Hi, All. Sorry for my late response. 2007/7/8, Jorge Torres <jorge.torres.maldonado@gmail.com>:

If you want Linux on top of Pistachio you should take a look at wombat, which is a paravirtualized Linux initially running on top of pistachio, you may want to check the ok-labs.com web site for this.

Thank you for information, I'll try it one of these days. I will use L4/Fiasco for the time being. 2007/7/8, Martin Pohlack <mp26@os.inf.tu-dresden.de>:

You seem to use the Dresden version of L4Linux which AFAIK only runs on L4/Fiasco, not (yet?) on L4Ka::Pistachio.

Please try it with L4/Fiasco.

Thanks, I tried to build L4Linux with L4Fiasco a short time ago. My build way is the same to Fiasco official page (http://os.inf.tu-dresden.de/fiasco/use.html). But the result was the same to before ... I'm glad if you read the following text about new errors. 2007/7/8, Adam Lackorzynski <adam@os.inf.tu-dresden.de>:

Please disable the "Run 'make headers_check' when building vmlinux" configuration option to make to error go away.

I've tried to build with setting disable it before, but I got following errors. -------------------------------------------------- ..... LD vmlinux SYSMAP System.map SYSMAP .tmp_System.map MODPOST vmlinux WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'rest_init' (at offset 0x40017e) and 'run_init_process' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4011ca) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4011f3) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x401252) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x401272) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4012b3) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4012eb) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401547) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text:l4env_v2p_init from .text between 'main' (at offset 0x40155b) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401634) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401680) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:initkmem_list3 from .text between 'set_up_list3s' (at offset 0x453c20) and '__kmem_cache_destroy' WARNING: vmlinux - Section mismatch: reference to .init.text:__alloc_bootmem from .text between 'vgacon_scrollback_startup' (at offset 0x4f4e7e) and 'vgacon_scrollback_update' WARNING: vmlinux - Section mismatch: reference to .init.data:logo_linux_clut224 from .text between 'fb_find_logo' (at offset 0x50089e) and 'fb_notifier_callback' WARNING: vmlinux - Section mismatch: reference to .init.data: from .text between 'l4fb_con_exit' (at offset 0x5032ca) and 'l4fb_dope_update_thread' WARNING: vmlinux - Section mismatch: reference to .init:.dyn from (offset 0x6d8010) WARNING: vmlinux - Section mismatch: reference to .init.text:.dyn from (offset 0x6d8050) WARNING: vmlinux - Section mismatch: reference to .init.data:.dyn from (offset 0x6d8054) WARNING: vmlinux - Section mismatch: reference to .init.setup:.dyn from (offset 0x7a1008) WARNING: vmlinux - Section mismatch: reference to .exit.text:.dyn from (offset 0x6d8060) COMPR vmlinux to vmlinuz26 LOCINST vmlinuz26 Building modules, stage 2. MODPOST 1388 modules WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'rest_init' (at offset 0x40017e) and 'run_init_process' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4011ca) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4011f3) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x401252) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x401272) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4012b3) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4012eb) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401547) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text:l4env_v2p_init from .text between 'main' (at offset 0x40155b) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401634) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401680) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:initkmem_list3 from .text between 'set_up_list3s' (at offset 0x453c20) and '__kmem_cache_destroy' WARNING: vmlinux - Section mismatch: reference to .init.text:__alloc_bootmem from .text between 'vgacon_scrollback_startup' (at offset 0x4f4e7e) and 'vgacon_scrollback_update' WARNING: vmlinux - Section mismatch: reference to .init.data:logo_linux_clut224 from .text between 'fb_find_logo' (at offset 0x50089e) and 'fb_notifier_callback' WARNING: vmlinux - Section mismatch: reference to .init.data: from .text between 'l4fb_con_exit' (at offset 0x5032ca) and 'l4fb_dope_update_thread' WARNING: vmlinux - Section mismatch: reference to .init:.dyn from (offset 0x6d8010) WARNING: vmlinux - Section mismatch: reference to .init.text:.dyn from (offset 0x6d8050) WARNING: vmlinux - Section mismatch: reference to .init.data:.dyn from (offset 0x6d8054) WARNING: vmlinux - Section mismatch: reference to .init.setup:.dyn from (offset 0x7a1008) WARNING: vmlinux - Section mismatch: reference to .exit.text:.dyn from (offset 0x6d8060) WARNING: drivers/atm/lanai.o - Section mismatch: reference to .init.text: from .text between 'sram_test_pass' (at offset 0x1b3) and 'sram_test_and_clear' WARNING: drivers/net/sis900.o - Section mismatch: reference to .init.text:sis900_mii_probe from .text between 'sis900_probe' (at offset 0x588) and 'sis900_default_phy' WARNING: drivers/net/sunhme.o - Section mismatch: reference to .init.text: from .text between 'happy_meal_pci_probe' (at offset 0x2cf8) and 'happy_meal_pci_remove' WARNING: drivers/net/tokenring/3c359.o - Section mismatch: reference to .init.text:xl_init from .text between 'xl_probe' (at offset 0x225) and 'xl_hw_reset' WARNING: "clear_user" [drivers/media/video/w9968cf.ko] undefined! WARNING: "l4lx_fpu_enabled" [drivers/md/xor.ko] undefined! WARNING: "l4lx_fpu_enabled" [drivers/md/raid456.ko] undefined! WARNING: "strnlen_user" [drivers/input/misc/uinput.ko] undefined! WARNING: "i8253_lock" [drivers/input/joystick/analog.ko] undefined! WARNING: "i8253_lock" [drivers/input/gameport/gameport.ko] undefined! make[1]: *** [__modpost] error 1 make: *** [modules] error 2 -------------------------------------------------- Does this header check always need to be off when I make L4Linux ? (Without it, we cannot build L4Linux ??) If so, what is these error's cause ? I feel that there is the issue in my L4 environment... Please theach if you like. Regards, Lin

Hi, Adam. Thank you for your quick response !

