- l4-hackers - example.com Site

DoS problem still existing?
by Robert Kaiser 19 May '06

19 May '06

Hello L4 Hackers, a long time ago (*), Jochen Liedtke described a potential denial-of-service attack against the L4 kernel by -IIRC- requesting a huge number of mappings, thereby exhausting kernel memory. Does this issue still exist in the current L4 implementations or has it been solved (How?). Kind Regards Rob (*) see http://i30www.ira.uka.de/research/documents/l4ka/preventing-denial-of-servi… -- Robert Kaiser <rkaiser(a)sysgo.com> SYSGO AG Tel.: +49-6136-9948-0 Am Pfaffenstein 14 Fax: +49-6136-9948-10 55270 Klein-Winternheim http://www.sysgo.com

2 1

con-ux bug?
by Wei Shen 12 May '06

12 May '06

Hi, I run con-ux with: ./fiasco -G 800x600@16 -m 128 -l names -l dm_phys -l con-ux The output is: >*con | Using fast memcpy. >*con | Character size is 8x14, font has 256 characters >*con | VESA reports 800x600@16 1600bpl (0000) [4096kB] >*con | Color mapping: red=11:5 green=5:6 blue=0:5 res=0:0 > ... .... >*con | Mapping I/O UX video mem c0000000 => 00400000+000000 [4096kB] >*con | Using MMXEXT for colorspace transform >KERNEL: 7.2 (tcb=201c1000) killed: >Unhandled trap >EAX 00000025 EBX 0000122b ECX 0000000a EDX 00000000 >ESI 00000320 EDI 00b0d73c EBP 012ffda8 ESP 012ffda4 >EIP 00b0dc7f EFLAGS 00200206 >CS 0002 SS 007b DS 007b ES 007b FS 0000 GS 0000 >trap 13 (General Protection), error 00000402, from user mode >(internal event regarding IDT gate descriptor no. 0x80) The backtrace: >main // l4con/server/src/main.c >do_switch >vc_clear // vc.c >vc_fill >pslim_fill // pslim.c >sw_fill In "sw_fill", "vc->do_drty" equals 11589436, which seems invalid. Maybe, it should be NULL? Maybe, we should add "new->do_drty = fg_do_drty;" to line 402 in main.c? Regards, Wei

3 8

L4Linux freezes
by Vilmos Nebehaj 12 May '06

12 May '06

Hi, I've compiled and installed Fiasco and L4Linux-2.6 and it works, except that when I start a simple test program that creates lots of pthreads, the whole L4Linux task freezes (simply pthread_create() a thousand threads after each other is enough to reproduce the problem). I can ping the machine from another host, but can't login through ssh, the console becomes totally unresponsive, etc. At the same time, the following message appears in the log: simplets| ERROR: task #23 doesn't match: client 10.03, owner 0.0 l4lx | l4ts_create_task(): failed (server=7.00, ret=-3, exc=0) Other L4 tasks continue running uncluttered. Do you have an idea why this happens? Is this a bug in Fiasco, or is it a misconfiguration problem? -vili

