- l4-hackers - example.com Site

Did not find VBE info block
by Shakthi Kannan 26 Sep '05

26 Sep '05

Greetings! I am trying to get the console using l4env and fiasco on an old x86 desktop. Is it like the bash shell? Images have been compiled from CVS sources. All images are booted from the grub present in the demo floppy. The grub entry is as follows: kernel (fd0)/l4/v2/rmgr -sigma0 modaddr 0x01500000 kernel (fd0)/l4/v2/main -noserial -nowait -nokdb kernel (fd0)/l4/v2/sigma0 kernel (fd0)/l4/v2/names kernel (fd0)/l4/v2/log kernel (fd0)/l4/v2/dm_phys kernel (fd0)/l4/v2/con The output that I get while booting is as follows: " con | OSKit support: using 128 KB at 0x00002000 as heap con | Characters size is 8x14, font has 256 characters con | vc_init_gr (): con | [8.2] l4con/server/src/vc.c: 315: vc_init_gr(): con | Did not find VBE info block in multiboot info con | Perhaps you have to upgrade GRUB, RMGR or oskit10_support con | GRUB has to set the video mode with the vbeset command con | con | Alternatively, try passing the --vbemode=<mode> switch con | ----PANIC, 'g' for exit " By console, does it correspond to serial port output or the VGA display? The desktop is a Sony Trinitron multiscan 15sx display. Any suggestions/ideas would be useful, Thanks, SK -- Shakthi Kannan, MS Software Engineer, Specsoft (Hexaware Technologies) [E]: shakthimaan(a)yahoo.com [M]: (91) 98407-87007 [W]: http://www.shakthimaan.com [L]: Chennai, India __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 5

Problems porting L4Linux 2.4 module to 2.6
by Derick Swanepoel 21 Sep '05

21 Sep '05

Hi, I'm trying to convert a custom kernel 2.4 module to 2.6. The most serious problem I'm having is that there are now some unresolved symbols with 2.6: *** Warning: "l4rm_detach" [/home/drops/DROPS-20050908/l4/pkg/test/examples/module.ko] undefined! *** Warning: "kmalloc" [/home/drops/DROPS-20050908/l4/pkg/test/examples/module.ko] undefined! *** Warning: "l4rm_do_attach" [/home/drops/DROPS-20050908/l4/pkg/test/examples/module.ko] undefined! *** Warning: "l4env_err_ipcstrings" [/home/drops/DROPS-20050908/l4/pkg/test/examples/module.ko] undefined! *** Warning: "l4lx_thread_shutdown" [/home/drops/DROPS-20050908/l4/pkg/test/examples/module.ko] undefined! *** Warning: "l4lx_thread_create" [/home/drops/DROPS-20050908/l4/pkg/test/examples/module.ko] undefined! *** Warning: "l4_sleep" [/home/drops/DROPS-20050908/l4/pkg/test/examples/module.ko] undefined! *** Warning: "names_waitfor_name" [/home/drops/DROPS-20050908/l4/pkg/test/examples/module.ko] undefined! I see that there is no l4_ksyms.c in l4linux 2.6 like there was in 2.4. Secondly, I get more unresolved symbols if the objects that are linked are specified at the end of the command line instead of the beginning (linking succeeds in both cases, however). The kernel build system puts the objects at the end while the DROPS build system puts them at the beginning. Does anyone have an idea why two otherwise equivalent linking commands would cause such different behaviour depending on the position of the object arguments? Here are the two command lines (the paths are slightly different because the kernel build system always outputs to the source directory: DROPS: gcc -o test_module.o test_dev.o main.o test_dm-client.o -r -nostdlib -nostartfiles -L../../../../../lib/x86_686/l4v2 -L/home/drops/DROPS-20050908/drops/lib/x86_686/l4v2 -L../../../../../lib/x86_686 -L/home/drops/DROPS-20050908/drops/lib/x86_686 -L../../../../../lib -L/home/drops/DROPS-20050908/drops/lib -T../../../../../lib/x86_686/main_rel.ld -ltest_lxk -ldm_generic -ldm_phys -ldm_mem kernel: gcc -r -nostdlib -nostartfiles -L/home/drops/DROPS-20050908/drops/lib/x86_686/l4v2 -L/home/drops/DROPS-20050908/drops/lib/x86_686 -L/home/drops/DROPS-20050908/drops/lib -T/home/drops/DROPS-20050908/l4/lib/x86_686/main_rel.ld -ltest_lxk -ldm_generic -ldm_phys -ldm_mem OBJ-x86_686-l4v2/mesg_dm-client.o -o test_module.o test_dev.o main.o Thanks, Derick

