users April 2017

users@lists.genode.org

22 participants
23 discussions

Now: Nova not maintained [Earlier]: Understanding Muen [Earlier] Re: Problem with make run/demo ....
by Shahbaz khan 19 Apr '17

19 Apr '17

Hi Nobody, Please don't mind my changing the subject title. On Tue, Apr 18, 2017 at 11:44 PM, Nobody III <hungryninja101@...9...> wrote: > By the way, NOVA doesn't seem to be maintained anymore, except a bit by > Genode developers. Is this observation correct? > My observation is that most of the innovations were achieved by 2012 in microkernel research so stable work is what we see in Fiasco.oc and Nova. The only problem was that these OSes could not be formally verified. Muen answers the problem by hardcore simplicity, while I believe SEL4 has practicality issues. Formal verification is not interesting for developers so somehow funds went somewhere else. Just a thought. Rest is driver development which is not a research and showcase problem and handed by industry rather than universities, TUD Genode people are the pioneers and can answer better. Regards, Shahbaz

2 2

Re: Now: Understanding Muen and base_hw kernel from docs, [Earlier] Re: Problem with make run/demo ....
by Shahbaz khan 19 Apr '17

19 Apr '17

Dear Norman, On 17.04.2017 19:53, Shahbaz khan wrote: > > VirtualBox closely interacts with the virtualization hardware. On > Muen, > > this interaction naturally has to go through the Muen SK. By using > > Genode as runtime for VirtualBox, Muen is able to leverage Genode's > > existing solution of the interaction of VirtualBox with a > > microkernel-based virtualization mechanism. > > > > > > Similar to karma-vmm, fiasco.oc and l4linux. > > I don't see the connection. My email was not related to Fiasco.OC, > Karma, or L4Linux. > I think I got the problem. Muen subject is not a subject like that of an OS. > > > Conceptually, a Muen partition is a hardware platform, similar to a > > board. Like on any board, you can run software directly (in > supervisor > > mode). But for running a complex software stack, or more than one > > application, one has to use an operating system (OS). Genode/base-hw > > plays this role. > > > > > > Simply put a build time think. Base-hw kernel means muen sk for genode > > runtime. > > Muen can be used without base-hw. > Base-hw can be used without Muen. > Second problem in my mind would be imagining the spartan environment provided by Muen to be fulfilled with Genode runtime minus microkernel. This would be as follows. Base-HW is a microkernel as you mentioned earlier. Like a layman I ask the name of this microkernel. Is it Nova? Again to my preliminary question ... we run a micro-kernel on microkernel to run virtualbox rather than virtualbox on Muen SK. > > But when combined, Genode's version of VirtualBox can run on top of Muen. > Muen is a micro-hypervisor. What is the nature of inter-subject communication? Considering we don't like pure hypervisor approach. I liked Nova's architecture much more and I would opt for Nova if formal verification could be as simple for it as in Muen. And the luxury of C++ too. Regards, Shahbaz

3 2

Genode, seL4 and VirtualBox
by Chris Rothrock 18 Apr '17

18 Apr '17

I'm working with the team at Critical Technologies and I am trying to get this combination to work - Genode with seL4 microkernel and the VirtualBox module (using TinyCore Linux). Let me first start off with saying that Genode, Nova and VirtualBox works flawlessly - I can boot (an Intel-based machine) and go into either a command line or GUI environment. The problem is when I try using seL4. I can compile this without any errors. The ISO image is made perfectly, it boots to this image on the same hardware - but that's where it stops. One time only I was able to see the loading screen of TinyCore Linux within VirtualBox but the boot time for this took about 45 minutes and I was unable to reproduce this boot. I guess my first question is this - does seL4 have the necessary virtualization support for this module? If so, how can I make this work reliably? -- Thank You, Chris Rothrock Senior System Administrator (315) 308-1637

3 3

Now: Suitable ARM alt. to Fiasco.OC [Earlier]: Understand Muen scenario, [Earlier] Re: Problem with make run/demo ....
by Shahbaz khan 18 Apr '17

18 Apr '17

> > Base-HW is unrelated to NOVA or any of the other kernels. It is a > > microkernel specifically developed for and tightly integrated with > > Genode. Think of it as an optional back end of Genode to run > > Genode-based systems directly on hardware (or virtual hardware like a > > Muen partition) without a third-party kernel. > > > > > > By the way it seems fiasco.oc is not being maintained. Is something > > better available, which supports both micrkernel based runtime and > > virtualization..? > > At Genode Labs, we primarily focus on NOVA, base-hw, and seL4. All these > kernels are third-generation modern microkernels [1]. For using Genode > on x86 PC hardware, NOVA is the best way to go right now. > > Genode is also able to run on Fiasco.OC but we don't actively facilitate > it. E.g., we dropped the support for L4Linux (Fiasco.OC-specific > paravirtualized Linux) earlier this year [2]. > What is the ARM alternative to Fiasco.OC? That is supporting micro-kernel subjects as well as virtual machines. Will it make sense if we (my team) supports Fiasco.OC if suitable microkernel is not available for ARM? In case of X86 Muen seems to be the best as I am interested in formal verification like everyone else. Fiasco.OC is troublesome when it comes to formal verification. Regards, Shahbaz Khan

