l4-hackers

l4-hackers@os.inf.tu-dresden.de

2 participants
2880 discussions

problems running hello in fiasco
by Fabian Sturm 26 Mar '03

26 Mar '03

Hello! I have a problem running the hello world example on fiasco. I downloaded the latest CVS version from 25.4 of fiasco and compiled it like it is described on the web. Except that the sigma0/rmgr binaries were in l4/bin/x86_586/l4v2 and not directly in l4/bin as described. I then created the following grub entry: title Fiasco, hello world root (hd0,4) kernel=/fiasco/rmgr -sigma0 module=/fiasco/main -nokdb module=/fiasco/sigma0 module=/fiasco/hello When I then boot into it I see some messages of the RMGR but then the system is frozen. Maybe someone knows what I should try next to track this down? Thanks a lot, Fabian Sturm My system is a P3 800Mhz 256MB ram notebook with the Intel i815 chipset. (Vaio vx88) The build environment was Debian Linux 3.0 with gcc 2.95.4. The boot messages were: RMGR: Loading /fiasco/sigma0 RMGR: Loading Kernel fiasco main -nokdb RMGR: check_l4_version 0x00300000 to 0x0044a244 RMGR: detected new_style DD-L4(v2)/Fiasco RMGR: configuring s0: (0x00070000, 0x00096000), start: 0x00070000 RMGR: reserve modules memory range: 0x00200000 to 0x0021053c RMGR: starting kernel /fiasco/main -nokdb proto=0x101ba0 @ 0x00300220 Booting kernel image at 0x302000 entry 0xf0001000 Enabling fully serial nested mode for pic Using the RTC on IRQ 8 (1khz) for scheduling (I hope I don't have introduced errors while copying that down)

2 1

Re: Problem compiling fiasco-ux: ld: entry.o: bad symbol index: 845556
by Udo A. Steinberg 24 Mar '03

24 Mar '03

On 24 Mar 2003 20:58:50 +0100 Rudolf Weber (RW) wrote: RW> Thank you very much, RW> anbei der readelf-output 4udo und das erzeugte entry.o für die echten RW> Programmierer ... RW> Relocation section '.rel.text' at offset 0x10d0 contains 13 entries: RW> Offset Info Type Sym.Value Sym. Name RW> 0000011b 00004101 R_386_32 00000000 base_trap_handler RW> 00000141 00004202 R_386_PC32 00000000 trap_dump_panic RW> 00000150 00004301 R_386_32 00000000 PAGE_FAULT_ADDR RW> 00000179 00004402 R_386_PC32 00000000 thread_page_fault RW> 000001cc 00004502 R_386_PC32 00000000 thread_handle_fputrap RW> 000001e7 00004702 R_386_PC32 00000000 thread_timer_interrupt RW> 000001f7 00004902 R_386_PC32 00000000 thread_timer_interrupt RW> 00000207 00004b02 R_386_PC32 00000000 thread_timer_interrupt RW> 00000211 00004c02 R_386_PC32 00000000 irq_interrupt RW> 00000324 00004d02 R_386_PC32 00000000 dispatch_syscall RW> 00000356 00004f02 R_386_PC32 00000000 call_ipc_short_cut RW> 00000360 0ce6f401 R_386_32 bad symbol index: 000ce6f4 RW> 00000374 00004d02 R_386_PC32 00000000 dispatch_syscall Hallo, Siehe vorletzte gequotete Zeile. Der generierte Code ist kaputt. Ich habe das Problem hier soeben mit gcc 2.95 und gcc 3.2 reproduziert. Bis ich genaueres weiß bzw. einen Patch habe, bitte "Assembler IPC Shortcut" in "make menuconfig" anschalten - dann funktioniert's. -Udo.

