- l4-hackers - example.com Site

Fiasco.OC API documentation
by Neal H. Walfield 28 Oct '13

28 Oct '13

Hi, I'm looking for documentation on Fiasco.OC's API. I started here: http://os.inf.tu-dresden.de/fiasco/doc.html This led me to this page: http://os.inf.tu-dresden.de/L4Re/doc/ This is mostly a theory page with some links to C++ API documentation, e.g.: http://os.inf.tu-dresden.de/L4Re/doc/group__l4__ipc__api.html Where should I be looking for a formal specification of the Fiasco.OC APIs similar to, say: http://www.cs.cmu.edu/afs/cs/project/mach/public/doc/osf/kernel_interface.ps Thanks for your help! :) Neal

2 2

FOSDEM microkernel devroom video recordings
by Julian Stecklina 18 Oct '13

18 Oct '13

Hello fellow microkernel hackers, the FOSDEM organizers are going to try to provide video recording equipment to all devrooms, but don't have the (wo)manpower to do the actual recording. To have recordings of the talks in the microkernel devroom, we need one volunteer. So if you planned on attending the devroom and are not afraid of a video camera, this is a great opportunity to contribute. Just send me a mail. Of course, every speaker is given a chance to opt out of recording his talk. Thanks, Julian PS. Don't forget to submit your talk until Dec 1st: https://os.inf.tu-dresden.de/pipermail/l4-hackers/2013/006068.html

1 0

Invitation to connect on LinkedIn
by Rahul Singhal 17 Oct '13

17 Oct '13

LinkedIn ------------ I'd like to add you to my professional network on LinkedIn. - Rahul Rahul Singhal Student at Indian Institute of Technology, Bombay Mumbai Area, India Confirm that you know Rahul Singhal: https://www.linkedin.com/e/-24ilnx-hmw5i1gw-4n/isd/17375721641/HSxn-9Vu/?hs… -- You are receiving Invitation to Connect emails. Click to unsubscribe: http://www.linkedin.com/e/-24ilnx-hmw5i1gw-4n/d2b6lZFP4STrtUgHOnSplZ2KmSz6E… (c) 2012 LinkedIn Corporation. 2029 Stierlin Ct, Mountain View, CA 94043, USA.

1 0

Client Server application with l4linux
by Wajidali Whowe 14 Oct '13

14 Oct '13

Dear all, I am interested in implementing a client-server application between an l4linux instance and an application running on top of u-kernel. I reviewed the client and server examples and thought writing a character driver for an l4linux which serves as either a client or a server. Nevertheless clearly the client / server examples are written in C++. Do you have any suggestion? Best Regards.

2 3

L4/Fiasco.OC & gcc-4.7
by Daniel Müller 14 Oct '13

14 Oct '13

Hello, are there any known issues with Fiasco.OC in conjunction with gcc-4.7? I've just upgrade from 4.6.3 to 4.7.3 and I see warnings of the form: KERNEL: Warning: nothing mapped: (Obj_space) from [0xffffffffffd97dd8/36]: 000000000000044c size: 0000000000000001 to [0xffffffffffd97d28/50] without changes to any of the previously working source codes. Going back to 4.6.3 fixes the issue. Thanks, Daniel

4 5

scanf is not working
by Raja Sekhar 10 Oct '13

10 Oct '13

Hello all, I'm trying to read input using scanf function. but it's not woring. what is the problem and how can i make it work. *Thanks & Regards* Raja Sekhar

2 1

issue while running l4re_fiasco on qemu-system-arm
by Raja Sekhar 10 Oct '13

10 Oct '13

Hello to all, I have cross compiled l4re-fiasco to panda_board and after running the image on Qemu. I'm getting blank screen and i'm not getting any console messages. and this is the command I used is as.. *$ *qemu-system-arm -M beagle -kernel /home/sets/l4re_work/Fiasco.OC-L4Re_arm/build_panda/L4Re/images/bootstrap.elf How to run l4re-fiasco on qemu-system-arm and what are the options i should provide to the command "qemu-system-arm". could please send me a immediate response..... * Thanks & Regards* Raja Sekhar

