- l4-hackers - example.com Site

demo cd question
by eric van tassell 17 Jan '11

17 Jan '11

Should the demo CD iso image from http://os.inf.tu-dresden.de/L4/LinuxOnL4/download.shtml boot in VirtualBox? It definitely does not on my alienware m17x hardware. I do have an older machine I can try. -- - evt (Eric van Tassell) twitter: evt_texelsoft linked-in: linkedin.com/in/evttxl

2 1

Fiasco.OC for PowerPC
by Wes 16 Jan '11

16 Jan '11

I keep seeing mention of there being a PowerPC WiP but can't seem to find it anywhere and the version in the snapshots and svn repos don't seem to have PowerPC as a target option. Where can I find this work? Cheers, Wes

2 2

Cannot determine L4 source directory
by eric van tassell 16 Jan '11

16 Jan '11

I've built L4Re and Fiasco without error but get the message below trying to build L4Linux. Does this indicate that CONFIG_L4_OBJ_TREE is set improperly? What should it point to? My directory structure looks like $ROOT /kernel /out #build dir /l4 /l4/out /l4linux /l4linux/out ERROR: Cannot determine L4 source directory Please properly set the path to your L4 build directory in the configuration menu. make[2]: *** [/home/evt/wrk/L4/TUD/forest1/src/l4linux/out/Makeconf.l4conf] Error 1 -- - evt (Eric van Tassell) twitter: evt_texelsoft linked-in: linkedin.com/in/evttxl

2 1

L4 fpages
by jason philip 16 Jan '11

16 Jan '11

I new to the L4 microkernel and do not understand how fpages works. What are the functions of fpages? Can them be used to extend the address space? Are they just for copy a region from one address space to other address space?

4 6

neophyte startup question
by eric van tassell 16 Jan '11

16 Jan '11

I'm intending to target an arm board in a few weeks but just to build and run the code does anyone have a suggestion as to the easiest way to get started? Do people usually use qemu to experiment with an x86 build? My goal is just to build the code and prove it boots, then move toward arm. Thanks. -- - evt (Eric van Tassell) twitter: evt_texelsoft linked-in: linkedin.com/in/evttxl

2 1

Thread IP initialization
by jason philip 15 Jan '11

15 Jan '11

How do initialize the IP or SP for thread or determine it location? In the "ThreadControl" system call (which used to create threads) it requires a IP and a SP, so where you get these values from? Are they implicitly determined by the kernel? I've briefly read pingpong source file but the calls made in the source do not seem to provide these values. And also how do I load program in order to execute it? I know this probably sounds like an easy question but I just started experimenting with l4 and I just getting to know things.

4 3

Re: Explanation of ARM TrustZone Support?
by Wesley Miaw 14 Jan '11

14 Jan '11

On Jan 13, 2011, at 3:00 AM, l4-hackers-request(a)os.inf.tu-dresden.de wrote: > Wesley Miaw(wesley@wesman.net)@2011.01.12 17:22:52 -0800: >> I was also thinking that L4Linux was run as the non-secure Linux via >> TrustZone but that was incorrect. L4Linux is for a paravirtualized >> security split while something like TZ-Linux is used for a full >> virtualized security split, with L4 as the "Hypervisor". Has TZ- >> Linux and >> its associated work been publicly released? > > No. Its just a prove of concept. However there exists a snapshot that > should build and run out of the box on a recent Realview/PBX board. > With the upcoming virtualization extensions for ARM this approach is > more > or less obsolete. I'd be interested in taking a look at this snapshot if you are willing to share it. I'm doing some investigation work into how to put together an execution environment like your proof of concept. Thanks, -- Wesley Miaw wesley(a)wesman.net

1 0

Explanation of ARM TrustZone Support?
by Wesley Miaw 13 Jan '11

13 Jan '11

Hi, I was wondering if someone could point me at documentation, or describe in a reply, how L4 Fiasco + L4Re makes use of the ARM TrustZone hardware? I see that support for TrustZone is listed listed as one of the features of Fiasco, but haven't been able to figure out exactly what that means. If built for an ARM chip with TrustZone support, does L4 automatically set up a TrustZone execution environment for each L4 server? Or do I need to tell it somewhere to put each server into its own environment? Are the hardware calls automatically made when switching between servers? How does use of TrustZone influence message passing between servers? I'm afraid all I've found in the source appears to be some code in the hardware porting layer that is a relative mystery to me. Thanks, -- Wesley Miaw wesley(a)wesman.net

3 4

Need for a copy of the original Assembly Code of L4
by 王彬彬 11 Jan '11

11 Jan '11

Who have the original assembly code of L4 could you please send me a copy? thanks very much!

