Re: [lfs-support] Slightly different output from gcc test in chapter 6.17.1
Hi, On Sat, Mar 02, 2013 at 06:56:29PM +0100, Niels Terp wrote: >Hi, >I'm doing the newly released version 7.3 on a OpenSuSE 12.3 host (32 bit). >In this chapter I get some of the output right, but in the wrong sequence: > The command grep -B4 '^ /usr/include' dummy.log >Should give this output: >#include <...> search starts here: >/usr/local/include >/usr/lib/gcc/i686-pc-linux-gnu/4.7.2/include >/usr/lib/gcc/i686-pc-linux-gnu/4.7.2/include-fixed >/usr/include >But in my case I get: >root:/sources/gcc-build# grep -B4 '^ /usr/include' dummy.log >#include <...> search starts here: >/usr/lib/gcc/i686-pc-linux-gnu/4.7.2/include >/usr/local/include >/usr/lib/gcc/i686-pc-linux-gnu/4.7.2/include-fixed >/usr/include >Does this mean anything ? >Niels The sequence is not a concern - just that the right include dirs are being searched. On my system, the order of the output is similair to yours with no problems. In other words, don't worry, grab your drink of choice and carry on. Best of luck ae -- http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] how to build newer glibc without root account?
Hello, On Mon, Feb 18, 2013 at 05:55:00PM +0800, Li Li wrote: > hi all > I am not sure whether this question is suitable here or not. sorry if not. > I have some binary application which need glibc at least 2.4 but > my linux box is 2.3.4. > I don't have root acount. is there any method to install a newer > glibc in my home directory and use some tricky to let the binary > application running using my glibc? > here is my environment: > $ uname -a Linux x 2.6.9-52bs #2 SMP Fri Jan 26 13:34:38 CST 2007 > x86_64 x86_64 x86_64 GNU/Linux > > $ gcc -v > --host=x86_64-redhat-linux Thread model: posix gcc version 3.4.5 > > $ /lib64/tls/libc.so.6 > GNU C Library stable release version 2.3.4, by Roland McGrath et al. It is possible to do. Since the system is older, and your needs are older as well, I would suggest following the install instructions for an older LFS: http://www.linuxfromscratch.org/lfs/view/6.3/chapter06/glibc.html chaning --prefix=/usr to --prefix= Once installed you can then add LD_CONFIG_PATH=/lib to the commandline when you run the binary, or the shell environemnt. Hope this points you in the right direction. Best of luck. ae -- http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] LFS 7.1, a problem
Hello, On Sat, Nov 10, 2012 at 02:40:05AM +0100, Jean-Philippe MENGUAL wrote: > Hi, > > As I finished a lfs 7.1 before 7.2 release, I try finishing before > updating. And I have a boot problem. > 1. My tmpfs entry in fstab is good. My dev entry too. > 2. CONFIG_TMPFS=y > 3. I have a /run dir > > 1st thing strange: I don't understand, in the book, "ln > -s /run/lock /var/lock" (chap6.5) which creates a broken symlink ( /run > is empty). Moreover, at boot, I get a message: "cannot find /dev in > fstab" then "can't find /run/var/bootlog". > > Could it be a kernel problem? A bootscript problem? Can I workaround or > fix? > > Thanks, I have a hunch that /run isn't being mounted as tmpfs, which would lead to the "can't find /run/var/bootlog" error. I'd look into your bootscripts and see if anything is out of order. Best of luck. ae -- http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Cannot boot to LFS with seperate boot partition
Hello,
On Mon, Jun 25, 2012 at 11:30:08AM -0500, Eleanore Boyd wrote:
> On 6/25/2012 11:24 AM, Yasser Zamani wrote:
> >*snip*
> >*/root@slax:~# cat /mnt/sda7/grub/grub.cfg/*
> >
> >/# Begin /boot/grub/grub.cfg/
> >/set default=0/
> >/set timeout=5/
> >/*# WHEN I SELECT THIS ENTRY IT TELLS "error: no argument
> >specified" BUT AFTER KEY PRESS ENTERS TO WINDOWS SUCCESSFULLY*/
> >/menuentry "Windows 7 (loader) (on /dev/sda3)" {/
> >/insmod part_msdos/
> >/insmod ntfs/
> >/set root='(hd0,msdos3)'/
> >/search --no-floppy --fs-uuid --set 125c5cdd5c5cbd63/
> >/chainloader +1/
> >/}//**/
> >
> *snip*
Errors like "error: no argument specified" usually point to something
missing. Try:
search --no-floppy --fs-uuid --set=root 125c5cdd5c5cbd63
Notice that '--set' needs the arguement '=root' in this case.