1 0

Now: Understanding Muen and base_hw kernel from docs, [Earlier] Re: Problem with make run/demo ....
by Shahbaz khan 18 Apr '17

18 Apr '17

On Fri, Apr 14, 2017 at 10:28 AM, Nobody III <hungryninja101@...9...> wrote: > I've been wondering, is there any reason not to use base-hw_x86_64_muen on > regular hardware? It seems to work fine in Qemu, and > Release notes mention Thinkpad X201.Muen website makes me cautious but if Qemu works then we are good for a start. Have your tried virtualbox? So bochs is not mandatory. > almost all Genode drivers are in userspace. Is there some limitation that > I'm not aware of? > > Well anything is easily possible with micro-kernel architecture in concept but I was curious whether we run base_hw kernel that is muen according to mainline repo's files. The diagram from release notes depict base_hw kernel on Muen SK. It still does not make sense to me. I will have to look later as I am working on some basic presentation on Muen and genode for some prospects. > > On Thu, Apr 13, 2017 at 1:38 PM, Shahbaz khan <shazalive@...9...> wrote: > >> >> >> On Wed, Apr 5, 2017 at 2:36 PM, Christian Helmuth < >> christian.helmuth@...1...> wrote: >> >>> On Wed, Apr 05, 2017 at 11:54:47AM +0500, Shahbaz khan wrote: >>> > I have started working with Genode, while I am interested in foc based >>> > development and it seems the Genode efforts also prefer foc as I have >>> been >>> > trying run/demo for builds. >>> >>> I would not go along with telling that we "prefer" base-foc as most of >>> our scenarios run on all supported kernels. Prominent exceptions are >>> support for hardware virtualization (only on NOVA and hw/Muen) and >>> TrustZone (hw only). If you address ARM and x86 alike I'd suggest you >>> try out base-hw. >>> >> >> I got a bit acquainted with the release notes but I still find it >> difficult to comprehend what really base-hw means when we say "Genode base >> -hw" on Muen SK. >> >> I am having some build issues (I do get an image.elf but build does not >> complete) and have yet not looked into earlier commits and branches so I >> would appreciate if you can let me know what actually runs over Muen SK in >> Genode base-hw kernel VM. Does it mean we run another kernel like nova >> between Muen and virtualbox. >> >> I really like the idea but if we have another kernel below virtualbox >> than why not run virtualbox on Muen as a subject? I am unable to criticize >> clearly due to the mentioned confusion. >> >> Thanks. >> >> Regards, >> Shahbaz >> >> ------------------------------------------------------------ >> ------------------ >> Check out the vibrant tech community on one of the world's most >> engaging tech sites, Slashdot.org! http://sdm.link/slashdot >> _______________________________________________ >> genode-main mailing list >> genode-main(a)lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/genode-main >> >> > > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, Slashdot.org! http://sdm.link/slashdot > _______________________________________________ > genode-main mailing list > genode-main(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/genode-main > > -- Shahbaz Khan