...

Does this header check always need to be off when I make L4Linux ?

I fixed that by now. (Update the source.)

I checked out the updated source and made sure the "header check" option is off in default. Thank you !

To fix that for now please just don't use modules. I fixed a couple of those above internally but I cannot fix all of them, e.g. the gameport driver looks suspicious to me.

I tried making L4Linux with the kernel configuration which was changed into like the following, then the make sequence was completed. Loadable module support ---> - [ ] Enable loadable module support Device Drives ---> - Input device support ---> - [ ] Joysticks ---> - Hardware I/O ports ---> - [ ] Gameport support But the "Section mismatch" errors in "MODPOST vmlinux" remain yet. -------------------------------------------------- MODPOST vmlinux WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'rest_init' (at offset 0x40017e) and 'run_init_process' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4011ca) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4011f3) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x401252) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x401272) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4012b3) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:boot_command_line from .text between 'main' (at offset 0x4012eb) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401547) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text:l4env_v2p_init from .text between 'main' (at offset 0x40155b) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401634) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'main' (at offset 0x401680) and 'l4x_linux_main_exit' WARNING: vmlinux - Section mismatch: reference to .init.data:initkmem_list3 from .text between 'set_up_list3s' (at offset 0x450a0c) and '__kmem_cache_destroy' WARNING: vmlinux - Section mismatch: reference to .init.text:__alloc_bootmem from .text between 'vgacon_scrollback_startup' (at offset 0x760f86) and 'vgacon_scrollback_update' WARNING: vmlinux - Section mismatch: reference to .init.data:logo_linux_clut224 from .text between 'fb_find_logo' (at offset 0x76c91a) and 'fb_notifier_callback' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'rivafb_init' (at offset 0x77cd9e) and 'nv3Busy' WARNING: vmlinux - Section mismatch: reference to .init.data: from .text between 'l4fb_con_exit' (at offset 0x7b20fa) and 'l4fb_dope_update_thread' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'happy_meal_pci_probe' (at offset 0x8b3aec) and 'happy_meal_pci_remove' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'sis900_probe' (at offset 0x8d8c6e) and 'sis900_default_phy' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'xl_probe' (at offset 0x96dde5) and 'xl_hw_reset' WARNING: vmlinux - Section mismatch: reference to .init.data: from .text between 'cpia_pp_attach' (at offset 0xa8bed0) and 'cpia_pp_init' WARNING: vmlinux - Section mismatch: reference to .init.data: from .text between 'cpia_pp_attach' (at offset 0xa8bf26) and 'cpia_pp_init' WARNING: vmlinux - Section mismatch: reference to .init.data: from .text between 'cpia_pp_init' (at offset 0xa8bf6d) and 'cpia_usb_complete' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text between 'sram_test_pass' (at offset 0xba6e23) and 'sram_test_and_clear' WARNING: vmlinux - Section mismatch: reference to .init.data: from .text between 'block2mtd_setup' (at offset 0xd73a93) and 'block2mtd_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .text after '__kprobes_text_start' (at offset 0x129cc1f) WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'alsa_card_dummy_exit' (at offset 0x1ad586e) and 'alsa_card_virmidi_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'alsa_card_virmidi_exit' (at offset 0x1ad5873) and 'alsa_card_mtpav_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'snd_mts64_module_exit' (at offset 0x1ad5894) and 'alsa_opl3_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'alsa_card_mpu401_exit' (at offset 0x1ad58ba) and 'alsa_vx_core_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'alsa_card_es18xx_exit' (at offset 0x1ad58e8) and 'alsa_card_opl3sa2_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'alsa_card_opl3sa2_exit' (at offset 0x1ad58ed) and 'alsa_cs4231_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'alsa_card_cs423x_exit' (at offset 0x1ad58f3) and 'alsa_cs4236_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'alsa_card_sb16_exit' (at offset 0x1ad5915) and 'alsa_sb16_exit' WARNING: vmlinux - Section mismatch: reference to .init.text: from .exit.text between 'alsa_card_sb16_exit' (at offset 0x1ad591c) and 'alsa_emu8000_exit' WARNING: vmlinux - Section mismatch: reference to .init:.dyn from (offset 0x171a010) WARNING: vmlinux - Section mismatch: reference to .init.text:.dyn from (offset 0x171a044) WARNING: vmlinux - Section mismatch: reference to .init.data:.dyn from (offset 0x171a048) WARNING: vmlinux - Section mismatch: reference to .init.data:.dyn from __setup_str_no_halt (offset 0x171a0c0) WARNING: vmlinux - Section mismatch: reference to .init.setup:.dyn from __setup_mca_pentium (offset 0x171a0c4) WARNING: vmlinux - Section mismatch: reference to .init.text:.dyn from no_387 (offset 0x171a0c8) WARNING: vmlinux - Section mismatch: reference to .init.data:.dyn from __setup_str_no_387 (offset 0x171a0cc) WARNING: vmlinux - Section mismatch: reference to .init.setup:.dyn from __setup_no_387 (offset 0x171a0d0) WARNING: vmlinux - Section mismatch: reference to .exit.data:.dyn from (offset 0x171a058) WARNING: vmlinux - Section mismatch: reference to .exit.text:.dyn from (offset 0x1c52004) COMPR vmlinux to vmlinuz26 LOCINST vmlinuz26 [root@lin l4linux-2.6]# -------------------------------------------------- After make, the vmlinux(about 101.7MB) and the vmlinuz26(about 32.8M) are created. Are these errors always printed when we make L4Linux ? (I don't think so...) And I'm trying to boot the made L4Linux and Fiasco with GRUB. My GRUB doesn't have "modaddr" command, so I want to use "single-image-mode" written in ML log (http://www.mail-archive.com/l4-hackers@os.inf.tu-dresden.de/msg02591.html). I'm reading "README" and "modules.list" now, but I cannot understand how to use it yet. I'll try harder... Regards, Lin

...

