users

users@lists.genode.org

2 participants
2322 discussions

Genode at FOSDEM (5th of February)
by Norman Feske 08 Feb '22

08 Feb '22

Hello everyone, during the upcoming weekend, the (virtual) FOSDEM 2022 conference will be held. It will feature another instance of our beloved Developer Room for Microkernel and Component-based Systems, this year organized by Martin Decky and Sebastian Sumpf: https://fosdem.org/2022/schedule/track/microkernel_and_component_based_os/ Among the packed schedule, the following talk is specifically about Genode: 11:50 CET "Genode meets the Pinephone" by Norman Feske https://fosdem.org/2022/schedule/event/nfeske/ In this talk, I will present our recent work of combining Genode with the Pinephone. It will also feature a short demo of the current state. The entire FOSDEM is free for everyone interested. Just tune in! Cheers Norman -- Dr.-Ing. Norman Feske Genode Labs https://www.genode-labs.com · https://genode.org Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth

1 1

Build errors while adding a new func to Region Map
by Sid Agrawal 04 Feb '22

04 Feb '22

Hi, I am using Genode with seL4 backend for my research. I am trying to add a new RPC method to RegionMap. It is meant as a simple exercise in understanding how the RPC framework in Genode works. The function name is "print_regions" and for now it does absolutely nothing. It takes in a void and returns a void. Unfortunately, I am running into some issues while adding this new RPC call to RM. I think I am missing something basic. First I ran into an issue where there was a mismatch between the ABI and library(3). I fixed it to the best of my ability and then got an error during library loading(4). Steps to reproduce and details of errors below. My diff is available here: Diff: https://github.com/sid-agrawal/genode/commit/26a79e7f1a77b99fb3b555198a29c6… HTML: https://github.com/sid-agrawal/genode/commit/26a79e7f1a77b99fb3b555198a29c6… It is just upstream genode + 1 commit (my changes) My platform is Ubuntu 20.04.1 on x86-64 To reproduce: git clone https://github.com/genodelabs/genode.git genode-question cd genode-question ./tool/ports/prepare_port sel4 ./tool/ports/prepare_port grub2 ./tool/create_builddir x86_64 # Apply patch wget https://github.com/sid-agrawal/genode/commit/26a79e7f1a77b99fb3b555198a29c6… git apply 26a79e7f1a77b99fb3b555198a29c638fd510bf4.diff cd build/x86_64 make run/thread KERNEL=sel4 Here is what I did in my diff. 1. I started off by adding pure virtual func, GENODE_RPC, and an associated change in GENODE_RPC_INTERFACE to Genode::Region_map. 2. Then I went on to fix the others files as the build errors popped up. This was basically in classes that inherited Region_Map. 3. Then I made a change to repos/base/lib/symbols/ld to work around the following error: Error: size of '_ZTVN6Genode17Region_map_clientE' symbol violates ld ABI (72 bytes in ABI, 80 bytes in library) make[3]: *** [/home/siagraw/genode/repos/base/mk/lib.mk:244: ld-sel4.lib.checked] Error 255 make[2]: *** [var/libdeps:200: ld-sel4.lib] Error 2 I am least sure of this change. I am unsure if this is a file I should even be modifying for something as simple as adding a new RPC method. But I did change it to get past the build error. 4. Then my "make run/thread KERNEL=sel4 " failed with the following error: Program lib/ld/sel4/ld-sel4 make[1]: Leaving directory '/home/siagraw/genode/build/x86_64' genode build completed using 'core-sel4.a' as 'core.a' using 'ld-sel4.lib.so' as 'ld.lib.so' /usr/local/genode/tool/21.05/bin/../lib/gcc/x86_64-pc-elf/10.3.0/../../../../x86_64-pc-elf/bin/ld: debug/core-sel4.a(region_map_client.o):(.data.rel.ro._ZTVN6Genode17Region_map_clientE[_ZTVN6Genode17Region_map_clientE]+0x40): u' collect2: error: ld returned 1 exit status while executing "exec [cross_dev_prefix]g++ {*}$arch -nostdlib {*}[core_ld_opts] -Wl,-z -Wl,max-page-size=0x1000 -Wl,-Ttext=$link_address -Wl,-gc-sections -Wl,-nost..." (procedure "build_core" line 18) invoked from within "build_core debug/$core_obj {} [run_dir].core [core_link_address]" (procedure "build_core_image" line 16) invoked from within "build_core_image $binaries" (procedure "run_boot_dir" line 19) invoked from within "run_boot_dir $binaries" (procedure "build_boot_image" line 35) invoked from within "build_boot_image "core ld.lib.so init test-thread"" (file "/home/siagraw/genode/repos/base/run/thread.run" line 48) invoked from within "source $include_name" ("foreach" body line 6) invoked from within "foreach include_name [get_cmd_arg --include ""] { # first check if the include name is absolute if {[string first "/" $include_name] == 0} { puts ..." (file "/home/siagraw/genode/tool/run/run" line 1115) make: *** [Makefile:405: run/thread] Error 1 I am at a loss on how to proceed from here and any help would be very useful. Perhaps I have missed another place where I should be adding my new function definition or perhaps the symbols need to be manually updated. As always, thanks for the help. Best, Sidhartha Agrawal

