- l4-hackers - example.com Site

the syscall stack frame and Context::regs method
by watermirror 28 Apr '15

28 Apr '15

hi, i am reading the code about system call. after the sys_ipc_wrapper (in file kern/syscalls.cpp) is called.it get the thread's pointer by current_thread and then get the Syscall_frame. the code is like this: ... Thread *curr=current_thread(); Syscall_frame *f = curr->regs(); ... i can not understand the relationship between curr->regs() and the system call's stack frame.

2 1

about sysenter
by watermirror 28 Apr '15

28 Apr '15

hi ,in the file :l4re/pkg/l4sys/include/ARCH-x86/ipc-invoke.h, the comments says: We use ecx to store the user-level espand save eip onto the stack. but where could if find the corresponding code?

2 1

the first instruction in entry_sys_fast_ipc_c
by watermirror 28 Apr '15

28 Apr '15

hi, after entered the entry_sys_fast_ipc_c through sysenter, why the first instruction is 'pop %esp'?

2 1

How to solve the problem about multiple definition of clk_enable?
by Youren Shen 28 Apr '15

28 Apr '15

Hi, L4 hackers, To solve the problem about "undefined reference to `clk_get_rate'", I select COMMON_CLK in config L4_PLATFORM_GENERIC. But when I build the L4Linux this occurs: multiple definition of `clk_enable' l4linux/arch/l4/kernel/arch-arm/l4.c:63: first defined here So, Should I disable L4_CLK_NOOP in L4_PLATFORM_GENERIC? Thank you very much! Best Regards! Youren Shen

2 1

Re: How L4Linux Linked the header file?
by Youren Shen 28 Apr '15

28 Apr '15

Hi, Adam, Sorry to disturb you again. I find the reason of this problem is because the "config GENERIC_CLOCKEVENTS" is removed from file arch/arm/Kconfig after Kernel 3.4. As a result, the GENERIC_CLOCKEVENTS is not defined in the arch/l4/kconfig.arm. And the CONFIG_SMP depends on GENERIC_CLOCKEVENTS. So the CONFIG_GENERIC_CLOCKEVENTS and CONFIG_SMP will be unmask in build processes. Is there any way to get previous version of L4Linux? For instance, If I can get the source code of the L4Linux 3.4, it will be very helpful. Thank you very much for your help! On Sun, Apr 26, 2015 at 12:18 AM, Youren Shen <shenyouren(a)gmail.com> wrote: > Hi, Adam > I'm not so sure how to reply the mail in L4-hackers-list, So I CC to both > the List mail address and thread mail address. > > Thank you for your reply, and that really helps. > > I find that those error occurred because the CONFIG_SMP and > CONFIG_GENERIC_CLOCKEVENTS_BUILD is not set in .config. > > I try to config them by 'make menuconfig' but there is none entry about > CONFIG_SMP or CONFIG_GENERIC_CLOCKEVENTS_BUILD appeared in menuconfig. So > I change the .config file directly, but once I invoke "make", the .config > will be rewrite. > > Would you mind give me any guide about how to set them? > > Thank you very much. > >> >> >> Hi, >> >> On Sat Apr 18, 2015 at 22:26:15 +0800, Youren Shen wrote: >> > I'm trying to port L4Linux 3.14 to Linux Kernel 3.4. But when I complied >> > the kernel, this error happened: >> > >> /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/main.c: >> > In function 'cpu0_startup': >> > >> /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/main.c:2124:2: >> > error: implicit declaration of function 'set_my_cpu_offset' >> > [-Werror=implicit-function-declaration] >> > >> > I checked main.c file: >> > >> > #include <asm/unistd.h> >> > #include <asm/percpu.h> >> > #include <asm/generic/l4lib.h> >> > >> > So I add one percpu.h file with definition of 'set_my_cpu_offset' in >> > >> /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/include/asm/arch-arm/, >> > however this error still happened. >> > >> > Then I add the definition of this function to >> > >> /home/shenyouren/workspace/kernel/kernel_exynos5250/include/linux/percpu.h, >> > then the error message will not show. >> > >> > Also, discard the change in $(src)/include/linux/percpu.h, then I change >> > the main.c file to include absolute path of percpu.h. >> > >> > #include <asm/unistd.h> >> > #include >> > >> </home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/include/asm/arch-arm/percpu.h> >> > #include <asm/generic/l4lib.h> >> > >> > It passed the complied to. >> >> When you put a header file in arch/l4/include/asm/arch-arm/ including it >> via <asm/foo.h> should work. >> >> > So that is say, the #include <asm/percpu.h> is reference to >> > src/include/linux/percpu.h. That really strange. >> >> So, include <linux/percpu.h> in main.c right away? >> >> The L4 part of L4Linux you're looking at is for 3.14 and you apply it to >> Linux 3.4, please expect that things are different and need adaption. >> >> > I do not change the Makefile in arch/l4, and I really confused about >> this >> > error. >> > Now this problem occurs to me again, when complied the >> > arch/l4/kernel/timer.c file, this error happened: >> > >> > >> /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/timer.c:153:35: >> > warning: 'struct clock_event_device' declared inside parameter list >> > [enabled by default] >> > >> /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/timer.c:153:35: >> > warning: its scope is only this definition or declaration, which is >> > probably not what you want [enabled by default] >> > >> > but the definition of this to file should be defined in clockchips.h >> which >> > is include in timer.c: >> > >> > #include <linux/timex.h> >> > #include <linux/clockchips.h> >> > #include <linux/irq.h> >> > >> > Thank you for your reading, I just want to know what's the rule of >> L4Linux >> > header file linking, and solve this problem. >> > I'd be so appreciated if anyone can give me some help! >> >> There's a #ifdef CONFIG_GENERIC_CLOCKEVENTS_BUILD in clockchips.h. >> Not defined? >> >> >> >> Adam >> -- >> Adam adam(a)os.inf.tu-dresden.de >> Lackorzynski http://os.inf.tu-dresden.de/~adam/ >> >> -- > Best Regards. > Youren Shen. > -- Best Regards. Youren Shen.