2 2

Re: TCP/IP between L4Linux and L4 task
by Bjoern Doebel 11 May '06

11 May '06

2 1

Re: TCP/IP between L4Linux and L4 task
by Stefan Scheler 11 May '06

11 May '06

Hi, > Sure. Both, L4Linux and your app need to be compiled without native > hardware support. Instead, use a software network switch like the one > provided by l4/pkg/ore. L4Linux needs to compiled with the ORe driver > and your app needs to send/receive packets through ORe. Well, sounds good. So I'm currently trying to integrate the latest e1000 driver (v7.0.33) into ORe as my e1000 adapter is not supported by the current one but encountered the following problem: The updated driver wants to alloc some pages using alloc_page(). This results in a undefined reference to "_alloc_pages" when linking ore: ==> Linking ore LD_PRELOAD=/home/sts/L4_full/l4/tool/gendep/libgendep.so GENDEP_TARGET=ore GENDEP_BINARY=ld gcc -o ore ../../../../../lib/x86_586/crt0.o main.o openclose.o clients.o recv.o send.o netdevice.o irq_handling.o rxtx_string.o debug.o util.o events.o checksum.o xmit_lock.o worker.o local_deliver.o dummy_dsi.o -Wl,--whole-archive ../../../server/lib/net/OBJ-x86_586-l4v2/liblinuxnetdrv.a -Wl,--no-whole-archive -Ttext=0x01800000 -L../../../server/lib/linuxemul/OBJ-x86_586-l4v2 -L../../../../../lib/x86_586/l4v2 -L/home/drops/lib/x86_586/l4v2 -L../../../../../lib/x86_586 -L/home/drops/lib/x86_586 -L../../../../../lib -L/home/drops/lib -T../../../../../lib/x86_586/main_stat.ld -llinuxemul -levents -llinuxnetdrv -ldde_linux -lomega0 -lio -ll4rm -lore_server-worker -lore_server -lore_notify -llist_alloc -lparsecmdline /usr/lib/gcc/i686-pc-linux-gnu/3.4.6/libgcc.a -static -lgeneric_ts -ll4env -ll4rm -ldm_generic -ldm_mem -lthread -lsemaphore -llogserver_capsule -lnames -ll4util_root -ll4util -lsigma0 -lroot.o -ll4env -ll4env_err -ll4rm -ldm_generic -ldm_mem -lthread -lslab -ll4sys -nostdlib -Wl,--nostdlib -luc_c -luclibc_support -u mmap_anon -lc_be_simple_mem -lc_be_mmap -lc_be_mmap_util -lc_be_l4env_start_stop -lc_be_minimal_log_io -luc_be_simple_sleep -ll4env -llogserver_capsule -ll4rm -ldm_generic -ldm_mem -lthread -lgeneric_ts -luc_c /usr/lib/gcc/i686-pc-linux-gnu/3.4.6/libgcc.a -luc_c -ll4sys -Ttext=0x01800000 -Wl,-gc-sections -Wl,--warn-common ../../../server/lib/net/OBJ-x86_586-l4v2/liblinuxnetdrv.a(e1000_main.o): In function `e1000_alloc_rx_buffers_ps': /home/sts/L4_full/l4/pkg/ore/server/lib/net/OBJ-x86_586-l4v2/../../../../server/include/linux_headers/linux/mm.h:454: undefined reference to `_alloc_pages' collect2: ld returned 1 exit status I suppose I have to link something else? Any hints appreciated. BTW is ORe an acronym for something? What does it stand for? Stefan

3 4

[Fwd: Re: TCP/IP between L4Linux and L4 task]
by Bjoern Doebel 11 May '06

11 May '06

3 4

Re: Fiasco hello world problem..
by raghavendra pai 11 May '06

11 May '06

---------------------------------------------------------------------- Message: 1 Date: Mon, 8 May 2006 13:48:25 +0200 From: Adam Lackorzynski <adam(a)os.inf.tu-dresden.de> Subject: Re: Fiasco hello world problem.. To: l4-hackers(a)os.inf.tu-dresden.de Message-ID: <20060508114825.GI1784(a)os.inf.tu-dresden.de> Content-Type: text/plain; charset=iso-8859-1 On Mon May 08, 2006 at 15:04:27 +0530, raghavendra pai wrote: > I was trying to run hello world program in Fiasco . > I downloaded the hello world from the CVS and did the steps as given in > the site. > > http://os.inf.tu-dresden.de/fiasco/use.html > > but I was trying to boot using my pc the hello world program . > the grub entry in grub.conf file was > > title hello world > module /fiasco-boot/main -nokdb > module /fiasco-boot/sigma0 > module /fiasco-boot/roottask > module /fiasco-boot/hello > > When i booted it the following was displayed > > Booting "Hello World" > module /fiasco/main -nokdb > > Error No:20 Multiboot kernel must be loaded before modules.. > > What is the cause of error.. >You forgot 'kernel bootstrap', i.e. the grub entry needs to look like: >title hello world > kernel /fiasco-boot/bootstrap > module /fiasco-boot/main -nokdb > module /fiasco-boot/sigma0 > module /fiasco-boot/roottask > module /fiasco-boot/hello I tried to locate the bootstrap file and i got this much entry /root/l4/l4/kernel/fiasco/build/.bootstrap.cc.d /root/l4/l4/kernel/fiasco/build/bootstrap.o /root/l4/l4/kernel/fiasco/src/boot/amd64/bootstrap.cc /root/l4/l4/kernel/fiasco/src/boot/amd64/bootstrap.ld /root/l4/l4/kernel/fiasco/src/boot/ia32/bootstrap.cc /root/l4/l4/kernel/fiasco/src/boot/ia32/bootstrap.ld /root/l4/l4/kernel/fiasco/src/kern/arm/bootstrap.cpp /root/l4/l4/kernel/fiasco/src/test/boottask/bootstrap-boottask.cc /root/l4/l4/tool/dice/bootstrap /root/l4/my/build/dir/.bootstrap.cc.d /root/l4/my/build/dir/bootstrap.o which of the one is to be copied into /fiasco-boot folder . I tried using /root/l4/l4/tool/dice/bootstrap but it tells no file is there Wht is the problem ? Am i entered the wrong Regards Pai

