users

users@lists.genode.org

7 participants
2332 discussions

Service denied when server is updated
by Stefan Thöni 10 Jul '18

10 Jul '18

Hello Genodians, I observed that when a runtime configuration with a app already present (downloaded and unpacked) accessing a service on new server (version) is deployed, the app gets a Service_denied exception while the new server (version) is being downloaded, verified or unpacked. Wouldn't it be better to start all components in the runtime configuration only when all possible downloading is complete? Otherwise we either need to make components tolerant to Service_denied exceptions or detect such error conditions and restart the runtime entirely. Cheers Stefan -- Freundliche Grüsse Stefan Thöni Senior Security Architect +41 79 610 64 95 gapfruit AG Baarerstrasse 135 6300 Zug https://gapfruit.com

2 3

Sculpt built from current master - availability of packages
by Tomasz Gajewski 07 Jul '18

07 Jul '18

Hi, after release Sculpt TC I've made another attempt to use it. I've downloaded official image and attempted to boot on my computers. On one it rebooted soon after splash screen probably due to issue with IOMMU investigated some time ago in issue 2700 to which I hope to get back to in near future. On another one it booted but failed to initiate ethernet controller. I've checked that it is an intel i219v-2 which is handled by dde_ipxe in newer version than the one in genode I've made some changes that resulted in commit: https://github.com/tomga/genode/commit/7ce2291796c2752974c182e42aebfc8d8d31… that allowed to get ip address from my router so it that networking is ok. This commit contains a simple change to intel.c patch and three changes to hashes that were needed to build sculpt after this change. Is it normal that each change needs update of hashes of all dependent components? Do I correctly understand that this will force many changes of hashes in repository? Next I attempted to perform next step from Sculpt TC manual and uncommented part from config/deploy about fonts_fs and it attempted to download some packages from depot.genode.org. Files that were attempted to be downloaded do not exist on this server. There are similar but with different timestamps e.g. download of .../fonts_fs/2018-07-02.tar.xz was attempted but the latest one in this location is 2018-06-12.tar.xz This date, if I correctly understand, comes from: b485caf33c (Christian Helmuth 2018-06-25 12:51:03 +0200 1) 2018-07-02 a31d98a7dfb8cd7913ec38b0dfe080ae3382f8b6 So the second question is: should I assume that if this "version" is in the git repository it should be available on depot.genode.org? If not, what would you suggest to proceed? Regards Tomasz Gajewski

2 3

Error after running run/printf. I built hw_imx6
by Prakash A 06 Jul '18

06 Jul '18

Hi, I am getting error while making a run/printf and my build platform is hw_imx6 and genode version 13.02. *$ make run/printf* .... . . .Skip target core/vea9x4/trustzone because it requires platform_vea9x4 trustzone Library platform Library cxx Library kernel_interface Library base-common Library perf_counter Program core/imx6/no_trustzone/core ASSEMBLE boot_modules.o LINK core make[1]: Leaving directory '/home/prakash-a/secureos/tool/Genode_imx6' SHOULD I BUILD UIMAGE BUILD UIMAGE *spawn qemu-system-arm -nographic -m 64 -kernel var/run/printf/image.elf* *qemu-system-arm: -kernel var/run/printf/image.elf: No machine specified, and there is no default* *Use -machine help to list supported machines* *Error: Spawned process died unexpectedly* *Makefile:237: recipe for target 'run/printf' failed* *make: *** [run/printf] Error 253* I dont know what is the machine for imx6and how to include her. Regards, Prakash A

2 1

sel4: Need to have python module 'future' installed
by Paul Dufresne 06 Jul '18

06 Jul '18

Is it my system that is supposed to give Python modules, or is it supposed to be part of the toolchain? I get it on: paul@paul-OptiPlex-745:~/genode$ ./tool/ports/prepare_port sel4 /home/paul/genode/repos/base-sel4/ports/sel4.port:11: *** Need to have python module 'future' installed.. Arrêt. /home/paul/genode/tool/ports/mk/prepare_single_port.mk:38: recipe for target '_install_in_port_dir' failed make[1]: *** [_install_in_port_dir] Error 2 tool/ports/prepare_port:33: recipe for target 'sel4' failed make: *** [sel4] Error 2 paul@paul-OptiPlex-745:~/genode$

2 1

depot/build creates target directory with missing dependencies
by Guido Witmond 06 Jul '18

06 Jul '18

Hi Genodians, I think I start to get the hang of the depot system. However, I get this behaviour with missing depencencies: $ tool/depot/create guido/src/feather created guido/src/feather/2018-07-04-c $ tool/depot/build guido/bin/x86_64/feather/2018-07-04-c KEEP_BUILD_DIR=1 ... Library description file libc_lwip.mk is missing Library description file libc_lwip_nic_dhcp.mk is missing created guido/bin/x86_64/feather/2018-07-04-c But that directory is completely empty. Running the command a second time give me: make[1]: Nothing to be done for 'all'. As there are missing dependencies, I'd expect the second run to give the same error as the first run, not a message that everything is fine. To retry I currently rm -r depot/guido and create and build again or add the REBUILD=1 flag. My questions: - How can I add these lwip-libraries included? I don't see any examples of these included in used_apis. - What's going wrong that it mentions the missing deps but still creates the target directory? Cheers, Guido