But the "Section mismatch" errors in "MODPOST vmlinux" remain yet.

Most of those are from Linux, don't take them too serious.

Ok, I see. I wrote "error" instead of "WARNING", but it was a mistake. Sorry...

...

After make, the vmlinux(about 101.7MB) and the vmlinuz26(about 32.8M) are created.

Over 100MB is too much. Can you please post your .config from the L4Linux build directory?!

My ".config" file may be too strange for you because I am a beginner in configuring or building kernels. I probably added many vain components to the kenel unconsciously.

...

And I'm trying to boot the made L4Linux and Fiasco with GRUB. My GRUB doesn't have "modaddr" command, so I want to use

I recommend using the -modaddr switch to bootstrap which does the same as the modaddr GRUB command.

Thanks, but I'd already gained the GRUB which has the "modaddr" command by using the patch (grub-0.97-os.1.diff.gz) you made. And I tried to boot L4Linux/Fiasco(+l4dope) with it, however it failed. The kernel configuration is equal to the posted file. The "menu.lst" in the "/boot/grub" directory and the loader configuration file "linux26.cfg" are as follows. [/boot/grub/menu.lst] ------------------------------ title L4Linux2.6/Fiasco root (hd0,0) kernel /boot/L4Linux/bootstrap modaddr 0x06000000 module /boot/L4Linux/fiasco -nowait -nokdbi -serial_esc -comspeed 115200 -comport 1 module /boot/L4Linux/sigma0 module /boot/L4Linux/roottask task modname "bmodfs" attached 4 modules module /boot/L4Linux/events module /boot/L4Linux/names --events module /boot/L4Linux/log --events module /boot/L4Linux/dm_phys --events module /boot/L4Linux/simple_ts -t 300 --events module /boot/L4Linux/rtc --events module /boot/L4Linux/l4io --noirq --events module /boot/L4Linux/bmodfs module /boot/L4Linux/vmlinuz26 module /boot/L4Linux/libloader.s.so module /boot/L4Linux/linux26.cfg module /boot/L4Linux/libld-l4.s.so module /boot/L4Linux/l4dope --l4io module /boot/L4Linux/loader --fprov=BMODFS linux26.cfg vbeset 0x117 ----------------------------------------------------- [linux26.cfg] --------------------------------------- verbose 0 task "vmlinuz26" "mem=256M video=l4fb root=/dev/hda1" all_sects_writable allow_vga ----------------------------------------------------- After GRUB loads modules, a black mouse cursor and a "Linux console" window are displayed (the window often aren't). But I cannot use my keyboard and mouse (the cusor won't move and keys produce nothing). Besides, the console message stops on the way (stop points are different every time like following). ------------------------------------------------------------------ ... l4fb: DOpE framebuffer at 0x12005000, size 600k l4fb: mode is 640x480x16, linelength=1280, pages=0 l4fb: directcolor: size=0:5:6:5, shift=0:11:5:0 Console: switching to colour frame buffer device 80x30 <--- [stop] l4fb0: l4fb L4 frame buffer device (refresh: 100ms) vga16fb: mapped to 0x000a0000 vga16fb: mapped to 0x000a0000 fb1: VGA16 VGA frame buffer device isapnp: Scanning for PnP cards... isapnp: No Plug & Play device found RocketPort device driver module, version 2.09, 12-June-2003 No rocketport ports found; unloading driver. Cyclades driver 2.4 <--- [stop] built Jul 11 2007 05:10:13 <--- [stop] ------------------------------------------------------------------ I cannot understand the cause of this behavior... I'm reading the ML logs, but I don't solve it yet. May I request your advice ? Regards, Lin

Hi, Adam. Sorry for my late answer.

...

My ".config" file may be too strange for you because I am a beginner in configuring or building kernels. I probably added many vain components to the kenel unconsciously.

Lots of stuff switched on (too much). Please use the attached config as a basis, this one should be much better.

Thanks too much, I'm going to compare my bad configuration with the given ... By Using it, it takes about 8 minutes to make. The size of "vmlinux" is about 17MB, and "vmlinuz32" is about 5MB. (It takes about 45 minutes to make and the created binaries are too large if I use my old configuration !)

...

Thanks, but I'd already gained the GRUB which has the "modaddr" command by using the patch (grub-0.97-os.1.diff.gz) you made. And I tried to boot L4Linux/Fiasco(+l4dope) with it, however it failed. The kernel configuration is equal to the posted file. The "menu.lst" in the "/boot/grub" directory and the loader configuration file "linux26.cfg" are as follows.

