users

users@lists.genode.org

2285 discussions

hash sum check failed for atari800 port
by Wouter van Oijen 22 Nov '18

22 Nov '18

Hello, I was just trying to prepare the "atari800" port (from genode-world) but the hash sum check failed (see logging below). I can easily fix the hash to make it work at my system, but I'm surprised that the current hash is only 40 characters (instead of 64 characters that are expected for a sha256sum). If it should be a sha256sum then the port will obviously not work for anybody, and should fail in any automated regression test. I have the feeling I'm missing something here... If I'll update the hash, will that break the build on another system? Why is the current hash only 40 characters long? Regards, Wouter LOGGING: vagrant@ubuntu-bionic:~/genode$ ./tool/ports/prepare_port atari800 atari800 update src/app/atari800 Error: Hash sum check for xformer failed /home/vagrant/genode/tool/ports/mk/install.mk:197: recipe for target 'xformer.file' failed make[2]: *** [xformer.file] Error 1 /home/vagrant/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 'atari800' failed make: *** [atari800] Error 2 vagrant@ubuntu-bionic:~/genode$ grep xformer repos/world/ports/atari800.port DOWNLOADS := atari800.git xformer.archive URL(xformer) := http://prdownloads.sf.net/atari800/xf25.zip SHA(xformer) := e65923fd3a4845c1c72d1677db0fda969187eb42 DIR(xformer) := src/app/atari800/xformer vagrant@ubuntu-bionic:~/genode$ sha256sum -b - < contrib/atari800-*/xf25.zip 98ae0ad10413dd6f35ed80f5662dba6d790def70c7829046e52012a03b574b8a *-

2 3

use sd_card driver in imx53_qsb_tz platform
by lzSun 22 Nov '18

22 Nov '18

Hello ,Genode Community I have written an email before which has a similar question to this email, but I haven't receive any response yet. Because solving this problem is very important to my work , so I decide to put it out again and hope to receive some help. The thing is I want to use sd_card driver in i.mx53 quick start board with trustzone. I know Martin said " you should not run 'sd_card' with Genodes Trustzone feature-set (platform 'hw_imx53_qsb_tz'), because Genode will then declare the ESDHC-IRQ to be unsecure to enable Linux to drive SD cards."[1]. But unfortunately, my work is just needs to use sd_card driver with trustzone. So I wonder is this possible? I am now using the latest Genode version(18.08). Form my point of view, if ESDHC-IRQ is still to be unsecure, this's two ways to solve this problem: 1. set ESDHC-IRQ to be secure. In this way ,Genode OS in secure world can directly drive the SD card. Also, the linux in the normal world must be modified to the stat that could not drive SD card. If this is right, could you provide me some hint to this method? 2. forward the sd card request to linux in the normal world.So all the sd card operation will be performed by linux. However, I think this will introduce more world switch between two worlds and you should provide a linux driver to serve for Genode's file operation. I will be grateful if you could give me some help, thanks a lot! ------------------ Best wishes

2 1

Re: optimization levels in genode
by lzSun 22 Nov '18

22 Nov '18

Hi,Stefan >The run-script you are using is an automated test, which remote-controls the test board by using >Tcl/expect. Therefore, if you use the unmodified run-script to control the Linux instance via serial line >manually, it won't work. The 'expect' mechanism holds control over the serial line I/O. That means >everything you type in will not be sent to the other side. >If you want to use the test-scenario in a non-automated fashion, you either replace the very first >"run_genode_until ..." with "run_genode_until forever", or you simply do not use the run-script to >connect to your test-board, but use a terminal emulator like "picocom" directly. >Please, first evaluate whether it accepts input events with "direct" serial-line access. I'm wondering that >you see a _blinking_ terminal but no input events are shown. Sorry,but I can not fully understand what you mean. Now I am using i.mx53 quick start borad and the borad is connected to my PC through it's UART DB9 connector. I managed this through a USB to serial converter. Also , I use a serial debug tool that is "minicom" to get the output of the board and input my command.Through the minicom , I can input command to linux , such as "ls mkdir wget... etc" , also ,I can open the kernel module that I implemented to issue "smc" to switch to tz_vmm to execute some code. It works fine when I use the genode 18.05. However ,when I changed to the hard-float version that you provided , I found this odd problem. Also, I have found that even I din't input any command the linux will get stuck after a period of time after startup. I have tried to replace the "run_genode_until..." with "run_genode_until forever" but the result haven't changed. This problem has bothered me for a few days, I will be very grateful if you could give me some help. ------------------ Best wishes

