users

users@lists.genode.org

1 participants
2310 discussions

Design review for Native File system support - ext2
by Janani k 13 May '15

13 May '15

Hi, I am trying to port an existing native file-system(ext2) to Genode. So, I started porting this in the same place as rump_ext2 so that this can be set up fast to get quick running of the code. I designed it so as to call our native implementation calls instead of rump calls. To get the block sessions working and basic ext2 functionalities checking, I did quick code work to get it up and running. Sorry didn't follow the Genode specifications in coding. I want get suggestions & feedback's on this design and then I will align code to Genode specifications and move further. I am attaching the patch file. And I will list out major things I did in this work to give an overview. 1. It contains totally 3 parts. one which has all Genode related code which used to deal with rump also and handle registry things, so I kept this one as is so as to utilize it again. Second part File-system interface which serves as an abstraction of all the extended file-system types(ext2,ext3,ext4). Third part ext2 this contains the implementation of ext2. 2. All the calls previously to rump will be replaced by this new ext2 implementation. 3. Till now I could get it up running. In this, I linked creating directory or file code, remove a directory, listing of directory entries works. This you can see in main.cc file comments where this gets called. 4. Since I used ram data space the changes I did to the ext2.raw are not persistent. So I removed 3 lines in rump_ext2.run which create ext2.raw file again and again when this demo is run and instead I made it persistent. And I created some directories before hand in this. 5. For mounting it and checking contents use :: $ sudo mount -t ext2 -o loop,rw ./build.lx/bin/ext2.raw /mnt/ext2/ 6 For checking ext2 structure from outside use :: $ dumpe2fs build.lx/bin/ext2.raw Entire development is done running Genode on linux. For compiling and running ext2 use :: make -C build.lx/ run/rump_ext2 Thanks & regards, Janan

3 6

Improved window management
by Nobody III 13 May '15

13 May '15

Genode's window manager is very simple and lacks the following functionality: minimize/maximize/restore close window force kill It seems like the nitpicker service should be modified to support whichever of these functions it doesn't already. I'm unsure of how exactly force kill could be implemented--especially since not all users can be considered trustworthy. How do we fit users into the Genode security model? Users need to be given reasonable authority--complete authority in some cases. Any ideas?

2 4

Segfaults while calling Rpc_methods from client.h
by Aditya Kousik 12 May '15

12 May '15

Hello, Calling Rpc_<method>(args, ...) from Session_client causes segfault 11 when I try to pass anything other than a default datatype. I'm trying to pass an object of std::string, and it prints the string inside the <method> just fine, but when I make the call<Rpc_method>(st), it breaks. I tried working around this by creating a char* pointer instead. Print inside this method works and the rpc to session_component method also reaches, but whatever pointer I receive in the session_component is garbage data. How do I fix the RPC calls? Thanks, Aditya.

2 4

Dynamic linker error?
by Aditya Kousik 11 May '15

11 May '15

Hello, I'm trying to run a modified version of the hello_tutorial to test some libraries that I've recently ported. While I am able to build the binaries, and the boot image, init gives me the following error:

3 8

libcrypto/openssl : Which to use
by Aditya Kousik 09 May '15

09 May '15

To use libcrypto++, I found online that openssl will suffice. But while there is a port file for both libraries, there's no openssl.mk in lib/mk. Grepping the mk files I saw a line in libcrypto.inc which calls select_from_ports,openssl. So, is it the case that we prepare_port for openssl but add LIBS += libcrypto in our target.mk files? Thanks, Aditya

4 16

Re: Help with RaspberryPI USB on Fiasco.OC+Genode
by Reinier Millo Sánchez 07 May '15