2 4

Problem with hybrid L4Linux task
by Mathias Krause 19 Sep '05

19 Sep '05

Hello Hackers, I've got a problem with a hybrid Linux Task. I want to create a little tool to tranfser data between two L4Linuxes easily. A tool similar to the famous netcat but using IPC instead of UDP/TCP. So far so good. Server is listen for client and client is sending data to the server (using long IPC transfering data as dwords). The problem is that the data doesn't arrive at the server task. There only an empty IPC message gets received (no IPC error). After I've played around a while with debug messages I got strange results. Using debug messages (simple printf()) inbetween and the transfer succedes! Digging a little deeper got me to the point that it must be a problem with the compiler. Using 'gcc -O0' generates working code too - 'gcc -O2' doesn't. Trying another compiler (gcc version 4.0.1 (Debian 4.0.1-2)) with -O2 got me also working code. So my question is if really my gcc (version 3.3.5 (Debian 1:3.3.5-13)) is broken or the includes under l4/l4sys/.../ipc.h. Maybe the inline assembly isn't that clean? But just a guess. My GNU assembly knowlegde isn't that good. Greets, Mathias PS: I'm runing Debian sarge over here.

2 4

Compiling DROPS' dietlibc on FreeBSD
by Julian Stecklina 15 Sep '05

15 Sep '05

Hello, after changing minimal parts of dietlibc's build procedure, I get as far as cd pkg/dietlibc ; gmake actually doing something but it fails with the following error: ... Compiling lib/__v_printf.o LD_PRELOAD=/home/blitz/src/l4/tool/gendep/libgendep.so GENDEP_TARGET=lib/__v_printf.o GENDEP_BINARY=cc1 gcc33 -c -nostdinc -DRAM_BASE=0x0 -DUSE_DIETLIBC=y -DSYSTEM_x86_686 -DARCH_x86 -DCPUTYPE_686 -DL4API_ -I../include -I.. -I../../../idl/OBJ-x86- -I/include -gstabs+ -g -O2 -fno-strict-aliasing -march=i686 -Wall -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations /home/blitz/src/l4/pkg/dietlibc/lib/dietlibc/lib/__v_printf.c -o lib/__v_printf.o /home/blitz/src/l4/pkg/dietlibc/lib/dietlibc/lib/__v_printf.c:2:20: stdarg.h: No such file or directory In file included from /home/blitz/src/l4/pkg/dietlibc/lib/dietlibc/lib/__v_printf.c:7: ../dietstdio.h:12:20: stdarg.h: No such file or directory In file included from ../dietstdio.h:52, from /home/blitz/src/l4/pkg/dietlibc/lib/dietlibc/lib/__v_printf.c:7: ../include/stdio.h:7:20: stdarg.h: No such file or directory In file included from ../dietstdio.h:52, from /home/blitz/src/l4/pkg/dietlibc/lib/dietlibc/lib/__v_printf.c:7: The include directories look bogus, but I am quite puzzled where they come from. Maybe my directory layout is somehow different from what is expected? Any hint is greatly appreciated. Regards, -- Julian Stecklina Lisp nearing the age of 50 is the most modern language out there. GC, dynamic, reflective, the best OO model extant including GFs, procedural macros, and the only thing old-fashioned about it is that it is compiled and fast. -- Kenny Tilton, comp.lang.python