2 1

make fun: optimize depot/create by using depot/download
by Roman Iten 06 Jul '18

06 Jul '18

Hello, Merely as an experiment to avoid building already published/uploaded archives, I tried to integrate depot/download into the depot/create front-end. Having basically zero make experience, I came up with [1]. If the environment variable DOWNLOAD is set, it 'seems' to do what I intend: - extract API/source/raw archives (make sure we use the correct version) - download as many archives as possible (best-effort) - build everything else, that couldn't be downloaded Still I have some questions: - Can you think of any (corner) cases which would result in an incomplete/inconsistent build regarding the content of the working directory? - How can I deal with the following warning (when calling depot/download): make[2]: warning: jobserver unavailable: using -j1. Add '+' to parent make rule. - Is there a simple way to prevent the ugly log output as it may contain (expected) error messages like: Error: failed to download 'https://...' make[2]: Target '...' not remade because of errors. - Anything else I didn't consider? Comments are highly appreciated, thanks! Roman [1] https://github.com/rite/genode/commit/1b50ac73dc3b4986c1df1694b88ecc49757d0…

1 0

double-fault handler?
by Paul Dufresne 06 Jul '18

06 Jul '18

I am far from an expert, but I remember from expensive 386 programming book I bought so long ago, that certain things would cause an exception, like page not present, division by zero, opcode not found... and when that this arrive, a specific exception handler is run for the specific exception. If during the execution of an exception handler, an exception happens, then the double-fault exception is called. If finally an exception happens during the execution of the double-fault exception, then the CPU itself decide the OS is so buggy, that it have to handle the thing itself, by rebooting the CPU. What I don't like about the fact that Genode would reboot is that it suggest to me Genode did not care about writing double-fault handler, or wrote a buggy one, and seems to not have tested it. What does Genode double-fault handler is supposed to do? Sorry for the ... pedantic tone, in reality I really I does not know so much about programming, and appreciate very much the Genode way of doing things in general.

1 0

foc without kernel debugger
by Johannes Kliemann 05 Jul '18

05 Jul '18

Hi all, I tried to use Fiasco.OC without the kernel debugger, respectively with the following config changes: > diff --git a/repos/base-foc/config/x86_64.kernel b/repos/base-foc/config/x86_64.kernel > index 07159da20d..8b8b1c8c50 100644 > --- a/repos/base-foc/config/x86_64.kernel > +++ b/repos/base-foc/config/x86_64.kernel > @@ -56,12 +56,7 @@ CONFIG_STACK_DEPTH=y > # CONFIG_IRQ_SPINNER is not set > # CONFIG_WATCHDOG is not set > CONFIG_SERIAL=y > -CONFIG_JDB=y > -CONFIG_JDB_LOGGING=y > -CONFIG_JDB_DISASM=y > -CONFIG_JDB_GZIP=y > -# CONFIG_JDB_ACCOUNTING is not set > -CONFIG_JDB_MISC=y > +# CONFIG_JDB is not set > CONFIG_POWERSAVE_GETCHAR=y > # CONFIG_WARN_NONE is not set > CONFIG_WARN_WARNING=y It compiles without any errors but the execution fails with multiple messages of: > KERNEL: Warning: nothing mapped: (Obj_space) from [0xffffffff81fd1e90/ffffffffffffffff]: 000000000000000a size: 0000000000000001 to [0xffffffff81fd1d80/ffffffffffffffff] Are there any further requirements to use foc without the kernel debugger? Regards, Johannes

3 3

Error with implementing custom IRQ session
by Harsha Sharma 05 Jul '18

05 Jul '18

Hello, I am trying to implement a custom irq session, using the RPC call functions from genode irq_session header files. The code is here : https://gist.github.com/hs256/e6a72788b7471316e04e242fbf4e82e4 Can anyone, please suggest what can be the reasons for this kind of error : "undefined reference to vtable and typeinfo" ? Thank you very much. Regards, Harsha Sharma

1 0

Checkpoint/restore of four specific capabilities
by Joos, Lukas 05 Jul '18

05 Jul '18

Dear Genode community, I am currently trying to extend a checkpoint/restore mechanism based on Fiasco.OC and Genode 16.08 that has been discussed on this mailinglist before. [1] [2] During my analysis of the existing code I was able to identify four capabilities that do not show up in the child's capability space after a restore: CAP1: Fiasco.OC GATE representing Genode::Rpc_entrypoint::_exit_cap CAP2: Fiasco.OC GATE representing Genode::Entrypoint::_signal_proxy_cap CAP3: Fiasco.OC IRQ representing Genode::Signal_source_rpc_object::_blocking_semaphore (Fiasco.OC specific) CAP4: Fiasco.OC GATE representing "Capability<Constructor> constructor_cap" local to the Genode::Entrypoint::Entrypoint(Env &env) constructor scope Based on my current understanding I'd say that CAP4 is not needed after a component has been constructed. CAP1 may be needed when a component is destroyed and CAP2 and CAP3 are needed for the signalling subsystem to work correctly. Regarding the Fiasco.OC specific CAP3 there is also the problem that it has to be attached to the "signal handler" thread with a system call (l4_irq_attach). The current idea (which has also been discussed previously [3]) is to delegate the thread and IRQ capabilities to core, which then invokes the syscall with the translated kcap values. My questions are: 1) Which of those four capabilities actually need to be present in a restored child? 2) Do they have to be in the capability map? 3) Do they have to be in the capability space? 4) Does the memory where the Genode::Capability<RPC_INTERFACE> is stored inside the child component have to be altered after the restore? If you have any ideas on how to restore one or more of those capabilities I'd love to hear all about it. Thank you for your time. Sincerely, Lukas Joos [1] https://lists.genode.org/pipermail/users/2016-August/004703.html or https://sourceforge.net/p/genode/mailman/message/35322604/ [2] https://lists.genode.org/pipermail/users/2016-December/004863.html or https://sourceforge.net/p/genode/mailman/message/35529306/ [3] https://lists.genode.org/pipermail/users/2017-November/005611.html

1 1

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users