2 4

RE: L4Env and L4 version X.2.
by Volkmar Uhlig 11 May '06

11 May '06

> -----Original Message----- > From: Adam Lackorzynski > Sent: Wednesday, May 10, 2006 7:14 PM > To: l4-hackers(a)os.inf.tu-dresden.de > Subject: Re: L4Env and L4 version X.2. > > On Thu May 04, 2006 at 18:26:56 -0500, sethb wrote: > > Is there ongoing work to port L4Env to L4 version X.2? > > None that I know of. > > > How much work would it be? > > Quite a bit, I guess. If you have an IDL compiler you can > then go through every module and port it over, probably > redesigning some parts of it. The X.2 interface was designed such that a mapping of V2 -> X.2 is possible via wrapper functions. The fixed thread/task model can be emulated and ex-regs etc can be mapped as well. Main problem is the IDL compiler; not sure if Dice supports X.2... Ron? - Volkmar

3 2

L4Env and L4 version X.2.
by sethb 11 May '06

11 May '06

Hello Everyone, Is there ongoing work to port L4Env to L4 version X.2? How much work would it be? Thanks, Seth

3 2

The question about flexpage
by 강후영 11 May '06

11 May '06

HI! I have some questions about flexpage. What i want to do is to share memory between L4linux "A" and L4linux "B". So i tried to use flexpage method but this method is not available. Another tests sharing memory between L4linux "A" and L4 server are good work with flexpage. Is it possible to share memory between between L4linux "A" and L4linux "B"? best regards.

2 1

TCP/IP between L4Linux and L4 task
by Stefan Scheler 10 May '06

10 May '06

Hello, is it possible to connect from an L4Linux process to a FLIPS-based L4 task using TCP/IP? And if so, how? Thanks, Stefan

1 0

Fiasco hello world problem..
by raghavendra pai 08 May '06

08 May '06

Hi , I was trying to run hello world program in Fiasco . I downloaded the hello world from the CVS and did the steps as given in the site. http://os.inf.tu-dresden.de/fiasco/use.html but I was trying to boot using my pc the hello world program . the grub entry in grub.conf file was title hello world module /fiasco-boot/main -nokdb module /fiasco-boot/sigma0 module /fiasco-boot/roottask module /fiasco-boot/hello When i booted it the following was displayed Booting "Hello World" module /fiasco/main -nokdb Error No:20 Multiboot kernel must be loaded before modules.. What is the cause of error.. Regards Pai..

