Re: [lfs-support] 8.1-systemd Ch. 8.3 Linux-4.12.7: Error during make?
On 09/18/2017 11:43 PM, Hans Malissa wrote: Hi, I’m working my way through LFS 8.1-systemd. I got all the way to 8.3.1 Linux-4.12.7 without any problem, but now I’m stuck during kernel compilation. After 'make mrproper’ and ‘make menuconfig', I run ‘make'. Compilation runs ok for quite some time, but eventually it fails with a message: ... Setup is 17532 bytes (padded to 17920 bytes). System is 5281 kB CRC ed1061fd Kernel: arch/x86/boot/bzImage is ready (#1) Building modules, stage 2. MODPOST 3534 modules ERROR: “__devm_regmap_init_mmio_clk” [sound/soc/codecs/snd-soc-msm8916-digital.ko] undefined! make[1]: *** [scripts/Makefile.modpost:91: __modpost] Error 1 make: *** [Makefile:1217: modules] Error 2 I’ve tried a few times, starting from a clean source every time. I don’t understand what’s going wrong here. I was assuming that the .config created with ‘make menuconfig’ should work in any case? I initially created .config by copying the config-file from the host distribution (debian), running ‘make oldconfig’, and then adjusting the .config to my liking, so it should work on my system. Most likely there are lots and lots of modules selected that I don’t really need. This should not be a problem, but if it’s some obscure module that I don’t need which causes the compile error, then I’d gladly remove it. Thanks a lot, Your host distro's configuration has almost no chance of doing anything more than panicking on you as it will most likely be unable to find the rootfs. You took the time to build LFS, then took a shortcut on the kernel? :-/ It'll take some time and effort, but start from a clean config and put in at least a little time exploring the various options, you'll be surprised how much you can learn poking around menuconfig. If really short on time, try local{yes,mod}config then tweak that, make sure that SATA and EXT are built-in (Y), not modules (M). No reason to build 2000 modules when you need ~40...especially for the Snapdragon 410. Well, that is, unless you've just managed to LFS your low-end smartphone, in which case, I'll need to ask you a few questions! :-) --DJ -- http://lists.linuxfromscratch.org/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page Do not top post on this list. A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail? http://en.wikipedia.org/wiki/Posting_style
[lfs-support] 8.1-systemd Ch. 8.3 Linux-4.12.7: Error during make?
Hi, I’m working my way through LFS 8.1-systemd. I got all the way to 8.3.1 Linux-4.12.7 without any problem, but now I’m stuck during kernel compilation. After 'make mrproper’ and ‘make menuconfig', I run ‘make'. Compilation runs ok for quite some time, but eventually it fails with a message: ... Setup is 17532 bytes (padded to 17920 bytes). System is 5281 kB CRC ed1061fd Kernel: arch/x86/boot/bzImage is ready (#1) Building modules, stage 2. MODPOST 3534 modules ERROR: “__devm_regmap_init_mmio_clk” [sound/soc/codecs/snd-soc-msm8916-digital.ko] undefined! make[1]: *** [scripts/Makefile.modpost:91: __modpost] Error 1 make: *** [Makefile:1217: modules] Error 2 I’ve tried a few times, starting from a clean source every time. I don’t understand what’s going wrong here. I was assuming that the .config created with ‘make menuconfig’ should work in any case? I initially created .config by copying the config-file from the host distribution (debian), running ‘make oldconfig’, and then adjusting the .config to my liking, so it should work on my system. Most likely there are lots and lots of modules selected that I don’t really need. This should not be a problem, but if it’s some obscure module that I don’t need which causes the compile error, then I’d gladly remove it. Thanks a lot, Hans -- http://lists.linuxfromscratch.org/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page Do not top post on this list. A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail? http://en.wikipedia.org/wiki/Posting_style
Re: [lfs-support] Ensuring a generic x86_64 build
On Mon, Sep 18, 2017 at 06:02:49PM -0500, Bruce Dubbs wrote: > > AFAIK, this is it only package in LFS, but it may not be for packages in > BLFS. There is at least one package in multimedia where you need to disable > some optimizations manually. For example gst-libav requires yasm. Indeed, > any package that uses yasm/nasm is a candidate for problems as those are > required for processor instructions not available in C or C++. > I suspect that many multimedia packages in BLFS (and outside it) will do their best to use as many optimizations as possible - that is based on old experience on ppc where the optimizations differed between what apple called a G3 and a G4 (750 and 7455, I think). Generally, multimedia packages seem to turn everything up to 11 - on x86_64 most of this is probably common to current CPUs. The original post mentioned -mtune=skylake : looking at 'info gcc' it was not obvious to me whether that would be set to what is fastest on skylake, or alternatively to using things that might not be available on older CPUs (basically, the -march | -mtune variation: I forget which is which, and after looking at the maze of different very similar CPU names for recent AMD I was not minded to research this ;) But if people need to detune packages, I'm sure that binary distributions (particularly fedora, debian-unstable) might sometimes show examples. And maybe gentoo from the source distros. Of course, fedora tends to ignore many AV packages (licensing/patent fears). Or, just build it on the good machine, ship the full install (including headers) to the lesser machine, and rebuild there if it breaks. Final thought: if you really need the optimizations, you probably need a more potent CPU :-( I used a K6-2 when I first came here, and I went through all the likely CFLAGS, but none made a noticeable improvement. ĸen -- Truth, in front of her huge walk-in wardrobe, selected black leather boots with stiletto heels for such a barefaced truth. - Unseen Academicals -- http://lists.linuxfromscratch.org/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page Do not top post on this list. A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail? http://en.wikipedia.org/wiki/Posting_style