I believe that the lines
set root='(hd0,msdos3)'
and
search --no-floppy --fs-uuid --set=root 125c5cdd5c5cbd63
are redundant, but I am not sure. Hopefully this can be verified by someone
else.
> I see at least one problem: Windows has a problem being on anything
> other than the first partition. I remember seeing somewhere that
> Grub has a way to trick bootmgr to think it's on the first
> partition, but I can't remember the method. Also, I didn't think
> that "msdos#" was used to define a partition that's FAT32 or NTFS.
>
> Elly
If needed to 'trick' Windows into believing it is on the first
partition, add this line right above 'chainloader +1' :
drivemap -s (hd0) ${root}
>> /dev/sda1 1 9 72261 de Dell Utility
>> /dev/sda2 101315104857607 HPFS/NTFS
>> /dev/sda31315 14674 1073098517 HPFS/NTFS
Dell Utility is on the first partition. Probably it came this way when
the computer was bought.
> # WHEN I SELECT THIS ONE IT TELLS "error: file not found" BUT AFTER KEY
> # PRESS ENTERS TO LFS SUCCESSFULLY
> menuentry "GNU/Linux, Linux 3.4.1-lfs-SVN-20120617" {
> insmod ext3
> set root=(hd0,7)
>linux /vmlinuz-3.4.1-lfs-SVN-20120617 root=/dev/sda6 ro
> }
I must admit that I find grub2 still a bit of a mystery. It seems that what
I thought earlier was in error. I was able to reproduce
"file not found" on my system having the kernel from one
partion booting LFS on another and eventually got it working.
>> set root=(hd0,7)
> This line is correct as it tells grub to look for your boot partition.
I now disagree with myself on this one, upon further research:
http://fedoraproject.org/wiki/GRUB_2
> This next line needs a little tweaking:
>> linux /vmlinuz-3.4.1-lfs-SVN-20120617 root=/dev/sda7 ro
Actually, I think it needs a bit more tweaking.
Okay, try this:
menuentry "GNU/Linux, Linux 3.4.1-lfs-SVN-20120617" {
insmod ext3
set root=(hd0,6)
linux (hd0,7)/vmlinuz-3.4.1-lfs-SVN-20120617 root=/dev/sda6 ro
}
Hope this works.
Okay, if I understand things better:
1) set root=(hd0,6) (grub.cfg), root=/dev/sda6 (kernel option) and /dev/sda6 /
(/etc/fstab)
should all 'match'
2) if the kernel is not on the set root=(hdX,X) partition, it can be specified,
eg.
linux (hdX,X)/whever-it-is/linux_kernel
Best of luck!
ae
--
My Blog: http://elian001.wordpress.com
--
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page
Re: [lfs-support] Cannot boot to LFS with seperate boot partition
Hello,
On Thu, Jun 21, 2012 at 09:43:00AM +0430, Yasser Zamani wrote:
>
> Hi there,
>
> When I reboot to LFS with following /etc/fstab and /boot/grub/grub.cfg,
> after selecting the boot menu, it tells me "error: file not found"
This means that grub (the bootloader) hasn't found the linux kernel.
> # Begin /etc/fstab
> # file system mount-point type options dump fsck
> # order
> /dev/sda6 / ext3 defaults1 1
> /dev/sda7 /boot ext3 defaults1 1
> # Begin /boot/grub/grub.cfg
> set default=0
> set timeout=5
> insmod ext3
> set root=(hd0,7)
> menuentry "GNU/Linux, Linux 3.4.1-lfs-SVN-20120617" {
> linux /boot/vmlinuz-3.4.1-lfs-SVN-20120617 root=/dev/sda6 ro
> }
> Then I tried following /boot/grub/grub.cfg instead; the result was getting
> two
> "file not found"s but after pressing key something takes place on screen then
> the pc halts and CAPS LOCK and NUM LOCK's LEDs are flashing!
