September 2004 - l4-hackers - example.com Site

28 Sep '04

Hi all, I am having a problem with the new name_server from the l4vfs package. I load it the name_server module with grub: kernel (nd)/fiasco/rmgr -sigma0 task modname "L4Linux task" modaddr 0x02000000 module (nd)/fiasco/fiasco -nowait -nokdb -nofancy -serial -comspeed 115200 -comport 1 module (nd)/fiasco/sigma0 module (nd)/fiasco/names module (nd)/fiasco/log module (nd)/fiasco/dm_phys module (nd)/fiasco/l4io --noirq module (nd)/fiasco/tftp -i -s 192.168.1.4 module (nd)/fiasco/simple_ts module (nd)/fiasco/l4exec module (nd)/fiasco/l4dope --l4io module (nd)/fiasco/name_server module (nd)/fiasco/loader (nd)/fiasco/dope-l4linux.cfg vbeset 0x117 and it produces the following debug output: name_ser| main(): 2 ... (0)/ file1 (56.1) test3 (5)/ c (8)/ b (7)/ a (6)/ test2 (4)/ test1 (3)/ server (2)/ linux (1)/ name_ser| main(): 3 ... Q1: Where did the name_server get these names? I wanted to test the name_server module so I ran "name_server_test" which always crashes with output: tftp | Loading (nd)/fiasco/name_server_test [549kB] loader | "(nd)/fiasco/name_server_test" is a valid binary image loader | Setting libpath to (nd)/fiasco/ exec | name_server_test: Loading exec | name_server_test: Has no dynamic info exec | name_server_test: Saved 20647 bytes of symbols exec | name_server_test: library "libloader.s.so" not found loader | name_server_test: Starting sigma0-style application exec | name_server_test: Packed 14374 bytes of symbols exec | name_server_test: Packed 69740 bytes of lines loader | name_server_test,#14: Entry at 00009c74 => 00980000 loader | name_server_test,#14: Started run | . n_s_test| main(): Results follow below (should be as predicted (123) == '123'). n_s_test| main(): Volume stuff: ******************* n_s_test| main(): Name server thread id (8.2) 'D.02' name_ser| l4vfs_basic_name_server_thread_for_volume_component(): Before: l4vfs_ name_ser: basic_name_server_thread_for_volume_component() name_ser| l4vfs_basic_name_server_thread_for_volume_component(): After: l4vfs_b name_ser: asic_name_server_thread_for_volume_component() n_s_test| main(): Volume server for NS volume (8.2) 'D.02' name_ser| l4vfs_basic_name_server_thread_for_volume_component(): Before: l4vfs_ name_ser: basic_name_server_thread_for_volume_component() name_ser| l4vfs_basic_name_server_thread_for_volume_component(): After: l4vfs_b name_ser: asic_name_server_thread_for_volume_component() n_s_test| main(): Volume server for other volume (0.0) '7FF.7F' name_ser| l4vfs_name_space_provider_register_volume_component(): register_volum name_ser: e: 5.0 name_ser| L4RM: [PF] read at 0x012ffd14, eip 012003f9, src D.02 name_ser| [D.0] l4rm/lib/src/pagefault.c:75:__unknown_pf(): name_ser| unhandled page fault I seem to get a lot of Double PF when loading my own or test programs. Any suggestions welcome! Thanks Leon

6 19

Returned mail: see transcript for details
by xaoveeyoyobb＠spammotel.com 25 Sep '04

25 Sep '04

The original message was received at Sat, 25 Sep 2004 06:04:01 -0400 from localhost [127.0.0.1] ----- The following addresses had permanent fatal errors ----- <xaoveeyoyobb(a)spammotel.com> (reason: 550 <xaoveeyoyobb(a)spammotel.com> Benutzer hat das Speichervolumen ueberschritten. / User has exhausted allowed storage space.) ----- Transcript of session follows ----- ... while talking to mx-ha01.web.de.: >>> DATA <<< 550 <xaoveeyoyobb(a)spammotel.com> Benutzer hat das Speichervolumen ueberschritten. / User has exhausted allowed storage space. 550 5.1.1 <xaoveeyoyobb(a)spammotel.com>... User unknown <<< 503 Valid RCPT TO <recipient> must precede DATA 451 4.0.0 Cannot open hash database /etc/mail/aliases.db: Invalid argument

1 0

Returned mail: see transcript for details
by xaoveeyoyobb＠spammotel.com 24 Sep '04