07 May '15

Hi Norman I've been testing the USB driver in the RaspberryPI assigning priorities to each processes, but continue losing events. I have also tried changing the top value of cnt, but still the same result. I will continue doing some independent tests of the driver, mouse, keyboard and network independently. If you have any other suggestions? Best regards > Hi Reinier, > >> I have checked the kernel warning about USB's irq, using base_hw and >> base_foc, and have noted (per 8000 need_trigger_sof calls): >> (base_hw) - kicked 5, filtered aproximatelly 7600 and triggered >> aproximatelly 250 IRQs >> (base_foc) - kicked 3, filtered aproximatelly 7700 and triggered >> aproximatelly 140 IRQs >> There is a big difference in the triggered and filtered irqs. Any idea >> or suggestion?? > SOF interrupts are filtered unless the USB driver explicitly calls > 'schedule_sof_interrupt' and thereby propagates the next > frame-to-deliver via the GUID scratch register to the kernel. If, for > some reason, the USB driver is not scheduled timely enough, it might > miss the scheduling of a frame. This can happen if the USB driver runs > concurrently with another program that consumes a lot of CPU time. In > such a scenario, the scheduling of the USB driver depends on the > scheduler of the kernel. Since Fiasco.OC has a different scheduler than > base-hw, the effect ultimately depends on the kernel. One way to work > around it is to run the USB driver at a high priority. I.e., by > assigning a priority of "-1" to all components except for the USB and > timer drivers. For an example of how to use priorities in Genode, please > have a look at the gems/run/wm.run script. Look out for the > 'prio_levels' and 'priority' attributes. > > Note that this is just a guess. If the effect also occurs when not > running a CPU hungry program besides the USB driver, the guess is most > likely wrong. > >> I don't understand the use of cnt variable (need_trigger_sof). Why use >> 160 as counter top?? > This mechanism ensures that an interrupt is delivered every now and then > - even if the USB driver has not scheduled a frame. There are 8000 IRQs > coming in per second. With the value of 160, an interrupt is forwarded > to the USB driver every 20 milliseconds, which appeared to me as a > low-enough rate to have a negligible effect on the system's performance. > >> I'm waiting fix this problem to upload the patch to GitHub. >> I have tested the USB driver using two versions of Fiasco.OC, the >> default Fiasco.OC version (https://github.com/ssumpf/foc.git) and a more >> recently version (https://github.com/skalk/foc.git), but have the same >> problem. >> >> Furthermore I have tested succesfully the support of Genode with >> Fiasco.OC as microkernel for RaspberryPI using the autopilot tests: >> (util_mmio,ldso,timer,lwip,rm_fault,rom_blk,tar_rom,noux,libc_ffat,libc_vfs,timed_semaphore,signal,sub_rm,python,thread_join,failsafe,noux_tool_chain_auto, >> seoul-auto,resource_request,xml_generator,blk_cache,rump_ext2,thread,pthread, >> netperf_lwip, netperf_lwip_usb30, netperf_lwip_bridge, netperf_lxip, >> netperf_lxip_usb30, netperf_lxip_bridge) >> Now I'm trying to test l4linux and other more complex examples. > -- Lic. Reinier Millo SÃ¡nchez Centro de Estudios de InformÃ¡tica Universidad Central "Marta Abreu" de Las Villas Carretera a CamajuanÃ Km 5 1/2 Santa Clara, Villa Clara, Cuba CP 54830 "antes de discutir ... respira; antes de hablar ... escucha; antes de escribir ... piensa; antes de herir ... siente; antes de rendirte ... intenta; antes de morir ... vive"

2 5

Strictness in syntax of Makefiles
by Aditya Kousik 07 May '15

07 May '15

Hello all, I'm still learning my ropes at writing Makefiles and make descriptor files. I can't seem to get the filter-out option working properly. Despite filling out FILTER_OUT = somefile.cpp SRC_CC += $(filter-out $(FILTER_OUT),$(notdir $(wildcard $(SRC_DIR)/subdirs/*.cpp))) the build system continues to add these to the set of files to be built. Next, most of the screen is bombarded with unused-variable and unused-function warnings. I tried to suppress them as mentioned in the build system manual with the CC_OPT += -Wunused-function -Wunused-variable Build system still throws the above warnings. Are the spacing and/or use of commas crucial to the build system? Thanks and regards, Aditya

1 1

ARM TrustZone Demo in i.MX53 SABRE Tablet
by JaeminPark 06 May '15

06 May '15

Today, I got Freescale i.MX53 SABRE Tablet and I’d like to make demo as described “Shortcut” in README (https://github.com/skalk/genode/blob/i.MX53_tablet_demo/os/src/server/vmm/i… <https://github.com/skalk/genode/blob/i.MX53_tablet_demo/os/src/server/vmm/i…>). After boot-up the tablet, I failed to load image.bin into the device. It could be a ridiculous question, but please tell me which point I did wrong. My procedure was as follows : 1) Download image.bin from URL in https://github.com/skalk/genode/blob/i.MX53_tablet_demo/os/src/server/vmm/i… <https://github.com/skalk/genode/blob/i.MX53_tablet_demo/os/src/server/vmm/i…> 2) Setup sdcard as follows : (sdcard is mounted in /dev/sdb) sudo parted /dev/sdb --script mklabel msdos sudo parted /dev/sdb --script mkpart primary ext4 5M 100% sudo dd if=$(U-BOOT-DIR)/u-boot.imx of=/dev/sdb seek=2 bs=512 conv=fsync sudo fdisk -l /dev/sdb sudo mkfs.ext4 /dev/sdb1 sudo udisks --mount /dev/sdb1 sudo cp image.bin /media/371a8dd7-58c1-4beb-aed3-cb18fb40f7eb/ sudo udisks –-unmount /dev/sdb1 3) insert sdcard into MMC adapter, and then insert MMC (sdcard) into the tablet 4) After power up the tablet, U-BOOT prompt (u-boot from MMC is not used), I did as follows : MX53-SMD U-Boot > ext2load mmc 0:1 0x70800000 image.bin (hang forever) So, I couldn’t load image.bin into the tablet, thus I couldn’t boot demo image in the tablet. If possible, (for the beginner in this field) please let me know full procedures to set up the MMC (sdcard) for this demo.