2 1

[EMSCB] L4Linux
by Oskar Senft 05 May '06

05 May '06

Hi! Nachdem alles lief habe ich nochmal ein Update der DROPS-Sourcen gemacht und bekomme jetzt folgenden Fehler beim Starten von L4Linux: l4lx | l4x_hybrid_return: Invalid hybrid return for 13.00 (0x00823630, d, 18 l4lx : e, 1, 804ba5e, eacff102)! l4lx | l4x_hybrid_return: Currently running: 00.00 Siehe auch angehängtes Logfile. Was ist denn diesmal falsch? Viele Grüße Oskar. PS: Sorry, irgendwie ging die Mail bei mir nicht richtig raus; falls doch: sorry für das Doppel-Posting! l4lx | ======> L4Linux 2.6 starting... <======== l4lx | Linux version 2.6.16-l4 (osk@gamma) (gcc version 3.3.6) #1 Fri May 5 l4lx : 09:58:11 CEST 2006 l4lx | Binary name: vmlinuz26-l4linux-2.6.15-svpnserver l4lx | Kernel command line (8 args): mem=20M noreplacement root=1:0 load_ram l4lx : disk=1 ramdisk_size=15000 l4env_rd=svpn-ramdisk.gz panicblink=0 lang= l4lx : de l4lx | Image: 00400000 - 0072e000 [3256 KiB]. l4lx | Areas: Text: 00400000 - 00645000 [2324kB] (a bit longer) l4lx | Data: 00645000 - 00690520 [301kB] l4lx | Initdata: 00694000 - 006c2000 [184kB] l4lx | BSS: 006c2000 - 0072c0a0 [424kB] l4lx | l4lx_thread_create: Created thread 0f.03 (server) l4lx | l4env_linux_startup thread 3. l4lx | main thread will be 0f.03 l4lx | l4env_register_pointer_section: addr = 00692000 size = 638976 l4lx | sec-w-init: virt: 0x00692000 to 0x0072dfff [624 KiB] l4lx | Noncontiguous region for sec-w-init l4lx | sec-w-init: Number of physical regions: 2, 638976 Bytes l4lx | sec-w-init: 1: Phys: 0x001d4000 to 0x001df000, Size: 45056 l4lx | sec-w-init: 2: Phys: 0x0146f000 to 0x01500000, Size: 593920 l4lx | main thread: received startup message. l4lx | Main thread running, waiting... l4lx | setup_l4env_memory: Forcing superpages for main memory l4lx | Main memory size: 20MB l4lx | Main memory: virt: 0x00800000 to 0x01bfffff [20480 KiB] l4lx | Main memory: Number of physical regions: 1, 20971520 Bytes l4lx | Main memory: 1: Phys: 0x03000000 to 0x04400000, Size: 20971520 l4lx | Filling lower ptabs... l4lx | Done (1069 entries). l4lx | l4env_rd_path: svpn-ramdisk.gz l4lx | Loading: svpn-ramdisk.gz l4lx | INITRD: Size of RAMdisk is 5015KiB l4lx | RAMdisk from 01c00000 to 020e5c00 [5015KiB] l4lx | l4lx_thread_create: Created thread 0f.04 (timer.i0) l4lx | l4lx_thread_create: Created thread 0f.05 (Idler) l4lx | l4x_hybrid_return: Invalid hybrid return for 13.00 (0x00823630, d, 18 l4lx : e, 1, 804ba5e, eacff102)! l4lx | l4x_hybrid_return: Currently running: 00.00