> What do these LEDs flashing mean?
This means that the kernel was found, started running and
then hung. Usually this happens when the kernel doesn't find the root
partition. It also means that you are a little closer to finding a
solution because this time grub found the linux kernel.
> # Begin /boot/grub/grub.cfg
> set default=0
> set timeout=5
> insmod ext3
> set root=(hd0,7)
> menuentry "GNU/Linux, Linux 3.4.1-lfs-SVN-20120617" {
> linux /vmlinuz-3.4.1-lfs-SVN-20120617 root=/dev/sda7 ro
> }
> Could you please help me to understanding followings:
> Which one takes place at first? Processing and applying
> /etc/fstab or processing /boot/grub/grub.cfg. If second one takes place at
> first so how it
> finds out that where it should look for /boot?
The GRand Unified Bootloader (grub) processes /boot/grub/grub.cfg first.
After the kernel has booted, the system bootscripts read /etc/fstab.
> If above two question's answers can not teach me the solution, so could you
> please help
> me to sort this problem out?
> Thanks in advance!
Let's have a look at these lines from your second try:
> set root=(hd0,7)
This line is correct as it tells grub to look for your boot partition.
This next line needs a little tweaking:
> linux /vmlinuz-3.4.1-lfs-SVN-20120617 root=/dev/sda7 ro
^ ^^^
| |
| |
| This is where I think the error is.
This part is good.
root=/dev/XXX should match the root in /etc/fstab. In your case this would
be /dev/sda6.
Try:
linux /vmlinuz-3.4.1-lfs-SVN-20120617 root=/dev/sda6 ro
Also, you will find editing the /boot/grub/grub.cfg easier with vim.
There is a little bit of a learning curve, so finding an online tutorial
is a good idea.
Best of luck!
ae
--
My Blog: http://elian001.wordpress.com
--
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page
Re: [lfs-support] Package management
Hello, On Sun, Mar 25, 2012 at 10:04:56PM -0500, Eleanore Boyd wrote: > Has anyone thought of this: when you download a package, store it in the > sources folder used for building the system, and organize it to personal > preferences and tastes? Have a look at the hints page for some ideas: http://www.linuxfromscratch.org/hints/read.html Quite a few people have added some form of package management to their LFS. Currently, I'm partial to dpkg. > It seems like a good idea to me Give it a try and see how it goes. > am I giving out the idea on the wrong list? > Elly It may be more appropriate to post this type of chatter on the LFS chat mailing list. Have some fun putting together your LFS system! ae -- My Blog: http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] error: No usable m4 in $PATH or /usr/5bin
Hello, On Mon, Mar 19, 2012 at 03:48:28PM +, gdsi wrote: > Hello. > Few days ago I wrote that did not might mount the LFS-partition > after chroot. Now this ok. But when I perform ./configure > --prefix=/usr --enable-cxx --enable-mpbsd from 6.14. GMP-5.0.4 and > was being issued: > ... > . > checking for suitable m4... configure: error: No usable m4 in $PATH > or /usr/5bin (see config.log for reasons). Double check how your $PATH is set up. /usr/5bin feel like it is a typo - perhaps it should be /usr/bin or /usr/sbin. > But I installed M4 in ch5, I seem. > If I forgot M4 earlier, may do this now? Yes, install M4 and then carry on. Best of luck. ae -- My Blog: http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: 5.31. Texinfo-4.13a