2 1

FOSDEM 2014 - CfP: Microkernels / Component-based OSes devroom
by Julian Stecklina 09 Oct '13

09 Oct '13

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 FOSDEM 2014 - CfP: Microkernels / Component-based OSes devroom Fellow Operating System Hackers, built on the success of the microkernel devroom in the last years, FOSDEM 2014 [1] will host a devroom for developers of open-source microkernel- and component-based operating systems. The devroom is scheduled for Sunday, February 2nd, 2014. We are looking for talks in all areas related to microkernels and component-based operating systems. Possible subjects include, but are not limited to: - introduction of a specific OS or framework, - design of subsystems and general architecture of an OS, - hardware support, - success stories, experiences and lessons learned, - security and robustness, - improving the toolchain, - maintainance and testing, - higher-level languages for OS implementation, - community building. To submit a talk, please create an account at the conference's submission system [2] and select the devroom as the talk's track. Should there be any problems, please send a mail to the devroom mailing list [3]. In your submission, be sure to include: - Title and approximate length of the talk (in the General tab), - the correct track (also in the General tab), - a short abstract (in the Description tab). Talks should not exceed 45 minutes. To help us schedule, please indicate whether your talk is flexible in its length. Please submit your proposals before December 1st. We will announce acceptance and hopefully a rough schedule before Christmas. Of course, there are other ways to contribute and make this event a success, for example by doing a demo of your system. Small devices are no problem, but please check back with us, if you plan to bring larger equipment. Please forward this Call for Participation to anyone interested in participating! See you in Brussels, Julian Stecklina [1] http://fosdem.org/2014 [2] https://penta.fosdem.org/submission/ [3] https://lists.fosdem.org/listinfo/microkernel-devroom -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.14 (GNU/Linux) iEYEARECAAYFAlJVRToACgkQ2EtjUdW3H9k18wCg2Bk9GUuSoB/zoInMyj4Vtmgw 19UAoJQNcTh5funOqyg6ezRz7RY9sEeI =/Kfk -----END PGP SIGNATURE-----

1 0

L4Linux on AMD64 (again)
by Alexander Tarasikov 03 Oct '13

03 Oct '13

Hi, l4-hackers! This message is mainly directed to Adam because most of the work on the latest revisions of l4linux was done by him. I have noticed that AMD64 support is present in most of the arch-l4 code, so I have tried compiling the l4linux kernel in the amd64 mode. I have uncommented the “default y” switch in the KConfig to enable CONFIG_X86_64. Then, I fixed some minor compilation issues: * entry.S did not compile. I had to replace the bitwise or with summation, and replaced the X86_EFLAGS_BIT1 with the integer constant (well, I think I could’ve replaced it with X86_EFLAGS_FIXED) * replaced BIT1 with FIXED in some file, I think, head64.c * added the definition of the early page table count to head64.c (it *is* defined in pgtable-64.h, but I was lazy to figure out why it was not properly included) Then, I applied the amd64 patch to L4Re bootstrap (the one in the mailing list) and raised the modaddr to 0x4000000 to work around region conflicts. I have also launched qemu with 512M memory. First of all, the kernel crashed at the SLAB allocator because the cache pointer was uninitialized. Actually, it nanged in the “arch-l4/kernel/main.c” before calling start_kernel. The reason was that the “head_64.c” contained the code to clean the “early page tables” which wrote to the CR3 register. I have commented the writing to the CR3 and it allowed to move a bit further. Subsequently, I have commented out all the code in the file. SLAB allocator fails to allocate because the function “setup_cpu_cache” in the “mm/slab.c” file has the “init_refok” attribute which means it gets placed into the “ref” section. Selecting SLOB allocator makes it go further. Then, it crashed at the “console_init” call. Console_init calls con_init in “drivers/tty/vt.c”, which freezes at the “pr_info” statement (for some weird reason, it does not jump into the printk() from pr_info). It started working after I lowered the default printk level, so it’s a minor issue. Some more crashes were related to the debug/kmemleak init calls, vfs cache setup which I commented out. Eventually, I had the kernel output the debug via the ttyLv0, not via LOG_printf. It complained about some memory which was not within e820 range. The kernel did not fully boot, because the “init_rest” routine is also marked with “init_refok”. So, I think that only the code from the text section works and the pointers to other sections are invalid. Hope my findings will be useful. Now, I have some questions which could save a few hours of debugging. I would be very grateful if you could answer them: * Is amd64 support planned soon? * Could you explain which part of page table manipulation functions are needed, and which can be commented out? Should the page directory be written somewhere in vCPU instead of CR3? * What load_addr should be used? Zero or 0x100000 (I suspect the former). vG8tQK?z,P-UC`a -- Regards, Alexander

