tools' in gcc specs file on '
Adjusting the toolchain' and all debug messages
for program interpreter are mention /lib/ld-linux.so.2 and not
/tools/lib/ld-linux.so.2.
A strange thing i have noticed is that all binaries of the
packages before 6.17 GCC-6.2.0, have correct paths
on ldd command. For example:

    ldd /usr/bin/file
  linux-gate.so.1 (0xb773a000)
  libmagic.so.1 => /usr/lib/libmagic.so.1 (0xb771)
  libz.so.1 => /lib/libz.so.1 (0xb76f5000)
  libc.so.6 => /lib/libc.so.6 (0xb753e000)
  /lib/ld-linux.so.2 (0xb773b000)

But after GCC-6.2.0 all the libraries point to 'tools'. For example:

    ldd /bin/bzip2
  linux-gate.so.1 (0xb7747000)
  libbz2.so.1.0 => not found
  libc.so.6 => /tools/lib/libc.so.6 (0xb758e000)
  /tools/lib/ld-linux.so.2 (0xb7748000)

When i execute "LD_DEBUG=libs ldd /usr/bin/file" a binary
compiled before GCC-6.2.0 i find that it searches
in the beginning the system path that refers to a directory
relative to "tools".

LD_DEBUG=libs ldd /usr/bin/file
   11241:    find library=libdl.so.2 [0]; searching
   11241:     search cache=/tools/etc/ld.so.cache
   11241:     search
path=/tools/lib/tls/i686/sse2:/tools/lib/tls/i686:/tools/lib/tls/sse2:/tools/lib/tls:/tools/lib/i686/sse2:/tools/lib/i686:/tools/lib/sse2:/tools/lib 
    (system search path)

   11241:      trying file=/tools/lib/tls/i686/sse2/libdl.so.2
   11241:      trying file=/tools/lib/tls/i686/libdl.so.2
   11241:      trying file=/tools/lib/tls/sse2/libdl.so.2
   11241:      trying file=/tools/lib/tls/libdl.so.2
   11241:      trying file=/tools/lib/i686/sse2/libdl.so.2
   11241:      trying file=/tools/lib/i686/libdl.so.2
   11241:      trying file=/tools/lib/sse2/libdl.so.2
   11241:      trying file=/tools/lib/libdl.so.2
   11241:
   11241:    find library=libc.so.6 [0]; searching
   11241:     search cache=/tools/etc/ld.so.cache
   11241:     search path=/tools/lib        (system search path)
   11241:      trying file=/tools/lib/libc.so.6
   11241:
   11241:    calling init: /tools/lib/libc.so.6
   11241:    calling init: /tools/lib/libdl.so.2
   11241:    initialize program: /bin/bash
   11241:    transferring control: /bin/bash
   11241:
   11241:    find library=libnss_files.so.2 [0]; searching
   11241:     search cache=/tools/etc/ld.so.cache
   11241:     search path=/tools/lib 

Re: [lfs-support] LFS 7.10 toolchain problem WAS LFS 7.10 - libreadline.so.6 not found error - when running make in 6.40 Perl-5.24.0

[Thanos Baloukas]

On 05/04/2018 12:38 μμ, Νίκος Λαμπόβας wrote:
On 25 February 2018 at 11:35, Thanos Baloukas 

On 02/25/2018 10:58 AM, Νίκος Λαμπόβας wrote:

Thank you for the quick answer.
As said in the email subject the book is LFS, the version is
7.10 and the Host is a virtual machine with Ubuntu 12.04 LTS.
The Oracle VM Virtualbox is running on Windows 10.


...
snip
...


         when running make in 6.40 in Perl i get the following
error a
         lot of times:

         sh: error while loading shared libraries: libreadline.so.6:
         cannot open shared object file: No such file or directory

         Before running make for 6.33 Bash this was the output
of ldconfig:

         ldconfig -p | grep libreadline
           libreadline.so.6 (libc6) => /lib/libreadline.so.6
           libreadline.so (libc6) => /usr/lib/libreadline.so

         After running make of 6.33 Bash i executed "exec /bin/bash
         --login +h" as the doc says the new bash with readline
was in place.

         root:/# echo $0
         /bin/bash
         root:/# ldd /bin/bash
           linux-gate.so.1 (0xb77c7000)
           libreadline.so.6 => /lib/libreadline.so.6
(0xb7783000)
           libhistory.so.6 => /lib/libhistory.so.6 (0xb7778000)
           libncursesw.so.6 => /lib/libncursesw.so.6
(0xb770d000)
           libdl.so.2 => /lib/libdl.so.2 (0xb7708000)
           libc.so.6 => /lib/libc.so.6 (0xb7551000)
           /tools/lib/ld-linux.so.2 (0xb77c8000)


At that point the loader should be the one from /lib and not from
/tools/lib . Perhaps you did not do
mv -vf /usr/bin/bash /bin
on bash installation and /bin/bash is still a link to /tools/bin/bash
which was created on 'Creating Essential Files and Symlinks'.
In chroot do
ls -l /bin/bash /bin/sh





Hi thanos,  "xronia polla" (greek wishes for easter holidays)


Hi Nikos, "xronia polla"

Firstly, is there a reason you are installing the outdated LFS 7.10
and not the latest 8.2? Did you check carefully with version-check.sh
if your ubuntu host fulfills the host system requirements?

I have built lfs all over again until "6.40 perl" for watching all 
warnings and possible package errors.
The commands you mentioned on 'Creating Essential Files and Symlinks' 
were executed successfully.
Also there is no reference of 'tools' in gcc specs file on ' Adjusting 
the toolchain' and all debug messages
for program interpreter are mention /lib/ld-linux.so.2 and not 
/tools/lib/ld-linux.so.2.
A strange thing i have noticed is that all binaries of the packages 
before 6.17 GCC-6.2.0, have correct paths

on ldd command. For example:

   ldd /usr/bin/file
     linux-gate.so.1 (0xb773a000)
     libmagic.so.1 => /usr/lib/libmagic.so.1 (0xb771)
     libz.so.1 => /lib/libz.so.1 (0xb76f5000)
     libc.so.6 => /lib/libc.so.6 (0xb753e000)
     /lib/ld-linux.so.2 (0xb773b000)

But after GCC-6.2.0 all the libraries point to 'tools'. For example:

   ldd /bin/bzip2
     linux-gate.so.1 (0xb7747000)
     libbz2.so.1.0 => not found
     libc.so.6 => /tools/lib/libc.so.6 (0xb758e000)
     /tools/lib/ld-linux.so.2 (0xb7748000)

When i execute "LD_DEBUG=libs ldd /usr/bin/file" a binary compiled 
before GCC-6.2.0 i find that it searches
in the beginning the system path that refers to a directory relative to 
"tools".


LD_DEBUG=libs ldd /usr/bin/file
  11241:    find library=libdl.so.2 [0]; searching
  11241:     search cache=/tools/etc/ld.so.cache
  11241:     search 
path=/tools/lib/tls/i686/sse2:/tools/lib/tls/i686:/tools/lib/tls/sse2:/tools/lib/tls:/tools/lib/i686/sse2:/tools/lib/i686:/tools/lib/sse2:/tools/lib
     (system search path)

  11241:      trying file=/tools/lib/tls/i686/sse2/libdl.so.2
  11241:      trying file=/tools/lib/tls/i686/libdl.so.2
  11241:      trying file=/tools/lib/tls/sse2/libdl.so.2
  11241:      trying file=/tools/lib/tls/libdl.so.2
  11241:      trying file=/tools/lib/i686/sse2/libdl.so.2
  11241:      trying file=/tools/lib/i686/libdl.so.2
  11241:      trying file=/tools/lib/sse2/libdl.so.2
  11241:      trying file=/tools/lib/libdl.so.2
  11241:
  11241:    find library=libc.so.6 [0]; searching
  11241:     search cache=/tools/etc/ld.so.cache
  11241:     search path=/tools/lib        (system search path)
  11241:      trying file=/tools/lib/libc.so.6
  11241:
  11241:    calling init: /tools/lib/libc.so.6
  11241:    calling init: /tools/lib/libdl.so.2
  11241:    initialize program: /bin/bash
  11241:    transferring control: /bin/bash
  1124

Re: [lfs-support] LFS 7.10 - libreadline.so.6 not found error - when running make in 6.40 Perl-5.24.0

[Thanos Baloukas]

On 02/25/2018 10:58 AM, Νίκος Λαμπόβας wrote:
Please don't top post

Hello Thano,

Thank you for the quick answer.
As said in the email subject the book is LFS, the version is 7.10 and 
the Host is a virtual machine with Ubuntu 12.04 LTS.

The Oracle VM Virtualbox is running on Windows 10.

In Perl when running make i get the following error which occurs a lot 
of times (more than 50); right after when make runs miniperl:


cc -c -DPERL_CORE -fwrapv -fno-strict-aliasing -pipe 
-fstack-protector-strong -I/usr/local/include -D_LARGEFILE_SOURCE 
-D_FILE_OFFSET_BITS=64 -std=c89 -O2 -Wall 
-Werror=declaration-after-statement -Wextra -Wc++-compat -Wwrite-strings 
time64.c
cc -c -DPERL_CORE -fwrapv -fno-strict-aliasing -pipe 
-fstack-protector-strong -I/usr/local/include -D_LARGEFILE_SOURCE 
-D_FILE_OFFSET_BITS=64 -std=c89 -O2 -Wall 
-Werror=declaration-after-statement -Wextra -Wc++-compat -Wwrite-strings 
miniperlmain.c

cc -fstack-protector-strong -L/usr/local/lib -o miniperl \
     opmini.o perlmini.o  gv.o toke.o perly.o pad.o regcomp.o dump.o 
util.o mg.o reentr.o mro_core.o keywords.o hv.o av.o run.o pp_hot.o sv.o 
pp.o scope.o pp_ctl.o pp_sys.o doop.o doio.o regexec.o utf8.o taint.o 
deb.o universal.o globals.o perlio.o perlapi.o numeric.o mathoms.o 
locale.o pp_pack.o pp_sort.o caretx.o dquote.o time64.o  miniperlmain.o  
-lm /tools/lib/libcrypt.a
./miniperl -w -Ilib -Idist/Exporter/lib -MExporter -e '' || sh -c 
'echo >&2 Failed to build miniperl.  Please run make minitest; exit 1'
*sh: error while loading shared libraries: libreadline.so.6: cannot open 
shared object file: No such file or directory*

./miniperl -Ilib -f write_buildcustomize.pl <http://write_buildcustomize.pl>
./miniperl -Ilib configpm
written lib/Config.pod
updated lib/Config.pm
updated lib/Config_heavy.pl
./miniperl -Ilib make_ext.pl <http://make_ext.pl> 
cpan/Archive-Tar/pm_to_blib  MAKE="make" LIBPERL_A=libperl.a

make[1]: Entering directory '/sources/perl-5.24.0/cpan/Archive-Tar'
Makefile out-of-date with respect to ../../lib/Config.pm ../../config.h
Cleaning current config before rebuilding Makefile...
make -f Makefile.old clean > /dev/null 2>&1
"../../miniperl" "-I../../lib" "-I../../lib" Makefile.PL 
"INSTALLDIRS=perl" "INSTALLMAN1DIR=none" "INSTALLMAN3DIR=none" 
"PERL_CORE=1" "LIBPERL_A=libperl.a"
*sh: error while loading shared libraries: libreadline.so.6: cannot open 
shared object file: No such file or directory
sh: error while loading shared libraries: libreadline.so.6: cannot open 
shared object file: No such file or directory

*

Thank you


On 24 February 2018 at 23:36, Thanos Baloukas <mailto:baloukast...@sch.gr>> wrote:


On 02/24/2018 10:37 PM, Νίκος Λαμπόβας wrote:

Hello,

when running make in 6.40 in Perl i get the following error a
lot of times:

sh: error while loading shared libraries: libreadline.so.6:
cannot open shared object file: No such file or directory

Before running make for 6.33 Bash this was the output of ldconfig:

ldconfig -p | grep libreadline
  libreadline.so.6 (libc6) => /lib/libreadline.so.6
  libreadline.so (libc6) => /usr/lib/libreadline.so

After running make of 6.33 Bash i executed "exec /bin/bash
--login +h" as the doc says the new bash with readline was in place.

root:/# echo $0
/bin/bash
root:/# ldd /bin/bash
  linux-gate.so.1 (0xb77c7000)
  libreadline.so.6 => /lib/libreadline.so.6 (0xb7783000)
  libhistory.so.6 => /lib/libhistory.so.6 (0xb7778000)
  libncursesw.so.6 => /lib/libncursesw.so.6 (0xb770d000)
  libdl.so.2 => /lib/libdl.so.2 (0xb7708000)
  libc.so.6 => /lib/libc.so.6 (0xb7551000)
  /tools/lib/ld-linux.so.2 (0xb77c8000)


At that point the loader should be the one from /lib and not from
/tools/lib . Perhaps you did not do
mv -vf /usr/bin/bash /bin
on bash installation and /bin/bash is still a link to /tools/bin/bash
which was created on 'Creating Essential Files and Symlinks'.
In chroot do
ls -l /bin/bash /bin/sh


I dont know why i am getting this error. Any help?

Thanks a lot
Nikos

Which book? Which version? Which host?


--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] LFS 7.10 - libreadline.so.6 not found error - when running make in 6.40 Perl-5.24.0

[Thanos Baloukas]

On 02/24/2018 10:37 PM, Νίκος Λαμπόβας wrote:

Hello,

when running make in 6.40 in Perl i get the following error a lot of times:

sh: error while loading shared libraries: libreadline.so.6: cannot open 
shared object file: No such file or directory


Before running make for 6.33 Bash this was the output of ldconfig:

ldconfig -p | grep libreadline
     libreadline.so.6 (libc6) => /lib/libreadline.so.6
     libreadline.so (libc6) => /usr/lib/libreadline.so

After running make of 6.33 Bash i executed "exec /bin/bash --login +h" 
as the doc says the new bash with readline was in place.


root:/# echo $0
/bin/bash
root:/# ldd /bin/bash
     linux-gate.so.1 (0xb77c7000)
     libreadline.so.6 => /lib/libreadline.so.6 (0xb7783000)
     libhistory.so.6 => /lib/libhistory.so.6 (0xb7778000)
     libncursesw.so.6 => /lib/libncursesw.so.6 (0xb770d000)
     libdl.so.2 => /lib/libdl.so.2 (0xb7708000)
     libc.so.6 => /lib/libc.so.6 (0xb7551000)
     /tools/lib/ld-linux.so.2 (0xb77c8000)

I dont know why i am getting this error. Any help?

Thanks a lot
Nikos


Which book? Which version? Which host?

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Configure: error: cannot compute suffix of object files: cannot compile LFS 8.1

[Thanos Baloukas]

On 02/01/2018 04:08 μμ, Pavel Drankov wrote:

Hi,

This is my first time writing anything here.

I faced an issuewhile compiling GCC 7.2.0 according to chapter 5.5 [1].


Hi Pavel

When you are asking about such problems it will be helpful to post the 
output of version-check.sh. See below.

