- l4-hackers - example.com Site

hello/l4/fiasco.oc for realview pbxa9
by Alexis Fajardo Moya 27 Mar '14

27 Mar '14

Hello, recently I tried to run hello program in top of l4/fiasco.oc for exynos 4412 platform, but qemu doest'n have any platform that support it (at least my installed version) none empty machine collie Collie PDA (SA-1110) nuri Samsung NURI board (Exynos4210) smdkc210 Samsung SMDKC210 board (Exynos4210) connex Gumstix Connex (PXA255) verdex Gumstix Verdex (PXA270) highbank Calxeda Highbank (ECX-1000) midway Calxeda Midway (ECX-2000) integratorcp ARM Integrator/CP (ARM926EJ-S) (default) kzm ARM KZM Emulation Baseboard (ARM1136) mainstone Mainstone II (PXA27x) musicpal Marvell 88w8618 / MusicPal (ARM926EJ-S) n800 Nokia N800 tablet aka. RX-34 (OMAP2420) n810 Nokia N810 tablet aka. RX-44 (OMAP2420) sx1 Siemens SX1 (OMAP310) V2 sx1-v1 Siemens SX1 (OMAP310) V1 cheetah Palm Tungsten|E aka. Cheetah PDA (OMAP310) realview-eb ARM RealView Emulation Baseboard (ARM926EJ-S) realview-eb-mpcore ARM RealView Emulation Baseboard (ARM11MPCore) realview-pb-a8 ARM RealView Platform Baseboard for Cortex-A8 realview-pbx-a9 ARM RealView Platform Baseboard Explore for Cortex-A9 akita Akita PDA (PXA270) spitz Spitz PDA (PXA270) borzoi Borzoi PDA (PXA270) terrier Terrier PDA (PXA270) lm3s811evb Stellaris LM3S811EVB lm3s6965evb Stellaris LM3S6965EVB tosa Tosa PDA (PXA255) versatilepb ARM Versatile/PB (ARM926EJ-S) versatileab ARM Versatile/AB (ARM926EJ-S) vexpress-a9 ARM Versatile Express for Cortex-A9 vexpress-a15 ARM Versatile Express for Cortex-A15 xilinx-zynq-a9 Xilinx Zynq Platform Baseboard for Cortex-A9 z2 Zipit Z2 (PXA27x) For that reason I try to test hello/l4/fiasco.oc in realview pbxa9 platform. I create build directories for l4 and fiasco, configure and compile them for realview pbxa9 platform. Then I execute the following command: ~/build/l4.realview$ make -j2 elfimage E=hello MODULE_SEARCH_PATH=~/build/foc.realview SYSTEM_TARGET=/usr/local/arm-2012.03/bin/arm-none-linux-gnueabi- and this was the result: make[1]: se ingresa al directorio «/home/afmoya/src/l4» =========== Updating RAM_BASE for platform rv_pbx to 0x70000000 ========= ... Regenerating RAM_BASE settings make[2]: se ingresa al directorio «/home/afmoya/src/l4/pkg/sigma0/server/src» make[2]: atención: el servidor de tareas no está disponible: se utilizará -j1. Añada `+' a la regla padre del make. make[3]: se ingresa al directorio «/home/afmoya/build/l4.realview/pkg/sigma0/server/src/OBJ-arm_armv7a-l4f» ... Building Makefile.inc make[3]: se sale del directorio «/home/afmoya/build/l4.realview/pkg/sigma0/server/src/OBJ-arm_armv7a-l4f» make[3]: se ingresa al directorio «/home/afmoya/build/l4.realview/pkg/sigma0/server/src/OBJ-arm_armv7a-l4f» ==> Linking sigma0 ==> "sigma0" built ==> Installing sigma0 to local build-tree make[3]: se sale del directorio «/home/afmoya/build/l4.realview/pkg/sigma0/server/src/OBJ-arm_armv7a-l4f» make[2]: se sale del directorio «/home/afmoya/src/l4/pkg/sigma0/server/src» make[2]: se ingresa al directorio «/home/afmoya/src/l4/pkg/moe/server/src» make[2]: atención: el servidor de tareas no está disponible: se utilizará -j1. Añada `+' a la regla padre del make. make[3]: se ingresa al directorio «/home/afmoya/build/l4.realview/pkg/moe/server/src/OBJ-arm_armv7a-l4f» ... Building Makefile.inc make[3]: se sale del directorio «/home/afmoya/build/l4.realview/pkg/moe/server/src/OBJ-arm_armv7a-l4f» make[3]: se ingresa al directorio «/home/afmoya/build/l4.realview/pkg/moe/server/src/OBJ-arm_armv7a-l4f» ==> Linking moe ==> "moe" built ==> Installing moe to local build-tree make[3]: se sale del directorio «/home/afmoya/build/l4.realview/pkg/moe/server/src/OBJ-arm_armv7a-l4f» make[2]: se sale del directorio «/home/afmoya/src/l4/pkg/moe/server/src» make[2]: se ingresa al directorio «/home/afmoya/src/l4/pkg/bootstrap/server/src» make[3]: se ingresa al directorio «/home/afmoya/build/l4.realview/pkg/bootstrap/server/src/OBJ-arm_armv7a» ... Building .general.d ... Building Makefile.inc Building entry "hello". Merging image /home/afmoya/build/foc.realview/fiasco to mod00 [344kB] Merging image /home/afmoya/build/l4.realview/bin/arm_armv7a/l4f/sigma0 to mod01 [340kB] Merging image /home/afmoya/build/l4.realview/bin/arm_armv7a/l4f/moe to mod02 [2396kB] Merging image /home/afmoya/build/l4.realview/bin/arm_armv7a/l4f/l4re to mod03 [1227kB] Merging image /home/afmoya/build/l4.realview/bin/arm_armv7a/l4f/hello to mod04 [887kB] make[3]: se sale del directorio «/home/afmoya/build/l4.realview/pkg/bootstrap/server/src/OBJ-arm_armv7a» make[3]: se ingresa al directorio «/home/afmoya/build/l4.realview/pkg/bootstrap/server/src/OBJ-arm_armv7a» ... Generating bootstrap.ld ... Compiling exec.o ... Compiling module.o ... Compiling ARCH-arm/crt0.o ... Compiling region.o ... Compiling startup.o ... Compiling init_kip_v2.o ... Compiling init_kip_v4.o ... Compiling libc_support+.o ... Compiling patch.o ... Compiling koptions.o ... Compiling platform/rv.o ... Compiling loader_mbi.o ... Compiling ARCH-arm/reboot.o ==> Linking bootstrap.elf ==> Post-processing bootstrap.elf ==> "bootstrap.elf" built ==> Installing bootstrap.elf in image directory ==> Installing bootstrap.elf to local build-tree ==> Installing bootstrap_hello to local build-tree ==> Installing bootstrap_hello.elf to local build-tree ==> Installing bootstrap_hello in image directory ==> Installing bootstrap_hello.elf in image directory Image size(s) in bytes: bootstrap_hello.elf: 971420 Start address: 0x71000000 --> Build-Nr: 2 make[3]: se sale del directorio «/home/afmoya/build/l4.realview/pkg/bootstrap/server/src/OBJ-arm_armv7a» make[2]: se sale del directorio «/home/afmoya/src/l4/pkg/bootstrap/server/src» make[1]: se sale del directorio «/home/afmoya/src/l4» After that I execute: ~/build/l4.realview/images$ qemu-system-arm -M realview-pbx-a9 -kernel bootstrap_hello.elf -m 256 -serial mon:stdio -nographic this was the result of the execution: oss: Could not initialize DAC oss: Failed to open `/dev/dsp' oss: Reason: No such file or directory oss: Could not initialize DAC oss: Failed to open `/dev/dsp' oss: Reason: No such file or directory audio: Failed to create voice `lm4549.out' L4 Bootstrapper Build: #2 mié mar 26 13:02:05 CDT 2014, 4.6.3 Scanning up to 256 MB RAM Memory size is 256MB (70000000 - 7fffffff) RAM: 0000000070000000 - 000000007fffffff: 262144kB Total RAM: 256MB mod04: 710cb000-710e4580: hello mod03: 710b1000-710ca460: l4re mod02: 71077000-710b062c: moe mod01: 7106d000-71076378: sigma0 mod00: 71017000-7106cedc: fiasco Moving up to 5 modules behind 71100000 moving module 00 { 71017000-7106cedb } -> { 711ce000-71223edb } [351964] moving module 03 { 710b1000-710ca45f } -> { 71100000-7111945f } [103520] moving module 04 { 710cb000-710e457f } -> { 71100000-7111957f } [103808] ...and that was all. No "Hello world". After that I kill the qemu process. What 's the problem with my procedure? Sorry for the huge amount of lines of my email. Best regards! Alexis.