2 1

Re: error in bootstrap when building amd64
by Joel Nider 01 Oct '13

01 Oct '13

Hi, > > ? ==> Linking bootstrap32.elf > > /usr/bin/ld: skipping incompatible > > /usr/lib/gcc/x86_64-linux-gnu/4.6/libgcc.a when searching for -lgcc > > /usr/bin/ld: cannot find -lgcc > > collect2: ld returned 1 exit status > > make[4]: *** [bootstrap32.elf] Error 1 > > > Any ideas how to fix it? > > I guess a 32bit version of libgcc might be missing on your system. It is not clear to me why I need a 32-bit bootstrap image when I am trying to build a 64-bit kernel. Is it possible for me to disable/skip this particular file? Everything else seems to build fine. Thanks, Joel

2 1

L4Re::Util::Dataspace_svr
by Daniel Müller 01 Oct '13

01 Oct '13

Hello, L4Re::Util::Dataspace_svr's copy functionality is not very flexible. It only works for local dataspaces which I believe is insufficient. The problem is the "if (!src_cap.id_received())" construct in the L4Re::Dataspace_::Copy case in the dispatch method. Imo this should be something comparable to the validate_ds() method used in region_mapping_svr. Daniel

2 1

ARMV7 virtualization extensions support
by Karoline Haus 01 Oct '13

01 Oct '13

Good afternoon, I'd like to understand if the Fiasco kernel already supports ARM's virtualization extensions (like it supports VT-x and AMD-V), and if not, is that something you are working on? Many thanks for any pointers. Karoline

2 1

error in bootstrap when building amd64
by Joel Nider 01 Oct '13

01 Oct '13

Hi, I'm having trouble building the amd64 L4Re image. I'm building on Ubuntu 12.04 64-bit (kernel 3.2.0-29); make 3.81; gcc 4.6.3 (default) I'm using l4re-core-2013091917 (not the snapshot) that I got from here: http://os.inf.tu-dresden.de/download/snapshots-oc/l4re-core-2013091917.tar.… Everything builds fine until the 'bootstrap' package: === Building package "bootstrap" === make[1]: Entering directory `/home/joel/projects/l4re-core-2013091917/src/l4/pkg/bootstrap' make[2]: Entering directory `/home/joel/projects/l4re-core-2013091917/src/l4/pkg/bootstrap/server' make[3]: Entering directory `/home/joel/projects/l4re-core-2013091917/src/l4/pkg/bootstrap/server/src' make[4]: Entering directory `/home/joel/projects/l4re-core-2013091917/obj/l4/amd64/pkg/bootstrap/server/src/OBJ-amd64_K8' make[5]: Entering directory `/home/joel/projects/l4re-core-2013091917/src/l4/pkg/bootstrap/server/src/ARCH-amd64/libc32' make[6]: Entering directory `/home/joel/projects/l4re-core-2013091917/obj/l4/amd64/pkg/bootstrap/server/src/ARCH-amd64/libc32/OBJ-amd64_K8' make[6]: Leaving directory `/home/joel/projects/l4re-core-2013091917/obj/l4/amd64/pkg/bootstrap/server/src/ARCH-amd64/libc32/OBJ-amd64_K8' make[5]: Leaving directory `/home/joel/projects/l4re-core-2013091917/src/l4/pkg/bootstrap/server/src/ARCH-amd64/libc32' ==> Linking bootstrap32.elf /usr/bin/ld: skipping incompatible /usr/lib/gcc/x86_64-linux-gnu/4.6/libgcc.a when searching for -lgcc /usr/bin/ld: cannot find -lgcc collect2: ld returned 1 exit status make[4]: *** [bootstrap32.elf] Error 1 Any ideas how to fix it? Thanks, Joel

