July 2015 - openSUSE ARM - openSUSE Mailing Lists

[opensuse-arm] Attempt to run KIWI on unsupported architecture 'armv8l'
by Dirk Müller 12 Apr '17

12 Apr '17

Hi, this is a known OBS bug, asked the admins to deploy a fix shortly. Greetings, Dirk -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

3 3

[opensuse-arm] Samsung Chromebook current tumbleweed images
by Misha Komarovskiy 02 Sep '15

02 Sep '15

Hello, Just tried latest 316/318 Tumbleweed images for Samsung Chromebook and noticed that u-boot wont auto load boot.scr now and output contain alot errors about extlinux pxeboot. If i load boot.scr manualy, with ext2load mmc 1:2 0x50000000 boot.scr and then source 0x50000000 boot starts fine. Also lcd screen stays black after kernel start, probably some drm module still missing. --- Best Regards, Misha Komarovskiy zombahatgmaildotcom -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

4 32

[opensuse-arm] Support for device-tree overlays
by Stefan Bruens 08 Aug '15

08 Aug '15

Hi everyone, as the RaspberryPI kernel is nowadays "configured" by the device tree, there are some issues for the current openSUSE images. By default the dtbs as shipped disable any optional hardware, which is a good thing. On the other hand things like SPI are no longer working. Raspbian tackles this with so called overlays and some help from the binary start.elf bootloader: 1. start.elf reads the dtb from the fat boot partition a) a file defined in config.txt, "device_tree=foo.dtb" b) a hardcoded hardware specific default file, "bcm2708-rpi-<variant>.dtb" 2. if the config.txt specifies "dtoverlay=foo-overlay" 1. the boot loader reads "overlays/foo-overlay.dtb" from the fat partition 2. the overlay is patched into the flattened device tree 3. the boot loader puts the device tree at address 0x100 (override with "device_tree_address=0xabcd1234" 4. the boot loader loads the specified kernel As openSUSE does not directly jump into the kernel, but starts u-boot, this does not work here. Issues: 1. u-boot does not have overlay support 2. u-boot loads its own dtb from storage, e.g. bcm2835-rpi-b.dtb 3. The u-boot boot.scr is loaded at address 0, and overwrites the boot loader provided dtb. The 3rd issue can be worked around with a "device_tree_address=0x1000" in config.txt. I tried to replace the "loadfdt" function in the boot.scr with a "fdt move 0x1000 0x2000000". Although the move succeeds, the kernel start crashes afterwards: ---- 4443840 bytes read in 1041 ms (4.1 MiB/s) 6946368 bytes read in 1403 ms (4.7 MiB/s) Copying FDT ... Kernel image @ 0x1000000 [ 0x000000 - 0x43cec0 ] data abort pc : [<17f63d68>] lr : [<17f644e0>] reloc pc : [<0001dd68>] lr : [<0001e4e0>] sp : 17b48a50 ip : 17b521e9 fp : 17f51c54 r10: 17f91354 r9 : 17b49ee8 r8 : 17f922bc r7 : 17f922b8 r6 : 17b521e0 r5 : 00000003 r4 : 33554432 r3 : 00000000 r2 : 00000010 r1 : 00000000 r0 : 33554432 Flags: nZCv IRQs off FIQs off Mode SVC_32 Resetting CPU ... ---- instead of: ---- 4443840 bytes read in 1268 ms (3.3 MiB/s) 6946368 bytes read in 1753 ms (3.8 MiB/s) 8127 bytes read in 56 ms (141.6 KiB/s) Kernel image @ 0x1000000 [ 0x000000 - 0x43cec0 ] ## Flattened Device Tree blob at 02000000 Booting using the fdt blob at 0x2000000 Using Device Tree in place at 02000000, end 02004fbe ---- So either this approach needs more work, or a completely different approach is needed to get overlays to work. Any ideas? Kind regards, Stefan -- Stefan Brüns / Bergstraße 21 / 52062 Aachen home: +49 241 53809034 mobile: +49 151 50412019 -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