Hello, On Sun, Oct 16, 2011 at 01:30:56PM -0500, knothea...@gmail.com wrote: > Greetings, > I am trying to install LFS 6.8 using an Fedora 15. I am at the 5.31. > Texinfo-4.13a install and it is failing on the make install area. > > Any assistance would be great. > > Thanks. > > Knothead00 > > Check-version.sh > *** > bash, version 4.2.10(1)-release > /bin/sh -> /bin/bash > Binutils: version 2.21.51.0.6-6.fc15 20110118 > bison (GNU Bison) 2.4.3 > /usr/bin/yacc -> /usr/bin/yacc > bzip2, Version 1.0.6, 6-Sept-2010. > Coreutils: 8.10 > diff (GNU diffutils) 2.8.1 > find (GNU findutils) 4.5.9 > GNU Awk 3.1.8 > /usr/bin/awk -> /bin/gawk > gcc (GCC) 4.6.1 20110908 (Red Hat 4.6.1-9) > GNU C Library stable release version 2.14 > grep (GNU grep) 2.9 > gzip 1.4 > Linux version 2.6.40.6-0.fc15.i686 > (mockbu...@x86-17.phx2.fedoraproject.org) (gcc version 4.6.1 20110908 > (Red Hat 4.6.1-9) (GCC) ) #1 SMP Tue Oct 4 00:51:19 UTC 2011 > m4 (GNU M4) 1.4.16 > GNU Make 3.82 > patch 2.6.1 > Perl version='5.12.4'; > GNU sed version 4.2.1 > tar (GNU tar) 1.26 > Texinfo: makeinfo (GNU texinfo) 4.13 > > > make[1]: Leaving directory `/mnt/lfs/sources/texinfo-4.13/install-info' > Making install in info > make[1]: Entering directory `/mnt/lfs/sources/texinfo-4.13/info' > make install-am > make[2]: Entering directory `/mnt/lfs/sources/texinfo-4.13/info' > gcc -g -O2 -o ginfo dir.o display.o dribble.o echo-area.o filesys.o > footnotes.o gc.o indices.o info-utils.o info.o infodoc.o infomap.o > m-x.o man.o nodemenu.o nodes.o search.o session.o signals.o terminal.o > tilde.o variables.o window.o doc.o ../lib/libtxi.a > ../gnulib/lib/libgnu.a > terminal.o: In function `terminal_goto_xy': > /mnt/lfs/sources/texinfo-4.13/info/terminal.c:235: undefined reference > to `tgoto' It looks like you are missing ncurses from 5.14. Try (re)installing it and then come back to texinfo. Best of luck, ae -- My Blog: http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Need Help - boot fails
Hello, On Sun, Sep 25, 2011 at 01:56:28PM -0400, scrat wrote: > I have completed the LFS-6.8 book to the chapter on rebooting to the new > LFS system. > > The kernel appears to load but fails at the point of: > > [ 1.433897] Freeing unused kernel memory: 400k freed > It stops there and a hard reset is required. > >>> You need to recompile your kernel. You need to work on your > >>> kernel .config and develop a working kernel that can boot your computer > >>> _without_ modules. > >> Ok I will look at the kernel. > >> I used a config file that had booted before on this machine > > One thing you can probably do is use Shift-PgUp to look at earlier > > screens to see if there are any error messages that might have scrolled off. > > One suspicious thing is the [ 1.433897] number. My system is not > > particularly slow, but it has > > [3.081309] Freeing unused kernel memory: 612k freed > > The next things that follow on my system are the initialization of the > > usb HW. I agree with Andy that the kernel config look suspicious. > > Remember to not use modules for the first time through. > > -- Bruce > It is a laptop dual core AMD 1600MHz > I tried a kernel without modules and it failed at the same place. > What does the > [ 1.433897] mean? If you have the framebuffer enabled, try turning it off and see what happens. Device Drivers ---> Graphics support ---> < > Support for frame buffer devices Sometimes if there is a mismatch between the driver and video card, the kernel will boot, but the screen will remain black. Best of luck, ae -- My Blog: http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: coreutils fails
Hi Robert, On Fri, Aug 05, 2011 at 11:41:15AM -0500, robert wrote: > Any suggestions? > I'm working at 6.22. Coreutils-8.10 > > == > 1 of 15 tests failed >> snip << > make[1]: Leaving directory `/sources/coreutils-8.10/tests' > make: *** [check-root] Error 2 > > selected log info follows: > > FAIL: cp/sparse-fiemap (exit: 1) > > > df: Warning: cannot read table of mounted file systems > FilesystemType 1K-blocks Used Available Use% Mounted on > --10768488 1741780 8479720 18% / >> snip << I was able to reproduce this failure by having an empty /etc/mtab file, which df looks for. Updating the mtab file with mount -f / and then running the testsuite again took care of this error. I think you can safely carry on with the next instructions in the book without any concern for this error. Relax, don't worry. Have a homebrew. (Or whatever liquid is best for you!) Best of luck, ae -- My Blog: http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: more glibc problems