...


and then:

make

While doing it I were login as /lfs/. And here the error appeared:

make[3]: Leaving directory '/mnt/lfs/sources/gcc-build/libbacktrace'
make[2]: Leaving directory '/mnt/lfs/sources/gcc-build/libbacktrace'
make[2]: Entering directory '/mnt/lfs/sources/gcc-build/libcpp'
test -f config.h || (rm -f stamp-h1 && make stamp-h1)
make[2]: Leaving directory '/mnt/lfs/sources/gcc-build/libcpp'
make[2]: Entering directory '/mnt/lfs/sources/gcc-build/libdecnumber'
make[2]: Nothing to be done for 'all'.
make[2]: Leaving directory '/mnt/lfs/sources/gcc-build/libdecnumber'
make[2]: Entering directory '/mnt/lfs/sources/gcc-build/gcc'
g++ -no-pie   -g -O2 -DIN_GCC  -DCROSS_DIRECTORY_STRUCTURE
  -fno-exceptions -fno-rtti -fasynchronous-unwind-tables -W -Wall
-Wno-narrowing -Wwrite-strings -Wcast-qual -Wmissing-format-attribute
-Woverloaded-virtual -pedantic -Wno-long-long -Wno-variadic-macros
-Wno-overlength-strings   -DHAVE_CONFIG_H -static-libstdc++
-static-libgcc  -o xg++ \
   gcc.o gcc-main.o ggc-none.o cp/g++spec.o driver-i386.o
libcommon-target.a \
libcommon.a ../libcpp/libcpp.a
  ../libbacktrace/.libs/libbacktrace.a ../libiberty/libiberty.a
../libdecnumber/libdecnumber.a
g++: fatal error: unknown spec function 'gt'
compilation terminated.
../../gcc-7.2.0/gcc/cp/Make-lang.in:59: recipe for target 'xg++' failed
make[2]: *** [xg++] Error 1
make[2]: Leaving directory '/mnt/lfs/sources/gcc-build/gcc'
Makefile:4214: recipe for target 'all-gcc' failed
make[1]: *** [all-gcc] Error 2
make[1]: Leaving directory '/mnt/lfs/sources/gcc-build'
Makefile:879: recipe for target 'all' failed
make: *** [all] Error 2

Googling for sometimes I decided to switch to /root/ and continue.
Executing make again:


Running the commands on chapter 5 while not being logged as user lfs
is wrong. You must follow exactly the book instructions. Running make
on the same source tree that had previously failed is wrong anyway.
First you have to find the cause of the error and fix that.
Then remove the whole source tree including the build directory,
extract the tarball again and continue.

Google suggests broken environment, so that makes me suspect that
something went wrong on
http://www.linuxfromscratch.org/lfs/view/stable/chapter04/settingenvironment.html

As user lfs run the command 'env'
The output must be only the variables referred in the aforementioned page.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Mistake in LFS book

[Thanos Baloukas]

On 10/10/2017 10:59 μμ, Pierre Labastie wrote:

On 10/10/2017 21:35, Thanos Baloukas wrote:

On 10/10/2017 07:29 μμ, Bruce Dubbs wrote:

Bruce Dubbs wrote:

Pierre Labastie wrote:



On 10/10/2017 13:55, Николай Сосницкий wrote:

10.10.2017, 14:42, "Pierre Labastie" :

On 10/10/2017 13:11, Николай Сосницкий wrote:

  Hello everybody!
  I found little mistake in LFS book (both sysvinit and systemd):
  in Capter 6 section Coreutils the last command
  "mv -v /usr/bin/{head,sleep,nice,test,[} /bin"
  has wrong operand "[".
  This mistake is not present in 7.5 and early versions and
present in
  7.6 and later include current stable:

http://www.linuxfromscratch.org/lfs/downloads/stable/LFS-BOOK-8.1-NOCHUNKS.html#ch-system-coreutils




  P. S. Sorry for my english, please.

"[" is the name of an executable. It does the same thing as "test".
Note
that it is not used much, since "[" is built in the sh or bash
shell.
But since we move "test", we move "[".


Hello, Pierre!
I'm sorry for this. I asked in #lfs-support IRC cannel and have
advice
to post in this mailing list.


Don't be sorry Nickolay: this is exactly what lfs-support is for.
Asking
questions! Actually, the text in the sentence above the command you
cite
may be misleading, since it mention only "head", "sleep", and
"nice". It
should also mention "test" and "["... Also, I doubt "[" is useful in
/bin,
since bash has it built in.


bash does include [ as an internal command, but I'm not sure if all
other
shells do.  However testing dash and zsh indicate that they do.  I
really
don't think /bin/[ is needed at all any more.  We can say the same
thing
about /bin/test.

I suspect the only reason for these commands any more are for really
old
scripts that may hard code the path for these commands.

Maybe ash needs them.  ash is used during installation/rescue of
debian. I
do not think that is a good choice any more.


A little more research...  busybox/ash does indeed need them.  I don't
think lfs needs them, but they come for 'free' with coreutils.


In regard to their utility on lfs, they are specified by POSIX also

http://pubs.opengroup.org/onlinepubs/9699919799/utilities/test.html

so I think that lfs needs them as long as it follows this standard,
as is stated on

http://www.linuxfromscratch.org/lfs/view/development/prologue/standards.html




Hmm, I'd say that the question is whether they should be moved from
/usr/bin to /bin, not whether they should be on the system. Applications
in /bin are for early boot process before /usr is mounted when it is on
a separate partition. In LFS, even in early boot, sh/bash is used, and
test and [ are built in commands, so neither test nor [ are needed in
/bin. But they can be in /usr/bin, it does not hurt... Concerning POSIX,
I am not sure it is specified that test and [ should be commands
independent on the shell or can be built in. If the second, removing
"test" and "[" would still be conformant if bash is present...

But my point was just that the sentence above the "mv" command on the
coreutils page should mention "test" and "[" ;)


Your point was obvious to me, but Bruce wrote:

A little more research...  busybox/ash does indeed need them.  I don't
think lfs needs them, but they come for 'free' with coreutils.

which is another matter, and my answer was about that.
Anyway, I think that Bruce clarified things in his last message.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Mistake in LFS book

[Thanos Baloukas]

On 10/10/2017 07:29 μμ, Bruce Dubbs wrote:

Bruce Dubbs wrote:

Pierre Labastie wrote:



On 10/10/2017 13:55, Николай Сосницкий wrote:

10.10.2017, 14:42, "Pierre Labastie" :

On 10/10/2017 13:11, Николай Сосницкий wrote:

  Hello everybody!
  I found little mistake in LFS book (both sysvinit and systemd):
  in Capter 6 section Coreutils the last command
  "mv -v /usr/bin/{head,sleep,nice,test,[} /bin"
  has wrong operand "[".
  This mistake is not present in 7.5 and early versions and
present in
  7.6 and later include current stable:

http://www.linuxfromscratch.org/lfs/downloads/stable/LFS-BOOK-8.1-NOCHUNKS.html#ch-system-coreutils



  P. S. Sorry for my english, please.

"[" is the name of an executable. It does the same thing as "test".
Note
that it is not used much, since "[" is built in the sh or bash shell.
But since we move "test", we move "[".


Hello, Pierre!
I'm sorry for this. I asked in #lfs-support IRC cannel and have advice
to post in this mailing list.


Don't be sorry Nickolay: this is exactly what lfs-support is for. Asking
questions! Actually, the text in the sentence above the command you cite
may be misleading, since it mention only "head", "sleep", and "nice". It
should also mention "test" and "["... Also, I doubt "[" is useful in
/bin,
since bash has it built in.


bash does include [ as an internal command, but I'm not sure if all other
shells do.  However testing dash and zsh indicate that they do.  I really
don't think /bin/[ is needed at all any more.  We can say the same thing
about /bin/test.

I suspect the only reason for these commands any more are for really old
scripts that may hard code the path for these commands.

Maybe ash needs them.  ash is used during installation/rescue of
debian. I
do not think that is a good choice any more.


A little more research...  busybox/ash does indeed need them.  I don't
think lfs needs them, but they come for 'free' with coreutils.


In regard to their utility on lfs, they are specified by POSIX also

http://pubs.opengroup.org/onlinepubs/9699919799/utilities/test.html

so I think that lfs needs them as long as it follows this standard,
as is stated on

http://www.linuxfromscratch.org/lfs/view/development/prologue/standards.html

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Anyone tried the AMD Ryzen 7 1700 processor on LFS?

[Thanos Baloukas]

On 07/08/2017 07:09 μμ, Bruce Dubbs wrote:

Thanos Baloukas wrote:

On 06/08/2017 07:56 μμ, Ken Moffat wrote:

On Sun, Aug 06, 2017 at 06:38:45PM +0300, Thanos Baloukas wrote:

On 06/08/2017 05:57 μμ, Alan Feuerbacher wrote:

On 8/5/2017 10:58 PM, Thanos Baloukas wrote:


I'm going to buy a new system to speed up lfs builds. From what I
read
the new AMD Ryzen 7 1700 would be a good choice for this job. Has
anyone
tested this processor on lfs?


In early May I bought a new ASUS TUF motherboard, Intel 3770
processor,
and 32G of ram. LFS builds quickly on this new system. I would think
that the new AMD processor would be similar.


Thanks. My concern is not that much on speed, various benchmarks
I've seen show it's fast enough. What I have in mind is stability
and compatibility. Segmentation faults, software produced not running
on other machines and such.


For not running on other machines, my usual answer is to blame gmp.


Except from gmp, I thought that until gcc catch up with Ryzen,
such problems would arise.


If the problem was illegal instruction, I would agree with you, but it's
hard to see how the compiler that works with Intel and other AMD
processors would case segmentation faults.


My suspicion was that if Ryzens have new instructions, and gcc did not
fully adjust yet, some packages compiled on a Ryzen could use some
optimization options that might cause illegal instruction trouble
on other cpus. That is a totally hypothetical and may be totally wrong
as well. Also it's not related at all to the thread that Ken posted and
the problems it describes.


In addition, the rather long thread I read indicated that the problem
was erratic.  Users would repeatedly compile mesa (or other packages) in
a loop and it would fail at random iterations.  That points to HW.

My best guess that the problem is some sort of complex timing problem
between the processor and it's support chips.

   -- Bruce




--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Anyone tried the AMD Ryzen 7 1700 processor on LFS?

[Thanos Baloukas]

On 06/08/2017 07:56 μμ, Ken Moffat wrote:

On Sun, Aug 06, 2017 at 06:38:45PM +0300, Thanos Baloukas wrote:

On 06/08/2017 05:57 μμ, Alan Feuerbacher wrote:

On 8/5/2017 10:58 PM, Thanos Baloukas wrote:


I'm going to buy a new system to speed up lfs builds. From what I read
the new AMD Ryzen 7 1700 would be a good choice for this job. Has anyone
tested this processor on lfs?


In early May I bought a new ASUS TUF motherboard, Intel 3770 processor,
and 32G of ram. LFS builds quickly on this new system. I would think
that the new AMD processor would be similar.


Thanks. My concern is not that much on speed, various benchmarks
I've seen show it's fast enough. What I have in mind is stability
and compatibility. Segmentation faults, software produced not running
on other machines and such.


For not running on other machines, my usual answer is to blame gmp.


Except from gmp, I thought that until gcc catch up with Ryzen,
such problems would arise.


Nowadays I always use the fsf config scripts in gmp:

cp -v  config{fsf,}.guess
cp -v config{fsf,}.sub

and then run configure.  I don't _expect_ to run binaries on a
different machine, but this seems a safer thing to do.

For Ryzen, I hope to get one - but I need to sort out various things
first (e.g. my KVM switches and monitors all use D-SUB, I don't see
me coming up with a quick solution to that).  But in the last week I
remember seeing a post on lkml about problems with Ryzen, and it
seems there are a lot of reports.

One thread starts at
https://community.amd.com/thread/215773?start=0&tstart=0
and it looks as if segfaults are common.

Other reports, e.g. at phoronix, note that some users are seeing the
problems and others aren't.

ĸen


Thanks for the link, it's very enlightening. It seems that I'll have to
delay the order until things clear up. A chasing game is happening
between users, AMD people, bios manufacturers, and kernel and gcc
developers, and I'd rather not get into that.

Opting in favor of Intel's safety is still an option, but 8 cores,
16 threads, at 65w TDP, for about 300 euro is very tempting
and I think it deserves some waiting.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Anyone tried the AMD Ryzen 7 1700 processor on LFS?

[Thanos Baloukas]

On 06/08/2017 05:57 μμ, Alan Feuerbacher wrote:

On 8/5/2017 10:58 PM, Thanos Baloukas wrote:


I'm going to buy a new system to speed up lfs builds. From what I read
the new AMD Ryzen 7 1700 would be a good choice for this job. Has anyone
tested this processor on lfs?


In early May I bought a new ASUS TUF motherboard, Intel 3770 processor,
and 32G of ram. LFS builds quickly on this new system. I would think
that the new AMD processor would be similar.


Thanks. My concern is not that much on speed, various benchmarks
I've seen show it's fast enough. What I have in mind is stability
and compatibility. Segmentation faults, software produced not running
on other machines and such.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

[lfs-support] Anyone tried the AMD Ryzen 7 1700 processor on LFS?

[Thanos Baloukas]

Hi

I'm going to buy a new system to speed up lfs builds. From what I read
the new AMD Ryzen 7 1700 would be a good choice for this job. Has anyone
tested this processor on lfs?

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] LFS 7.10, Chapter 6: Inetutils compile error

[Thanos Baloukas]

On 17/02/2017 12:45 μμ, Markus wrote:

Hey folks,

Thanks again for help on my first posting..

Meanwhile I've made it to chapter 6 and ironed out some problems I've caused
myself.

I had to start over once  with the backup I've made at the end of chapter 5,
but now everything seems correctly set up.

But...
When trying to build inetutils 1.9.4, I get this error message:

=
In file included from system.c:23:0:
system/linux.c: In function 'pnd_read':
system/linux.c:415:15: error: 'PATH_PROCNET_DEV' undeclared (first use in this
function)
fp = fopen (PATH_PROCNET_DEV, "r");
^~~~
system/linux.c:415:15: note: each undeclared identifier is reported only once
for each function it appears in
In file included from system.c:23:0:
system/linux.c: In function 'linux_if_nameindex':
system/linux.c:917:24: error: 'PATH_PROCNET_DEV' undeclared (first use in this
function)
content = read_file (PATH_PROCNET_DEV, &length);
 ^~~~
make[3]: *** [Makefile:1516: system.o] Error 1
make[3]: Leaving directory '/sources/inetutils-1.9.4/ifconfig'
make[2]: *** [Makefile:1536: all-recursive] Error 1
make[2]: Leaving directory '/sources/inetutils-1.9.4/ifconfig'
make[1]: *** [Makefile:1491: all-recursive] Error 1
make[1]: Leaving directory '/sources/inetutils-1.9.4'
make: *** [Makefile:1428: all] Error 2
root:/sources/inetutils-1.9.4#
=

First I thought I've made a mistake again, but so far I couldn't find anything
suspicious...

With a quick web search I found a patch for v1.9.1 ( https://git.congatec.com/
yocto/meta-openembedded/commit/3402bfac6b595c622e4590a8ff5eaaa854e2a2a3 ) but
I don't want to apply it (or add the missing definition by hand) before asking
the gurus ;-)

I hope I didn't mess something up (again) that may be the cause for this
problem :-)

Thanks, Markus



Is proc mounted on $LFS/proc?