5 9

[opensuse-arm] armv6 kernel update 4.1.3
by Michael Ströder 06 Aug '15

06 Aug '15

1 3

[opensuse-arm] Cubietruck 13.2 regression
by Andreas Färber 05 Aug '15

05 Aug '15

Hello, After upgrading 13.2 from kernel-lpae-3.16.7-7 to -11/-21 (why is it installing two new kernels btw?) I no longer have network. Diagnosing this is difficult because, as reported previously, the kernel does not provide any serial output to ttyS0. The LED lights up, which is done by my boot.local script, so it is coming up to some degree, just not far enough to login via ssh, and it's not reachable by ping either. I doubt this is just the dtb getting out of sync again? Originally it was a U-Boot 2014.10, but same with 2015.07 (and still working with a downgraded kernel/initrd). Regards, Andreas -- SUSE Linux GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GF: Felix Imendörffer, Jane Smithard, Dilip Upmanyu, Graham Norton; HRB 21284 (AG Nürnberg) -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

2 3

[opensuse-arm] Raspberry Pi 1 Model B+, No USB without b-plus dtb
by Evan McClain 20 Jul '15

20 Jul '15

Hi list, After rebasing my pi setup on a recent openSUSE image (since I was using raspberrypi branch kernels), it took me longer than I'm willing to admit to realize that the B+ needs to use the B+ dtb for things like USB to work. The kernel does boot with the rpi-b.dtb, but since USB isn't up the network devices aren't up and you can't plug in a keyboard to do anything about it. The ethernet MAC, serial, and revision still aren't being picked up, but this time I'm going with the ifcfg-eth0 change instead of switching kernels and editing the boot args. Have a lot of fun, Evan

3 2

[opensuse-arm] What repository is recommended for images for Raspberry Pi's?
by Freek de Kruijf 20 Jul '15

20 Jul '15

I am a bit confused about which repositories should be used for Raspberry Pi Tumbleweed images. Currently I have: for RPi1: http://download.opensuse.org/repositories/devel:/ARM:/Factory:/Contrib:/Ras… http://download.opensuse.org/repositories/devel:/ARM:/Factory:/Contrib:/Ras… for RPi2: http://download.opensuse.org/repositories/devel:/ARM:/Factory:/Contrib:/Ras… http://download.opensuse.org/repositories/devel:/ARM:/Factory:/Contrib:/Ras… -- fr.gr. member openSUSE Freek de Kruijf -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

1 1

[opensuse-arm] Build327.8 for rpi1 broken
by Michael Ströder 15 Jul '15

15 Jul '15

HI! openSUSE-Tumbleweed-ARM-JeOS-raspberrypi.armv6l-1.12.1-Build327.8 does not work. It soon runs into a "rebootException" after start. FWIW openSUSE-Tumbleweed-ARM-JeOS-raspberrypi.armv6l-1.12.1-Build327.6 seems to work. Could someone look at the difference(s)? Ciao, Michael.

2 1

[opensuse-arm] Buiding Libreoffice 5.0
by Bill Merriam 14 Jul '15

14 Jul '15

I have for many months been trying to build Libreoffice on ARM with no success. There is a working build of it on Debian so it must be possible. I suspect the first thing they did is remove a whole bunch of obscure options that are hardly every used. This would remove pre-requisite packages and reduce the build time. I am trying now to build LO 5.0. It runs for a day and then starts over again. Can someone explain why this never builds? I am going to look at the package on Debian and see what options they use and how they break it into sub-packages. Maybe I will try building the Debian package on the openSuse Build System to see if there is a problem with the build system. Any thoughts on this would be appreciated. Bill -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

4 4

[opensuse-arm] Corrupted files in recent Builds for RPi1
by Freek de Kruijf 13 Jul '15

