- l4-hackers - example.com Site

L4Linux and AMD64
by Zephiris 09 Apr '06

09 Apr '06

I haven't been looking around at things very long, but I was noticing, while Fiasco largely supports AMD64 processors in 64-bit mode, L4Linux appears to specifically avoid it. I couldn't find any references for this, will it ever be supported? It's something I was excited about/meaning to try, but as my system is setup with 64-bit binaries in mind, it looks like the only way to test it would be for L4Linux to support that.

2 1

Re: Building the environment fails
by Alexander Boettcher 05 Apr '06

05 Apr '06

1 0

Re: Building the environment fails
by Alexander Boettcher 05 Apr '06

05 Apr '06

1 0

Building the environment fails
by Stefan Schulz 05 Apr '06

05 Apr '06

Sorry for using this ML as my personal helpdesk, but i believe there is a problem in the build process for l4env. I checked the current version out from CVS, and used "make config" to configure. While "make" was running, the process failed with ... Compiling ARCH-all/libc/misc/internals/__uClibc_main.o LD_PRELOAD=/l4/tmp/l4/tool/gendep/libgendep.so GENDEP_TARGET=ARCH-all/libc/misc/internals/__uClibc_main.o GENDEP_BINARY=cc1 gcc -c -nostdinc -include ../../../lib/uclibc/ARCH-all/include/libc-symbols.h -DNDEBUG -D_LIBC -DSTATIC -DSYSTEM_x86_586 -DARCH_x86 -DCPUTYPE_586 -DL4API_ -DCONFIG_L4_CALL_SYSCALLS -DCONFIG_L4_ABS_SYSCALLS -I../ARCH-x86/include -I../ARCH-all/include -I../../../../../include/x86 -I../../../../../include -I/usr/lib/gcc/i686-pc-linux-gnu/3.4.5/include -fno-builtin -gstabs+ -g -O2 -fno-strict-aliasing -march=i586 -Wall -Wstrict-prototypes -fno-common /l4/tmp/l4/pkg/uclibc/lib/uclibc/ARCH-all/libc/misc/internals/__uClibc_main.c -o ARCH-all/libc/misc/internals/__uClibc_main.o /l4/tmp/l4/pkg/uclibc/lib/uclibc/ARCH-all/libc/misc/internals/__uClibc_main.c:2:49: l4/sys/consts.h: No such file or directory /l4/tmp/l4/pkg/uclibc/lib/uclibc/ARCH-all/libc/misc/internals/__uClibc_main.c:15: error: `L4_PAGESIZE' undeclared here (not in a function) make[4]: *** [ARCH-all/libc/misc/internals/__uClibc_main.o] Error 1 make[4]: Leaving directory `/l4/tmp/l4/pkg/uclibc/lib/uclibc/OBJ-x86_586' make[3]: *** [OBJ-x86_586] Error 2 make[3]: Leaving directory `/l4/tmp/l4/pkg/uclibc/lib/uclibc' make[2]: *** [virt_dep] Error 2 make[2]: Leaving directory `/l4/tmp/l4/pkg/uclibc' make[1]: *** [uclibc.headers] Error 2 make[1]: Leaving directory `/l4/tmp/l4/pkg' make: *** [pkg] Error 2 Even if i "fixed" this error (by manually defining L4_PAGESIZE", similar problems crept up all over the build, far too many to manually fix. I tried this build on an IBM T43, using current sources, gcc3.4.5 and make 3.80 Is the build process really broken, or am i doing something wrong?

1 0

Make hello stops with an error
by Ricard de Benito Panillo 04 Apr '06

04 Apr '06

Hi, the following error arises making 'hello': cc1: note: obsolete option -I- used, please use -iquote instead In file included from ../libkern/x86/base_cpu_init.c:22: ../flux/x86/base_trap.h:105: error: array type has incomplete element type make[3]: *** [base_cpu_init.o] Error 1 gcc version 4.0.2 20051125 (Red Hat 4.0.2-8) on Fedora Core 4 Any hint? it seems that the compiler versions is newer than expected. Best regards, Ricard de Benito

