Re: [blfs-support] /dev/fb0 not being created on boot
On Fri, Nov 22, 2013 at 07:05:14PM +, Richard Melville wrote: Ken, I'm using vesafb on a web server with no Xorg, and I just use the console. I realise that my kernel was quite old but as I like to check every configuration option (often because of new hardware) it takes a long time to configure a new kernel and becomes incredibly boring towards the end :-( Therefore, I usually limit my upgrades to about one per year, or when I can be bothered. OK, but in that case I suggest that you go for early stable kernels (.1, .2, etc). You said you had been using 3.7-rc8 - that probably turned out to be totally good for your uses, but sometimes even .0 releases still have issues in a few places. Also, the config you use in a late -rc will normally not provide any extra questions for released / stable kernels in the same series if you use 'make oldconfig'. Interestingly, as 3.12.1 is now available I downloaded the source, copied the config across, and ran 'make oldconfig'. As you predicted there were no extra questions to answer so I proceeded to compile. Strangely, it exited with an error when it reached a *keyboard* configuration. I know I should have investigated further, but by this time I was so frustrated that I returned to 3.12 as I knew it was good. I can't remember the last time I had a kernel fail at the compile stage. I'll investigate further when I have more time. Bruce, my framebuffer config was much like yours but with one exception: I had CONFIG_X86_SYSFB=y. This was stopping vesafb loading and stopping /dev/fb0 being created. I've removed that option, reconfigured, and now it all works as expected. Interesting. That option works ok on intel (I'm running with it at the moment), I'll try to remember that for the future - my server also runs with CONFIG_FB_VESA and vga=792 (it's a radeon RS780, when I got it I had no experience with modern ATI hardware and totally failed to get the radeon framebuffer to work - 80x25 is too restrictive for me!) - for the moment I'm running 3.10 (LTS) there. Hmm -- I'm using an Intel board. Regarding vga=792, that still works for me. If I substitute video=1024x768 the command is ignored an I get a large, ugly font. I'm currently using grub-2.0, so I can't understand what the problem is likely to be. Any ideas? Richard Stick with vga=792 since it still works ? Any idea how large the font is, or how many pixels in the screen size when you boot with video=1024x768 ? It's the correct font that I've set but without the resolution setting; It looks ugly at that size on the screen. When the box boots, you get the font from its bios. But when the LFS bootscripts start to run it ought to switch to your specified font (provided the setup supports it, e.g. my own 12x22 is only supported on framebuffers and I've not tried it with Vesa, my screen there is physically 1024x768 so I use an 8x16 font on that). So, basically work out what sort of console font *size* will suit you, then try setfont in a spare tty and see if any of the available fonts look ok _and_ provide the character coverage you require. My own LatGrkCyr are intended for white/pale text on a black background and I'm told that at least one of them looks awful with dark text on a pale background : there is a balance between getting adequate coverage, shapes which do not offend our own particular sensibilities, and being able to distinguish the various accents and diacritical markings - for some uses, noting that a glyph is e.g. letter a with accent may be good enough, but others may wish to be able to see at a glance what sort of accent is present, particularly in unfamiliar languages. ?en I'm using LatGrkCyr and I really like it. On my console screen at vga=792 it looks good. So everything is now working but some questions remain unanswered. Thanks for your help Ken. Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: [blfs-support] /dev/fb0 not being created on boot
Ken, I'm using vesafb on a web server with no Xorg, and I just use the console. I realise that my kernel was quite old but as I like to check every configuration option (often because of new hardware) it takes a long time to configure a new kernel and becomes incredibly boring towards the end :-( Therefore, I usually limit my upgrades to about one per year, or when I can be bothered. Bruce, my framebuffer config was much like yours but with one exception: I had CONFIG_X86_SYSFB=y. This was stopping vesafb loading and stopping /dev/fb0 being created. I've removed that option, reconfigured, and now it all works as expected. Regarding vga=792, that still works for me. If I substitute video=1024x768 the command is ignored an I get a large, ugly font. I'm currently using grub-2.0, so I can't understand what the problem is likely to be. Any ideas? Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] BLFS Version 7.4 is released
Excellent news :-) However, may I just point out that there's a broken libungif link on the Emacs-24.3 page. Maybe it should point here: http://directory.fsf.org/wiki/Libungif Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Booting LFS Error Kernel Panic
On Thu, 2012-12-20 at 14:53 +, Richard Melville wrote: I think that was understood; when they said that it was stupid it was surely meant that there could be some confusion in the use of similar terms. Possibly, though if they'd understood it, you'd think they'd have mentioned the by-partuuid directory, instead of claiming that the gdisk tool was the only way of working out the partition UUID. Simon. It seems a little churlish to pick holes in what is essentially a good article, and, indeed, one that supplied the answer to a question on this list. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Booting LFS Error Kernel Panic
Now it would be nice for it to work using UUIDs so the booting can be independent of host system. You need to use an initrd of that. See BLFS. -- Bruce Would't using GPT instead of MBR be a viable alternative? Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] keyboard-1.15.3 errors on backspace with uk keymap
I've got a few files at http://www.linuxfromscratch.org/~ken/ in the keyboard-items and fonts directories - note that LatGrkCyr-8x16 is a 512-ish character font and ships in kbd. It comes from the sigma fonts there which are very much roll your own but do allow a 256 character font if that is what you need. ?en Thanks for the help and the link Ken; I'll have a play when I have more time. I'm still using vga=792 on the grub kernel boot line to get the right (for me) sized screen fonts. Is that still acceptable or is there an alternative? Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] keyboard-1.15.3 errors on backspace with uk keymap
On Sat, Dec 15, 2012 at 01:00:38PM +, Richard Melville wrote: When I use backspace in the terminal/console and then re-type I get white blocking. I'm fairly sure that I installed the patches when I built the keyboard package. Any advice? It's really annoying. Richard I suppose white blocks might be a result of a console font which cannot display the glyph it was asked for. In a unicode font, that situation ought to show an inverse question mark (black-on-white for normally white-on-black text), but many fonts cannot do that. However, that doesn't explain why the backspace isn't effective. The backspace patch only changes this for a few keymaps which still gave Backspace instead of Delete - the last time I looked (some time before 1.15.3, so something might have slipped in), all of the other keymaps shipped in the package already did this. What do you have in /etc/sysconfig/console ? ?en Ken -- thanks for the reply. I changed the font setting over the weekend and now it seems to be OK. The problem was the following: typing worked OK, and if I made a typo and wanted to delete with the backspace key, deletion worked OK, however, when I began to type again that's when I saw the white blocking. I know very little about fonts, keymaps, unicode, etc. What I would like to do is set up a unicode environment but I'm not sure how to go about it, although I'll probably only be using an accented e, an umlaut/diaeresis, and a euro symbol in addition to a uk keymap. The following are the console parameters of /etc/sysconfig/rc.site (I'm not using /etc/sysconfig/console) and I've left my original font setting in, but commented out:- # Console parameters UNICODE=0 KEYMAP=uk #KEYMAP_CORRECTIONS=euro2 #FONT=default8x16 FONT=lat1-16 -m 8859-1 #LEGACY_CHARSET= Thanks for your help. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
[lfs-support] keyboard-1.15.3 errors on backspace with uk keymap
When I use backspace in the terminal/console and then re-type I get white blocking. I'm fairly sure that I installed the patches when I built the keyboard package. Any advice? It's really annoying. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Strsnge udev-181 behaviour
Richard Melville wrote:
I have one Ethernet adapter (Intel 82574L Gigabit) but udev has found two
complete with MAC addresses. The phantom version is installed on eth0
and
the real version is installed on eth1. I've searched the system for the
phantom MAC address but I cannot find any reference to it. Has anybody
else experienced this behaviour? The only other case I can find is
somebody on a Raspberry Pi list who experienced the same phantom creation
with udev and his wireless adapter.
Sometimes a Gigabit adapter will have tho connections. Once the chip is
designed, it's probably just as easy to produce that as a separate chip
with one interface.
Thanks Bruce, I suppose that's the logical explanation as, AFAIK, only the
manufacturer can embed a MAC address in the hardware.
What are the contents of /etc/udev/rules.d/70-persistent-net.rules?