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Error ld on building bzip2 Chapter 6.18 LFS v.7.10

[Thanos Baloukas]

On 30/01/2017 11:19 πμ, Sofyan Sugianto wrote:

Hello, today im trying to build LFS, but got error on chapter 6.18 LFS 7.10
is this error harmless for continue to the next step? please help me
what should i do if this harmful to continue

root:/sources/bzip2-1.0.6# make -f Makefile-libbz2_so
gcc -shared -Wl,-soname -Wl,libbz2.so.1.0 -o libbz2.so.1.0.6
blocksort.o huffman.o crctable.o randtable.o compress.o decompress.o
bzlib.o
/usr/bin/ld: blocksort.o: relocation R_X86_64_32 against
`.rodata.str1.1' can not be used when making a shared object;
recompile with -fPIC
/usr/bin/ld: compress.o: relocation R_X86_64_32 against
`.rodata.str1.8' can not be used when making a shared object;
recompile with -fPIC
/usr/bin/ld: decompress.o: relocation R_X86_64_32S against `.rodata'
can not be used when making a shared object; recompile with -fPIC
/usr/bin/ld: bzlib.o: relocation R_X86_64_32S against symbol
`BZ2_crc32Table' can not be used when making a shared object;
recompile with -fPIC
/usr/bin/ld: final link failed: Nonrepresentable section on output
collect2: error: ld returned 1 exit status
make: *** [Makefile-libbz2_so:38: all] Error 1


Is this the output you get after 'make -f Makefile-libbz2_so' ?
The following lines must be above the lines you posted:

gcc -fpic -fPIC -Wall -Winline -O2 -g -D_FILE_OFFSET_BITS=64 -c blocksort.c
gcc -fpic -fPIC -Wall -Winline -O2 -g -D_FILE_OFFSET_BITS=64 -c huffman.c
gcc -fpic -fPIC -Wall -Winline -O2 -g -D_FILE_OFFSET_BITS=64 -c crctable.c
gcc -fpic -fPIC -Wall -Winline -O2 -g -D_FILE_OFFSET_BITS=64 -c randtable.c
gcc -fpic -fPIC -Wall -Winline -O2 -g -D_FILE_OFFSET_BITS=64 -c compress.c
gcc -fpic -fPIC -Wall -Winline -O2 -g -D_FILE_OFFSET_BITS=64 -c decompress.c
gcc -fpic -fPIC -Wall -Winline -O2 -g -D_FILE_OFFSET_BITS=64 -c bzlib.c

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Build Texinfo-6.1 against Ncurses-6.0 ( wide-character)

[Thanos Baloukas]

On 06/01/2017 02:23 πμ, Roger Frost wrote:

Greetings,

I'm at LFS version 7.10, section 5.32. Texinfo-6.1

The problem I'm having is during the configuration:

configure: WARNING: Could not find a terminal library among tinfo
ncurses curses termlib termcap terminfo
configure: WARNING: The programs from `info' directory will not be built.


It's only a warning, I'd say to ignore it and proceed.
See the the build log on

http://www.linuxfromscratch.org/lfs/build-logs/7.10-rc1/logs/062-texinfo-6.1

The warning is there too.


Deviations from the book:
1) I'm building a multilib LFS 7.10 system based (loosely) on CLFS
Version 3.0.0-SYSVINIT-x86_64-Multilib.

Host system:
Second generation multilib (B)LFS 7.7, also based on CLFS.

Package Configuration:
PERL=/usr/bin/perl\
./configure   \
--prefix=/tools   \
--libdir=/tools/lib64 \
--build=${CLFS_HOST}  \
--host=${CLFS_TARGET}

Environment:
lfs:/mnt/clfs/sources/texinfo-6.1$ env
AS=x86_64-unknown-linux-gnu-as
AR=x86_64-unknown-linux-gnu-ar
CLFS_TARGET32=i686-pc-linux-gnu
CLFS_HOST=x86_64-cross-linux-gnu
TERM=xterm
CLFS_TARGET=x86_64-unknown-linux-gnu
OLDPWD=/mnt/clfs/sources
LC_ALL=POSIX
LFS=/mnt/clfs
BUILD64=-m64
CLFS=/mnt/clfs
BUILD32=-m32
PATH=/tools/bin:/cross-tools/bin:/bin:/usr/bin
LD=x86_64-unknown-linux-gnu-ld
PWD=/mnt/clfs/sources/texinfo-6.1
STRIP=x86_64-unknown-linux-gnu-strip
LFS_TGT=x86_64-unknown-linux-gnu
PS1=\u:\w\$
CXX=x86_64-unknown-linux-gnu-g++ -m64
SHLVL=1
HOME=/home/lfs
RANLIB=x86_64-unknown-linux-gnu-ranlib
CC=x86_64-unknown-linux-gnu-gcc -m64
_=/tools/bin/env

Other observations:
1) If I do 'ln -sv /tools/lib64/libncursesw.so.6.0
/tools/lib64/libncurses.so' the configure script won't complain, but the
build fails (probably because libncursesw is not binary compatible with
libncurses, and texinfo still requires the latter).
2) I don't believe I can install the regular and wide-character versions
side by side, can I? The header files are different, and one would
overwrite the other...

Bottom Line:
I presume it is possible to build Textinfo-6.1 against wide-character
Ncurses 6.0, since it's in the book and all. I feel like I'm missing
something very obvious here... Has anyone else experienced this issue?

I'll be happy to try anything, and provide any output that might be helpful.

Thank you for your time,
Roger



--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] how to test rw/ro/&c of a file [Was: "I don't need to know"]

[Thanos Baloukas]

On 13/11/2016 03:02 πμ, Paul Rogers wrote:

I don't understand what you are trying to do. If you want to check
if a filesystem is mounted ro or rw you may want to test this trivial
C program that utilizes the statvfs(3) glibc function. It needs
a mountpoint as argument and prints ro or rw respectively.

Build with
gcc check_ro_fs.c -o check_ro_fs

and use it on bash like that:

if test `./check_ro_fs ` == rw; then ; else echo "readonly filesystem"; fi



check_ro_fs.c
#include 
#include 

int main(int argc, char *argv[])
{
struct statvfs fs_stat;

if ( argc != 2 || statvfs(argv[1], &fs_stat) == -1 )
return 1;

if ( fs_stat.f_flag == 0 )
puts("no flags");
else
printf("%s\n", fs_stat.f_flag & ST_RDONLY != 0 ? "ro" : "rw");

return 0;
}


I installed it.  On the target system with linux-2.6.17 & glibc-2.3.4, I
get "no flags".  It works on the host system, linux-2.6.32 &
glibc-2.11.1, but I was only able to test there after booting was
complete.  I ran diff on both sys/statvfs.h and they were the same.
Other than that, well, I'm not a Linux/GNU developer.


I'm not a developer either. I was curious and learned few things.
The problem is rather your old glibc and kernel versions. There have
been changes over  the years in statvfs() implementation.
In statvfs(3) man page says:

In glibc versions before 2.13, statvfs() populated the bits of the
f_flag field by scanning the mount options shown in /proc/mounts.
However, starting with Linux 2.6.36, the underlying statfs(2) system
call provides the necessary information via the f_flags field, and
since glibc version 2.13, the statvfs() function will use information
from that field rather than scanning /proc/mounts.

Why were you able to test on host only after booting was complete?
What was the problem?

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] how to test rw/ro/&c of a file [Was: "I don't need to know"]

[Thanos Baloukas]

On 11/11/2016 08:51 μμ, Paul Rogers wrote:


Does 'mount -l [...]' give you accurate info?


No, it does not.  Inserted in mountkernfs even before proc is mounted:
/dev/hda9 / (rw)
Mount seems to think, "the root is always rw".  Anyway, not usable.

As my first message said, interestingly, at that point "cat
/proc/1/mounts" has two mounts obviously referring to the root, one rw,
one
ro.  Again, as I said there, I think that should remain "off limits".
There should be a better way.



Or, you might just try a write, to see if the fs is read-only or
read-write:

if touch /fsrwtestfile 2>/dev/null; then


That blasts a console message that it's a read-only filesystem,
exactly the message I'm trying to silence.  As I said before it
seems to be a direct console write from the kernel, not stderr.
That's why I tried dropping the log-level.  But it seems dangerous
to go too low.


I don't understand what you are trying to do. If you want to check
if a filesystem is mounted ro or rw you may want to test this trivial
C program that utilizes the statvfs(3) glibc function. It needs
a mountpoint as argument and prints ro or rw respectively.

Build with
gcc check_ro_fs.c -o check_ro_fs

and use it on bash like that:

if test `./check_ro_fs ` == rw; then it>; else echo "readonly filesystem"; fi


--
Thanos
#include 
#include 

int main(int argc, char *argv[])
{
struct statvfs fs_stat;

if ( argc != 2 || statvfs(argv[1], &fs_stat) == -1 )
return 1;

if ( fs_stat.f_flag == 0 )
puts("no flags");
else
printf("%s\n", fs_stat.f_flag & ST_RDONLY != 0 ? "ro" : "rw");

return 0;
}
-- 
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Compiling binutils, pass 2

[Thanos Baloukas]

On 03/11/2016 05:27 πμ, Jared lima wrote:

am i supposed to delete the already preset "build" directroy that was
created for doing the pass 1 compiling for binutils or do i create a new
one along side the already existent one using a similar name?



Follow the book. On chapter 5.3. General Compilation Instructions
section <> it says:
...
...
d. Change back to the sources directory.
e. Delete the extracted source directory unless instructed otherwise.

That means that you have to remove the $LFS/sources/binutils-2.27 dir.

When we create a build directory and run the commands from there,
there is the possibility files to be created outside that, so we remove
the whole source directory we extracted with the build directory
we created.

Don't reuse the same binutils source tree for pass 2 removing only
the build directory.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Probably simple binutils installation issue

[Thanos Baloukas]

On 25/09/2016 06:53 μμ, smith.aaron wrote:

Hi,
After much experimentation and reading around I have ended up back with
the same problem I had originally, so I am replying to your queries
regarding my original problem, which is that I can't compile binutils-2.2.7.
Thanks so much for your help. I hope I have emailed this to the right
people, I am reading up on mailing list etiquette too!
Thanks again,
Aaron

Thanos said:

g"The error in config.log should be after the line:

checking for C compiler default output file name

Copy and paste that part."

The 'checking for  compiler version' line occurs twice in my config.log.
First time I assume is fine:

configure:4377: checking for C compiler version
configure:4386: gcc --version >&5
gcc (Debian 5.3.1-13) 5.3.1 20160323

Second time the checking for ompiler version occurs is where I guess the
problem is:

configure:4417: checking for C compiler default output file name
configure:4439: gccconftest.c  >&5
gcc: error trying to exec 'cc1': execvp: No such file or directory
configure:4443: $? = 1
configure:4480: result:
configure: failed program was:
| /* confdefs.h */
| #define PACKAGE_NAME ""
| #define PACKAGE_TARNAME ""
| #define PACKAGE_VERSION ""
| #define PACKAGE_STRING ""
| #define PACKAGE_BUGREPORT ""
| #define PACKAGE_URL ""
| /* end confdefs.h.  */


Thanos/Bruce posted similar:
"Also, do this test:

$ cat > test.c << "EOF"
int
main ()
{

;
return 0;
}
EOF

$ gcc test.c

You should have an a.out executable.

$ ./a.out

$ echo $?

This should output 0 "

Here is what happened at this point:

lfs@kali:/mnt/lfs/sources/binutils-2.27/build$ cat > test.c << "EOF" > int > main () > { >  
> ; > return 0; > } > EOF lfs@kali:/mnt/lfs/sources/binutils-2.27/build$ gcc test.c gcc: error trying to exec 
'cc1': execvp: No such file or directory lfs@kali:/mnt/lfs/sources/binutils-2.27/build$ ./a.out bash: ./a.out: No such 
file or directory lfs@kali:/mnt/lfs/sources/binutils-2.27/build$ echo $? 127


Thanks again!

Hi everyone,

Have run into difficulty trying to install binutils at the start
of LFS chapter 5. I spent the last two days slowly re-reading up
to chapter five, googling around, and correcting any mistakes I
found, but to no avail. I found plenty online close to
describing my specific symptoms, but nothing that exactly
addressed it, or enabled me to figure the problem out.

First here is the terminal output where I encounter the problem:

!Y+lfs@kali:/mnt/sda7/sources/binutils-2.27/build$
../configure --prefix=/tools\
 >  --with-sysroot=$LFS\
 >  --with-lib-path=/tools/lib \
 >  --target=$LFS_TGT  \
 >  --disable-nls  \
 >  --disable-werror
checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
checking target system type... i686-lfs-linux-gnu
checking for a BSD-compatible install... /bin/install -c
checking whether ln works... yes
checking whether ln -s works... yes
checking for a sed that does not truncate output... /bin/sed
checking for gawk..Unless. gawk
checking for gcc... gcc
checking for C compiler default output file name...
configure: error: in `/mnt/sda7/sources/binutils-2.27/build':
configure: error: C compiler cannot create executables
See `config.log' for more details.
lfs@kali:/mnt/sda7/sources/binutils-2.27/build$




Here is the output of version-check.sh. Hopefully I am just
including this for reference/to show everything is ok in this
regard...

bash, version 4.3.46(1)-release
/bin/sh -> /bin/bash
Binutils: (GNU Binutils for Debian) 2.26.1
bison (GNU Bison) 3.0.4
/usr/bin/yacc -> /usr/bin/bison.yacc
bzip2,  Version 1.0.6, 6-Sept-2010.
Coreutils:  8.25
diff (GNU diffutils) 3.3
find (GNU findutils) 4.7.0-git
GNU Awk 4.1.3, API: 1.1 (GNU MPFR 3.1.4-p2, GNU MP 6.1.1)
/usr/bin/awk -> /usr/bin/gawk
gcc (Debian 6.1.1-11) 6.1.1 20160802
g++ (Debian 6.1.1-11) 6.1.1 20160802
(Debian GLIBC 2.23-5) 2.23
grep (GNU grep) 2.25
gzip 1.6
Linux version 4.6.0-kali1-686-pae (de...@kali.org) (gcc
version 5.4.0 20160609 (Debian 5.4.0-6) ) #1 SMP Debian
4.6.4-1kali1 (2016-07-21)
m4 (GNU M4) 1.4.17
GNU Make 4.1
GNU patch 2.7.5
Perl version='5.22.2';
sed (GNU sed) 4.2.2
tar (GNU tar) 1.29
texi2any (GNU texinfo) 6.1
xz (XZ Utils) 5.1.0alpha
g++ compilation OK

Re: [lfs-support] No Networking in Post Boot

[Thanos Baloukas]

On 14/09/2016 10:28 μμ, Tim Tassonis wrote:

On 09/14/2016 06:37 PM, Thanos Baloukas wrote:

On 14/09/2016 04:31 μμ, Thanos Baloukas wrote:

Did the network interfaces naming policy change? I thought eth* names
are not used anymore.


Me and my terrible english!
s/are not used/were not used/
I think. :)


Well, that's definitely a very minor mistake


I made a major mistake, sorry. eth* names are still used.

https://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames/

Thanks for pointing that.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] No Networking in Post Boot

[Thanos Baloukas]

On 14/09/2016 04:31 μμ, Thanos Baloukas wrote:

On 14/09/2016 10:52 πμ, Rob wrote:

LFS-7.10-Systemd.
In section 7.2.1.2 the book talks about creating
"/etc/systemd/network/10-eth0-dhcp.network"
. Then, in BLFS Chapter 14, - DHCPCD, it talks about disabling
systemd-networkd.service.
So, does the file referenced in LFS 7.2.1.2  still need to be
created if you install dhcpcd? Because I'm getting persistent
messages, when I try running
"systemctl start dhcpcd@eth0"


Did the network interfaces naming policy change? I thought eth* names
are not used anymore.