13 Jul '15

When trying to build an updated package which was not available in the repository I needed the file: /usr/lib/perl5/5.20.1/armv6l-linux-thread-multi/Fcntl.pm which gives gives an error Right after putting the image on the SD card, I mounted the lxroot partition on /mnt/ll and got the following on the command: # ls -l /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux-thread-multi/ ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/Fcntl.pm: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/Config.pod: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/Config_git.pl: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/.packlist: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/Encode.pm: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/GDBM_File.pm: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/DynaLoader.pm: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/Config.pm: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/Config_heavy.pl: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/DB_File.pm: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/Errno.pm: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/Cwd.pm: Invoer-/uitvoerfout ls: kan geen toegang krijgen tot /mnt/ll/usr/lib/perl5/5.20.1/armv6l-linux- thread-multi/B.pm: Invoer-/uitvoerfout totaal 428 [....] -r--r--r-- 1 root root 20519 23 jun 17:39 encoding.pm -????????? ? ? ? ? ? Errno.pm -????????? ? ? ? ? ? Fcntl.pm drwxr-xr-x 3 root root 4096 9 jul 04:38 File [....] -- fr.gr. member openSUSE Freek de Kruijf -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

2 3

Re: [opensuse-arm] Experiences with Build317.3 for RPi2
by Dirk Müller 09 Jul '15

09 Jul '15

Hi, > Sorry for the slow response on this, I did not have time to look into this > recently. It appears that the issue is that the first boot does not detect > the MMC anymore, so it doesn't find the boot partition and goes into a > rescue reboot. I have reverted the commit that most likely is causing this > issue and am going to debug it further in the next day(s). hopefully the > next image build after build 317.6 should be working again. took slightly longer than anticipated since the kernel didn't build with gcc 5.x anymore, so I had to revert to gcc 4.8, but the image now works again, and I've moved all changes to :Staging and debug that image. Greetings, Dirk -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

1 0

[opensuse-arm] Cross-compiling for ARM
by Stefan Bruens 07 Jul '15

07 Jul '15

Hi, I tried to cross-compile some stuff for ARMv7hl on a x86_host, but have some issues. The installed compiler/binutils are: cross-armv7hl-gcc49-4.9.0+r211729-2.1.7.x86_64 cross-arm-binutils-2.25.0-215.1.x86_64 both from the devel:gcc OBS repository The first issue is the assembler (as) is not found by the compiler. According to "armv7hl-suse-linux-gnueabi-gcc-4.9 -v --print-search-dirs" it uses: /usr/lib64/gcc/armv7hl-suse-linux-gnueabi/4.9/ /usr/lib64/gcc/armv7hl-suse-linux-gnueabi/ /usr/lib64/gcc/armv7hl-suse-linux-gnueabi/4.9/../../../../armv7hl-suse-linux- gnueabi/bin/armv7hl-suse-linux-gnueabi/4.9/ /usr/lib64/gcc/armv7hl-suse-linux-gnueabi/4.9/../../../../armv7hl-suse-linux- gnueabi/bin/ and then falls back to $PATH. I can work around this by setting the GCC env var COMPILER_PATH to /usr/arm- suse-linux-gnueabi/bin/ , where the binutils are located. The problem here obviously is the different architecture in the gcc and binutils package, as the last search dir entry points to /usr/armv7hl-suse- linux-gnueabi/bin/ instead of /usr/arm-suse-linux-gnueabi/bin The second issue it hit is a missing "unwind.h". The x86_64 gcc provides one for its arch in /usr/lib64/gcc/x86_64-suse-linux/4.9/include/unwind.h, but the armv7hl gcc is missing it. Kind regards, Stefan -- Stefan Brüns / Bergstraße 21 / 52062 Aachen home: +49 241 53809034 mobile: +49 151 50412019 work: +49 2405 49936-424 -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

3 2

[opensuse-arm] Experiences with Build317.3 for RPi2
by Freek de Kruijf 05 Jul '15

05 Jul '15

Still the same problem as before. The system boots the openSUSE kernel, which apparently crashes. After which it boots again. The partitioning of the SD card does not seem to changed. -- fr.gr. member openSUSE Freek de Kruijf -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

1 0

[opensuse-arm] Build326.4 for RPi1 does not produce a proper ext4 partition
by Freek de Kruijf 04 Jul '15

04 Jul '15

I copied this image to my SD card and tried to mount the ext4 partition. Using parted showed that it it was not a proper ext4 partition. Did not know how to make it right without destroying the content. -- fr.gr. member openSUSE Freek de Kruijf -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

