users

users@lists.genode.org

1 participants
2310 discussions

Re: SMC on i.mx6
by Vincent Raes 12 Mar '15

12 Mar '15

Hello, Thank you for the advice. I performed some checks to see in what exact mode the 'smc' instruction is executed and it is supervisor mode which is a privileged mode so no problems there. This leads me to believe there is indeed something which goes wrong with the instruction itself. Since I do not have access to a JTAG debugger, trying to find out what kind of exceptions are taken seems to be the best course of action. As for what stop means, it is literally just that, it stops. I get all the output I would expect right up to the point where the 'smc' instruction is issued. After that, nothing happens. At least not in the normal world since I have tried running a different thread next to the one that performs the hypercall. This secondary thread should periodically print some output but remains silent all the time. The secure world does not give any signs of life either which is the main reason I'm posting here. As far as I can see, this non responsiveness from the secure world may make it hard to get the base-hw kernel that's running there to actually see what exception is taken but I'll give it a shot. And I am using 2 different genode kernels in the secure and normal world. The normal kernel is treated like the Linux kernel in the tz_vmm example. I simply manually place the binary image of the kernel in the bin folder so the run script can package it with the secure kernel. Best regards, Vincent On 03/12/2015 07:48 AM, Stefan Kalkowski wrote: >Hello Vincent, > >On 03/11/2015 08:44 PM, Vincent Raes wrote: >> Hello, >> >> I am facing some issues with the 'SMC' instruction on the i.mx6. >> I can run the tz_vmm example, based on the veax9 code since they both >> have a TZASC, on the board and the world switch from secure to normal >> world works as intended. However when trying to switch back from normal >> world to secure world, execution simply hangs after the 'smc' assembler >> instruction. >> It was my understanding that this instruction should trap to the secure >> world and then the tz_vmm continues to work with that. In my case >> everything simply stops. > > That is right, the 'smc' instruction should provoke an exception taken > to the secure world's exception vector. Are you sure that the > instruction is executed in the normal world? What does stop mean? To get > a picture what is going on you either need to debug it via a JTAG > debugger, or by instrumenting the base-hw kernel of the secure-side to > see what kind of exceptions are taken. A good starting-point would be > the 'Kernel::Vm' class' 'exception' function. In Genode 15.02 you can > find this function in here: > > > https://github.com/genodelabs/genode/blob/master/repos/base-hw/src/core/spe… > > If you use another version, you have to grep for it. > >> >> I am using Genode in both normal and secure world since I'm mainly going >> for a proof of concept at this point. In order to execute the 'smc' >> instruction, I implemented a public syscall so the kernel will be the >> one to give the command which should ensure that the processor is in a >> privileged mode when completing the call. >> >> Should I perhaps add a new core service to base-hw to make the call to >> the kernel? And if so, are there any pointers as to how to go about >> this? The only real advantage this method offers as far as I can see is >> that it is better for Genode's access control policy and gives more >> control to which applications can call for the 'smc' instruction. >> > > The other possibility why it doesn't work would be that the 'smc' > instruction is not really executed in privileged mode. To be really sure > about it, you might print the CPSR register and have a look at the five > least-significant bits to validate the current mode. > > Having a public syscall that is not private to core should not be the > problem in here. That is more a question of the desired architecture, > but it should function in both ways. > > Just out of curiosity: if you use Genode on both sides of the TrustZone, > do you use one and the same kernel/core for both worlds? And if so, how > do you distinguish between both when booting the kernel? > > Regards > Stefan > >> >> Best regards, >> Vincent >> >> >> ------------------------------------------------------------------------------ >> Dive into the World of Parallel Programming The Go Parallel Website, sponsored >> by Intel and developed in partnership with Slashdot Media, is your hub for all >> things parallel software development, from weekly thought leadership blogs to >> news, videos, case studies, tutorials and more. Take a look and join the >> conversation now. http://goparallel.sourceforge.net/ >> >> >> >> _______________________________________________ >> genode-main mailing list >> genode-main@...52... >> https://lists.sourceforge.net/lists/listinfo/genode-main >> > > -- > Stefan Kalkowski > Genode Labs > > http://www.genode-labs.com/ · http://genode.org/

1 0

How can I port Genode to Exynos4 platform?
by pyl 12 Mar '15

12 Mar '15

Hi everyone, I can run genode(base-foc) on the Pandaboard ES, and now I want to port genode to Odroid-Q2(exynos4412), but there is no script to run , I don't know what need I do and how to configure genode to make it work on Odroid. Can you help me? Thanks Cooper