-- Bruce
Contents of /etc/udev/rules.d/70-persistent-net.rules:-
# This file was automatically generated by the /lib/udev/write_net_rules
# program, run by the persistent-net-generator.rules rules file.
#
# You can modify it, as long as you keep each rule on a single
# line, and change only the value of the NAME= key.
# PCI device 0x8086:/sys/devices/pci:00/:00:1c.0/:01:00.0
(e1000e)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:22:4d:7c:be:b7, ATTR{dev_id}==0x0, ATTR{type}==1,
KERNEL==eth*, NAME=eth0
# PCI device 0x8086:/sys/devices/pci:00/:00:1c.0/:01:00.0
(e1000e)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*,
ATTR{address}==00:22:4d:9a:f4:89, ATTR{dev_id}==0x0, ATTR{type}==1,
KERNEL==eth*, NAME=eth1
Richard
--
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page
Re: [lfs-support] vi for chapter 6
Or just create your scripts with cat blah EOF Then if you have errors use sed or perl to fix them! Sincerely, William Harrington A little off-topic but I've pondered this for a long time: in the LFS book why is EOF always in quotes; I've found EOF without quotes to work just fine. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Test failures in automake-1.12.3 : sorted!
On Sun, Sep 30, 2012 at 12:32:22AM +0100, Ken Moffat wrote: and I'll open a ticket for this possible fix to t/python-missing.sh. Normally, I'd just upload the patch, but I'd prefer to get confirmation that it fixes the problem. More on -dev when I've created a ticket. Forget that - the testsuite fix was applied upstream in August, and 1.12.4 came out in September (#3185) so fixing the 1.12.3 tests is a waste of time. However, I do think that perhaps we shouldn't discourage people from running the automake testsuite on modern processors : sure, with -j1 it takes an excessively long time, but with modern SMP it runs much quicker if you use -j4. ?en Hi Ken I'm not sure I understand what's going on here. When I tested automake-1.12.2 *without* python installed I received no such error:- == Testsuite summary for GNU Automake 1.12.2 == # TOTAL: 2852 # PASS: 2648 # SKIP: 164 # XFAIL: 40 # FAIL: 0 # XPASS: 0 # ERROR: 0 == Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] GCC-4.7.1-Pass 2 MPC configure fails
On Wed, Aug 08, 2012 at 07:01:02PM +0100, Richard Melville wrote: I realise that I'm building the dev edition, but my host is Linux Mint Cinnamon 64 bit and the host requirements appeared to fit better. Also it looked as though the dev edition was at a reasonably stable stage. I'm building a 64 bit edition on a 64 bit host (OS and hardware). The failure is:- checking for MPFR... no configure: error: libmpfr not found or uses a different ABI (including static vs shared). make[1]: *** [configure-mpc] Error 1 make[1]: Leaving directory `/mnt/lfs/sources/gcc-build' make: *** [all] Error 2 Everything has built fine up to this stage and the sanity checks were OK. MPFR and GMP have compiled OK with the libraries installed in .libs. I've even checked to make sure that the MPFR libraries were 64 bit, and now I've run out of ideas. I'd be really grateful for any help. I've tried rebuilding GCC four times now with the same result. Richard Richard - since this is still bugging you, I've come back to your original post. I notice one thing which nobody has mentioned: *when* you get this error, look at the *appropriate* config.log file [ gcc, like binutils, runs configure in multiple directories]. Even if this error happens/happened when running 'make', the error was within one of the configure scripts - gcc builds everything several times, and each time it configures the directories within it. Whenever configure fails (I usually point this out for cannot create executables messages), the key to understanding the problem is to find the appropriate config.log file, open that up in 'view' (or 'less'), search for 'a different ABI', and then look for the error messages in the lines before that. Probably, an error from gcc or ld. Once you have the error message, there are two possibilities: 1. it will indicate an error you made, and perhaps be blindingly obvious (I've had that when I was building for multiple archs and accidentally fell through to passing some ppc-only options in my CFLAGS :) - if so, please give the list a brief summary of what went wrong so that the next person who eventually does that can fix it. or, more likely: 2. Something new, which needs to be addressed. The ABI in the message reminds me of a past problem with gmp where, if CFLAGS were set, a processor capable of running 64-bit code would default to building 64-bit even though the rest of hte build was 32-bit. But, the variability in your results suggests this is not something easy like that. Bruce and Ken -- thanks for the replies and apologies for my tardy response. The fact is that I finished the build a while ago with this problem being pretty much the only real issue. I deleted all the packages and files so I don't have anything to work with in terms of tracking down exactly what happened. I've tried rebuilding GCC a couple of times since but I've been unable to replicate the failure. I'm fairly certain that it was a bug of some sort (too many others have reported the same issue) but now I have no proof. I realise that without the logs of the failed build it's pointless discussing the issue, unless. of course, the next person to experience the problem is willing to share their log files. Anyway, thanks again for your help. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] GCC-4.7.1-Pass 2 MPC configure fails
Richard Melville wrote: Finally -- some recognition that there is a potential problem with this build. I disagree that it works okay for x86 and x86_64 because I reported here that the MPC configure error appeared randomly on my x86_64 build, as, indeed, others have. I've already suggested building outside the tree in chapter 5, as we do in chapter 6. I still believe that there is some sort of race condition happening by building the three packages in the tree at the same time, but I've been too busy of late to test for that. Have you seen the problem with 'make -j1' Yes I probably build lfs more than anyone else and have never seen this issue. -- Bruce I originally tried with make -j2, but when it failed I set make -j1 having first created a newly untarred copy; It still failed, and failed a further couple of times before finally building, apparently at random, that is, with me doing nothing different from the other occasions with make -j1 set. It's weird. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] GCC-4.7.1-Pass 2 MPC configure fails
William Harrington wrote: Sometimes you can't build mpfr mpc and gmp within the gcc source tree for some targets. We found that out in CLFS. That's we we don't build gmp mpc and mpfr within the tree. Works okay for x86 and x86_64, however, when you start building for other targets, it becomes hairy. You may want to try to build gmp mpfr and mpc separately. I've been mostly quietly keeping up with this question on several lists: is it better to build those programs in the source tree or separately? There's a discussion somewhere (on a gcc list?) that comes down strongly on the side of building them in the source tree. It even asks why anyone would want to do it differently. Yet the programs contain instructions for tuning, which requires them to be built separately. During the 1 1/2 years I've been playing around with compiling gcc and all of the LFS programs, I've experimented with building separately many times. No problem building them, but gcc usually fails to find at least one of them. Now, I'm a real newbie when it comes to all this, but if anyone knows why gcc can't seem to find the programs, I'd sure like to know. I'm trying to understand all the ins and outs of everything covered by LFS and a lot more besides. Alan Finally -- some recognition that there is a potential problem with this build. I disagree that it works okay for x86 and x86_64 because I reported here that the MPC configure error appeared randomly on my x86_64 build, as, indeed, others have. I've already suggested building outside the tree in chapter 5, as we do in chapter 6. I still believe that there is some sort of race condition happening by building the three packages in the tree at the same time, but I've been too busy of late to test for that. Richard -- 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 7.2 GCC pass 1
On 09/03/2012 03:53 PM, Israel Silberg wrote: checking for MPFR... no configure: error: libmpfr not found or uses a different ABI (including static vs shared). make[2]: *** [configure-stage1-mpc] Error 1 make[2]: Leaving directory `/mnt/lfs/sources/gcc-build' make[1]: *** [stage1-bubble] Error 2 make[1]: Leaving directory `/mnt/lfs/sources/gcc-build' make: *** [all] Error 2 For what it's worth, I ran into the same problem but found that it was just a typographical error on the last line of the configure command, where --with-mpfr-lib is defined. Since it's the last line you may have truncated it during the cut/paste or something. One thing is just to try find /mnt/lfs/sources/gcc-build -name 'libmpfr*' -print and see if that gives you anything. If so, check to see if the path is what you have as the value for --with-mpfr-lib. Another thing you can do is look at /mnt/lfs/sources/gcc-build/mpc/config.log. Search for libmpfr not found and if you look above that a page or two you'll see the gcc command that's testing for libmpfr. It's trying to compile a program called conftest. See if the paths given in the -L directives there match what you gave for --with-mpfr-lib. You can even extract the code for conftest.c (it's down below the error message if I remember right) and try to build it yourself in /mnt/lfs/sources/gcc-build/mpc/ using the gcc command line from config.log. Somewhere in the process you should see something that gives you a clue! Tim Thanks Tim, those tips will be really useful for anybody who encounters that problem again. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Symlinking /etc/mtab to /proc/mounts
