users June 2016

users@lists.genode.org

28 participants
30 discussions

16.05 custom kernel call fails to get session capability.
by Bob Stewart 28 Jun '16

28 Jun '16

Hi, I'm moving from release 15.11 to 16.05 on base-hw and am having a "service denied" issue when trying to get a connection to a custom service I created to handle pin muxing on TI Sitara processors. In this family of processors, access to a "Control Module" is need to enable subsystems, set up clock frequencies, and set the mode for gpio pins which have multiple functions dependent upon the mode they are set to. Write access to the Control Module is only allowed in a privileged mode so I had to create custom kernel calls to access the appropriate SoC registers. I did this in release 14.05 by adding the code for unrestricted kernel calls to base-hw and it has been working without issue through various releases until I stared to move to 16.05. The log output I'm getting is: Starting kernel ... kernel initialized Genode 16.05 <local changes> int main(): --- create local services --- int main(): --- start init --- int main(): transferred 503 MB to init int main(): --- init created, waiting for exit condition --- [init] Could not open ROM session for module "ld.lib.so" [init -> ctrl_module] ---437x Control Module Server : Setting Pin Muxes--- [init -> ctrl_module] pin mux capability [init -> timer] Maximum timeout 1717986000 us [init -> timer] --- am437x timer driver --- [init -> 437x_platform_client] --- 437x platform being configured -- [init -> ctrl_module] Uncaught exception of type 'Genode::Parent::Service_denied' [init -> ctrl_module] Warning: abort called - thread: ep [init] virtual void Genode::Child_policy::exit(int): child "ctrl_module" exited w ith exit value 1 [init -> 437x_platform_client] Enabled dmtimer 3 freq = 1 [init -> 437x_platform_client] Enabled dmtimer 4 freq = 1 ....(other messages unrelated to the issue follow) The uncaught exception is being generated in my "pinmux_session" connection code when a capability for a session is requested, as shown the following in the following code snippet from pinmux_connection.h namespace Genode { struct Pinmux_connection; } struct Genode::Pinmux_connection : Connection<Pinmux_session>, Pinmux_session_client { /** * Issue session request * * \noapi */ Capability<Pinmux_session> _session(Parent &parent, char const *label, long priority, unsigned long affinity) { Genode::printf("pin mux capability\n"); return***/session(parent,/**/ /**/ "priority=0x%lx, affinity=0x%lx, ram_quota=16K, label=\"%s\"",/**/ /**/ priority, affinity, label); (exception occurs here)/* } /** * Constructor * * \param label initial session label * \param priority designated priority of the Pinmux * \param affinity which physical CPU the VM should run on top of */ Pinmux_connection(Genode::Env &env, const char *label = "", long priority = Cpu_session::DEFAULT_PRIORITY, unsigned long affinity = 0) : Connection<Pinmux_session>( env, _session(env.parent(), label, priority, affinity)), Pinmux_session_client(cap()) { } /** * Constructor * * \noapi * \deprecated Use the constructor with 'Env &' as first * argument instead */ Pinmux_connection(const char *label = "", long priority = Cpu_session::DEFAULT_PRIORITY, unsigned long affinity = 0) : Connection<Pinmux_session>(_session(*env()->parent(), label, priority, affinity)), Pinmux_session_client(cap()) { Genode::printf("ctrl mod connection\n");} }; This code differs from the 15.11 code in that the request for a session now requires a "parent" argument in 16.05. Looking at similar kernel call codes like the vm calls, my changes to my custom calls for 16.05 appear to be correct. I can see in the kernel code (parent.h) where the exception is occurring but I have no clue as to why it occurs. Any hints would be greatly appreciated. Thanks, Bob Stewart

3 3

Recursive Nitpicker
by Michael Ullrich 28 Jun '16

28 Jun '16

Hallo, I'm currently trying to modularize a Qt-based ChatClient. The basic idea is to isolate the parts of some guy-components from each other. I want to use Nitpicker for this. In generally the Client itself should run in a trusted nitpicker session. Now I want to start another nitpicker-sessions inside the quote-client. Meaning by that, I want for example a trusted input-field or a trusted label, isolated from the remaining qt-app. So isolate several graphical components from each other. Is it in generally a good idea, to use nitpicker in that way? Has someone tried something similar, or is there an reference implementation for that? With kind regards, Michael Ullrich