On Tue, Jul 26, 2011 at 09:44:13AM -0400, Bill Cunningham wrote: > Andrew Elian wrote: > > On Mon, Jul 25, 2011 at 10:19:04PM -0400, Bill Cunningham wrote: > >> I received this error message when my compilation finally broke. > >> I was doing well for a long time. > >> > >> > >> /mnt/lfs/g/nscd/nscd.o: In function `nscd_open_socket': > >> /mnt/lfs/glibc-2.13/nscd/nscd.c:442: undefined reference to > >> `__stack_chk_guard' > >> > >> Bill > > > > I ran into this odd error myself some time ago when I (just for fun) > > built LFS 6.8 using a much older LFS. > > > >> From http://lkml.indiana.edu/hypermail/linux/kernel/0902.1/01077.html > > "It seems that even with the same gcc versions, gcc built against > > libc w/o TLS support generates __stack_chk_guard, so if you mix the > > two flavors, the has-stack-protector check can be compiled on > > machines w/TLS while some other files end up being built on > > machines w/o TLS support thus circumventing the support check." > > > > I think there are two options: adding libc_cv_ssp='no' to configparms > > or > > adding -lssp for the linker. Having gone with option two I found I > > also > > needed to then supply libssp (from /tools) on the host system. Both > > of > > those seem to be hacks. > > Now does this have anything with not building libssp back either > when 4.5.2 was being built? I built it with gcc-4.6.0 too if that means > anything. > > It might be better to rebuild gcc (in Chapter 5) with > > -fno-stack-protection, however, I've not tried that myself. I just double checked the build scripts I used. I think you are doing Chapter 5.7, whereas I was looking at Chapter 5.10 gcc pass 2. > Remember libssp was disabled earlier. Was it built? If not how can > you link statically to it? Actually, I created a soft link to libssp and hacked a Makefile to include -lssp if there was an error. Really, this is a bad hack and I do not recommend it. I am not sure of the reasons for the undefined reference to `__stack_chk_guard'. I have a hunch it is related to the glibc/gcc build process. This doesn't seem to be a problem when building on a newer linux host. It seems to me you will have two major difficulties with your current LFS build: 1) your build host is old and 2) using an amd64 with a 32 bit linux. Not in itself a bad thing, but as you have discovered, some packages will do strange things and fail to compile. I will strongly recommend the x86_64-Pure64 Cross Linux From Scratch book as it deals with building on a host which is very different than the target. Give development version (as it is the most up to date) a try and see how it goes. Keep at it, it will make sense eventually. ae -- My Blog: http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: more glibc problems
On Mon, Jul 25, 2011 at 10:19:04PM -0400, Bill Cunningham wrote: > I received this error message when my compilation finally broke. I was > doing well for a long time. > > > /mnt/lfs/g/nscd/nscd.o: In function `nscd_open_socket': > /mnt/lfs/glibc-2.13/nscd/nscd.c:442: undefined reference to > `__stack_chk_guard' > > Bill I ran into this odd error myself some time ago when I (just for fun) built LFS 6.8 using a much older LFS. >From http://lkml.indiana.edu/hypermail/linux/kernel/0902.1/01077.html "It seems that even with the same gcc versions, gcc built against libc w/o TLS support generates __stack_chk_guard, so if you mix the two flavors, the has-stack-protector check can be compiled on machines w/TLS while some other files end up being built on machines w/o TLS support thus circumventing the support check." I think there are two options: adding libc_cv_ssp='no' to configparms or adding -lssp for the linker. Having gone with option two I found I also needed to then supply libssp (from /tools) on the host system. Both of those seem to be hacks. It might be better to rebuild gcc (in Chapter 5) with -fno-stack-protection, however, I've not tried that myself. Best of luck, ae -- My Blog: http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: error log