Me and my terrible english!
s/are not used/were not used/
I think. :)


It says that it fails because of a dependency, but it doesn't say
what the dependency is.
"systemctl status dhcpcd@eth0"
says:
"... / start failed with result 'depencency'.
Again, it doesn't tell you what the dependency is.
"systemctl edit dhcpcd@eth0"
shows an empty file.
I'm a little stuck here. Where do I go next?
Kernel configuration is correct and I'm pretty sure it's loading the
hardware, since LFS 7.8 was running on the same metal, and I
just copied over the 4.2 configuration and did
"make oldconfig"
to build 4.7.2.







--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] No Networking in Post Boot

[Thanos Baloukas]

On 14/09/2016 10:52 πμ, Rob wrote:

LFS-7.10-Systemd.
In section 7.2.1.2 the book talks about creating
"/etc/systemd/network/10-eth0-dhcp.network"
. Then, in BLFS Chapter 14, - DHCPCD, it talks about disabling
systemd-networkd.service.
So, does the file referenced in LFS 7.2.1.2  still need to be
created if you install dhcpcd? Because I'm getting persistent
messages, when I try running
"systemctl start dhcpcd@eth0"


Did the network interfaces naming policy change? I thought eth* names
are not used anymore.


It says that it fails because of a dependency, but it doesn't say
what the dependency is.
"systemctl status dhcpcd@eth0"
says:
"... / start failed with result 'depencency'.
Again, it doesn't tell you what the dependency is.
"systemctl edit dhcpcd@eth0"
shows an empty file.
I'm a little stuck here. Where do I go next?
Kernel configuration is correct and I'm pretty sure it's loading the
hardware, since LFS 7.8 was running on the same metal, and I
just copied over the 4.2 configuration and did
"make oldconfig"
to build 4.7.2.




--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] devtmpfs device node creation

[Thanos Baloukas]

On 12/09/2016 04:05 πμ, William Harrington wrote:

On Mon, 12 Sep 2016 00:43:18 +0300
Thanos Baloukas  wrote:


Could someone please point me to some source that could clear things up?


Yes I can,

DEVTMPFS

https://lwn.net/Articles/331818/


Thanks for that. It was very interesting and helpful.
It was in my google search results, but the "The return of devfs"
title misguided me.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] devtmpfs device node creation

[Thanos Baloukas]

On 12/09/2016 02:31 πμ, Bruce Dubbs wrote:

Thanos Baloukas wrote:

Following another thread on the list I searched a little about
when the nodes are created and found

https://lwn.net/Articles/330985/

where it states that after the rootfs is mounted by the kernel,
the populated tmpfs is mounted at /dev. The article is not new,
so at least this was the initial design.

Then reading the book on

http://www.linuxfromscratch.org/lfs/view/systemd/chapter07/udev.html

in section 7.3.2.2. Device Node Creation first paragraph, I think
implies that the nodes are created after the tmpfs is mounted on /dev

Could someone please point me to some source that could clear things up?


No it doesn't imply that.  I suppose we could s/be created with/have/ in
the 2nd sentence, but that's pretty picky.

   -- Bruce


Thanks for clarifying that. The change you mentioned or
s/initially be/have been initially/
(which I'm not sure it's correct in grammar and syntax)
would make the wording much more clear for me.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

[lfs-support] devtmpfs device node creation

[Thanos Baloukas]

Following another thread on the list I searched a little about
when the nodes are created and found

https://lwn.net/Articles/330985/

where it states that after the rootfs is mounted by the kernel,
the populated tmpfs is mounted at /dev. The article is not new,
so at least this was the initial design.

Then reading the book on

http://www.linuxfromscratch.org/lfs/view/systemd/chapter07/udev.html

in section 7.3.2.2. Device Node Creation first paragraph, I think
implies that the nodes are created after the tmpfs is mounted on /dev

Could someone please point me to some source that could clear things up?

Thanks

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] LFS and BLFS Version 7.10 are released

[Thanos Baloukas]

Congratulations to all the team members and many thanks for the hard work!

On 08/09/2016 01:19 πμ, Bruce Dubbs wrote:

The Linux From Scratch community is pleased to announce the release of
LFS Version 7.10, LFS Version 7.10 (systemd), BLFS Version 7.10, and
BLFS Version 7.10 (systemd).

This release is a major update to both LFS and BLFS.

The LFS release includes updates to glibc-2.24, binutils-2.27, and
gcc-6.2.0. In total, 29 packages were updated, fixes made to
bootscripts, and changes to text have been made throughout the book.

The BLFS version includes approximately 800 packages beyond the base
Linux From Scratch Version 7.9 book. This release has over 810 updates
from the previous version including numerous text and formatting changes.

Major changes to BLFS include the removal of Qt4 and KDE4.  The systemd
version of BLFS now include the full GNOME desktop.

One internal note is that through the efforts of several editors, the
book sources have combined the versions of the System V and systemd into
a single set of xml files but still provides for separate generation of
the books.  This allows the 80% of the two books that are common to be
maintained efficiently.

Thanks for this ambitious release go to many contributors.  Notably:

DJ Lucas
Pierre Labastie
Ken Moffat
Douglas Reno

You can read the books online[0]-[3], or download[4]-[7] to read locally.

Please direct any comments about this release to the LFS development
team at lfs-...@linuxfromscratch.org or blfs-...@linuxfromscratch.org.
Registration for the mailing lists is required to avoid junk email.

   -- Bruce Dubbs
  LFS

[0] http://www.linuxfromscratch.org/lfs/view/7.10/
[1] http://www.linuxfromscratch.org/blfs/view/7.10/
[2] http://www.linuxfromscratch.org/lfs/view/7.10-systemd/
[3] http://www.linuxfromscratch.org/blfs/view/7.10-systemd/

[4] http://www.linuxfromscratch.org/lfs/downloads/7.10/
[5] http://www.linuxfromscratch.org/blfs/downloads/7.10/
[6] http://www.linuxfromscratch.org/lfs/downloads/7.10-systemd/
[7] http://www.linuxfromscratch.org/blfs/downloads/7.10-systemd/



--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] When is /dev populated?

[Thanos Baloukas]

On 07/09/2016 09:16 μμ, Thanos Baloukas wrote:

On 07/09/2016 08:39 μμ, Bruce Dubbs wrote:

William Harrington wrote:

On Wed, September 7, 2016 03:14, Samuel Tyler wrote:

Sorry - I should have said - I am using systemd.




On Wed, Sep 07, 2016 at 11:26:55AM +1000, Samuel Tyler wrote:

When is /dev populated? When I reboot, it says /dev/sda1 is

unavailable.

/dev only contains console and null.

LFS 7.9



Review the README in the systemd source or in
/usr/share/doc/systemd-x.y.z/README and make sure your kernel config is
set according to the required/recommended kernel options. Systemd
populates /dev when it mounts devtmpfs or if the kernel automounts it.


Actually, I think that systemd/udev generally does not populate /dev.
The kernel should do that via CONFIG_DEVTMPFS.  The only thing udev
(part of systemd) does is change permissions, usr/group ownership, and
create symlinks.

   -- Bruce


I was puzzled about that. According to

http://cateee.net/lkddb/web-lkddb/DEVTMPFS.html

you are right.


Do I understand it right?

The kernel creates a tmpfs and populates it with device nodes.
That tmpfs is mounted on /dev either by the kernel it's self
if CONFIG_DEVTMPFS_MOUNT is set, or by systemd if it's not set.

If that is correct, the answer to the question "When is /dev populated?"
is "when that tmpfs is mounted on it (on /dev). And since this mount
can be done by systemd, systemd can populate /dev.
So William is right too.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] When is /dev populated?

[Thanos Baloukas]

On 07/09/2016 08:39 μμ, Bruce Dubbs wrote:

William Harrington wrote:

On Wed, September 7, 2016 03:14, Samuel Tyler wrote:

Sorry - I should have said - I am using systemd.




On Wed, Sep 07, 2016 at 11:26:55AM +1000, Samuel Tyler wrote:

When is /dev populated? When I reboot, it says /dev/sda1 is

unavailable.

/dev only contains console and null.

LFS 7.9



Review the README in the systemd source or in
/usr/share/doc/systemd-x.y.z/README and make sure your kernel config is
set according to the required/recommended kernel options. Systemd
populates /dev when it mounts devtmpfs or if the kernel automounts it.


Actually, I think that systemd/udev generally does not populate /dev.
The kernel should do that via CONFIG_DEVTMPFS.  The only thing udev
(part of systemd) does is change permissions, usr/group ownership, and
create symlinks.

   -- Bruce


I was puzzled about that. According to

http://cateee.net/lkddb/web-lkddb/DEVTMPFS.html

you are right.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Sanity Check Failed in GCC Pass 2

[Thanos Baloukas]

On 04/09/2016 06:55 μμ, Rob wrote:

In conducting the sanity check in section 5.10:
echo 'int main(){}' > dummy.c
cc dummy.c
readelf -l a.out | grep ': /tools'
cc dummy.c
got the following output
/usr/bin/ld: 
/mnt/lfs/tools/bin/../lib/gcc/i686-pc-linux-gnu/6.2.0/../../../crti.o: 
unrecognized relocation (0x2b) in section `.init'
/usr/bin/ld: final link failed: Bad value
collect2: error: ld returned 1 exit status
I am following LFS 7.10 RC1. I do not understand enough of the inner workings 
of linking to debug this issue. The pages on google are seemingly directed at 
experienced programmers--something which I am not.
What went wrong here?


Check lfs user's bash history to be sure that you run
the book's commands correctly. Check if you used the correct
tarball versions also. You can do this with command "history|less" on
a terminal as user lfs, or open /home/lfs/.bash_history with an editor.

If you don't find something, post some info for the host
and the output of version-check.sh

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Force i386

[Thanos Baloukas]

On 04/09/2016 11:12 πμ, Rob wrote:

Thanos Baloukas  wrote:
http://lists.linuxfromscratch.org/pipermail/lfs-support/2016-September/050333.html

In that thread, you linked to a script collection, on github. The instructions 
in there did not work (not surprising), and asking google just linked to a 
bunch of pages about installing multiarch packages using various distro package 
managers. Not useful here.
So, I gave up on trying to build a multilib toolchain.



Maybe you want to try CLFS.

http://trac.clfs.org/
http://clfs.org/view/

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Force i386

[Thanos Baloukas]

On 04/09/2016 08:14 πμ, Rob wrote:

I can't seem to get multilib GCC to compile properly. It stops haflway through 
compiling with
"checking dynamic linker characteristics... configure: error: Link tests are not 
allowed after GCC_NO_EXECUTABLES."
and won't go any further. Sounds like something went wrong with binutils, but 
I'm not sure how to fix.
So, is there any way to force i686 on a x86_64 system? Perhaps change $LFS_TGT 
to i386-linux-gnu?


You don't mention what changes to the instructions led to that error.
Did you read the replies to

http://lists.linuxfromscratch.org/pipermail/lfs-support/2016-September/050333.html

Multilib is not simple. It needs changes to the whole toolchain -
binutils, gcc, glibc. The changes may be configure options,
variables, patches etc.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Multilib support

[Thanos Baloukas]

On 03/09/2016 09:35 πμ, Hazel Russman wrote:

On Sat, 3 Sep 2016 02:21:44 +0300
Thanos Baloukas  wrote:


On 02/09/2016 10:06 μμ, Rob wrote:

In chapter 6 of LFS, we have
"--disable-multilib"
specified in the configure instructions. What will happen if I don't include 
that? I have some programs that I want to run which are 32 bit, and require 32 
bit libs.


AFAIK omitting this is needed to build a multilib compiler,
but it's not enough. It needs more to achieve this. Try Google.

--
Thanos
--

I don't think you can do multilib in LFS. You need to use CLFS.

Hazel


This is what I meant saying "it needs more to achieve this".
Not sure about CLFS. Sometime ago Armin's instructions on
https://github.com/elkrejzi
worked for me. I don't know if they still do.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Multilib support

[Thanos Baloukas]

On 02/09/2016 10:06 μμ, Rob wrote:

In chapter 6 of LFS, we have
"--disable-multilib"
specified in the configure instructions. What will happen if I don't include 
that? I have some programs that I want to run which are 32 bit, and require 32 
bit libs.


AFAIK omitting this is needed to build a multilib compiler,
but it's not enough. It needs more to achieve this. Try Google.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Probably simple binutils installation issue

[Thanos Baloukas]

On 02/09/2016 09:26 μμ, smith.aaron wrote:

Hi everyone,

Have run into difficulty trying to install binutils at the start of LFS
chapter 5. I spent the last two days slowly re-reading up to chapter
five, googling around, and correcting any mistakes I found, but to no
avail. I found plenty online close to describing my specific symptoms,
but nothing that exactly addressed it, or enabled me to figure the
problem out.

First here is the terminal output where I encounter the problem:

!Y+lfs@kali:/mnt/sda7/sources/binutils-2.27/build$ ../configure
--prefix=/tools\
 >  --with-sysroot=$LFS\
 >  --with-lib-path=/tools/lib \
 >  --target=$LFS_TGT  \
 >  --disable-nls  \
 >  --disable-werror
checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
checking target system type... i686-lfs-linux-gnu
checking for a BSD-compatible install... /bin/install -c
checking whether ln works... yes
checking whether ln -s works... yes
checking for a sed that does not truncate output... /bin/sed
checking for gawk... gawk
checking for gcc... gcc
checking for C compiler default output file name...
configure: error: in `/mnt/sda7/sources/binutils-2.27/build':
configure: error: C compiler cannot create executables
See `config.log' for more details.
lfs@kali:/mnt/sda7/sources/binutils-2.27/build$




Here is the output of version-check.sh. Hopefully I am just including
this for reference/to show everything is ok in this regard...

bash, version 4.3.46(1)-release
/bin/sh -> /bin/bash
Binutils: (GNU Binutils for Debian) 2.26.1
bison (GNU Bison) 3.0.4
/usr/bin/yacc -> /usr/bin/bison.yacc
bzip2,  Version 1.0.6, 6-Sept-2010.
Coreutils:  8.25
diff (GNU diffutils) 3.3
find (GNU findutils) 4.7.0-git
GNU Awk 4.1.3, API: 1.1 (GNU MPFR 3.1.4-p2, GNU MP 6.1.1)
/usr/bin/awk -> /usr/bin/gawk
gcc (Debian 6.1.1-11) 6.1.1 20160802
g++ (Debian 6.1.1-11) 6.1.1 20160802
(Debian GLIBC 2.23-5) 2.23
grep (GNU grep) 2.25
gzip 1.6
Linux version 4.6.0-kali1-686-pae (de...@kali.org) (gcc version
5.4.0 20160609 (Debian 5.4.0-6) ) #1 SMP Debian 4.6.4-1kali1
(2016-07-21)
m4 (GNU M4) 1.4.17
GNU Make 4.1
GNU patch 2.7.5
Perl version='5.22.2';
sed (GNU sed) 4.2.2
tar (GNU tar) 1.29
texi2any (GNU texinfo) 6.1
xz (XZ Utils) 5.1.0alpha
g++ compilation OK


Finally, here is the config.log mentioned in the terminal. I think I cut
and paste the right bit of it. Overall it was very big and I am not sure
of mailing list etiquette, so tried to just get the relevant bit. I am
confident I have entered any other commands correctly (except that of
course I am having this problem, so...!)


lfs@kali:/mnt/sda7/sources/binutils-2.27/build$ ../configure
--prefix=/tools\
 >  --with-sysroot=$LFS\
 >  --with-lib-path=/tools/lib \
 >  --target=$LFS_TGT  \
 >  --disable-nls  \
 >  --disable-werror
checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
checking target system type... i686-lfs-linux-gnu
checking for a BSD-compatible install... /bin/install -c
checking whether ln works... yes
checking whether ln -s works... yes
checking for a sed that does not truncate output... /bin/sed
checking for gawk... gawk
checking for gcc... gcc
checking for C compiler defa


Any help will be greatly appreciated! Thanks for the book and community
- although I have run into trouble here I think it is going to be of
great help for me.