2 6

Have you modified L4 Kernel command line ?
by Taeung 27 Mar '14

27 Mar '14

Hello, L4 hackers :-) May I ask you a question about L4 linux ? Have you modified ' l4x_rd=rom/ramdisk-arm.rd root=1:0 ramdisk_size=4000 init=/bin/sh' in L4 kernel command ? I want to change it , as if the part of original linux's kernel command line is "root=/dev/mmcblk1p1 rootwait rw" Because I want to assign 'root' to '/dev/mmcblk1p1' or '/dev/mtdblock0' instead of 'rom/ramdisk-arm.rd'. To what I want is achieved, how to modify L4 linux kernel command line ?

2 1

Compiling L4Linux for Pandaboard ARM Error
by Erry Pradana Darajati 25 Mar '14

25 Mar '14

Hello L4 Hackers, i am doing a project that involve L4 and Pandaboard Rev.B1 So far, i am successful to boot Fiasco with L4Re in it. And now, i'm moving on to boot the L4Linux to Pandaboard I'm following the tuts from wiki tudos quickstart. But when i'm try to *make* the L4Linux, it gave me an error like this : =============================================== GEN /media/chiqiVstorage/l4resnap/src/l4linux/TArev2/Makefile CHK include/generated/uapi/linux/version.h CC scripts/mod/empty.o HOSTCC scripts/mod/mk_elfconfig gcc: error: unrecognized argument in option '-mabi=aapcs-linux' gcc: note: valid arguments to '-mabi=' are: ms sysv CC scripts/mod/devicetable-offsets.s gcc: error: unrecognized command line option '-mno-thumb-interwork' make[3]: *** [scripts/mod/empty.o] Error 1 make[3]: *** Waiting for unfinished jobs.... gcc: error: unrecognized argument in option '-mabi=aapcs-linux' gcc: note: valid arguments to '-mabi=' are: ms sysv gcc: error: unrecognized command line option '-mno-thumb-interwork' make[3]: *** [scripts/mod/devicetable-offsets.s] Error 1 make[2]: *** [scripts/mod] Error 2 make[1]: *** [scripts] Error 2 make[1]: *** Waiting for unfinished jobs.... make: *** [sub-make] Error 2 ================================================================ Is this error caused by my GCC ? i am using GCC from the default repository of ubuntu 13.10 And also, when i'm try to *make mrproper *it gave me error : ERROR: Cannot determine L4 source directory Please properly set the path to your L4 build directory in the configuration menu. But i'm sure i set in the wrong path Is there any solution about my problems above, I am appreciating any help and answer Thank you L4 Hackers -- *Regards,* *Erry Pradana Darajati | Student of Computer Engineering* *Politeknik Elektronika Negeri Surabaya / EEPIS, Indonesia*

2 1

Can use other root file system instead of ramdisk(ramdisk-arm.rd) ?
by Taeung 25 Mar '14

25 Mar '14

Hello , L4 hackers !!:-) Would you mind if I ask you other questions ? I've done booting "L4 linux + L4/Fiasco" on Freescale imx6q_sabre_sd. I'm using ramdisk of which name is 'ramdisk-arm.rd'. I've got the file from URL (os.inf.tu-dresden.de/download/ramdisk-arm.rd). But it is not enough to install and run X-window. There is not enough package. I really want to use 'apt-get' on L4 Linux. 1. Can is what I want possible on L4 linux ? 2. Can use other root file system instead of ramdisk(ramdisk-arm.rd) ? 3. Whatever packages which is put in a ramdisk, is it no matter ? If I put ubuntu packages in a ramdisk, is it fine? - Taeung -

2 1

Is wrong my kernel command line, on this situation ? (Booting L4 linux)
by Taeung 24 Mar '14

24 Mar '14

Hello, every l4-hacker. I've succeeded booting "Hello world + L4/Fiasco" on Freescale i.MX6q_sabre_sd because big help of Admed, Tim, Matthias. And then, I'm trying to boot "L4 Linux + L4/Fiasco" on my board. I've configured for L4 linux build. Such as.. |System type -> ARM system type (L4 Linux) Build type (Use ARMv7) | And then, I've successfully build it. I've did "make uimage E='L4Linux ARM' ..." ,so I've got uimage. And I try to boot it ,using 'uImage' and 'u-boot.bin'. But I did not boot it on my board. It was stoped at "l4linux | panic: going to sleep forever, bye" ==================== ... (eclipsis) Trying to unpack rootfs image as initramfs... rootfs image is not initramfs (junk in compressed archive); looks like an initrd INITRD: Freeing memory. l4x: Checks passed. NetWinder Floating Point Emulator V0.97 (double precision) msgmni has been set to 126 io scheduler noop registered io scheduler deadline registered io scheduler cfq registered (default) L4 serial driver ttyLv0 at MMIO 0x1 (irq = 211, base_baud = 230400) is a L4 l4ser_shm: L4 shared mem serial driver l4cdds: No name given, not starting. brd: module loaded l4bdds: No name given, not starting. mousedev: PS/2 mouse device common for all mice TCP: cubic registered NET: Registered protocol family 17 Minicom2.6.2RAMDISK: ext2 filesystem found at block 0 RAMDISK: Loading 3072KiB [1 disk] into ram disk... done. EXT4-fs (ram0): couldn't mount as ext3 due to feature incompatibilities EXT4-fs (ram0): mounting ext2 file system using the ext4 subsystem EXT4-fs (ram0): mounted filesystem without journal. Opts: (null) VFS: Mounted root (ext2 filesystem) readonly on device 1:0. Freeing unused kernel memory: 112K (022ea000 - 02306000) Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000004 CPU: 0 PID: 1 Comm: sh Not tainted 3.13.0-l4-svn46 #1 [<0200b6ec>] (unwind_backtrace+0x0/0xf8) from [<02005308>] (show_stack+0x10/0x14) [<02005308>] (show_stack+0x10/0x14) from [<0222a17c>] (panic+0x78/0x1c0) [<0222a17c>] (panic+0x78/0x1c0) from [<02017d28>] (do_exit+0x724/0x7cc) [<02017d28>] (do_exit+0x724/0x7cc) from [<02017ea0>] (do_group_exit+0x40/0xa0) [<02017ea0>] (do_group_exit+0x40/0xa0) from [<02021754>] (get_signal_to_deliver+0x1a4/0x474) [<02021754>] (get_signal_to_deliver+0x1a4/0x474) from [<02006738>] (do_signal+0x78/0x3d4) [<02006738>] (do_signal+0x78/0x3d4) from [<02006b08>] (do_work_pending+0x74/0xcc) [<02006b08>] (do_work_pending+0x74/0xcc) from [<02009cf4>] (l4x_vcpu_entry_c+0x18ec/0x1e3c) [<02009cf4>] (l4x_vcpu_entry_c+0x18ec/0x1e3c) from [<00094e9c>] (0x94e9c) panic: going to sleep forever, bye l4linux | panic: going to sleep forever, bye ==================== I guess that the problem is somthing about ramdisk(root file system) and kernel command line might be wrong. Referentially I've used ramdisk file(ramdisk-arm.rd) and my kernel command line "setenv bootargs_mmc 'setenv bootargs ${bootargs} root=/dev/mmcblk1p1 rootwait rw video=mxcfb1:dev=ldb,LDB-XGA,if=RGB666 video=mxcfb0:dev=hdmi,1920x1080M@60,if=RGB24'". How do you think about that? Is wrong my kernel command line,on this situation ? - Taeung -