5 6

[EMSCB] L4Linux
by Oskar Senft 05 May '06

05 May '06

Hi! Nachdem alles lief habe ich nochmal ein Update der DROPS-Sourcen gemacht und bekomme jetzt folgenden Fehler beim Starten von L4Linux: l4lx | l4x_hybrid_return: Invalid hybrid return for 13.00 (0x00823630, d, 18 l4lx : e, 1, 804ba5e, eacff102)! l4lx | l4x_hybrid_return: Currently running: 00.00 Siehe auch angehängtes Logfile. Was ist denn diesmal falsch? Viele Grüße Oskar. -- Sirrix AG security technologies - http://www.sirrix.com Oskar Senft eMail: o.senft(a)sirrix.com Tel +49 (681) 936 251 - 119 Fax +49 (681) 936 251 - 500 get public key from keyserver Fingerprint 6FC6 3E44 233F 7F4B 33C7 A8D0 2EDA F3BA B5A3 0BA4 This message may contain confidential and/or privileged information. If you are not the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. l4lx | ======> L4Linux 2.6 starting... <======== l4lx | Linux version 2.6.16-l4 (osk@gamma) (gcc version 3.3.6) #1 Fri May 5 l4lx : 09:58:11 CEST 2006 l4lx | Binary name: vmlinuz26-l4linux-2.6.15-svpnserver l4lx | Kernel command line (8 args): mem=20M noreplacement root=1:0 load_ram l4lx : disk=1 ramdisk_size=15000 l4env_rd=svpn-ramdisk.gz panicblink=0 lang= l4lx : de l4lx | Image: 00400000 - 0072e000 [3256 KiB]. l4lx | Areas: Text: 00400000 - 00645000 [2324kB] (a bit longer) l4lx | Data: 00645000 - 00690520 [301kB] l4lx | Initdata: 00694000 - 006c2000 [184kB] l4lx | BSS: 006c2000 - 0072c0a0 [424kB] l4lx | l4lx_thread_create: Created thread 0f.03 (server) l4lx | l4env_linux_startup thread 3. l4lx | main thread will be 0f.03 l4lx | l4env_register_pointer_section: addr = 00692000 size = 638976 l4lx | sec-w-init: virt: 0x00692000 to 0x0072dfff [624 KiB] l4lx | Noncontiguous region for sec-w-init l4lx | sec-w-init: Number of physical regions: 2, 638976 Bytes l4lx | sec-w-init: 1: Phys: 0x001d4000 to 0x001df000, Size: 45056 l4lx | sec-w-init: 2: Phys: 0x0146f000 to 0x01500000, Size: 593920 l4lx | main thread: received startup message. l4lx | Main thread running, waiting... l4lx | setup_l4env_memory: Forcing superpages for main memory l4lx | Main memory size: 20MB l4lx | Main memory: virt: 0x00800000 to 0x01bfffff [20480 KiB] l4lx | Main memory: Number of physical regions: 1, 20971520 Bytes l4lx | Main memory: 1: Phys: 0x03000000 to 0x04400000, Size: 20971520 l4lx | Filling lower ptabs... l4lx | Done (1069 entries). l4lx | l4env_rd_path: svpn-ramdisk.gz l4lx | Loading: svpn-ramdisk.gz l4lx | INITRD: Size of RAMdisk is 5015KiB l4lx | RAMdisk from 01c00000 to 020e5c00 [5015KiB] l4lx | l4lx_thread_create: Created thread 0f.04 (timer.i0) l4lx | l4lx_thread_create: Created thread 0f.05 (Idler) l4lx | l4x_hybrid_return: Invalid hybrid return for 13.00 (0x00823630, d, 18 l4lx : e, 1, 804ba5e, eacff102)! l4lx | l4x_hybrid_return: Currently running: 00.00

1 0

error build docs?
by Colin Burgess 05 May '06

05 May '06

What verison of latex do I need to build the docs? My suse 10 system seems to have pdfeTeX installed as latex, and this spits out a lot of errors... Thanks, Colin

2 2

[EMSCB] L4Linux + IPC
by Oskar Senft 03 May '06

03 May '06

Hi all! We are using different Deamons running in L4Linux providing IPC services to other L4 task. With the most current DROPS CVS checkout, the L4Linux user processes freeze completely, as soon as an IPC call is executed which causes the process to sleep (eg. DICE IPC server loop). The L4Linux kernel seems to run, as it answers Ethernet/IP "ping" requests. A connection to a running SSH daemon no longer is possible and active SSH connection freeze, too. => Only L4Linux userland processes are frozen ?!? This did not happen with other version of L4Linux (eg. from January) and does not depend on the program we run in L4Linux - it happens with different applications that provide IPC services. Is this a problem with our settings for L4Env / L4Linux? Or could it be a bug in L4Linux? Do you have a test program that should run? Regards Oskar.

