Re: OT: Too much RAM ?
32bit kernel, right?. Probably it's the kernel remaping mem. PAE is quite slow. It's an old tale ;). On 11/30/09, tho...@equinox.homelinux.org wrote: > Hi all, > > this some kind of Off-Topic since it has nothing to do with LFS at all. I > send this to this list because I know that there are quite a lot geeks and > experienced people out there who may have a hint for me. > > Have you ever seen a machine which has too much RAM installed? We have a > brand new HP (DL380G6) server here with 48G RAM installed. It has an p410i > and a p800 SAS controller connected to a MSA70 with 1TB usable disk space. > There are 2 Quad-Core-Xeon (X5560) with hyperthreading (a total of 16 CPUs). > > As an OS there is s SuSE SLES10 installed and as DB-Software (because of > which we have this machine) Oracle10g. > > The problem: > When we create a file which is 32GByte large (for instance using dd > if=/dev/null of=test.dat...) the copy process runs very fast up to round > about 29-30GB slows down quite immediatly and the workload increases up to > loadavg=35 or so. It finally results in a non-responding machine where we > can only press the reset-button. > > What we did: > Adding the mentioned 30GB plus the rest what the OS itself is using it > adds up to quite exact 32GB. We removed 6 of the 12 RAM modules (every has > 4GB) so that the machine now has less that 32GB RAM (exact 24GB). The dd > command runs quite fast until the file was created and nothing special is > seen on the machine - everything worked as expected. > > Do you have an idea what the 32GB-limit could be? Or what could make the > machine to behave the way it does? > > Our next step will be to test that all with the 64-bit stuff but it should > work on 32bit too, isn't it? > > Any ideas? > > Thank you! > > Thomas > > -- > http://linuxfromscratch.org/mailman/listinfo/lfs-dev > FAQ: http://www.linuxfromscratch.org/faq/ > Unsubscribe: See the above information page > -- Ismael Luceno -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Grub2
Bruce Dubbs escribió: > Ismael Luceno wrote: > No, sorry. I don't see any posts from you since 2007. Yep, I used to use LFS, but went to debian because I had no time to maintain my system. Now I got a new job, less hours and more comfortable, so it lets me to dedicate more time to this stuff :). > What are you going to rewrite? The script, but I see it's not needed, sorry for the noise. -- Ismael Luceno My Blog - http://ismaell.wordpress.com/ InitNG- http://www.initng.org Ohloh - http://www.ohloh.net/accounts/ismael Gitorious - http://gitorious.org/~ismaell Contact me at: JID: ismael.luc...@gmail.com IRC: ismaell at irc.freenode.net () ascii ribbon campaign - against html e-mail /\ www.asciiribbon.org - against privative attachments signature.asc Description: OpenPGP digital signature -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Grub2
Bruce Dubbs escribió: > Reference: http://wiki.linuxfromscratch.org/lfs/ticket/2093 > > I noticed that a new beta version of grub2 was released last week. > > http://grub.enbug.org/ > > 2. configure errors out if you don't have ruby installed. It is only used > in > configure to generate the Makefile. The ruby file is only about 400 lines > long. > I would hope (but am not optimistic) that it won't be required to make the > final grub2 release. > Hi. I hope somebody still remembers me :P. If it's not too complex, I could rewrite it ;). I will look at it right now :). -- Ismael Luceno My Blog - http://ismaell.wordpress.com/ InitNG- http://www.initng.org Ohloh - http://www.ohloh.net/accounts/ismael Gitorious - http://gitorious.org/~ismaell LinkedIn - http://www.linkedin.com/in/ismaell My resume - http://ismael.initng.org/cv25.pdf Contact me at: JID: ismael.luc...@gmail.com IRC: ismaell at irc.freenode.net () ascii ribbon campaign - against html e-mail /\ www.asciiribbon.org - against privative attachments signature.asc Description: OpenPGP digital signature -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Default filesystem
