users

users@lists.genode.org

1 participants
2310 discussions

Segmentation fault in free_io_mem
by Johannes Kliemann 13 Mar '18

13 Mar '18

Hi all, while testing the IO_MEM session on Linux with ACPI I noticed that the acpi_drv segfaults when freeing its io memory on repos/os/src/drivers/acpi/acpi.cc:1304 [1]. The crash happens then in repos/base/include/base/allocator.h:319 at > operator delete (obj, dealloc); I noticed that this is called many times without problems but not from free_io_mem. Unfortunately I have to admit that my C++ knowledge is insufficient to really grasp what happens in this expression. Is there any difference in how io memory is freed or should there be any (at least on Linux)? Regards, Johannes [1]: https://github.com/jklmnn/genode/blob/c2e9d35010a8a3c4f79d51d022cb3e5beb6f0… [2]: https://github.com/jklmnn/genode/blob/c2e9d35010a8a3c4f79d51d022cb3e5beb6f0…

2 2

Quota Preservation
by Nobody III 13 Mar '18

13 Mar '18

I sent this last year, but nobody responded, and I still need an answer. The RAM preservation feature in init is very nice; I used to have issues with init running out of memory. However, the RAM reservation applies after first reserving the RAM for all of the child processes based on their quotas. That may be fine in most cases, but it doesn't work properly when you give a child process a huge RAM quota to ensure that it gets all of the available RAM. In that situation, the RAM preservation happens after all of the available RAM has been reserved for the child processes, leaving no RAM available for init to preserve for itself. Something needs to change here. Changing the ordering is the simplest solution, but there may be a better solution. When we reserve all the remaining RAM for a particular process, we supply a huge RAM quota that is much larger than the process actually needs. It seems that the solution would be to split the quota numbers into a minimum (amount *reserved*) and a maximum (amount *available*). It would work very nicely for standard desktop workloads, especially when web browsers are involved. For example, I'd like to ensure that chromium has access to up to 2G of RAM, if it is available, but I don't want to reserve 2G of RAM specifically for chromium. Would this solution work well? Or is there something that I'm missing here?

2 6

Sculpt not accepting input from touchpad
by Nobody III 13 Mar '18

13 Mar '18

When I run the Sculpt scenario on my laptop, the touchpad doesn't work, unlike in the old Turmvilla scenario. How can I fix this?

2 2

menu_view: does not get focus from parent
by Boris Mulder 08 Mar '18

08 Mar '18

Hi All, I have an application which is a custom version of launcher. It has its own nitpicker session as well as a menu_view_slave that is used to display a context menu if a certain element is clicked, similar to the launcher. In my run script, I used the focus ROM policy in nitpicker such that this launcher is focused when it's turned on in a similar fashion to the leitzentrale from sculpt. So far, everything has worked well. However, when I updated my genode version from 17.11 to 18.02, the context menu never gets the focus, even when it's up. Nitpicker does produce a hover report for the menu, and the menu is displayed correctly and fades when the mouse is no longer over it, just as in launcher. But the focus is never transferred to the menu, and I cannot click on the buttons in the menu, nor do they change color when I hover them (which used to be the case). The policies in nitpicker and the focus handling should be correct as they did work on 17.11. I checked if anything important was changed in both the launcher and the nitpicker after 17.11, but I could not find any change that I thought was relevant for this problem. I am also currently not able to run the launcher or menu_view run scripts. menu_view.run crashes with an out of ram error and launcher gives a stale screen with "menu" on the top left but does not do anything else. Are you able to run those tests? Do you have any idea where to look for? What change to 18.02 could affect the focusing of a context menu? -- 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 3

32 bit overflow in timer
by Johannes Kliemann 06 Mar '18

06 Mar '18

Hi all, I noticed a 32 bit overflow in the Timer::Connection sleep functions and triggers. Calling usleep or trigger_once with an hour (3 600 000 000 us) works but calling it with two hours (7 200 000 000 us) causes something to overflow and stop the timer after 2 905 032 704 us which equals 7 200 000 000 & (2^32 - 1). As far as I have checked this these functions use the unsigned long which is indeed 64 bit but I think at some point only 32 bit are used. Is this intended behaviour? If so I think it would make sense to give out a warning at least. Regards, Johannes

4 7

Sculpt EA On VirtualBox
by John J. Karcher 06 Mar '18

06 Mar '18

Hello, I have downloaded, built, and begun testing Sculpt EA, following the instructions linked to in the 18.02 release notes. Everything went well during the build process. Since I primarily run everything I do in VirtualBox (and don't have the recommended hardware anyway), I created a VM as close to the recommendations as possible. It booted fine from a virtual CD, the system came up using the VESA framebuffer driver, the mouse worked, and I was able to follow the instructions until I got to the part about storage. Unfortunately, there are no "<device>" nodes under "<block_devices>" in "/report/drivers/block_devices". I have the VM set up with a "SATA" controller and a single virtual drive. Has anyone tried this? Do you have any hints on how to diagnose what is going wrong? Also, I modified the config to use wired networking (using a virtual Intel-based NIC), but I couldn't think of a way to easily test whether it is working or not. (I'm sure I'm missing something obvious here, but I appreciate any hints.) If I can get those two items working, it looks like I'll be in good shape! Thanks, -- John J. Karcher devuser@...544...