1 1

Poblem compiling fiasco-ux: ld: entry.o: bad symbol index: 845556
by Rudolf Weber 23 Mar '03

23 Mar '03

Hello, I tried to compile the current fiasco-ux: make[1]: Entering directory `/home/rw/l4/20030322/l4/kernel/fiasco/build-ux' ==> Linking kernel.image g++ -Wl,--gc-sections,-T../src/kernel.ux.ld -static -o kernel.image boot_info.o boot_info-ux.o cpu-ia32-common.o cpu-ux.o mapdb.o region.o space_context-ia32-ux.o space_context-ux.o space.o space-ia32-ux.o space-ux.o space_index.o space_index_util.o stack.o kmem-ia32-ux.o kmem-ux.o kmem_alloc.o kmem_alloc-ia32-ux.o slab_cache_anon.o kmem_slab_simple.o kmem_slab.o switch_lock.o thread_lock.o helpin g_lock.o cpu_lock.o cpu_lock-generic.o timer.o timer-ia32.o timer-ux.o pic.o pic-ux.o thread_state.o sched.o sched-ia32-ux.o context.o context-ia32-ux.o context-ux.o threadid.o sender.o receiver.o thread.o thread-ia32-ux.o thread-ux.o thread-v2.o thread-syscall.o thread-syscall-v2.o thread-ipc.o thread-msg.o thread-msg-ux.o thread-list.o thread-nolog.o thread-user-ux.o thread-nosmas.o kernel_thread.o kernel_thread-std.o kernel_thread-ux.o thread_util.o kdb_ke.o dirq.o dirq-ia32-ux.o fpu.o fpu-ia32-ux.o fpu-ux.o fpu_alloc.o fpu_state.o map_util.o map_util-ia32-ux.o irq.o irq-ipc.o irq-ux.o globals.o globals-ia32-ux.o main-ia32-ux.o main-ux.o config.o config-ux.o config-ia32-ux.o usermode.o mapped_alloc.o trampoline.o loader.o kern_types.o vmem_alloc.o vmem_alloc-ia32-ux.o vmem_alloc-ux.o panic.o paging.o paging-ia32-ux.o entry_frame.o entry_frame-ia32-common-v2.o startup.o emulation.o hostproc.o task.o task-no-auto-map-kip.o irq_alloc.o libc_support.o long_msg.o long_msg-ux.o entry.o sighandler.o libk.a libamm.a liblmm.a libabi.a libjabi.a libcxx.a libkern.a -lutil /usr/bin/ld: entry.o: bad symbol index: 845556 entry.o: could not read symbols: File format not recognized My System: [rw@fetzi build-ux]$ gcc -v Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/3.2/specs Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --disable-checking --host=i386-redhat-linux --with-system-zlib --enable-__cxa_atexit Thread model: posix gcc version 3.2 20020903 (Red Hat Linux 8.0 3.2-7) [rw@fetzi build-ux]$ as -v GNU assembler version 2.13.90.0.2 (i386-redhat-linux) using BFD version 2.13.90.0.2 20020802 [rw@fetzi build-ux]$ ld -v GNU ld version 2.13.90.0.2 20020802 (from Redhat 8.0) Who can fix it ? Greetings Rudolf -- -- --------------------------------------------------------------- Rudolf Weber http://www.infnet.verein.de/~rw E-Mail: rfweber(a)tesionmail.de

3 2

about l4rm
by Yu Neil 22 Mar '03

22 Mar '03

hello every one: where i can find the source code of l4rm ( l4 region maper ) ? and should it work well on x86? Neil Yu _________________________________________________________________ 享用世界上最大的电子邮件系统― MSN Hotmail。 http://www.hotmail.com

2 1

(no subject)
by Yu Neil 20 Mar '03

20 Mar '03

hello every one: where i can find the source code of l4rm ( l4 region maper ) ? and should it work well on x86? Neil Yu _________________________________________________________________ 免费下载 MSN Explorer: http://explorer.msn.com/lccn/