Fix escribió: > On 4/9/07, Ismael Luceno <[EMAIL PROTECTED]> wrote: > >> My system is somewhat deviated, so a normal LFS may take a bit more, >> but the gettys/dm will be up as soon as possible, that's the beauty >> of initng, it does it without any effort :). > > InitNG is great __idea__. However, I know two men at least, who tried > to use it on a LFS system with no success. Can't you help, please? > The problems come from our scripts, not InitNG itself, we want to drop them, maintaining generic scripts is a big trouble, and they aren't as fast as they could be, because of the generic-ness :(. I can help to find the bugs in our scripts, but i prefer to have scripts specific for lfs. -- Ismael Luceno <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> InitNG maintainer and project lead - http://www.initng.org Registered Linux User #439653 - http://counter.li.org LFS User #17162- http://www.linuxfromscratch.org SourceMage GNU/Linux User - http://www.sourcemage.org IRC: ismaell @ irc.freenode.net #initng #uruguay #cross-lfs Jabber ID: [EMAIL PROTECTED] Blog: http://ismaell.wordpress.com GPG Key ID: EC8E5C9A GPG Key Fingerprint: 1356 7578 232E CCA6 D16D 46A8 FE6C 58D3 EC8E 5C9A signature.asc Description: OpenPGP digital signature -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Default filesystem
Dan Nicholson escribió: > On 2/3/07, TheOldFellow <[EMAIL PROTECTED]> wrote: >> What I need is something it can't handle, like Udev for several months a >> year ago, or a new booting scheme... > > This is actually something I want to bring up. Our booting is dog > slow. Maybe it's time to look into making improvements. We could > replace init with init-ng or upstart. Or, we could just work to > parallelize the bootscripts like is done on RedHat and SuSE. I think > this has been brought up before. > > -- > Dan For desktop systems it doesn't matter how long it takes to boot, what matters is how long it takes to load a getty or a dm ;). In my current system, an Pentium III @ 450MHz, the runlevel takes 8 secs to be up, but gettys are available at 4 secs, or less, i'm not sure... My system is somewhat deviated, so a normal LFS may take a bit more, but the gettys/dm will be up as soon as possible, that's the beauty of initng, it does it without any effort :). -- Ismael Luceno <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> InitNG maintainer and project lead - http://www.initng.org Registered Linux User #439653 - http://counter.li.org LFS User #17162- http://www.linuxfromscratch.org SourceMage GNU/Linux User - http://www.sourcemage.org IRC: ismaell @ irc.freenode.net #initng #uruguay #cross-lfs Jabber ID: [EMAIL PROTECTED] Blog: http://ismaell.wordpress.com GPG Key ID: EC8E5C9A GPG Key Fingerprint: 1356 7578 232E CCA6 D16D 46A8 FE6C 58D3 EC8E 5C9A signature.asc Description: OpenPGP digital signature -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Make bootscripts more POSIX compliant
Matthew Burgess escribió: On Tuesday 20 February 2007 06:35, Dan Nicholson wrote: : 1. POSIX compliant bootscripts - I'm all for this. If anyone wants to install a different shell as /bin/sh they should be able to without compromising their ability to boot their system without errors. Dan, did you know that dash(1) has a '-n' option: "If not interactive, read commands but do not execute them. This is useful for checking the syntax of shell scripts." This might make debugging/testing POSIX compatibility substantially quicker :-) 2. Parallel bootscripts. Whilst the benefits and drawbacks of doing this are unclear at the moment (or at least I think they are), I'm all for having these worked on until such a point where folks interested in it can come to the list with a hard-sell on why LFS should integrate them :-) If DJ and everyone else are happy with having them in contrib, that's fine by me. If you'd rather have a svn branch set up, just holler and it'll be yours. 3. Replace sysvinit. Again, I've not seen any convincing arguments why we'd want to do this but if someone wants to work on it then I can set up a branch for such work to be carried out. I want to integrate InitNG, however I'm very short of time right now, so if someone wants that too, I will try to help!. signature.asc Description: OpenPGP digital signature -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: fsck + localtime in BIOS and POSIX compliance