3 2

Benchmarking Genode TrustZone
by Tiago Brito 28 Jun '16

28 Jun '16

Hi, I want to benchmark the execution of a function running in the secure world of the TZ_VMM scenario in the i.MX53 QSB. I have added a syscall to Linux which allows me to trigger a world switch from a user program running in Linux. In this program I have a function which allocates a buffer and processes it (each buffer position is changed in some way). This same function is coded inside TZ_VMM. This is what I'm testing: 1. Inside my user program in Linux I use gettimeofday before and after the execution of the function in order to get the amount of milliseconds in between. This is my NW test. 2. Inside my user program in Linux I use gettimeofday to get the start time, then I execute the syscall which in turn does a world switch. Then the function is executed inside the SW and it returns to the user program inside Linux. After this I call another gettimeofday in order to get the amount of milliseconds of execution. The problem is that test 1 is giving me about 90 ms of real time execution, but test 2 gives me about 40 ms. I suspect it might be a problem with Linux virtualization in the TZ_VMM example, which may be causing a drift in Linux's clock once it loses control to the SW. What I mean is, when there isn't a syscall triggering the SMC, Linux can count time just fine, but once the control is lost to the secure world the clock inside Linux becomes inconsistent and doesn't count time while the secure world is executing. Is this right? Since I really need to benchmark a scenario similar to this I think that the best alternative is to offload the time functionality to Genode (SW). I create another syscall which is responsible for starting a timer inside Genode, then I call the SMC syscall which processes the buffer in the SW, then I call the time syscall again and check the difference. When I want to benchmark the NW function I follow the same steps as before. Will this work as intended? I'm thinking that this alternative may suffer from the same problem as before if Genode's time clock becomes inconsistent whenever Linux is being executed in NW. Do you know any other way to benchmark a world switch + processing + world switch scenario? Is there any timer I can execute inside TZ_VMM? Thanks in advance, Tiago

4 9

l4linux fails on fiasco.oc 64bit
by john stokes 27 Jun '16

27 Jun '16

i tried to run l4linux over foc_x86_32 platefrom it ran well. now after building the build directory for foc_x86_64 i am agian trying to run the run script of l4linux from the build directory then i am getting this error I have changed the <spec>.conf and of the ports-foc and other tarhget.mk for my foc_x86_64 but it is not working the error is : the same thing works fine for foc_x86_32 Program drivers/nic/nic_drv Program drivers/platform/spec/x86/platform_drv Library alarm Library timer Program drivers/timer/timer Program drivers/uart/kdb/spec/foc/kdb_uart_drv Program drivers/uart/spec/i8250/uart_drv Program init/init Library l4lx COMPILE l4_util_util.o In file included from /home/rishabh/genode-16.05/repos/ports-foc/src/lib/l4lx/l4_util_util.cc:18:0: /home/rishabh/genode-16.05/repos/ports-foc/include/l4/util/util.h: In function ‘void Fiasco::l4_touch_ro(const void*, unsigned int)’: /home/rishabh/genode-16.05/repos/ports-foc/include/l4/util/util.h:31:34: error: cast from ‘const void*’ to ‘unsigned int’ loses precision [-fpermissive] bptr = (const char*)(((unsigned)addr) & L4_PAGEMASK); ^ /home/rishabh/genode-16.05/repos/ports-foc/include/l4/util/util.h:32:34: error: cast from ‘const void*’ to ‘unsigned int’ loses precision [-fpermissive] eptr = (const char*)(((unsigned)addr+size-1) & L4_PAGEMASK); ^ /home/rishabh/genode-16.05/repos/ports-foc/include/l4/util/util.h: In function ‘void Fiasco::l4_touch_rw(const void*, unsigned int)’: /home/rishabh/genode-16.05/repos/ports-foc/include/l4/util/util.h:45:28: error: cast from ‘const void*’ to ‘unsigned int’ loses precision [-fpermissive] bptr = (char*)(((unsigned)addr) & L4_PAGEMASK); ^ /home/rishabh/genode-16.05/repos/ports-foc/include/l4/util/util.h:46:34: error: cast from ‘const void*’ to ‘unsigned int’ loses precision [-fpermissive] eptr = (const char*)(((unsigned)addr+size-1) & L4_PAGEMASK); ^ make[3]: *** [l4_util_util.o] Error 1 make[2]: *** [l4lx.lib] Error 2 make[1]: *** [gen_deps_and_build_targets] Error 2 make[1]: Leaving directory `/home/rishabh/genode-16.05/build/foc_x86_64' Error: Genode build failed make: *** [run/l4linux] Error 252

