September 2014 - l4-hackers - example.com Site

Graphical desktop, l4con, mag
by teclis High Elf 02 Jun '17

02 Jun '17

Guten Tag, I've got the Hello app runing with fiasco and l4re in qemu, how do I get the graphical UI and console running that I see on the screen shots page? Do I need to use l4con and mag? I don't see an entry with l4con in modules.list. danke * <http://german.about.com/library/media/Audio/gutentag.wav>*

6 15

shared dataspace for l4re_kernel/ registering additional caps in ned
by Stark, Josef 12 Mar '15

12 Mar '15

Hello all, for a project, I want to be able to load (elf-)binaries not only from the filesystem, but also from the network (receive binary, then start it.) For this, I will write a L4 app (let's call it "manager" for now) which handles the network part and then tells ned to start the received binary via a special "filename". I already modified ned so that its LUA-interface is exposed to other apps via a capability. But here comes the first problem: Normally you specify the capability table of a task and an IPC channel in the lua-config file, which ned reads. Ned then prepares the channel and the caps table and starts the task, which then can register the capability. Otherwise, registering fails (the example would do a printf("Could not register my service, is there a 'calc_server' in the caps table?\n");) So... how do I specify which capabilities ned itself can have/register? Currently, I need to have a lua-config which tells ned to start another instance of ned with the required permission to register the capability. But is there any way to avoid having two ned instances and making the first (and only) instance able to register the cap? Second problem: I found out, that ned (also moe) doesn't start the desired app directly but instead starts the l4re binary (source located in src/l4/pkg/l4re_kernel), passing it the filename of the desired binary. l4re, once started, gets the dataspace of this file and then ultimately starts the app. For my purpose, I would modify l4re_kernel so that if it detects the special "filename" mentioned above, it does not read the file from the filesystem but loads the binary that the "manager" received over the network. To get the data from the manager to the l4re_kernel instance, I thought that a shared dataspace or a shared memory area l4shmc_area_t like in prdocons.c example would be suitable. However, again, both of these things normally require special parameters in the lua config file (as can be seen in the examples), but I have no clue how to do this to l4re_kernel since it is not loaded from such a file, but loaded implicitly each time a new task is started. (so I guess this makes it even harder than the problem with ned since ned is only started once.) Does anyone have ideas on how to solve these? If someone finds that I phrased this too opaque, please let me know. Best regards, Josef Stark

3 16

Ned,moe, l4linux interaction
by Masti Ramya Jayaram 01 Oct '14

01 Oct '14

Dear all, I see that in the default l4linux configuration, it uses ned, moe to start up. However, I am not sure about the run-time interaction between them. More specifically, a. Does l4linux need them during its run-time? For example, does l4linux go directly to sigma0 for memory or does it go through ned? b. Do ned and moe get killed once they start l4linux or not? Thanks, ramya

3 3

Difference between capability and memory mapping
by Yuxin Ren 01 Oct '14

01 Oct '14

Hi All, I found in the kernel, the capability mapping and memory mapping have very similar code path. But I found their performance difference is around 1000 cycles. Is this true? If it is true, why do they have such large difference? If this gap does not make sense, it should be a bug in my test. Thank you very much. Best, Yuxin

2 3

Repository Update
by Adam Lackorzynski 29 Sep '14

29 Sep '14

Hi, the repository and snapshots have been updated and I'm happy to announce a couple of new features: - L4Re supports ARM virtualization (HYP mode) now. There's a 'vmm' package that allows to run unmodified Linux, however, it might still have some rough edges. - With that ARM exception error codes have been changed to use HYP's HSR style, also for non-HYP kernels! - gcc-4.9 support, including libstdc++, and preparations for gcc-5 - EFI is available as a boot method now, both in 32 and 64 bit modes - x86 power management has been improved - Stackprotector support is available in L4Re now - The console server 'cons' has been added - Fiasco runs on more ARM targets, including Xilinx Zynq (Zedboard, Parallela), OMAP5, Allwinner (Cubieboard, Cubietruck) - More examples have been added - 3rd-party packages have been updated - L4Linux has been upgraded to version 3.16. Adam -- Adam adam(a)os.inf.tu-dresden.de Lackorzynski http://os.inf.tu-dresden.de/~adam/