Richard Melville wrote: I've noticed that this was recommended way back on LFS 3.3 but now seems to have been dropped. As all the distros appear to have caught up with LFS by having this symlink what are the current views here on creating it? I've noticed some discussion on the dev list in January but it seems to have been inconclusive. The last I looked at it, doing the symlink make the output of the mount command break some scripts. For instance, instead of having /dev/sda12 on / type ext3 (rw) it would give /dev/root / ext3 rw,... 0 0 /dev/root does not provide any useful information, especially since /dev/root does not exist in /dev. /dev/sda12 does tell me which partition is mounted as root. -- Bruce Thanks Bruce, I'll check that out. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
[lfs-support] Symlinking /etc/mtab to /proc/mounts
I've noticed that this was recommended way back on LFS 3.3 but now seems to have been dropped. As all the distros appear to have caught up with LFS by having this symlink what are the current views here on creating it? I've noticed some discussion on the dev list in January but it seems to have been inconclusive. Richard -- 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 7.2 GCC pass 1
On Tue, Sep 4, 2012 at 4:23 AM, Richard Melville richard.melvill...@googlemail.com wrote: I extracted all of these packages from within the GCC-4.7.1 folder snip I'd still be interested to know why we build GMP, MPC, and MPFR inside GCC except on the final build where they are built separately. Richard During pass 1, GCC requires the GMP, MPC and MPFR libraries, but we don't want GCC to get these libraries from the host. GCC searches for the libraries either via the regular search paths *or* inside it's own source tree. We install a temporary copy inside the GCC source tree to take advantage of this, and thus allow GCC to not be contaminated with host libraries During pass 2, we are in a protected chroot environment, so we are no longer concerned about the host. So other programs in pass 2 can take advantage of the GMP, MPC and MPFR libraries later in the build, we install them before GCC instead of in the source tree. -- -- - Steve Crosby Thanks for the reply Steve ( and Eleanor earlier). Picking up on what Bruce said about the possibility of race conditions relating to building GCC with MAKEFLAGS set to -j 1, I'm wondering if there may be a race condition affecting the GCC build with GMP, MPC, and MPFR building inside the GCC directory at the same time. I have no proof for this; it's just a hypothesis, but I was wondering what others may think. There is *definitely* a problem where the GCC build sometimes fails at the same point each time (checking for MPFR), and then builds OK on a random attempt. I'm not aware of the problem ever occurring on the final build of GCC where GMP, MPC, and MPFR are built outside the GCC directory. Richard -- 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 7.2 GCC pass 1
On 2012-09-05 10:43, Richard Melville wrote: Thanks for the reply Steve ( and Eleanor earlier). ?Picking up on what Bruce said about the possibility of race conditions relating to building GCC with MAKEFLAGS set to -j 1, Im wondering if there may be a race condition affecting the GCC build with GMP, MPC, and MPFR building inside the GCC directory at the same time. ?I have no proof for this; its just a hypothesis, but I was wondering what others may think. ?There is *definitely* a problem where the GCC build sometimes fails at the same point each time (checking for MPFR), and then builds OK on a random attempt. ?Im not aware of the problem ever occurring on the final build of GCC where GMP, MPC, and MPFR are built outside the GCC directory. Richard I've been bashing away at building LFS for a VERY long time, and done many many builds of 7.2, this problem has not hit me once, Well aren't you the lucky one. If you took the trouble to look back over the mailing list you would see that a number of people have experienced the error. I would suggest you stop building GCC with MAKEFLAGS set to -j 1, as was suggested. -- Jasmine Iwanek If you spent less time hectoring people and more time reading the posts you would know that I'm quite aware of the issues surrounding the setting of MAKEFLAGS. What is your problem? Richard -- 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 7.2 GCC pass 1
On 2012-09-05 15:34, Baho Utot wrote: On 09/05/2012 09:55 AM, Jasmine Iwanek wrote: Leaping before looking is what I do well and it has taught me a great deal. Following a path by others may be a very good guide, but to truly learn requires ones to deviate from the beaten path and strike out on your own. How else can you create a truly giant mess in which to learn from? Like taking LFS and adding pacman packager. By scripting your builds you learn a great deal about linux and admin. One also has the opportunity to learn some debugging skills. Scripted builds also give one repeatability once they are working. I have scripted my LFS builds and incorporating the pacman package manager. I started with 6.8 and I am currently completing 7.2. I did so that I can confirm that my scripts produce a proper build, i.e. it was tested over the four builds which gave me the opportunity to weed out non apparent errors. I then took those same packages produced by the build and installed them onto 5 other machines so I could check to see if the build was generic for the i686 and x86_64 platforms. I now have a solid platform in which to create a distribution system ( as well building BLFS ) as for the computers under my care. I have learned many things. I still think that helping others even if they have failed to follow the book is a worthy goal as it shows where the book my be improved. Who knows by some not following the book new things are learned? Helping others is always good. Oh, I agree with you fully, don't get me wrong, but people should be starting at the start, not the end. -- Jasmine Iwanek What is that supposed to mean? Really, if you have nothing useful to say then don't say anything. Thanks for your positive post Baho. Richard -- 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 7.2 GCC pass 1
Walter Webb wrote: I just joined this list and can't respond properly. I had a different file not found than Israel Silberg. I unset MAKEFLAGS and retried, and it worked. I'm glad you got it to build, but that's the conclusion of a simple empiricist. It's like me saying that because I went to the kitchen and made a cup of tea before it built successfully, then it must have been the tea. I also unset MAKEFLAGS from -j 2 and it *didn't* work for me. Good point. Using -j 1 can cause problems in some packages. It can cause race conditions that sometimes cause a failure. I suppose we can put a warning about this in the gcc sections, but we'd need it in three places. -- Bruce There's already a warning near the beginning of the book. Richard -- 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 7.2 GCC pass 1
I extracted all of these packages from within the GCC-4.7.1 folder and the configure and make are from gcc-build Here is the output for ls -lah of gcc-4.7.1 lfs@kitt-Lenovo-Product:/mnt/lfs/sources/gcc-4.7.1$ ls -lah total 11M drwxr-xr-x 33 lfs lfs 4.0K Sep 3 09:25 . drwxrwxrwt 4 lfs lfs 4.0K Sep 3 09:27 .. -rw-r--r-- 1 lfs lfs 38K Jul 4 2003 ABOUT-NLS -rw-r--r-- 1 lfs lfs 18K Jul 14 2005 COPYING -rw-r--r-- 1 lfs lfs 26K Jul 14 2005 COPYING.LIB -rw-r--r-- 1 lfs lfs 3.3K Apr 9 2009 COPYING.RUNTIME -rw-r--r-- 1 lfs lfs 35K Jul 17 2007 COPYING3 -rw-r--r-- 1 lfs lfs 7.5K Jul 17 2007 COPYING3.LIB -rw-r--r-- 1 lfs lfs 524K Jun 14 11:27 ChangeLog -rw-r--r-- 1 lfs lfs 3.2K May 13 2004 ChangeLog.tree-ssa drwxr-xr-x 2 lfs lfs 4.0K Jun 14 11:48 INSTALL -rw-r--r-- 1 lfs lfs 58 Jun 14 11:48 LAST_UPDATED -rw-r--r-- 1 lfs lfs 22K Feb 17 2012 MAINTAINERS -rw-r--r-- 1 lfs lfs 6.0M Jun 14 13:01 MD5SUMS -rw-r--r-- 1 lfs lfs 25K Jan 2 2012 Makefile.def -rw-r--r-- 1 lfs lfs 1.4M May 16 18:54 Makefile.in -rw-r--r-- 1 lfs lfs 67K May 16 18:54 Makefile.tpl -rw-r--r-- 1 lfs lfs 579K Jun 14 11:48 NEWS -rw-r--r-- 1 lfs lfs 815 Oct 10 2009 README drwxr-xr-x 7 lfs lfs 4.0K Jun 14 11:35 boehm-gc -rwxr-xr-x 1 lfs lfs 3.7K Aug 22 2009 compile drwxr-xr-x 2 lfs lfs 4.0K Jun 14 11:35 config -rw-r--r-- 1 lfs lfs 25K Mar 22 2011 config-ml.in -rwxr-xr-x 1 lfs lfs 44K Jun 6 2011 config.guess -rwxr-xr-x 1 lfs lfs 15K Feb 13 2011 config.rpath -rwxr-xr-x 1 lfs lfs 35K Nov 2 2011 config.sub -rwxr-xr-x 1 lfs lfs 458K Feb 2 2012 configure -rw-r--r-- 1 lfs lfs 101K Feb 2 2012 configure.ac drwxr-xr-x 5 lfs lfs 4.0K Jun 14 11:35 contrib -rwxr-xr-x 1 lfs lfs 19K Aug 22 2009 depcomp drwxr-xr-x 3 lfs lfs 4.0K Jun 14 11:48 fixincludes drwxr-xr-x 17 lfs lfs 20K Sep 3 09:27 gcc drwxr-xr-x 14 lfs lfs 4.0K May 6 14:20 gmp drwxr-xr-x 2 lfs lfs 4.0K Jun 14 11:37 gnattools drwxr-xr-x 2 lfs lfs 