2 1

Re: ARM L4Linux Kernel Panic
by Wesley Miaw 06 Jan '11

06 Jan '11

Hi Adam, I don't know what caused the problem, but I have figured out how to work around it. I was compiling with the "Sourcery G++ Lite 2010.09-50" build. Your README indicates "Sourcery G++ Lite 2009q3-67" is known to work. I downgraded to that version and now L4linux starts up without any errors. Thanks, -- Wesley Miaw wesley(a)wesman.net

2 1

Re: ARM L4Linux Kernel Panic
by Wesley Miaw 05 Jan '11

05 Jan '11

Hi Adam, > On Mon Jan 03, 2011 at 16:04:01 -0800, Wesley Miaw wrote: >> I am attempting to run the L4linux ARM build from the Fiasco + L4Re + L4linux snapshot l4re-snapshot-2010111519.tar.bz2 but L4linux is kernel panicking during startup. Here's my build information and a snippet of the >> error output. I'd appreciate any help trying to figure out what's going wrong and resolving the problem. >> GCC_PATH_PREFIX=/opt/sourcery-g++-lite/bin/arm-none-linux-gnueabi- L4ARCH=arm >> make setup >> -> selected ARM realview >> make >> cd obj/l4linux/arm-up >> make arm_defconfig >> make menuconfig >> -> set L4 build path obj/l4/arm-rv >> -> set cross compile prefix arm-none-linux-gnueabi- >> -> turned on ARM EABI + old EABI binaries options >> make >> cd obj/l4/arm-rv >> make image E="L4linux ARM" >> qemu-arm-l4 images/bootstrap.elf >> Got through to Ned: loading file: 'rom/l4lx.cfg' without any apparent issues. L4linux started, but got a bunch of bad page errors: >> BUG: Bad page state in process swapper pfn:0431c >> page:00488380 count:4752264 mapcount:0 mapping:00488390 index:0x488390 page flags: 0x488380(slab|owner_priv_1|arch_1|tail|buddy|mlocked) These went through, and then it looked like the kernel was proceeding, but >> eventually: > I think the "bad page state" messages are the cause for the following oops. I tried to reproduce that by following your steps but I failed, i.e. it always works for me. Could you give me some more info, like all of the Linux output so that I see gcc version used etc. Qemu is which version? I'll try attaching the startup output as a text file, so as to avoid a gigantic email body. Here's the Linux and QEMU info: Ubuntu 10.04 LTS Linux lgux-wmiaw2 2.6.32-25-generic #45-Ubuntu SMP Sat Oct 16 19:48:22 UTC 2010 i686 GNU/Linux QEMU PC emulator version 0.12.3 (qemu-kvm-0.12.3), Copyright (c) 2003-2008 Fabrice Bellard I have the QEMU Ubuntu packages installed: qemu-kvm-extras 0.12.3+noroms-0ubuntu9.3 vgabios 0.6c-2ubuntu1 qemu-common 0.12.3+noroms-0ubuntu9.3 qemu-kvm 0.12.3+noroms-0ubuntu9.3 Here's the Sourcery G++ version info: $ arm-none-linux-gnueabi-gcc -v Using built-in specs. COLLECT_GCC=arm-none-linux-gnueabi-gcc COLLECT_LTO_WRAPPER=/opt/sourcery-g++-lite/bin/../libexec/gcc/arm-none-linux-gnueabi/4.5.1/lto-wrapper Target: arm-none-linux-gnueabi Configured with: /scratch/nathan/arm-lite/src/gcc-4.5-2010.09/configure --build=i686-pc-linux-gnu --host=i686-pc-linux-gnu --target=arm-none-linux-gnueabi --enable-threads --disable-libmudflap --disable-libssp --disable-libstdcxx-pch --enable-extra-sgxxlite-multilibs --with-arch=armv5te --with-gnu-as --with-gnu-ld --with-specs='%{save-temps: -fverbose-asm} %{funwind-tables|fno-unwind-tables|mabi=*|ffreestanding|nostdlib:;:-funwind-tables} -D__CS_SOURCERYGXX_MAJ__=2010 -D__CS_SOURCERYGXX_MIN__=9 -D__CS_SOURCERYGXX_REV__=50 %{O2:%{!fno-remove-local-statics: -fremove-local-statics}} %{O*:%{O|O0|O1|O2|Os:;:%{!fno-remove-local-statics: -fremove-local-statics}}}' --enable-languages=c,c++ --enable-shared --enable-lto --enable-symvers=gnu --enable-__cxa_atexit --with-pkgversion='Sourcery G++ Lite 2010.09-50' --with-bugurl=https://support.codesourcery.com/GNUToolchain/ --disable-nls --prefix=/opt/codesourcery --with-sysroot=/opt/codesourcery/arm-none-linux-gnueabi/libc --with-build-sysroot=/scratch/nathan/arm-lite/install/arm-none-linux-gnueabi/libc --with-gmp=/scratch/nathan/arm-lite/obj/host-libs-2010.09-50-arm-none-linux-gnueabi-i686-pc-linux-gnu/usr --with-mpfr=/scratch/nathan/arm-lite/obj/host-libs-2010.09-50-arm-none-linux-gnueabi-i686-pc-linux-gnu/usr --with-mpc=/scratch/nathan/arm-lite/obj/host-libs-2010.09-50-arm-none-linux-gnueabi-i686-pc-linux-gnu/usr --with-ppl=/scratch/nathan/arm-lite/obj/host-libs-2010.09-50-arm-none-linux-gnueabi-i686-pc-linux-gnu/usr --with-host-libstdcxx='-static-libgcc -Wl,-Bstatic,-lstdc++,-Bdynamic -lm' --with-cloog=/scratch/nathan/arm-lite/obj/host-libs-2010.09-50-arm-none-linux-gnueabi-i686-pc-linux-gnu/usr --with-libelf=/scratch/nathan/arm-lite/obj/host-libs-2010.09-50-arm-none-linux-gnueabi-i686-pc-linux-gnu/usr --disable-libgomp --enable-poison-system-directories --with-build-time-tools=/scratch/nathan/arm-lite/install/arm-none-linux-gnueabi/bin --with-build-time-tools=/scratch/nathan/arm-lite/install/arm-none-linux-gnueabi/bin Thread model: posix gcc version 4.5.1 (Sourcery G++ Lite 2010.09-50) And my local /usr/bin/gcc: $ gcc -v Using built-in specs. Target: i486-linux-gnu Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.4.3-4ubuntu5' --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 --program-suffix=-4.4 --enable-nls --enable-clocale=gnu --enable-libstdcxx-debug --enable-plugin --enable-objc-gc --enable-targets=all --disable-werror --with-arch-32=i486 --with-tune=generic --enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu --target=i486-linux-gnu Thread model: posix gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) -- Wesley Miaw wesley(a)wesman.net