You posted the configure output, not config.log
config.log is created by the configure script,
not redirecting it's output.

The error in config.log should be after the line:

checking for C compiler default output file name

Copy and paste that part.

Also, do this test:

$ cat > test.c << "EOF"
int
main ()
{

;
return 0;
}
EOF

$ gcc test.c

You should have an a.out executable.

$ ./a.out

$ echo $?

This should output 0

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] dmesg does not create .log

[Thanos Baloukas]

On 07/07/2016 05:02 μμ, Marcos Pansani wrote:

OK folks

After compiling the LFS 7.9-systemd, and after the first boot, I noticed
that is not created the kernel log file in the var / log folder.

The dmesg has been properly installed by util-linux package, when I go
with the dmesg command in the console, it shows on the screen the log
boot, but the file would have to be created in the folder var / log does
not exist 

Does anyone know how to solve?, The file is very useful to look for
problems in the sitema boot.

Thank you

Marcos



The explanation for that is in /var/log/README

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Error kernel panic LFS 7.9-systemd

[Thanos Baloukas]

On 07/07/2016 06:00 πμ, Marcos Pansani wrote:

All right with you...

I continue with kernel panic issue after compiling LFS 7.9-systemd.

I started again from scratch and redid the book again, but I did not
succeed ... :(

I continue with the errors in the boot:

Starting Init: / sbin / init exists but could not execute it (error -8)
Starting Init: / bin / sh exists but could not execute it (error -8)
kernel panic -not syncing: not working init found. Try passing init =
option to kernel

I went with the following commands in the chroot:

root: / # ldd / bin / sh
 linux-vdso.so.1 (0x7ffc126cd000)
 libreadline.so.6 => /lib/libreadline.so.6 (0x7f7d8ff64000)
 libhistory.so.6 => /lib/libhistory.so.6 (0x7f7d8fd5c000)
 libncursesw.so.6 => /lib/libncursesw.so.6 (0x7f7d8faf2000)
 libdl.so.2 => /lib/libdl.so.2 (0x7f7d8f8ee000)
 libc.so.6 => /lib/libc.so.6 (0x7f7d8f54d000)
 /lib64/ld-linux-x86-64.so.2 (0x7f7d901a9000)
root: / # ldd / sbin / init
 linux-vdso.so.1 (0x7ffd4a567000)
 libcap.so.2 => /lib/libcap.so.2 (0x7f9951232000)
 librt.so.1 => /lib/librt.so.1 (0x7f995102a000)
 libattr.so.1 => /lib/libattr.so.1 (0x7f9950e26000)
 libkmod.so.2 => /lib/libkmod.so.2 (0x7f9950c11000)
 libmount.so.1 => /lib64/libmount.so.1 (0x7f99509cb000)
 libpthread.so.0 => /lib/libpthread.so.0 (0x7f99507ae000)
 libc.so.6 => /lib/libc.so.6 (0x7f995040d000)
 /lib64/ld-linux-x86-64.so.2 (0x7f9951436000)
 liblzma.so.5 => /lib/liblzma.so.5 (0x7f99501e8000)
 libz.so.1 => /lib/libz.so.1 (0x7f994ffcf000)
 libblkid.so.1 => /lib64/libblkid.so.1 (0x7f994fd8f000)
 libuuid.so.1 => /lib64/libuuid.so.1 (0x7f994fb8b000)

That appears to be all satisfied ...

I tried running the init by chroot, the following message appears:

root: / # init
Could not find an alternative telinit implementation to spawn.

Does this help in something to try to unravel this mystery?

Is it by chance, some kernel configuration could interfere with this error?

Thank you

Marcos



There are some weird spaces in the error messages and the commands:

Starting Init: / sbin / init exists but could not execute it (error -8)
root: / # ldd / bin / sh

Where do they come from?

Pierre suggested to run the file command but you must have missed that.
file /lib/systemd/systemd
file /bin/bash

Error 8 is exec format error
What is the host? Is it x86-64?
The relevant kernel CONFIGs must be

CONFIG_64BIT=y
CONFIG_X86_64=y
CONFIG_X86=y
CONFIG_OUTPUT_FORMAT="elf64-x86-64"

Are the settings root= in /boot/grub/grub.cfg
and / (root partition) in /etc/fstab consistent?
That's all I can think of.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Error kernel panic LFS 7.9-systemd

[Thanos Baloukas]

On 05/07/2016 09:14 μμ, Marcos Pansani wrote:

Alright guys, I'm in need of your help  Sorry for my bad English.

I'm compiling the Linux From Scratch - Version 7.9-systemd, but I found
a problem in the way.

At the Chapter 7. Basic System Configuration, when I give the following
commands:

timedatectl set-local-rtc 1
timedatectl set-time -MM-DD HH:MM:SS
timedatectl set-timezone TIMEZONE
timedatectl list-timezones

I returned the following error for all of them: "failed to create bus
connection: No such file or directory"

Move forward in the construction book, and after completing, when I boot
the system, gave kernel panic.

The following error on the boot screen, did not yield to copy the log
file, because it was not created:

- Starting Init: /sbin/init exists but couldn't execute it (error -8)
- kernel panic -not syncing: not working init found. Try passing init=
option to kernel


I appreciate the help

Thank you



Post the output of the following commands in chroot

ls -l /sbin/init
ls -l `readlink -f /sbin/init`
ldd /sbin/init

and check if all the libraries listed by ldd exist on your LFS system.
If you are on x86_64 check if the /lib64 -> lib and /usr/lib64 -> lib
symlinks are in place.
Did you deviate from the book somehow?

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Missing Check for Install of kbd-2.0.3??

[Thanos Baloukas]

On 27/06/2016 08:50 μμ, Shomper, Keith A wrote:

Working out of the LFS 7.9 Book. Got to section 6.58 Kbd-2.0.3 w/o any
problems. When running configure command,

/PKG_CONFIG_PATH=/tools/lib/pkgconfig ./configure --prefix=/usr
--disable-vlock/

received the following error:

/checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for a thread-safe mkdir -p... /bin/mkdir -p
checking for gawk... gawk checking whether make sets $(MAKE)... yes
checking for pkg-config..08:26shomper... /usr/bin/pkg-config
checking pkg-config is at least version 0.9.0... yes
checking for CHECK... no
configure: error: Package requirements (check >= 0.9.4) were not met:
No package 'check' found
Consider adjusting the PKG_CONFIG_PATH environment variable if you
installed software in a non-standard prefix.
Alternatively, you may set the environment variab08:27shomperles
CHECK_CFLAGS
and CHECK_LIBS to avoid the need to call pkg-config.
See the pkg-config man page for more details./

There was an online post

referencing this same problem (Sep 2013), but the answer was adjust
PKG_CONFIG_PATH to what is already referenced in LFS book

I see that Check0.10 is part of the Chapter 5 sources, but not called
for in Chapter 6. Is this an omission? Should I install check-0.10.0 and
re-try install kbd-2.0.3?


Hi Keith

If you haven't done already, check if check-0.10.0 installation in 
/tools is ok.

In chroot do

ls -l /tools/lib/pkgconfig/check.pc && cat /tools/lib/pkgconfig/check.pc
ls -l /tools/lib/libcheck*

The results should be the same outside chroot too,
if the /tools -> $LFS/tools symlink on the host is ok.

--
Thanos

--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] shutting down systemd

[Thanos Baloukas]

On 30/05/2016 08:00 μμ, Thef keo wrote:

I agree with your answer :
"Maybe the problem is that you mixed instructions by the sysvinit and
systemd books."

I started with the SystemV LFS Book but it was a mistake. Then I
followed the SystemD LFS Book and I started again the whole Chapter 6.

There was little differences between systemv and systemd. With the
systemV, you install 3 additionnals packages : Sysklogd-1.5.1,
Sysvinit-2.88dsf, Eudev-3.1.5

And for systemd you need 2 more : Systemd-229, D-Bus-1.10.6

Like I said, I started again the whole Chapter 6 and so, I should not
have errors, but I have errors.
However, I removed all installed packages from the first installation.

And I never stoped dbus because it never worked. I tried to reinstall
dbus and systemd but without success.

What should I have to do now ?


You keep top posting. You should write your reply after the mail
you're replying to. I think systemd runs the scripts in /etc/init.d/
trying to be compatible with sysvinit. So maybe what you have in there
conflicts with some systemd units. Boot your host,
rm -rf /mnt/lfs/etc/init.d/*
and then try to boot LFS. Maybe others have something better to suggest.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] shutting down systemd

[Thanos Baloukas]

On 30/05/2016 01:48 μμ, Thanos Baloukas wrote:

On 30/05/2016 12:40 πμ, Thef keo wrote:


 > On 27/05/2016 10:36 μμ, Thef keo wrote:
 >  > Hello,
 >  >
 >  > I followed LFS 7.9 Book.
 >  > I finished installing the system and I tried to solve à
problem :
 >  > During the last command shutdown -r now, I get the
following result:
 >  > Failed to talk to init daemon .
 >  >
 >  > I tried several commands to correct this problem like
systemctl
 > poweroff
 >  > but it returns : Failed to get D-Bus connection: Unknown
error -1.
 >  >
 >  > I restarted the installation of systemd but this has not
solved the
 >  > problem.
 >  >
 >  > Do I have a problem with a package or is it something
else ?
 >  >
 > I would also check if something is wrong with the dbus
installation.
 > Were the commands ran as root?
 > What's the output of
 > echo $DBUS_SESSION_BUS_ADDRESS
 > ?
 > Date: Fri, 27 May 2016 15:43:13 -0500
 > From: Bruce Dubbs 
 > I cannot help, but please make sure the message title is
descriptive
 > when posting.
 >
 >
 > Hi,
 >
 > Yes, I ran the commands as root, and I even tried reinstalling
dbus but
 > it doesn't solve the problem.
 >
 > When I tried echo $DBUS_SESSION_BUS_ADDRESS, this is no message
(the
 > variable has an empty value)
 >
 > So what I need to do is to declare this variable ?
 >
 >
 > Resenting because I replied to the wrong message and confused
the OP.
 >
 > Receiving the list's mails batched makes it difficult to reply.
 > Maybe you should go to your mailing list account's preferences

 > and disable that. On LFS mailing lists bottom posting is
preferred,
 > so please use this style.

 > DBUS_SESSION_BUS_ADDRESS should be set automatically.
 > When I login as root to a linux console, it is set to
 > unix:path=/run/user/0/bus

 > As a next step I suggest to run `ps -ef|grep dbus`
 > and then `journalctl -u dbus` and `journalctl --user|grep -i bus`
 > hopefully will give a clue on what's wrong.

>--
>Thanos


Thanks again for your answer and I'm sorry for missing Bruce's advice
but I think that I fixed the object problem of the mail. I also turned
off some preferences in my mailing list account.



But you didn't choose No to the question:
Would you like to receive list mail batched in a daily digest?
If you do that you will receive the list mails individually
and it will be a lot easier to reply. If you are receiving them batched
you have to edit the subject and trim to reply. See how I trimmed yours
in this reply.


I did one command before all your steps, it's ' /etc/init.d/dbus status '
and I get the following message :
[FAIL] dbus is not running ... failed !


What is /etc/init.d/dbus? Is it sysvinit bootscript?
I thought you were using systemd. The commands to control the
dbus service on a system running systemd are

systemctl stop dbus
systemctl start dbus
systemctl status dbus

and some more you can find in systemctl man page.


but you should only use `systemctl status dbus` and not stop because
systemd depends heavily on dbus.


Maybe the problem is that you mixed instructions
by the sysvinit and systemd books.


Then I tried to active dbus service but I always get my error "Failed to
get D-Bus connection: Unknown error -1"





After that, I tried your steps.
For the last command ' journalctl --user|grep -i bus ' I got " No
journal files were founded. "

Please, you can see my other results in the attachment.






--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] shutting down systemd

[Thanos Baloukas]

On 30/05/2016 12:40 πμ, Thef keo wrote:


 > On 27/05/2016 10:36 μμ, Thef keo wrote:
 >  > Hello,
 >  >
 >  > I followed LFS 7.9 Book.
 >  > I finished installing the system and I tried to solve à
problem :
 >  > During the last command shutdown -r now, I get the
following result:
 >  > Failed to talk to init daemon .
 >  >
 >  > I tried several commands to correct this problem like
systemctl
 > poweroff
 >  > but it returns : Failed to get D-Bus connection: Unknown
error -1.
 >  >
 >  > I restarted the installation of systemd but this has not
solved the
 >  > problem.
 >  >
 >  > Do I have a problem with a package or is it something else ?
 >  >
 > I would also check if something is wrong with the dbus
installation.
 > Were the commands ran as root?
 > What's the output of
 > echo $DBUS_SESSION_BUS_ADDRESS
 > ?
 > Date: Fri, 27 May 2016 15:43:13 -0500
 > From: Bruce Dubbs 
 > I cannot help, but please make sure the message title is
descriptive
 > when posting.
 >
 >
 > Hi,
 >
 > Yes, I ran the commands as root, and I even tried reinstalling
dbus but
 > it doesn't solve the problem.
 >
 > When I tried echo $DBUS_SESSION_BUS_ADDRESS, this is no message (the
 > variable has an empty value)
 >
 > So what I need to do is to declare this variable ?
 >
 >
 > Resenting because I replied to the wrong message and confused the OP.
 >
 > Receiving the list's mails batched makes it difficult to reply.
 > Maybe you should go to your mailing list account's preferences

 > and disable that. On LFS mailing lists bottom posting is preferred,
 > so please use this style.

 > DBUS_SESSION_BUS_ADDRESS should be set automatically.
 > When I login as root to a linux console, it is set to
 > unix:path=/run/user/0/bus

 > As a next step I suggest to run `ps -ef|grep dbus`
 > and then `journalctl -u dbus` and `journalctl --user|grep -i bus`
 > hopefully will give a clue on what's wrong.

>--
>Thanos


Thanks again for your answer and I'm sorry for missing Bruce's advice
but I think that I fixed the object problem of the mail. I also turned
off some preferences in my mailing list account.



But you didn't choose No to the question:
Would you like to receive list mail batched in a daily digest?
If you do that you will receive the list mails individually
and it will be a lot easier to reply. If you are receiving them batched
you have to edit the subject and trim to reply. See how I trimmed yours
in this reply.


I did one command before all your steps, it's ' /etc/init.d/dbus status '
and I get the following message :
[FAIL] dbus is not running ... failed !


What is /etc/init.d/dbus? Is it sysvinit bootscript?
I thought you were using systemd. The commands to control the
dbus service on a system running systemd are

systemctl stop dbus
systemctl start dbus
systemctl status dbus

and some more you can find in systemctl man page.

Maybe the problem is that you mixed instructions
by the sysvinit and systemd books.


Then I tried to active dbus service but I always get my error "Failed to
get D-Bus connection: Unknown error -1"





After that, I tried your steps.
For the last command ' journalctl --user|grep -i bus ' I got " No
journal files were founded. "

Please, you can see my other results in the attachment.



--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] shutting down systemd (was lfs-support Digest, Vol 500, Issue 1)

[Thanos Baloukas]

On 29/05/2016 01:01 μμ, Thef keo wrote:


Message: 1
Date: Fri, 27 May 2016 21:36:53 +0200
From: Thef keo mailto:flofl...@gmail.com>>
To: lfs-support@lists.linuxfromscratch.org
<mailto:lfs-support@lists.linuxfromscratch.org>
Subject: [lfs-support] lfs-support Digest
Message-ID:

Hello,

I followed LFS 7.9 Book.
I finished installing the system and I tried to solve à problem :
During the last command shutdown -r now, I get the following result:
Failed to talk to init daemon .