2 1

Question about the doc of L4Re
by lwf408498027 28 Sep '14

28 Sep '14

Hi hackers, I am learning the fiasco's source code recently, and I saw the doc in http://l4re.org/doc/ does not match with the code in l4re-snapshot-2014053111 such as: in class scheduler, there has some function: L4_msg_tag sys_run(L4_fpage::Rights, Syscall_frame *f, Utcb const *utcb); L4_msg_tag sys_idle_time(L4_fpage::Rights, Syscall_frame *f, Utcb *utcb); L4_msg_tag sys_info(L4_fpage::Rights, Syscall_frame *f, Utcb const *iutcb, Utcb *outcb); but in the document, the function are not the same: l4_msgtag_t info (l4_umword_t *cpu_max, l4_sched_cpu_set_t *cpus, l4_utcb_t *utcb=l4_utcb()) const throw () l4_msgtag_t run_thread (Cap< Thread > const &thread, l4_sched_param_t const &sp, l4_utcb_t *utcb=l4_utcb()) const throw () l4_msgtag_t idle_time (l4_sched_cpu_set_t const &cpus, l4_utcb_t *utcb=l4_utcb()) const throw () So, Is this doc out of date? and if the answer is yes, where can i find some fresh doc? if no, How to use the doc correctly? Thanks, Kevin

2 1

L4_SYSCALL_INVOKE is a negative PC value？
by Zachary 28 Sep '14

28 Sep '14

hello everyone, In cache.h file(..\l4\pkg\l4sys\include\ARCH-arm\cache.h), a piece of code is as follows： l4_cache_op_arm_call(unsigned long op, unsigned long start, unsigned long end) { ... __asm__ __volatile__ ("@ l4_cache_op_arm_call(start) \n\t" "mov lr, pc \n\t" "mov pc, %[sc] \n\t" "@ l4_cache_op_arm_call(end) \n\t" : "=r" (_op), "=r" (_start), "=r" (_end) : [sc] "i" (L4_SYSCALL_MEM_OP), "0" (_op), "1" (_start), "2" (_end) : "cc", "memory", "lr" ); } My question is whether the code means that PC jumps to the address of L4_SYSCALL_MEM_OP(actual value is -0x00000010)? If it is true, then it means the address is 0xFFFFFFF0 ? My second question is whether the address 0xFFFFFFF0 corresponds to the sys_arm_mem_op in the following definition? (..\kernel\fiasco\src\kern\arm\asm_entry.h) /*********************************************************************/ /* The syscall table stuff */ /*********************************************************************/ .macro GEN_SYSCALL_TABLE .align 4 .global sys_call_table sys_call_table: .word sys_kdb_ke .word sys_kdb_ke .word sys_ipc_wrapper .word sys_arm_mem_op .word sys_invoke_debug_wrapper .word sys_kdb_ke .word sys_kdb_ke .word sys_kdb_ke .word sys_kdb_ke .word sys_kdb_ke .word sys_kdb_ke .endm Thank you so much. Best, Zachary

2 3

What does idle thread do
by Yuxin Ren 27 Sep '14

27 Sep '14

Hi All, I want to know what kernel idle thread does. Does it just execute "halt" instruction or anything else? Where is code for the idle thread? Thank you so much. Best, Yuxin

2 3

cannot find -lld-l4
by ba_f 25 Sep '14

25 Sep '14

Hello l4-hackers, I get an error when trying to compile, respectively linking my program. /opt/gcc-linaro-arm-linux-gnueabihf-4.8-2014.03_linux/bin/arm-linux-gnueabihf-ld: cannot find -lld-l4 make[3]: *** [myProg] Fehler 1 My Program does not include this lib. It comes from the L4-make process (prog.mk). I have already compiled smaller programs, without problems. BTW before that missing lib error, i get a lot of "multiple definition of...". Any ideas? I use Debian 7 32bit. Thanks

3 4

Does unmap() flush tlb on other cores
by Yuxin Ren 23 Sep '14

23 Sep '14

Hi All, I have question about tlb flush. When a memory page is unmaped, do we always flush tlb of all other cores? Or do we have some way to track which core to flush? If so, how do we track that information? And it would be better if some one can show the code in the kernel to do that. Thanks a lot. Yuxin

2 2