users

users@lists.genode.org

2 participants
2342 discussions

additional question from "memcpy, strlen etc."
by Peter Nguyen 29 Jan '10

29 Jan '10

As an addition to the previous post by me, I have some pre-existing C code that I want to use in a Genode server I have created, and I don't really want to change this code. Currently, i'm not able to utilize functions like printf, memcpy etc in the C code i've written. What header files can I include to ensure that functions like memcpy and printf work in my code? From what I can see, the build process does not recognize the standard header files at all. Peter

4 3

Further response
by Peter Nguyen 28 Jan '10

28 Jan '10

Hi, In doing some more investigating, for some reason, that message seems to pop up when i instantiate a thread. Before this thread, i instantiate 3 threads, which execute fine, yet when i instantiate this 4th thread, it doesn't work. Furthermore, when i only instantiate this 4th thread (ie. commenting out the 3 previous thread instantiations), this thread works fine. Peter

2 1

qt4 on genode 9.11 on l4ka:pistachio
by Tang, Shuo 22 Jan '10

22 Jan '10

Hi, I am trying to get qt4 on genode 9.11 to work. I am using pistachio. When I try to run a iso using qemu, it halt and give a pagefault on CPU 0. Does anyone have a grub menu list file and genode config file for a qt4 demo? I have no problem running simple genode application on pistachio such as scout. Thanks, Shuo

2 1

Thread IDs
by Peter Nguyen 20 Jan '10

20 Jan '10

Hi, So I've done some more messing around with genode (running on top of OKL4), more specifically creating some threads, and testing the functionality in general. I'm wondering how i can go about obtaining the thread's ID and printing it out. I've noticed that there are functions that are within the Thread class that can get me a Native_thread_id, but currently I can't seem to do much with it. Speaking of the native thread, is that type related to L4? Can I actually get a numerical value for that ID? Peter

2 3

Build errors
by Peter Nguyen 18 Jan '10

18 Jan '10

Hi, In running genode with the "hello" app, I keep getting the following message: [init] starting hello_server with quota=262144 [init] using unique child name "hello_server" [init] Warning: Could not open file "hello_server" at ROM service. I also get the same thing with the hello client. Does this mean that the hello app is not loading at all? What do I need to do to fix this? Thanks Peter N

2 1

Broken assertion
by Peter Nguyen 05 Jan '10

05 Jan '10

Hi, In the wiki describing how to build Genode on OKL4, does the bug alert refer to following message: An error occurred: [Errno 2] No such file or directory: 'config'? I've followed the build instructions, and i've changed the value as suggested as the quick fix, but I still get this error. As usual, any help would be greatly appreciated. Peter

2 1

Cannot run genode
by Peter Nguyen 04 Jan '10

04 Jan '10

Hi, I'm trying to boot genode on OKL4 natively. I followed the instructions for building genode via the following link: http://genode.org/community/wiki/revisions/GenodeOnOKL4.6/?searchterm=build I've tried building version 9.05, but when i try to boot it up, i keep getting a grub error (Error 28: Selected item cannot fit into memory). Does anyone have any ideas why this is occurring? Peter N

3 2

Announcement: Genode OS Framework version 9.11 released
by Stefan Kalkowski 25 Nov '09

25 Nov '09

The new release 9.11 of the Genode OS Framework is available. The current release comprises a whole bunch of new functionality with respect to libraries, OS services and device drivers, like the Light-weight IP stack, Qt4 Webkit and USB storage, just to mention a few. Moreover, we incorporated new generic interfaces and made parts of Genode more efficient especially with respect to memory consumption. Apart from functional extensions and improvements, the new release contains initial support for the ARM architecture. As cherry on the cake, we release an adapted version of OKLinux running on top of the Genode/OKL4 platform. As a consequence, native Genode applications run side-by-side with unmodified Linux binaries. Release-notes summary for version 9.11 -------------------------------------- * Base framework * Signal-framework refinements * Binary-size optimizations * Paravirtualized Linux on top of Genode on OKL4 * Initial support for the ARM architecture * Operating-system services and libraries * Packet-streaming interface * Fully integrated shared-library support * Light-weight IP stack * Zero-footprint runtime for Spark/Ada programs * Port of popular libraries such as Freetype2 and Jpeg * Device drivers * Support for USB storage * PS/2 mouse scroll-wheel support * NIC driver service * Qt4 Webkit * Support for Jpeg and Freetype2 * Applications * Xvfb integration into Genode on Linux * Improved configurability of native applications You can find the complete release notes for the version 9.11 here: http://genode.org/documentation/release-notes/9.11/ The new release is available to the project's subversion repository and at the Genode download page: http://genode.org/download/latest-release Regards -- Stefan Kalkowski Genode Labs Developer http://genode-labs.com