I tried several commands to correct this problem like systemctl poweroff
but it returns : Failed to get D-Bus connection: Unknown error -1.

I restarted the installation of systemd but this has not solved the
problem.

Do I have a problem with a package or is it something else ?

Regards,

Florent
--

Message: 2
Date: Fri, 27 May 2016 23:34:29 +0300
From: Thanos Baloukas mailto:baloukast...@sch.gr>>
To: lfs-support@lists.linuxfromscratch.org
<mailto:lfs-support@lists.linuxfromscratch.org>
Subject: Re: [lfs-support] lfs-support Digest
Message-ID: <5748af55.8040...@sch.gr <mailto:5748af55.8040...@sch.gr>>
Content-Type: text/plain; charset=UTF-8; format=flowed

On 27/05/2016 10:36 μμ, Thef keo wrote:
 > Hello,
 >
 > I followed LFS 7.9 Book.
 > I finished installing the system and I tried to solve à problem :
 > During the last command shutdown -r now, I get the following result:
 > Failed to talk to init daemon .
 >
 > I tried several commands to correct this problem like systemctl
poweroff
 > but it returns : Failed to get D-Bus connection: Unknown error -1.
 >
 > I restarted the installation of systemd but this has not solved the
 > problem.
 >
 > Do I have a problem with a package or is it something else ?
 >
I would also check if something is wrong with the dbus installation.
Were the commands ran as root?
What's the output of
echo $DBUS_SESSION_BUS_ADDRESS
?
Date: Fri, 27 May 2016 15:43:13 -0500
From: Bruce Dubbs mailto:bruce.du...@gmail.com>>
To: LFS Support List mailto:lfs-support@lists.linuxfromscratch.org>>
Subject: Re: [lfs-support] shutting down systemd
Message-ID: <5748b161.70...@gmail.com <mailto:5748b161.70...@gmail.com>>
Content-Type: text/plain; charset=UTF-8; format=flowed

I cannot help, but please make sure the message title is descriptive
when posting.


Hi,

Yes, I ran the commands as root, and I even tried reinstalling dbus but
it doesn't solve the problem.

When I tried echo $DBUS_SESSION_BUS_ADDRESS, this is no message (the
variable has an empty value)

So what I need to do is to declare this variable ?



Resenting because I replied to the wrong message and confused the OP.

Receiving the list's mails batched makes it difficult to reply.
Maybe you should go to your mailing list account's preferences
and disable that. On LFS mailing lists bottom posting is preferred,
so please use this style.

DBUS_SESSION_BUS_ADDRESS should be set automatically.
When I login as root to a linux console, it is set to
unix:path=/run/user/0/bus

As a next step I suggest to run `ps -ef|grep dbus`
and then `journalctl -u dbus` and `journalctl --user|grep -i bus`
hopefully will give a clue on what's wrong.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] shutting down systemd (was lfs-support Digest, Vol 500, Issue 1)

[Thanos Baloukas]

On 27/05/2016 11:43 μμ, Bruce Dubbs wrote:

Thanos Baloukas wrote:

On 27/05/2016 10:36 μμ, Thef keo wrote:

Hello,

I followed LFS 7.9 Book.
I finished installing the system and I tried to solve à problem :
During the last command shutdown -r now, I get the following result:
Failed to talk to init daemon .

I tried several commands to correct this problem like systemctl poweroff
but it returns : Failed to get D-Bus connection: Unknown error -1.

I restarted the installation of systemd but this has not solved the
problem.

Do I have a problem with a package or is it something else ?


I would also check if something is wrong with the dbus installation.
Were the commands ran as root?
What's the output of
echo $DBUS_SESSION_BUS_ADDRESS


I cannot help, but please make sure the message title is descriptive
when posting.

Yes, I ran the commands as root, and I even tried reinstalling dbus but
it doesn't solve the problem.

When I tried echo $DBUS_SESSION_BUS_ADDRESS, this is no message (the
variable has an empty value)

So what I need to do is to declare this variable ?


Hi Florent

You seem to missed that Bruce correctly changed the thread's subject.
His message is included, see above. Receiving the list's mails batched
makes it difficult to answer. Maybe you should go to your mailing list
account's preferences and disable that. On LFS mailing lists
bottom posting is preferred, so please use this style.

Now to your issue. DBUS_SESSION_BUS_ADDRESS should be set
automatically. When I login as root to a linux console, it is set to
unix:path=/run/user/0/bus

As a next step I suggest to run `ps -ef|grep dbus`
and then `journalctl -u dbus` and `journalctl --no-pager|grep dbus`
hopefully will give a clue on what's wrong.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] lfs-support Digest

[Thanos Baloukas]

On 27/05/2016 10:36 μμ, Thef keo wrote:

Hello,

I followed LFS 7.9 Book.
I finished installing the system and I tried to solve à problem :
During the last command shutdown -r now, I get the following result:
Failed to talk to init daemon .

I tried several commands to correct this problem like systemctl poweroff
but it returns : Failed to get D-Bus connection: Unknown error -1.

I restarted the installation of systemd but this has not solved the
problem.

Do I have a problem with a package or is it something else ?


I would also check if something is wrong with the dbus installation.
Were the commands ran as root?
What's the output of
echo $DBUS_SESSION_BUS_ADDRESS
?

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] (LFS 7.9) 5.10. GCC-5.3.0 - Pass 2 error loading plugin: /lib64/libc.so.6: version `GLIBC_2.14' not found

[Thanos Baloukas]

On 03/05/2016 04:47 μμ, Madhusudhan Acharya wrote:

Hi

I am trying to build a new system using CentOS 6.7 as the
base by referring LFS-BOOK-7.9 for the build purpose. In
the stage "5.10. GCC-5.3.0 - Pass 2", i am seeing below
error while running make command

/lfs/tools/bin/../lib/gcc/x86_64-lfs-linux-gnu/5.3.0/../../../../x86_64-lfs-linux-gnu/bin/ld:
 /lfs/sources/gcc-5.3.0/build/./gcc/liblto_plugin.so: error loading plugin: 
/lib64/libc.so.6: version `GLIBC_2.14' not found (required by 
/lfs/sources/gcc-5.3.0/build/./gcc/liblto_plugin.so)

All required packages seems to be present in the system

[root@maddy-linux-dev ~]# bash version-check.sh
bash, version 4.1.2(1)-release
/bin/sh -> /bin/bash
Binutils: version 2.20.51.0.2-5.43.el6 20100205
bison (GNU Bison) 2.4.1
yacc is /usr/bin/yacc - 1.9 20070509
bzip2,  Version 1.0.5, 10-Dec-2007.
Coreutils:  8.4
diff (GNU diffutils) 2.8.1
find (GNU findutils) 4.4.2
GNU Awk 3.1.7
/usr/bin/awk -> /bin/gawk
gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-16)
g++ (GCC) 4.4.7 20120313 (Red Hat 4.4.7-16)
(GNU libc) 2.12
grep (GNU grep) 2.20
gzip 1.3.12
Linux version 2.6.32-573.el6.x86_64 (mockbu...@c6b9.bsys.dev.centos.org) (gcc 
version 4.4.7 20120313 (Red Hat 4.4.7-16) (GCC) ) #1 SMP Thu Jul 23 15:44:03 
UTC 2015
m4 (GNU M4) 1.4.13
GNU Make 3.81
patch 2.6
Perl version='5.10.1';
GNU sed version 4.2.1
tar (GNU tar) 1.23
version-check.sh: line 42: makeinfo: command not found
xz (XZ Utils) 4.999.9beta
g++ compilation OK


Am i missing anything here? Can someone suggest me how i can solve this issue?

thanks in advance
Maddy


Your host seems to not meet the requirements in:

gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-16)
g++ (GCC) 4.4.7 20120313 (Red Hat 4.4.7-16)
version-check.sh: line 42: makeinfo: command not found

The book says that the minimum requirement is GCC-4.7
You don't have Texinfo also.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Alsa Drivers Keep loading, even blacklisted

[Thanos Baloukas]

On 05/02/2016 09:05 μμ, Rob wrote:

I ask here because module blacklisting is mentioned in the lfs book.
I have 3 cards on my system: an onboard and two PCI cards. I want the
onboard one to not load. It is too quiet and I can barely hear it; so it
is therefore useless for my purposes.
It is an Intel board, so in /etc/modprobe.d/blacklist.conf, I put
blacklist snd_hda_intel*


On 3.19.1 kernel the module name is snd-hda-intel with `-', not `_'
Are you sure you are using the correct module name?


That didn't appear to entirely work, however. Here is  an entry from
aplay -l:
card 0: Intel [HDA Intel], device 0: ALC662 rev1 Analog [ALC662 rev1
Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: Intel [HDA Intel], device 1: ALC662 rev1 Digital [ALC662 rev1
Digital]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
And here is from cat /proc/asound.cards:
0 [Intel  ]: HDA-Intel - HDA Intel
  HDA Intel at 0xfdff8000 irq 28
How can I really stop this card from loading?
Thank you all for your help on both lists. I'm almost upup and running
to my satisfaction.




--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Why so few gcc test failures in lfs log?

[Thanos Baloukas]

On 15/01/2016 03:24 μμ, Richard wrote:

On 01/14/2016 03:29 PM, Thanos Baloukas wrote:

On 14/01/2016 03:27 πμ, Richard wrote:


Your results are very similar to mine. I used a
1.7 GHz P4 and 3GB RAM.

I wonder if there is a way to disable the gcc tests
that are just going to time out anyway. There is
some information here about running tests selectively
but I after reading it I still don't know how to
go about it:


https://gcc.gnu.org/install/test.html



Perhaps there is a way to disable them, but why?
As I see it, the number of the failing tests is not significant
compared to the total tests' number. Bruce's explanation that
gcc developers are not concerned with old systems seems convincing.
And finally, as long as the resulting system is robust - as all of my
lfs installations have been - I think that there is no reason to worry.



I thought about this mostly because of the time it takes
me to do the tests. I didn't time them exactly, but I
think it took close to 24 hours to do the gcc tests. So
any reduction in that amount of time would be welcome.



I see. I counted 24 timeout warnings in the log. No matter
what the timeout limit is, it's useless to run those tests
just to fail. On my next gcc build I'll consider removing them.
Lets wait for Douglas' results.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Why so few gcc test failures in lfs log?

[Thanos Baloukas]

On 14/01/2016 03:27 πμ, Richard wrote:


Your results are very similar to mine. I used a
1.7 GHz P4 and 3GB RAM.

I wonder if there is a way to disable the gcc tests
that are just going to time out anyway. There is
some information here about running tests selectively
but I after reading it I still don't know how to
go about it:


https://gcc.gnu.org/install/test.html



Perhaps there is a way to disable them, but why?
As I see it, the number of the failing tests is not significant
compared to the total tests' number. Bruce's explanation that
gcc developers are not concerned with old systems seems convincing.
And finally, as long as the resulting system is robust - as all of my
lfs installations have been - I think that there is no reason to worry.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Why so few gcc test failures in lfs log?

[Thanos Baloukas]

On 12/01/2016 04:42 μμ, Richard wrote:

On 01/12/2016 09:14 AM, Douglas R. Reno wrote:



I am going to try building on a 2.4GHz P4 with 1.5GB of RAM soon. Do you
want me to post my results for GCC here?

Douglas R. Reno





I would like to see a summary of the test results.
I have 3GB RAM, which is the maximum that can be
installed on my MB.



On a Pentium4 2.4GHz with 1.5GB ram like Douglas', gcc 4.9.2:

  === g++ tests ===

Running target unix
FAIL: g++.dg/asan/asan_test.C  -O2  AddressSanitizer_HugeMallocTest 
Ident((char*)malloc(size))[-1] = 0 output pattern test

WARNING: program timed out.
FAIL: g++.dg/cilk-plus/CK/catch_exc.cc  -O1 -fcilkplus execution test
WARNING: program timed out.
FAIL: g++.dg/cilk-plus/CK/catch_exc.cc  -O3 -fcilkplus execution test
WARNING: program timed out.
FAIL: g++.dg/cilk-plus/CK/catch_exc.cc  -g -fcilkplus execution test
WARNING: program timed out.
FAIL: g++.dg/cilk-plus/CK/catch_exc.cc  -g -O2 -fcilkplus execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -O1 execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -O1 execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -O3 execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -O3 execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -g execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -g execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -g -O2 execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -g -O2 execution test
FAIL: g++.dg/ipa/pr61160-3.C -std=gnu++98 execution test
FAIL: g++.dg/ipa/pr61160-3.C -std=gnu++11 execution test
FAIL: g++.dg/ipa/pr61160-3.C -std=gnu++1y execution test
XPASS: g++.dg/tls/thread_local-order2.C -std=c++11 execution test
XPASS: g++.dg/tls/thread_local-order2.C -std=c++1y execution test


  === g++ Summary ===

# of expected passes87019
# of unexpected failures   16
# of unexpected successes  2
# of expected failures 443
# of unsupported tests 2798


  === gcc tests ===

Running target unix
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -g  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -g  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -O1  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -O1  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -O2 -std=c99  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -O2 -std=c99  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -O2 -ftree-vectorize 
execution test

WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -O2 -ftree-vectorize 
execution test

WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -O3 -g  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -O3 -g  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/spawning_arg.c  -O3 -flto -g  execution test
WARNING: program timed out.
FAIL: c-c++-common/cilk-plus/CK/steal_check.c  -O3 -flto -g  execution test

  === gcc Summary ===

# of expected passes105536
# of unexpected failures   12
# of expected failures 249
# of unsupported tests 1217


  === libstdc++ tests ===

...
Running 
/build/tmp/gcc-4.9.2/libstdc++-v3/testsuite/libstdc++-dg/conformance.exp ...
FAIL: 20_util/enable_shared_from_this/cons/constexpr.cc (test for excess 
errors)

FAIL: 20_util/shared_ptr/cons/constexpr.cc (test for excess errors)
FAIL: 20_util/unique_ptr/cons/constexpr.cc (test for excess errors)
FAIL: 20_util/weak_ptr/cons/constexpr.cc (test for excess errors)
FAIL: 30_threads/future/cons/constexpr.cc (test for excess errors)
FAIL: 30_threads/shared_future/cons/constexpr.cc (test for excess errors)
Running 
/build/tmp/gcc-4.9.2/libstdc++-v3/testsuite/libstdc++-prettyprinters/prettyprinters.exp 
...


  === libstdc++ Summary ===

# of expected passes9837
# of unexpected failures   6
# of expected failures 41
# of unsupported tests 274

  === libatomic Summary ===

# of expected passes44
# of unsupported tests 5

 === libgomp Summary ===

# of expected passes3198
# of unsupported tests 35

  === libitm Summary ===

# of expected passes26
# of expected failures 3
# of unsupported tests 1

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally r

Re: [lfs-support] LFS 7.8 Chapter 5 script

[Thanos Baloukas]

On 11/01/2016 01:09 πμ, Bruce Dubbs wrote:

Thanos Baloukas wrote:

On 11/01/2016 12:10 πμ, Ken Moffat wrote:


Thanos, thanks for killing this spurious "it appeared to fail"
thread.  I've just looked at one of my own 7.8 logs - I had assumed
that configure ended with that warning, but in my case it appeared
during configure's run.