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Bryan Kadzban escribió: > Jack Brown wrote: >> I already am turing off UTC in /etc/sysconfig/clock. > > I have no idea if this is what this setting is for or not, but what > happens if you (1) have a 32-bit kernel, and (2) enable CONFIG_APM, and > then disable CONFIG_APM_RTC_IS_GMT? > > Of course if you run 64-bit, that won't help, and even if you run 32-bit > it still may not help. (Like I said, I'm not sure if that's what that > is for.) But it sounds like it *may* be helpful, even if the setclock > script doesn't get moved. > I think the script should be moved. This remembers me that I had problems because some of the InitNG scripts use some programs that aren't available until /usr is mounted, but they should be on POSIX compliant systems. Some of those programs are awk, sort, and uniq. As many distros have this problem, we're working on some workarounds, until they are fixed... The problem is that in some cases the workaround is waiting for /usr to be mounted, and InitNG is targeted for speed, so that's not good. - -- Ismael Luceno <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> InitNG maintainer and project lead - http://www.initng.org Registered Linux User #439653 - http://counter.li.org LFS User #17162- http://www.linuxfromscratch.org Jabber ID: [EMAIL PROTECTED] IRC: ismaell @ irc.freenode.org #initng GPG Key ID: EC8E5C9A GPG Key Fingerprint: 1356 7578 232E CCA6 D16D 46A8 FE6C 58D3 EC8E 5C9A -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.3 (GNU/Linux) iD8DBQFFwAB7/mxY0+yOXJoRAvivAJ4wXqAmfrZuvosvfUT1B+vsPXKfsACguZEA GRfDflRhzEM511Vrk42oWRU= =Y/po -END PGP SIGNATURE- -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Fwd: LFS Script suggestion
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Bruce Dubbs escribió: > Dan Nicholson wrote: >> On 1/29/07, Bruce Dubbs <[EMAIL PROTECTED]> wrote: >>> Dan Nicholson wrote: >>>> On 1/28/07, Alexander E. Patrakov <[EMAIL PROTECTED]> wrote: >>>>> Tushar Teredesai wrote: >>>>>> cat /proc/version | head -n1 | cut -d" " -f1-3,5-7 >>>> Pet peeve. Don't use cat to create input streams when the shell is >>>> perfectly capable on it's own with <. >>> vs >>> >>>> sed -n 's/.*gcc version \([^() ]*\)[() ].*/\1/p;q' < /proc/version >>> >From an educational view, what's wrong with showing a different way of >>> doing things. >>> >>> Actually Tush's version above is much clearer to me than your sed, >>> although I don't think the head -n1 is needed (it may be in some cases, >>> but not for my systems). >> OK, but it still needs to be fixed to work on Debian. Here's the >> output from Alexander's version string: >> >> [ 5:28 PM [EMAIL PROTECTED] cat foo | head -n1 | cut -d" " -f1-3,5-7 >> Linux version 2.6.18-3-686 2.6.18-7) ([EMAIL PROTECTED]) (gcc >> >> It might be best to cat the whole file like he mentioned unless you >> want to have a much more heavy duty parser. > > OK, I see now. Debian adds a field to their string. In that case a sed > is definitely needed. > >>From one of mine (Obviously not LFS): > > Linux version 2.6.9-42.0.3.EL_lustre.1.5.97smp ([EMAIL PROTECTED]) (gcc > version 3.4.4 20050721 (Red Hat 3.4.4-2)) #1 SMP Fri Jan 12 17:22:43 MST > 2007 > > And Alex: > > Linux version 2.6.18-3-686 (Debian 2.6.18-7) ([EMAIL PROTECTED]) (gcc > version 4.1.2 20061115 (prerelease) (Debian 4.1.1-20)) #1 SMP Mon Dec 4 > 16:41:14 UTC 2006 > > How about: > > $ sed -r 's/.*(gcc version [01234567890\.]+).