2 1

Re: l4linux
by Taeung 24 Mar '14

24 Mar '14

Hello, Tim I'm glad. Tim Your help save me !!! Thank you very much !!!! :-) I've succeed booting "L4 Linux + L4/Fiasco" on my board(Freescale i.MX6 Quad SABRE SD). After I readjustmy thread for booting, I will write document for booting it on my board. The document might help L4/Fiasco beginner or not, I'll share the document for their. Thank you again !! - Taeung - On 03/24/2014 07:33 PM, Tim Komarov wrote: > Hello, TaeWoong! > I've returned after the weekend. So, my congratulations! I guess that yo don't need my environment anymore :-) > > I suppose that your configuration of L4Linux is wrong. > > Try to build L4Linux like that (arm_defconfig, not your manual configuration): > > $mkdir builddir > > $ make O=pathtobuilddir L4ARCH=arm CROSS_COMPILER=arm-linux- arm_defconfig > > Change path to l4re-tree: > $ make O=pathtobuilddir L4ARCH=arm CROSS_COMPILER=arm-linux- menuconfig > > Build: > $ make O=pathtobuilddir L4ARCH=arm CROSS_COMPILER=arm-linux- -j5 > > It may help you!

1 0

Can you tell me function name that means 'Entry Point: 0x11000000' ?
by Taeung 24 Mar '14

24 Mar '14

Hello, l4 hacker. I'm trying to boot "Hello world / L4/ Fiasco" on Freescale i.MX6 Quad SABRE SD board(Cortex-A9, ARMv7). I'm using 'u-boot' and 'uImage'. So, i build uimage by "make uimage E=hello ~...". And i've try to booting it on Freescale i.MX6q_sabre_sd board. But i did not it because stoped at "Starting kernel ..." !! MMC read: dev # 2, block # 2048, count 8192 ... 8192 blocks read: OK ## Booting kernel from Legacy Image at 10800000 ... Image Name: L4 Image #2 Image Type: ARM Linux Kernel Image (uncompressed) Data Size: 868352 Bytes = 848 kB Load Address: 11000000 Entry Point: 11000000 Loading Kernel Image ... OK OK Starting kernel ... uimage's file name is "bootstrap_hello.uimage". "bootstrap_hello.uimage" has information such as "bootstrap_hello.uimage: u-boot legacy uImage, L4 Image #2, Linux/ARM, OS Kernel Image (Not compressed), 868352 bytes, Thu Mar 20 13:14:42 2014, Load Address: 0x11000000, Entry Point: 0x11000000, Header CRC: 0xBEA251F3, Data CRC: 0xA0217B7D" I have a question. Can you tell me function name that means 'Entry Point: 0x11000000' ?

3 5

Fwd: Can you tell me function name that means 'Entry Point: 0x11000000' ?
by TaeWoong Song 23 Mar '14

23 Mar '14

Begin forwarded message: > From: Taeung <treeze.taeung(a)gmail.com> > Subject: Re: Can you tell me function name that means 'Entry Point: 0x11000000' ? > Date: March 23, 2014 12:07:38 PM GMT+09:00 > To: Matthias Lange <matthias.lange(a)kernkonzept.com> > > Thank you , Matthias :-) > > Fortunately , Ahmed in L4 hackers give me a patch about UART on Freescale imx6q sabre_sd. > > But i question , again. > > I have known that entry point mean the address of the execution of the image. > But i really want to know that.. > > What is concreate function name mapping the address in L4 re source code ? > > - Taeung - > > On 03/22/2014 07:02 AM, Matthias Lange wrote: >> Hi, >> >>> On 21 Mar 2014, at 11:33, Taeung <treeze.taeung(a)gmail.com> wrote: >>> >>> Hello, l4 hacker. >>> >>> >>> I'm trying to boot "Hello world / L4/ Fiasco" on Freescale i.MX6 Quad SABRE SD board(Cortex-A9, ARMv7). >>> >>> I'm using 'u-boot' and 'uImage'. >>> So, i build uimage by "make uimage E=hello ~...". >>> And i've try to booting it on Freescale i.MX6q_sabre_sd board. >>> But i did not it because stoped at "Starting kernel ..." !! >> Probably you selected the wrong UART, so the system is not printing something on the serial line. Make sure you have the right UART driver for your platform and make sure you have selected the right one. Usually there are more than one UARTs and only one is connected to you serial line. You can also check Linux to see which UART it uses. >> >>> MMC read: dev # 2, block # 2048, count 8192 ... 8192 blocks read: OK >>> ## Booting kernel from Legacy Image at 10800000 ... >>> Image Name: L4 Image #2 >>> Image Type: ARM Linux Kernel Image (uncompressed) >>> Data Size: 868352 Bytes = 848 kB >>> Load Address: 11000000 >>> Entry Point: 11000000 >>> Loading Kernel Image ... OK >>> OK >>> >>> Starting kernel ... >>> >>> >>> uimage's file name is "bootstrap_hello.uimage". >>> "bootstrap_hello.uimage" has information such as >>> "bootstrap_hello.uimage: u-boot legacy uImage, L4 Image #2, Linux/ARM, OS Kernel Image (Not compressed), 868352 bytes, >>> Thu Mar 20 13:14:42 2014, Load Address: 0x11000000, Entry Point: 0x11000000, Header CRC: 0xBEA251F3, Data CRC: 0xA0217B7D" >>> >>> I have a question. >>> >>> Can you tell me function name that means 'Entry Point: 0x11000000' ? >> In the uimage format the entry point is defined as the address where uboot should start, as the name implies, the execution of the image. >> >> >> Matthias. >

2 1

compile l4re for arm
by Waldo 21 Mar '14

21 Mar '14

Hi, i am running the hello world example for the arm realview (exynos4412), and i don't how create a elfimage to run the example, i compile the fiasco and l4re for arm, but the elfimage is not generated.