2 1

An in-depth look into the ARM virtualization extensions
by Stefan Kalkowski 12 Mar '15

12 Mar '15

Hello everyone, as already announced, Genode's recent release 15.02 introduces virtualization support for ARM to our custom base-hw kernel. Due to the brisk demand regarding our prior article about ARM TrustZone, we decided to provide in-depth information about our virtualization work again. If you are interested in this topic, you may consider the following article, which was published right now: http://genode.org/documentation/articles/arm_virtualization I hope it provides a helpful insight into ARM virtualization in general and our related work in particular. Best regards Stefan -- Stefan Kalkowski Genode Labs http://www.genode-labs.com/ · http://genode.org/

1 0

Progress of bringing Genode to seL4
by Norman Feske 12 Mar '15

12 Mar '15

Hello, I have written up the first of a series of articles about using seL4 as base platform for the Genode OS framework: http://genode.org/documentation/articles/sel4_part_1 I hope that it contains useful bits of information for novices of seL4 and Genode alike. Even though the article outlines a few hurdles, the overall experience of using the kernel had been overly positive so far. For those of you interested in tracking the progress of the ongoing work, here is the corresponding topic branch: https://github.com/nfeske/genode/commits/sel4 Cheers Norman -- Dr.-Ing. Norman Feske Genode Labs http://www.genode-labs.com · http://genode.org Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth

3 3

SMC on i.mx6
by Vincent Raes 12 Mar '15

12 Mar '15

Hello, I am facing some issues with the 'SMC' instruction on the i.mx6. I can run the tz_vmm example, based on the veax9 code since they both have a TZASC, on the board and the world switch from secure to normal world works as intended. However when trying to switch back from normal world to secure world, execution simply hangs after the 'smc' assembler instruction. It was my understanding that this instruction should trap to the secure world and then the tz_vmm continues to work with that. In my case everything simply stops. I am using Genode in both normal and secure world since I'm mainly going for a proof of concept at this point. In order to execute the 'smc' instruction, I implemented a public syscall so the kernel will be the one to give the command which should ensure that the processor is in a privileged mode when completing the call. Should I perhaps add a new core service to base-hw to make the call to the kernel? And if so, are there any pointers as to how to go about this? The only real advantage this method offers as far as I can see is that it is better for Genode's access control policy and gives more control to which applications can call for the 'smc' instruction. Best regards, Vincent

2 1

imx53 trustzone boot problem
by Ofer Hasson 12 Mar '15

12 Mar '15

Hello, I'm trying to boot genode on the secure world and linux on normal world of the iMX53-QSB board (freescale). The problem I'm encountering is: [init -> tz_vmm] Start virtual machine ... and then various errors, either it hangs, or undefined instruction, or Pagefault in core thread (pager_activation): ip=70055228 fault=20ef8904 failed to communicate thread event My steps leading to this: Downloaded http://genode.org/files/images/imx53_qsb/initrd.gz and cloned https://github.com/skalk/linux.git git checkout imx53-tz Kernel compile: using http://www.codesourcery.com/sgpp/lite/arm/portal/package4573/public/arm-non… make CROSS_COMPILE=/opt/codesourcery/bin/arm-none-linux-gnueabi- ARCH=arm imx5_android_tz_defconfig make CROSS_COMPILE=/opt/codesourcery/bin/arm-none-linux-gnueabi- ARCH=arm uImage copied the initrd and uImage into the micro sd card, changed the name of uImage into linux I cloned https://github.com/genodelabs/genode.git working on the master branch, current commit hash: 54ef87c Created build directory: tool/create_builddir hw_imx53_qsb_tz BUILD_DIR=hw_imx53_qsb_tz I added "SPECS += trustzone" to etc/specs.conf in the build directory Compiled: make run/tz_vmm Created u-boot image: /opt/gcc-linaro-arm-none-eabi-4.9-2014.09_linux/bin/arm-none-eabi-objcopy -O binary var/run/tz_vmm/image.elf var/run/tz_vmm/image.bin mkimage -A arm -O linux -T kernel -C none -a 0x70010000 -e 0x70010000 -d var/run/tz_vmm/image.bin var/run/tz_vmm/uImage Next, i copy the image to the micro sd card. On the u-boot shell: ext2load mmc 0:1 0x71000000 uImage bootm 0x71000000 OUTPUT: ## Booting kernel from Legacy Image at 71000000 ... Image Name: Image Type: ARM Linux Kernel Image (uncompressed) Data Size: 9355264 Bytes = 8.9 MiB Load Address: 70010000 Entry Point: 70010000 Verifying Checksum ... OK Loading Kernel Image ... OK Starting kernel ... kernel initialized Genode 15.02 <local changes> int main(): --- create local services --- int main(): --- start init --- In Rom_connection(): /home/hassonof/arm-tz/devel/boards/iMX53/genode-original/repos/base/include/rom_session/con7 filename = init, label = <NULL> int main(): transferred 246 MB to init int main(): --- init created, waiting for exit condition --- [init] Could not open file "ld.lib.so" [init] In Rom_connection(): /home/hassonof/arm-tz/devel/boards/iMX53/genode-original/repos/base/include/rom_sess7 [init] filename = config, label = <NULL> [init] parent provides [init] service "ROM" [init] service "RAM" [init] service "IRQ" [init] service "IO_MEM" [init] service "CAP" [init] service "PD" [init] service "RM" [init] service "CPU" [init] service "LOG" [init] service "SIGNAL" [init] service "VM" [init] In Rom_connection(): /home/hassonof/arm-tz/devel/boards/iMX53/genode-original/repos/base/include/rom_sess7 [init] filename = tz_vmm, label = tz_vmm [init] child "tz_vmm" [init] RAM quota: 3932160 [init] ELF binary: tz_vmm [init] priority: 0 [init -> tz_vmm] Start virtual machine ... and then errors described at the beginning. I was looking in the code and dumping some output to find the problem, at repos/os/src/server/tz_vmm/include/vm_base.h: _load_kernel() function, at lines #52-53 the memcpy crashes. the arguments being passed to him are: destination: 80008000 source: f0001000, size: 3203072 what am i doing wrong ? Thanks Ofer