2 1

Re: How L4Linux Linked the header file?
by Youren Shen 25 Apr '15

25 Apr '15

Hi, Adam I'm not so sure how to reply the mail in L4-hackers-list, So I CC to both the List mail address and thread mail address. Thank you for your reply, and that really helps. I find that those error occurred because the CONFIG_SMP and CONFIG_GENERIC_CLOCKEVENTS_BUILD is not set in .config. I try to config them by 'make menuconfig' but there is none entry about CONFIG_SMP or CONFIG_GENERIC_CLOCKEVENTS_BUILD appeared in menuconfig. So I change the .config file directly, but once I invoke "make", the .config will be rewrite. Would you mind give me any guide about how to set them? Thank you very much. > > > Hi, > > On Sat Apr 18, 2015 at 22:26:15 +0800, Youren Shen wrote: > > I'm trying to port L4Linux 3.14 to Linux Kernel 3.4. But when I complied > > the kernel, this error happened: > > > /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/main.c: > > In function 'cpu0_startup': > > > /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/main.c:2124:2: > > error: implicit declaration of function 'set_my_cpu_offset' > > [-Werror=implicit-function-declaration] > > > > I checked main.c file: > > > > #include <asm/unistd.h> > > #include <asm/percpu.h> > > #include <asm/generic/l4lib.h> > > > > So I add one percpu.h file with definition of 'set_my_cpu_offset' in > > > /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/include/asm/arch-arm/, > > however this error still happened. > > > > Then I add the definition of this function to > > > /home/shenyouren/workspace/kernel/kernel_exynos5250/include/linux/percpu.h, > > then the error message will not show. > > > > Also, discard the change in $(src)/include/linux/percpu.h, then I change > > the main.c file to include absolute path of percpu.h. > > > > #include <asm/unistd.h> > > #include > > > </home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/include/asm/arch-arm/percpu.h> > > #include <asm/generic/l4lib.h> > > > > It passed the complied to. > > When you put a header file in arch/l4/include/asm/arch-arm/ including it > via <asm/foo.h> should work. > > > So that is say, the #include <asm/percpu.h> is reference to > > src/include/linux/percpu.h. That really strange. > > So, include <linux/percpu.h> in main.c right away? > > The L4 part of L4Linux you're looking at is for 3.14 and you apply it to > Linux 3.4, please expect that things are different and need adaption. > > > I do not change the Makefile in arch/l4, and I really confused about this > > error. > > Now this problem occurs to me again, when complied the > > arch/l4/kernel/timer.c file, this error happened: > > > > > /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/timer.c:153:35: > > warning: 'struct clock_event_device' declared inside parameter list > > [enabled by default] > > > /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/timer.c:153:35: > > warning: its scope is only this definition or declaration, which is > > probably not what you want [enabled by default] > > > > but the definition of this to file should be defined in clockchips.h > which > > is include in timer.c: > > > > #include <linux/timex.h> > > #include <linux/clockchips.h> > > #include <linux/irq.h> > > > > Thank you for your reading, I just want to know what's the rule of > L4Linux > > header file linking, and solve this problem. > > I'd be so appreciated if anyone can give me some help! > > There's a #ifdef CONFIG_GENERIC_CLOCKEVENTS_BUILD in clockchips.h. > Not defined? > > > > Adam > -- > Adam adam(a)os.inf.tu-dresden.de > Lackorzynski http://os.inf.tu-dresden.de/~adam/ > > -- Best Regards. Youren Shen.

