- l4-hackers - example.com Site

Re: FLIPS issue
by Tiago Jorge 12 Apr '05

12 Apr '05

Christian Helmuth wrote: >Hello, > >it's me again. Could you please send me all source files for client and >server code under Linux as attachment with working broadcast code? So I >may get a chance to test it. > >Cheers > hello... sorry i'm bothering you again. first issue: i've been "investigating" the code and following the calls that all procedures do during their execution. And i cannot find where setsockopt in l4vfs is implemented :(. I don't see anyware the method l4vfs_net_io_setsockopt_component(...) implemented. Just its template. second issue: i've managed to get a serial cable to redirect output to another terminal, and i saw something i thought i saw earlier but didn't managed to figured it out. During flips boot it gives a strange error that i can't figure it out... ########################################################## flips-0 | l4dde_mm_init(): Using ... *flips-0 | 640 kB at 0x00280000 (vmem) *flips-0 | 1024 kB in 1 regions (kmem) *flips-0 | Initializing RT netlink socket *flips-0 | lo: I'm up now. *flips-0 | [9.2] ../../include/__macros.h:57:DICE_ERR(): *flips-0 | Error: call failed (ret -8 "item not found", exc 0) --- maybe this i *flips-0 : s okay for you! *flips-0 | Linux Tulip driver version 0.9.15-pre12 (Aug 9, 2002) *flips-0 | tulip0: MII transceiver #1 config 1000 status 786d advertising 05e1. *flips-0 | eth0: ADMtek Comet rev 17 at 0x1800, 00:30:05:62:64:4D, IRQ 11. *flips-0 | NET4: Linux TCP/IP 1.0 for NET4.0 *flips-0 | IP Protocols: ICMP, UDP, TCP *flips-0 | IP: routing cache hash table of 128 buckets, 3Kbytes *flips-0 | si_meminfo(): called (1) *flips-0 | TCP: Hash tables configured (established 128 bind 170) *flips-0 | CORBA_alloc(): size=1024, 1 pointers used *flips-0 | CORBA_alloc(): got block at 0x00180004 *flips-0 | CORBA_alloc(): size=1024, 2 pointers used *flips-0 | CORBA_alloc(): got block *flips-0 | flips_session_thread(): Flips(session_thread): new thread_id 9.8 *flips-0 | *flips-0 | CORBA_alloc(): size=1024, 3 pointers used *flips-0 | CORBA_alloc(): got block at 0x00180814 *flips-0 | CORBA_alloc(): size=1024, 4 pointers used *flips-0 | CORBA_alloc(): got block at 0x00180c1c *minifcfg| print_flags(): FLAGS for eth0: BROADCAST MULTICAST *minifcfg| print_flags(): FLAGS for eth0: UP BROADCAST RUNNING MULTICAST *minifcfg| print_flags(): FLAGS for eth0: BROADCAST MULTICAST *minifcfg| print_flags(): FLAGS for eth0: UP BROADCAST RUNNING MULTICAST *minifcfg| ifconfig(): eth0: inet 10.10.5.197 *minifcfg| *minifcfg| print_flags(): FLAGS for eth0: BROADCAST *minifcfg| ifconfig(): eth0: netmask 255.255.0.0 *minifcfg| main(): register at names *minifcfg| main(): eternal sleep *flips-0 | flips_session_thread(): Flips(session_thread): new thread_id 9.a *flips-0 | *flips-0 | CORBA_alloc(): size=1024, 5 pointers used *flips-0 | CORBA_alloc(): got block at 0x00181024 *flips-0 | CORBA_alloc(): size=1024, 6 pointers used *flips-0 | CORBA_alloc(): got block at 0x0018142c *testebro| main(): socket send error *testebro| main(): size_sent = -1 *testebro| *flips-0 | eth0: Setting full-duplex based on MII#1 link partner capability of 4 *flips-0 : 1e1. ################################################### *flips-0 | Error: call failed (ret -8 "item not found", exc 0) --- maybe this i *flips-0 : s okay for you! I don't know what this means, but my menu.lst looks like this... is it something missing? ########################################################### title Testes root (hd0,0) kernel /home/tiago/fiasco_builddir/rmgr -sigma0 modaddr 0x02000000 module /home/tiago/fiasco_builddir/main -nokdb -nowait -serial_esc -comspeed 115200 -comport 1 module /home/tiago/fiasco_builddir/sigma0 module /home/tiago/fiasco_builddir/names module /home/tiago/fiasco_builddir/dm_phys module /home/tiago/fiasco_builddir/l4io module /home/tiago/fiasco_builddir/name_server module /home/tiago/fiasco_builddir/flips-lxdrv module /home/tiago/fiasco_builddir/mini_ifconfig eth0 10.10.5.197 255.255.0.0 10.10.255.255 module /home/tiago/fiasco_builddir/testebroadcast ########################################################### is there any server missing? thanks for your help so far Tiago