3 3

Google Summer of Code - DEADLINE IS TODAY!
by Crypto Stick 21 Mar '14

21 Mar '14

Hi! Crypto Stick is an open source secure USB device for data encryption and authentication. OpenCoin is an open source system for anonymous cash. We (both projects) are aiming to collaborate on building a L4-based server system which would act as an eCash issuer server. Now, we got accepted at the upcoming Google Summer of Code 2014, which means that Google will sponsor a student working on this project during the summer with $5500. If this sounds interesting for you, please get in touch with us! Also, please share the word with fellow students. THE APPLICATION DEADLINE IS TODAY 19:00 UTC! Sorry for this last minute announcement. More details are available at: https://www.assembla.com/spaces/cryptostick/wiki/Ideas Best regards, Jan

1 0

L4 Fiasco's scheduling in Multi-Core
by Ganis Zulfa Santoso 20 Mar '14

20 Mar '14

Hi l4-hackers, Suppose we have 2 cores (core 0 and core 1) and we have 2 tasks (task A and task B). And I assign all threads of task A to run in core 0 and all threads of task B to run in core 1. How L4 Fiasco's scheduler manages this? Will the tasks run together or they have to switch from one to another? >From what I understand, the tasks don't run together regardless which core they are located, is it correct? Thank you. Best regards, Ganis

3 2

compiling L4RE for ARM
by Waldo 18 Mar '14

18 Mar '14

hello, i am a beginner in this topic, but you can tell me how i can compile L4re for ARM ?

2 1

I was not able to run "Hello world" on my board(imx6).
by taeung 18 Mar '14

18 Mar '14

Hello, L4 hackers. I have questions about booting "Hello world" on myboard. Please help me. I'm try to boot "Hello world + L4re + fiasco" on my board of which name is 'freescale i.MX6 Quad SABRE SD'(Cortex-A9,ARMv7). I built L4re and Fiasco. So, i got L4 build tree and fiasco image. Although i have built uimage file by "make uimage" and and i try to "make qemu", i was not able to run "hello world". (For reference,i have run original linux of which name is Linux/ARM, OS Kernel ,Linux-3.0.35-2508-g54750ff on my board(Freescale imx6q_sabre_sd)) <1. Result of the try booting on my board after "make uimage"> ======================================================================== MMC read: dev # 2, block # 2048, count 8192 ... 8192 blocks read: OK ## Booting kernel from Legacy Image at 10800000 ... Image Name: L4 Image #3 Image Type: ARM Linux Kernel Image (uncompressed) Data Size: 1224704 Bytes = 1.2 MB Load Address: 11000000 Entry Point: 11000000 Verifying Checksum ... OK Loading Kernel Image ... OK OK Starting kernel ... ======================================================================== <1-1. My Question > Why stop at "Starting kernel..." ? Is wrong my environment variables in u-boot ? (If it is the problem, how fix u-boot environment variables from '1-4'? ) Or, my toolchains(linaro gcc 4.6 (64bit) ) are a wrong choice ? <1-2. Fiasco : What i configured for myboard > ======================================================================== Target configuration ---> Architecture (ARM processor family) Platform (Freescale i.MX) Freescale i.MX (i.MX6) CPU (ARM Cortex-A9 CPU) [ ] Enable alignment check (NEW) [ ]Execution Model (Standard mode) [ ] Enable the deprecated 'swp' instruction (NEW) [*] Enable L2 Cache [ ] Enable FPU co-processor [ ] Enable CPU errata workarounds (NEW) Kernel options ---> [*] Enable multi processor support (4) Maximal supported number of CPUs (NEW) [*] Fine-grained CPU time Scheduler (Fixed priority scheduler) ======================================================================== <1-3. L4 re : What i configured for myboard> ======================================================================== Target Architecture 3. ARM architecture (BUILD_ARCH_arm) (NEW) CPU variant 9. ARMv7A type CPU (CPU_ARM_ARMV7A) (NEW) Platform Selection 7. Freescale i.MX6 (PLATFORM_TYPE_imx6) (NEW) Automatically determine internal gcc preprocessor names (INT_CPP_NAME_SWITCH) [Y/n] (NEW) Y Automatically determine internal ld names (INT_LD_NAME_SWITCH) [Y/n] (NEW) Y Strip binaries on install (BID_STRIP_PROGS) [N/y/?] (NEW) N GCC: Omit Frame-pointers (BID_GCC_OMIT_FP) [N/y/?] (NEW) N Build doc directories (BID_BUILD_DOC) [N/y/?] (NEW) N Enable RELEASE flag (RELEASE_MODE) [N/y/?] (NEW) N ======================================================================== <1-4. U-boot : what i set environment variable for my board> (u-boot-mx6q-sabresd.bin is supported from freescale home page.) ======================================================================== MX6Q SABRESD U-Boot > setenv loadaddr 0x10800000 MX6Q SABRESD U-Boot > setenv bootargs_base 'setenv bootargs console=ttymxc0,115200' MX6Q SABRESD U-Boot > setenv bootargs_mmc 'setenv bootargs ${bootargs} root=/dev/mmcblk1p1 rootwait rw video=mxcfb1:dev=ldb,LDB-XGA,if=RGB666 video=mxcfb0:dev=hdmi,1920x1080M@60,if=RGB24' MX6Q SABRESD U-Boot > setenv bootcmd_mmc 'run bootargs_base bootargs_mmc;mmc dev 2;mmc read ${loadaddr} 0x800 0x2000;bootm' MX6Q SABRESD U-Boot > setenv bootcmd 'run bootcmd_mmc' MX6Q SABRESD U-Boot > saveenv Saving Environment to MMC... Writing to MMC(2)... done MX6Q SABRESD U-Boot > run bootcmd ======================================================================== <2. Result of the second try "make qemu"> ======================================================================== root@taeung-MS-7798:/path/to/obj2/tudos# make qemu E=hello-cfg MODULE_SEARCH_PATH=/path/to/obj2/fiasco/ make[1]: Entering directory `/home/taeung/Hello_test_l4/src/l4' ... Generating bootstrap.ld Building entry "hello-cfg". Merging image /path/to/obj2/fiasco//fiasco to mod00 [360kB] Merging image /path/to/obj2/tudos/bin/arm_armv7a/l4f/sigma0 to mod01 [351kB] Merging image /path/to/obj2/tudos/bin/arm_armv7a/l4f/moe to mod02 [2365kB] Merging image /path/to/obj2/tudos/bin/arm_armv7a/l4f/l4re to mod03 [1228kB] Merging image /path/to/obj2/tudos/bin/arm_armv7a/l4f/ned to mod04 [2387kB] Merging image /path/to/obj2/tudos/bin/arm_armv7a/l4f/hello.cfg to mod05 [1kB] Merging image /path/to/obj2/tudos/bin/arm_armv7a/l4f/hello to mod06 [896kB] ... Compiling exec.o ... Compiling module.o ... Compiling ARCH-arm/crt0.o ... Compiling region.o ... Compiling startup.o ... Compiling init_kip_v2.o ... Compiling init_kip_v4.o ... Compiling libc_support+.o ... Compiling patch.o ... Compiling koptions.o ... Compiling platform/imx.o ... Compiling loader_mbi.o ... Compiling ARCH-arm/reboot.o ==> Linking bootstrap.elf ==> Post-processing bootstrap.elf ==> "bootstrap.elf" built ==> Installing bootstrap.elf in image directory ==> Installing bootstrap_hello-cfg in image directory ==> Installing bootstrap_hello-cfg.elf in image directory Image size(s) in bytes: bootstrap_hello-cfg.elf: 1257856 Start address: 0x11000000 --> Build-Nr: 4 ==> Installing bootstrap.elf to local build-tree ==> Installing bootstrap_hello-cfg to local build-tree ==> Installing bootstrap_hello-cfg.elf to local build-tree QEmu-cmd: qemu-system-arm -kernel /path/to/obj2/tudos/images/bootstrap.elf qemu: fatal: Trying to execute code outside RAM or ROM at 0x11000000 R00=00000000 R01=00000000 R02=00000000 R03=00000000 R04=00000000 R05=00000000 R06=00000000 R07=00000000 R08=00000000 R09=00000000 R10=00000000 R11=00000000 R12=00000000 R13=00000000 R14=00000000 R15=11000000 PSR=400001d3 -Z-- A svc32 s00=00000000 s01=00000000 d00=0000000000000000 s02=00000000 s03=00000000 d01=0000000000000000 s04=00000000 s05=00000000 d02=0000000000000000 s06=00000000 s07=00000000 d03=0000000000000000 s08=00000000 s09=00000000 d04=0000000000000000 s10=00000000 s11=00000000 d05=0000000000000000 s12=00000000 s13=00000000 d06=0000000000000000 s14=00000000 s15=00000000 d07=0000000000000000 s16=00000000 s17=00000000 d08=0000000000000000 s18=00000000 s19=00000000 d09=0000000000000000 s20=00000000 s21=00000000 d10=0000000000000000 s22=00000000 s23=00000000 d11=0000000000000000 s24=00000000 s25=00000000 d12=0000000000000000 s26=00000000 s27=00000000 d13=0000000000000000 s28=00000000 s29=00000000 d14=0000000000000000 s30=00000000 s31=00000000 d15=0000000000000000 FPSCR: 00000000 bash: line 3: 5158 Aborted (core dumped) $qemu -kernel /path/to/obj2/tudos/images/bootstrap.elf make[1]: *** [qemu] Error 134 make[1]: Leaving directory `/home/taeung/Hello_test_l4/src/l4' make: *** [qemu] Error 2 ======================================================================== <2-1. My Question > Qemu support cpu such as cortex-A9 and ARMv7. But it not support mahine information for Freescale i.MX6 quad sabre sd. Because this is the reason, Can't boot "Helloworld + L4re+Fiasco" by qemu ? Or Is there other reason ? Thank you, if you help whatever. /Regards, Taeung Song | Student of Computer Engineering Hansung University./