[/boot/grub/menu.lst] ------------------------------ title L4Linux2.6/Fiasco root (hd0,0) kernel /boot/L4Linux/bootstrap modaddr 0x06000000 module /boot/L4Linux/fiasco -nowait -nokdbi -serial_esc -comspeed 115200 -comport 1 module /boot/L4Linux/sigma0 module /boot/L4Linux/roottask task modname "bmodfs" attached 4 modules module /boot/L4Linux/events module /boot/L4Linux/names --events module /boot/L4Linux/log --events module /boot/L4Linux/dm_phys --events module /boot/L4Linux/simple_ts -t 300 --events module /boot/L4Linux/rtc --events module /boot/L4Linux/l4io --noirq --events module /boot/L4Linux/bmodfs module /boot/L4Linux/vmlinuz26 module /boot/L4Linux/libloader.s.so module /boot/L4Linux/linux26.cfg module /boot/L4Linux/libld-l4.s.so module /boot/L4Linux/l4dope --l4io module /boot/L4Linux/loader --fprov=BMODFS linux26.cfg vbeset 0x117 -----------------------------------------------------

[linux26.cfg] --------------------------------------- verbose 0 task "vmlinuz26" "mem=256M video=l4fb root=/dev/hda1" all_sects_writable allow_vga -----------------------------------------------------

After GRUB loads modules, a black mouse cursor and a "Linux console" window are displayed (the window often aren't). But I cannot use my keyboard and mouse (the cusor won't move and keys produce nothing). Besides, the console message stops on the way (stop points are different every time like following).

------------------------------------------------------------------ ... l4fb: DOpE framebuffer at 0x12005000, size 600k l4fb: mode is 640x480x16, linelength=1280, pages=0 l4fb: directcolor: size=0:5:6:5, shift=0:11:5:0 Console: switching to colour frame buffer device 80x30 <--- [stop] l4fb0: l4fb L4 frame buffer device (refresh: 100ms) vga16fb: mapped to 0x000a0000 vga16fb: mapped to 0x000a0000 fb1: VGA16 VGA frame buffer device isapnp: Scanning for PnP cards... isapnp: No Plug & Play device found RocketPort device driver module, version 2.09, 12-June-2003 No rocketport ports found; unloading driver. Cyclades driver 2.4 <--- [stop] built Jul 11 2007 05:10:13 <--- [stop] ------------------------------------------------------------------

I cannot understand the cause of this behavior... I'm reading the ML logs, but I don't solve it yet.

Please try with the updated L4Linux configuration. I can't spot an obvious mistake in the configuration you posted (except it's -nokdb instead of -nokdbi)

I updated L4Linux (using the attached config) and retried, but it also stopped. To see the setup sequences, I tried using the serial output way. It always stops here: ------------------------------------------------------------------------------- l4lx | ======> L4Linux 2.6 starting... <======== l4lx | Linux version 2.6.21-l4 (khayashi@khayashi1.lab.namikilab.tuat.ac.jp) l4lx : (gcc version 3.4.6 20060404 (Red Hat 3.4.6-4)) #7 Fri Jul 13 06:05:2 l4lx : 5 JST 2007 l4lx | Binary name: vmlinuz26 l4lx | Kernel command line (4 args): console=ttyS1,115200 init=1 mem=256M ro l4lx : ot=/dev/hda1 l4lx | Image: 00400000 - 0079d000 [3700 KiB]. l4lx | Areas: Text: 00400000 - 006d1000 [2884kB] (a bit longer) l4lx | Data: 006d1000 - 006ef2cc [120kB] l4lx | Initdata: 006f2000 - 0072e0a0 [240kB] l4lx | BSS: 00730000 - 0079aee0 [427kB] l4lx | l4lx_thread_create: Created thread 0d.03 (tamer) l4lx | Using tamed mode. l4lx | l4env_linux_startup thread 4. l4lx | l4lx_thread_create: Created thread 0d.04 (server) l4lx | main thread will be 0d.04 l4lx | l4env_register_pointer_section: addr = 00400000 size = 3788800 l4lx | sec-w-init: virt: 0x00400000 to 0x0079cfff [3700 KiB] l4lx | Noncontiguous region for sec-w-init l4lx | sec-w-init: Number of physical regions: 5, 3788800 Bytes l4lx | sec-w-init: 1: Phys: 0x002e2000 to 0x00381000, Size: 651264 l4lx | sec-w-init: 2: Phys: 0x01419000 to 0x01500000, Size: 946176 l4lx | sec-w-init: 3: Phys: 0x001ed000 to 0x002d0000, Size: 929792 l4lx | sec-w-init: 4: Phys: 0x01344000 to 0x01400000, Size: 770048 l4lx | sec-w-init: 5: Phys: 0x01288000 to 0x01300000, Size: 491520 l4lx | main thread: received startup message. l4lx | Main thread running, waiting... l4lx | setup_l4env_memory: Forcing superpages for main memory l4lx | Main memory size: 256MB l4lx | Main memory: virt: 0x00800000 to 0x107fffff [262144 KiB] l4lx | Main memory: Number of physical regions: 1, 268435456 Bytes l4lx | Main memory: 1: Phys: 0x07800000 to 0x17800000, Size: 268435456 l4lx | Filling lower ptabs... l4lx | Done (1087 entries). l4lx | l4lx_thread_create: Created thread 0d.05 (timer.i0) l4lx | l4lx_thread_create: Created thread 0d.06 (Idler) <---[here!] ------------------------------------------------------------------------------- I read Log and found a state which is similar to mine, but it seems not to be resolved. If I use the DOpE graphic system, the mouse cursor is able to be moved but the first window don't appear in this time. I think that it also stops here when I use the DOpE. How do you think ? Regards, Lin

Please enable EARLY_PRINTK in the L4Linux configuration (under Kernel hacking) and add "earlyprintk=yes" to the Linux kernel command line. This should give us more output.