2 2

Kernel Module Pointers for L4Linux on Fiasco
by Julian Grizzard 11 Apr '05

11 Apr '05

Anyone have any pointers on using kernel modules with L4Linux? Here's the setup: i386 L4 Fiasco V.2 L4Linux third party kernel modules We've been playing with various modules trying to get them to work with the L4Linux kernel. Some modules compile fine against the L4Linux source; others give warnings. Further, some load fine but do not work properly; others complain about missing symbols. It is conceivable that we have non-L4Linux related issues but wanted to check and see if there is anything special to know about building modules for L4Linux? Specifically, what are possible reasons that a kernel module would work on a stock i386 arch Linux kernel but not work on the same version L4 arch Linux kernel? Thanks! -Julian

2 2

Various things on log_net package
by Rene Wittmann 11 Apr '05

11 Apr '05

Hi, While trying to get the log server working for logging to network, I found some problems/bugs... And here is my report: L4DIR/pkg/log_net/server/src/tcpip.c: l.96 : in #define INITDRV(drv) the called function can give an error, but this error is just ignored (fct. init_ethernet always returns 0) l.264: (closing etherdev on error) if you select a wrong driver in init_ethernet, the system boot will fail with strange output over and over again (maybe just logging to normal console and go on booting is more desired??) L4DIR/pkg/log/server/src/logserver.c l.45: initializing buffer for net output if the buffer size is not given explicitly as command line switch the later buffer size is 0(!! not 4096 as expected). Therefore the logging to the network will not work. In order to prevent this you should put some lines right after parsing command line parameters in main(), e.g.: // if flushing to net && no buffer_size specified, use an initial value if (flush_to_net && !buffer_size){ buffer_size = 4096; } (or you should document it somewhere) BTW: to use the correct driver in init_ethernet for vmware, you should use "vlance" as network setting in vmware and "INITDRV(lance)" in init_ethernet, the drivers for 3com cards are not called 3c??? as I expected but can be used with "INITDRV(etherlink<??>)" with <??>="","2","3","16","plus" or "INITDRV(vortex)". In my opinion this should be documented somewhere. (in case it is already, and I was just too stupid to find it, ignore this remark...) Problem when using the logging to network within a program that uses the parse_cmd library I can't explain why this is the case, but when you use the log server compiled for network output and afterwards start an application that uses "parse_cmdline" the logging will not go to network, but to local console (even if you specify "-nolocal" on the log server). If you're interested in this bug, I can send you my sample code which is just a modification of rt_sched example in the hello-package Regards, Rene

2 2

L4Linux Kernel Thread IDs
by Leon Brits 08 Apr '05

08 Apr '05

Hi all, I have a L4Linux kernel module which creates a kernel thread, but the new thread and the module has the same L4 thread ID [l4_myself()] while their PIDs are different as usual. Is this correct? Thanks Leon

2 2

con deprecated?
by Stefan Kalkowski 08 Apr '05

08 Apr '05

hello list, i want to do a new build of the l4-tree after building a new gcc. i got the problem, that packages like proxygon, that uses l4con-headers don't pass through, because the con-package is marked as deprecated since 2 days and it's Makefile is modified, so that it won't build. I think it's only a naming convention problem and there should be a l4con replacement? Or is there any other reason, why con is obsolet? greetings stefan -- Stefan Kalkowski GNU PGP-Key: https://www.inf.tu-dresden.de/~sk852421