4.0K Jun 14 11:35 include -rwxr-xr-x 1 lfs lfs 14K Aug 22 2009 install-sh drwxr-xr-x 2 lfs lfs 4.0K Jun 14 11:34 intl drwxr-xr-x 2 lfs lfs 4.0K Jun 14 11:48 libada drwxr-xr-x 4 lfs lfs 4.0K Jun 14 11:48 libcpp drwxr-xr-x 4 lfs lfs 4.0K Jun 14 11:37 libdecnumber drwxr-xr-x 7 lfs lfs 4.0K Jun 14 11:48 libffi drwxr-xr-x 4 lfs lfs 4.0K Jun 14 11:37 libgcc drwxr-xr-x 9 lfs lfs 4.0K Jun 14 11:48 libgfortran drwxr-xr-x 6 lfs lfs 4.0K Jun 14 11:47 libgo drwxr-xr-x 4 lfs lfs 4.0K Jun 14 12:02 libgomp drwxr-xr-x 4 lfs lfs 4.0K Jun 14 11:35 libiberty drwxr-xr-x 4 lfs lfs 4.0K Jun 14 13:01 libitm drwxr-xr-x 15 lfs lfs 4.0K Jun 14 11:37 libjava drwxr-xr-x 3 lfs lfs 4.0K Jun 14 11:34 libmudflap drwxr-xr-x 4 lfs lfs 4.0K Jun 14 11:47 libobjc drwxr-xr-x 5 lfs lfs 4.0K Jun 14 12:49 libquadmath drwxr-xr-x 3 lfs lfs 4.0K Jun 14 11:48 libssp drwxr-xr-x 11 lfs lfs 4.0K Jun 14 11:34 libstdc++-v3 -rwxr-xr-x 1 lfs lfs 3.3K Sep 20 2007 libtool-ldflags -rw-r--r-- 1 lfs lfs 258K Nov 21 2011 libtool.m4 -rw-r--r-- 1 lfs lfs 1.8K Sep 26 2008 ltgcc.m4 -rw-r--r-- 1 lfs lfs 244K Jan 13 2011 ltmain.sh drwxr-xr-x 2 lfs lfs 4.0K Jun 14 11:35 lto-plugin -rw-r--r-- 1 lfs lfs 12K Dec 5 2009 ltoptions.m4 -rw-r--r-- 1 lfs lfs 4.3K Sep 26 2008 ltsugar.m4 -rw-r--r-- 1 lfs lfs 703 Dec 5 2009 ltversion.m4 -rw-r--r-- 1 lfs lfs 6.0K Dec 5 2009 lt~obsolete.m4 drwxr-xr-x 2 lfs lfs 4.0K Jun 14 11:37 maintainer-scripts -rwxr-xr-x 1 lfs lfs 12K Aug 22 2009 missing -rwxr-xr-x 1 lfs lfs 2.2K Jul 22 2000 mkdep -rwxr-xr-x 1 lfs lfs 3.5K Aug 22 2009 mkinstalldirs -rwxr-xr-x 1 lfs lfs 2.6K Feb 12 2011 move-if-change drwxr-xr-x 6 lfs lfs 4.0K Jul 19 15:46 mpc drwxr-xr-x 9 lfs lfs 4.0K Jul 3 18:02 mpfr -rwxr-xr-x 1 lfs lfs 2.3K Jul 14 2005 symlink-tree -rwxr-xr-x 1 lfs lfs 6.1K Aug 22 2009 ylwrap drwxr-xr-x 11 lfs lfs 4.0K Jun 14 11:34 zlib I know it's frustrating -- it's the same bug that I and numerous others have experienced, but none of the team will accept that it's a bug. It took me four or five attempts to get GCC to build, but my problem was at pass 2. At one attempt I even copied and pasted the whole instruction set and it still failed at the same point that you've found. Bruce suggested wrapping the commands in a script so you can see what's happening, and that seems like a good idea, however I had already built it by then by doing nothing different -- just trying it yet again. I'd still be interested to know why we build GMP, MPC, and MPFR inside GCC except on the final build where they are built separately. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Perl-5.16.1 test failures in Ch 6 SVN-20120816
Thanks, but perhaps not necessary - it seems to be a problem at my end (see Bruce's response, and my reply to that). In particular, the run as a regular user seems NOT to be the key. ?en -- das eine Mal als Trag?die, das andere Mal als Farce Probably not of much use to you then, but as I ran the tests last night as root here are the results:- == Testsuite summary for GNU Automake 1.12.2 == # TOTAL: 2852 # PASS: 2648 # SKIP: 164 # XFAIL: 40 # FAIL: 0 # XPASS: 0 # ERROR: 0 == Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Perl-5.16.1 test failures in Ch 6 SVN-20120816
Richard Melville wrote: Failed 2 tests out of 2202, 99.91% okay. ../cpan/IO-Compress/t/105oneshot-zip-only.t ../cpan/Time-Local/t/Local.t I'm guessing that this is not a problem. Any views appreciated. That's a problem we are working right now. It's a timezone installation issue. From your comments, you appear to be pretty new to LFS. Why are you not using the stable version? -- Bruce Thanks Bruce. I'm a little rusty with LFS as I'm revisiting it after some years. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] Perl-5.16.1 test failures in Ch 6 SVN-20120816
On Mon, Aug 20, 2012 at 12:04:49AM +0100, Ken Moffat wrote: Unfortunately, this was unlogged and scrolled out of my term's buffer - it then died with an EPERM trying to create test-suite.log.tmp so I've now started it again, after chown me ../automake-1.12.3. So, in effect that is chown -R some-normal-user ../automake-1.12.x If you are interested, compare what we do for the coreutils tests. [ if you aren't, I understand ] ?en -- das eine Mal als Trag?die, das andere Mal als Farce Thanks for the really detailed replies Ken -- it's much appreciated. I'm building this when I have the spare time; I'll see if I can run the tests tonight and get back to you. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] coreutils-8.17 tests run as nobody fails
Thanks Bruce -- the nobody test suite now runs but all tests fail owing to mv and grep not being found :-( Maybe I should just move on. That would be best for you until I get this fixed. The problem is that we are using a different version of su in Chapter 6 than we used to use. The old version was from coreutils. The new version is from shadow and the behavior is different. Specifically it changes PATH and /tools/bin is missing. I need to correct the path for these tests to work. Thanks Bruce -- what was the thinking behind using SU from Shadow instead of Coreutils? Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
[lfs-support] Perl-5.16.1 test failures in Ch 6 SVN-20120816
Failed 2 tests out of 2202, 99.91% okay. ../cpan/IO-Compress/t/105oneshot-zip-only.t ../cpan/Time-Local/t/Local.t I'm guessing that this is not a problem. Any views appreciated. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] coreutils-8.17 tests run as nobody fails
Richard Melville wrote: Some help with this would be great -- I just can't understand it. I ran the tests as root which ran OK. I've added the temporary group and changed permissions but when I run:- su nobody -s /bin/bash -c make RUN_EXPENSIVE_TESTS=yes -k check || true It returns:- bash: make: command not found. That's due to an issue associated with changes I made about 12 hours ago. Try: su nobody -s /bin/bash -c TZ=UCT0 /tools/bin/make RUN_EXPENSIVE_TESTS=yes -k check || true -- Bruce Thanks Bruce -- the nobody test suite now runs but all tests fail owing to mv and grep not being found :-( Maybe I should just move on. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] upgrading 2-year-old sys
Hi! I had been using LFS for half a year until I realized that keeping it up-to-date is a pain. I feel I'd like to have some hard work again:), so now I'm trying to upgrade that system. As far as I know It could be a failure, but definately not an easy task... I'm confused about how to upgrade the toolchain. I believe constructing a temp. system can be skipped, as I have a functional LFS. I'm doing the project chrooting into LFS. I have updated the arithmetic packages that gcc need, and binutils. Ok, to make it short I don't know the order of the packages to update. Any clue would be appreciated! Why do you want to upgrade? Is something not working or do you just want the latest of each package. I have to agree -- I'm still running a venerable LFS 6.1.1 build which is quite capable of supporting, for example, the latest Erlang and Postgres packages. I do have a question, however, regarding vulnerabilities in old packages. Does anybody know of a good website that lists vulnerabilities as they are found. That would enable us to replace just those packages in old builds that represent a security risk. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] upgrading 2-year-old sys
On Fri, Aug 17, 2012 at 11:32:06AM +0100, Richard Melville wrote: I do have a question, however, regarding vulnerabilities in old packages. Does anybody know of a good website that lists vulnerabilities as they are found. That would enable us to replace just those packages in old builds that represent a security risk. The best I can do is to point you to what I added to BLFS : http://www.linuxfromscratch.org/blfs/view/svn/postlfs/vulnerabilities.html ?en -- das eine Mal als Trag?die, das andere Mal als Farce Thanks Ken -- that was really helpful. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] GCC-4.6.2 on 32 bit Mint13 mpfr error
Just went through this step in linux mint 32bit in vmware and found no issues. That's because it's an intermittent bug. I've just had the same problem using Linux Mint Cinnamon 64 bit. Sometimes the error message appears and sometimes it doesn't. If you look back through the posts in the mailing list you will see that a number of people have experienced this bug irrespective of host used. Richard Explain the exact commands used at the command line and any environment variables you may have changed. Sincerely, William Harrington -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
[lfs-support] Google Chrome disappears after running script to remove $LFS/dev/shm symlink
Can anybody tell me why the above happens? I'm using Chrome on the host to follow the book. Chrome won't restart and I'm now using Firefox. I noticed that the symlink was also removed from the host /dev directory; should that be so? I'm guessing that's why Chrome halted. I thought that the script would just remove the $LFS/dev symlink. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] GCC-4.7.1-Pass 2 MPC configure fails