4 9

[Fwd: Re: Inspect L4Linux Memory]
by Julian Grizzard 15 Sep '05

15 Sep '05

Forgot to copy to list...message below -------- Original Message -------- Subject: Re: Inspect L4Linux Memory Date: Tue, 13 Sep 2005 14:12:14 -0400 From: Julian Grizzard <grizzard(a)ece.gatech.edu> To: Marcus Voelp <voelp(a)os.inf.tu-dresden.de> References: <432214B5.8010109(a)ece.gatech.edu> <43225D54.7040009(a)os.inf.tu-dresden.de> Marcus and all, I like option two better: sharing the dataspace that provides L4Linux with memory. After studying docs under l4/pkg/dm_phys/doc/dm_phys/html, seems like first I need to make an l4dm_share call from the L4Linux server. Then, I should be able make an l4dm_map call from the isolated L4 task. Does this sound like the right general approach? Also, I have a question about the examples under l4/pkg/dm_phys/examples. Take list_ds. There is a call to "if_l4dm_generic_list_call". Where is this call defined? It's not clearly locatable in the API? Is this an IDL thing? I'm not too familiar with IDL. Thanks! -Julian Marcus Voelp wrote: > Hi, > there are actually two nice ways how you can access L4Linux kernel > memory. The first is to install a driver to L4Linux which creates an L4 > thread that in turn would read out the memory when requested to do so. > The second is mapping the L4Linux memory to your external observer by > sharing the dataspace that provides L4Linux with memory. > > The first solution obviously is not suited when reading out much data > and you also rely on the thread to live (i.e., if Linux crashes there is > the chance that the driver thread crashes as well). The second solution > provides a faster read access and survives also kernel crashes. > > Regards > Marcus > > Julian Grizzard wrote: > >> Hi all, >> >> I am running L4Linux 2.6 on Fiasco using L4Env. I would like to build >> an isolated L4 task that can inspect the contents of the L4Linux task >> state. Any pointers on the best way to do this? Conceivably, I could >> map the contents of the L4Linux task into my isolated L4 task to read >> the memory, but I'm not sure the best way to do this. >> >> Thanks! >> >> -Julian >> >> _______________________________________________ >> l4-hackers mailing list >> l4-hackers(a)os.inf.tu-dresden.de >> http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers >> >> >

3 2

Inspect L4Linux Memory
by Julian Grizzard 10 Sep '05

10 Sep '05

Hi all, I am running L4Linux 2.6 on Fiasco using L4Env. I would like to build an isolated L4 task that can inspect the contents of the L4Linux task state. Any pointers on the best way to do this? Conceivably, I could map the contents of the L4Linux task into my isolated L4 task to read the memory, but I'm not sure the best way to do this. Thanks! -Julian

2 1

Re: Reboot does not work anymore!
by Derick Swanepoel 08 Sep '05

08 Sep '05

Very sorry, this should've gone to the list. ---------- Forwarded message ---------- From: Derick Swanepoel <dswanepoel(a)gmail.com> Date: 08-Sep-2005 12:21 Subject: Re: Reboot does not work anymore! To: Frank Mehnert <fm3(a)os.inf.tu-dresden.de> On 8/17/05, Frank Mehnert <fm3(a)os.inf.tu-dresden.de> wrote: > On Wednesday 17 August 2005 15:48, Leon Brits wrote: > > > Should not does any matter. But how do you set the video mode, via GRUB > > > vbeset of via --vbemode parameter of l4con? > > > > grub > > Hmm. Very strange. When you reboot via the l4con console you most probably > mean the pkg/loader/example/run program. Could you please disassemble that > binary > > objdump -ld loader/example/run/OBJ-x86_586-l4v2/run | less > > and search for pc_reset, right? That function should come from oskit10 and > we did not change anything there. Hi, I'm having the same problem (CVS 20050829). Rebooting does not work from "run" or from JDB. It gets as far saying: Return reboots, "k" enters L4 kernel debugger... Rebooting. Shutting down... but then nothing happens. This is on a Pentium 4 machine. Strangely, however, it works on a Xeon. For what it's worth, here is the disassembled pc_reset(): 013082f0 <pc_reset>: pc_reset(): ../kern/x86/pc/reset.c:30 13082f0: 55 push %ebp 13082f1: 89 e5 mov %esp,%ebp 13082f3: 53 push %ebx 13082f4: 50 push %eax 13082f5: bb 63 00 00 00 mov $0x63,%ebx ../kern/x86/pc/reset.c:34 13082fa: a1 ec 05 31 01 mov 0x13105ec,%eax 13082ff: 66 c7 80 72 04 00 00 movw $0x1234,0x472(%eax) 1308306: 34 12 ../kern/x86/pc/reset.c:39 1308308: 83 ec 0c sub $0xc,%esp 130830b: 68 fe 00 00 00 push $0xfe 1308310: e8 1b 00 00 00 call 1308330 <kb_command> ../kern/x86/pc/reset.c:37 1308315: 83 c4 10 add $0x10,%esp 1308318: 4b dec %ebx 1308319: 79 ed jns 1308308 <pc_reset+0x18> ../kern/x86/pc/reset.c:45 130831b: 0f 01 1d 9e 04 31 01 lidtl 0x131049e ../kern/x86/pc/reset.c:46 1308322: cc int3 1308323: 8b 5d fc mov 0xfffffffc(%ebp),%ebx 1308326: c9 leave 1308327: c3 ret Thanks, Derick