3 6

Genode Compilation Flags - Optimization
by Tiago Brito 23 Jun '16

23 Jun '16

Hi, I've been trying to benchmark two similar programs, one running in the NW and another running in the SW of a i.MX53 QSB. I've adapted the tz_vmm example to run the SW program and the NW program is running on top of Linux. I compiled my NW program with the -O3 optimization flag and I'm wondering if the code in tz_vmm can also be compiled with this flag in order to have the same optimization flag in both scenarios. If it's possible, what do I have to change in order to achieve this? Thanks in advance! Tiago

3 2

Feather: A very lightweight web service for Genode platforms.
by Guido Witmond 22 Jun '16

22 Jun '16

Fellow Genodians, At Heeselicht, I've been working on my web server. With all your help, I got it working. It's single threaded, slow as molassis and - after fixing some of my null-pointer errors - now runs for more than 5 seconds. Actually, it's quite stable. It has been running for a day and night without hiccups. However, it's slow, I get up to 1500 kB/s when serving a 1.4 MB file. Short files (300 bytes) get served at 5.19 kB/s. So there is plenty of room for improvement given the specs of the hardware. The site: - http://eccentric-authentication.org Source code: - github.com/gwitmond/feather Platform: - Fiasco.OC with kernel debugger enabled - Genode 16.2 framework. Hardware: - HP ProLiant DL120 G6 (2011) - Pentium G6950 @ 2.80GHz processor - 4GB ECC RAM - Network: on board Broadcom Nextreme BCM5723 with a 100Mb/s uplink - No disk used, the site lives in a tar file loaded by grub at boot It was a pleasure creating this, I hope someone might find it useful too. With regards, Guido Witmond.

5 8

libport usage and extra inclusion line
by Menno Valkema 22 Jun '16

22 Jun '16

Hi Everyone, We're using several libraries in our product, and most of them just work by adding a line like: LIBS = libc However in some cases this is not sufficient. For example lwip requires an additional include directive: INC_DIR += $(REP_DIR)/src/lib/lwip/include (line from repos/libports/src/test/lwip/http_srv_static/target.mk) Leaving out this line will result in a compilation error of a local include file. Looking at various library ports (mbedtls, polarssl, curl, and probably more) we see similar behavior. Usually some library configuration header file should be added manually to the include path to make things work. Adding this extra line for a default library configuration feels redundant, so we're wondering we might do something we're not supposed to do, or should be done in a different way? Thanks, Menno -- Cyber Security Labs B.V. | Gooimeer 6-31 | 1411 DD Naarden | The Netherlands

4 5

booting genode component along with linux on x86
by RISHABH PANDEY 22 Jun '16

22 Jun '16

can someone help me , by explaining the steps required for booting the genode component along with linux on x86 base plateform(i ma not able to find iso file)???

1 0

create_builddir error
by RISHABH PANDEY 21 Jun '16

21 Jun '16

when using command create_builddir it shows command not found. how to resolve this error

2 1

"focus" module
by Matt Brown 17 Jun '16

17 Jun '16

Hello Genode experts, I am still pretty new to Genode so sorry if this is another trivial question. I am trying to run a script that runs a terminal session in a window. At runtime I'm getting the error 'Could not open ROM session for module "focus"'. I understand that this means that the "focus" module was not loaded into memory at boot but I cannot figure out was "focus" is. I have tried searching for focus through the Genode directories but no luck, (with the notable exception of qt libraries). I also tried adding focus to the boot models list with no luck. Any help would be greatly appreciated. Thanks, Matt

2 1

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users June 2016