November 2013 - l4-hackers - example.com Site

NOVA, NUL, snapshotting
by Sartakov A. Vasily 13 Jan '14

13 Jan '14

Hello. Does the NOVA/NUL support snapshotting of VMes? -- Sartakov A. Vasily

5 21

Dynamic creation and passing of IPC channels
by Robert Kaiser 13 Dec '13

13 Dec '13

Hi all, I have two tasks running and communicating through an IPC connection (basically the clntsrv demo). Now I would like to extend this a bit: Using an ipc_call(), the client (actually: multiple clients) should be able to register with a server to receive a wakeup IPC at individually different points in time in the future. (First, flawed idea to do this was to *not* use the combined reply_and_wait call on the server side but instead to only do a wait() call and then postpone the corresponding reply() until when the client should be woken up. However, I learned that this will not work because the implicit reply capability will be lost as soon as another ipc call is done by the server.) Next idea now is to create another IPC channel in the client and to pass (via IPC) the send capability for that channel to the server. The client shall than block in an ipc_receive() on that channel and the server can wake it by sending to that channel. I guess this should be possible and for a seasoned Fiasco.OC/L4Re expert it is probably quite easy to do, but I can't seem to find how to: - create an IPC channel at client/server run time (so far, all IPC channels I used were prepared by Ned through lua scripts) - pass the send capability for such a channel to another task via IPC Can anyone point me to some helpful documentation or maybe an example I could cannibalize to implement this (preferably in C, but C++ code would surely be helpful too)? Thanks in advance for any help! Robert

4 6

serial_drv failed
by Raja Sekhar 28 Nov '13

28 Nov '13

Hello all, I'm trying to run serial_drv on x86 . But I'm getting the following error...... *serial-drv: virtual base at:0x3000* *L4Re[rm]: mapping for page fault failed with -34 @3030 pc=10008e8* *L4Re: unhandled exception: pc=0x10008e8* What migh be the reason....... *Thanks,* Raja Sekhar

2 1

FOSDEM microkernel devroom submission deadline nears (Dec 1st)
by Julian Stecklina 27 Nov '13

27 Nov '13

Hello fellow microkernel hackers, the deadline for submitting talks to our devroom in Brussels is December 1st. If you haven't already, consider submitting a talk. Details are in the original CfP mail: https://os.inf.tu-dresden.de/pipermail/l4-hackers/2013/006068.html Submissions are done via FOSDEM's Pentabarf: https://penta.fosdem.org/submission/ If Pentabarf doesn't work for you, please send your proposal or any other questions regarding the devroom to: microkernel-devroom(a)lists.fosdem.org See you in Brussels, Julian

1 0

[ASK] Make L4 + Fiasco ISO to boot in pandaboard
by Erry Pradana Darajati 25 Nov '13

25 Nov '13

Hello L4hackers, I have a project from my college to build and configure fiasco with l4re to boot and run in Pandaboard. So far i have already build the kernel and l4. But i am having trouble to making ISO for l4re. i follow the instruction using *grub2iso *but it unsuccessful. The Error says : *This mode can only be used with architectures x86 and amd64*. My question is, how can i make ISO for Pandaboard ARM. Thank you for the answer, i am really new for this subject, so i am really appreciate any help -- *Regards,* *Erry Pradana Darajati | Student of Computer Engineering* *Politeknik Elektronika Negeri Surabaya*