2 1

Sawmill's dataspaces and the Hurd's physmem
by Neal H. Walfield 06 Sep '05

06 Sep '05

During my presentation of the design of the Hurd on L4 to the Dresden group, Lars Reuther asked me if I had considered Sawmill's dataspace model and if so why I had rejected it. My answer was that we want to be able to catch any errors at the time of data acquisition and that relying on a file system server to provide a mapping to the data does not guarantee this: a malicious server can unmap mappings and refuse to remap them; or if the server is shorter lived than the client, the the data becomes inaccessible to the client when the server exits. Lars asked for more details. Unfortunately, I failed to provide a coherent and complete argument: my problem was mainly that I came to this conclusion fairly early in the design process and had since forgotten too many details of Sawmill's architecture to reconstruct my argumentation. I've recently revisited some related issues and am now in a position to better answer Lars's question. My primary reference for the Sawmill framework is [1]. I think I may be able to best explain the problem with an illustration: consider a server providing access to a file system backed by a disk. In the Sawmill dataspace model [1], the file system server is a dataspace manager which likely provides a dataspace for each file. When a task wants to use a file, it first identifies the dataspace associated with the file (e.g. gets a capability to the file from the DM) and attaches it to its address space (e.g. tells its pager to associate a portion of the VM with the capability). "After an attach, the region mapping forwards all page fault requests in that region to the dataspace manager. The dataspace manager resolves the faults with map or grant operations." I understand this to mean that a client depends on a DM to: - provide mappings to data - provide resources backing the data When a client requests some data from a dataspace, the DM provides a mapping to the client. The client can proceed to use the data, however, at any point, the server could cause the mapping to be unmapped and possibly render the data inaccessible. The implication is that the client must either trust the server to always provide the mapping or be prepared to recover should the data disappear. The latter approach can be simplified by making a physical copy of data before committing to using it (which can be done by interposing a second DM between the DM and the client). General use of this tactic means a lot of cycles will be spent copying bytes and a reduction in the amount of physical memory sharing in the system. DMs appear to use their own resources to fetch and store data as well as to hold data (neither [1] nor [2] mentions any mechanism for the client to specify to the DM what memory/data space to read data into). I assume that the normal mode of operation is that a file system DM has a certain amount of physical memory available from a "physical memory" DM which it uses to hold data from backing store. Once that memory is exhausted, it must choose some page to evict. There are several problems with this model: because DMs allocate resources on behalf of clients resources are allocated with the priority of the DM and resource accounting is extremely difficult. We know this from our experience with the Hurd on Mach. Moreover, the DM controls the paging policy; not the clients who are actively using the memory. To control the availability of memory, it would seem that a client would again have to copy the data. [ Physmem DM ] | v [ FS DM ] | v [ client ] The framework that I have developed for the Hurd avoids these dependencies on the file system servers. The physical memory manager ("physmem") is part of the TCB. physmem provides capabilities to so-called containers which identify memory reservations (either specific, e.g. a specific set of frames, or general, e.g. a specific number of frames). Given a container capability, the holder can map the contents or logically copy the contains to or from a second container. When a task wants to read data from a file, it passes a container capability to the file system server. The file system server stores the data in the container (if it already read it into memory then it can logically copy it). Then it returns to the client. If the task wants a mapping to the data, it requests one from physmem. Thus, tasks on the Hurd do not depend on file system server to provide mappings. After the read is complete, the task knows that the data is either available or not available. If the file system server exits, this does not affect the data that the client has. [ physmem ] / \ |_ _| [ client ] [ FS ] Because the client passes containers to the server, the server does not allocate memory to store the data on behalf of the client. (The server may need other resources such as CPU time, I/O bandwidth and state, however, we have other provisions for that.) Thus memory allocations are directly attributed to the client which is vital to the correct functioning of accounting. Also, when the client has exhausted its memory quota, it must free some memory before it can allocate a container with the required reserve. Thus, clients fully control their own paging policy. One of the goals of the Hurd framework is to minimize the number of dependencies that a client has on the behavior of servers. This is important to us because clients often interact with other users's servers which may be malicious. Another goal is to more directly link consumers of resources with resource allocators. Our observation is that many applications are hurt due to policies such as the eviction scheme imposed by the OS. Moreover, applications such as garbage collectors and multimedia applications could benefit from knowing how much real resource is actually available to them. (Applications should work in harmony with the mechanisms provided to them and the policies imposed on them; they should not have to work around them.) Thanks. I have tried to be as concise as possible which means I may have missed some details. I am particularly interested in the thoughts of the Sawmill and DROPS developpers. Neal [1] http://l4ka.org/publications/2001/sawmill-framework.pdf [2] http://os.inf.tu-dresden.de/l4env/doc/l4env-concept/l4env.pdf