Re: [lfs-support] Ensuring a generic x86_64 build
Paul Rogers wrote: I noticed this note about GMP: "The default settings of GMP produce libraries optimized for the host processor." Using the default configure settings in the summary of build options I see "Host type: skylake-pc-linux-gnu". I then see gcc is using "-mtune=skylake" when compiling the GMP library. Now if I use configfsf.guess as mentioned in the LFS book, it will build with "Host type: x86_64-pc-linux-gnu". So I've made a note to recompile GMP when I move it to the older system. Isn't that a PITA? But check out the "fat" option. Not really. Do you want performance or portability? The user decides. In this case the default is performance which I, for one, like. Beyond GMP, is anyone aware of other LFS packages which, when using the default compilation options, could lead to binaries that are tightly coupled to the host build's processor? I'd like my build to be as generic as possible so that it can run on any other x86_64 system. I do too, just got into trouble over this and had to rebuild. Apparently that is the only package (so far!) to be so presumptuous. AFAIK, this is it only package in LFS, but it may not be for packages in BLFS. There is at least one package in multimedia where you need to disable some optimizations manually. For example gst-libav requires yasm. Indeed, any package that uses yasm/nasm is a candidate for problems as those are required for processor instructions not available in C or C++. -- Bruce -- http://lists.linuxfromscratch.org/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page Do not top post on this list. A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail? http://en.wikipedia.org/wiki/Posting_style
Re: [lfs-support] Ensuring a generic x86_64 build
> I noticed this note about GMP: > > "The default settings of GMP produce libraries optimized for the host > processor." > > Using the default configure settings in the summary of build options I > see "Host type: skylake-pc-linux-gnu". I then see gcc is using > "-mtune=skylake" when compiling the GMP library. Now if I use > configfsf.guess as mentioned in the LFS book, it will build with "Host > type: x86_64-pc-linux-gnu". So I've made a note to recompile GMP when > I move it to the older system. Isn't that a PITA? But check out the "fat" option. > > Beyond GMP, is anyone aware of other LFS packages which, when using > the default compilation options, could lead to binaries that are > tightly coupled to the host build's processor? I'd like my build to > be as generic as possible so that it can run on any other x86_64 > system. I do too, just got into trouble over this and had to rebuild. Apparently that is the only package (so far!) to be so presumptuous. -- http://lists.linuxfromscratch.org/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page Do not top post on this list. A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail? http://en.wikipedia.org/wiki/Posting_style
Re: [lfs-support] 5.3. General Compilation Instructions - I made a mistake
Am 2017-09-16 um 19:07 schrieb Pierre Labastie: On 16/09/2017 12:07, David Brandl wrote: Hi! I made a mistake in 5.3 - there you have to check, if your $LFS shows to the right directory. As I mounted my partition with lfs to the /home/~user~ directory I overread the part in 4.4 where |LFS=/mnt/lfs | So now my $LFS is targeting to /mnt/lfs - that´s not right. What I´m thinking about is, that I´m user lfs with his default shell and so on. What if i change the $LFS now - can I get problems at this point? Thanks a lot! Whatever the location of your partition with lfs, in this example /home//lfs if I understand correctly, that location should be contained into the LFS variable: --- LFS=/home//lfs --- Note that this setting should be made permananent by putting the same instruction in lfs user's ".bashrc" file. Also, always check that it is set as well when running instructions as root; use "su" or "sudo", not "su -". As for your question: you can change LFS now, but be sure you have put nothing into /mnt/lfs. If you have, you should move or copy the content of /mnt/lfs to /home//lfs. And do not forget to change .bashrc. Pierre Hi! Thanks - yes I understand, that I should have everything on /home/~user~/lfs, where I have the partition mounted. But my fear was, that I failed something, as I work as an other user in the shell now. Where is the difference now? When I open the .bashrc with nano, the file is empty. How do I open it correctly? Thanks a lot! | | -- http://lists.linuxfromscratch.org/listinfo/lfs-support FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page Do not top post on this list. A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail? http://en.wikipedia.org/wiki/Posting_style