1 0

ARM L4Linux Kernel Panic
by Wesley Miaw 04 Jan '11

04 Jan '11

Hi, I am attempting to run the L4linux ARM build from the Fiasco + L4Re + L4linux snapshot l4re-snapshot-2010111519.tar.bz2 but L4linux is kernel panicking during startup. Here's my build information and a snippet of the error output. I'd appreciate any help trying to figure out what's going wrong and resolving the problem. GCC_PATH_PREFIX=/opt/sourcery-g++-lite/bin/arm-none-linux-gnueabi- L4ARCH=arm make setup -> selected ARM realview make cd obj/l4linux/arm-up make arm_defconfig make menuconfig -> set L4 build path obj/l4/arm-rv -> set cross compile prefix arm-none-linux-gnueabi- -> turned on ARM EABI + old EABI binaries options make cd obj/l4/arm-rv make image E="L4linux ARM" qemu-arm-l4 images/bootstrap.elf Got through to Ned: loading file: 'rom/l4lx.cfg' without any apparent issues. L4linux started, but got a bunch of bad page errors: BUG: Bad page state in process swapper pfn:0431c page:00488380 count:4752264 mapcount:0 mapping:00488390 index:0x488390 page flags: 0x488380(slab|owner_priv_1|arch_1|tail|buddy|mlocked) These went through, and then it looked like the kernel was proceeding, but eventually: l4lx_irq_init: creating IRQ thread on cpu 0 irq_thread: Starting IRQ thread on CPU 0 Console: colour dummy device 80x30 console [ttyLv0] enabled Calibrating delay loop... 161.79 BogoMIPS (lpj=808960) pid_max: default: 32768 minimum: 301 Mount-cache hash table entries: 512 l4linux | EXvcada: pfa=1 pc=12eda8 l4linux | l4x_setup_die_utcb 2863 Internal error: Boom!: 410017 [#1] last sysfs file: Modules linked in: CPU: 0 Tainted: G B (2.6.36-l4 #1) PC is at __wake_up_common+0x2c/0x98 LR is at __wake_up+0x40/0x50 pc : [<0012eda8>] lr : [<0012f1c0>] psr: 0000001f sp : 0035dea0 ip : ffffffff fp : 0035decc r10: 00000003 r9 : 0035def8 r8 : 00000000 r7 : 00488208 r6 : 00000001 r5 : 0035def8 r4 : 00000001 r3 : 00000000 r2 : 00000001 r1 : 00000003 r0 : ffffffff Flags: nzcv IRQs on FIQs on Mode SYS_32 ISA ARM Segment kernel Process swapper (pid: 0, stack limit = 0x0035c270) Stack: (0x0035dea0 to 0x0035e000) dea0: 000000d0 00000001 03c00dd0 00000001 0035def8 00000001 00000003 00488208 dec0: 00000000 00000000 0035def4 0012f1c0 0035def8 00000001 00387508 03c00dd0 dee0: 00368528 0035df08 00368528 002da0e8 00000000 0014d164 03c00edc 00000003 df00: 0035df08 001dbd2c 00000000 0e4e1c02 62656572 00400000 0420c400 001de490 df20: 0420e9e0 00368558 0032d1fc 00400000 00000000 00195454 0420e9e0 00000000 df40: 00000000 00388818 00000000 00000000 00000000 0010abec 00000000 00109da4 df60: 00387648 00001000 00003dcb 00109dc8 00000000 001097e4 00000000 00000000 df80: 00000000 003874f0 00003dcb 00109a98 00000000 0038769c 00000001 00109f18 dfa0: 00385d94 00000001 004884c0 00100628 001000d0 0037809c 0040f000 00114fec dfc0: 00000000 0037809c 0040f000 00102b10 346c0000 67752078 2d657461 00000030 dfe0: 00000000 00000000 00000000 00000000 0035dff4 0040f000 00000000 00000000 [<0012eda8>] (__wake_up_common+0x2c/0x98) from [<0012f1c0>] (__wake_up+0x40/0x50) [<0012f1c0>] (__wake_up+0x40/0x50) from [<0014d164>] (__wake_up_bit+0x28/0x30) [<0014d164>] (__wake_up_bit+0x28/0x30) from [<001dbd2c>] (sysfs_get_inode+0x10c/0x1a0) [<001dbd2c>] (sysfs_get_inode+0x10c/0x1a0) from [<001de490>] (sysfs_get_sb+0xd4/0x190) [<001de490>] (sysfs_get_sb+0xd4/0x190) from [<00195454>] (vfs_kern_mount+0x54/0x120) [<00195454>] (vfs_kern_mount+0x54/0x120) from [<0010abec>] (sysfs_init+0x60/0xbc) [<0010abec>] (sysfs_init+0x60/0xbc) from [<00109dc8>] (mnt_init+0x8c/0x1ac) [<00109dc8>] (mnt_init+0x8c/0x1ac) from [<00109a98>] (vfs_caches_init+0xe0/0x110) [<00109a98>] (vfs_caches_init+0xe0/0x110) from [<00100628>] (start_kernel+0x22c/0x294) [<00100628>] (start_kernel+0x22c/0x294) from [<00102b10>] (l4x_linux_startup+0x1d0/0x208) [<00102b10>] (l4x_linux_startup+0x1d0/0x208) from [<00000000>] (0x0) Code: e1570000 e50b2028 e1a08003 e59b9004 (e41c400c) ---[ end trace 1b75b31a2719ed1c ]--- Kernel panic - not syncing: Attempted to kill the idle task! [<00125b4c>] (unwind_backtrace+0x0/0xf0) from [<002d1130>] (panic+0x58/0x178) [<002d1130>] (panic+0x58/0x178) from [<00139b2c>] (do_exit+0x5c4/0x600) [<00139b2c>] (do_exit+0x5c4/0x600) from [<0011ee9c>] (die+0x2a8/0x384) [<0011ee9c>] (die+0x2a8/0x384) from [<00000000>] (0x0) panic: going to sleep forever, bye l4linux | panic: going to sleep forever, bye Thanks, -- Wesley Miaw wesley(a)wesman.net