*/\1/' /proc/version This string depends on the locale in use when the kernel was compiled, so it will not work on all systems, it needs to be more generic: $ cat /proc/version Linux version 2.6.19.1 ([EMAIL PROTECTED]) (gcc versión 4.0.3) #1 Fri Dec 15 02:19:26 UYT 2006 $ sed -r 's/.*\((gcc .+ [01234567890\.]+)\).*/\1/' /proc/version gcc versión 4.0.3 - -- Ismael Luceno <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> InitNG maintainer and project lead - http://www.initng.org Registered Linux User #439653 - http://counter.li.org LFS User #17162- http://www.linuxfromscratch.org Jabber ID: [EMAIL PROTECTED] IRC: ismaell @ irc.freenode.org #initng GPG Key ID: EC8E5C9A GPG Key Fingerprint: 1356 7578 232E CCA6 D16D 46A8 FE6C 58D3 EC8E 5C9A -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.3 (GNU/Linux) iD8DBQFFvsW5/mxY0+yOXJoRAuUyAKCS0pUO1++r/MxcO8i7o75f4VZcpgCfcDym 9LOP2JIHHrhikEUTh3C0WoM= =B7Tw -END PGP SIGNATURE- -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Problem with upgrading ncurses-5.x to ncurses-5.x+1
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 The last time I built LFS 6.2, I got a problem when compiling ncurses (related to tic); the host system was LFS 6.1. Now, I tried to update from 5.4 to 5.6, and I got the same problem. The solution was simple: rm -f /lib/libncurses.so.* ldconfig I hope it helps :) - -- Ismael Luceno <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> Try InitNG - http://www.initng.org Jabber ID: [EMAIL PROTECTED] IRC: ismaell @ irc.freenode.org #initng GPG Key ID: EC8E5C9A GPG Key Fingerprint: 1356 7578 232E CCA6 D16D 46A8 FE6C 58D3 EC8E 5C9A -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.3 (GNU/Linux) iD8DBQFFqqXh/mxY0+yOXJoRAtdYAJ9jhoPZyHrcZ0suV1JRMSixjD/bygCgvcuG qHMCECliR7J3/xDK7g9UYcE= =3kkI -END PGP SIGNATURE- -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Use of unionfs to build lfs in stage 5 en 6.
Stef Bon escribió: > Hello all, > > well I've posted a howto on howtoforge with something about LFS: > > http://www.howtoforge.com/ihlfs_full_control_over_what_youre_installing > > It's about the control you can get over a installation (or everything that > does modifications on your system) with the use of unionfs and chroot. > > You all know that the mainwebsite www.linuxfromscratch.org is not there > anymore? > > > But now something else: > > I've been working with the unionfs in combination with the chroot and I > thought we also could use it for building LFS. > > In stead of creating a tools directory on the new partition where the system > will be installed, and creating a link to it, > > you can also do: > > mount -t unionfs -o dirs=/mnt/lfs:/=ro unionfs /mnt/union > > and > > mount --bind /dev /mnt/union/dev > mount -t devpts devpts /mnt/union/dev/pts > mount -t tmpfs shm /mnt/union/dev/shm > mount -t sysfs sysfs /mnt/union/sys > mount -t proc proc /mnt/union/proc > > After these steps make the new "virtual" environment complete with a chroot: > > chroot /mnt/union /bin/bash --login > > Now you can compile everything in this chrooted environment as normal, thus > without the /tools directory: so with normal options like --prefix=/usr in > stead of the --prefix=/tools. As pointed out by Vladimir, this isn't a good idea. > > This works - I think - in step 5. When everything is done in chapter 5, and > the chroot session is ended, in /mnt/lfs there should be a temporary > system, with all the normal paths. I've not tested this, but there is a > hint about this: > > http://www.nl.linuxfromscratch.org/hints/downloads/files/package_management_using_trip.txt > > In chapter 6 the unionfs is not necessary anymore. And the tools directory > is not used, which is making things easier. But there the unionfs may be > used for packagemanagement. > > > What do you think? > The existing situation works, but this new construction looks good. The last two systems I built, I used castfs (from SourceMage), and worked very well. It was designed with that purpose on mind, and is better because it's a FUSE fs. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Use of unionfs to build lfs in stage 5 en 6.