1 0

Problem beim starten von l4linux auf fiasco
by Rudolf Weber 15 Mar '03

15 Mar '03

Hallo, Vorbemerkung: Das starten von glinux bringt einen Fehler überlappender Module, Der Fiasco beginnt da, wo glinux liegt, folgende Änderung habe ich gemacht: --- ./l4/kernel/fiasco/src/Makerules.BOOT.vorrudi 2003-02-25 11:34:43.000000000 +0100 +++ ./l4/kernel/fiasco/src/Makerules.BOOT 2003-03-15 16:13:58.000000000 +0100 @@ -7,7 +7,7 @@ $(BOOT): kernel.o $(OBJ_BOOT) $(LIBKERN) $(MINILIBC) @echo " ... Building boot image" $(LINK_MESSAGE) - $(VERBOSE)$(LD) -Ttext 0x300000 -o $@ $^ $(LIBGCC) \ + $(VERBOSE)$(LD) -Ttext 0x400000 -o $@ $^ $(LIBGCC) \ -defsym checksum_ro=0x`cut -f1 -d' ' $(CHECKSUM)` \ -defsym checksum_rw=0x`cut -f2 -d' ' $(CHECKSUM)` chmod 755 $@ Das starten von grub aus: title L4 fiasco x586 l4v2 L4Linux kernel /l4-20030315-x586/rmgr -sigma0 module /l4-20030315-x586/fiasco.400000 module /l4-20030315-x586/sigma0 module /l4-20030315-x586/glinux root=/dev/hda2 Der glinux startet dann auch, und ganz zum Schluß : Warning: unable to open an initial console Root: ptc is NULL for 00080030 (virt 00080030) and pdir 00400000 at eip: 0050c11b Die Warnung kommt aus dem l4linux als ein open("/dev/console/") schief geht. (offenbar ging wohl das mounten von / nicht :-() Was will mir der rest sagen ? pdir sieht ganz so aus wie die Start-Adresse von Fiasco ? Was habe ich nur falsch gemacht ? Gruß Rudolf Weber -- -- --------------------------------------------------------------- Rudolf Weber http://www.infnet.verein.de/~rw E-Mail: rfweber(a)tesionmail.de

2 1

RE: Problem beim starten von l4linux auf fiasco
by Volkmar Uhlig 15 Mar '03

15 Mar '03

If you have overlapping modules you have no chance whatsoever. So you better fix that ;) I'm not too familiar with the FIASCO kernel boot process, but usually the problem is GRUB, not the kernel. GRUB loads the modules after all loaded elf sections of the initial kernel (which is RMGR in your case). RMGR then tries to load the modules as ELF files and figures out that some elf sections collide with the module locations and barks. This was the reason to add the modaddr feature to GRUB allowing to relocate the load address in main mem. Simply try sticking a modaddr=0x<whatevermemoryyouhave-16MB> after the kernel=rmgr line. Note, that this is only available in the Dresden version of GRUB, not the GNU GRUB (at least AFAIK). So bake yourself a boot floppy from http://os.inf.tu-dresden.de/~fm3/grub.html. The Linux error message is due to the kernel-debugger console in L4Linux. At the end Linux tries to open the console but since it does not support all required features Lx simply falls back to VGA. So this error can be ignored even if it sounds scary. - Volkmar PS: Please post in English -- most subscribers of this list don't understand a word of German (except maybe "Bier" ;) > -----Original Message----- > From: Rudolf Weber [mailto:rfweber@tesionmail.de] > Sent: Saturday, March 15, 2003 8:21 PM > To: l4-hackers(a)os.inf.tu-dresden.de > Subject: Problem beim starten von l4linux auf fiasco > > > Hallo, > > Vorbemerkung: > Das starten von glinux bringt einen Fehler überlappender Module, > Der Fiasco beginnt da, wo glinux liegt, > folgende Änderung habe ich gemacht: > --- ./l4/kernel/fiasco/src/Makerules.BOOT.vorrudi 2003-02-25 > 11:34:43.000000000 +0100 > +++ ./l4/kernel/fiasco/src/Makerules.BOOT 2003-03-15 > 16:13:58.000000000 +0100 > @@ -7,7 +7,7 @@ > $(BOOT): kernel.o $(OBJ_BOOT) $(LIBKERN) $(MINILIBC) > @echo " ... Building boot image" > $(LINK_MESSAGE) > - $(VERBOSE)$(LD) -Ttext 0x300000 -o $@ $^ $(LIBGCC) \ > + $(VERBOSE)$(LD) -Ttext 0x400000 -o $@ $^ $(LIBGCC) \ > -defsym checksum_ro=0x`cut -f1 -d' ' $(CHECKSUM)` \ > -defsym checksum_rw=0x`cut -f2 -d' ' $(CHECKSUM)` > chmod 755 $@ > > Das starten von grub aus: > title L4 fiasco x586 l4v2 L4Linux > kernel /l4-20030315-x586/rmgr -sigma0 > module /l4-20030315-x586/fiasco.400000 > module /l4-20030315-x586/sigma0 > module /l4-20030315-x586/glinux root=/dev/hda2 > > Der glinux startet dann auch, und ganz zum Schluß : > > Warning: unable to open an initial console > Root: ptc is NULL for 00080030 (virt 00080030) > and pdir 00400000 at eip: 0050c11b > > Die Warnung kommt aus dem l4linux als ein open("/dev/console/") schief > geht. (offenbar ging wohl das mounten von / nicht :-() > > Was will mir der rest sagen ? pdir sieht ganz so aus wie die > Start-Adresse von Fiasco ? > > Was habe ich nur falsch gemacht ? > > Gruß > Rudolf Weber