2 1

Problems booting Pistachio
by Ashford Nichols 26 Dec '10

26 Dec '10

I recently built the microkernel and tried to boot it using "qemu" but it doesn't work. qemu just seems to not even seem to recognize the image as a bootable image. I followed the instructions given on the site and everything worked, right up to the point where it doesn't boot. I must say though that I did slightly tweak the steps given on the site to create the image file since the original ones didn't work for me. I am really new to qemu so I don't have the slightest clue as to why this isn't working. If anyone experienced this problem or at least know how to resolve this please help me. The step I followed in order to create the image follows: dd if=/dev/zero of=l4-image.img bs=512 count=8640mke2fs l4-image.imgmkdir /media/mount_pointmount -o loop l4-image.img /media/mount_pointmkdir /media/mount_point/bootmkdir /media/mount_point/boot/grubcp -aR grub_stages/* /media/mount_point/boot/grubcp -aR modules/* /media/mount_pointumount /media/mount_point

1 0

L4Re: Thread local storage
by Taru Karttunen 22 Dec '10

22 Dec '10

Hello The following program crashes: #include <stdio.h> #include <pthread.h> __thread int i; int main(void) { puts("Hello World!"); printf("address of i=%p\n", &i); printf("i=%d\n", i); return 0; } with: MOE: loading 'rom/hello2' Hello World! address of i=0xb30001fc L4Re: unhandled exception: pc=0x100013c How should TLS initialization be done? This is with L4_MULTITHREADED = y Dump of assembler code for function main: 0x01000100 <+0>: push %ebp 0x01000101 <+1>: mov %esp,%ebp 0x01000103 <+3>: and $0xfffffff0,%esp 0x01000106 <+6>: sub $0x10,%esp 0x01000109 <+9>: movl $0x1017b20,(%esp) 0x01000110 <+16>: call 0x100dab0 <puts> 0x01000115 <+21>: movl $0x1,(%esp) 0x0100011c <+28>: call 0x100f9a0 <sleep> 0x01000121 <+33>: mov %gs:0x0,%eax 0x01000127 <+39>: movl $0x1017b2d,(%esp) 0x0100012e <+46>: add $0xfffffffc,%eax 0x01000133 <+51>: mov %eax,0x4(%esp) 0x01000137 <+55>: call 0x100da80 <printf> 0x0100013c <+60>: mov %gs:0xfffffffc,%eax 0x01000142 <+66>: movl $0x1017b3e,(%esp) 0x01000149 <+73>: mov %eax,0x4(%esp) 0x0100014d <+77>: call 0x100da80 <printf> 0x01000152 <+82>: xor %eax,%eax 0x01000154 <+84>: leave 0x01000155 <+85>: ret So how to get this working? ps. I have a rudimentary port of gccgo on L4Re working, and am polishing it over the holidays. - Taru Karttunen