1 0

(no subject)
by deng.chao1＠zte.com.cn 24 Apr '15

24 Apr '15

Hi: I am reading the code of fiasco kernel.I have a question about the function "Context::schedule". At line 894 in kernel/fiasco/src/kern/context.cpp： "while (EXPECT_FALSE(current_cpu != access_once(&_home_cpu)))" As I know, the variable "_home_cpu" indicates the CPU number which the thread that invokes the “schedue ” as a member function is running on。And the variable “current cpu” indicates the CPU number now this code fragment is running on。 The condition test tells me that it seems the class Context's member function "schedule" will run on a different cpu some time . If it does so, some extra work will need to be done in the "schedule" function. I have searched all of the fiasco kernel sources, but I did not find this special context. So would anybody please tell me whether there has a situation that the "schedule" function will run on different CPU， why does it happen？ Thanks！

2 1

where is the call_bootstrap?
by 水镜 24 Apr '15

24 Apr '15

Hi, in the file main-ia32-32.cpp, there's a line(line 60): call call_bootstrap where could i find the implemention of the call_bootstrap?

2 1

Static_object in fiasco kernel
by 水镜 23 Apr '15

23 Apr '15

hi, what is the Static_object in fiasco kernel used for?

2 1

How L4Linux Linked the header file?
by Youren Shen 23 Apr '15

23 Apr '15

Hi, I'm trying to port L4Linux 3.14 to Linux Kernel 3.4. But when I complied the kernel, this error happened: /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/main.c: In function 'cpu0_startup': /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/main.c:2124:2: error: implicit declaration of function 'set_my_cpu_offset' [-Werror=implicit-function-declaration] I checked main.c file: #include <asm/unistd.h> #include <asm/percpu.h> #include <asm/generic/l4lib.h> So I add one percpu.h file with definition of 'set_my_cpu_offset' in /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/include/asm/arch-arm/, however this error still happened. Then I add the definition of this function to /home/shenyouren/workspace/kernel/kernel_exynos5250/include/linux/percpu.h, then the error message will not show. Also, discard the change in $(src)/include/linux/percpu.h, then I change the main.c file to include absolute path of percpu.h. #include <asm/unistd.h> #include </home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/include/asm/arch-arm/percpu.h> #include <asm/generic/l4lib.h> It passed the complied to. So that is say, the #include <asm/percpu.h> is reference to src/include/linux/percpu.h. That really strange. I do not change the Makefile in arch/l4, and I really confused about this error. Now this problem occurs to me again, when complied the arch/l4/kernel/timer.c file, this error happened: /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/timer.c:153:35: warning: 'struct clock_event_device' declared inside parameter list [enabled by default] /home/shenyouren/workspace/kernel/kernel_exynos5250/arch/l4/kernel/timer.c:153:35: warning: its scope is only this definition or declaration, which is probably not what you want [enabled by default] but the definition of this to file should be defined in clockchips.h which is include in timer.c: #include <linux/timex.h> #include <linux/clockchips.h> #include <linux/irq.h> Thank you for your reading, I just want to know what's the rule of L4Linux header file linking, and solve this problem. I'd be so appreciated if anyone can give me some help! -- Best Regards. Youren Shen.