Stef Bon escribió: > Alexander E. Patrakov wrote: > >> Stef Bon wrote: >>> What do you think about unionfs? >>> The existing situation works, but this new construction looks good. >>> >> Unionfs was used with LFS-6.1 LiveCDs. It has been dropped then, for a >> reason: too buggy, especially on SMP. Try running "make check" for glibc >> on unionfs. Or (for stress testing) try installing the same package >> (e.g., Vim) 1000 times repeatedly in /usr on LFS-6.1.1 CD. It will just >> hang or produce an oops. Many unionfs releases could not even boot the >> CD (i.e., produced an oops as early as during boot). > That's not good. Thanks anyway for you info. > Are these bugs reported to the team developing Unionfs? > >> The new DM-based CDs handle this stress test just fine, but DM is not >> suitable for package management because it works on block device level, >> not filesystem level. >> > > What is DM? Device Mapper. > > Stef Bon > -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Binutils testsuite
Alexander E. Patrakov escribió: > Robert Connolly wrote: >> I noticed -O3 causes tests to fail, too, with binutils-2.17. >> > Does "make CFLAGS= check" fix this case also? > It's the same problem as with -Os, and yes, that fixes it. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Future udev rules
Bryan Kadzban escribió: > Bruce Dubbs wrote: >> I was once told that the card closest to the power supply would come >> up first because it is in the lowest numbered pci slot. > > I am not sure, but I do not believe this is the case anymore. (Or if it > is, it's not supposed to be relied on.) ISTR reading something about > the PCI device discovery order changing several kernels ago. > It's not true. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Sysvinit - modified init.c "sending all processes" string
Bryan Kadzban escribió: > Ismael Luceno wrote: >> I think "Sending processes controlled by init the TERM signal..." is >> better; as processes started by bootscipts, and bootscripts >> themselves are not "controlled" by init. > > Good point. > > I could see a user thinking that bootscripts *were* controlled by init > (I did) -- but they really aren't. At least, not in a manner that would > affect these TERM/KILL actions (which is sort of the problem; if you > already know what processes are being killed, any correct wording should > work, but otherwise you might get more confused). Using Bruce's > "controlled directly" wording should work, though. > > I suppose it probably depends on how much we want to tell the user about > exactly how sysvinit works through this message. If it's important that > the user know that the affected processes will all be listed in > /etc/inittab, then the "controlled directly by /etc/inittab" wording > would be better. If it doesn't matter, the "controlled directly by > init" version would work just as well. > > What do you think about Bruce's "controlled directly by /etc/inittab"? > It's wrong, because /etc/inittab is a config file, so it doesn't control anything. The user can get confused with that too... -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: Sysvinit - modified init.c "sending all processes" string
Bryan Kadzban escribió: > Bruce Dubbs wrote: >> If a process is running that was started with the action ONCE and has >> not terminated (e.g. sulogin), it may be killed also. This makes >> your wording incorrect. > > That's right; I must have not realized that when looking through the > inittab manpage. For that matter, WAIT, POWERWAIT, POWEROKWAIT, and > maybe even KBREQUEST might be killed also, if they're still running. > And possibly even others; hmm. I think the vast majority of the time > it'll be only RESPAWNs, but that doesn't make the wording correct. > >> Perhaps the wording should be "Sending processes controlled by >> inittab ..." > > That sounds good to me. If we could fit a "directly" in there, that'd > be even better, though (e.g. "started directly from inittab"). Let's > see (I've got Thunderbird set up to wrap at 72 characters): > > Sending processes started directly from /etc/inittab the TERM signal... > I think "Sending processes controlled by init the TERM signal..." is better; as processes started by bootscipts, and bootscripts themselves are not "controlled" by init. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: lfs-6.2pre1 test report