1 0

In Freescale i.MX6q_sabre_sd,The problem booting Helloworld + L4 + Fiasco
by taeung 17 Mar '14

17 Mar '14

Hello, L4 hackers. I have one question. please help me.. I have failed booting 'L4 Linux + L4 + Fiasco' on my board(Freescale i.MX6 Quad SABRE SD : Cortex-A9 , ARMv7) So, i'm try to boot 'Helloword + L4 + Fiasco' on my board Such as.. |$ svn cathttp://svn.tudos.org/repos/oc/tudos/trunk/repomgr | perl - inithttp://svn.tudos.org/repos/oc/tudos fiasco l4re | 1. L4 Re : configure and build (Success) |||==========================================================================| $ mkdir -p /path/to/obj/tudos |$ make O=/path/to/obj/tudos oldconfig $ make O=/path/to/obj/tudos -j2| |========================================================================== | |2. Fiasco : configure and build (Success)| || |========================================================================== $ cd /somedir/tudos/src/kernel/fiasco $ make BUILDDIR=/path/to/obj/fiasco $ cd /path/to/obj/fiasco $ make menuconfig $ make ========================================================================== 3. But,when i try to 'make uimage' or 'make qemu', error is happened. such as.. ========================================================================== root@taeung-MS-7798:/path/to/obj2/tudos# make uimage E=hello-cfg MODULE_SEARCH_PATH=/path/to/obj2/fiasco/ make[1]: Entering directory `/home/taeung/Hello_test_l4/src/l4' =========== Updating RAM_BASE for platform imx6 to 0x10000000 ========= ... Regenerating RAM_BASE settings Library dependencies missing: libc_minimal cxx_util cxx_io cxx_base cxx_io_kdebug l4sys-direct l4sys in directory '/home/taeung/Hello_test_l4/src/l4/pkg/sigma0/server/src'; aborting. /home/taeung/Hello_test_l4/src/l4/mk/binary.inc:348: *** Cflags dependencies missing: libc_minimal cxx_util cxx_io cxx_base cxx_io_kdebug l4sys-direct l4sys in directory '/home/taeung/Hello_test_l4/src/l4/pkg/sigma0/server/src'; aborting.. Stop. make[2]: *** [/path/to/obj2/tudos/pkg/sigma0/server/src/OBJ-arm_armv7a-l4f] Error 2 make[1]: *** [check_and_adjust_ram_base] Error 2 ========================================================================== |<< My Question >> Although i search of '|Library dependencies missing: libc_minimal cxx_util cxx_io cxx_base cxx_io_kdebug l4sys-direct l4sys|' in google.com, i could'nt get information about this error. I don't know why the problem is happened updating |'RAM_BASE for platform imx6 to 0x10000000'. | || Please any help.. /Regards, Taeung Song | Student of Computer Engineering Hansung University./

2 2

L4Linux + L4re + Fiasco to boot in Freescale i.MX6q_sabre_sd
by taeung 17 Mar '14