Hi Bill, On Sat, Jul 23, 2011 at 10:43:51PM -0400, Bill Cunningham wrote: > nptl/sysdeps/unix/sysv/linux/bits/local_lim.h:39:26: fatal error: > linux/limits.h: No such file or directory > compilation terminated. ...snip... > make: *** [all] Error 2 > I'm not quite sure what is wrong with the above but I am trying to compile > glibc-2.14 with the development code. One I don't understand and deviated > from the online book a little was concerning copying the kernel headers. > With the 2.4 version I am familiar with you simply created in your > usr/include directory two directory called linux and asm. Linux kernel > headers when into "linux" and your arch's asm code went into "asm". There is > now called a directory "asm-generic" It has been completing changed from the > 2.4 kernel. This is my first workings with 2.6. How do I copy those kernel > headers over for a x86_64 or a i386. My old linux is a 32 bit version but I > would like to build a 64 bit lfs because I have an Athlon 3200+. 64 bit. > > Bill With the more recent 2.6 kernel, the installation of the header files has changed. Quoted from LFS-6.8, 5.6.1. Installation of Linux API Headers: make mrproper make headers_check make INSTALL_HDR_PATH=dest headers_install cp -rv dest/include/* /tools/include Also, it is easier to build a 64 bit LFS on a 64 bit host. If that is not an option (or you are up to the challenge) you could cross compile the 64 bit LFS on a 32 bit host. It is a bit tricky and I would suggest looking into Cross Linux From Scratch. Best of luck, ae -- My Blog: http://elian001.wordpress.com -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Testing failure: GCC-4.5.2
On Wed, Jun 22, 2011 at 07:42:40AM -0500, robert wrote:
> I've reposted this ... previous post possibly lost with earlier similarly
> titled
> thread.
>
Hi Robert,
> Original Message
> Subject: 6.16. GCC-4.5.2
> Date: Tue, 21 Jun 2011 07:36:04 -0500
> From: robert
> To: LFS Support List
>
> This is disconcerting. While results seem to be "acceptable", the Error 1,
> Error
> 2 messages I don't like. Any comments? Should I proceed? fall back?
I snipped out some of the details...
> === libmudflap Summary ===
>
> # of expected passes 1414
> # of unexpected failures 13
> === libgomp Summary ===
>
> # of expected passes 1029
>
> === g++ Summary ===
>
> # of expected passes 22526
> # of expected failures149
> # of unsupported tests115
>
> === gcc tests ===
>
>
> Running target unix
>
> === gcc Summary ===
>
> # of expected passes 61949
> # of expected failures166
> # of unsupported tests542
> /sources/gcc-build/gcc/xgcc version 4.5.2 (GCC)
>
>
> === libgomp Summary ===
>
> # of expected passes 1029
>
> === libmudflap Summary ===
>
> # of expected passes 1414
> # of unexpected failures 13
> === libstdc++ tests ===
>
>
> Running target unix
>
> === libstdc++ Summary ===
>
> # of expected passes 7075
> # of expected failures95
> # of unsupported tests339
For myself, I would proceed forward, especially if the number of
failures is similar to the results others have gotten. I don't think
I've ever seen the gcc testsuite return zero failures.
>
>[Requesting program interpreter: /lib/ld-linux.so.2]
>
> /usr/lib/gcc/i686-pc-linux-gnu/4.5.2/../../../crt1.o succeeded
> /usr/lib/gcc/i686-pc-linux-gnu/4.5.2/../../../crti.o succeeded
> /usr/lib/gcc/i686-pc-linux-gnu/4.5.2/../../../crtn.o succeeded
>
> #include <...> search starts here:
> /usr/local/include
> /usr/lib/gcc/i686-pc-linux-gnu/4.5.2/include
> /usr/lib/gcc/i686-pc-linux-gnu/4.5.2/include-fixed
> /usr/include
>
> SEARCH_DIR("/usr/i686-pc-linux-gnu/lib")
> SEARCH_DIR("/usr/local/lib")
> SEARCH_DIR("/lib")
> SEARCH_DIR("/usr/lib");
>
> attempt to open /lib/libc.so.6 succeeded
> found ld-linux.so.2 at /lib/ld-linux.so.2
Looks like your toolchain is set up properly. I would suggest that you
carry on.