1 0

Sharing locks between several address spaces
by Sven Fülster 20 Oct '09

20 Oct '09

Hi there, I'm working on two applications which require communication over shared memory. Therefore, I've adapted some of your datastructures (such as the slab allocator, semaphores and locks) so that they are safe to be shared between different address spaces, by adding smart pointers and taking care of having all shared components (like the Applicant list of the locks, the applicants and the locks themselves) within the shared memory. I have now a little problem with the locks. I have a thread in address space 1 which performs a lock operation on a lock in shared memory. The lock operation is (by default) mapped to the L4_Stop syscall (I work with the Okl4-Kernel). Another thread in address space 2 performs a wake-operation on the same lock (where the thread id is deposited in the applicant), which is mapped to L4_ThreadSwitch which is again mapped to the L4_ExchangeRegisters syscall. This operation does not wake the first thread. In detail, the L4_ThreadWasHalted call which is performed upon L4_ThreadSwitch returns false, hence the thread which tried the wake operation remains spinning in the forever-loop (this is a nice opportunity for a denial-of-service-attac of a malicious client by the way...). However, I've started to investigate by replacing the L4_ThreadSwitch syscall by L4_WaitNotify and the wake-operation by L4_Notify. The result was the same. Checking the error code of the L4_Notify syscall with L4_ErrorCode showed that the call had been aborted due to L4_ErrInvalidThread, respectively L4_ErrNonExist (2). Now I wonder if a thread is generally allowed to invoke syscalls on threads which don't belong to its address space (maybe I need some capability? What exactly is env()->pd_session()->bind_thread() for?). I've also noticed that the thread ids I get from env()->cpu_session()->state() or from thread_get_my_native_id() differ from those the kernel debugger shows (what exactly do you do when you ask Okl4::__L4_TCR_ThreadWord(Genode::UTCB_TCR_THREAD_WORD_MYSELF) for a thread id? Is that the global thread id?) Perhaps you have an idea what's going wrong here (meanwhile I'd add L4_MutexControl to core :o) ). With kind regards Sven -- Sven FÃ¼lster

2 1

Re: Split console with two keyboards
by Christian Helmuth 16 Oct '09

16 Oct '09

Hello Shakthi, On Fri, Oct 16, 2009 at 12:16:21AM +0530, Shakthi Kannan wrote: > Hi, > > I would like to know if it is possible in any of the L4* systems to > have a setup like the following: > > * The console (framebuffer) to be split into two frames/windows, say > top and bottom halves, but, each one should be used/accessed > independently of each other. > > * With two keyboards connected to this single system, one should be > able to use (shell) one of the keyboards with the top window, while > the other should use the bottom one. > > Appreciate any thoughts/inputs in this regard. Thanks! I don't know of any L4-based system, which supports this setup out-of-the-box. IMO Genode [1] and L4Env [2] have all you need to build such a system on your own: PS/2 drivers, USB HID drivers, and frame buffer console/GUI. Both options implicate some hacking on your part as multi-head scenarios were not addressed so far. BTW, Genode seemlessly supports several L4 kernels, namely OKL4, Fiasco, and Pistachio. That's a nice feature if you're not certain which kernel fits your needs best and want to postpone the final decision :-) [1] http://www.genode.org [2] http://www.inf.tu-dresden.de/index.php?node_id=1431 Regards -- Christian Helmuth Genode Labs http://www.genode-labs.com/ · http://genode.org/

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users