In my log configure ends with:

configure: WARNING: Could not find a terminal library among tinfo ncurses
curses termlib termcap terminfo
configure: WARNING: The programs from `info' directory will not be built

on lines  529 and 530.
I don't remember if I even noticed them. It was scripted build.


Not that it matters.  It's a warning, not an error.  And what we need
here are the libraries for linking.  The terminal library references in
chapter 5 are not meaningful for building and they are installed in
Chapter 6 anyway.


My point wasn't that the warnings matter because they are at the end.
I just wanted to spot the difference between Ken's log and mine.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] LFS 7.8 Chapter 5 script

[Thanos Baloukas]

On 11/01/2016 12:10 πμ, Ken Moffat wrote:


Thanos, thanks for killing this spurious "it appeared to fail"
thread.  I've just looked at one of my own 7.8 logs - I had assumed
that configure ended with that warning, but in my case it appeared
during configure's run.


In my log configure ends with:

configure: WARNING: Could not find a terminal library among tinfo 
ncurses curses termlib termcap terminfo

configure: WARNING: The programs from `info' directory will not be built

on lines  529 and 530.
I don't remember if I even noticed them. It was scripted build.

 Not quite sure how James even managed to

notice it, he must have kept a log, or at least looked at config.log
(which would be deleted when the source directory was deleted by his
script).

ĸen



--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] LFS 7.8 Chapter 5 script

[Thanos Baloukas]

On 10/01/2016 11:23 μμ, Read, James C wrote:


I have run it several times and it gets to completion as long as the sources 
directory is as expected with no major errors except for the Texinfo warnings.

I can only conclude that the book is either

a) wrong
b) a bit of an online joke to see how many times people will try this before 
they just give up
c) the book isn't an online joke and some peculiarity about my host system 
wasn't forseen by the book which works well for others

If nobody is going to test the script I put up. I guess I'll never know which 
of the above is the problem.


On lfs-7.7 with ncurses-5.9, texinfo-5.2 and an lfs-svn as host
I had the same warnings with you on chapter 5 and that didn't cause any 
problem.


--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] LFS 7.8 Gcc second pass problem

[Thanos Baloukas]

On 08/01/2016 05:11 μμ, Read, James C wrote:

No, && is the logical AND. When it's between two commands,
the second is executed only if the first succeeds.
I don't understand why you used it since you don't know what it does.


That's what I thought it does. And that's why I used it. I put it at the end of 
all commands so that if any of them failed I would know about it.

What I don't really understand is how this then causes variables to go out of 
scope.


I hope William will bear with me.

CC=$LFS_TGT-gcc &&
AR=$LFS_TGT-ar &&
RANLIB=$LFS_TGT-ranlib &&
../binutils-2.25.1/configure --prefix=/tools --disable-nls 
--disable-werror --with-lib-path=/tools/lib --with-sysroot


What bash understands with that is:
set CC var AND if succeeds set AR AND if succeeds set RANLIB
AND if succeeds execute configure. The  three vars' settings
and the configure are separate commands.
They are not in one command line.
configure has #! /bin/sh on top. That means that a new shell will start
that will NOT have CC AR and RANLIB in it's environment because of the &&'s.

CC=$LFS_TGT-gcc\
AR=$LFS_TGT-ar \
RANLIB=$LFS_TGT-ranlib \
../binutils-2.25.1/configure   \
--prefix=/tools\
--disable-nls  \
--disable-werror   \
--with-lib-path=/tools/lib \
--with-sysroot

What bash understands with that is:
set CC AR and RANLIB and pass them to the environment of the shell that
will execute configure because they are in configure's command line.
The `\' is line continuation.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] LFS 7.8 Gcc second pass problem

[Thanos Baloukas]

On 08/01/2016 04:31 μμ, Read, James C wrote:



Hope it is clear.


So, the effect of && is like starting a new bash session? Is that why they fall 
out of scope?


No, && is the logical AND. When it's between two commands,
the second is executed only if the first succeeds.
I don't understand why you used it since you don't know what it does.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] cannot stat './modules.builtin', kernel make error

[Thanos Baloukas]

On 20/12/2015 07:18 μμ, Joshua Huffman wrote:

Greetings.  This is my 3rd attempt at LFS.  I will actually finish it this time.
I have encountered trouble during my build, and despite google searches, I 
haven't been able to work around the issue.

LFS version 7.8
I am building linux kernel version 4.2.8, as per the note in section 3.2, which states to use the latest 4.2.x version number.  
The entire build has gone very smoothly up until section 8.3.1 - Installation of Kernel.  I began with "make mrproper". 
 I then ran "make allmodconfig", thinking extra modules won't hurt anything. "make LANG=en_US LC_ALL= 
menuconfig" as per the book to double check and make adjustments. "make" completed successfully, but when I run 
"make modules_install", I get the following error:

cp: cannot stat './modules.builtin': No such file or directory
Makefile:1125: recipe for target '_modinst_' failed
make: *** [_modinst_] Error 1


Here is the referenced portion of the Makefile:
1123 PHONY += _modinst_
1124 _modinst_:
1125 @rm -rf $(MODLIB)/kernel
1126 @rm -f $(MODLIB)/source
1127 @mkdir -p $(MODLIB)/kernel
1128 @ln -s `cd $(srctree) && /bin/pwd` $(MODLIB)/source
1129 @if [ ! $(objtree) -ef  $(MODLIB)/build ]; then \
1130 rm -f $(MODLIB)/build ; \
1131 ln -s $(CURDIR) $(MODLIB)/build ; \
1132 fi

My host system is a Slackware 14.1 full install (minus KDE).
I can post the output of the host version script from the book's section vii if 
anyone would like to see it.  Any other relevant information can be provided 
upon request.

I have tried several times with no success.  Makefiles and make errors confuse me very badly. 
 If I have read & interpreted correctly, though, the make command exits when it doesn't 
find the file modules.builtin.  I tried "touch"-ing the file, but that doesn't work 
either.  I am at a loss.  Has anyone experienced this?  Should I just start over using the 
kernel from the standard wget-list?  Anyone's thoughts are appreciated.

Joshua

Looking in the Makefile of my linux-3.19.1 I see that there is a 
modules.builtin target

that should be done with make. I would try to see what happens with

make modules.builtin

Also, a

find . -name "modules.builtin"

from the top source directory here, finds a ./modules.builtin that you 
probably don't have,

and many others in various module directories. Do you have them?
And a last thought, did you use -jx with make? If you did, did you try 
without it?

HTH

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] makefile:376: recipe for target 'perl' failed

[Thanos Baloukas]

On 19/12/2015 09:24 μμ, Thanos Baloukas wrote:

On 19/12/2015 07:59 μμ, Jose Angel Fernandez wrote:




It should be that /bin/bash, not /bin/dash is the active shell.



I have the same problem and reading the messages I've found that my link
was to dash instead of bash. Which is the best way to solve the problem?
Recompiling from pass 1 or pass 2?



It would be safer to start from the beginning.


Sorry, wrong expression.
There are people that changed the /bin/sh symlink after they had started 
the built
and because of that they had failures later. So I advise to definitely 
start from the beginning.


--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] makefile:376: recipe for target 'perl' failed

[Thanos Baloukas]

On 19/12/2015 07:59 μμ, Jose Angel Fernandez wrote:




It should be that /bin/bash, not /bin/dash is the active shell.



I have the same problem and reading the messages I've found that my link
was to dash instead of bash. Which is the best way to solve the problem?
Recompiling from pass 1 or pass 2?



It would be safer to start from the beginning.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Problem in section 5.3 of lfs book

[Thanos Baloukas]

On 17/11/2015 04:18 μμ, Joseph Hesse wrote:


In the second "Important" box in section 5.3, item 3 says that I should
use tar to extract the packages in /mnt/lfs/sources. Am I reading this
right? There are 27 .tar.gz files, 11 .tar.bz2 files and 32 tar.xz
packages. Extracting each one from the command line seems error prone. I
am thinking about creating a script that would extract each package. Is
this a good approach?



If you mean that it's difficult to type correctly the packages' names,
use bash completion. Go to /mnt/lfs/sources directory and type for example

tar -xf gc

and press the tab key. It will print the file names in the current
directory that start with gc, or if there is only one it will complete
the command line.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Problem in section 5.3 of lfs book

[Thanos Baloukas]

On 17/11/2015 04:18 μμ, Joseph Hesse wrote:


Also, I had to stop work and shut down my host system.  When I restarted
it I had a gdm that displayed my name and the name "lfs". If I selected
"lfs" nothing happened,  I was back to the gdm prompt. The only way I
can be user "lfs" is to log in as myself and "su - lfs". Does this seem OK?

The lfs user was created for the LFS book's needs, and you only have to 
be lfs when installing packages by the book, that is on a terminal.

I don't see a reason to start a graphical session with gdm as user lfs.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] 5.33

[Thanos Baloukas]

On 04/11/2015 10:17 μμ, Michael Havens wrote:

Okay, I do not know what I did but some commands work (cd) and some
don't (cp, tar, mv).


Just a <> doesn't help.
What are the error messages you got by the commands or by the shell?

I'm thinking my bash_profile has someting to do

with it.  I did:

cat > ~/.bashrc << "EOF"
set +h
umask 022
LFS=/mnt/lfs
LC_ALL=POSIX
LFS_TGT=$(uname -m)-lfs-linux-gnu
PATH=/tools/bin:/bin:/usr/bin
export LFS LC_ALL LFS_TGT PATH
EOF
(as instructed earlier)
and I'm thinking 'set +h' is the culprit. Please, how should I proceed?



--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] export

[Thanos Baloukas]

On 29/10/2015 12:11 πμ, Michael Havens wrote:

The $LFS is set. I'm trying to set another variable
sources=$LFS/sources. it refuses to stay set.

lfs@debian:~$ cat ~/.bash_profile
exec env -i HOME=$HOME TERM=$TERM PS1='\u:\w\$ ' /bin/bash
export LFS=/mnt/lfs
export sources=$LFS/sources
lfs@debian:~$ exit
exit
root@debian:~# su - lfs
lfs@debian:~$ cat ~/.bash_profile
exec env -i HOME=$HOME TERM=$TERM PS1='\u:\w\$ ' /bin/bash
export LFS=/mnt/lfs
export sources=$LFS/sources
lfs@debian:~$ cd $sources
lfs@debian:~$ pwd
/home/lfs
lfs@debian:~$ nano ~/.bash_profile
lfs@debian:~$ cat ~/.bash_profile
exec env -i HOME=$HOME TERM=$TERM PS1='\u:\w\$ ' /bin/bash
export LFS=/mnt/lfs
export sources=/mnt/lfs/sources
lfs@debian:~$ cd $sources
lfs@debian:~$ exit
exit
root@debian:~# su - lfs
lfs@debian:~$ cd $sources
lfs@debian:~$ cd $LFS/sources
lfs@debian:/mnt/lfs/sources$



Chapter 4.4. Setting Up the Environment:

The exec env -i.../bin/bash command in the .bash_profile file
replaces the running shell with a new one with a completely empty
environment, except for the HOME, TERM, and PS1 variables.
...
The new instance of the shell is a non-login shell, which does not read 
the /etc/profile or .bash_profile files, but rather reads the .bashrc 
file instead. Create the .bashrc file now:


Then it says to set the lfs user's environment variables in .bashrc

As I understand it, the lines in .bash_profile after the
exec env -i.../bin/bash command do nothing. Put them in .bashrc
as the book says.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] 5.10

[Thanos Baloukas]

On 28/10/2015 05:25 μμ, Michael Havens wrote:



On Wed, Oct 28, 2015 at 11:00 AM, Pierre Labastie
mailto:pierre.labas...@neuf.fr>> wrote:

Two solutions:
- Use ln -sf (force)
- remove the link command from your script...


Thank you Pierre force option it is.
As for my first run having already created it not really. It ran for lie
10 seconds and then barfed up the rest of the script in errors followed
with the link error. Wait it is just a link! It is linking to the
program. It is linking to the recompiled program if I recompile it. I
get it!


If I got it right, a command in your script failed and the script 
continued. If that's the case, the rest commands naturally failed,

because each one depends on the previous to be executed with no errors.
Thus it is desirable in build scripts and scripts in general to exit
when a command fails, so that the rest commands are not executed
just to through out errors who make it harder to track down the first
one that failed. To accomplish that you can make the first line

#!/bin/bash -e

or add the line "set -e" under #!/bin/bash

and your scripts will exit when a command fails.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Configure error in step 5.8 Libstdc++-5.2.0 of LFS ver7.8

[Thanos Baloukas]

On 12/10/2015 06:55 μμ, Bruce Dubbs wrote:

Jun Hu wrote:



another thing to note is the configure command contains
"--with-gxx-include-dir=/tools/$LFS_TGT/include/c++/5.2.0", however
on my
system there is no such directory:
"""
lfs at vm-svr:~$ echo $LFS_TGT
x86_64-lfs-linux-gnu

lfs at vm-svr:~$ ls -l /tools/$LFS_TGT/
total 8
drwxr-xr-x 2 lfs lfs 4096 Oct 11 15:44 bin
drwxr-xr-x 3 lfs lfs 4096 Oct 11 15:44 lib
"""


The /tools/$LFS_TGT/include/c++/5.2.0 should have been installed by gcc.


I didn't install lfs 7.8 and I may be wrong but I think that
this directory is installed by Libstdc++-5.2.0 .
If I grep for "/tools/i686-lfs-linux-gnu/include/c++/4.9.2"
on tools buildlogs on my x86 lfs 7.7 systemd installation
it finds references only in libstdc++-build.log .
Furthermore gcc-5.2.0/libstdc++-v3/configure has:
--with-gxx-include-dir=DIR  installation directory for include files


Following error in config.log (in $LFS/sources/gcc-build) seems also
suggest the specified include-dir is not right.
"conftest.c:10:19: fatal error: stdio.h: No such file or directory


That's your problem.  /usr/include/stdio.h should be installed by glibc.


Is this a typo? s/usr/tools/

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] [6.?] Forgot which section I was.

[Thanos Baloukas]

On 02/07/2015 11:17 πμ, Thanos Baloukas wrote:

On 02/07/2015 09:15 πμ, Elias Polytarchos wrote:

On Thu, Jul 2, 2015 at 7:19 AM, Mateus Furquim mailto:furquim@gmail.com>> wrote:

Hi everyone,

I was doing Linux From Scratch a couple of months ago, but can't
remember in which section of the chapter 6 I was.

Is there anyway to know which are the packages I've already
installed? (Other than ls'ing and looking at programs installed)

I mean, is there a tool or shell script that I could use to search
all the installed packages on LFS?

Sincerely,

Mateus

Hi!

While I am not aware of a tool that will indicate the installed packages
per se, most of the packages store information in the
${INSTALL_PREFIX}/lib/pkgconfig/.

So, by checking out these directories one should be able to
approximately determine which packages have been installed.



Something else that could help is to check bash history.
If your HISTSIZE value is big enough and you didn't use much the console
during the last period, you can see the commands you executed
when doing LFS (e.g. tar xf) and find out the last installed package.
Even if you are using buildscripts, you can see the last ran.


Sorry, I didn't think you're in chroot and I'm not sure if you can use
that in chroot.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] [6.?] Forgot which section I was.

[Thanos Baloukas]

On 02/07/2015 09:15 πμ, Elias Polytarchos wrote:

On Thu, Jul 2, 2015 at 7:19 AM, Mateus Furquim mailto:furquim@gmail.com>> wrote:

Hi everyone,

I was doing Linux From Scratch a couple of months ago, but can't
remember in which section of the chapter 6 I was.