2 1

How to compile the fiasco kernel test directory?
by Liu Xiaodong 23 Apr '15

23 Apr '15

Hi,I got a tar ball(l4re-snapshot-2014092821.tar.bz2) from fiasco website. I see a directory for test under the fiasco kernel source path: /l4re-snapshot-2014092821/src/kernel/fiasco/src/test/. But I can't compile them in the normal way, such as that you discrbed in the website http://os.inf.tu-dresden.de/fiasco/build.html. I really want to do some kernel tests on fiaso. If you can tell me how to compile the fiasco kernel test directory, it will be very helpful, and I will appreciate it. By the way, the test source code seems to not be included completely. Such as /l4re-snapshot-2014092821/src/kernel/fiasco/src/test/space/space.t.cc #include <iostream> #include <iomanip> #include <cassert> using namespace std; #include <flux/page.h> #include "space.h" int main() { // // Make task 4 the chief of task 5 // assert (space_index_t (5).lookup () == 0); assert (space_index_t (5).chief () == 0); assert (space_index_t (5).set_chief (space_index_t(0), space_index_t(4))); assert (space_index_t (5).lookup () == 0); assert (space_index_t (5).chief () == 4); ...... } I can't find headfile "space.h" in the tar ball, and there is no define about "space_index_t" too. can you tell me why? thanks again! -------------------------------------------------------- ZTE Information Security Notice: The information contained in this mail (and any attachment transmitted herewith) is privileged and confidential and is intended for the exclusive use of the addressee(s). If you are not an intended recipient, any disclosure, reproduction, distribution or other dissemination or use of the information contained is strictly prohibited. If you have received this mail in error, please delete it and notify us immediately.

2 3

How to do quickly in order to starting fiasco with a console shell?
by zhou.li9＠zte.com.cn 23 Apr '15

23 Apr '15

How to do quickly in order to starting fiasco with a console shell? After copying src/l4/conf/examples/x86-fb.list to src/l4/conf/modules.list,I use QEMU to boot,the boot process appears to be normal: ....... fbdrv | Starting server loop mag | mapped frame buffer at 0x200000 mag | View::Info: mag | flags: 0 mag | size: 1280x1024 pos: 0, 0 mag | bytes_per_line: 2560 mag | buffer_offset: 0 mag | RGBA(2): 5(11):6(5):5(0):0(0) mag | memory 200000-600000 mag | Plugin: Mag_client service started mag | Plugin: Frame-buffer service started mag | L4INPUT native mode activated mag | L4INPUT: !!! W A R N I N G !!! mag | L4INPUT: Please, do not use Fiasco's "-esc" with L4INPUT. mag | L4INPUT: !!! W A R N I N G !!! mag | serio: i8042 AUX port at 0x60,0x64 irq 12 mag | serio: i8042 KBD port at 0x60,0x64 irq 1 mag | input: ImExPS/2 Generic Explorer Mouse on isa0060/serio1 mag | input: AT Translated Set 2 keyboard on isa0060/serio0 mag | connect "ImExPS/2 Generic Explorer Mouse", isa0060/serio1/input0 mag | connect "AT Translated Set 2 keyboard", isa0060/serio0/input0 spectrum| x:640 y:480 bit/pixel:16 bytes/line:1280 But How can I input some shell command? -------------------------------------------------------- ZTE Information Security Notice: The information contained in this mail (and any attachment transmitted herewith) is privileged and confidential and is intended for the exclusive use of the addressee(s). If you are not an intended recipient, any disclosure, reproduction, distribution or other dissemination or use of the information contained is strictly prohibited. If you have received this mail in error, please delete it and notify us immediately.