5 8

MIKES 2006 - Call for Papers
by Ihor Kuz 03 May '06

03 May '06

Call for Papers: MIKES 2006 First International Workshop on MicroKernels for Embedded Systems To be held on August 15 2006, in Sydney, Australia, co-located with the 12th IEEE RTCSA (http://www.rtcsa.org) Microkernels and microkernel-based operating systems are experiencing an increased popularity in the field of embedded systems. Despite a period of low popularity a decade ago, there is a new realisation, in academia and industry, of the usefulness of microkernels. This is especially true in the embedded systems domain where the reduced size and increased dependability of microkernel-based systems have made them particularly attractive. The MIKES workshop will bring together researchers and developers from industry and academia to discuss: * emerging novel concepts, * analysis approaches, and * case studies in the area of microkernels. MIKES 2006 seeks papers that describe original research in the area of microkernels. This includes but is not limited to: * case studies * real-time behaviour * temporal analysis * formal/functional verification * portability * API proposals * security and safety Papers describing new ideas, promising approaches, experiences with practical systems are considered particularly appropriate. Papers will be 6 pages in IEEE conference paper format. Papers should be submitted electronically at http://www.easychair.org/MIKES2006/ in PDF or postscript format. The material must be unpublished and not under submission elsewhere. Proceedings will be published as a technical report by National ICT Australia (NICTA). IMPORTANT DATES Please note the submission deadline has been pushed back to disentangle it from concurrent deadlines. Submissions deadlines: May 20th Notification of acceptance: June 12th Camera ready papers: July 7th ORGANISERS Gernot Heiser, NICTA/UNSW, Australia Stefan M. Petters, NICTA/UNSW, Australia Ihor Kuz, NICTA/UNSW, Australia PROGRAM COMMITTEE Kevin Elphinstone, UNSW, Australia (chair) Andrew Tannenbaum, Vrije Universiteit, Amsterdam, Netherlands Gerwin Klein, NICTA/UNSW, Australia Hermann Haertig, TU Dresden, Germany Jonathan S. Shapiro, The Johns Hopkins University, USA Neil Audsley, University of York, UK Sebastian Schoenberg, Intel, Hillsboro, USA Volkmar Uhlig, IBM Watson, USA RELATED EVENTS Also of interest to some attendees of MIKES might be the International Workshop on System Verification (http://www.cse.unsw.edu.au/~formalmethods/events/svws-06/) being held in Sydney (just before MIKES and RTCSA) on August 7 and 8, 2006.

1 0

Getting Started...
by Colin Burgess 25 Apr '06

25 Apr '06

Hi All, I tried following the build instructions for building L4-Linux 2.6 but to no avail. I am running Suse Linux 2.6.13-15.8-default I checked out l4linux-2.6-env, and typed make config, and selected a DROPS_DIR of /home/cburgess/drops I then typed make and it eventually failed to find l4/sys/types.h trying to build log-server Are there other modules I need to checkout? Do I also need the l4linux-2.6 module, and should it be in the same toplevel dir as the l4linux-2.6-env module? Thanks, Colin

3 4

About l4.sec schedule
by Wei Shen 24 Apr '06

24 Apr '06

Hi, Is there any schedule for the first release of l4.sec kernel? Maybe, in several months? Or before the end of this year? How do you consider the compatibility of l4.sec and l4env? Thanks, Wei

1 0

fiasco-ux does not build
by Stefan Scheler 21 Apr '06

21 Apr '06

Hello, I get the following error building fiasco-ux: ==> Linking kernel.image /usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.6/../../../../i686-pc-linux-gnu/bin/ld: errno: TLS definition in /usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.6/../../../libc.a(errno.o) section .tbss mismatches non-TLS reference in boot_info-ux.o /usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.6/../../../libc.a: could not read symbols: Bad value collect2: ld returned 1 exit status make[2]: *** [kernel.image] Error 1 make[2]: Leaving directory `/home/sts/hellomodule/l4/kernel/fiasco/build' make[1]: *** [all] Error 2 make[1]: Leaving directory `/home/sts/hellomodule/l4/kernel/fiasco/build' make: *** [all] Error 2 Real: 91.12s User: 74.26s System: 11.10s Percent: 93% Cmd: make So is this an gcc, glibc or binutils issue in the first place? And which versions are supposed to be able to build fiasco-ux? I'm using binutils 2.16.92 and glibc 2.4 and various gcc versions btw. Thanks in advance Stefan