1 1

Re: [opensuse-arm] Experiences with Build325.4 for RPi1
by Freek de Kruijf 01 Jul '15

01 Jul '15

Op woensdag 1 juli 2015 13:53:47 schreef Alexander Graf: > On 30.06.15 17:43, Freek de Kruijf wrote: > > Op dinsdag 30 juni 2015 11:49:09 schreef u: > >> On 06/30/15 11:24, Freek de Kruijf wrote: > >>> Op dinsdag 30 juni 2015 10:57:31 schreef Alexander Graf: > >>>> On 06/30/15 10:55, Freek de Kruijf wrote: > >>>>> Earlier a second boot of a system generated with this build for > >>>>> Raspberry > >>>>> Pi 1B did not work. Now I have a system that works and can be booted > >>>>> again. > >>>>> > >>>>> However trying to start a module with yast (ncurses), I tried network, > >>>>> software repositories and software all end in the following error > >>>>> message: > >>>>> > >>>>> /sbin/yast: line 426: 19947 Segmentation fault $ybindir/y2base > >>>>> $module "$@" "$SELECTED_GUI" $Y2_GEOMETRY $Y2UI_ARGS > >>>>> > >>>>> However 19947 in the above message differs with the type of module > >>>>> called > >>>>> and with the locale used. > >>>>> > >>>>> The repository still does not contain the ruby 2.2 modules for > >>>>> libstorage. > >>>>> I build these modules, which went OK. I noticed gcc5 was used. > >>>>> > >>>>> I will continue with this build to configure the applications I use. > >>>> > >>>> Do you think you could gdb the y2base binary to figure out where > >>>> exactly > >>>> it segfaults? > >>> > >>> Alex, I have no experience with gdb. I know it is used to debug and one > >>> can > >>> set breakpoints and step through the code, but that's all I know. > >> > >> No worries, there's always a first time :). In this case, all we need is > >> to attach to the program and then do 'bt' when we see the segfault. > >> > >>> I tried to run yast in one console and did "gdb -p process-id" in > >>> another > >>> after installing yast2-core-debugger. However it likes to have > >>> yast2-core- > >>> debuginfo which is not available. After giving a c in gdb and selecting > >>> software in the ncurses console, gdb continues with the following > >>> output: > >>> > >>> (gdb) c > >>> Continuing. > >>> Detaching after fork from child process 20159. > >>> [Thread 0xb601f450 (LWP 20077) exited] > >>> [Thread 0xb631b450 (LWP 20076) exited] > >>> [Inferior 1 (process 20075) exited normally] > >> > >> This means that the process was doing just fine, no segfault. So maybe a > >> different PID was the one broken? Which PID did you attach to? According > >> to the log, "y2base" is the application that faults. > >> > >> If you can't find it, try to attach strace -f to the yast binary to > >> figure out which binary gets executed. The strace should also at least > >> print out which PID segfaulted. > > > > Attached you will find strace.2. I can't find anything about a segfault. > > Search for this line: > > [pid 25753] --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, > si_addr=0x10} --- > > That's where the segfault occurs :). I entered the following bug in bugzilla: https://bugzilla.opensuse.org/show_bug.cgi?id=936788 The problem is apparently in the package yast2-pkg-bindings. -- fr.gr. member openSUSE Freek de Kruijf -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

1 0

[opensuse-arm] Experiences with Build326.2 for RPi1 and Build316.1 for RPi2
by Freek de Kruijf 01 Jul '15

01 Jul '15

The Build for RPi1 works, but has the same problem as with the previous build when trying to use YaST with ncurses. The Build for RPi2 does not work. The same problem. After booting the openSUSE image the system boots from the start after some time and repeats this. So apparently the openSUSE image crashes. Inspecting the micro-SD card after this process shows no changes in the partitioning of this card. -- fr.gr. member openSUSE Freek de Kruijf -- To unsubscribe, e-mail: opensuse-arm+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-arm+owner(a)opensuse.org

1 0