3 2

schedule system call
by Ashford Nichols 20 Dec '10

20 Dec '10

In version 2 of the interface under the schedule system call, it says that the call can be used be schedulers. However, in the next sentence it goes on to say that the calling thread must reside in the same address space as the scheduler. Also, it goes on to say that an error code will be set if the calling thread is not the scheduler. So is the "schedule" system call made by the scheduler or by a thread that simply resides in the same address space as the scheduler?

1 0

L4 pager
by Ashford Nichols 19 Dec '10

19 Dec '10

I am new to L4 microkernels and I am reading the reference manual on version 2 of the interface. From what I've read so far the manual suggests that there's more than one pager, but if this is so what is the purpose of more than one pager?

2 2

GCC 4.5.1 and L4Re
by Taru Karttunen 17 Dec '10

17 Dec '10

Hello Has anyone gotten L4Re to work with GCC 4.5.1 on amd64? I am having issues with just a simple hello world modules.list: entry taru roottask moe --debug=all --l4re-dbg=all --init=rom/hello module l4re module hello Grub loads all the modules ok, but then the qemu just exits without anything being printed from hello. Any hints would be appreciated. - Taru Karttunen

2 3

Several questions
by Nuno Santos 10 Dec '10

10 Dec '10

Hi I've managed to build and run Fiasco with one instance of L4Linux. At this point I have several questions: - I believe that an application running in l4linux can communicate with an application sitting on fiasco. Is there a simple example I could look at? - As a follow up, how can we allow multiple l4linux instances to communicate using IPC? This post from 2005 [1] has done precisely what I need. Would it be possible to share that code? - I'd like to use and deploy a tpm driver directly on top of L4 (not directly accessible to l4linux). Is it possible to obtain that driver? How to set it up? This post [2] mentions the stpm package, but I couldn't find it in the l4re snapshot. - Is it possible to plug a TPM emulator to work directly on fiasco? It's mostly for experimentation and testing on machines without a physical TPM. - I believe that it's possible to boot fiasco with a trusted bootloader. [2] mentions oslo, but again I couldn't find the code in the snapshot. Do you know where can I find it? Do you know if trusted grub can be used as a viable alternative? I've looked extensively for online information, but I confess I'm a bit overwhelmed. Your guidance here would be fantastic. I really appreciate your help. Thanks! --Nuno [1] http://www.mail-archive.com/l4-hackers@os.inf.tu-dresden.de/msg01709.html [2] http://www.mail-archive.com/l4-hackers@os.inf.tu-dresden.de/msg03861.html

3 2

Problem with building L4Linux
by Nuno Santos 07 Dec '10

07 Dec '10

Hi I'm having trouble with building l4linux. First, I had the same problem as reported in this [1] post; then I tried to build l4linux from the latest snapshot as suggested in [2] by following the steps in the readme file ('make setup' and 'make'). If I setup the target as x86-32, at some point the build process fails with the following output (without further description of the error): " make[6]: Leaving directory `/local/l4re-snap/l4re-snapshot-2010111519/src/l4/pkg/valgrind/build' make[5]: Leaving directory `/local/l4re-snap/l4re-snapshot-2010111519/src/l4/pkg/valgrind' make[4]: Leaving directory `/local/l4re-snap/l4re-snapshot-2010111519/src/l4/pkg' make[3]: *** [pkg] Error 2 make[3]: Leaving directory `/local/l4re-snap/l4re-snapshot-2010111519/src/l4' make[2]: *** [x] Error 2 make[2]: Leaving directory `/local/l4re-snap/l4re-snapshot-2010111519/obj/l4/x86' Error building the L4Re 'obj/l4/x86' variant. " If I setup to the amd64 target, the build process fails with a compilation error: " /tmp/camlasm564f4e.s: Assembler messages: /tmp/camlasm564f4e.s:580: Error: junk `lazy_force(%rip)' after expression /tmp/camlasm564f4e.s:580: Error: suffix or operands invalid for `lea' /tmp/camlasm564f4e.s:785: Error: junk `lazy_force(%rip)' after expression /tmp/camlasm564f4e.s:785: Error: suffix or operands invalid for `lea' Assembler error, input left in file /tmp/camlasm564f4e.s make[8]: *** [stream.cmx] Error 2 ... make[4]: Leaving directory `/local/l4re-snap/l4re-snapshot-2010111519/src/l4/pkg' make[3]: *** [pkg] Error 2 make[3]: Leaving directory `/local/l4re-snap/l4re-snapshot-2010111519/src/l4' make[2]: *** [x] Error 2 make[2]: Leaving directory `/local/l4re-snap/l4re-snapshot-2010111519/obj/l4/amd64' Error building the L4Re 'obj/l4/amd64' variant. " The problematic assembly instruction is: leaq $25lazy_force(%rip), %rax I'm using the snapshot from Nov 15th, on a machine running linux 2.6.32, gcc 4.3.2. Any ideas on how to fix it? Thanks for your help! --Nuno [1] http://www.mail-archive.com/l4-hackers@os.inf.tu-dresden.de/msg04380.html [2] http://www.mail-archive.com/l4-hackers@os.inf.tu-dresden.de/msg04383.html