3 4

Re: Using a second UART to control a serial device
by Pflaum, Clemens 20 Apr '15

20 Apr '15

Thank you for your fast response, I've adapted your code to my needs (using the uart_pxa driver in drivers-frst). It compiles and starts up fine on the BananaPi but sadly I'm not getting any response on the uart3 pins... It seems to have a problem in the l4\pkg\drivers-frst\uart\src\uart_pxa.cc when calling the Uart_16550::startup(regs). I get a few wierd boxlike characters on the console and then "failed to startup uart regs". This is my init() : #define IRQ_NUM 86 #define UART_BASE 0x01C28C00 . . . bool Maestro_server::init() { l4_addr_t virt_base = 0; if (l4io_request_iomem((l4_addr_t)UART_BASE, 0x1000, L4IO_MEM_NONCACHED, &virt_base)) { printf("maestro-drv: request io-memory from l4io failed.\n"); return false; } printf("maestro-drv: virtual base at:%lx\n", virt_base); L4::Io_register_block_mmio *regs = new L4::Io_register_block_mmio(virt_base); _uart = new (malloc(sizeof(L4::Uart_16550))) L4::Uart_16550((unsigned long) 115200); if(!(_uart->startup(regs))){ printf("failed to startup uart regs!"); } if(!(_uart->change_mode(0x00 | 0x03 | 0x00, 115200))){ printf("failed to change uart mode!"); } the rest of it is equal to the serial-drv example. Any idea what i couldve done wrong? regard, Clemens

1 0

Using multiple NICs within one application
by Stark, Josef 19 Apr '15

19 Apr '15

Hello, is it possible to use more than one network card in one L4Re-application and how would I do that? I would guess from the example lwip lua-script that I had to create a separate shm-area and/or communication channel (for ankh-sessions) for each device that my app wants to access, but I don't know how I would supply these additional channels to ankh and if this would be enough to enable multi-NIC networking. Any help would be appreciated. Regards, Josef

1 0

Re: Using a second UART to control a serial device
by Reinier Millo Sánchez 18 Apr '15

18 Apr '15

Hi Clemens I have been testing something similar in other board (YSE5250-Exynos5). First you must guarantee that the UART device driver is compatible with the hardware or you must write an specific device driver. I have tried to make a GPS device driver using the S3C2410 UART based on the serial-drv package. To connect the uart on the user land with my server i have configured the HW.Device in a vbus with the IO server using this configuration: -- File: yse5250.devs local Hw = Io.Hw local Res = Io.Res Io.hw_add_devices { UART3 = Hw.Device { hid = "UART3"; Res.irq(86); Res.mmio(0x12C30000, 0x12C3ffff); } } To create the vbus, i have created the file yse5250.io: -- File: yse5250.io Io.add_vbusses { -- Create a virtual bus for a client and give access to UART3 uart03 = Io.Vi.System_bus(function () dev = wrap(Io.system_bus():match("UART3")); end); } To run my server using the UART3 on the vbus have used this configuration -- gps-drv.cfg -- Run GPS Driver for YSE5250 (Exynos 5250) require("L4"); -- Shortcuts local ld = L4.default_loader; local vbus_uart03 = ld:new_channel(); local platform_ctl = ld:new_channel(); local gps_srv = ld:new_channel(); -- Start the IO Server ld:start( { caps = { sigma0 = L4.cast(L4.Proto.Factory, L4.Env.sigma0):create(L4.Proto.Sigma0); icu = L4.Env.icu; platform_ctl = platform_ctl:svr(); uart03 = vbus_uart03:svr(); }, log = { "IO Server", "green" }, l4re_dbg = L4.Dbg.Warn, }, "rom/io rom/yse5250.devs rom/yse5250.io"); -- Start the GPS Driver Server conected to UART3 ld:start( { caps = { gps_srv = gps_srv:svr(); vbus = vbus_uart03; }, log = {"GPS Driver","blue"} }, "rom/gps_drv"); The server code is very similar to: /pkg/serial-drv/server/src/main.cc But in my case i have adjusted the UART to use the S3C2410 uart. Best regards On 04/18/2015 06:00 AM, l4-hackers-request(a)os.inf.tu-dresden.de wrote: > Message: 1 > Date: Fri, 17 Apr 2015 12:03:46 +0200 > From: "Pflaum, Clemens"<clemens.pflaum(a)mytum.de> > To:l4-hackers@os.inf.tu-dresden.de > Subject: Using a second UART to control a serial device > Message-ID:<4a79357eec3b3359e8b09e16a3fe0994(a)roundcube.lrz.de> > Content-Type: text/plain; charset=US-ASCII; format=flowed > > Hi L4 Hackers, > Im currently running Fiasco.OC with L4Re on my BananaPi. I'm trying to > control a serial device (Pololu Servo Controller) by sending a small > series of Bytes over a second UART ( not UART 0 since im getting the > console from there ). Via documentation I've found out that UART 3 is at > 0x01C28C00 and is a U6_16550A, so I'm hoping the driver in > /pkg/drivers-frst/uart/src/uart_pxa.cc will be compatible with that. > Where and how would I have to go about initiating my UART 3? By writing > a new platform for BPi in /pkg/bootstrap/server/src/platform? The > function set_stdio_uart(&_uart); made me think its only the "console" > UART. Or do I have to instantiate some kind of Hw.Device like the gpio > in /pkg/io/config/arm-omap3.devs and put it on a new vBus like in > /pkg/io/doc/example.io ? > > Or am I thinking way too complicated and it is as simple as editing the > /pkg/serial-drv/server/src/main.cc : > > bool > Serial_drv::init() > { > int irq_num = 37; > // How can i find out what this number refers to/do i have to > change it at all? > l4_addr_t phys_base = 0x1000a000; > // change to 0x01C28C00 > #if 0 > int irq_num = 74; > l4_addr_t phys_base = 0x49020000; > #endif > l4_addr_t virt_base = 0; > > if (l4io_request_iomem(phys_base, 0x1000, L4IO_MEM_NONCACHED, > &virt_base)) > { > printf("serial-drv: request io-memory from l4io failed.\n"); > return false; > } > printf("serial-drv: virtual base at:%lx\n", virt_base); > > L4::Io_register_block_mmio *regs = new > L4::Io_register_block_mmio(virt_base); > _uart = new (malloc(sizeof(L4::Uart_pl011))) L4::Uart_pl011(24019200); > //change to L4::Uart_pxa(what does this number mean?) > _uart->startup(regs); > > > Thanks for any answers in advance. > > Regards, > Clemensc...(a)mytum.de -- Lic. Reinier Millo Sánchez Centro de Estudios de Informática Universidad Central "Marta Abreu" de Las Villas Carretera a Camajuaní Km 5 1/2 Santa Clara, Villa Clara, Cuba CP 54830 "antes de discutir ... respira; antes de hablar ... escucha; antes de escribir ... piensa; antes de herir ... siente; antes de rendirte ... intenta; antes de morir ... vive"