Richard Melville wrote: I realise that I'm building the dev edition, but my host is Linux Mint Cinnamon 64 bit and the host requirements appeared to fit better. Also it looked as though the dev edition was at a reasonably stable stage. I'm building a 64 bit edition on a 64 bit host (OS and hardware). The failure is:- checking for MPFR... no configure: error: libmpfr not found or uses a different ABI (including static vs shared). make[1]: *** [configure-mpc] Error 1 make[1]: Leaving directory `/mnt/lfs/sources/gcc-build' make: *** [all] Error 2 Everything has built fine up to this stage and the sanity checks were OK. MPFR and GMP have compiled OK with the libraries installed in .libs. I've even checked to make sure that the MPFR libraries were 64 bit, and now I've run out of ideas. I'd be really grateful for any help. I've tried rebuilding GCC four times now with the same result. Are you sure you changed to the gcc-4.7.1 directory before tar -Jxf ../mpfr-3.1.1.tar.xz mv -v mpfr-3.1.1 mpfr tar -Jxf ../gmp-5.0.5.tar.xz mv -v gmp-5.0.5 gmp tar -zxf ../mpc-1.0.tar.gz mv -v mpc-1.0 mpc -- Bruce Thanks for the really quick response Bruce -- much appreciated. I'm 99.9% certain that I was in gcc-4.7.1; It's a routine -- untar and then cd. Even if I forgot once (which is unlikely) I don't believe that I would have forgotten on all four occasions; I always double check everything. I've now attempted building gcc for the fifth time and it has built OK; this is really weird. I've trawled the web and noticed that others have had the same error message in MPC configure without finding a positive answer. Could it be a strange intermittent bug? I'm not trying to shift responsibility from my own actions, but I just can't see what I did differently the fifth time around. Thanks again. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: [lfs-support] GCC-4.7.1-Pass 2 MPC configure fails
Hi Bruce I've now completed the temp build successfully, but digging around in the file system to try and track down that error I've noticed that I have the following directories under $LFS/tools (in addition to all the others of course):- x86_64-lfs-linux-gnu x86_64-unknown-linux-gnu That can't be right can it? They seem to hold duplicated content. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
[lfs-support] GCC-4.7.1-Pass 2 MPC configure fails
I realise that I'm building the dev edition, but my host is Linux Mint Cinnamon 64 bit and the host requirements appeared to fit better. Also it looked as though the dev edition was at a reasonably stable stage. I'm building a 64 bit edition on a 64 bit host (OS and hardware). The failure is:- checking for MPFR... no configure: error: libmpfr not found or uses a different ABI (including static vs shared). make[1]: *** [configure-mpc] Error 1 make[1]: Leaving directory `/mnt/lfs/sources/gcc-build' make: *** [all] Error 2 Everything has built fine up to this stage and the sanity checks were OK. MPFR and GMP have compiled OK with the libraries installed in .libs. I've even checked to make sure that the MPFR libraries were 64 bit, and now I've run out of ideas. I'd be really grateful for any help. I've tried rebuilding GCC four times now with the same result. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Clock Problems
Mykal Fink wrote:- I replaced the battery and the behavior didn't change. But at least, for a very small outlay, you can now rule out battery problems, and you don't have to worry about losing time when the box is unplugged. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Clock Problems
I agree with everything that's been said, but why not just *buy the battery*; then you'll have no time concerns whatsoever. In the UK they cost from about £1 upwards, depending on the type. I really can't see what the problem is. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Clock Problems
Sorry, I meant to say no time problems whatsoever regarding the battery. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Clock Problems
On Friday 11 December 2009 12:49:52 Johnneylee Rollins wrote: I am use to old hardware (i486DX) having problems keeping time on the hardware clock. But isn't the system clock a separate thing? I am losing about 4 min on the system clock for every 10 minutes of real time. I've googled around for clock drift information. What I found suggests that a system under heavy load with the 2.6.x kernel on certain hardware might show this symptom. I've yet to try it, but I've read that adding clock=pit noapic nolapic to the boot parameters should fix it. Is this something that will affect an LFS build? I don't like the idea of finding out towards the end that it will. That is my main concern. Should I ignore the clock issue? Is this something I should concern myself about? Any advice would be welcome. Thanks in Advance, Mykal Funk I'm not sure about a permanent fix, but a script to update the time with a ntp server might help. I'm not sure of the best method for offline use unless someone can absolve this issue with a more permanent solution. You may use the hwclock command periodically (that is in cron job) to help keeping your system time accurate. It's a very old motherboard; a dying cmos battery will affect the hardware clock which in turn will affect the system clock. You could try replacing the battery (usually a coin cell); they're not very expensive. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Stormy Peters and the Gnome Foundation
Just a quick report back. Although a good night was had by all, Stormy probably wasn't the right person to ask about Gnome technical issues as her post is mainly managerial. In reply to Jason she did say that Nautilus was very much in active development and that The Gnome Foundation was very keen to get input from users. In reply to Alan she said that her own view about Mono was that she wouldn't want to say that people shouldn't use it but she, personally, felt that it was best avoided. In reply to my question about GConf she answered, quite honestly, that she did not know enough about it. Although I wasn't able to glean much information regarding the above questions it was really good to meet Stormy Peters and to be able to discuss the work of The Gnome Foundation with her. Thanks again for your questions. Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Stormy Peters and the Gnome Foundation
Thanks to Simon, Alan and Jason for the feedback; I'll put the two questions to Stormy. Simon, I take your point but my only thoughts on the relationship between Windows and Gnome was that they both have registries and they can both become corrupted. Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: UDEV - Not Leaving Well Enough Alone
Running old computers is often touted as the green option. It's a fact that the two most vulnerable components in such computers are the power supply and the hard disk. Having had both of these components fail at various times on oldish boxes I can only infer that those on this mailing list continually running *very* old computers must be continually replacing these components. This is hardly an economic pursuit as new components for old computers are always much more expensive than their modern counterparts (RAM for instance). If, instead of new parts, old parts are being recycled then the failure rate on a particular computer must be even greater. Finally, old computers are far less efficient than new ones in terms of power consumed. So, they're large, noisy, power hungry, and expensive to maintain. To conclude then, I really can't see the attraction. Recycle them and treat yourself to something new, small, quiet, and relatively powerful. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: UDEV - Not Leaving Well Enough Alone
Simon Geard said:- Because an external DVD writer costs on average three times what an internal one does, and offers roughly half the read and write speeds. It also adds clutter to my desk, and adds to the mess of cabling down the back of the desk, not to mention the inevitable bulky AC/DC adapter taking up several wall sockets. Simon. Mine (a very small slimline drive) is powered from the USB port. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: UDEV - Not Leaving Well Enough Alone
Bruce Dubbs wrote:- That works for you, but for most people, it's far easier to use a usb thumb drive with capacities in GB to do the same thing. Some very old systems do not have usb connections, but many, if not most, newer systems do not have a floppy drive. Parallel printer connections have gone away too. -- Bruce Bruce, i have to say that I agree, but I'm not sure who is using those old systems (that) do not have usb connections. When all we had in the way of removable media was floppy disks then there was no choice. However, they were unreliable then just as they are today. I remember buying packets of floppy disks only to find that when I started to use them some were faulty. Today we have a whole array of removable media from which to choose. It seems churlish, therefore, to select something that is built on outdated and unreliable technology. If I had my way I would round up all the floppy disks in the world and burn them, thus doing everybody a favour. And while I was at it I would throw all those clunky old fax machines onto the same fire. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Grub-1.97 problems
I kept building kernels, and /boot partition kept filling up, and eventually I switched to just using a /boot directory on the root /. I've found that LVM is excellent for managing partitions that need to be resized. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Upgrading udev