2 1

Hello, world!
by Wouter van Oijen 20 Nov '18

20 Nov '18

This is my first message to this mailinglist, so Hello, world! ;-) I'm new to Genode and trying to learn the basics. Our goal is to develop a setup that demonstrates the application of Genode to create a powerful separation between critical and non-critical code. Genode looks really promising for this! To get a better understanding of the build environment and Noux in particular, I'm trying to run a simple "Hello, world" application on Noux. However, I don't know how to achieve this yet. The Porting Guide [1] doesn't help me much, as I don't think the example is really applicable to my use case. It raises the following questions: * Is it necessary to create a "port file" for any application I want to run on Noux (even if it's just a simple application consisting of one source file that is already part of the build directory)? * How can I add files generated by the Genode build system to the Noux environment, so that I can call my "hello" binary from bash in Noux? Any help or a working example would be much appreciated. It would help me much if someone could explain me how to extend the "run/noux_bash" [2] example to include my hello-world application, if this is right way to continue. [1] https://genode.org/documentation/developer-resources/porting_noux_packages [2] https://github.com/genodelabs/genode/blob/master/repos/ports/run/noux_bash.… Kind regards, Wouter van Oijen Lead Designer T +31(0)182594000 | E wouter.van.oijen(a)technolution.nl A Burgemeester Jamessingel 1, P.O. Box 2013, 2800 BD Gouda, The Netherlands | W technolution.com This e-mail is intended exclusively for the addressee(s), and may not be passed on to, or made available for use by any person other than the addressee(s). Technolution B.V. rules out any and every liability resulting from any electronic transmission.

2 4

sd_card_drv quota exceeded for Panda
by Matthias Adams 20 Nov '18

20 Nov '18

Hello everybody, I am a student from Munich and I am currently working on building some bare metal code for the Pandaboard M3 co-processor, while a Genode system, using the L4 kernel, is running on the A9. In order to load the bare metal application form the SD card I read some blocks using the sd_card driver. During the initialization of the driver the following output is produced: init] child "sd_card_drv" [init] RAM quota: 104722432 [init] ELF binary [init] priority: 0 [init] provides service Block [init] ram_quota=104722432 Warning: Quota exceeded! amount=81920, size=4069, consumed=81920 The Quota waring is coming from the allocator guard in 'repos/base/include/base/allocator_guard.h', but I have no idea which quota could have been exceeded, because the ram_quota should definitely be enough when reading only a few KiB. I would be grateful for any hints or ideas on why this is happening and how to solve this issue. Thanks, Matthias Adams