2 5

RasPi 3 build problem
by Ken.Dickey 05 Mar '18

05 Mar '18

Greetings, I git cloned Genode, downloaded last version of tools, and am following the instructions in doc/getting_started.txt I seem to be close but am lacking usb. Tried both sel4 and foc. I am totally unfamiliar with the build tools. Can some kind person help me out? Thanks much, -KenD vvv======================================vvv ## Did: ## ./tool/ports/prepare_port foc ## ./tool/create_builddir rpi ## -- then uncommented 'libports' directive in etc/build.conf Linux:x86.64:/mnt/sda4/genode >>> make -C build/rpi KERNEL=foc run/demo make: Entering directory '/mnt/sda4/genode/build/rpi' including /mnt/sda4/genode/tool/run/boot_dir/foc including /mnt/sda4/genode/repos/os/run/demo.run building targets: core init drivers/timer server/nitpicker app/pointer app/status_bar app/global_keys_handler app/nit_focus server/liquid_framebuffer app/launchpad app/scout test/nitpicker server/nitlog drivers/framebuffer drivers/input server/report_rom server/rom_filter drivers/usb drivers/gpio drivers/platform spawn make core init drivers/timer server/nitpicker app/pointer app/status_bar app/global_keys_handler app/nit_focus server/liquid_framebuffer app/launchpad app/scout test/nitpicker server/nitlog drivers/framebuffer drivers/input server/report_rom server/rom_filter drivers/usb drivers/gpio drivers/platform make[1]: Entering directory '/mnt/sda4/genode/build/rpi' Error: target 'drivers/usb' does not exist Makefile:214: recipe for target 'error_missing_targets' failed make[1]: *** [error_missing_targets] Error 1 make[1]: Leaving directory '/mnt/sda4/genode/build/rpi' Error: Genode build failed Makefile:312: recipe for target 'run/demo' failed make: *** [run/demo] Error 252 make: Leaving directory '/mnt/sda4/genode/build/rpi' Linux:x86.64:/mnt/sda4/genode >>> uname -a Linux tranquil 4.13.0-36-generic #40~16.04.1-Ubuntu SMP Fri Feb 16 23:25:58 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux ^^^=======================================^^^

4 5

Bender and GRUB1 "legacy"
by Valery V. Sedletski 05 Mar '18

05 Mar '18

Hi, I tried to boot Genode scenarios from my bootable flash stick. It uses GRUB 1 "Legacy".Â Tried to boot Genode/Nova with bender with GRUB "legacy" (actually, GRUB4DOS). I have the following entries in menu.lst: titleÂ Sculpt scenario (Genode/Nova) kernel /genode/bender module /genode/hypervisor hypervisor iommu serial novpid novga module /genode/sculpt/image.elf image.elf This scenario just hangs when booting. Does this work with GRUB2 only? In GRUB2, I see loading the "multiboot2.mod" module. Is this a new version of Multiboot protocol? Could Bender work with older version of multiboot protocol? Or, is it possible to use Genode/Nova without Bender somehow with GRUB1? Also, is it possible for Genode build system to avoid assembling all modules into image.elf, so I could load them separately as multiple modules? (at least, older Genode versions did so) 10nx in advance! WBR, valery

4 19

keyboard/keymap issues
by Nobody III 05 Mar '18

05 Mar '18

In Sculpt, I have noticed 2 major keyboard issues: 1. The caps lock and shift keys are treated identically, which gives unexpected results (symbols instead of numbers) when a user types numbers with caps lock enabled. 2. The delete, end, and home keys don't work properly in bash. They are equivalent to typing the '~' character. Those keys work fine in vim, however.

2 1

Sculpt questions, briefly answered
by Norman Feske 05 Mar '18

05 Mar '18

Hello everyone, it's really great to see your interest in Sculpt! Hereby, I'd like to clarify a few things about Sculpt that were raised in the discussion. Question: Is any kernel supported other than NOVA? No. Please stick to NOVA because it is the only kernel used with Sculpt at Genode Labs. (this may change throughout the year) Question: How about using disk partitions? We deliberately left this topic out of the scope of Sculpt EA because it is not essential for a minimum-viable general-purpose setup. It goes without saying that Sculpt will support partitions down the road. You can use part_blk manually but it is not the official Sculpt path today. Question: Improving the detection of the default block device? The current heuristics is overly simplistic and of course not here to stay. Our current idea is to let the drivers subsystem scan the available block devices for one that hosts a partition with a magic label like "GENODE". Question: Do we plan to introduce any boot-time options? No. Genode's core has no boot-time arguments. User-provided configuration enters the picture at the level of the init component. Question: How to debug driver-related issues? If Sculpt does not run out of the box, please start experimenting with running simpler scenarios on your hardware first. E.g., Sculpt's drivers subsystem can be executed without the rest of Sculpt using the gems/run/driver_manager.run script, which - in contrast to Sculpt - outputs driver messages to core's LOG service that prints the message to the serial line. Question: What is the scope of features of the drivers subsystem? Since the drivers subsystem is not managed or configured by the user, it must stay as simple and generic as possible. Everything that can in principle be realized as part of the runtime subsystem should be part of the runtime, not the drivers subsystem. E.g., part_blk does not belong into the drivers subsystem. Question: What is the vision behind the Leitzentrale? Eventually, all actions that we perform manually via Noux as of today will become available via a graphical user interface. Under the hood, this user interface will "emulate" the user's manual interaction with the report and config file systems. Once all (common) boot-time operations are covered by the graphical user interface, Noux will be removed from the boot image. Question: How will the end result of Sculpt look like? We don't know. Sculpt EA is just a stepping stone that we use to explore the design space. Cheers Norman -- Dr.-Ing. Norman Feske Genode Labs https://www.genode-labs.com Â· https://genode.org Genode Labs GmbH Â· Amtsgericht Dresden Â· HRB 28424 Â· Sitz Dresden GeschÃ¤ftsfÃ¼hrer: Dr.-Ing. Norman Feske, Christian Helmuth

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users