I need to upgrade udev-056 to a more recent version, say 122, on a older LFS. This, I understand, can be tricky. I figured if I chrooted into the system from elsewhere, deleted the existing version and reinstalled it would work. What do you think, any advice on this matter appreciated? MAC Hi Cliff I've been working away from home for some months so I don't have access to my LFS/BLFS boxes. As I mentioned before, I upgraded LFS 6.11 from what was an early version of udev (which needed hotplug) to a relatively recent version. I had no problems that I can remember; i think that I just overwrote the old version and removed hotplug, although maybe I removed both first -- not sure. . My advice though is to work on a copy of the OS -- that way if something does go wrong then you have other copies to fall back on. It also means that you don't have to work on a live version. i usually have different copies that I can boot into from grub. i can run from the master and mount other copies (directories or files) on a temporary mount point; trying different upgrades on each. If it doesn't work out (if you can't boot into that copy) then you can delete that copy and you still have the master. Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: Asus EeePC Laptop
Adrian Fisher wrote:- I don't really like the interface that comes with it but that is not the reason I bought it. I bought it with the intention of wiping it and putting my own system on there. After you have it successfully dual-booting Xandros/LFS you can wipe the Xandros partition. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Asus EeePC Laptop
Adrian Fisher wrote:- I want to put LFS on my ASUS Eee PC Laptop (40GB SSD) but it has no CD/DVD drive and I have no external one. While it already has Linux on it it is a minimal installation as it has no compiler and no means of installing software manually, other than the few packages Asus saw fit to make available for it. It doesn't have a compiler but it does have a terminal (Ctrl+Alt+T), a browser, and even wget; so no, you can't build software on it but you can install binaries of your choice as root. I have an EeePC for note-taking at events, and some web surfing and email, so I haven't done much with it. Maybe you could create another partition and install LFS on that partition from a USB flash drive. Then you could adjust grub accordingly -- just an idea. However, storage space may be a problem. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: userspace error
Bruce Dubbs wrote:- In fact, I have: /dev/sda5/ ext3 defaults 1 1 /dev/sda7/home ext3 defaults 1 2 /dev/sda3/boot ext3 defaults 1 2 /dev/sda9/opt ext3 defaults 1 2 /dev/sdb1/usr/src ext3 defaults 1 2 /dev/sdb2/home/vmware ext3 defaults 1 2 /dev/sda6swap swap pri=1 0 0 Bruce, I've just noticed that you have /boot listed in your fstab shown above, but this is not necessary as it is called from Grub. IMHO it is better that /boot remains unmounted whilst the system is running making it more secure. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: userspace error
Ken Moffat wrote:- If /boot is a separate filesystem, you can use the version of grub installed by your host system. I would always recommend a separate /boot partition whatever the build. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: userspace error
Ken Moffat wrote:- If /boot is a separate filesystem, you can use the version of grub installed by your host system. I would always recommend a separate /boot partition whatever the build. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: glibc-2.5.1 installation fails
William Immendorf wrote: BUT, if stabilaty on recent systems is your goal, you should use 6.4. William, it's no good just repeating the same thing like a mantra -- show us the proof. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Swapon not working
Hello, this really does sound like it isn't enabled in your kernel. The option is CONFIG_SWAP, or under the name Support for paging of anonymous memory (swap). zgrep CONFIG_SWAP /proc/config.gz ( or substitute grep and your .config if you didn't create /proc/config.gz ) Yeah, sorry for the noise. Not sure what I was doing with grep -- my only excuse is that it was Friday evening and I was tired. Kernel is now re-configured and re-built and all is OK. Thanks Lauri and Ken. Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: kernel configuration and installation
That implies you are using LFS-6.2. I'm afraid I think glibc-2.3.6 is now regarded as very old. I don't know what to recommend - LFS-6.3 is about to become old, hopefully within the next 3 weeks, but as I said in a different thread I expect there to be a *lot* of breakage with the package versions currently in BLFS. Meanwhile, I see that you've just reached the final stage of your build :-( Hi Ken You were close -- it's actually LFS-6.1.1 with errata, and glibc-2.3.4 and gcc-3.4.3. The thing is I wanted to try an experiment; I wanted to see if an oldish version of LFS could be used to build an up-to-date fully-functioning OS with GUI, and I think that I've succeeded. I've had very few problems with LFS-6.1.1 -- there was an issue with glibc-2.3.4 and Xorg but DJ had come up with a workaround. I think that the only other modification that I made was to upgrade gettext to a newer version. Kernel headers have only been a problem with VLC and by disabling DCCP support it built OK. The finished product is a stripped down gnome environment with openbox and the AWN launcher. It has recent versions of scribus, ganttproject, gnucash, gimp, inkscape, dia, speedcrunch (calculator), cornice (image viewer), and the latest VLC, thunderbird 2, and firefox-3.0.3. I'm just finishing open office version 3 to (more or less) complete the build, which seems quite stable so far. The major problem is now the hardware. I wanted to use a low power, fanless, CPU and chose a VIA mini-ITX board with a 600MHz processor. It was quite fast when I was in text mode, but now that I have all the graphics assembled it has slowed to a crawl. I'm going to have to rethink it. Anyway, my point is that (exploits aside) older versions of LFS can be put to good use. I have read, although I am not qualified to comment, that gcc-3.x.x produces better, and more concise, code than gcc-4.x.x. When somebody with only a rudimentary knowledge of Linux begins their first LFS build it can be quite daunting and, more importantly here, take a long time. To be told that their build is now out-of date after spending hours and hours crafting it is, I believe, a little dispiriting. I realise that we have to move on but I do feel that a better balance can be struck. I didn't mean for this to sound too critical. I believe that the LFS/BLFS project is an incredible learning environment. It has certainly taught me a great deal about Linux and I hope that it has a long life ahead of it. I'd like to thank everybody involved in it. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: kernel configuration and installation
I've never looked at VLC. Looking at /usr/include/linux seems a reasonable thing for a configure script to be doing. Which kernel headers did you use when building glibc, and what is now reported to be missing-and-required ? Ken I believe that my original kernel headers were 2.6.12. The headers that VLC was looking for were the DCCP related ones ones that are available with the 2.6.26.5 kernel that I had just installed. According to the kernel devs, or at least last time I _heard_ (hearsay) anything about the subject, the answer was that the VLC maintainers failed to include the necessary kernel headers in the distribution tarball and provide a runtime check of the kernel for the necessary feature(s). I'm not certain if this is still current practice, and would appreciate a confirmation on that. DJ I'm not sure if it is still current practice but it appears to be -- I was building the very latest version. Thanks to both of you for the help, but I'm still unclear as to whether there is a workaround when the necessary headers aren't included with the package. Thanks Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Swapon not working
Hi I've managed without a swap partition until now, but I'm trying to build open office and the build is failing at the last minute due to lack of memory. So now I've made a swap partition of 2 gig and run mkswap. I've also amended the fstab. All seems fine until I run swapon -a which returns Function not implemented. I've tried making a swap file rather than a partition but I get the same output. If I reboot to run the script at boot-time I can see the same output. I've tried various sizes of swap -- no difference. Could it be a udev problem? I upgraded to udev-124 some time ago but everything else works fine. Running free -m shows swap but with all zeros, of course. Maybe swapon is corrupted -- I just don't know. I have no /proc/swaps file. When is this generated? I was wondering whether swap has to be enabled in the kernel but I can find no reference to it in the kernel .config. I'd be really grateful for some help as I'm right at the final stage of my BLFS build. Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: kernel configuration and installation