2 1

sd_card driver for i.mx53 time out
by lzSun 19 Nov '18

19 Nov '18

Hi,all I mentioned in previous email that I am trying to use the file system in the tz_vmm demo and use the SD card as a storage device which is formatted as ext2. Now I have included sd_card driver, part_blk, rump_fs etc. in the tz_vmm.run script which I show it in [1]. But when I try to create a file in tz_vmm, I get this error: [init -> tz_vmm] --- start test --- [init -> tz_vmm] current working directory : / [init -> sd_card_drv] Error: wait till issuing a new command is allowed timed out I have no idea how to fix this problem, so I want to ask for your help. I also provide my test code in [2]. In addition, I saw in this link: https://sourceforge.net/p/genode/mailman/genode-main/thread/CABDqRrJei5p5Jy… , Martin said "Be aware, that you should not run 'sd_card' with Genodes Trustzone feature-set (platform 'hw_imx53_qsb_tz'), because Genode will then declare the ESDHC-IRQ to be unsecure to enable Linux to drive SD cards. Use the platform 'hw_imx53_qsb' instead when issuing 'create_builddir'. I know he is targeting genode 15.08, so I want to know how about this situation in Genode's latest version? If genode can directly drive the SD card now , does this meas the SD card can not accessed by linux because the ESDHC-IRQ is configured to be secure ? Thanks for your help in advance! [1] part of tz_vmm.run script: <start name="platform_drv"> <resource name="RAM" quantum="3M"/> <provides> <service name="Regulator"/> <service name="Platform"/> </provides> </start> <start name="sd_card_drv"> <resource name="RAM" quantum="3M"/> <provides><service name="Block"/></provides> </start> <start name="part_blk"> <resource name="RAM" quantum="10M" /> <provides><service name="Block" /></provides> <route> <service name="Block"><child name="sd_card_drv" /></service> <any-service><parent/><any-child/></any-service> </route> <config> <policy label_prefix="rump_fs" partition="1" writeable="yes"/> </config> </start> <start name="rump_fs" caps="200"> <resource name="RAM" quantum="16M"/> <provides><service name="File_system"/></provides> <config fs="ext2fs"> <policy label_prefix="" root="/" writeable="yes"/> </config> <route> <service name="Block"> <child name="part_blk"/> </service> <any-service> <parent/> <any-child/> </any-service> </route> </start> <start name="tz_vmm" caps="400"> <resource name="RAM" quantum="160M"/> <config> <vfs> <tar name="darknet.tar"/> <dir name="dev"> <log/> <inline name="rtc">2000-01-01 00:00</inline> </dir> <fs/> </vfs> <libc stdout="/dev/log" stderr="/dev/log" rtc="/dev/rtc" /> </config> [2] test code: int fd; ssize_t n; off_t offset; static char buf[16384]; static char str[] = "hello,world!"; printf("--- start test ---\n"); getcwd(buf, sizeof(buf)); printf("current working directory : %s\n", buf); fd = open("file.txt",O_WRONLY|O_CREAT); if (fd == -1) { perror("open"); return; } n = write(fd, str, sizeof (str)); printf("write %d bytes.\n",n); close(fd); printf("--- test finished ---\n"); ------------------ Best wishes

1 0

Re: optimization levels in genode
by lzSun 16 Nov '18

16 Nov '18

Hello,Stefan >Just out of curiousity, are you still using the tz_vmm.run script? And >if so, did you changed it to do: "run_genode_until forever", or does >it still use "run_genode_until <some_condition> <timeout>"? Yes, I am using tz_vmm.run script, and it sitll use the condition like this: # execute and wait for console run_genode_until {.*\/ #.*} 220 set serial_id [output_spawn_id] # wait for network to settle down send -i $serial_id "sleep 5\n" # test network send -i $serial_id "wget http://genode.org/\n" run_genode_until "Connecting to genode.org" 10 $serial_id run_genode_until {.*\/ #.*} 30 $serial_id Also, I tried to modify it to "run_genode_until forever" but I encountered the same problem.Please correct me if there are something wrong with those settings. ------------------ Best wishes

2 1

Request for feedback about "Genode Component Public License"
by Norman Feske 16 Nov '18

16 Nov '18

Dear Genode community, today, I'd like to present an idea that developed over the past three years during discussions with various contributors and users about the prospect of a Genode-based business ecosystem. As you certainly know, Genode Labs pursues a dual-licensing business model, which allows us to fund the development of Genode as an independent team. Our licensing business is enabled by the combination of the AGPLv3 as a strong copyleft license with the library-like nature of Genode. Until today, this model is limited to our company. To foster a sustainable ecosystem around Genode, we wish to enable others to pursue a similar business model while maintaining the spirit of open collaboration and free software. We eventually crafted a new license called "Genode Component Public License" (Genode CPL) specifically for components developed by 3rd parties, outside of Genode Labs. Please find the rationale, license text, and FAQ under the following link (which is not officially reachable from the website yet): https://genode.org/documentation/articles/component_public_license Before I go forward with publicly announcing the article at genode.org, I would very much appreciate your feedback, in particular: * How does the rationale and approach given in the description resonate with you? * Do you see any potential for controversies or uncertainties? * If you are in the business of producing proprietary software, would the Genode CPL create an incentive to open up your products without putting your business at risk? I'm very much aware that licensing is a delicate topic. To avoid hot-heated controversies and mis-interpretation, I encourage everyone to carefully read the "Questions and Answers" Section of the article before commenting. I'm looking forward to your feedback! Best regards 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