1 0

Using a second UART to control a serial device
by Pflaum, Clemens 17 Apr '15

17 Apr '15

Hi L4 Hackers, Im currently running Fiasco.OC with L4Re on my BananaPi. I'm trying to control a serial device (Pololu Servo Controller) by sending a small series of Bytes over a second UART ( not UART 0 since im getting the console from there ). Via documentation I've found out that UART 3 is at 0x01C28C00 and is a U6_16550A, so I'm hoping the driver in /pkg/drivers-frst/uart/src/uart_pxa.cc will be compatible with that. Where and how would I have to go about initiating my UART 3? By writing a new platform for BPi in /pkg/bootstrap/server/src/platform? The function set_stdio_uart(&_uart); made me think its only the "console" UART. Or do I have to instantiate some kind of Hw.Device like the gpio in /pkg/io/config/arm-omap3.devs and put it on a new vBus like in /pkg/io/doc/example.io ? Or am I thinking way too complicated and it is as simple as editing the /pkg/serial-drv/server/src/main.cc : bool Serial_drv::init() { int irq_num = 37; // How can i find out what this number refers to/do i have to change it at all? l4_addr_t phys_base = 0x1000a000; // change to 0x01C28C00 #if 0 int irq_num = 74; l4_addr_t phys_base = 0x49020000; #endif l4_addr_t virt_base = 0; if (l4io_request_iomem(phys_base, 0x1000, L4IO_MEM_NONCACHED, &virt_base)) { printf("serial-drv: request io-memory from l4io failed.\n"); return false; } printf("serial-drv: virtual base at:%lx\n", virt_base); L4::Io_register_block_mmio *regs = new L4::Io_register_block_mmio(virt_base); _uart = new (malloc(sizeof(L4::Uart_pl011))) L4::Uart_pl011(24019200); //change to L4::Uart_pxa(what does this number mean?) _uart->startup(regs); Thanks for any answers in advance. Regards, Clemens c...(a)mytum.de

1 0

the kernel's output
by 水镜 17 Apr '15

17 Apr '15

hi , there're a lot of printf in fiasco kernel, but i can not seen any output on the screen. how could i turn these things on?