I enabled them and found new messages from kernel. ---------------------------------------------------------------------- ..... Linux version 2.6.21-l4 () (gcc version 3.4.6 20060404 (Red Hat 3.4.6-4)) #9 Sat Jul 14 05:08:25 JST 2007 CPU: Vendor unknown, using generic init. CPU: Your system may be unstable. dm_phys: 0000000000800000 - 0000000010800000 (usable) 256MB memory available. Zone PFN ranges: DMA 0 -> 512 Normal 512 -> 67584 early_node_map[1] active PFN ranges 0: 0 -> 67584 Allocating PCI resources starting at 20000000 (gap: 10800000:ef800000) Built 1 zonelists. Total pages: 67056 Kernel command line: earlyprintk=yes console=ttyS1,115200 init=1 mem=256M root=/dev/hda1 Initializing CPU#0 PID hash table entries: 2048 (order: 11, 8192 bytes) Detected 1995.013 MHz processor. disabling early console l4lx | l4lx_thread_create: Created thread 0d.05 (timer.i0) l4lx | l4lx_thread_create: Created thread 0d.06 (Idler) ---------------------------------------------------------------------- Is the message "CPU: Your system may be unstable." abnormal ? I'm running L4Linux on LV-673 (with Pentium M cpu). (http://www.commell.com.tw/Product/SBC/LV-673.HTM) Of course, I set the "target processor" Pentium M in fiasco and L4Linux configurations. I attached the log which all of serial output are written in. I could not find any messages which is relate to this freeze. Do you have anything about it ? Regards, Lin

...

I attached the log which all of serial output are written in. I could not find any messages which is relate to this freeze. Do you have anything about it ?

The kernel command line says that the kernel output goes to the second serial interface. Obviously you won't see more Linux kernel messages in the Fiasco debug log as well as in any Dope window. Do you see something on ttyS1? If you want to see more messages in the debug log, enable the "serial driver" under "stub drivers" in L4Linux and replace the console=... option with console=ttyLv0 If you leave it out or write console=tty0 the output will go to the Dope window.

There is no message on ttyS1. I changed "console=ttyS1,1152000" to "console=ttyLv0" and enable serial console support on L4Linux configration. The result is the following (when using DOpE). -------------------------------------------------- ..... Linux version 2.6.21-l4 () (gcc version 3.4.6 20060404 (Red Hat 3.4.6-4)) #11 Sat Jul 14 15:48:16 JST 2007 CPU: Vendor unknown, using generic init. CPU: Your system may be unstable. dm_phys: 0000000000800000 - 0000000010800000 (usable) 256MB memory available. Zone PFN ranges: DMA 0 -> 512 Normal 512 -> 67584 early_node_map[1] active PFN ranges 0: 0 -> 67584 Allocating PCI resources starting at 20000000 (gap: 10800000:ef800000) Built 1 zonelists. Total pages: 67056 Kernel command line: earlyprintk=yes console=ttyLv0 mem=256M video=l4fb root=/dev/hda1 Initializing CPU#0 PID hash table entries: 2048 (order: 11, 8192 bytes) Detected 1995.006 MHz processor. disabling early console Console: colour dummy device 80x25 l4ser_init_ports l4ser: input not enabled! Dentry cache hash table entries: 65536 (order: 6, 262144 bytes) Inode-cache hash table entries: 32768 (order: 5, 131072 bytes) Memory: 259456k/270336k available (2346k kernel code, 2536k reserved, 710k data, 244k init, 0k highmem) virtual kernel memory layout: fixmap : 0xffffd000 - 0xfffff000 ( 8 kB) vmalloc : 0x18000000 - 0x20000000 ( 128 MB) lowmem : 0x00000000 - 0x10800000 ( 264 MB) .init : 0x00700000 - 0x0073d0c0 ( 244 kB) .data : 0x0064a9c4 - 0x006fc34c ( 710 kB) .text : 0x00400000 - 0x0064a9c4 (2346 kB) l4lx_irq_timer_startup(0) l4lx | l4lx_thread_create: Created thread 0e.05 (timer.i0) timer_irq_thread: Starting timer IRQ thread. Calibrating delay using timer specific routine.. 3990.83 BogoMIPS (lpj=19954193) Mount-cache hash table entries: 512 Compat vDSO mapped to 0040e000. CPU: GenuineIntel Intel(R) Pentium(R) M processor 2.00GHz stepping 08 NET: Registered protocol family 16 PCI: Using configuration type 1 Setting up standard PCI resources SCSI subsystem initialized PCI: Probing PCI hardware PCI quirk: region 0400-047f claimed by ICH6 ACPI/GPIO/TCO PCI quirk: region 0480-04bf claimed by ICH6 GPIO PCI: Transparent bridge - 0000:00:1e.0 PCI: Bus #04 (-#07) is hidden behind transparent bridge #03 (-#04) (try 'pci=assign-busses') Please report the result to linux-kernel to fix this permanently PCI: Ignore bogus resource 6 [0:0] of 0000:00:02.0 PCI: Bridge: 0000:00:1c.0 IO window: b000-bfff MEM window: d0000000-d00fffff PREFETCH window: 24000000-240fffff PCI: Bridge: 0000:00:1c.1 IO window: c000-cfff MEM window: d0100000-d01fffff PREFETCH window: 24100000-241fffff PCI: Bus 4, cardbus bridge: 0000:03:0d.0 Time: tsc clocksource has been installed. IO window: 0000d000-0000d0ff IO window: 0000d400-0000d4ff PREFETCH window: 20000000-23ffffff MEM window: 28000000-2bffffff PCI: Bridge: 0000:00:1e.0 IO window: d000-dfff MEM window: d0200000-d02fffff PREFETCH window: 20000000-23ffffff NET: Registered protocol family 2 l4lx | l4lx_thread_create: Created thread 0e.06 (Idler) l4dope | input: ImPS/2 Logitech Wheel Mouse on isa0060/serio1 l4dope | input: AT Raw Set 2 keyboard on isa0060/serio0 l4dope | input: PC Speaker l4dope | input: L4 input event injector l4dope | L4 input event injector: IRQ handler up l4dope | connect "ImPS/2 Logitech Wheel Mouse", isa0060/serio1/input0 l4dope | connect "AT Raw Set 2 keyboard", isa0060/serio0/input0 l4dope | connect "PC Speaker", isa0061/input0 l4dope | connect "L4 input event injector", l4/sys l4dope | SharedMem(alloc): hl.raw=e0000, id=69, size=800 l4dope | SharedMem(get_adr): address = 0x6000 l4dope | SharedMem(alloc): hl.raw=e0000, id=6a, size=800 l4dope | SharedMem(get_adr): address = 0x7000 l4dope | SharedMem(alloc): hl.raw=e0000, id=6b, size=800 l4dope | SharedMem(get_adr): address = 0xd000 l4dope | SharedMem(alloc): hl.raw=e0000, id=6c, size=800 l4dope | SharedMem(get_adr): address = 0xe000 l4dope | SharedMem(alloc): hl.raw=e0000, id=6d, size=800 l4dope | SharedMem(get_adr): address = 0xf000 l4dope | SharedMem(alloc): hl.raw=e0000, id=6f, size=800 l4dope | SharedMem(get_adr): address = 0x16000 l4dope | SharedMem(alloc): hl.raw=e0000, id=70, size=800 l4dope | SharedMem(get_adr): address = 0x17000 l4dope | SharedMem(alloc): hl.raw=e0000, id=75, size=484 l4dope | SharedMem(get_adr): address = 0x21000 l4dope | dope: paddr=c0000000 size=7872KiB l4dope | Mapped video memory at c0000000 to 00800000+000000 [7872kB] via L4IO l4dope | mapping: vaddr=0x800000 size=8060928(0x7b0000) offset=0(0x0) l4dope | Frame buffer base: 0x800000 l4dope | Resolution: 1024x768x16 l4dope | Bytes per scanline: 2048 l4dope | Current video mode is 1024x768 red=11:5 green=5:6 blue=0:5 res=0:0 (stop here) -------------------------------------------------- Without DOpE, the messages of "l4dope" (last 32 lines) don't exist and others are the same. I'm not able to find any problems in these messages... It seems there is no too serious message in this Log. Could I request your opinion again ? Regards, Lin