2 1

About the use of C++11 features
by Johannes Schlatow 20 Sep '13

20 Sep '13

Hi, I just noticed that the current snapshot (2013-08-19) doesn't compile with gcc4.4 anymore, which is apparently due to some C++11 incompatibilities (see below). Based on the snapshot from April, I got the Fiasco.OC and L4Re port for the LEON3 (sparc) up and running and would like to merge this with the newest snapshot. Unfortunately, the LEON3 cross compiler (bcc) is based on gcc4.4. So...what's your policy about supporting oldish cross-compilers? And, by the way, how do you feel about including the LEON3 code in your repository? Cheers Johannes Here is the condensed output of a gcc4.4 compilation run: src/kernel/fiasco/src/lib/libk/cxx/function:34: error: ‘cxx::functor<Res(Args ...)>::functor(cxx::functor<Res(Args ...)>&&)’ cannot be defaulted src/kernel/fiasco/src/kern/map_util.cpp: In member function ‘void Mu::Auto_tlb_flush<Mem_space>::global_flush()’: src/kernel/fiasco/src/kern/map_util.cpp:78: error: expected primary-expression before ‘[’ token src/kernel/fiasco/src/kern/map_util.cpp:78: error: expected primary-expression before ‘)’ token src/kernel/fiasco/src/kern/map_util.cpp:741: error: expected primary-expression before ‘[’ token src/kernel/fiasco/src/kern/map_util.cpp:741: error: expected ‘(’ before ‘*’ token src/kernel/fiasco/src/kern/map_util.cpp:741: error: ‘m’ was not declared in this scope src/kernel/fiasco/src/kern/map_util.cpp:741: error: expected ‘(’ before ‘size’

2 1

Booting on Raspberry Pi
by Robert Kaiser 17 Sep '13

17 Sep '13

Hello, i'm currently trying to get Fiasco.OC + L4Re up and running on a Raspberry Pi, but no luck :-( I have gotten as far as building a bootstrap_hello.elf file. According to previous posts here, I should just copy this file to the first partition of a linux-bootable SD card, add an entry like "kernel=bootstrap_hello.elf" to config.txt, and I should see some messages at the serial port as the Pi boots L4. Alas, my Pi stays silent. If I then simply remove the added line from config.txt, leaving everything else the same, Linux boots from the card alright and I can see its messages on the serial port. Thus, I'm confident that my hardware set-up should be OK. I am not absolutely sure wether I got the building of the boot file right. What surprises me is that it is an ELF file: I have previously followed the "Baking Pi" tutorial at http://www.cl.cam.ac.uk/projects/raspberrypi/tutorials/os/. There I used to boot *raw* binaries which expected to be loaded at 0x8000. In contrast, this file here displays as an ELF file with .text starting at 0x01000000. Is that how it should be? The building process also creates a file named "bootstrap_hello", but that is simply a copy of bootstrap_hello.elf. What's the point of this? Thanks a lot for any help Robert Kaiser

2 10

Is mapping database still in use now?
by Sheng-Wei Cheng 15 Sep '13

15 Sep '13

Hi hackers, I've read the paper “µ-kernel Memory Management” and know that, in theory, mapping database is designed for quickly unmaping a page which is recursively mapped to several address spaces. However, is it still in use now in popular implementations like Fiasco.OC? If that is the case, is it used intensively when virtualization on microkernel is involved? (for example, L4Linux etc) Best regards, Lawrence