2 1

I have new problem!
by chenggh04＠st.lzu.edu.cn 04 Apr '06

04 Apr '06

>Can you please give more details on what you did? I cannot really say >anything helpful on a generic "it doesn't work" statement. Thank adam lackorzynski(<adam(a)os.inf.tu-dresden.de>) for your help. Now I resolve last problem and the compilation fiasco in slackware (last time is debian) is ok. The execution of hello example is also good! Now I am ready for experiencing the fiasco-ux. I compile the package but runs badly. The error information is as follows: Fiasco-UX on Linux 2.4.31 (i686) Mapped 64 MB Memory + 0 KB Framebuffer + 0 MB Input Area on FD 3 Loading Module 0x00090000-0x00094e60 [sigma0-ux] Failed to open file: roottask But according to the instructions of Fiasco-ux website the information should be like this: Fiasco-UX on Linux 2.4.31 (i686) Mapped 64 MB Memory + 0 KB Framebuffer + 0 MB Input Area on FD 3 Loading Module 0x00070000-0x000955c0 [sigma0-ux] Loading Module 0x00100000-0x001b9ce0 [rmgr-ux] ... The addresses of the sigma0-ux are different! My configuration of fiasco is as follows: Target System Options >(UX) Target Platform (IA-32) Target CPU Family (K7) Target processor >(L4V2) Kernel ABI Version [ ] TCB size of 4k [ ] Fine-grained CPU time [Y] Graphical console (requires SDL library!) [Y] Network support Kernel Debugging >[Y] Generate inline code [ ] Do not compile assertions [Y] Compile without frame pointer [ ] Measure stack depth of threads [ ] Sanity checks in low level allocator [ ] Sanity checks at syscall entry/exit [Y] Compile with gstabs+ debugging information [Y] JDB kernel debugger [ ] JDB extended logging [ ] JDB accounting [ ] Miscellaneous JDB modules (ANY) Runtime warning level Compiling and Building >gcc C compiler g++ C++ compiler [ ] Verbose build process [ ] Do additional checks at build time and then make VERBOSE=@ according to the instruction of Fiasco-ux website but I don't find the option "Assembler IPC shortcut" . Please help me! Cheng

4 3

Why don't I have /bin/ directory in L4
by chenggh04＠st.lzu.edu.cn 03 Apr '06

03 Apr '06

Hi,Everybody: I am a beginner for L4. when I compile the L4 without any errors but finally I don't have the bin directory which inlcudes hello, rmgr and sigma0. Maybe I think there is lack of some packages in my install package. But the package is downloaded from the Fiasco official WebSite. I don't know why! Please help me! Chenggh

2 1

Lovelace Web-Site
by Frederic BOYER 30 Mar '06

30 Mar '06

By the way, Lovelace has now a little web-site (I've just began to do it) and there is a mailing list. Please visit The Lovelace web-site. http://lovelace.rochebrune.org Regards, Frédéric. "This message, including any attachments may contain confidential and privileged material; it is intended only for the person to whom it is addressed. Its contents do not constitute a commitment by Credit Agricole Cheuvreux except where provided for in a written agreement. Credit Agricole Cheuvreux assumes no liability or responsibility for the consequences arising out of a delay and/or loss in transit of this message, or for corruption or other error(s) arising in its transmission and for any misuse or fraudulent use which may be made thereof. If you are not the intended recipient, please contact us and abstain from any disclosure, use or dissemination. To the extent that this message contains research information and/or recommendations, these are provided on the same basis as Credit Agricole Cheuvreux's published research and the recipient must have regard to all disclosures and disclaimers contained therein."

2 1

L4Linux without libloader?
by Daniel Vandersee 23 Mar '06

23 Mar '06

Hi there, is it possible to compile L4Linux without the need of libloader.s.so (e.g. by using other libs) to have it statically linked? Thanks, Daniel

4 7

Reloading L4Linux problem
by Jugwan Eom 17 Mar '06

17 Mar '06

Hi. I encountered problems reloading L4Linux. After shutdown loaded L4Linux, I tried reloading the L4Linux (same configuration) using 'l' command in l4con but it didn't work(network device probe failed, lost ide disk interrupt). During booting L4linux showed "irq_thread: RMGR denied IRQ 14: code 0x1". Because I use l4con, I can see the system memory on bottom right. After shutdown loaded L4Linux, it still shows the same value before. Using 'a' command in l4con memory region used by previous L4Linux seem to be still allocated. IMHO, this is the source to fail to reloading L4Linux. Because memory region by used L4Linux includes the DMA region, reloading L4Linux cannot control the system device. Is this a bug in L4Linux or other L4Env server? Also, see the attached log. Best regards, -- Jugwan Eom

4 7