2 1

Using Trustzone on mx53 and L4Linux
by panton 21 Nov '13

21 Nov '13

Hello Hackers, I am trying to run the example on l4/examples/sys/vm-tz on my imx53 qsb. I get something running but I am pretty sure it is not correct. I have found a message called "Full virtualization with Trustzone" that make me wonder several questions and I hope you can help me. 1- The vm-tz example is supposed to run a L4Linux on NS world. What I don't understand in which format should be the linux-image (or any other app you want to run). I use a gz compress linux image (the same I use for running "L4Linux ARM" example), but I think it should be a bin file (not even a ELF). 2- The vm-tz example make use of l4sigma0_map_iomem() to request a memory mapping, but why not a l4sigma0_map_mem() if what you want is RAM mapping? The serial exit I get is: SIGMA0: Dump of all resource maps RAM:------------------------ [0:70000000;70000fff] [0:70063000;7008ffff] [0:70097000;70097fff] [0:7009f000;7013ffff] [4:70140000;7016cfff] [0:7016d000;7016ffff] [4:70170000;70186fff] [0:70187000;7100cfff] [4:7100d000;7100dfff] [0:7100e000;710fffff] [4:71100000;7182cfff] [0:7182d000;8effffff] IOMEM:---------------------- [0:0;6fffffff] [0:90000000;ffffffff] MOE: Hello world MOE: found 499832 KByte free memory MOE: found RAM from 70000000 to 8f000000 MOE: allocated 496 KByte for the page array @0x7009f000 MOE: virtual user address space [0-bfffffff] MOE: rom name space cap -> [C:501000] BOOTFS: [716eb000-71704494] [C:503000] l4re BOOTFS: [71705000-7171e674] [C:504000] vm-tz BOOTFS: [7171f000-7182c73c] [C:505000] io BOOTFS: [71100000-711000bd] [C:506000] arm-rv.io BOOTFS: [71101000-713ea4b4] [C:507000] vmlinuz.arm (My gz compressed linux image) BOOTFS: [713eb000-716eb000] [C:508000] ramdisk-arm.rd MOE: cmdline: moe --init=rom/vm-tz MOE: Starting: rom/vm-tz MOE: loading 'rom/vm-tz' And next is repeated forever without change. VM run VM exit pc:20007000 cpsr:00000013 r0:00000000 r1:0000033b r2:20007100 r3:00000000 r4:00000000 r5:00000000 r6:00000000 r7:00000000 r8:00000000 r9:00000000 r10:00000000 r11:00000000 r12:00000000 lr_svc:00000000 sp_svc:00000000 spsr_svc:00000000 Thanks in advanced for your answers, if any :). Pablo Anton.

2 1

Cons
by Daniel Müller 11 Nov '13

11 Nov '13

Hi, Can someone please explain in a few sentences what the purpose of the cons application (l4/pkg/cons) is? Is it (just) a multiplexer for output? Thanks, Daniel

2 1

question about process generation
by Alexander Aring 08 Nov '13

08 Nov '13

Hi all, I like to implement two processes(if this is the correct word for l4). I want to implement two elf binaries, each of them have a main function. These two processes should communicate over IPC. Is this possible? I have in my modules.list: entry krusty kernel fiasco -serial_esc roottask moe rom/krusty.cfg module foo ... so my init process is 'moe' which should start one than more processes, how can I do that if it possible? In l4 examples of 'ipc' I only saw solution with threads, is this the correct way to initiate new processes? - Alex

2 1

about the snd-base/hot-spot of flexpage mapping
by 冯希顺 08 Nov '13

08 Nov '13

hello everbody. I have a question about the snd-base/hot-spot of flexpage mapping feature. in many l4 documents, there are a lot of words about the snd-base/hot-spot concept, it seem that are a complex mechanism to calculate the send base when the sender's flexpage have different size compares to receiver's flexpage, I have read a lot of documents(l4 user manual/l4 reference manual etc.) about this mechanism, but I can not figure out it... Could you tell why l4 needs this mechanism, and what situation that this mechanism adapt to? And I have another question to Dresden University guys: what is the l4 version of Fiasco.OC based? Thank for your help! Naus

2 1

some question bout L4 memory management
by yakijahn 07 Nov '13

07 Nov '13

Hello L4 hackers Now a days I am studying the memory manage Mechanisms of fiasco kernel and L4RE, and I have some questions: 1) I see that when a process need some memory, a dataspace will be invoked and MOE will retrieve pages from alloc_list as kernel mapping sender, and attach region to region map. if process free a region of memory , the region will be dettached from region map and link the memory back to alloc_list, right? The problem I hold is when and how memory be recycled when a process terminated ? Cause I didn't see system use alloc_list :: free() function when a process had ended, in the process I malloc() about 1 MB memory and didn't free them, but it seems that the memory pre-allocated not been link back to alloc list, could you please tell me when and how will system recycle these memory? If possible, please tell me which code folders is it there. 2) Could you please tell me where are codes(files) that responsible for process virtual memory space management, such as how vitual address be assigned to a variable or pointer. Sometimes i fell to an unhandled page fault when I free pointers, because the page fault address cant be find in region map. Thank you very much for kindly answering Best Reguard YuFen Chang National Tsing-Hua University

2 1