Dan Nicholson escribió: On 7/21/06, Randy McMurchy <[EMAIL PROTECTED]> wrote: Matthew Burgess wrote these words on 07/21/06 16:51 CST: > Dan Nicholson wrote: > >> Does anyone know what creates /etc/blkid.tab? > > e2fsprogs You may wish to 'man blkid' for more information. OK, so I read it and ran the utility. I see what it does now. Now I'm curious about where in the LFS build through jhalfs this happens. libblkid actually does the work, so it could be mount or e2fsck or something else. I suspect mount, like Bruce says. Yes, it's done by mount. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: lfs-2.6-pre1 test report
The first time i built ncurses-5.5 tic complained about some unresolved symbols. I built ncurses-5.5 again (this time on LFS-6.2-pre1), and tic doesn't complain anymore, maybe i did something that caused the problem, or maybe the problem is related to the host. I will try to recreate the problem with an LFS-6.1.1 host. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: lfs-6.2pre1 test report
Dan Nicholson escribió: On 7/20/06, Ismael Luceno <[EMAIL PROTECTED]> wrote: I got a failure while installing ncurses-5.5, tic complains about an undefined symbol. So I'm using 5.4, until I had time to check that. Could you show the error? As I recall, this comes up from time to time and the fix is simple. I don't remember, and I deleted the logs. I will build it next weekend, if I have time. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: lfs-6.2pre1 test report
Miguel Bazdresch escribió: Hello, This is a report on my experience with LFS 6.2-pre1 so far. Arch is i386 (pentium3, 800MHz). I built it with jhalfs, which went well except for an initial glitch almost certainly caused by me. Host is an LFS system built from SVN as of last november, slightly suspicious because of said problems with jhalfs. Ch. 6 tests ran without any problems. Only unexpected failures with gcc were the usual libmudflap ones. I got the common annex.c Error 1 with glibc. The system boots and I have installed the following blfs packages: Python-2.4.3 X-6.9.0 atk-1.11.4 cairo-1.0.4 ctags-5.6 dhcp-3.0.4 expat-2.0.0 firefox-1.5.0.4 fontconfig-2.3.2 freetype-2.1.10 getmail-4.4.4 gkrellm-2.2.9 glib-2.10.3 gtk+-2.8.18 jpegsrc.v6b libIDL-0.8.6 libpng-1.2.8 libxml2-2.6.22 mutt-1.5.11 openssl-0.9.8b pango-1.12.3 pkg-config-0.20 postfix-2.2.3 procmail-3.22 screen-4.0.2 tcl-8.4.13 tiff-3.8.2 tk-8.4.13 vim-7.0 vte-0.12.0 wget-1.10.2 which xchat-2.6.6 xfce-4.3.90.2 zip-2.31 Everything seems to work. I have yet to test my scanner and printer, but I'm unlikely to have a chance to set them up before the weekend. Sucessfully built on an Pentium III Katmai with LFS-6.1 as host. Additional Packages: abiword-2.4.4 alsa-lib-1.0.9 alsa-utils-1.0.9a arj-3.10.22 asclock-2.0.12 atk-1.11.4 audiofile-0.2.6 cairo-1.0.4 cdparanoia-III-alpha9.8 cdrtools-2.01 clamav-0.88.2 cmake-2.2.3 cups-1.2svn-r4699 dhcpcd-2.0.5 dosfstools-2.11 eject-2.1.4 enlightenment-0.16.7.2 esound-0.2.36 espgs-8.15.2 expat-1.95.8 firefox-1.5.0.4 flac-1.1.2 fontconfig-2.3.2 freetype-2.1.10 fribidi-0.10.7 gaim-1.5.0 gdbm-1.8.3 geany-0.5 giflib-4.1.3 gimp-2.2.11 gimp-print-4.2.7 glib-1.2.10 glib-2.10.2 glibmm-2.8.3 gnumeric-1.6.3 gnupg-1.4.3 goffice-0.2.1 gtk+-1.2.10 gtk+-2.8.17 gtkmm-2.8.2 hdparm-6.1 hexedit-1.2.12 icecast-2.3.1 id3v2-0.1.11 imlib-1.9.15 imlib2-1.2.1 initng-0.6.7 ispell-3.3.02 joe-3.3 lcms-1.15 libao-0.8.6 libart-lgpl-2.3.17 libcroco-0.6.1 libebml-0.7.6 libexif-0.6.9 libfame-0.9.1 libglade-2.5.1 libglademm-2.6.1 libgnomecanvas-2.12.0 libgnomecups-0.2.2 libgnomeprint-2.12.1 