l4-hackers August 2014

l4-hackers@os.inf.tu-dresden.de

20 participants
35 discussions

Re: Mapping memory into process address space
by Björn Döbel 19 Aug '14

19 Aug '14

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 >> Also, do not call the allocator twice within the function call, >> because then you would obtain two different dataspaces and this >> is not what you want here. Better do >> >> L4:Cap<Dataspace> x = >> L4Re::Util::cap_alloc.alloc<L4::Dataspace>(); long alloc = >> L4Re::Env::env()->mem_alloc.alloc(size, x); if (!x.is_valid()) { >> // handle error } > > Okay, I've gotten the allocation of the dataspace to work correctly > (at least as far as I can tell), but I'm still not sure how to map > the resulting dataspace into the newly created process. I'm trying > something like this: > > L4::Cap<Dataspace> stack = > L4Re::Util::cap_alloc.alloc<L4::Dataspace>(); < allocate memory > > task->map(L4Re::This_task, stack.fpage(), 0); > > but it still freezes. Is the fpage() call still waiting on IPC, or > should I be using a different value for snd_base? The snd_base parameter to the map() call is the address in the target address space. Nevertheless, I'm wondering while this would hang. Are you sure this happens in the map() operation? Bjoern -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Icedove - http://www.enigmail.net/ iEYEARECAAYFAlPzSo4ACgkQP5ijxgQLUNmVHACfYrqIXq5opXk2e2nzac6fe3AB a10AnitnVNIpOIBPQOhqV1RGMD7j5Z5K =xWcr -----END PGP SIGNATURE-----

1 0

Adding a new scheduling algorithm to Fiasco.OC
by Valentin Hauner 19 Aug '14

19 Aug '14

Hi, I'm planning to add a new scheduling algorithm to Fiasco.OC. I will implement EDF. First of all, is there some handout written by you main developers that copes with the issues of implementing EDF in Fiasco.OC? What I've achieved so far: I've created the files sched_context-edf.cpp and ready_queue_edf.cpp in ./kernel/fiasco/src/kern, using the corresponding source files for your fixed-priority scheduler as templates. I've named the config directive used in ./kernel/fiasco/src/Modules.* files analogous to yours, that is 'sched_edf' in my case. Now, I'm beginning to modify sched_context-edf.cpp and ready_queue_edf.cpp to fit the requirements of EDF, especially the en- and dequeuing as well as the 'dominates' method. Are there any other major source files that I do have to alter? Thanks in advance! Best regards, Valentin

2 17

I share a document to access HW on L4 Linux.
by Taeung 18 Aug '14

18 Aug '14

Thank you L4 Hackers. You have help me in securing Ethernet on L4 Linux. So, I succeeded it. I’m particularly grateful to Martin Schröder, Wei jin, Adam Lackorzynski, Matthias Lange. Although there is already a document 'L4 Runtime Environments : Io, the Io server' a link of which is http://os.inf.tu-dresden.de/L4Re/doc/io.html, I made a guide document for accessing HW on L4 Linux because of ethernet example. (I've used ARM platform Freescale i.MX6 Quad SABRESD ) This document is https://www.dropbox.com/s/abj9ao5uk9r3w72/The_Guide_to_access_HW_on_L4Linux… I don’t know whether this document can help someone in accessing HW on L4Linux. But I want to share this document in order to repay for l4hackers's kindness. - Taeung -

3 5

compiling and running l4android-arm in qemu
by cem akpolat 18 Aug '14

18 Aug '14

Dear all, I have compiled l4android, l4 and fiasco for arm architecture and now I am attempting to run elf image in qemu.However, I see only a black screen in qemu. There is any io messages. What could be the reason? My qemu version is 1.7.1. The images that I have after making elfimages are: bootstrap.elf bootstap_L4LinuxARM bootstap_L4LinuxARM.elf The used command for running image: $qemu-system-arm -M realview-pbx-a9 -m 2048 -kernel bootstrap_L4LinuxARM.elf -serial stdio Thanks in advance, Sincerely, Cem Akpolat