3 3

Stringitem
by BERTRAND Joel 04 Dec '10

04 Dec '10

Hello, I think I have misunderstood StringItem IPC but I don't know where is my mistake. My roottask spawns a thread that sends a stringitem to roottask. I have checked with debugger that this stringitem contains my string : si = L4_StringItem(str.length_trim, (void *) str.c); L4_Clear(&msg); L4_Append(&msg, si); L4_Set_Label(&msg, CALL$PRINT); L4_Load(&msg); L4_Call(L4_Pager()); str is a struct vms$string : struct vms$string { vms$pointer length; vms$pointer length_trim; unsigned char *c; }; My roottask receives this IPC in following function: void sys$loop() { int running; L4_ThreadId_t partner; L4_MsgBuffer_t buffer; L4_MsgTag_t tag; L4_Msg_t msg; running = 1; L4_Accept(L4_MapGrantItems(L4_CompleteAddressSpace) + L4_StringItemsAcceptor); tag = L4_Wait(&partner); while(running) { L4_Clear(&msg); L4_Store(tag, &msg); if ((tag.raw & L4_REQUEST_MASK) == L4_PAGEFAULT) { sys$pagefault(partner, L4_Get(&msg, 0), L4_Get(&msg, 1), tag.raw); } else { switch(L4_Label(tag)) { case CALL$PRINT: notice("Here!\n"); break; default: PANIC(1, notice(IPC_F_UNKNOWN "unknown IPC from $%lX " "with label $%lX\n", L4_ThreadNo(partner), L4_Label(tag))); } } tag = L4_ReplyWait(partner, &partner); } return; } When system boots, I can intercept on serial line: ... %SYS-F-PAGEFLT, pagefault request from $3C at $0000000000800110 %SYS-F-PAGEFLT, pagefault request from $3C at $000000000010FFF8 %SYS-F-PAGEFLT, pagefault request from $3C at $00000000008015C0 %SYS-F-PAGEFLT, pagefault request from $3C at $0000000000801C10 Here! %IPC-F-UNKNOWN, unknown IPC from $0 with label $0 Panic at sys$loop, sys/sys_loop.c line 73 Have a nice day ! Backtrace: <00> [$000000000104F5B8] -> $0000000001004ED5 (sys$loop) <01> [$000000000104F838] -> $0000000001005813 (main) Pagefault messages are written by sys$pagefault() subroutine. As sys$loop() function writes 'Here!', it receives an IPC with CALL$PRINT label. First question: I don't know how read this stringitem. I have tried with msgbuff without any result. I suppose I have misunderstood how StringItem IPC works. Second question: I don't understand last line: "%IPC-F-UNKNOWN, unknown IPC from $0 with label $0". I suppose this error comes from a bad stringitem IPC. I haven't found any stringitem IPC example. Help is welcome. Regards, JB

1 1

Some questions about pager
by BERTRAND Joel 29 Nov '10

29 Nov '10

Hello, It's my first post on this mailing list. A long time ago, I have started to write a VMS-clone (http://www.freevms.net) This opensourced project was sleeping for a long time because monolithic kernel isn't good enough to write a clone of this system. Thus, I have restarted this project with a L4 microkernel (exactly pistachio L4/X2). Now, I have written a roottask that tries to launch VMS$INIT, first thread required to boot all system. I have reuse some works done for iguana project (and fixed some bugs). When VMS$INIT is started, this thread sends two IPC to its pager : vms$pagefault(addr:1001730, s:8000) [priv=4] -> default addresse is $1001730 (start address of VMS$INIT thread) ... vms$pagefault(addr:47fe8, s:2000) [priv=4] I'm not sure that I have understand how should work a pager. I have uploaded my pager here : http://www.freevms.net/trac/browser/sources/freevms/vm/vm_pagefault.c All sources are available with : git clone git://rayleigh.systella.fr/FreeVMS.git and I have uploaded a disk image that runs with qemu (with a x86/64) : http://www.systella.fr/~bertrand/freevms.img.bz2 My pager tries to find a physical free page to map virtual memory. I think that memory is correctly mapped and I don't understand why pager receives other request with same virtual address. I suppose I have done a mistake. Any idea ? Thanks in advance for all explanations, JB