17 Mar '14

Hello L4 hackers. I have two question (My question 1, 2 ). please help me. I'm trying to boot 'L4Linux + L4re + Fiasco' in Freescale i.MX6 Quad SABRE SD board (Cortex-A8, ARMv7-A). Such as.. 0. Get Source and install required package. |$ cd /somedir/tudos $ svn cat http://svn.tudos.org/repos/oc/tudos/trunk/repomgr | perl - init http://svn.tudos.org/repos/oc/tudos fiasco l4re l4linux_requirements| 1. L4 Re : configure and build (Success) |||==========================================================================| $ mkdir -p /path/to/obj/tudos |$ make O=/path/to/obj/tudos oldconfig $ make O=/path/to/obj/tudos -j2| |==========================================================================| 2. Fiasco : configure and build (Success) |==========================================================================| ||$ cd /somedir/tudos/src/kernel/fiasco |$ make BUILDDIR=/path/to/obj/fiasco $ cd /path/to/obj/fiasco $ make menuconfig $ make| |==========================================================================| 3. L4 linux : configure and build. (Failure) (Here, i have trouble to build l4 linux after configure according to my board ) |==========================================================================| |$ cd /somedir/tudos $ svn co http://svn.tudos.org/repos/oc/l4linux/trunk l4linux $ cd l4linux $ mkdir /path/to/obj/l4linux $ make L4ARCH=arm CROSS_COMPILE=arm-linux- O=/path/to/obj/l4linux arm_defconfig # In the menu, in "L4Linux configuration", set "L4 tree build directory" to /path/to/obj/tudos $ make L4ARCH=arm CROSS_COMPILE=arm-linux- O=/path/to/obj/l4linux menuconfig ||==========================================================================| To configure according to my board(Freescale i.MX6q_sabre_sd), I was able to choice two kinds. <The first > |System type -> ARM system type (L4 Linux) Build type (Use ARMv7) | <Result of the first > 1) Sucess build L4 linux and i got 'vmlinuz.arm' 2) 'make uImage' according to module.list |========================================================================== entry L4Linux_xs_2 kernel fiasco -nokdb roottask moe rom/l4linux.cfg module l4linux.cfg module l4re module ned module vmlinuz.arm module ramdisk-arm.rd| |==========================================================================| 3) i make SD Card bootable by u-boot.bin and bootstrap.uImage 4) Booting L4Linux + L4re + Fiasco in Freescale i.MX6q_sabre_sd But, i have problem to boot it. =========================================================================== U-Boot 2009.08 (Aug 16 2013 - 12:04:20) CPU: Freescale i.MX6 family TO1.2 at 792 MHz Thermal sensor with ratio = 180 Temperature: 44 C, calibration data 0x5764ba7d mx6q pll1: 792MHz mx6q pll2: 528MHz mx6q pll3: 480MHz mx6q pll8: 50MHz ipg clock : 66000000Hz ipg per clock : 66000000Hz uart clock : 80000000Hz cspi clock : 60000000Hz ahb clock : 132000000Hz axi clock : 264000000Hz emi_slow clock: 132000000Hz ddr clock : 528000000Hz usdhc1 clock : 198000000Hz usdhc2 clock : 198000000Hz usdhc3 clock : 198000000Hz usdhc4 clock : 198000000Hz nfc clock : 24000000Hz Board: i.MX6Q-SABRESD: unknown-board Board: 0x63012 [POR ] Boot Device: SD I2C: ready DRAM: 1 GB MMC: FSL_USDHC: 0,FSL_USDHC: 1,FSL_USDHC: 2,FSL_USDHC: 3 In: serial Out: serial Err: serial i2c: I2C3 SDA is low, start i2c recovery... I2C3 Recovery success Found PFUZE100! deviceid=10,revid=11 Net: got MAC address from IIM: 00:04:9f:02:70:b5 FEC0 [PRIME] Hit any key to stop autoboot: 0 mmc2 is current device MMC read: dev # 2, block # 2048, count 17408 ... 17408 blocks read: OK ## Booting kernel from Legacy Image at 10800000 ... Image Name: L4 Image #7 Image Type: ARM Linux Kernel Image (uncompressed) Data Size: 7680000 Bytes = 7.3 MB Load Address: 11000000 Entry Point: 11000000 Verifying Checksum ... OK Loading Kernel Image ... OK OK Starting kernel ... ========================================= <My question 1.> Why hang at 'Starting kernel ...' ? Setting environment variable and bootcmd , i have thought the setting is no problem. This kernel image is wrong? <The Second > |System type ---> ARM system type (Allow multiple platforms to be selected)| Multiple platform selection ---> ARMv7 based platforms (Cortex-A, PJ4, Scorpion, Krait) Build type Freescale i.MX family Freescale i.MX support ---> i.MX6 Quad/DualLite support <Result of the second > i could not build l4 linux because.. GEN /path/to/obj/l4linux/Makefile scripts/kconfig/conf --silentoldconfig Kconfig ERROR: CONFIG_ARM_PATCH_PHYS_VIRT must not be enabled. make[1]: *** [check_linux_config] Error 1 make: *** [sub-make] Error 2 <My question 2.> Why can not build L4 linux according to this config ? and What is CONFIG_ARM_PATCH_PHYS_VIRT ? Please any help.. /Regards, Taeung Song | Student of Computer Engineering Hansung University. /

2 1

Re: Problem in Linking program
by Irvanda Kurniadi 13 Mar '14

13 Mar '14

Case closed. I found that the __assert_fail is not supported in uclibc. Then, I tried to change __assert_fail with __assert and it works. But I'm still curious about using glibc in L4re. I try to compile my program using libgcc but it also doesn't work. Can it cause any problem if I use glibc compiler than any of requires_libs using uclibc compiler? just wondering irvanda

2 1

Problem in Linking program
by Irvanda Kurniadi 12 Mar '14

12 Mar '14

Hi, I'm trying to build my program in L4/Fiasco. This program requires libtiff, MPI, and libstep library. I've ported those all libraries. But, when I built my program, I got a problem in linking program. The compilation just work fine but when entering the linking part, I got this error message: ==> Linking sobel-step ........ main-sobel-V4.cpp_processed.o: In function `main': /home/l4re-mpich/src/l4/pkg/sobel-step/server/src/main-sobel-V4.cpp_processed.c:57: undefined reference to `__assert_fail' /home/l4re-mpich/src/l4/pkg/sobel-step/server/src/main-sobel-V4.cpp_processed.c:56: undefined reference to `__assert_fail' ......... This "__assert_fail" should be available in libstep.so and I've put the LDFLAGS = -lstep on my makefile. Here is my Makefiel anyway: PKGDIR ?= ../.. L4DIR ?= $(PKGDIR)/../.. SYSTEMS = x86-l4f amd64-l4f include $(L4DIR)/mk/Makeconf ifneq ($(SYSTEM),) # check whether used gcc has the omp.h file available OMP_H_PATH := $(shell $(CC) -print-file-name=include/omp.h) ifeq ($(strip $(OMP_H_PATH)),include/omp.h) $(info [32m $(CC) does not have omp.h header file available, skipping.[0m) else TARGET = $(if $(filter 4.2 4.3 4.4 4.5 4.6 4.7,$(GCCVERSION)),sobel-step) REQUIRES_LIBS = libgomp libtiff libmpich libmpich_f77sup libstep libc libc_support_misc endif endif SRC_C = main-sobel-V4.cpp_processed.c sobel-omp-V4.cpp_processed.c pictureInterface.c util.c CFLAGS += -fopenmp -I. -g -Wall LDFLAGS += -rpath -ltiff -rpath -lstep PRIVATE_INCDIR = $(SRC_DIR) vpath %.c $(SRC_DIR) include $(L4DIR)/mk/prog.mk Previously I compiled the program and there are so much missing shared libraries and the debug message suggested to add the -rpath. I updated my LDFLAGS by adding the -rpath. Here is the error message before I add the LDFLAGS with -rpath: ==> Linking sobel-step .... ld: warning: libl4z.so, needed by /home/l4re-mpich/obj/l4/x86/lib/x86_586/l4f/libtiff.so, not found (try using -rpath or -rpath-link) ld: warning: libgfortran.so, needed by /home/l4re-mpich/obj/l4/x86/lib/x86_586/l4f/libstep.so, not found (try using -rpath or -rpath-link) ....... ld: final link failed: Bad value I'm not sure whether this is causing another problem by adding the -rpath. I mean, Is it the proper way to fix the missing shared library? Because the "__asser_fail" is supposed to be found in the libstep.so but I still got the undefined reference error message. So perhaps there are several things to wrap up this linking problem. If you by any chance find any misplace or missing configuration in my makefile or you have any idea how to fix it please let me know. Thanks Regards, Irvanda