2 1

vbus problem
by vijay kumar Godavarthy 15 Sep '13

15 Sep '13

Dear all I am using the following code. vbus = l4re_env_get_cap("vbus"); if (l4_is_invalid_cap(vbus)) { printf("Couldn't obtain vbus capability\n"); I am getting the error "Couldn't obtain vbus capability". How to overcome this problem. -- vijay kumar

2 1

How to start DoPe and clients ?
by Rudolf Weber 15 Sep '13

15 Sep '13

Hello, I want to start dope, an fbterminal, and an fb_spectrum, (using l4re-snapshot-2013060718) with the following Start-Script: -- vim:set ft=lua: loader = L4.default_loader; -- Start io vbus_l4linux = loader:new_channel(); local io_buses = { input = loader:new_channel(); fbdrv = loader:new_channel(); }; loader:start( { caps = { sigma0 = L4.cast(L4.Proto.Factory, L4.Env.sigma0):create(L4.Proto.Sigma0); icu = L4.Env.icu; input = io_buses.input:svr(); fbdrv = io_buses.fbdrv:svr(); l4linux = vbus_l4linux:svr(); }, }, "rom/io rom/x86-legacy.devs rom/l4lx-x86.io"); -- Start fb-drv (but only if we need to) local fb = L4.Env.vesa; if (not fb) then fb = loader:new_channel(); loader:start({ caps = { fb = fb:svr(), vbus = io_buses.fbdrv }, log = { "fbdrv", "r" }, l4re_dbg = L4.Dbg.Warn, }, "rom/fb-drv -m 0x117"); end local gui_caps = { svc = loader:new_channel(), dope = loader:new_channel(), }; -- Start dope loader:start( { caps = { vbus = io_buses.input, fb = fb, svc = gui_caps.svc:svr(), dope = gui_caps.dope:svr(), }, }, "rom/l4dope"); local term_srv = loader:new_channel(); print("Start FBterminal "); e=loader:start( { caps = { fb = gui_caps.svc:create(L4.Proto.Goos, "g=640x480"), term = term_srv:svr(), }, log = { "fbterm", "b" }, l4re_dbg = L4.Dbg.Warn, },"rom/fbterminal"); print("Start fb_spectrum"); e2 = loader:start({ caps = { fb = gui_caps.svc:create(L4.Proto.Goos, "g=320x480"), }, log = { "spectrum", "b" }, l4re_dbg = L4.Dbg.Warn, }, "rom/ex_fb_spectrum_cc"); print("fbterminal exited with: " .. e:wait()); print("ex_fb_spectrum exited with: " .. e2:wait()); -------------------- Result: dope starts, but neither the fbterminal nor the ex_fb_specturm_cc is visible. (The kerneldebugger shows, that the Tasks are not existent ...) What is wrong ? Thank you very much Rudolf P.S: I collect my good experience on http://www.infnet.verein.de/betriebssysteme/kleinkernarch/l4/l4re_conf_2013… with mag it was successfull: http://www.infnet.verein.de/betriebssysteme/kleinkernarch/l4/l4re_conf_2013…

2 1

L4Re::Util::Dataspace_svr
by Daniel Müller 15 Sep '13

15 Sep '13

Hi again, I believe the Dataspace_svr dispatch function is missing a case for L4Re::Dataspace_::Allocate (as declared in l4/pkg/l4re/include/dataspace-sys.h) -- I receive this opcode and since there is no handling the result is a return of -L4_ENOSYS (which will make subsequent processing fail). As a quick fix the attached patch works but there should probably also be a hook (virtual function) in the Dataspace_svr class (and maybe some more sophisticated default processing, not sure). Thanks, Daniel