2 2

base-hw: Cpu_lazy_state and alignment constraints
by Adrian-Ken Rueegsegger 10 Mar '15

10 Mar '15

Hello, While implementing lazy FPU loading/storing for the x86 base-hw port I came across an issue with regards to data structure alignment. The implementation is similar to the one of base-hw cortex A9 and uses the Cpu_lazy_state class to store the FPU state, see [1]. To use the fx[save|rstor] instructions, the storage area must be 16-byte aligned which I set as alignment of the Cpu_lazy_state class [2]. Checking the alignment of the data structure during runtime showed however, that the constraint is not correctly adhered to. I suspect that the Kernel::Cpu_job class, which contains the lazy state [3], is allocated without consideration for the required alignment. I have worked around the issue by manually ensuring the correct alignment [4] but I consider it a hack would like to get rid of it. Since I am not quite sure how Cpu_job is allocated and what its life cycle looks like, I would be glad if someone could give me some pointers to verify my suspicion and give a suggestion on how one would go about ensuring the proper alignment of all instances of Cpu_lazy_state. Regards, Adrian [1] - https://github.com/codelabs-ch/genode/blob/9a2a85c807a695a91088b6ab6e0d0de4… [2] - https://github.com/codelabs-ch/genode/blob/9a2a85c807a695a91088b6ab6e0d0de4… [3] - https://github.com/codelabs-ch/genode/blob/hw_x86-fpu/repos/base-hw/src/cor… [4] - https://github.com/codelabs-ch/genode/commit/996ffa6407effaa45488ddf3e41986…

2 1

Re: Documentation - request for feedback
by Norman Feske 09 Mar '15

09 Mar '15

Hi Stephan, > My experience with such documentation (considering that I am maintaining > a 1500+ design documentation of the Linux kernel) is that references to > the code are helpful. For example: when you talk about a capability in > section 3.1, just add a hint which C++ object such capability is stored > in. With this, a reader now as an entry into the code and can start to > understand the code. > > Note, I am not looking for comprehensive code snippets -- just a > reference to the code structs / objects. that is a good advice. I actually deliberately tried to keep Chapter 3 void of "implementation smell" to avoid distracting the reader from understanding the architectural concepts. But I agree that the reader might desire to the draw the connection to the actual code. Would the addition of cross references to the corresponding subsections of Chapter 7 (Functional specification) achieve the desired effect? Chapter 7 will contain the actual API documentation (of course referring to the source code). This way, the text in Chapter 3 would not directly but indirectly refer to the source code. Best regards Norman -- Dr.-Ing. Norman Feske Genode Labs http://www.genode-labs.com · http://genode.org Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth

3 3

How to run two_linux_panda
by Zachary 05 Mar '15

05 Mar '15

Hi all, I am interested in the work of Genode about muti_linux on pandaboard[1]. So, I downloaded the image from that page, and run it correctly. I try to build it by myself but get some troubles. Firstly, there is no two_linux_panda.run script in the latest release of Genode, so I got one form genode 12.08. In order to build it , I changed 'omap4_fb' to 'fb_drv', and delete 'lib_log.lib.so'(The building log says can't find it). After that, I got an image.elf and run it on pandaboard, but it can't not work well. This is the log of UART output: ... Genode 14.11 int main(): --- create local services --- int main(): --- start init --- int main(): transferred 989 MB to init int main(): --- init created, waiting for exit condition --- Quota exceeded! amount=28672, size=4096, consumed=28672 [init] upgrading quota donation for Env::CPU (8192 bytes) Quota exceeded! amount=36864, size=4096, consumed=36864 [init] upgrading quota donation for Env::CPU (8192 bytes) Quota exceeded! amount=45056, size=4096, consumed=45056 [init] upgrading quota donation for Env::CPU (8192 bytes) [init -> sd_card_drv] --- OMAP4 SD card driver --- [init -> nic_bridge] Could not open file "config" [init -> nic_bridge] Could not obtain config file [init -> blocktest_term] int main(int, char**): --- terminal service started --- Insufficient quota for transfer: init -> nic_bridge have 1564672, need 3301376 [init] not enough quota for a donation of 3301376 bytes [init -> nic_bridge] resource_request: ram_quota=3301376 [init -> nic_bridge] virtual void Genode::Platform_env::release(): used before 8 [init -> nic_bridge] virtual void Genode::Platform_env::release(): used after f2 [init] child "nic_bridge" requests resources: ram_quota=3301376 [init -> fb_drv] Could not open file "config" [init -> fb_drv] Could not obtain config file [init -> http_term] int main(int, char**): --- terminal service started --- Insufficient quota for transfer: init -> part_blk have 1564672, need 4206592 [init] not enough quota for a donation of 4206592 bytes [init -> part_blk] resource_request: ram_quota=4206592 [init -> part_blk] virtual void Genode::Platform_env::release(): used before fr6 [init -> fb_drv] int main(int, char**): using default configuration: HDMI@...305... [init -> part_blk] virtual void Genode::Platform_env::release(): used after fre0 [init] child "part_blk" requests resources: ram_quota=4206592 [init -> vmlinux.1] Booting L4Linux ... [init -> vmlinux.1] ======> L4Linux starting... <======== [init -> vmlinux.1] Linux version 3.9.0-l4-gf2eebae (czzcz@...306...) (gcc5 [init -> vmlinux.1] Binary name: vmlinux [init -> vmlinux.1] This is an AEABI build. [init -> vmlinux.1] Linux kernel command line (2 args): mem=54M l4x_rd=initrd.gz [init -> vmlinux.1] CPU mapping (l:p)[0]: Image: 02000000 - 02500000 [5120 KiB]. [init -> vmlinux.1] Areas: Text: 02000000 - 023dc000 [3952kB] (a bit longer) [init -> vmlinux.1] Data: 023dc000 - 024091b8 [180kB] [init -> vmlinux.1] Initdata: 023b4000 - 023da9c0 [154kB] [init -> vmlinux.1] BSS: 0240a000 - 02446de8 [243kB] [init -> vmlinux.2] Booting L4Linux ... [init -> usb_drv] Services::Services(): Could not read screen resolution in cone [init -> usb_drv] Services::Services(): No <storage> config node found - not ste [init -> usb_drv] Services::Services(): No <raw> config node found - not startie [init -> usb_drv] Warning: No USB controllers enabled. [init -> usb_drv] Use <config (u/e/x)hci="yes"> in your 'usb_drv' configuration [init -> sd_card_drv] CID: 0x97009861 0x38caa862 0x44303847 0x02544d53 [init -> sd_card_drv] RCA: 0x895a [init -> vmlinux.2] ======> L4Linux starting... <======== [init -> vmlinux.2] Linux version 3.9.0-l4-gf2eebae (czzcz@...306...) (gcc5 [init -> vmlinux.2] Binary name: vmlinux [init -> vmlinux.2] This is an AEABI build. [init -> vmlinux.2] Linux kernel command line (2 args): mem=54M l4x_rd=initrd.gz [init -> sd_card_drv] SD card detected [init -> sd_card_drv] capacity: 7600 MiB [init -> vmlinux.2] CPU mapping (l:p)[0]: Image: 02000000 - 02500000 [5120 KiB]. [init -> vmlinux.2] Areas: Text: 02000000 - 023dc000 [3952kB] (a bit longer) [init -> vmlinux.2] Data: 023dc000 - 024091b8 [180kB] [init -> vmlinux.2] Initdata: 023b4000 - 023da9c0 [154kB] [init -> vmlinux.2] BSS: 0240a000 - 02446de8 [243kB] [init -> part_blk] Partition 1: LBA 62 (78058 blocks) type: c [init -> part_blk] Partition 2: LBA 78120 (46872 blocks) type: c [init -> part_blk] Partition 3: LBA 124992 (46872 blocks) type: c [init -> nitpicker] Could not open file "config" [init -> nitpicker] Could not obtain config file [init -> nitpicker] create session with args: label="linux.1", ram_quota=28672 [init -> nitpicker] create session with args: label="linux.2", ram_quota=28672 [init -> nitpicker] create session with args: label="http_fb", ram_quota=28672 [init -> nitpicker] create session with args: label="blocktest_fb", ram_quota=22 [init -> linux.1] using xywh=(1,1,510,382) [init -> linux.2] using xywh=(513,513,510,382) [init -> http_fb] using xywh=(1,1,510,376) [init -> blocktest_fb] using xywh=(513,513,510,376) [init -> http_term] Could not open file "config" [init -> http_term] Could not obtain config file [init -> blocktest_term] Could not open file "config" [init -> blocktest_term] Could not obtain config file [init -> http_term] cell size is 8x16 [init -> blocktest_term] cell size is 8x16 [init -> http_term] create terminal session [init -> blocktest_term] create terminal session [init -> http_term] new terminal session: [init -> http_term] framebuffer has 510x376 pixels [init -> http_term] character size is 8x16 pixels [init -> http_term] terminal size is 63x23 characters [init -> blocktest_term] new terminal session: [init -> blocktest_term] framebuffer has 510x376 pixels [init -> blocktest_term] character size is 8x16 pixels [init -> blocktest_term] terminal size is 63x23 characters [init -> nic_bridge] Invalid session request, no matching policy [init -> vmlinux.1] l4io_get_root_device: Not implemented yet! [init -> vmlinux.1] Device scan: [init -> vmlinux.1] l4io_iterate_devices: Not implemented yet! [init -> vmlinux.1] Device scan done. [init -> vmlinux.2] l4io_get_root_device: Not implemented yet! [init -> vmlinux.2] Device scan: [init -> vmlinux.2] l4io_iterate_devices: Not implemented yet! [init -> vmlinux.2] Device scan done. [init] vmlinux.1: no route to service "Timer" [init -> vmlinux.1] C++ runtime: Genode::Parent::Service_denied [init] vmlinux.2: no route to service "Timer" [init -> vmlinux.1] void* abort(): abort called [init -> vmlinux.2] C++ runtime: Genode::Parent::Service_denied [init -> vmlinux.2] void* abort(): abort called [init] virtual void Genode::Child_policy::exit(int): child "test-libc_ffat" exi1 [init] virtual void Genode::Child_policy::exit(int): child "test-lwip_httpsrv" 1 -------------------------------------------------------------- Only test-lwip_httpssrv shows on the The HDMI moniter, and it says : [test-lwip_httpssrv] Could not open file "config" [test-lwip_httpssrv] Could not obtain config file [test-lwip_httpssrv] no VFS configured [test-lwip_httpssrv] DHCP timed out [test-lwip_httpssrv] We got no IP address! I want to ask: 1. What is 'lib_log.lib.so'? Can I delete it? 2. Is there a document about configuration of run script? 3. How to run the two_linux_panda correctly? Best, Zachary [1] http://genode.org/documentation/articles/pandaboard

2 5

implementation of dummies methods
by Mohammad Hamad 02 Mar '15

02 Mar '15

Hi all, in libc/src/lib/libc there is Dummies.cc file which contains many methods.Actually i am trying to port ISAKMP which use most of them ( such as fork, getuid ,socketpair, sendmsg and etc) . i was able to compile it but i face many no implementation for " method" message , so I think i need to implement all these methods, am i right ?! from where i should start, should i use new libc plugin or i should implement them in specific places. thank you in advance . best,

4 6

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

users