2 1

the ldscripts of L4Re
by 水镜 16 Apr '15

16 Apr '15

hi, the L4Re has three ldscripts: main_dyn.ld main.rel.ld main.stat.ld what are the differences?

2 1

How to track execution process of fiasco user mode applications？
by zhou.li9＠zte.com.cn 16 Apr '15

16 Apr '15

I found that fiasco does not support GDB, in order to track execution process of fiasco user mode applications,what method can be used? -------------------------------------------------------- ZTE Information Security Notice: The information contained in this mail (and any attachment transmitted herewith) is privileged and confidential and is intended for the exclusive use of the addressee(s). If you are not an intended recipient, any disclosure, reproduction, distribution or other dissemination or use of the information contained is strictly prohibited. If you have received this mail in error, please delete it and notify us immediately.

2 1

mmu in bootstrap
by 水镜 16 Apr '15

16 Apr '15

hi, i want to know, whether the bootstrap in l4re(x86) will initialize the MMU, setup pagetable? i am trying to debug the fiasco with gdb and qemu. when i attach to the qemu before system startup, i can check memory like 0x1100000.after the system startup, the gdb tell me can not access memory if i check 0x1100000 again.

2 1

IPC stream magic
by teclis High Elf 11 Apr '15

11 Apr '15

I have a complex nested structure that looks something like this: typedef struct struct_A { int i; struct struct_B[5]; int j; } using an IPC stream I do: struct_A a; ios.put(a); and then in the server process I do: struct_A b; ios.get(b); ( ios is a L4::Ipc::Iostream ) and magically my complex nested structure reappears with all the data intact in the server process running in another address space. As the IPC Streams are not well documented would someone be able to explain to me how my complex nested structure got automatically marshalled in the client and unmarshalled in the server process?