Re: rump_fs mount file system failed
by lzSun 14 Nov '18

14 Nov '18

Hi,Christian Thanks for your help. Now, the file system can be mounted. However, there's still errors in my test.That is File_system-session creation failed. Here's the output message: [init -> part_blk] Partition 1: LBA 10240 (7733248 blocks) type: 83 [init] child "part_blk" announces service "Block" [init -> part_blk] session opened at partition 1 for 'rump_fs -> ' [init -> rump_fs] Using ext2fs as file system [init -> rump_fs] asserting rump kernel 14253 KB of RAM [init -> rump_fs] rump: /genode: file system not clean; please fsck(8) [init] child "rump_fs" announces service "File_system" [init -> test-file] Error: File_system-session creation failed (ram_quota=163840, cap_quota=12, tx_buf_size=131072, label="", root="", writeable=1) [init -> test-file] Error: failed to create <fs> VFS node [init -> test-file] --- start test --- [init -> test-file] Error: plugin()->open("file.txt") failed [init -> test-file] open [init -> test-file] : [init -> test-file] No such file or directory I don't know whether it's because there's some mistake in my configuration. Can you give me some hint? The run scripts is just like that I showed you before. And here's test file: #include <base/attached_rom_dataspace.h> #include <libc/component.h> /* libc includes */ #include <dirent.h> #include <fcntl.h> #include <stdio.h> #include <stdlib.h> #include <sys/stat.h> #include <sys/uio.h> #include <unistd.h> #include <string.h> #include <errno.h> static void test() { int fd; ssize_t n; off_t offset; static char buf[100]; static char str[] = "hello,world!"; printf("--- start test ---\n"); fd = open("file.txt",O_WRONLY|O_CREAT); if (fd == -1) { perror("open"); return; } n = write(fd, str, sizeof (str)); printf("write %d bytes.\n",n); close(fd); fd = open("file.txt", O_RDONLY); if (fd == -1) { perror("read open"); return; } n = read(fd, buf, sizeof (str)); printf("read %d bytes: ", n); for (size_t i = 0; i < sizeof (str); i++) printf("%c ", buf[i]); printf("\n"); close(fd); printf("--- test finished ---\n"); return; } struct Main { Main(Genode::Env &env) { Genode::Attached_rom_dataspace config_rom { env, "config" }; Libc::with_libc([&] () { test(); }); env.parent().exit(0); } }; void Libc::Component::construct(Libc::Env &env) { static Main main(env); } ------------------ Best wishes

2 1

Re: optimization levels in genode
by lzSun 14 Nov '18

14 Nov '18

Hello,Stefan Thanks for your effort! I have followed your advice to check to genode 18.08. I used the patch you provided to me: https://github.com/skalk/genode/tree/imx53-hf-neon At first, I just build the whole project as before. But there's an error at building stage: can't read "cc_arch": no such variable while executing "exec [cross_dev_prefix]gcc -print-libgcc-file-name {*}$cc_arch" (procedure "build_core" line 8) invoked from within Then, I add the "cc_arch" to the file: ./tool/run/run, like this: set asm_src [generate_boot_modules_asm [run_dir]/genode $modules] + set cc_arch "-mfloat-abi=hard -mfpu=neon" # determine the libgcc set libgcc [exec [cross_dev_prefix]gcc -print-libgcc-file-name {*}$cc_arch] After this, I can build the whole project successfully and measure the performance gain on the i.mx53 qsb board. However, there's still a problem. When my program have done and return to normal world, the system seems stuck. From the serial port, I can only see the linux's cursor blinking but I couldn't input any command.The serial port's output message like this: [init -> tz_vmm] task finished in 14588ms / # There are no errors which just like a command have executed normally. Do you have any idea about this? Thank you very much. ------------------ Best wishes

2 1

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users