1 0

RE: enabling/disabling trampoline
by Volkmar Uhlig 15 Mar '03

15 Mar '03

This sounds like a problem with RMGR. Try either running L4Linux not using the "flat clan structure" (make xconfig) or enable the "Jochen compatibility/bug mode" for Hazelnut. Have you tried running the binaries of the provided RC2 tarball? They also contain a working L4Linux binary. - Volkmar > -----Original Message----- > From: Cristiano Ligieri Pereira [mailto:cpereira@ics.uci.edu] > Sent: Thursday, March 13, 2003 9:01 PM > To: Espen Skoglund > Cc: l4-hackers > Subject: Re: enabling/disabling trampoline > > > > I'm not managing to get hazelnut working properly. It boots > but then when > it creates the Linux panics with the following messages: > > Kernel panic: failed to create ping pong task > In swapper task = no syncing > --- Linux panic --- > > Might be something stupid but I can't figure it out. > Any suggestion? > > Cristiano.

7 8

enabling/disabling trampoline
by Cristiano Ligieri Pereira 13 Mar '03

13 Mar '03

Hi all, I understand that the trampoline mechanism slows down the execution because it adds one more level of "indirection" when handling system calls on l4-linux. I would like to run some experiments with/without it but I'm not quite sure on how to achieve it,i.e., disable and enable it back again. Do I have to recompile the whole kernel? Do I have to recompile only the Linux application I will run on the top of l4-linux? What exactly do I have to do? I'm pasting part of a message from mailing list in which Frank Mehnert mention something about a patch, which I could not find. Any help is appreciated. Thanks, Cristiano. > Additional speedup for L4Linux could be achieved by patching the > syscalls in the libc by direct jumping into the emulib of the process > preventing the trampoline mechanism (int 0x80 => general protection => > int 0x30 => l4linux server). I don't know if such a patch if floating > around somewhere. See linux22/arch/l4/x86/emulib/int_entry.S, function > entry13. ------------------------------------------------------------ Cristiano Ligieri Pereira - http://www.ics.uci.edu/~cpereira

5 10

a article about l4 vm
by Yu Neil 28 Feb '03

28 Feb '03

hello: where i can find the article: "L4's Recursive Virtual Memory Model, Design and Implementation Issues of a L4 V4 Mapping Database" ? there is no link at www.l4ka.org Neil Yu _________________________________________________________________ 与联机的朋友进行交流，请使用 MSN Messenger: http://messenger.msn.com/cn

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

l4-hackers