But the DOpE window is opened? And is there something in the window?

The DOpE window doesn't appear (while I use the ".config" given from you). The cause of it is not clear...

I have no clear idea right now what could be wrong. Could you post your .config, that you are using right now, again?

The attached configuration is used right now, which uses the DOpE. The kernel configuration file and the GRUB menu are the following. [config] --------------------------------------------------------------------------- verbose 0 task "vmlinuz26" "earlyprintk=yes console=ttyLv0 mem=256M video=l4fb root=/dev/hda1" all_sects_writable allow_vga ------------------------------------------------------------------------------------ [menu.lst] ------------------------------------------------------------------------------- title L4Linux2.6/Fiasco (DOpE) root (hd0,0) kernel /boot/L4Linux/bootstrap modaddr 0x06000000 module /boot/L4Linux/fiasco -nowait -nokdb -serial_esc -comspeed 115200 -comport 1 module /boot/L4Linux/sigma0 module /boot/L4Linux/roottask task modname "bmodfs" attached 4 modules module /boot/L4Linux/names module /boot/L4Linux/log module /boot/L4Linux/dm_phys module /boot/L4Linux/simple_ts -t 300 module /boot/L4Linux/rtc module /boot/L4Linux/l4io --noirq module /boot/L4Linux/bmodfs module /boot/L4Linux/vmlinuz26 module /boot/L4Linux/config module /boot/L4Linux/libloader.s.so module /boot/L4Linux/libld-l4.s.so module /boot/L4Linux/l4dope --l4io module /boot/L4Linux/loader --fprov=BMODFS config vbeset 0x117 ------------------------------------------------------------------------------------------

You could also try to disable PCI just for a test and see if it gets to the point where Linux wants to mount the root partition and can't because no driver is there. Maybe making earlyprintk to earlyprintk=keep makes a difference too.

I disabled PCI and retried, but the setup stops at the same line (some PCI related　messages disappear). And it seem that the sequence doesn't come to the step of mounting root partition yet　because the results are the same though I change the "root=..." value. Regards, Lin

Ok, so some other thing. Can you please leave out DOpE and try without? Please add VGA and PS/2/Keyboard support to your L4Linux configuration. You can also use the x86-native_defconfig configuration to achieve the same. Hopefully the behaviour will be different with such a setup. If it still hangs I would like to see a Alt-Print-p and Alt-Print-t (SysRQs...).

I see nothing obvious wrong in your attached .config.

Thanks for your checking .config. I quitted using DOpE, and enabled VGA and keyboard like the following. ###### L4Linux Kernel Configuration ###### o Device Drivers ---> o Input device support ---> o [*] Keyboards ---> o --- Keyboards o <*> AT keyboard o Graphics support ---> o <*> Support for frame buffer devices o <*> VGA 16-color graphics support o [*] VESA VGA graphics support o Console display driver support ---> o [*] VGA text console ########################################## But nothing changed. Then, typing "Alt+SysRq+p" and "Alt+SysRq+t" (dump ?) return nothing. It is not clear whether the keyboard is recognized or not. (Only the "NumLock" key works.) Are there any other works ? Regards, Lin

Ok, looks like I need to have a look. Can you make all binaries and files that you load with GRUB + the menu.lst entry available to me? Then I'll have a look myself...