2 5

L4Linux booting failed
by chenggh04＠st.lzu.edu.cn 18 Apr '06

18 Apr '06

Hi: I compile the L4Linux with plain mode! But when I boot L4Linux it is wrong. The information is as following: RNGR: Stage1 loading (hd0,1)/boot/fiasco_L4/sigma0 loading kernel (hd0,1)/boot/fiasco_L4/main -nowait -nokdb -serial_esc -compseed The module 00:00300000-00390408: (hd0,1)/boot/fiasco_L4/main -nowait -nokdb -serial overlaps with module loaded by bootloader at 00100000-001caec0: RMGR 00:0020f000-002661ac: (hd0,1)/boot/fiasco_L4/main -nowait 01:00267000-00275925: (hd0,1)/boot/fiasc_L4/sigma0 ==>02:00276000-004d8950: (hd0,1)/boot/fiasco_L4/vmlinuz.V2 no-scroll no 03:00409000-014d9000:(hd0,1)/boot/fiasco_L4/drops-rd.rd The reason for this error message is that the modules as loaded by GRUB Interface with their final position as defined by the linker. Most often you can solve this problem by inserting the command "modaddr 0x02000000" immediately as next line after the "kernel=" line of you menu.lst. Note that you have to use a special version of GRUB which is avaiable from http://os.inf.tu-dresden.de/drops/download.html The grub about L4Linux is : title fiasco_L4Linux root (hd0,1) kernel (hd0,1)/boot/fiasco_L4/rmgr -sigma0 task modname "vmlinuz" module modaddr 0x02000000 module (hd0,1)/boot/fiasco_L4/main -nowait -nokdb -serial_esc -comspee d 115200 -comport 1 module (hd0,1)/boot/fiasco_L4/sigma0 module (hd0,1)/boot/fiasco_L4/vmlinuz.V2 no-scroll no-hlt l4irqack=lin ux load_ramdisk=1 ramdisk_size=16384 root=/dev/ram module (hd0,1)/boot/fiasco_L4/drops-rd.rd I don't know what is wrong! Please help me! Thank you!

3 2

Strange L4Linux behavior
by Jugwan Eom 14 Apr '06

14 Apr '06

Hi, L4Linux fails to run correctly with the SCSI and network device under heavy load. My system has a Adaptec AIC7902 SCSI adapter (one SCSI hard disk is connected) and a E1000 network adapter. When the two devices is used at the same time under heavy load (some benchmarks to SCSI disk and transferring files using ftp), SCSI adapter driver prints error message and dumps device's status. and then system is not work. Actually, this error doesn't happen immediately after using the two devices. It works during some seconds (about 10~20secs). I could not capture all the message by SCSI adapter driver printed on console because it is very fast and vast. Anyway, the message means "PCI interrupt error". In L4Linux, the two devices share same IRQ number. IMHO, this is one of the source of this behavior. To use SCSI in L4Linux, I use dm_phys with '--isa=0x00800000' option. Two different IRQ handling methods (one thread per IRQ and multiple IRQs) got the same result. Is there anyone can help me? Regards, -- Jugwan Eom

2 3

Fwd: attempted fix for asm in oskit cpuid.c
by spooq 12 Apr '06

12 Apr '06

*sigh* ---------- Forwarded message ---------- From: spooq <spoooq(a)gmail.com> Date: Apr 11, 2006 11:12 PM Subject: Re: attempted fix for asm in oskit cpuid.c To: Ronald Aigner <ra3(a)os.inf.tu-dresden.de> I see dietlibc's printf is not given a good review on http://www.and.org/vstr/printf_comparison... On 4/11/06, spooq <spoooq(a)gmail.com> wrote: > Ok, I'll start using one of those. Is there any real difference between the two? > > Id also like to know what would need to be done to port L4Env to a new > arm-based arch, eg video card driver (not supported by existing linux > driver), touchscreen, mmu. Is there a list somewhere of interfaces > that need to be implemented? > > btw, just spooq is fine > > -- > spooq > > On 4/11/06, Ronald Aigner <ra3(a)os.inf.tu-dresden.de> wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > > Hash: SHA1 > > > > Hello spoo(o)q, > > > > spooq wrote on 10.04.2006 19:36 this: > > > I tried to fix some errors in cpuid.c ... wouldnt rely on it, my asms > > > not very good and I havent tested it beyond seeing that it compiles. > > > Basically it just makes the asm respect gcc using the ebx as a pic > > > reg. Not sure about the \n\t stuff, it made it work tho. > > Nice job. However, oskit is no longer supported as a libc for L4Env > > applications. Please use either dietlibc or uclibc. > > > > Thanks, keep it up. > > Ron. > > - -- > > Mit freundlichen Gruessen / with regards > > ra3 @ inf.tu-dresden.de > > http://os.inf.tu-dresden.de/~ra3/ > > -----BEGIN PGP SIGNATURE----- > > Version: GnuPG v1.4.2.2 (GNU/Linux) > > > > iD8DBQFEPCGivCdOf9l7ipgRAv9RAKCEMCFLEx1o0ehBLG+f5J+6I9/dIgCgwjdV > > SzC8KfKBWcPvexF8A8vzadw= > > =1wJw > > -----END PGP SIGNATURE----- > > >