1 0

seL4 what exactly is in the kernel
by Taylor Bioniks 11 Mar '14

11 Mar '14

Like all L4 kernels, it provides basic mechanisms which must be in the kernel for security reasons. These include low-level interrupt and exception handlers (which essentially convert these events into IPC messages to user-level handlers), a notion of address spaces into which you can map frames (basically an abstraction of page tables), a notion of kernel-scheduled (and context-switched) threads, a notion of a protection domain (capability tree) and a communication and synchronisation mechanism (IPC endpoints). Other L4 kernels have similar abstractions. Please explain what you mean by notion? Do you mean that it has code to deal with the user mode memory management and such? What is different in seL4 is that all memory management is done at user level. After booting up and allocating and initialising its static data structures (scheduling queues, the kernel stack and a few globals) the kernel sets up an initial user address space with a single thread, and hands it the rights to all remaining memory (in the form of "Untyped" capabilities). When performing an operation that requires kernel data structures (eg creating a thread requires creating a thread control block, TCB) the user has to provide these, by "retyping" some Untyped memory into a kernel object type. This makes it kernel memory (inaccessible to userland), but the owner of that memory can always revoke it (revert it to Untuped, thus nuking the kernel object). so is the single kernel stack in it's own address space or is it somehow mapped or shared with other address spaces? There's a single kernel stack, which is in use whenever kernel code executes. The terms "kernel mode" or "user mode" make no sense for data - they refer to the processor mode that is active when particular code is executing. However, the kernel stack is, of course, a kernel object, only accessible in kernel mode. While the kernel stack (like the scheduler queues) is somewhat special, in that it is allocated by the kernel at boot time, in terms of accessibility it's no different from other "kernel objects", such as PTs or TCBs, that are allocated by userland. so kernel objects are stored in this user mode kernel stack? Okay, so the kernel never actually has a stack in kernel mode, and it never allocates space for itself does this mean that the kernel does not grow? And if so does that eliminate the requirement that I've seen on X86 machines where they have 1GB for the kernel and 3GB for the user programs Final thing is I've been pointed to the university site a few times and I've been told there are slides with audio on there, however, I can never find them even after mirroring the site and looking through it. Where might I find these?

2 1

Problem when booting L4 Fiasco/Re
by Sören Jonsson 11 Mar '14

11 Mar '14

Hello, I have now successfully compiled the L4 Fiasco.OC and your runtime environment. I made the hello.iso bootable image as well, and made an attempt to run it under VMWare workstation. My problem is that it don't work. It stops during boot, with the following information: I had to provide two images, in order to capture all information. The information above "Total RAM: 255MB" disapers very fast otherwise. Is this enough information to explain what is happening, or can I provide more information using other methods? With my best regards, Sören Jonsson Sweden

2 1

seL4 what exactly is in the kernel
by Taylor Bioniks 08 Mar '14

08 Mar '14

What does the seL4 kernel actually provide in the kernel itself? I've read through documentation and it seems they just stop before they state exactly what the seL4 kernel does. So what I'd like to know is what features and components are in the kernel and what is outside of it? Does the kernel have a single kernel stack or multiple kernel stacks and is/are it/they in kernel mode or user mode? Are there any good diagrams that show how seL4 works with its user space components? Thanks in advance

2 1

Panic with l4Linux
by Ismael BARAKAT 07 Mar '14

07 Mar '14

Hello, I'm working on Fiasco, l4Re and l4Linux. I build the uImage for an arm architecture. When I boot it, I have a "panic: going to sleep forever, bye". Can someone help I don't know what caused the problem. This is the information I have on my console when I boot: L4 Bootstrapper Build: #10 Thu Mar 6 11:11:19 CET 2014, 4.7.3 Scanning up to 128 MB RAM Memory size is 128MB (80000000 - 87ffffff) RAM: 0000000080000000 - 0000000087ffffff: 131072kB Total RAM: 128MB mod07: 813f7000-816f7000: ramdisk-arm.rd mod06: 81125000-813f6480: vmlinuz.arm mod05: 81124000-81124189: l4Linux.cfg mod04: 810d1000-81123bc0: ned mod03: 810b7000-810d0458: l4re mod02: 81085000-810b6624: moe mod01: 8107b000-8108433c: sigma0 mod00: 81015000-8107a3a8: fiasco Moving up to 8 modules behind 81100000 moving module 00 { 81015000-8107a3a7 } -> { 817e2000-818473a7 } [414632] moving module 01 { 8107b000-8108433b } -> { 81848000-8185133b } [37692] moving module 02 { 81085000-810b6623 } -> { 81852000-81883623 } [202276] moving module 03 { 810b7000-810d0457 } -> { 816f7000-81710457 } [103512] moving module 04 { 810d1000-81123bbf } -> { 81711000-81763bbf } [338880] moving module 05 { 81124000-81124188 } -> { 81100000-81100188 } [393] moving module 06 { 81125000-813f647f } -> { 81101000-813d247f } [2954368] moving module 07 { 813f7000-816f6fff } -> { 813d3000-816d2fff } [3145728] moving module 03 { 816f7000-81710457 } -> { 816d3000-816ec457 } [103512] moving module 04 { 81711000-81763bbf } -> { 816ed000-8173fbbf } [338880] Scanning fiasco Scanning sigma0 Scanning moe rom/l4Linux.cfg Relocated mbi to [0x8100e000-0x8100e12a] Loading fiasco Loading sigma0 Loading moe find kernel info page... found kernel info page at 0x80002000 Regions of list 'regions' [ 80001000, 80001b3f] { b40} Kern fiasco [ 80002000, 80072fff] { 71000} Kern fiasco [ 80090000, 800965fb] { 65fc} Sigma0 sigma0 [ 80098000, 8009e17b] { 617c} Sigma0 sigma0 [ 80140000, 8016cacb] { 2cacc} Root moe [ 80170000, 80186f0f] { 16f10} Root moe [ 81000000, 810143f3] { 143f4} Boot bootstrap [ 8100e000, 8100e227] { 228} Root Multiboot info [ 81100000, 8173fbbf] { 63fbc0} Root Module API Version: (87) experimental Sigma0 config ip:80090100 sp:81013de4 Roottask config ip:80140260 sp:00000000 Starting kernel fiasco at 80001300 Hello from Startup::stage2 Cache config: ON ID_PFR[01]: 00001131 00000011 ID_[DA]FR0: 00010400 00000000 ID_MMFR[04]: 01100003 20000000 01202000 00000211 Welcome to Fiasco.OC (arm)! L4/Fiasco.OC arm microkernel (C) 1998-2013 TU Dresden Rev: rUnversioned director compiled with gcc 4.7.3 for Beagleboard [] Build: #1 Wed Mar 5 17:15:13 CET 2014 Calibrating timer loop... done. MDB: use page size: 20 MDB: use page size: 12 SIGMA0: Hello! KIP @ 80002000 allocated 4KB for maintenance structures SIGMA0: Dump of all resource maps RAM:------------------------ [0:80000000;80000fff] [0:80073000;8008ffff] [0:80097000;80097fff] [0:8009f000;8013ffff] [4:80140000;8016cfff] [0:8016d000;8016ffff] [4:80170000;80186fff] [0:80187000;8100dfff] [4:8100e000;8100efff] [0:8100f000;810fffff] [4:81100000;8173ffff] [0:81740000;86ffffff] IOMEM:---------------------- [0:0;7fffffff] [0:88000000;ffffffff] MOE: Hello world MOE: found 107500 KByte free memory MOE: found RAM from 80000000 to 87000000 MOE: allocated 112 KByte for the page array @0x80073000 MOE: virtual user address space [0-bfffffff] MOE: rom name space cap -> [C:501000] BOOTFS: [816d3000-816ec458] [C:503000] l4re BOOTFS: [816ed000-8173fbc0] [C:504000] ned BOOTFS: [81100000-81100189] [C:505000] l4Linux.cfg BOOTFS: [81101000-813d2480] [C:506000] vmlinuz.arm BOOTFS: [813d3000-816d3000] [C:507000] ramdisk-arm.rd MOE: cmdline: moe rom/l4Linux.cfg MOE: Starting: rom/ned rom/l4Linux.cfg MOE: loading 'rom/ned' Ned says: Hi World! Ned: loading file: 'rom/l4Linux.cfg' l4Linux | libio: Warning: Query of 'vbus' failed! l4Linux | PH 0 (t: 1) offs=00008000 vaddr=01000000 vend=01331b1c l4Linux | phys=01000000 ephys=01331b1c l4Linux | f_sz=002b33d4 memsz=00331b1c flgs=rwx l4Linux | PH 1 (t: 4) offs=002bb3b0 vaddr=012b33b0 vend=012b33d4 l4Linux | phys=012b33b0 ephys=012b33d4 l4Linux | f_sz=00000024 memsz=00000024 flgs=r-x l4Linux | PH 2 (t: 1685382481) offs=00000000 vaddr=00000000 vend=00000000 l4Linux | phys=00000000 ephys=00000000 l4Linux | f_sz=00000000 memsz=00000000 flgs=rwx l4Linux | Starting binary at 0x10002a0, argc=7 argv=0x80007f8c *argv=0xb1007ff0 argv0=rom/vmlinuz.arm l4Linux | External resolver is at 0xa8000794 l4Linux | ======> L4Linux starting... <======== l4Linux | Linux version 3.7.0-l4-svn30 (juliens@ubuntu) (gcc version 4.7.2 (Sourcery CodeBench Lite 2012.09-64) ) #1 Wed May 29 10:25:09 CEST 2013 l4Linux | Binary name: rom/vmlinuz.arm l4Linux | This is an AEABI build. l4Linux | Linux kernel command line (6 args): mem=64M console=ttyLv0 l4x_rd=rom/ramdisk-arm.rd root=1:0 ramdisk_size=4000 init=/bin/sh l4Linux | CPU mapping (l:p)[1]: 0:0 l4Linux | Image: 01000000 - 01400000 [4096 KiB]. l4Linux | Areas: Text: 01000000 - 01290000 [2624kB] (a bit longer) l4Linux | Data: 01290000 - 012aeee0 [123kB] l4Linux | Initdata: 01273000 - 0128ed7c [111kB] l4Linux | BSS: 012b33d4 - 01331b1c [505kB] l4Linux | Device scan: l4Linux | Device scan done. l4Linux | l4lx_thread_create: Created thread 413 (cpu0) (u:b3000e00, v:b3000c00, sp:01291fa4) l4Linux | main thread will be 413 l4Linux | l4x_register_pointer_section: addr = 01000000 size = 3350528 l4Linux | section-with-init: virt: 0x1000000 to 0x1331b1b [3270 KiB] l4Linux | section-with-init: Phys: 0x80241000 to 0x80572b1c, Size: 3349276 l4Linux | Main thread running, waiting... l4Linux | L4x: Setting superpages for main memory l4Linux | Main memory size: 64MB l4Linux | Main memory: virt: 0x1400000 to 0x53fffff [65536 KiB] l4Linux | Main memory: Phys: 0x81800000 to 0x85800000, Size: 67108864 l4Linux | Filling lower ptabs... l4Linux | Done (220 entries). l4Linux | l4x_register_pointer_section: addr = 01000000 size = 3350528 l4Linux | text: virt: 0x1000000 to 0x1331b1b [3270 KiB] l4Linux | text: Phys: 0x80241000 to 0x80572b1c, Size: 3349276 l4Linux | l4x_rd_path: rom/ramdisk-arm.rd l4Linux | Loading: rom/ramdisk-arm.rd l4Linux | INITRD: Size of RAMdisk is 3072KiB l4Linux | RAMdisk from 0d400000 to 0d700000 [3072KiB] l4Linux | panic: going to sleep forever, bye Kinds regards and thanks for the help, Ismaël Barakat