2 1

Re: L4Linux Kernel Thread IDs
by Frank Mehnert 06 Apr '05

06 Apr '05

Leon, On Wednesday 06 April 2005 09:57, Leon Brits wrote: > I have a L4Linux kernel module which creates a kernel thread, but the > new thread and the module has the same L4 thread ID [l4_myself()] while > their PIDs are different as usual. > > Is this correct? Which L4Linux version do you talk about? 2.4? (Please post those messages to l4-hackers since I'm too busy to answer all your questions.) Frank -- ## Dept. of Computer Science, Dresden University of Technology, Germany ## ## http://os.inf.tu-dresden.de/~fm3 ##

1 0

writing code for IPC servers
by Tiago Jorge 05 Apr '05

05 Apr '05

hello again. is it possible to code the server side of DICE generated stubs using C++ by only modifing the filename and deleting the #ifdef __cplusplus? or is DICE ready to generate C++ code? thanks Tiago

4 6

Some questions with regard to emul_l4rm.c
by Leon Brits 01 Apr '05

01 Apr '05

Hi all, I am trying to understand the workings of the emul_l4rm.c. Please allow me to explain how I understand the working of this file so that you can correct me. When calling l4rm_attach(), the function l4rm_do_attach() gets called. This function calls reserve_virtual_area() which uses mmap(). Mmap() searches for an available area (with a certain size) in the calling process's virtual memory. After finding such an area it remaps memory, allocated using kmalloc() by /dev/zero, to the newly found virtual memory space. Kmalloc() allocates memory from the l4linux dataspace and that piece of memory is now mapped into user space. Next, the call to l4dm_map_pages(), maps my specified dataspace's pages into this mapped virtual address. Q1: Why go thru the first memory mapping of the kmalloc-ed memory into user space if immediately after it is remapped again by l4dm? Seems like a waste of the l4linux dataspace. Q2: Can the free virtual memory area be determined by another method and then call l4dm_map_pages()? Q2: When l4_fpage_unmap() is called, is the previous mapping to the pages of the kmalloc-ed memory restored? Thanks for your time. Leon

2 1

L4Linux hangs on self-compiled fiasco
by Björn Fischer 31 Mar '05

31 Mar '05

Hello, I am trying to build the complete DROPS including fiasco, DoPE and L4Linux. As a starting point I had the demo CD from the OS research day 2004. At first I used the binaries from the cd (including the config files) and ran them on top of vmware. Modules were loaded successfully via tftp and worked (well, when accessing the cd, L4Linux had the prob that there was no cd anymore, but at least the kernel worked well). After that I compiled the cvs tree and replaced the binaries by the own-compiled ones. It seems that fiasco and DOpE work fine since it renders nice windows and so on. But when it comes to starting L4Linux it crashes - well, in fact it just "hangs" and does nothing anymore. The weird thing is that the bootmessages, the ones printed in the "\ m /" window do not end at a reproducable point. Sometimes it comes up to "======> L4Linux 2.6 starting <========", sometimes only up to "loader : vmlinuz26,#d: Continue at l4env_init (00014264, libloader.s.so)", sometimes it hangs in the middle of a message. These are my config-files: menu.lst: --------- title L4Linux 2.6 from this CD [DOpE] kernel (nd)/drops/rmgr -sigma0 task modname "bmodfs" module module module module module module module task modname "l4dope" boot_priority 0xA8 modaddr 0x02000000 module (nd)/drops/fiasco -nokdb -nowait -noserial module (nd)/drops/sigma0 module (nd)/drops/names module (nd)/drops/dmon module (nd)/drops/dm_phys module (nd)/drops/simple_ts -t 380 module (nd)/drops/rtc module (nd)/drops/l4io --noirq module (nd)/drops/bmodfs module (nd)/drops/libloader.s.so module (nd)/drops/run.bmodfs.cfg module (nd)/drops/run module (nd)/drops/vmlinuz26 module (nd)/drops/vscrtest module (nd)/drops/cfg/l4linux26.cd.k_us.cfg module (nd)/drops/drops-rd.rd module (nd)/drops/l4exec module (nd)/drops/l4dope --l4io module (nd)/drops/proxygon module (nd)/drops/backdrop ------------------------- drops/cfg/l4linux26.cd.k_us.cfg: -------------------------------- sleep 1 task "vmlinuz26" "mem=64M noreplacement root=1:0 load_ramdisk=1 ramdisk_size=4096 l4env_rd=drops-rd.rd lang=us" allow_cli priority 0xA0 -------------------------------- Does anybody know what kernel options I could try to fix this? Oder does anybody have any idea? Am I doing something wrong? I used gcc 2.95.3 to compile. Many thanks and greetings Björn Fischer