3 2

Re: Fwd: attempted fix for asm in oskit cpuid.c
by spooq 12 Apr '06

12 Apr '06

I think this list needs to be fixing its reply-tos... I wasnt the only one caught by it... On 4/12/06, Steffen Liebergeld <s1010824(a)inf.tu-dresden.de> wrote: > On Tue, 11 Apr 2006 23:14:08 +0100, spooq <spoooq(a)gmail.com> said: > > *sigh* ---------- Forwarded message ---------- From: spooq > > <spoooq(a)gmail.com> Date: Apr 11, 2006 11:12 PM Subject: Re: > > attempted fix for asm in oskit cpuid.c To: Ronald Aigner > > <ra3(a)os.inf.tu-dresden.de> > > > > I see dietlibc's printf is not given a good review on > > http://www.and.org/vstr/printf_comparison... > > That comparison is very old, and they tested version 0.22 of > dietlibc. The current version is 0.29 (have a look at > http://www.fefe.de/dietlibc/) > > -- > Microsoft is simply one example of a proprietary software developer, a > software developer that tries to subjugate users to keep them divided > and helpless. -- Richard M. Stallman >

2 1

attempted fix for asm in oskit cpuid.c
by spooq 11 Apr '06

11 Apr '06

Hello list, I tried to fix some errors in cpuid.c ... wouldnt rely on it, my asms not very good and I havent tested it beyond seeing that it compiles. Basically it just makes the asm respect gcc using the ebx as a pic reg. Not sure about the \n\t stuff, it made it work tho. --- virgin_code/oskit/libkern/x86/cpuid.c 2006-04-02 15:41:55.000000000 +0000 +++ cpuid.c 2006-04-10 17:14:57.000000000 +0000 @@ -39,12 +39,17 @@ unsigned i; cicount++; - asm volatile("cpuid" + asm volatile( + "movl %%ebx, %%edi\n\t" + "cpuid\n\t" + "movl %1, %%ebx\n\t" + "movl %%edi, %%ebx\n\t" : "=a" (ci[0]), - "=b" (ci[1]), + "=r" (ci[1]), "=c" (ci[2]), "=d" (ci[3]) - : "a" (2)); + : "a" (2) + : "edi"); for (i = 0; i < 4; i++) { @@ -90,21 +95,29 @@ * CPUID is supported, so use it. * First get the vendor ID string. */ - asm volatile("cpuid" + asm volatile( + "movl %%ebx, %%edi\n\t" + "cpuid\n\t" + "movl %1, %%ebx\n\t" + "movl %%edi, %%ebx\n\t" : "=a" (highest_val), - "=b" (*((int*)(out_id->vendor_id+0))), + "=r" (*((int*)(out_id->vendor_id+0))), "=d" (*((int*)(out_id->vendor_id+4))), "=c" (*((int*)(out_id->vendor_id+8))) - : "a" (0)); + : "a" (0) + : "edi"); /* Now the feature information. */ if (highest_val >= 1) { - asm volatile("cpuid" - : "=a" (*((int*)out_id)), - "=d" (out_id->feature_flags) - : "a" (1) - : "ebx", "ecx"); + asm volatile( + "movl %%ebx, %%edi\n\t" + "cpuid\n\t" + "movl %%edi, %%ebx\n\t" + : "=a" (*((int*)out_id)), + "=d" (out_id->feature_flags) + : "a" (1) + : "ecx","edi"); } /* Cache and TLB information. */ -- spooq

2 1