2 1

Seting Library search path
by Sören Jonsson 06 Mar '14

06 Mar '14

Hello, I am interested in L4 Fiasco as a virtualization environment. I am currently attempting to compile the L4 runtime environment, and I am getting the following error message: === Building package "bootstrap" === ==> Linking bootstrap32.elf /usr/bin/ld: skipping incompatible /usr/lib/gcc/x86_64-linux-gnu/4.7/libgcc.a when searching for -lgcc /usr/bin/ld: cannot find -lgcc collect2: error: ld returned 1 exit status make[5]: *** [bootstrap32.elf] Error 1 make[4]: *** [/home/osj/l4re/build/pkg/bootstrap/server/src/OBJ-amd64_amdfam10] Error 2 make[3]: *** [src] Error 2 make[2]: *** [server] Error 2 make[1]: *** bootstrap] Error 2 make: *** [pkg] Error 2 I have examined the problem, and can see that I have a library /usr/lib/gcc/x86_64-linux-gnu/4.7.2, as well as the /usr/lib/gcc/x86_64-linux-gnu/4.7. Can the problem be that the linker is attempting to link to a dated version of the library, and what is the best way to resolve this problem? Best regards, Sören Jonsson Sweden

2 1

Running application can't create a file
by Irvanda Kurniadi 03 Mar '14

03 Mar '14

Hi, Recently I'm using libtiff library on L4/Fiasco. I can use some function from that library. But when I'm trying to create an image by using libtiff function, I got this error message: TIFFOpen: output.tif: Cannot open. Could not open output.tif for writing MOE: task 1561539 exited with 42 Here is my configuration in module.list entry tiff_test2 roottask moe --init=rom/tiff_test2 module l4re module tiff_test2 I compile and run it over linux and it works properly. This application is supposed to create the output file and then open it by TIFFOpen function. So, I think there is a trick to launch this tiff application over L4/fiasco, so the kernel can permit the application to create a file. But, it's only my assumption and I'm not really sure about this. If by any chance you know something about it, please let me know. Thanks regards, Irvanda

2 3

Framebuffer in L4Re for Raspberry PI
by Peter Hanzel 03 Mar '14

03 Mar '14

Hello. I have compiled Fiasco.OC and L4Re for Raspberry PI. It boots ok with UART output. Now I want to add framebuffer device and output to monitor. Is this a work in progress or is it already implemented? or to be implemented? Thanks.

4 4

Genode 14.02 enables VirtualBox on NOVA
by Christian Helmuth 28 Feb '14

28 Feb '14

Today we released a new version of Genode, which, beside many other improvements, adds support for VirtualBox as VMM on NOVA. For years, we addressed virtualization in Genode beginning with OKLinux on the OKL4 kernel. Later, we added support for L4Linux on Fiasco.OC and the Vancouver/Seoul VMM on NOVA. All these solutions had strong points in distinct usage scenarios but we always missed support for a wide range of guest OSes, guest-host integration features, ease of use, active development, and, finally, support for MS Windows guests. Therefore, we are happy to have enabled VirtualBox on the NOVA microhypervisor now, which brings all these features and (due to NOVA's virtualization support) integrates perfectly with the architecture of Genode. Besides VirtualBox, the main features and improvements of the current release are: - Base framework - Improved stack management - Storage - NetBSD file systems ported via rump kernels - Revised block-driver framework - New block cache - New FUSE-based NTFS-3g support - Ported e2fsprogs to the Noux runtime - Low-level OS infrastructure - New pseudo file system as tracing front end - Unified interfaces for graphics - New status-reporting mechanism - Nitpicker support for dynamic screen resizing - Libraries and applications - Extended Noux runtime with POSIX signals - Dynamic ROM service - Platforms - Basic support ODROID XU More details about version 14.02 can be found in the release documentation here http://genode.org/documentation/release-notes/14.02 Best regards -- Christian Helmuth Genode Labs http://www.genode-labs.com/ · http://genode.org/ https://twitter.com/GenodeLabs · /ˈdʒiː.nəʊd/ Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth

1 0