libgnomeprintui-2.12.1 libgsf-1.14.0 libIDL-0.8.6 libjpeg-6b libmatroska-0.8.0 libmikmod-3.1.11 libmng-1.0.8 libogg-1.1.3 libpng-1.2.8 librsvg-2.14.3 libsigc++-2.0.17 libtheora-1.0alpha5 libusb-0.1.10a libvorbis-1.1.2 libxml2-2.6.24 libxslt-1.1.15 links-2.1pre21 Linux-PAM-0.99.3.0 mkvtoolnix-1.6.5 mp3info-0.8.4 mp3check-0.8.0 nasm-0.98.39 nss-3.11 openssl-0.9.8 p7zip-4.33 paco-1.9.6 pango-1.12.2 pciutils-2.2.1 pcre-6.6 pkg-config-0.20 popt-1.7 ppp-2.4.3 SDL-1.2.9 sox-12.17.8 speex-1.0.5 subversion-1.3.1 thunderbird-1.5.0.4 tiff-3.8.2 unzip-5.52 VisualBoyAdvance-1.7.2 vorbis-tools-1.1.1 wget-1.10.2 wine-0.9.14 wxGTK-2.6.2 xchat-2.4.3 xine-lib-1.1.1 xine-ui-0.99.4 XML-Parser-2.34 xmms-1.2.10 Xorg-6.9.0 zip-2.31 I got a failure while installing ncurses-5.5, tic complains about an undefined symbol. So I'm using 5.4, until I had time to check that. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: LFS Editors and Release Mgr
Dan Nicholson escribió: On 7/9/06, Jim Gifford <[EMAIL PROTECTED]> wrote: Why not make CLFS the main LFS? The two-stage build method has been proven and tested over years for the most popular arch, x86. I don't think it would be smart to throw that away yet. I have no problem increasing the visibility of CLFS as a viable option for the base system, but I'm not willing to abandon the two-stage build. It's been pretty quiet on LFS for a while, but I think with a release and some new activity getting the toolchain current, LFS will be just as relevant as ever. I agree with you Dan, but in the future the sysroot method will be a better option, I think. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Re: PROPOSAL -- new group to handle multi-project tasks
Randy McMurchy escribió: Gerard Beekmans wrote these words on 05/30/06 10:42 CST: The bootscripts were brought up as well as something this new group could take on. The udev problem we're trying to fix is different than the bootscript problem so people might have some issues with putting both tasks in the same group. DJ and Nathan for bootscripts. I am for everything you say, except the bootscripts. A BLFS bootscript is just as much a part of a package's instructions as any other part. BLFS editors need to be able to update/create bootscripts to do their job. BLFS bootscripts are different than the others. They are part of an individual package. The other bootscripts are all installed at once to the "system". Huge difference. I think LFS and CLFS can be merged, but BLFS is a completely different thing. Somebody has tried InitNG? -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
ncurses-5.5 doesn't install
tic complains about an undefined symbol (_nc_check_termtype2), when installing ncurses. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Test failures when compiling binutils with optimizations
Some test related to symbols will fail when compiling binutils with -O3 or -finline-functions (because of inlined functions), but binutils should be ok. It's a good idea to put a note about it, or maybe add instructions to patch the tests makefiles to use -fno-inline-functions. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Glibc will fail to compile in chapter 6
I noticed that in the chapter 6, glibc-2.3.6 will fail to compile, because the gcc specs patch is preventing glibc from including the kernel headers at /usr/include, adding the option --with-headers should solve the problem. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Stripping
I propose to change the stripping instructions from:
strip --strip-debug /tools/lib/*
strip --strip-unneeded /tools/{,s}bin/*
to:
strip -x /tools/lib/*
strip /tools/{,s}bin/*
"strip -x" strips all non-global symbols, not only debug ones,
which produces smaller libraries, and removing --strip-unneeded option
doesn't hurt anybody.
And maybe add a comment about the .note and .comment sections,
at least to say that they can be removed safely.
--
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
