On Fri Jul 13, 2007 at 14:57:57 +0900, Lin wrote:
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 !)
That's a lot better.
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 ?
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. Adam -- Adam adam@os.inf.tu-dresden.de Lackorzynski http://os.inf.tu-dresden.de/~adam/