Yes, I think you've missed the important thing ;) The kernel headers are what glibc was compiled against, and they should not be changed unless you upgrade glibc [ and before anyone misconstrues that, we *don't* support upgrading glibc - when the time comes, build a new system ]. Hi Ken My reading of Rob's post was that he was wondering why distros like Ubuntu could frequently update kernel headers when we are told not to. If this was not his question then I wouldn't mind some advise on this issue. The problem occurs when some packages insist on parsing /usr/include/linux. I had a problem recently when installing VLC. I had enabled DCCP in my new kernel and I wanted to build VLC with the required support. I had already tested DCCP and it was working OK, but the VLC build failed complaining about missing headers. When I checked the source code it was looking in /usr/include/linux, which surely must be bad practice. I can't see why arbitrary packages should be poking around in the kernel headers. Clearly, as my glibc was built against much older kernel headers its search was unsuccessful. I was wondering what the solution is here? Should we install the new kernel headers into a separate sub-directory and change the source code to point to the new sub-directory rather than to /usr/include/linux, or would this just not work? I'd appreciate your, or anybody else's, view on the subject. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Failure to boot
When booting my lfs installation, I get the following errors: swapon: cannot stat dev/sda3: no such file or directory fsck.ext3: No such file or directory while trying to open /dev/sda4 I suspect this is due to my grub configuration. Rather than installing grub, I added these lines to menu.lst on my host system: title LFS 6.3 root (hd0,3) kernel /boot/lfskernel-2.6.22.5 root=/dev/hda4 I'm assuming that the grub menu.lst that you edited on your host system is on sda2 (Ubuntu). If these are the files that you are using to boot LFS then your LFS entry on the second line should read root (hd0,1), *not* root (hda0,3). As far as I know grub only recognises the hd disk nomenclature, so even if the kernel sees your disks as sd grub will still see them as hd, so the above is correct. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Via epia ex15000g framebuffer @1024x768
howto put my framebuffer to right resolution.. i am using video=uvesafb:1024x768-32 but screen is too big to fit my display, seems that 1024x768 mode does not work?? can anyone help me? i have via epia ex15000g motherboard, kernel 2.6.25, lfs 6.3 I thought that uvesafb was still experimental. Maybe you are better off using vesafb, or try installing one of the via unichrome frame buffer drivers available from via or directfb. Vesafb, as it has already been pointed out, takes the decimal form when fed to the kernel at boot time, eg vga=792 equates to 1024x768-24; note *vga=* in this instance, not *video=*. I don't think that vesafb operates at a 32 bit colour depth. Richard Melville -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: grub hangs without error message on mac mini
I've just installed LFS SVN-20080423 on a new Mac Mini (Intel) but the computer doesn't start (no OSX, only one partition for Linux). All I see is: GRUB Loading stage1.5. GRUB loading, please wait... There is no error message, nothing. I'm a little bit puzzled, as this is not the first LFS system I install, but the first time I don't know what to do. I use grub-0.97 with the disk_geometry and the 256byte_inode patches from the development page. The partition to boot from is /dev/sda1 and in my grub menu is default 0 timeout 10 title LFS root (hd0,0) kernel /boot/lfskernel root=/dev/sda1 The file does exist and I installed grub in the grub shell with root (hd0,0) and setup (hd0). Just to make sure, I installed ubuntu (7.04) on the very same computer (they use grub too) and the system starts, without problem. So the Mac Mini does start some Linux with grub, it's just my installation. I have a MacBook running with LFS and grub, I think this is some similar hardware and there isn't a problem either. So, why is grub hanging there (I rebooted after 10 minutes)? What might I do to solve the problem? What is grub waiting for? Thanks for any help, Andreas Just a thought - have you copied over to the /boot/grub/ directory the correct 1.5 file in relation to the file system that you are using. They are all file system specific. Richard Melville -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: GRUB Problem
This is really maddening because I'm so close to finishing, I'm on section 8.4 of lfs 6.3, Making the LFS System Bootable and when I use grub it gives me Error 21: Selected disk does not exist. The host distribution is Ubuntu 7.10 and since I'm new to this I haven't deviated from the book. Thanks. Ben Without seeing your menu.lst file and what partition(s) you are using it's hard to comment. I assume that you are aware of the difference between the GRUB partition naming convention and that of Linux, e.g. (hd0,0) ==hda1. Also, I'm not sure what you mean by when I use GRUB; do you mean when you boot LFS? Richard Melville -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Grub Hangs
l 1. When Grub hangs, the screen freezes. None of the keys, including PgUp, let me scroll up. 2. No, I don't have a .config file for the machine. (I didn't know what you meant by .config, zcat, /proc/congif.gz and make oldconfig. After doing my homework, I have now learned how to configure the kernel using an old configuration file. I also noticed along the way that my kernel was set for an Athlon processor. I recompiled for Pentium MMX, but the problem persists, so tweaking does indeed seem tangential to it.) 3. The kernel version is 2.6.16.27. Would it be worthwhile trying Adrian Bunk's version 2.6.16.60? Is there anything I have to do to avoid a conflict with what I have done so far with 2.6.16.27, or can I simply unpack his version and follow the compilation steps as if starting from scratch? Edward Why not just use the latest stable kernel? I'm using 2.6.24 with LFS 6.2 and it works well. Richard Melville -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: xfdesktop from CBLFS
XFCE is no longer supported in BLFS (and you didn't even specify its version!). There is some activity on the XFCE mailing list, but it is mostly about the upcoming 4.6 release. Any idea why XFCE is no longer supported in BLFS? Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: No sound from speakers
On Tue 12 Sep 2006 Dan Nicholson wrote: The alsa.dev script was used when hotplug and udev were installed together. Hotplug handled dynamic devices, and it would use the alsa script placed in /etc/dev.d. Nowadays, udev has completely deprecated hotplug to the point where any of the dynamic actions are defined in udev rules. But, that happened in between LFS-6.1.1 and LFS-6.2. So, what version of LFS are you building on? I'm sorry I should have said. I'm building on LFS-6.1.1 which is why I'm using the BLFS-6.1 book. Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
No sound from speakers
Please can somebody help. I've reached the stage of desperation. I've compiled the via drivers and alsa into the kernel and installed the alsa library, plug-ins, utilities and alsa oss. I've run the speaker-test and played a wav file and everything on the terminal screen looks good. I have not been able to get any failures, but I still cannot get any sound from the speakers. I've tested the speakers and cable on another box and they are good. Does anybody know what the problem could be? It must be something simple that's eluding me. Thanks in advance. Richard -- http://linuxfromscratch.org/mailman/listinfo/blfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: (no subject)
When I built my LFS system, i got the GRUB error 18 message at stage 1.5. i looked this up and saw something about my disk was beyond the scope of the BIOS or something (the meaning of error 18). Isn't this the error referred to in the LFS book - the one it says to ignore. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Problem with building Perl-5.8.8 in section 6.2.2
The final build of Perl works fine but when I try to build autoconf The @INC path has the /tools directory hardwired into it. I've tried using grep in the Perl source tree to find where this occurs but have had no luck. I don't know why this would happen. Just to make sure, could you run the sanity check at the end of the GCC build in Ch. 6.12? There should be almost no references to /tools left after that. I suspect that it's finding that /tools/lib is your standard library directory. I can't quite narrow it down, but look at config.h after Configure. There are some variables in there that seem suspicious. Otherwise, you have to grovel through the the Configure script, which is equivalent to gouging your eyes out with your thumbs. -- Dan -- Sorry about the wrong title in my last posting. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: LFS 6.1.1 Chapter 6 Findutils make check errors
Ken, although I have followed the book to the letter and already have coreutils successfully installed and tested, the findutils test failures do seem to point to an incorrect path. All the failures in both xargs and locate were accompanied by the *no such file or directory* message. When I return on Wednesday I will post the findutils error messages, and apply some of your suggestions. Thanks again. Richard Now I feel really stupid. It has taken me months to get this far with lfs because I've had so little time to spare - I've been dipping in and out of it. That's my excuse anyway for suddenly realising that I had compiled and tested coreutils but forgotten to install it. Sorry for wasting your time you guys. I'll be more careful before I post again. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: LFS 6.1.1 Chapter 6 Findutils make check errors