3 3

Compiling the LOG server for net output
by Rene Wittmann 31 Mar '05

31 Mar '05

Hi, as I'm wanted to use the LOG server to log over the network, I changed the variable CONFIG_USE_TCPIP in the Makefile to 1. (in L4DIR/pkg/log/server/src) I tried to follow the instructions in the documentation of the LOG server (http://os.inf.tu-dresden.de/l4env/doc/log/p_faq.html#p_faq) So I tried to enable my driver in L4DIR/pkg/log/server/src/tcpip.c, but the file was not there! I'm pretty sure that I didn't remove it, so I guess it was not checked out... Is there a suitable file in the repository? If so, where? I didn't really care about that, so I tried to compile it, but of course it failed. (The error message: ==> Linking log LD_PRELOAD=/home/sirewitt/src/drops/l4/tool/gendep/libgendep.so GENDEP_TARGET=log GENDEP_BINARY=ld /usr/local/bin/gcc-2.95 -o log ../../../../../lib/x86_586/crt0.o logserver.o stuff.o flusher.o -Ttext=0x400000 -L../../../../../lib/x86_586/l4v2 -L/home/sirewitt/src/drops/drops/lib/x86_586/l4v2 -L../../../../../lib/x86_586 -L/home/sirewitt/src/drops/drops/lib/x86_586 -L../../../../../lib -L/home/sirewitt/src/drops/drops/lib -L../../../../../../oskit/lib -L/home/sirewitt/src/drops/drops/lib/oskit06 -T../../../../../lib/x86_586/main_stat.ld -static -lmain -lnames -ll4util -lparsecmdline -ll4serial -llog /usr/local/lib/gcc-lib/i686-pc-linux-gnu/2.95.2.1/libgcc.a -static -lmain -llogserver -lnames -ll4util -lrmgr -ll4sys -nostdlib -Wl,--nostdlib -lmc -loskit_support -ll4util -lrmgr -lkern -llmm -lmc /usr/local/lib/gcc-lib/i686-pc-linux-gnu/2.95.2.1/libgcc.a -lmc -Ttext=0x400000 -Wl,-gc-sections logserver.o(.text+0x3c9): In function `get_message': /drops/l4/pkg/log/server/src/logserver.c:204: undefined reference to `channel_get_next_fpage' logserver.o(.text+0x7b1): In function `main': /drops/l4/pkg/log/server/src/logserver.c:323: undefined reference to `channel_write' logserver.o(.text+0x7c1):/drops/l4/pkg/log/server/src/logserver.c:326: undefined reference to `channel_flush' logserver.o(.text+0x7d1):/drops/l4/pkg/log/server/src/logserver.c:329: undefined reference to `channel_close' logserver.o(.text+0x7e1):/drops/l4/pkg/log/server/src/logserver.c:336: undefined reference to `channel_open' flusher.o(.text+0xcb): In function `thread_loop': /drops/l4/pkg/log/server/src/flusher.c:105: undefined reference to `net_init' flusher.o(.text+0x16e):/drops/l4/pkg/log/server/src/flusher.c:123: undefined reference to `net_wait_for_client' flusher.o(.text+0x1bb):/drops/l4/pkg/log/server/src/flusher.c:146: undefined reference to `net_receive_check' flusher.o(.text+0x22f):/drops/l4/pkg/log/server/src/flusher.c:166: undefined reference to `client_socket' flusher.o(.text+0x38b): In function `do_flush_buffer': /drops/l4/pkg/log/server/src/flusher.c:242: undefined reference to `net_flush_buffer' collect2: ld returned 1 exit status make[3]: *** [log] Error 1 make[3]: Leaving directory `/home/sirewitt/src/drops/l4/pkg/log/server/src/OBJ-x86_586-l4v2' make[2]: *** [OBJ-x86_586-l4v2] Error 2 make[2]: Leaving directory `/home/sirewitt/src/drops/l4/pkg/log/server/src' make[1]: *** [src] Error 2 make[1]: Leaving directory `/home/sirewitt/src/drops/l4/pkg/log/server' make: *** [server] Error 2 ) So what do I have to do to get the net output working for the LOG server? Thanks, Rene

2 1