Best of luck,
ae
--
My Blog: http://elian001.wordpress.com
--
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page
Re: Shadow
Hello, On Sat, Jun 11, 2011 at 12:08:14AM +, Nick Amor wrote: > Hi, I'm working on LFS 6.8 and I cannot download the Shadow package. > When I try to use wget it returns the message "failed: connection refused". > I tried going to website to download directly and it said I didn't have > access on the download page. I have searched Google and still can't find > a download location. Please help. It seems that the download links have changed. In the meantime, you might find a copy on one of the LFS mirrors - I found it here: ftp://ftp.lfs-matrix.net/pub/lfs/lfs-packages/conglomeration/shadow In times past I've found it useful to do a linux google search with the package name and +Slackware. http://www.google.ca/linux I've also found gentoo to be useful as well - their mirrors also have tarballs and reading the ebuilds can be quite instructive. Best of luck, ae -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: can't boot a new kernel
On Wed, Mar 09, 2011 at 12:32:00PM +0800, ztq小名 wrote: > Hi. > I'm trying to install kernel 2.6.24.7 as a second kernel.I use these > commands: > - > make > > make modules_install > > cp -v arch/i386/boot/bzImage /boot/lfskernel-2.6.24.7 > > cp -v System.map /boot/System.map-2.6.24.7 > > cp -v .config /boot/config-2.6.24.7 > > install -d /usr/share/doc/linux-2.6.24.7 > > cp -r Documentation/* /usr/share/doc/linux-2.6.24.7 > --- > Then it's stuck while booting.There are no error reports in the picture. > > Can someone help me? > > Thanks Hello, Try removing vga=788 when you boot the kernel. If the kernel is booting past that point it should display some more messages. ae -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Need to undo toolchain adjustment
On Fri, Feb 11, 2011 at 10:48:27PM -0600, Mitch Embry wrote: > Hello, > > This is my first LFS build and all was going well until I went and > changed a command. I am building this system in virtualbox using the > latest LFS Live-CD, all the latest packages from lfs.org, and LFS-BOOK-6.7 > > I had completed the steps up to 5.9 in the book where I was starting > pass 2 of Binutils when I realized what I had done. Two steps back when > I was compiling Glibc I erroneously changed a command thinking I was > doing the right thing (against my better judgment). The command I edited > was the following: > > #begin command > > ../glibc-2.13/configure --prefix=/tools \ > --host=$LFS_TGT --build=$(../glibc-2.13/scripts/config.guess) \ > --disable-profile --enable-add-ons \ > --enable-kernel=2.6.22.5 --with-headers=/tools/include \ > libc_cv_forced_unwind=yes libc_cv_c_cleanup=yes > > #end command > > While typing this command I mistakenly assumed I needed to change the > '--enable-kernel=2.6.22.5' line to reflect the latest kernel that I had > downloaded from lfs.org not realizing that this line was in fact > referring to the host system. However, everything compiled and installed > correctly and I moved on to 5.8 Adjusting the Toolchain. Again all of > this completed without error, including the sanity-check. > > It was not until I started to compile Pass 2 of Binutils and the > compiler threw a "FATAL: Kernel too old" error that I realized what I > had done. Now, since I already have adjusted the toolchain and GCC's > dynamic linker has changed when I try to rebuild Glibc, make will not > complete (it hangs) and as a result I am stuck. > > What should I do to correct this? Can I undo the Toolchain adjustment so > I can recompile Glibc or do I have to start from square one again? Any > advice is greatly appreciated. Either way I've learned my lesson about > altering commands unless expressly directed to do so. > > Thanks. Hello, It seems that I've made this mistake a few times myself. You may consider updating the kernel on your host system (which looks like the liveCD). This is what I've done before. Also, it's my preferred solution for this mistake. Works out fine in the end. Or cleaning out the file system and starting over. Very time consuming and not much fun! (Think I've gone this route too, on occasion). Or by undoing the changes made to the gcc spec file and reverting the toolchain back and then rebuilding glibc. Time consuming and a fun way to learn, but ... it's been a long time since I've done such the details seem fuzzy... AE -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