2 2

Memory access in tz_vmm
by chirag garg 06 May '15

06 May '15

Hi all, I am working on i.MX6 Sabre Lite board. Total ram available in this board is 1GB (0x10000000-0x50000000). To demonstrate TrustZone, i tried to run tz_vmm with Genode in Normal world also. So i divide the available RAM as secure (0x10000000-0x30000000) and normal (0x30000000-0x50000000). On the normal world ram, secure world also have the access which is programmed in tzasc (tsc_380.h) driver. I tried to access normal world region (0x40000000) through IO_MEM from the secure world but when i am trying to create the iomem object for 0x40000000 region in tz_vmm. It gives me the following fault. I/O memory [40000000,50000000) not available Local MMIO mapping failed! [init -> tz_vmm] C++ runtime: Genode::Parent::Service_denied I have included this region in mmio_regions inside the platform_support.cc also. What could be the reason for this fault. ? Log : Starting kernel ... kernel initialized Core virtual memory allocator --------------------- Allocator 10135250 dump: Block: [00001000,00002000) size=00001000 avail=00000000 max_avail=00000000 Block: [00002000,00003000) size=00001000 avail=00000000 max_avail=00000000 Block: [00003000,00004000) size=00001000 avail=00000000 max_avail=ef832000 Block: [00004000,00005000) size=00001000 avail=00000000 max_avail=00000000 Block: [00005000,00006000) size=00001000 avail=00000000 max_avail=ef832000 Block: [00006000,00007000) size=00001000 avail=00000000 max_avail=00000000 Block: [00007000,10001000) size=0fffa000 avail=0fffa000 max_avail=ef832000 Block: [107be000,ffff0000) size=ef832000 avail=ef832000 max_avail=ef832000 => mem_size=4286783488 (4088 MB) / mem_avail=4286758912 (4088 MB) RAM memory allocator --------------------- Allocator 101341f4 dump: Block: [10000000,10001000) size=00001000 avail=00000000 max_avail=00000000 Block: [107be000,107bf000) size=00001000 avail=00000000 max_avail=00000000 Block: [107bf000,107c0000) size=00001000 avail=00000000 max_avail=00000000 Block: [107c0000,107c1000) size=00001000 avail=00000000 max_avail=0f83d000 Block: [107c1000,107c2000) size=00001000 avail=00000000 max_avail=00000000 Block: [107c2000,107c3000) size=00001000 avail=00000000 max_avail=0f83d000 Block: [107c3000,20000000) size=0f83d000 avail=0f83d000 max_avail=0f83d000 => mem_size=260321280 (248 MB) / mem_avail=260296704 (248 MB) IO memory allocator ------------------- Allocator 101362b8 dump: Block: [00000000,10000000) size=10000000 avail=10000000 max_avail=20000000 Block: [30000000,50000000) size=20000000 avail=20000000 max_avail=20000000 => mem_size=805306368 (768 MB) / mem_avail=805306368 (768 MB) IRQ allocator ------------------- Allocator 10137314 dump: Block: [00000000,00000400) size=00000400 avail=00000400 max_avail=00000400 => mem_size=1024 (0 MB) / mem_avail=1024 (0 MB) Genode 15.02 int main(): --- create local services --- int main(): --- start init --- : : [init] child "tz_vmm" [init] RAM quota: 3932160 [init] ELF binary: tz_vmm [init] priority: 0 I/O memory [40000000,50000000) not available Local MMIO mapping failed! [init -> tz_vmm] C++ runtime: Genode::Parent::Service_denied [init -> tz_vmm] void* abort(): abort called Regards Chirag Garg IIT Madras , Rise Lab

2 2

i.MX53 TrustZone cross world interactions
by Ofer Hasson 06 May '15

06 May '15

Hey, I'm using Freescale i,MX53 QSB. I'm trying to transfer data between the secure world and the normal world, at the moment, I manage to transfer data using the CPU registers (using Linux kernel module I wrote), and I manage to receive them on the hypervisor. I would like to transfer larger structure, for example a memory buffer, I'm allocating in the normal world kernel a buffer, I send the physical address (using __pa) to the secure world. My question is, how in the Genode hypervisor I can access this memory address ? Any idea will be helpful Thanks Ofer

3 3

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users