4 6

DROPS build success on FreeBSD
by Julian Stecklina 04 Sep '05

04 Sep '05

Hello, I managed to compile most (the documentation fails with latex errors) of DROPS using the attached patch on FreeBSD 5.4. Notes on how to build drops on L4 are attached as well. Notable difficulties in porting: * bash assumed to be in /bin/ (solved with sed) * assuming Linux-style /proc in gendep * assuming dynamically-linked gcc It might be worthwile to make kernel/fiasco/src/Makeconf include mk/Makeconf to avoid redundancy. I think the whole build process would be quite portable with some autoconf magic and with gendep removed. Also I did not test the result. Maybe tomorrow. Regards, -- Julian Stecklina (Of course SML does have its weaknesses, but by comparison, a discussion of C++'s strengths and flaws always sounds like an argument about whether one should face north or east when one is sacrificing one's goat to the rain god.) -- Thant Tessman

1 1

gendep on FreeBSD
by Julian Stecklina 03 Sep '05

03 Sep '05

Hello, in my quest to get the Fiasco tree buildable on FreeBSD I encountered the following problem (this is with gendep compiled with VERBOSE=1) [...] echo BID_LIBGENDEP=/home/blitz/src/l4/tool/gendep/libgendep.so >>./.Makeconf.bid.converted; echo 'BUILD_PIC += $(BUILD_LOADER_PICS)' >>>>./.Makeconf.bid.converted set -e; X="tmp.$$$RANDOM.c" ; echo 'int main(void){}'>$X ; rm -f $X.out ; LD_PRELOAD=/home/blitz/src/l4/tool/gendep/libgendep.so GENDEP_SOURCE=$X GENDEP_OUTPUT=$X.out gcc33 -c $X -o $X.o; test -e $X.out; echo INT_CPP_x86_NAME=`cat $X.out` >>./.Makeconf.bid.converted; rm -f $X $X.{o,out}; fopen(/proc/self/cmdline) real_fopen(/proc/self/cmdline) gendep: Cannot resolve fopen(): Unknown error: 0 It seems that RTLD magic is somehow broken, but I cannot say how... if someone has a hint, I am all ears. My patches to gendep are attached. Regards, -- Julian Stecklina (Of course SML does have its weaknesses, but by comparison, a discussion of C++'s strengths and flaws always sounds like an argument about whether one should face north or east when one is sacrificing one's goat to the rain god.) -- Thant Tessman

1 1