2 1

Problems with SVN co
by Manuel Dejonghe 10 Sep '13

10 Sep '13

Hi all, I subscribed to this list because last night and now, I have problems checking out the source code: After a 'svn co http://svn.tudos.org/repos/oc/l4linux/trunk l4linux' I am getting these kind of error messages: ... A l4linux/include/uapi/linux/netfilter/xt_statistic.h svn: In directory 'l4linux/include/uapi/linux/netfilter' svn: Can't open file 'l4linux/include/uapi/linux/netfilter/.svn/tmp/text-base/xt_CONNMARK.h.svn-base': No such file or directory ... A l4linux/include/uapi/linux/netfilter_ipv4/ipt_ttl.h svn: In directory 'l4linux/include/uapi/linux/netfilter_ipv4' svn: Can't open file 'l4linux/include/uapi/linux/netfilter_ipv4/.svn/tmp/text-base/ipt_ecn.h.svn-base': No such file or directory ... A l4linux/include/uapi/linux/netfilter_ipv4/ipt_CLUSTERIP.h svn: In directory 'l4linux/include/uapi/linux/netfilter_ipv4' svn: Can't open file 'l4linux/include/uapi/linux/netfilter_ipv4/.svn/tmp/text-base/ipt_ttl.h.svn-base': No such file or directory I have had that as well after only a couple of seconds or after some minutes, with varying milage of completeness, of course. What could be the reason for that ? Does anyone else have strange symptoms currently with the subversion ? thank you very much, lImbus

1 1

L4kproxy::Factory_svr
by Daniel Müller 09 Sep '13

09 Sep '13

Hello, Please find attached a minimal example illustrating a bug in Factory_svr. When executing this example the output is something like: Ned: loading file: 'rom/minimal.lua' Ned: ERROR: ned.lua:199: could not create process: Insufficient memory (: -12) The problem is in parsing the utcb_area parameter in the Factory_hndl::handle_task() method. A patch could be the attached libkproxy.diff which will make the task creation succeed. I would also like to point out that in my opinion there is no point in writing this functionality but never testing or using it (I have been unable to find a single client or test using it). Thanks, Daniel

2 2

Fosdem'14
by Sartakov A. Vasily 09 Sep '13

09 Sep '13

Hello, my dear friends. I am sorry if you receive multiple copies of this message (but I know, most of you will receive copies), but mailing list from uKernel dev room at FOSDEM'13 was destroyed (https://lists.fosdem.org/listinfo) Time rolls on, a FOSDEM'14 is not too far and now it is good time to submit a "uKernel and component based os" dev room. Who would like to organize dev room in this year? I would be proud to make this, but I heard someone from TUD would like to organize Dev room in this year. -- Sartakov A. Vasily sartakov(a)ksyslabs.org

2 2

L4::Cap::move
by Daniel Müller 06 Sep '13

06 Sep '13

Hello, I am wondering is there any particular reason why Cap::move() uses L4_FPAGE_RWX instead of a mask of all bits? l4_task_map(L4_BASE_TASK_CAP, L4_BASE_TASK_CAP, src.fpage(L4_FPAGE_RWX), snd_base(L4_MAP_ITEM_GRANT)); Shouldn't L4_FPAGE_RWX be replaced with L4_CAP_FPAGE_RWSD in order to not restrict rights artificially? I am seeing subtle differences in application behavior in a scenario of mine... Daniel

2 1

Cross-CPU TLB flush problem with Fiasco.OC (r56, r58; exynos5)
by Christian Prochaska 01 Sep '13

01 Sep '13

Hello, I had a situation where CPU 1 accessed the wrong (same as before) physical memory location after it was idle while CPU 0 unmapped the memory page and mapped a different page at the same virtual address afterwards. When a CPU is idle, its TLB gets marked as 'inactive' in the kernel and the CPU doesn't receive cross-CPU TLB flush requests during this time. It seems there is a TLB flush missing somewhere after CPU 1 leaves the idle thread, to compensate for any missed flush requests during the idle time. When I added a 'Mem_unit::tlb_flush()' call after 'enable_tlb(cpu)' in 'Kernel_thread::idle_op()' as a test, this problem did not occur anymore. Christian

2 1

Top runscript
by Daniel Müller 28 Aug '13

28 Aug '13

Hello, I've seen the l4/pkg/top package and find it to be quite interesting. Can someone point me to a lua-runscript to start a program to be monitored by it? What caps do I have to redirect? Thanks, Daniel

1 0