2 3

FOSDEM'22: Microkernel devroom is ready
by Sebastian Sumpf 04 Feb '22

04 Feb '22

Hi all, the "Microkernel and Component-based OS devroom" takes place tomorrow (05.02. from 10:40am - UTC+1). You can join either with Q&A or video only using the following links: With Q&A: https://chat.fosdem.org/#/room/#microkernel-devroom:fosdem.org Video only: https://live.fosdem.org/watch/dmicrokernel The schedule can be found under: https://fosdem.org/2022/schedule/track/microkernel_and_component_based_os Have a great weekend and we will be happy to see you, Sebastian

1 0

SYMBOLS / check_abi mechanism
by Roland Bär 03 Feb '22

03 Feb '22

2 1

SulptOS with Intel(R) Dual Band Wireless-AC 8265
by Lonnie Cumberland 01 Feb '22

01 Feb '22

Hello All, I am testing out Sculpt on my Intel NUC i7 and although it seems to boot, it runs rather slow as well as will not find the wireless network card that is built into the NUC (Intel(R) Dual Band Wireless-AC 8265) I am wondering if there are any updates on the wireless card detection for SculptOS. Also, I was thinking about the desktop and GUI (Nitpicker) and was wondering how hard it might be to change out the graphics to something more "themeable" like LVGL (https://lvgl.io/) but am still learning for which Genode might have these capabilities. Best Regards, Lonnie

3 2

Roadmap 2022
by Norman Feske 01 Feb '22

01 Feb '22

Dear Genode community, it is the time of the year again to reflect and make plans for the foreseeable future. Hereby, I'd like to kick off our traditional brainstorming about Genode's road map for the year ahead of us. What happened in the previous episode... ---------------------------------------- I have two perspectives to share. One is the look at Genode as our project, and the other one is my personal view. Let's start with the first one. When reviewing the past twelve months, I am immensely proud of the accomplishments of our team. Together, we conquered the territory of GPU support that was ridden with uncertainties and seemed almost impenetrable when we started. Now, our Intel GPU multiplexer has landed in Sculpt like it always belonged there. The next highlight was witnessing the puzzle pieces of our new Linux device-driver environment coming together, replacing former confusion and chaos with knowledge and order, ultimately uncovering the treasure of Linux drivers for Genode with very little friction. The third highlight was witnessing the growing sophistication of Genode-native workloads, with the media features of the Chromium-based browser on 64-bit ARM being the most impressive example. Apart from the apparent functional benefits for Genode and Sculpt OS, this is the long outstanding validation of some arguably risky design decisions we took many year ago, in particular the role and architecture of the VFS and its interplay with the libc. When reviewing the road map for 2021 [1], some items remained uncovered. In particular the seL4-related topics became stale. One year ago when I assembled the road map, there was a tangible prospect of a paying customer funding this work. However, those plans were repeatedly deferred and we don't know whether or when they will come to fruition. I was too optimistic. Also, there are some items that have seen healthy doses of progress - like the topics related to Ada/SPARK or Goa - but received less attention than anticipated. On the other hand, the four releases [2,3,4,5] of this year covered quite a few topics not advertised at the road map, e.g., webcam support, Xilinx Zynq, or RISC-V. Priorities shift. That's fine. [1] https://genode.org/about/road-map [2] https://genode.org/documentation/release-notes/21.02 [3] https://genode.org/documentation/release-notes/21.05 [4] https://genode.org/documentation/release-notes/21.08 [5] https://genode.org/documentation/release-notes/21.11 >From my personal perspective, I've wholly enjoyed the work on the Pinephone as documented in my "Pine fun" article series at Genodians.org [6]. Much of the enjoyment came from the _process_, in particular the co-development of the new DDE-Linux together with Stefan, the mutual cross-validation of ideas and code, and our joint sense of great care. Granted, feature-wise, I missed my original goal of being able to issue phone calls with Genode on the Pinephone by now. But the collateral effects of the work in terms of tooling (dts_extract), interfaces (Genode-C-API), and documentation ("Genode Platforms") deserved the attention they got. [6] https://genodians.org What's up for next year? ------------------------ I'm in full swing with the Pinephone. So I will keep moving full steam ahead. With the touchscreen and display tamed now, the next topics are telephony, mobile-data connectivity, Sculpt, browser, and a simple user interface. Over the year, I will increasingly focus on non-functional aspects as well, in particular power management (battery life) and quality of service (UI latency, audio latency). By the end of the year, I want to be able to casually use a video-chat solution like Jitsi on the phone. Besides the Pinephone, I am planning to simplify and solidify Genode's base framework by gradually removing complexity (like C++ exceptions [7]), increasing the strictness of the coding style (like the aftermath of [8]), and attending the most-neglected corners of our issue tracker [9]. [7] https://genodians.org/nfeske/2021-11-26-attempt-no-exceptions [8] https://genodians.org/nfeske/2021-12-07-wconversion [9] https://github.com/genodelabs/genode/issues What about you? --------------- My point of view outlined above is only one way to look at the picture. Now I would be interested in your perspective! What's your reflection of Genode's past year? What are the topics you deem as most interesting to work on? Do you already have tangible plans you can share with us? Are there road blocks that stand in the way of your plans? What is your vision of using Genode at the end of 2022? I hope that this posting spawns a fruitful discussion of potential topics for the next episode. Please be considerate to avoid dropping mere proposals or wish lists. It's best to present suggestions together with actionable steps that you are willing to take. In mid of January, I am going to update the official road map. Cheers, Norman -- Dr.-Ing. Norman Feske Genode Labs https://www.genode-labs.com · https://genode.org Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth

11 36

Intel's Memory Protection Keys
by Sid Agrawal 31 Jan '22

31 Jan '22

Hi, I want to use Intel's MPK in my project with Genode. A grep of the source code did not yield anything obvious. >From what I know about Intel's MPKs, I should be able to piggyback on the kernel's capability to manipulate the page table. The "keys" reside in the top 4 bits of the translation in the PTE. So maybe I do not need any special support from Genode or seL4, for that matter. But I just wanted to run it by the group to see if I missed something. Or if someone has tried this before. Best, Sidhartha Agrawal sid-agrawal.ca

1 0

how to allow execution of bash scripts inside genode
by Alexander Tormasov 14 Jan '22

14 Jan '22

based on combination of tool_chain_auto.run and ssh.run I am able to run an instance with toolchain and login into it I am able to create file using vim with some commands, like this: echo ok if I try to execute it inside genode bash (eg in /home/build) - I do receive the following: in ssh client : bash-4.4# vim c bash-4.4# ls -la total 1 -rwx------ 1 root 0 1004 Jan 1 2022 c drwx------ 1 root 0 144 Jan 1 1970 depot bash-4.4# ./c bash: ./c: No such file or directory bash-4.4# bash -c ./c bash: ./c: No such file or directory bash-4.4# bash -c c bash: c: command not found or, if I add this to beginning #!/bin/sh echo ok bash: ./c: /bin/bash: bad interpreter: No such file or directory in log: [init -> /bin/bash -> 12] Error: ROM-session creation failed (ram_quota=6144, cap_quota=3, label="/home/build/c") [init -> /bin/bash -> 12] Error: Could not open ROM session for "/home/build/c" [init -> /bin/bash -> 12] Warning: execve: executable binary inaccessible as ROM module I suspect that I need to allow execution via dynamic run file. or may be I need to add some attributes in run file for this dir? Sincerely, Alexander

2 3

lwext4_fs test fail on ftruncate()
by Alexander Tormasov 13 Jan '22

13 Jan '22

I am trying to run lwext4_fs.run from world repo it gives the following error on linux/x86_64 ... [init -> test-libc_vfs] calling open(file_name4, O_CREAT | O_WRONLY) file_name=test4.tst [init -> test-libc_vfs] open(file_name4, O_CREAT | O_WRONLY) succeeded [init -> test-libc_vfs] calling ftruncate(fd, 100) [init -> test-libc_vfs] ftruncate(fd, 100) succeeded [init -> test-libc_vfs] calling close(fd) [init -> test-libc_vfs] close(fd) succeeded [init -> test-libc_vfs] calling stat(file_name4, &stat_buf) file_name=test4.tst [init -> test-libc_vfs] stat(file_name4, &stat_buf) failed, ret=0, errno=2 [init -> test-libc_vfs] Error: Uncaught exception of type 'Test_failed' [init -> test-libc_vfs] Warning: abort called - thread: ep ^Cmake: *** [Makefile:396: run/lwext4_fs] Error 130 can this error could be ignored? is lwext4 port somhow usable in general on 21.11? can I utilise it as a main fs inside genode to store compilation results in rw mode? Sincerely, Alexander

2 2

native gcc and external image mount example
by Alexander Tormasov 13 Jan '22

13 Jan '22

Hi! I am considering native (not cross compilation) support of golang inside genode. I made two first steps - port golang runtime using cross compilation via genode toolchain and check it with tcp support inside qemu x86_64/arm64. Now I want to run gccgo and related binutils (files from gcc like gccgo/go1/as/ld/etc) natively inside genode using qemu. do anyone have an example to run gcc inside genode? I remember that sometimes ago it was mentioned, while I did not find any test in repos/ I can build it using "make noux/gcc_x86» - but any examples how to utilise it inside genode image? like to compile and run «helloworld.c» inside? Also I need a persistent partition (eg linux loopback image) to be mounted via qemu. Do we have any examples how I can create such image, attach it to qemu, mount/umount it inside? I need this to keep compiled code inside qemu. preferably with file system with journalling support if any (ext3/4? not talking about btrfs) Alexander

3 8

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users