2 1

Re: ARM with Qemu
by Nicolas VARONA 09 Apr '15

09 Apr '15

Hi, Thanks for your answer and your help. >Could also be the uart issue? But entering jdb always works for me. Even if I can't give inputs to the bare metal application on Qemu, I can also enter into the JDB. Is that mean that all interruptions are well handled? So the uart issue seems more probable. >Which kind of panic? I firstly tested my L4linux on Qemu for the Zinq board and the boot finishes well (with the problem of inputs that I told you before). Nonetheless, with the same image on the board, I have this message: [...] Freeing unused kernel memory: 116K (0230f000 - 0232c000) potentially unexpected fatal signal 11. CPU: 0 PID: 1 Comm: init Not tainted 3.16.0-l4 #152 task: 05c23ac0 ti: 05c26000 task.ti: 05c26000 PC is at 0x0 LR is at 0x0 pc : [<00000000>] lr : [<00000000>] psr: 60000010 sp : bf73cef0 ip : 00000000 fp : 00000000 r10: 00000000 r9 : 00000000 r8 : 00000000 r7 : 00000000 r6 : 00000000 r5 : 00000000 r4 : 00000000 r3 : 0000c50c r2 : 00000000 r1 : 00000000 r0 : 00000000 vcpu: b3000c00 vcpu-state: 00000001 Flags: nZCv IRQs on FIQs on Mode USER_32 ISA ARM Segment user CPU: 0 PID: 1 Comm: init Not tainted 3.16.0-l4 #152 [<0200c9c8>] (unwind_backtrace) from [<020055e8>] (show_stack+0x10/0x14) [<020055e8>] (show_stack) from [<0202844c>] (get_signal_to_deliver+0x204/0x490) [<0202844c>] (get_signal_to_deliver) from [<02006be8>] (do_signal+0x114/0x448) [<02006be8>] (do_signal) from [<0200a210>] (l4x_vcpu_entry_c+0xa94/0x1da8) [<0200a210>] (l4x_vcpu_entry_c) from [<00000000>] ( (null)) Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b CPU: 0 PID: 1 Comm: init Not tainted 3.16.0-l4 #152 [<0200c9c8>] (unwind_backtrace) from [<020055e8>] (show_stack+0x10/0x14) [<020055e8>] (show_stack) from [<022441a4>] (panic+0x7c/0x1e0) [<022441a4>] (panic) from [<0201e6a8>] (do_exit+0x708/0x7a0) [<0201e6a8>] (do_exit) from [<0201e80c>] (do_group_exit+0x3c/0x9c) [<0201e80c>] (do_group_exit) from [<02028390>] (get_signal_to_deliver+0x148/0x490) [<02028390>] (get_signal_to_deliver) from [<02006be8>] (do_signal+0x114/0x448) [<02006be8>] (do_signal) from [<0200a210>] (l4x_vcpu_entry_c+0xa94/0x1da8) [<0200a210>] (l4x_vcpu_entry_c) from [<00000000>] ( (null)) ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b panic: going to sleep forever, bye l4linux | panic: going to sleep forever, bye Thanks in advance for your help, Nicolas >Hi, > >On Mon Mar 02, 2015 at 15:10:04 +0100, Nicolas VARONA wrote: >> I am trying to set up a simulator with Qemu and l4linux ARM to >> simulate the Zynq board. I have successfully compiled Fiasco and L4 >> with the tool chain "arm-linux-gnueabi-gcc". I succeeded to launch >> l4linux but I have no input on l4linux when it's running on Qemu. >> Nevertheless, the fiasco jdb can be invoked. After reading the mailing >> list I changed the hello world program to take inputs and tried a bare >> metal application with l4re with the same result (I didn't modify the >module list). >> >> So, I tried to do the same thing with the board and I have inputs with >> the l4 bare metal application > >So I looked a bit at this and it seems to me that there's something strange >with uart rx in Qemu, or at least which only shows up in Qemu. >I need to investigate that further. > >> and a kernel panic for the l4linux. > >Which kind of panic? > >> I boot >> with the uboot image on it. Nonetheless, using uImage for qemu doesn't >> change anything: >> >> ./qemu-system-arm -M arm-generic-fdt-plnx -m 512M -kernel >> /path/bootstrap.uimage -dtb /path/devicetree.dtb -serial mon:stdio >> >> I tried to compile the u-boot from Xilinx and Qemu freezes when I >> launch the l4 application. > >Could also be the uart issue? But entering jdb always works for me. > >> To be sure I have done things right, I also tried the modified hello >> world program on x86 intel and it works well. Then I launched a >> Petalinux with Qemu and a uImage with uboot and it works. >> >> My Qemu is from Xilinx: QEMU emulator version 2.0.50 > >Oh, good to know. I tried the normal one.

2 12

Question about L4/Fiasco running the kernel with interrupts disable
by deng.chao1＠zte.com.cn 09 Apr '15

09 Apr '15

Hi all， I have read the wikipedia website of L4 kernel(http://en.wikipedia.org/wiki/L4_microkernel_family), it says that L4/Fiasco was designed "fully preemptible to achieve a low interrupt latency " at first, however "the complexities of a fully preemptible design resulted in later versions of Fiasco to return to the traditional L4 approach of running the kernel with interrupts disabled ". I am very interested in the kernel's interrupt latency. I have check the latest versions of Fiasco kernel in running, and it keeps interrupts disabled indeed. So I wonder whether there was any early version of L4/Fiasco which supported fully preemptible? Or was there any article talked about this ? Thanks！

2 1

Question about L4/Fiasco running the kernel with interrupts disable
by cui.yunfeng＠zte.com.cn 08 Apr '15

08 Apr '15

Hi all， I have read the wikipedia website of L4 kernel(http://en.wikipedia.org/wiki/L4_microkernel_family), it says that L4/Fiasco was designed "fully preemptible to achieve a low interrupt latency " at first, however "the complexities of a fully preemptible design resulted in later versions of Fiasco to return to the traditional L4 approach of running the kernel with interrupts disabled ". I am very interested in the kernel's interrupt latency. I have check the latest versions of Fiasco kernel in running, and it keeps interrupts disabled indeed. So I wonder whether there was any early version of L4/Fiasco which supported fully preemptible? Or was there any article talked about this ? Thanks！

1 0

about ".initcall.text" and ".initcall.data"
by 水镜 07 Apr '15

07 Apr '15

hi ,there are ".initcall.data" and ".initcall.data" in the kernel.ia32.ld, what are these two sections for?

2 1