Thanks, I attached the files to the mail sent to you only (because of its large size). It is a tar compressed archive (L4Linux.tar) and the size of it is about 13MB. o L4Linuxs : This directory includes binaries. The path is "/boot/L4Linuxs". o menu.lst : The GRUB menu list file (grub.conf). By the way, the L4Linux binary without DOpE made with my old configuration (like one I sent few days ago, which produces a large vmlinux binary) doesn't stop there. But it prints some errors on the way and then aborts (because of some modules ?). So, I think the cause of hanging is in the L4Linux configuration about some drivers, is it right ? Regards, Lin

It hangs for me as well. Rebuilding my L4Linux tree with your configuration and using gcc-4.1 works for me. Building with gcc-3.4 hangs at the exact same place. Could you try another gcc version (preferably 4.1)?

I rebuilded the L4Linux with gcc-4.1.1, and made sure the boot sequence with it doesn't hang ! Thanks. BUT, I get a new error (the sequence isn't completed). ############################# L4Linux Boot Log ############################# ..... l4lx | l4lx_thread_create: Created thread 0d.06 (Idler) <---(it stops here before) Time: tsc clocksource has been installed. IP route cache hash table entries: 4096 (order: 2, 16384 bytes) TCP established hash table entries: 16384 (order: 5, 131072 bytes) TCP bind hash table entries: 16384 (order: 4, 65536 bytes) TCP: Hash tables configured (established 16384 bind 16384) TCP reno registered io scheduler noop registered io scheduler anticipatory registered (default) io scheduler deadline registered io scheduler cfq registered L4 serial driver l4ser_init_ports l4ser: input not enabled! ttyLv0 at MMIO 0x1 (irq = 0) is a L4 RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2 ide: Assuming 50MHz system bus speed for PIO modes; override with idebus=xx hda: HTS541080G9SA00, ATA DISK drive l4lx | l4lx_irq_dev_startup_hw: Starting IRQ thread for IRQ 14. l4lx | l4lx_thread_create: Created thread 0d.07 (IRQ14) ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 hda: max request size: 512KiB hda: Host Protected Area detected. current capacity is 156299375 sectors (80025 MB) native capacity is 156301488 sectors (80026 MB) hda: Host Protected Area disabled. hda: 156301488 sectors (80026 MB) w/7538KiB Cache, CHS=16383/255/63 hda: cache flushes supported hda: hda1 hda2 hda3 TCP cubic registered NET: Registered protocol family 1 NET: Registered protocol family 17 kjournald starting. Commit interval 5 seconds EXT3 FS on hda1, internal journal EXT3-fs: mounted filesystem with ordered data mode. VFS: Mounted root (ext3 filesystem). modprobe: FATAL: Could not load /lib/modules/2.6.21-l4/modules.dep: No such file or directory udevd[73]: init_uevent_netlink_sock: bind failed: No such file or directory udevd[73]: main: error initializing netlink socket Restarting system. l4lx | Terminating L4Linux. simplets| Exit D.02 ############################################################################ The last 6 massages may be notable. I cannot estimate the cause of this error ... (It seems the cause is related to "netlink socket".) I found a ML log similar to this here : http://os.inf.tu-dresden.de/pipermail/l4-hackers/2006/002650.html but not solve yet. Do you have any idea ? Regards, Lin

Do you get a prompt when you add a 'init=/bin/sh' to the Linux kernel command line? What type of Red Hat are you using? Seeing this for the first time.

I don't get a prompt. I use "Fedora 7 (2.6.21-1.3194.fc7)" in "/dev/hda1". And the "/bin/sh" in it is GNU bash 3.2.9 (i686-redhat-linux-gnu). The kernel command line values and their massage logs are the following: ##################################################################### (1) task "vmlinuz26" "earlyprintk=yes console=ttyLv0 mem=256MB root=/dev/hda1" ..... VFS: Mounted root (ext3 filesystem). modprobe: FATAL: Could not load /lib/modules/2.6.21-l4/modules.dep: No such file or directory udevd[73]: init_uevent_netlink_sock: bind failed: No such file or directory udevd[73]: main: error initializing netlink socket Restarting system. l4lx | Terminating L4Linux. simplets| Exit D.02 (2) task "vmlinuz26" "earlyprintk=yes console=ttyLv0 mem=256MB init=/bin/sh root=/dev/hda1" ..... VFS: Mounted root (ext3 filesystem). Kernel panic - not syncing: Attempted to kill init! panic: going to sleep forever, bye l4lx | panic: going to sleep forever, bye #################################################################### What should I set to "init=" generally ? Without it, is the default init run automatically ? (but it looks make some errors in (1).) And, do programs such as "/bin/sh" which runs on Fedora run also on L4Linux ? Regards, Lin

...

What should I set to "init=" generally ? Without it, is the default init run automatically ? (but it looks make some errors in (1).)

Yes, if not setting it the default init will be started.

...

And, do programs such as "/bin/sh" which runs on Fedora run also on L4Linux ?

Yes, L4Linux is binary compatible with native Linux.

Ok, I see. Then, the init (of my Fedora on /sbin/init) should work on L4Linux normally, is it right ?

You should try a ramdisk first to make sure something works. You can get one of the L4Linux site.