2 8

errors getting python up-and-running
by lkcl 28 Nov '10

28 Nov '10

ok, sorry about the number of messages, i'm going rapidly through things, trying to get python and other stuff started. i've had to create a one-by-one "add the library, get the error, go on to next error" modules.list file, and when i get to libm_support.so i get "rom/libm_support.so is not an ELF file" so i did nm on it as you can see below. not sure what it means though. anyway here's the list of modules so far, there _must_ be a better way to do this - any clues? modaddr 0x002000000 kernel fiasco -serial_esc roottask moe rom/hello.lua module l4re module ned module hello.lua module hello module x86-fb.cfg module x86-fb.io module x86-legacy.devs module l4re module io module fbterminal module fb-drv module mag module python module libld-l4.so module libc_be_file_stdin.so module libc_be_sig.so module libevent.so module libsupc++.so module libm_support.so module lib4re-util.so module libdl.so lkcl@teenymac:~/src/l4ka/build-x86_64$ ls -altr lib/amd64_K8/l4f/libm_support.solrwxrwxrwx 1 lkcl lkcl 92 2010-11-26 05:09 lib/amd64_K8/l4f/libm_support.so -> /home/lkcl/src/l4ka/build-x86_64/pkg/libc_backends/lib/math/OBJ-amd64_K8-l4f/libm_support.so lkcl@teenymac:~/src/l4ka/build-x86_64$ ls -altr home/lkcl/src/l4ka/build-x86_64/pkg/libc_backends/lib/math/OBJ-amd64_K8-l4f/libm_support.so ls: cannot access home/lkcl/src/l4ka/build-x86_64/pkg/libc_backends/lib/math/OBJ-amd64_K8-l4f/libm_support.so: No such file or directory lkcl@teenymac:~/src/l4ka/build-x86_64$ ls -altr /home/lkcl/src/l4ka/build-x86_64/pkg/libc_backends/lib/math/OBJ-amd64_K8-l4f/libm_support.so -rwxr-xr-x 1 lkcl lkcl 11163 2010-11-26 05:09 /home/lkcl/src/l4ka/build-x86_64/pkg/libc_backends/lib/math/OBJ-amd64_K8-l4f/libm_support.so lkcl@teenymac:~/src/l4ka/build-x86_64$ nm !$ nm /home/lkcl/src/l4ka/build-x86_64/pkg/libc_backends/lib/math/OBJ-amd64_K8-l4f/libm_support.so 00000000002009a0 a _DYNAMIC 0000000000200b00 a _GLOBAL_OFFSET_TABLE_ w _Jv_RegisterClasses 0000000000200980 d __CTOR_END__ 0000000000200978 d __CTOR_LIST__ 0000000000200990 d __DTOR_END__ 0000000000200988 d __DTOR_LIST__ 0000000000000970 r __FRAME_END__ 0000000000200998 d __JCR_END__ 0000000000200998 d __JCR_LIST__ 000000006ffff000 A __L4_KIP_ADDR__ 0000000070000000 A __L4_STACK_ADDR__ 0000000000200b68 A __bss_start w __cxa_finalize 0000000000000830 t __do_global_ctors_aux 0000000000000690 t __do_global_dtors_aux 0000000000200b60 d __dso_handle 0000000000200b68 A _edata 0000000000200b78 A _end 0000000000000868 T _fini 00000000000005d0 T _init 0000000000000820 T _sin 0000000000000810 T _sinf 0000000000200b68 b completed.6341 U cos U cosf 0000000000200b70 b dtor_idx.6343 U exp2 00000000000007f0 T exp2f 0000000000000710 t frame_dummy 00000000000007e0 T log2f U sin 0000000000000790 T sincos 0000000000000740 T sincosf U sinf lkcl@teenymac:~/src/l4ka/build-x86_64$ ldd !$ ldd /home/lkcl/src/l4ka/build-x86_64/pkg/libc_backends/lib/math/OBJ-amd64_K8-l4f/libm_support.so linux-vdso.so.1 => (0x00007fff445d4000) libuc_c.so => not found lkcl@teenymac:~/src/l4ka/build-x86_64$ -- View this message in context: http://old.nabble.com/errors-getting-python-up-and-running-tp30308895p30308… Sent from the L4 mailing list archive at Nabble.com.

4 9

python libc_be_file_stdin missing
by lkcl 26 Nov '10

26 Nov '10

am making progress with the goals i've set: did "svn update libc_be_file_stdin" which returned silently... but then make O=.... python fails with missing libc_be_file_stdin.pc (?) any ideas? -- View this message in context: http://old.nabble.com/python-libc_be_file_stdin-missing-tp30308867p30308867… Sent from the L4 mailing list archive at Nabble.com.