Is there anyway to know which are the packages I've already
installed? (Other than ls'ing and looking at programs installed)

I mean, is there a tool or shell script that I could use to search
all the installed packages on LFS?

Sincerely,

Mateus

Hi!

While I am not aware of a tool that will indicate the installed packages
per se, most of the packages store information in the
${INSTALL_PREFIX}/lib/pkgconfig/.

So, by checking out these directories one should be able to
approximately determine which packages have been installed.



Something else that could help is to check bash history.
If your HISTSIZE value is big enough and you didn't use much the console
during the last period, you can see the commands you executed
when doing LFS (e.g. tar xf) and find out the last installed package.
Even if you are using buildscripts, you can see the last ran.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] e2fsprogs make error

[Thanos Baloukas]

On 23/06/2015 12:35 πμ, Sandeep Singh wrote:

Another thing I am not sure about is whether to keep the extracted
directories in the sources folders or to delete them. The book says to
delete them till otherwise indicated. I read on the linux questions
forum it is a good idea to keep them.
What would be the right thing to do?


Please do not top post. To follow the book of course.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] e2fsprogs make error

[Thanos Baloukas]

On 22/06/2015 08:43 μμ, Sandeep Singh wrote:

Hello,

I am using Arch linux as the host system and version 7.7 of the LFS
book. The check commands after gcc build in 6.17 came out spot on.
Now while making e2fsprogs I am getting this error

./../../lib/ext2fs/closefs.c:1:3: error: expected '=', ',', ';', 'asm'
or '__attribute__' before '.' token
  nt.*old_desc_blocks/s/int/blk64_t/
^


It seems to me that you didn't apply correctly the sed command.
I'd suggest to remove the source dir, extract the tarball again,
copy and paste the sed carefully and proceed.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Problem in the make of glibc-2.21 in LFS 7.7

[Thanos Baloukas]

On 12/04/2015 10:20 μμ, João wrote:

Hey guys, newbie here...

LFS version: LFS 7.7
Host System: Elementary OS Freya 64 bits (Ubuntu 14.04 based)

I'm having problems with the installation of glibc-2.21 in the chapter
5.7 of LFS 7.7. When I run the 'make' command to compile the package it
just return the output:

lfs@joao:/mnt/lfs/sources/glibc-build$ make
make: *** No targets specified and no makefile found.  Stop.

In the previous step, when I prepare Glibc for compilation with the
commands:

../glibc-2.21/configure \
   --prefix=/tools \
   --host=$LFS_TGT \
   --build=$(../glibc-2.21/scripts/config.guess) \
   --disable-profile \
   --enable-kernel=2.6.32 \
   --with-headers=/tools/include \
   libc_cv_forced_unwind=yes \
   libc_cv_ctors_header=yes \
   libc_cv_c_cleanup=yes

the output returned is:

checking build system type... x86_64-unknown-linux-gnu
checking host system type... x86_64-lfs-linux-gnu
checking for x86_64-lfs-linux-gnu-gcc... x86_64-lfs-linux-gnu-gcc
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether x86_64-lfs-linux-gnu-gcc accepts -g... yes
checking for gcc... gcc
checking for x86_64-lfs-linux-gnu-g++... x86_64-lfs-linux-gnu-g++
checking whether we are using the GNU C++ compiler... yes
checking whether x86_64-lfs-linux-gnu-g++ accepts -g... yes
checking for x86_64-lfs-linux-gnu-readelf... no
checking for readelf... readelf
configure: WARNING: using cross tools not prefixed with host triplet
checking for sysdeps preconfigure fragments... aarch64 alpha arm hppa
i386 m68k microblaze mips nios2 powerpc s390 sh sparc tile x86_64
checking whether x86_64-lfs-linux-gnu-gcc compiles in -mx32 mode by
default... no

configure: running configure fragment for add-on libidn
checking for assembler and linker STT_GNU_IFUNC support... no
checking whether .text pseudo-op must be used... yes
checking sysdep dirs... sysdeps/unix/sysv/linux/x86_64/64
sysdeps/unix/sysv/linux/x86_64 sysdeps/unix/sysv/linux/x86
sysdeps/unix/sysv/linux/wordsize-64 sysdeps/x86_64/nptl
sysdeps/unix/sysv/linux sysdeps/nptl sysdeps/pthread sysdeps/gnu
sysdeps/unix/inet sysdeps/unix/sysv sysdeps/unix/x86_64 sysdeps/unix
sysdeps/posix sysdeps/x86_64/64 sysdeps/x86_64/fpu sysdeps/x86/fpu
sysdeps/x86_64 sysdeps/x86 sysdeps/ieee754/ldbl-96
sysdeps/ieee754/dbl-64/wordsize-64 sysdeps/ieee754/dbl-64
sysdeps/ieee754/flt-32 sysdeps/wordsize-64 sysdeps/ieee754 sysdeps/generic
checking for a BSD-compatible install... /usr/bin/install -c
checking whether ln -s works... yes
checking whether as is GNU as... yes
checking whether ld is GNU ld... yes
checking for as... as
checking version of as... 2.24, ok
checking for ld... ld
checking version of ld... 2.24, ok
checking for gnumake... no
checking for gmake... no
checking for make... make
checking version of make... 3.81, ok
checking for gnumsgfmt... no
checking for gmsgfmt... no
checking for msgfmt... msgfmt
checking version of msgfmt... 0.18.3, ok
checking for makeinfo... makeinfo
checking version of makeinfo... 5.2, ok
checking for sed... sed
checking version of sed... 4.2.2, ok
checking for gawk... gawk
checking version of gawk... 4.0.1, ok
checking if x86_64-lfs-linux-gnu-gcc is sufficient to build libc... yes
checking for x86_64-lfs-linux-gnu-nm... no
checking for nm... nm
checking LD_LIBRARY_PATH variable... ok
checking whether GCC supports -static-libgcc... -static-libgcc
checking for bash... /bin/bash
checking for perl... /usr/bin/perl
checking for install-info... /usr/bin/install-info
checking for bison... no


bison should be in /usr/bin/
check your host system requirements


checking for libc-friendly stddef.h... yes
checking whether we need to use -P to assemble .S files... no
checking for .set assembler directive... no
checking for assembler gnu_unique_object symbol type... yes
checking for .previous assembler directive... yes
checking for .protected and .hidden assembler directive... yes
checking whether __attribute__((visibility())) is supported... no
configure: error: compiler support for visibility attribute is required

How can I compile the glibc-2.21 package properly?
Thanks for the support!



--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] Unclear where to execute make mrproper, Chapter 5.6

[Thanos Baloukas]

On 06/04/2015 12:29 πμ, rail shafigulin wrote:


No.  The commands are in the correct order.

The key phrase is "user-visible kernel headers".  When you extract
the tarball, you get *all* the headers.  The make command extracts
only those that we want.

   -- Bruce


Bruce,

I wasn't able to find mrproper target in any of the makefiles in
previously extracted packages. If all the command are in correct order,
where should I look for mrproper target?

Any help is appreciated.






linux-3.19.1 $ grep mrproper Makefile
no-dot-config-targets := clean mrproper distclean \
echo >&2 "  $(srctree) is not clean, please run 'make 
mrproper'"; \
# make mrproper  Delete the current configuration, and all generated files
# Directories & files removed with 'make mrproper'
# mrproper - Delete all generated files, including .config
mrproper: rm-dirs  := $(wildcard $(MRPROPER_DIRS))
mrproper: rm-files := $(wildcard $(MRPROPER_FILES))
mrproper-dirs  := $(addprefix _mrproper_,Documentation/DocBook scripts)
PHONY += $(mrproper-dirs) mrproper archmrproper
$(mrproper-dirs):
$(Q)$(MAKE) $(clean)=$(patsubst _mrproper_%,%,$@)
mrproper: clean archmrproper $(mrproper-dirs)
distclean: mrproper
	@echo  '  mrproper	  - Remove all generated files + config + various 
backup files'

@echo  '  distclean   - mrproper + remove editor backup and patch 
files'

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] configure: error: cannot compute suffix of object files: cannot compile

[Thanos Baloukas]

On 03/04/2015 05:22 πμ, rail shafigulin wrote:


If you hit for
"configure: error: cannot compute suffix of object files: cannot
compile"
on google, it finds many references to GMP/MPFR/MPC
Make sure you understood the note in the book above the commands
that extract those tarballs.

--
Thanos


Hey Thanos.

Don't get mad but I was not able to find your post online. I even
searched for  "configure: error: cannot compute suffix of object files:
cannot compile" Thanos, but wasn't able to find it. (If you don't mind
giving me a link to it, I'd appreciate it). There are plenty of other
posts that look like they have the same issue as I am but the solutions
suggested take additional steps which are not mentioned in the book. I
don't want to do that, not just yet. As per following instructions in
the books here is the 'ls -lah' of  my gcc -4.9.2 directory.


No mad at all. I just found the references I mentioned on google
and thought that I might help. It's not clear to me which post of mine
you are refering to. If you mean the google search findings,
the most typical is the following:
https://gcc.gnu.org/wiki/FAQ#Configuration_fails_with_.27.27configure:_error:_cannot_compute_suffix_of_object_files:_cannot_compile.27.27._What_is_the_problem.3F


I apologize in advance if my questions seem trivial or repetitive. I'm
trying to learn how to do this and I'm very appreciative of your help.


drwxr-xr-x  3 lfs  lfs  4.0K Oct 30 04:28 fixincludes
drwxr-xr-x 18 lfs  lfs   24K Apr  2 13:24 gcc
drwxr-xr-x 15 lfs  lfs  4.0K Mar 25  2014 gmp

...

drwxr-xr-x  6 lfs  lfs  4.0K Jan 15  2014 mpc
drwxr-xr-x  9 lfs  lfs  4.0K Mar 13  2013 mpfr



That shows that you uncompressed the tarballs in the right directory
and renamed them correctly.
Anyway, I think that the most wise thing to do, (if you haven't done
already), since you had the problem with the /bin/sh symlink
and you are in the begining of the installation, is to remove everything 
in /tools and redo Binutils-2.25 - Pass 1


--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] configure: error: cannot compute suffix of object files: cannot compile

[Thanos Baloukas]

On 02/04/2015 10:45 μμ, rail shafigulin wrote:


The gcc dev's caught you out ;-)  I think that what you have
attached is the top-level config.log.  That runs before 'make', but
then make runs configure in various directories.  In this case, it
seems to have failed in the libgcc directory.

The attached file shows error(s) which are normal.  For example,
configure creates a tiny test program to see if libisl exists and
the long output shows an apparent error, but the screen output from
the initial configure would have been:

checking for objdir... .libs
checking for version 0.10 of ISL... no
checking for version 0.11 of ISL... no
checking for version 0.12 of ISL... no
*** This configuration is not supported in the following subdirectories:
  target-libquadmath gnattools target-libada target-libgfortran
target-libgo target-libffi target-libbacktrace target-zlib
target-libjava target-libobjc target-libgomp target-libcilkrts
target-libatomic target-libitm target-libstdc++-v3
target-libsanitizer target-libvtv target-libssp target-boehm-gc
 (Any other directories should still work fine.)

So, what you now need to do is check the newest config.log (in
libgcc, I think), then search for 'cannot compute suffix of object
files:' e.g. in view or less, use the command
/cannot\ compute

and then look at the lines above, probably a screen or two earlier.
You should see the 'checking for suffix' message, a program
fragment, and a more-detailed error, probably from gcc.  That error
message should identify the problem.

Ken you were correct that the config.log I need is in libgcc and that
the error is there as well. Here is the snippet I believe that causes a
problem. Unfortunately I can't figure out what it means. Just in case
I'm also posting the contents of
/mnt/lfs/sources/gcc-build/x86_64-lfs-linux-gnu/libgcc/config.log. Any
help is appreciated.



If you hit for
"configure: error: cannot compute suffix of object files: cannot compile"
on google, it finds many references to GMP/MPFR/MPC
Make sure you understood the note in the book above the commands
that extract those tarballs.

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] gcc erorr: unrecognized command lin eoption -V, -qversion

[Thanos Baloukas]

On 02/04/2015 04:21 μμ, rail shafigulin wrote:

I'm new to LFS and using version 7.7 Current in Chapter 5.4.1


Welcome to LFS. I hope you enjoy it.


binutils configuration I'm getting the following errors:

gcc: error: unrecognized command line option '-V'
gcc: error: unrecognized command line option '-qversion'
g++: error: unrecognized command line option '-V'
g++: error: unrecognized command line option '-qversion'


The configure script does some tests and some of these fail.
They are not considered build errors.


Here is the output config.log. The outputs from version-check.sh and
library-check.sh are below as well. I'm not quite clear what is wrong.
Any help in resolving the issue is appreciated.

This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.

It was created by configure, which was
generated by GNU Autoconf 2.64.  Invocation command line was

   $ ../binutils-2.25/configure --prefix=/tools --with-sysroot=/mnt/lfs
--with-lib-path=/tools/lib
...
...
#define PACKAGE_URL ""
#define LT_OBJDIR ".libs/"

configure: exit 0


If configure exited 0 everything is ok. The same applies to make
and other commands. If in doubt, do 'echo $?'. It prints the last
command's exit status. If it's 0 the command succeeded.


Here is the output from version-check.sh:
bash, version 4.3.11(1)-release
/bin/sh -> /bin/dash


It should be /bin/sh -> /bin/bash


Binutils: (GNU Binutils for Ubuntu) 2.24
bison (GNU Bison) 3.0.2
/usr/bin/yacc -> /usr/bin/bison
bzip2,  Version 1.0.6, 6-Sept-2010.
Coreutils:  8.21
diff (GNU diffutils) 3.3
find (GNU findutils) 4.4.2
GNU Awk 4.0.1
/usr/bin/awk -> /usr/bin/gawk
gcc (Ubuntu 4.8.2-19ubuntu1) 4.8.2
g++ (Ubuntu 4.8.2-19ubuntu1) 4.8.2
(Ubuntu EGLIBC 2.19-0ubuntu6.6) 2.19
grep (GNU grep) 2.16
gzip 1.6
Linux version 3.13.0-48-generic (buildd@orlo) (gcc version 4.8.2 (Ubuntu
4.8.2-19ubuntu1) ) #80-Ubuntu SMP Thu Mar 12 11:16:15 UTC 2015
m4 (GNU M4) 1.4.17
GNU Make 3.81
GNU patch 2.7.1
Perl version=5.18.2
sed (GNU sed) 4.2.2
tar (GNU tar) 1.27.1
makeinfo (GNU texinfo) 5.2
xz (XZ Utils) 5.1.0alpha
g++ compilation OK

Here is the output from library-check.sh
libgmp.la : not found
libmpfc.la : not found
libmpc.la : not found



Good luck
--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Do not top post on this list.

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?

http://en.wikipedia.org/wiki/Posting_style

Re: [lfs-support] A small correction is necessary to some text

[Thanos Baloukas]

On 03/05/2014 01:54 μμ, Googlemail wrote:

In previous versions of LFS, util-linux was built early in Chapter 6. In version 7.5 it is built 
much later. The build parameters have been adjusted to reflect this but some of the explanatory 
material has not been updated to match. There are references to things not being built 
"because util-linux has already installed them". This should now read "because 
util-linux will install them later". The relevant sections are for procps, e2fsprogs and 
inetutils.

It's a minor grouse but I think it could cause a lot of confusion in novices 
(it certainly confused me and this is my second LFS build).

Just out of curiosity, why was it moved?

Hazel



If you look in the changelog, it says that the change fixes

wiki.linuxfromscratch.org/lfs/ticket/3467

--
Thanos
--
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page