Richard Melville wrote: Could you show the failures in glibc and findutils? Thanks for all your help. I'm away until Wednesday but, Dan, I have already posted the glibc test failures in the April 2006 archives. You can see them there. I stripped out some of the obvious ones like the maths test and left the ones that I could find no reference to anywhere. We had this on CLFS at one time, when we were building in a slightly different order than LFS. Our xargs failures were because /bin/echo didn't exist, and the locate failure seems to have been related to 'sort' - I think the installed script was using /tools/bin/sort (ouch!), but the tests used /usr/bin/sort. Our problem was because we built coreutils too late, but it only came to light with findutils-4.2.25 which has rewritten tests. Are you perhaps using newer versions of some of the packages ? Ken, although I have followed the book to the letter and already have coreutils successfully installed and tested, the findutils test failures do seem to point to an incorrect path. All the failures in both xargs and locate were accompanied by the *no such file or directory* message. When I return on Wednesday I will post the findutils error messages, and apply some of your suggestions. Thanks again. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
LFS 6.1.1 Chapter 6 Findutils make check errors
Hi I'm hoping somebody can help as I can find no other threads relating to this. I'm getting a lot of test failures in *locate* and *xargs*, but *find* was OK, and Findutils seems to compile OK. I can't see that this is too much of a problem (it doesn't appear that this package is a dependency for anything else) and I was thinking of ploughing on regardless. However, I'm concerned that it might be symptomatic of some other unseen problem. I did have a few failures in glibc but gcc test results were better than the book. If more detailed info is required I will reply with it. BTW I'm using the live CD and a VIA Eden fanless m/board with 512 MB of RAM. Thanks in advance Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
glibc-2.3.4 test failures in 6.1.1 chapter 6
I'd really welcome some comments on these test failures. I've stripped out those that I know have already been reported on these threads, but have been given the all clear. I've also stripped out the maths tests which I knew would fail as I am using a VIA processor. This leaves the following:- make[2] : *** [ /sources/glibc-build/posix/tst-regex2.out ] Error 1 make[1] : *** [ posix/tests ] Error 2 make[2] : *** [ /sources/glibc-build/misc/tst-tsearch.out ] Error 1 make[1] : *** [ misc/tests ] Error 2 make[2] : *** [ /source/glibc-build/rt/tst-timer5.out ] Error 1 make[1] : *** [ rt/tests ] Error 2 make[2] : *** [ /sources/glibc-build/elf/tst-align.out ] Error 1 make[1] : *** [ elf/tests ] Error 2 BTW I'm using the live CD. Thanks in anticipation. Richard Melville -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Glibc configure problems with LFS 6.1 after chroot
Hi I'd really appreciate some help from somebody. I'm trying to build LFS 6.1 with errata. Chapter 5 seems to have built OK, but glibc, the first package to be compiled in chapter 6, fails to configure, complaining that there is no such file as /tools/bin/gcc, when there clearly is. The sanity check fails, as does gcc -dumpmachine. However, if I exit the chroot environment and run the sanity check again, all is well. The sanity check returns /tools/lib/ld-linux.so.2 and gcc -dumpmachine returns i686-pc-linux-gnu. Is it a compiler problem? If so can I rebuild gcc-3.4.3 again at this stage? The glibc config.log reads *cannot compute suffix of object files: cannot compile*. I did run the test suite on gcc-3.4.3 in chapter 5, and apart from precompiled header failures it seemed OK. BTW I'm using the live cd. Many thanks in advance. Richard Melville -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Glibc configure problems with LFS 6.1 after chroot
On 3/22/06, Richard Melville richard at netvaluesystems.com
http://linuxfromscratch.org/mailman/listinfo/lfs-support wrote:
/
// The sanity check fails, as does gcc -dumpmachine. However, if I exit
// the chroot environment and run the sanity check again, all is well. The
// sanity check returns /tools/lib/ld-linux.so.2 and gcc -dumpmachine
// returns i686-pc-linux-gnu.
/
First, what's the exact error give when you attempt the sanity check
in the chroot?
Second, are you sure you have all the proper symlinks in place from
Creating Essential Symlinks?
Third, let's try a souped up sanity check similar to the one in the SVN book.
http://www.linuxfromscratch.org/lfs/view/development/chapter06/readjusting.html
echo 'main(){}' dummy.c
cc dummy.c -Wl,--verbose dummy.log
readelf -l a.out | grep 'ld-linux'
grep 'libc.so.6' dummy.log
grep 'SEARCH_DIR' dummy.log
grep 'crt[1in].*succeeded' dummy.log
grep 'found' dummy.log
That should help get to the bottom of this mess.
--
Dan
Hi Dan
In the chroot environment I can't even get past compiling dummy.c;
*cc dummy.c* returns the reply */tools/bin/cc: No such file or directory*
I've checked all the symlinks and they are installed.
Ken
The libraries all seem to be in place. I understand what you are saying
about being linked to a library on the host which disappears after entering
then chroot environment. I though that the sanity check was meant to guard
against this. How can I check whether this is the case?
--
http://linuxfromscratch.org/mailman/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page
Glibc configure problems with LFS 6.1 after chroot
OK, I pretty stumped. One last thing. Make sure that gcc itself is actually linked correctly. readelf -l /tools/bin/gcc | grep 'ld-linux' This should be /tools/lib/ld-linux.so.2 or obviously it won't reference the correct location in the chroot. -- Dan The output is /lib/ld-linux.so.2 rather than /tools/lib/ld-linux.so.2. Is this the problem, and if so how could it have happened? Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Perl 5.8.7 compilation error
Hi I'm building LFS 6.1 with the errata (perl 5.8.7 and the 5.8.7 libc patch). In chapter 5.32 after running *make perl utilities* the compilation fails complaining that there is no such file or directory as *lib/auto/posix/posix.a*. Any ideas. I'd appreciate the help. I'm using the live CD. Thanks in advance. Richard Sorry to everyone for the unnecessary post. I must have typed posix in lower case when compiling. My excuse - it was late and I was very tired. Richard Melville -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Perl 5.8.7 compilation error
Hi I'm building LFS 6.1 with the errata (perl 5.8.7 and the 5.8.7 libc patch). In chapter 5.32 after running *make perl utilities* the compilation fails complaining that there is no such file or directory as *lib/auto/posix/posix.a*. Any ideas. I'd appreciate the help. I'm using the live CD. Thanks in advance. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
GCC-3.4.3-Pass 2 Test Results
Hi I realise that the test results are not critical in the temporary system but I just need some confirmation that it is worth proceeding. Three of the four sets of my results were as good, if not better, than the results shown at the URL (which incidently now seem to have disappeared). My GCC Summary, however, seems a lot worse than those results. The figures are as follows:- === gcc Summary === # of expected passes 24485 # of unexpected failures 146 # of expected failures70 # of untested testcases 152 # of unsupported tests 200 /mnt.lfs/sources/gcc-build/gcc/xgcc version 3.4.3 Most, if not all, of the failures appear to be thus:- gcc.dg/pch/* I would welcome some comments. Thanks in advance. Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
GCC-3.4.3-Pass 2 Test Results
Dan I'm using the 6.1 live CD and I've downloaded the errata (obviously not relevant at this stage of the build). Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
GCC-3.4.3-Pass 2 Test Results
Dan Output from uname -r is 2.6.11.12 Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
PTY test
Hi This is my third attempt at building LFS 6.1 as I have not had enough time to complete before. Each time I have reached the PTY test after the installation of *dejagnu*, and issued the command *expect -c spawn ls* the shell just echoes *spawn ls*. Why is this? Thanks in advance Richard -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
Re: Patching expect-5.43
Richard Melville wrote: / Thanks Dave. Do you mean that if the Hunks aren't listed then they // have succeeded? In addition, an offset of 103 lines isn't *off by a // line or so.* Is this still acceptable?/ Yep. :) If the hunks aren't listed, they've succeeded. :) And the offset of 103 lines is fine... It just means the patch hasn't changed lately, while the code has... If the patch is rejected, it's something to worry about, but that hasn't happened in this case. :) If you're following the book, you're in good hands! BTW, I notice that you said you're following 6.1... You *may* want to switch to 6.1.1, or at least read the errata for 6.1 before continuing. :) Dave Thanks everybody for the help. Thanks Dave - I already read the errata and downloaded the extra patches together with Perl-5.8.7 and Zlib-1.2.3. -- http://linuxfromscratch.org/mailman/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/lfs/faq.html Unsubscribe: See the above information page