2 3

Ideas about interactive GUI
by Boris Mulder 15 Apr '17

15 Apr '17

Dear Genode Community, We have been brewing on the best way to implement an interactive menu interface. We would like to hear your opinions and ideas on this. Currently, only launcher provides this functionality. However, we had the following objectives in mind: - Decouple the menu state and the functionality. We think the menu should only be responsible for rendering the dialogs and registering input events. Whatever functionality is to be invoked based on this input should be delegated to other components. - Have a fully customizable menu. The model of what buttons and (nested) submenus exist should be configurable, and should be separated from the layout (such as how a button looks like) and from the current state (which menus are collapsed, etc). In order to shape this, we have the following ideas: - The configuration of the menu model can be specified in XML in its config, and be reconfigurable at runtime. Parts of this could be defined by other components as well. Examples of how this could look like is described in the attached xml file. - Events related to the menu either trigger menu actions (in which case the config dictates the proper response, such as showing or hiding a submenu) or trigger external actions. In the second case, the menu generates a report that for instance states a certain button has been clicked. Again, see the attachment. - Any (custom) component can read these reports and take certain actions. For example, in case programs are to be started with the menu, we can add an init component that takes its config from another rom module. A subsystem management component containing a more extensive state model would read the events from the menu, change init's config based on this, and monitor the state of started subsystems and perhaps add extra context menus for these subsystems. Another example is a shutdown button that talks to a component that can write the system state, and which could show a dialog with the options to power off or reset a systems. Perhaps it can try to gracefully stop running programs with disk state (such as virtualbox) by talking with the subsystem manager as explained above. In any case, the menu should be able to support any functionality by providing a generic interface to components such as the examples above. - The layout of the menu can be specified in the config as well, perhaps using (a subset of) CSS, or as it is now, with the styles database that contains pictures for each element. (buttons, bars, etc.) While looking at ways to implement this, we also encountered the following problems: -As it is currently with launcher, a lot of subsystem functionality is intertwined with the interactive menu. This makes implementing such a modification a bit more difficult than replacing the subsystem management part with a reporting part. One of the issues we face is that showing and hiding windows of other components is done through the session_control function in the nitpicker session interface, but this is limited to the session of the components and its children. Children of an init component therefore cannot be affected. One way to fix this is to add a policy in the window manager or nitpicker that allows the menu or some other component to execute session_control on certain or all sessions. Implementing session_control functionality in init itself does not look like a clean solution. Another issue is that init does not keep track of exited children. Launcher neatly closes the corresponding menu entries by installing a signal when the child exits, but init does not have such functionality now. Perhaps init could be modified such that it generates a report when a child exits? For now, if we would quit a child with another method than through the menu, we have no way of knowing if a child exited. I'd very much like to hear your comments on these ideas and on the best way to realize it (perhaps together), and for instance how you think we could resolve the above issues. -- Met vriendelijke groet / kind regards, Boris Mulder Cyber Security Labs B.V. | Gooimeer 6-31 | 1411 DD Naarden | The Netherlands +31 35 631 3253 (office)

2 1

Problem with make run/demo & run/hellofor fiasco.oc based build
by Shahbaz khan 14 Apr '17

14 Apr '17

Dear list, I have started working with Genode, while I am interested in foc based development and it seems the Genode efforts also prefer foc as I have been trying run/demo for builds. I am facing a few problems. Some that makes sense as a developer, while some are new to me and definitely need some hints to move forward. I have encountered the following errors: x86_64 (run/demo & run/hello):- Error: target 'core' does not exist. I was not able to figure out how to enable build core but it should have been there by default like other Genode builds. pba9 (run/demo):- The mouse pointer does not work. pba9 (run/hello):- [init] Error: Uncaught exception of type 'Genode::Parent::Service_denied' linux (run/hello):- [init] Error: Uncaught exception of type 'Genode::Parent::Service_denied' Does the uncaught exception mean that the capability list is not configured properly? I am onto figuring out the build issue of core till your replies. Regards, Shahbaz Khan R&D Engineer, IM|Sciences

3 4

Socket fs
by Boris Mulder 13 Apr '17

13 Apr '17

Hi all, I've managed to port one of our existing network applications to use the socket fs from 17.02 now, starting from the echo_udp scenario and got it working. It looks a lot better now. Following this, we would like to share our findings about the socket fs so far: 1. In the release notes you describe the support for async file I/O. However, it seems that while asynchronous I/O works now, synchronous I/O is no longer possible. When I remove the select() call in the echo scenario, the call to recvfrom() will fail saying the socket is not connected (even though it is an UDP socket and should not have the notion of connections at all). Of course this means it is not posix compatible; the intended behaviour is to block until data becomes available. 2. Currently there are two ways an application can use tcp/ip sockets from the lxip stack: By using libc sockets, and by directly opening a file system session and writing to the right socket files. For any new Genode application that does not already depend on libc, the first way means you unnecessarily add the entirety of libc to your component. So the second way seems preferred. However, when programming an application using a file system session directly adds a lot of code with many open()s, read()s and write()s that could be abbreviated into single calls such as connect() and recv(). Have you thought of providing a more convenient API (such an object-oriented Socket class that hides the vfs calls from the application programmer)? Or do you consider using the fs session interface to be better? If so, why? -- Met vriendelijke groet / kind regards, Boris Mulder Cyber Security Labs B.V. | Gooimeer 6-31 | 1411 DD Naarden | The Netherlands +31 35 631 3253 (office)

3 3

infradead.org seems to be down
by Johannes Kliemann 11 Apr '17

11 Apr '17

Hi, while trying to build Genode with dde_linux I noticed that the connection to infradead.org timed out, so I'm not able to retrieve https://www.infradead.org/~tgr/libnl/files/libnl-3.2.25.tar.gz Is there any other source for this file? JK

4 7

Intel fb debug output
by Johannes Kliemann 10 Apr '17

10 Apr '17

Hi, I want to enable the debug output in the Intel framebuffer (to see messages printed by DRM_DEBUG_KMS). This is usually done by adding the boot parameter `drm.debug=0xe`. How would I enable this in Genode with Nova? JK

3 3

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users April 2017