I had tried it but it also stops with the following message. ######################################################################## ..... VFS: Mounted root (ext2 filesystem). <---(mounted on RAM ?) INIT: version 2.84 booting <---(it is the init on RAMDISK ?) Loading /etc/console/boottime.kmap.gz Activating swap. System time was Wed Jul 18 20:06:17 UTC 2007. Setting the System Clock using the Hardware Clock as reference... System Clock set. System local time is now Wed Jul 18 20:06:19 UTC 2007. Checking all file systems... fsck 1.27 (8-Mar-2002) Setting kernel variables. Loading the saved-state of the serial devices... Mounting local filesystems... nothing was mounted Setting up IP spoofing protection: rp_filter. Configuring network interfaces: Operation failed. done. No IP found! at /etc/rcS.d/S40sethostname line 30. Starting portmap daemon: portmap. Setting the System Clock using the Hardware Clock as reference... System Clock set. Local time: Wed Jul 18 22:06:20 CEST 2007 Cleaning: /tmp /var/lock /var/run. INIT: Entering runlevel: 2 No user config loaded! <---(stops here...) ######################################################################## The last message (with "L4") is not found in Google... Regards, Lin

The login-prompt appears on tty1, so either in the dope window or on the vga screen, depending what you are using.

Does it appear on tty1 only ? I'm using serial console(ttyLv0) now, and see the same messages as tty1. In this case, should I see login-prompt on serial console too ? I have not gotten any prompts until now. My current settings is ... [L4Linux Configuration] Using given .config. (L4Linux DOpE - disable) (L4Linux serial console support - enable ) (Linux keyboard and VGA support - disable) (Linux serial support - disable) [GRUB menu.lst] ... module /boot/L4Linux/roottask task modname "bmodfs" attached 5 modules ... module /boot/L4Linux/bmodfs module /boot/L4Linux/libloader.s.so module /boot/L4Linux/libld-l4.s.so module /boot/L4Linux/vmlinuz26 module /boot/L4Linux/config module /boot/L4Linux/ramdisk ... [Kernel command line] modpath "/boot/L4Linux" task "vmlinuz26" "earlyprintk=yes load_ramdisk=1 ramdisk_size=16384 console=ttyLv0 mem=256M root=/dev/ram l4env_rd=/boot/L4Linux/ramdisk" all_sects_writable I'm using ramdisk (16MB) download from http://os.inf.tu-dresden.de/L4/LinuxOnL4/download/drops-rd.rd and gunziped. The result is the same as the last mail. (Stops at "No user config loaded!") Are there any problem in my settings ? Regards, Lin

...

I'm using serial console(ttyLv0) now, and see the same messages as tty1. In this case, should I see login-prompt on serial console too ?

No.

On Linux (and most Unices) there's a file called /etc/inittab which tells /sbin/init what to do. There are entries which define how the getties are started. The getties are those programs that give you the prompt. They have an argument that specifies the console they work on. The inittab in the ramdisk defines tty1, tty2, tty3, tty4, tty5 and tty6 for the getties. If you want to have the prompt on ttyLv0, you need to modify the /etc/inittab on the ramdisk accordingly.

...

I have not gotten any prompts until now. My current settings is ...

[L4Linux Configuration] Using given .config. (L4Linux DOpE - disable) (L4Linux serial console support - enable ) (Linux keyboard and VGA support - disable) (Linux serial support - disable)

That config gives you no tty1. You either need to enable dope to get a framebuffer and thus tty1 or without dope enable the keyboard and VGA to get tty1.

Thanks for your particular explanation. I have not seen the boot process of Linux in detail until now... I'll study harder. I enabled keyboard and VGA support in Linux configuration and changed "console=ttyLv0" to "console=tty1" in the kernel command line. Then, I got the login prompt on tty1 and could use bash of the ramdisk on L4Linux finally ! :) It seems to operate well. I also tried using DOpE on tty1 without keyboard and VGA, then it works. I would like to express to you my deepest gratitude for having spent much time on taking care of my work. Lastly, I have a small question. With keyboard and VGA support enabled, I get the following prompt on the serial console (not on tty1). (When I use DOpE, it doesn't appear because keyboard and VGA are disabled.) ######################################################################### ..... l4lx | l4env_rd_path: /boot/L4Linux/ramdisk l4lx | Loading: /boot/L4Linux/ramdisk l4lx | INITRD: Size of RAMdisk is 16384KiB l4lx | RAMdisk from 10800000 to 11800000 [16384KiB] l4lx | l4lx_thread_create: Created thread 0d.05 (timer.i0) l4lx | l4lx_thread_create: Created thread 0d.06 (Idler) l4lx | l4lx_irq_dev_startup_hw: Starting IRQ thread for IRQ 14. l4lx | l4lx_thread_create: Created thread 0d.07 (IRQ14) l4lx | l4lx_irq_dev_startup_hw: Starting IRQ thread for IRQ 12. l4lx | l4lx_thread_create: Created thread 0d.08 (IRQ12) l4lx | l4lx_irq_dev_startup_hw: Starting IRQ thread for IRQ 1. l4lx | l4lx_thread_create: Created thread 0d.09 (IRQ1) --l4io_search_mem_region nope----------------------------IP: 0040e87d [l4lx.server] (d.04) jdb: ######################################################################### If I type "g" on above prompt, the sequence re-runs and then I get the login prompt finally. The messages on tty1 at this time are the following. ######################################################################### ..... io scheduler noop registerd io scheduler anticipatory registered (default) io scheduler deadline registered io scheduler cfq registered vga16fb: mapped to 0x000a0000 ######################################################################### Is this input always needed ? I don't think so. Are there any problem in my kernel configuration or boot settings ? My now kernel configuration is the same as the recent mail (based on the given .config) :

###### L4Linux Kernel Configuration ######

o Device Drivers ---> o Input device support ---> o [*] Keyboards ---> o --- Keyboards o <*> AT keyboard o Graphics support ---> o <*> Support for frame buffer devices o <*> VGA 16-color graphics support o [*] VESA VGA graphics support o Console display driver support ---> o [*] VGA text console

##########################################

Thanks and Regards, Lin