2 2

compile error svn rev 27 on x86
by lkcl 26 Nov '10

26 Nov '10

hi folks, this is with defaults (x86) - just did make B= and make O= what's going on!!! === Building package "l4sys" === make[2]: Entering directory `/home/lkcl/src/l4ka/src/l4/pkg/l4sys' make[3]: Entering directory `/home/lkcl/src/l4ka/src/l4/pkg/l4sys/include' make[3]: Leaving directory `/home/lkcl/src/l4ka/src/l4/pkg/l4sys/include' make[3]: Entering directory `/home/lkcl/src/l4ka/src/l4/pkg/l4sys/lib' make[4]: Entering directory `/home/lkcl/src/l4ka/src/l4/pkg/l4sys/lib/src' make[5]: Entering directory `/home/lkcl/src/l4ka/build-x86/pkg/l4sys/lib/src/OBJ-x86_586-l4f' ... Compiling utcb.o ... Compiling errtostr.o ==> Archiving into libl4sys.a ==> "libl4sys.a" built make[5]: *** No rule to make target `/usr/lib/gcc/x86_64-linux-gnu/4.4.3/../../../../lib32//crtendS.o', needed by `libl4sys.so'. Stop. make[5]: Leaving directory `/home/lkcl/src/l4ka/build-x86/pkg/l4sys/lib/src/OBJ-x86_586-l4f' make[4]: *** [/home/lkcl/src/l4ka/build-x86/pkg/l4sys/lib/src/OBJ-x86_586-l4f] Error 2 -- View this message in context: http://old.nabble.com/compile-error-svn-rev-27-on-x86-tp30307856p30307856.h… Sent from the L4 mailing list archive at Nabble.com.

2 6

errors on running x86-fb example
by lkcl 26 Nov '10

26 Nov '10

after quite a bit of mucking about, i managed to get a version of examples conf x86-fb.list that semi-worked, showing here some errors under qemu. any idea what's going on? tia. l. fbdrv | Exception: Invalid argument: not an ELF binary: backtrace ... fbdrv | L4Re: FATAL: could not load binary IO | Io service mag | Exception: Invalid argument: not an ELF binary: backtrace ... mag | L4Re: FATAL: could not load binary IO | Hello from L4-ACPICA IO | Find root Pointer IO | AcpiOsMapMemory(40e, 2) = 40040e IO | AcpiOsMapMemory(9f400, 400) = 49f400 Ned[lua]: Hm, cpapbility 42d000 does not support the meta protocol: -2004 IO | AcpiOsMapMemory(e0000, 20000) = 4e0000 IO | Find root Pointer: f8950 IO | AcpiOsMapMemory(f8950, 24) = 4f8950 IO | ACPI: RSDP 0xf8950 00014 (v00 BOCHS ) IO | AcpiOsMapMemory(7ffde30, 24) = bfde30 IO | ACPI: RSDT 0x7ffde30 00034 (v01 BOCHS BXPCRSDT 00000001 BXPC 00000001) IO | AcpiOsMapMemory(7ffde30, 34) = bfde30 IO | AcpiOsMapMemory(7fffe70, 24) = bffe70 IO | ACPI: FACP 0x7fffe70 00074 (v01 BOCHS BXPCFACP 00000001 BXPC 00000001) IO | AcpiOsMapMemory(7fffe70, 74) = bffe70 IO | AcpiOsMapMemory(7ffdfd0, 24) = bfdfd0 IO | ACPI: DSDT 0x7ffdfd0 01E22 (v01 BXPC BXDSDT 00000001 INTL 20090521) IO | AcpiOsMapMemory(7fffe00, 24) = bffe00 IO | ACPI: FACS 0x7fffe00 00040 IO | AcpiOsMapMemory(7ffdf90, 24) = bfdf90 IO | ACPI: SSDT 0x7ffdf90 00037 (v01 BOCHS BXPCSSDT 00000001 BXPC 00000001) IO | AcpiOsMapMemory(7ffdeb0, 24) = bfdeb0 IO | ACPI: APIC 0x7ffdeb0 00072 (v01 BOCHS BXPCAPIC 00000001 BXPC 00000001) IO | AcpiOsMapMemory(7ffde70, 24) = bfde70 IO | ACPI: HPET 0x7ffde70 00038 (v01 BOCHS BXPCHPET 00000001 BXPC 00000001) IO | AcpiOsMapMemory(7ffdfd0, 1e22) = bfdfd0 Ned: ERROR: rom/hello.lua:56: attempt to call method 'create' (a nil value) IO | Parsing all Control Methods: -- View this message in context: http://old.nabble.com/errors-on-running-x86-fb-example-tp30308892p30308892.… Sent from the L4 mailing list archive at Nabble.com.

1 1