users

users@lists.genode.org

2285 discussions

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

Porting GNU libraries
by Aditya Kousik 05 May '15

05 May '15

Hello I'm trying to port libsqlite which is a dependency for another application (which I'm also trying to port). The build of this library is a one line compilation. But it requires a -ldl flag for dynamically loaded libraries (it uses dlopen() and dlfcn.h). I've tried getting the source for this, to no avail so far. Is there a way to link libraries directly from the linux host. Intuitively, that doesn't seem to work. Thoughts on how to move forward? Thanks and regards, Aditya

6 15

Run scripts not working on Raspberry Pi
by Aditya Kousik 05 May '15

05 May '15

Hello, I built a sample run/printf and run/nested_init to test serial output on the Raspberry Pi. Funnily enough, I got a console output when I built it a week back. The same compilation doesn't work now. I don't see any output at all. What could have gone wrong? (I copied the boot folder of the default firmware to my SD card) I pasted genode.img after running the objcopy command as well. Any input would be helpful. Thanks Aditya

2 2

Non-standard display resolution support
by Nobody III 30 Apr '15

30 Apr '15

I would like Genode to support my monitor's native resolution, 1920x1080, which my BIOS doesn't support. I have a Radeon HD 5450 video card. What is the easiest way to support the resolution? Do I need a Radeon HD-specific driver, or can I add additional display modes through another means?

1 0

Re: Develop TrustZone for iMX6
by bharath 30 Apr '15

30 Apr '15

> Hello bharath, > On April 29, 2015 6:55:01 AM CEST, bharath <bharathkr.bojja@...52...> wrote: >> please, follow Martin's advice and do not use the outdated RUN_OPT >> setting when using u-boot. If you replace it like this: > >> RUN_OPT += --include image/uboot > >> You will even find a uImage in your 'var/run/tz_vmm_mx6' folder. >Sorry I >> was not recognizing that you were using the old, invalid setting for >> RUN_OPT when writing my last mail. > Please follow Stefan's advice and extend RUN_OPT as described above. > Greets > -- > Christian Helmuth > Genode Labs Hello Helmuth, Yes I have extend RUN_OPT += --include image/uboot before giving make run/tz_vmm_mx6. Still uImage is not found in var/run/tz_vmm_mx6/ folder. Only image.elf and boot_modules.s were found in the folder. Please suggest. Thanks, Bharath

3 2

AHCI not working on real hardware
by Nobody III 30 Apr '15

30 Apr '15

Months ago, I had issues with the AHCI driver not working with part_blk. I tested a simple scenario with AHCI and a non-writing version of test-blk-cli. It didn't work (test-blk-cli-nowrite didn't give any output), but the AHCI driver gave its normal output. I tested a slightly modified config using rom_blk instead of ahci, and it worked just fine. Any ideas about what might be wrong?

2 3

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users