24 Sep '04

The original message was received at Fri, 24 Sep 2004 16:36:23 -0400 from localhost [127.0.0.1] ----- The following addresses had permanent fatal errors ----- <xaoveeyoyobb(a)spammotel.com> (reason: 550 <xaoveeyoyobb(a)spammotel.com> Benutzer hat das Speichervolumen ueberschritten. / User has exhausted allowed storage space.) ----- Transcript of session follows ----- ... while talking to mx-ha01.web.de.: >>> DATA <<< 550 <xaoveeyoyobb(a)spammotel.com> Benutzer hat das Speichervolumen ueberschritten. / User has exhausted allowed storage space. 550 5.1.1 <xaoveeyoyobb(a)spammotel.com>... User unknown <<< 503 Valid RCPT TO <recipient> must precede DATA 451 4.0.0 Cannot open hash database /etc/mail/aliases.db: Invalid argument

1 0

C++ and dietlibc
by Leon 23 Sep '04

23 Sep '04

Hi all, I am trying to port (test) a small C++ app, which uses sockets, to the L4 environment. Most of the FLIPS examples uses the mode=l4env_base which includes dietlibc paths. I've tried to compile the app against libstdc++ and dietlibc but they do not seem compatible. The C++ app successfully compiles when using l4env_freebsd, but as soon as I include the Linux24/include directory to get the socket headers then there is a lot of clashes (duplicate defines) with the Oskit. This is obviously the reason for dietlibc in FLIPS, but the problem then is that dietlibc is not C++ compatible. With "compatible" I mean that some of the functions (like putc) is defined and not declared in the stdio.h file of dietlibc, which is then actually undefined in the cstdio file when the functions are moved to the STD namespace. Q1: Is this true - any suggestions? Thanks Leon

2 2

Fw: Re: Question about the OS Group
by R. Tyler Ballance 20 Sep '04

20 Sep '04

Anyways, minor question regarding the L4-Linux (or Fiasco for that matter) First off, I have some decent Compaq Pentium 1 desktops back home I could easily commandeer for hacking on. (I'm in college, so home is a 5 hour trip away) Will it run fine? I know system builds will suck, but it should work fine right? Secondly, I understand fully how a microkernel works, i've seen the schematics, flow charts, etc, but I'm confused at how a driver, for example, would be coded. Is it something that one hacks into userland? (Is it portable across L4 Operating Systems? (L4Linux, Fiasco, etc) Or does L4 have some special way of handling drivers? I know for example, that the Mach "microkernel" that Apple has bastardized for Darwin handles networking and such in the userland (it's been a while since i've played around on Darwin however) -Tyler ----------------- R. Tyler Ballance NetBSD-FreeBSD

6 7

FLIPS and l4linux
by Leon 08 Sep '04

08 Sep '04

Hi all, I've got a running system and some small programs running on Fiasco with DoPE and l4linux-2.4. I want to investigate FLIPS to use as an extension to one of the programs (network capable). The difference between compiling L4 and all its packages now is that I must set the parameter LINUX24_DIR to point to a directory with vanilla kernel code (I used 2.4.27) instead of previously when the parameter pointed to the L4LINUX directory (l4linux-2.4). The last one is currently my working setup. Q1: Was I doing it wrong the whole time and was just lucky to get it working? Anyway the newly compiled FLIPS system starts to load as usual but as soon as l4linux is loaded is stops with a cursor blinking below the penguin with NO text (console) output. Here is the debug output: EXT3-fs: mounted filesystem with ordered data mode. VFS: Mounted root (ext3 filesystem) readonly. Warning: unable to open an initial console. simple_t| ERROR: task #13 doesn't match: client 11.03, owner 0.00 l4lx | l4ts_create_task(): failed (server=A.00, ret=-3, exc 0) The system is still operational and I can start my programs from the "doperun" terminal. Q2: Any suggestions on why my l4linux does not want to load anymore? Thanks Leon

4 5

New L4Env release
by Ronald Aigner 01 Sep '04

01 Sep '04

The TU Dresden OS team is happy to announce the next release of L4Env - the development environment for L4/Fiasco. Please visit the L4Env website http://os.inf.tu-dresden.de/l4env/ to download version 0.2 or browse the L4Env documentation. This new release contains our efforts to further improve L4Env, so you can look forward to enhanced functionality of, for instance, the loader or region manager, which make it easier to build applications for L4. Starting with this release we will also provide you with daily snapshots of the L4Env packages at http://os.inf.tu-dresden.de/l4env/. We would like to invite you and try L4Env 0.2 and help us improve it even more. Yours, the L4Env team.

1 0