- l4-hackers - example.com Site

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

Re: std::cin in Fiasco/L4Re (with 'make qemu')
by Stark, Josef 23 Sep '14

23 Sep '14

Hi, >I'm not sure where you enter this. >For me, the following works quite fine: > >#include <stdio.h> > >int main(void) >{ > printf("Press some key: "); > char c = getchar(); > printf("\nYour input: %c (%d)\n", c, c); > > printf("Enter a number: "); > int number; > scanf("%x", &number); > printf("Your number: %d\n", number); > > return 0; >} If I copy this code into the hello example, and change the Makefile accordingly: PKGDIR ?= ../.. L4DIR ?= $(PKGDIR)/../.. TARGET = hello SRC_C = main.c REQUIRES_LIBS = libc_be_file_stdin include $(L4DIR)/mk/prog.mk I can successfully run it IF moe runs hello directly (without Ned and a config, as a "make qemu E=hello" does). If, however, hello is run through a config file and therefore Ned, as in hello-cfg, it doesn't work. Execution does not halt and outputs invalid values: MOE: Starting: rom/ned rom/hello.cfg MOE: loading 'rom/ned' Ned says: Hi World! Ned: loading file: 'rom/hello.cfg' hello | Press some key: hello | Your input: � (-1) hello | Enter a number: Your number: -2147450972 How to make it work? Pass a certain capability to hello? Josef

2 1

l4/sys/syscalls.h: No such file or directory
by Valentin Hauner 18 Sep '14

18 Sep '14

Hi, I'm using the following example provided by Michael Hohmuth in 2000 to create new tasks and threads: http://os.inf.tu-dresden.de/pipermail/l4-hackers/2000/000384.html I've placed the code in a copy of the hello example, but added > REQUIRES_LIBS = l4sys to the Makefile. The directory structure is located at './l4/pkg/taskcreat'. Nevertheless, compilation is terminated due to a fatal error: > ./l4/pkg/taskcreat/server/src/main.c:5:29: fatal error: l4/sys/syscalls.h: No such file or directory Why is syscalls.h missing here? Best regards, Valentin

4 52

How to debug Fiasco Kernel
by cem akpolat 18 Sep '14

18 Sep '14

Dear all, I was searching how to debug the fiasco kernel through JDB. The following link gives a good approach how to it through minicom. http://os.inf.tu-dresden.de/l4env/doc/html/dev-overview/ However, even if I see the messages on the screen, is it possible to interact with JDB without having an exception? I looked at also the "Fiasco Kernel Debugger Manual", however, I couldn't figure out how to exactly interact with JDB. Until now I could only interact with JDB in case of any exception occurrence. Thanks in advance. Sincerely, Cem Akpolat

3 3

Trimming uclibc in l4
by Masti Ramya Jayaram 18 Sep '14

18 Sep '14

Dear all, I would like to extract the minimal subset of uclibc required by bootstrap, sigma and their dependencies only. My idea is to have these modules use the smaller libc while retain the usual uclibc build for the other packages. I intend to go about this by copying uclibc (say into uclibc_min) and then modifying the contrib_files_all.lst and contrib_files_x86.lst. Is this the correct way to do this? Also, I extracted all the package dependencies of bootstrap and sigma from the control files (below) bootstrap: drivers_uart drivers_of libc l4util cxx_io drivers-frst:libc l4util: crtn libc crtn: l4sys l4sys: ldscripts l4util: crtn libc cxx: l4sys l4util ulibc: l4sys sigma0: crtn l4sys l4util libsigma0 cxx_io libsigma0: l4sys Is this the correct way to do this as well? Thanks, Ramya

3 6

Physical memory allocation to L4linux
by Masti Ramya Jayaram 16 Sep '14

16 Sep '14

Hi all, I start l4linux by specifying 64M on memory in its configuration file. I would like this memory to be a fixed range of physical memory and I was wondering if there is a way to do it. I understand from my earlier posts that the image of l4linux is in the same location everytime because the boot loader puts it there. I am not referring to this memory. I am concerned with the 64M that will potentially get allocated to l4linux. Thanks in advance, ramya

3 13

Question about dec_lock_cnt method in Context class
by Yuxin Ren 15 Sep '14

15 Sep '14

Hi All, I have a question about the dec_lock_cnt method in Context class under multiple processor. In its implementation, it checks if thread's home cpu is equal to current cpu. If not, it does not unset "_running_under_lock" variable, even if the _lock_cnt is 0. Why does it check if home cpu is equal to current cpu? Consider the following case: Thread A is running on core 1 holding a helping lock, but is preempted by another high priority thread. Thread B is running on core 2, and tries to take the same helping lock. It finds the lock is held by A, so it helps A to run by migrating A to core 2. When A releases its lock, finding the current cpu is not its home cpu, so it does not set "_running_under_lock" to false. Then A continues to run on core 1 and tries to take that lock again. But now it can never take the lock, as the implementation of helping lock. PRIVATE inline bool NO_INSTRUMENT Switch_lock::set_lock_owner(Context *o) { bool have_no_locks = o->_lock_cnt < 1; if (have_no_locks) { assert_kdb (current_cpu() == o->home_cpu()); for (;;) { if (EXPECT_FALSE(access_once(&o->_running_under_lock))) continue; if (EXPECT_TRUE(mp_cas(&o->_running_under_lock, Mword(false), Mword(true)))) break; } } ... } Do I misunderstand anything here? Thank you very much. Best, Yuxin

2 8