2 1

Mapping memory into process address space
by Noah Zentzis 15 Aug '14

15 Aug '14

------------------------------------------------------------------------ I don't want to use libloader because I'm trying to create a new process with dynamically-generated assembly (and alter it while the process is running), so I can't really output an ELF binary. On 08/18/2014 08:50, Björn Döbel wrote: > The dataspace that prog_attach_ds() is called with is allocated by the > ELF loader in libloader/include/elf. > > reserve_utcb_area() is supposed to not attach a real dataspace but > reserve a virtual memory area that will not be used by the new > application's RM to attach any other memory, because this is where the > UTCBs will later be mapped to by the kernel iirc. I'm pretty sure I need to do this myself, since any calls to map() are causing the program to freeze (could this be because the task doesn't have its UTCB allocated yet?). I'm also not sure what to use for the destination offset of the L4::Task::map() call, so that might be the issue. > At this point your best way to do so is > to use the L4::Task::map() operation on the capability of your newly > created task. How do I specify what point in the target the memory maps to? It looks like the L4::Dataspace::map() would work, but without being able to map the capability, I can't really check. Thanks! - Noah Zentzis

3 3

Build system
by teclis High Elf 14 Aug '14

14 Aug '14

If I execute this command from the l4 build directory: make E=hello S=bootstrap In which Makefile or script is the S= processed (its a bit of a difficult thing to grep for :) thanks

2 3

Accessing dbg_info()->dbg_id() for threads in ready queue class
by Valentin Hauner 13 Aug '14

13 Aug '14

Hi, how can I access the identificator used in JDB's thread list in the enqueue method of my ready queue class? Following code snippet does not work: > include "thread.h" > [...] > Ready_queue_edf<E>::enqueue(E *i, bool) { > printf("%d", static_cast<Thread*>(i->context())->dbg_info()->dbg_id()); } It results in various compiler errors: > [...] > ./kernel/fiasco/src/kern/irq.cpp:504:11: error: ‘class Thread’ has no member named ‘home_cpu’ > ./kernel/fiasco/src/kern/irq.cpp:505:15: error: ‘class Thread’ has no member named ‘drq’ > [...] > ./kernel/fiasco/src/kern/tcboffset_in.h:11:3: error: ‘class Thread’ has no member named ‘_caller’ > ./kernel/fiasco/src/kern/tcboffset_in.h:14:3: error: ‘class Thread’ has no member named ‘_timeout’ What else do I have to include that it works? Best regards, Valentin

2 1

Mapping memory into process address space
by Noah Zentzis 13 Aug '14

13 Aug '14

I'm trying to launch a new process (without libloader) and I'm having trouble mapping memory to it. I've found the prog_attach_ds and prog_reserve_utcb_area methods in libloader and ned, but I'm not sure where they're getting the memory to assign or the region map for the task. Can I just do something like: L4Re::Util::Cap<L4Re::Dataspace> stack; L4Re::Env::mem_alloc()->alloc(0x10000, stack); or are there special steps I need to take when I want to give another process access to the same memory? Thanks! - Noah Zentzis

2 1

Question about IPI interrupt handler
by Yuxin Ren 11 Aug '14

11 Aug '14

Hi All, I cannot understand the logic of the IPI interrupt handler very well. I hope someone can help me. Thanks a lot. I am using intel IA32 multi core cpu. My scenario is one client and one server thread on different core. Client sends simple data-only IPC to the server. 1. The function Thread::handle_remote_requests_irq is the IPI interrupt handler, right? So it is running within the interrupt context? If so, 2. In this function, the variable "c" is set to the current Context. On the server core, when the server is blocked(waiting for IPC), what is current Context? The server thread or the idle thread? I have no other threads in the user level, and I assume there are no background threads in the kernel or L4Re. In addition, I guess each thread associates with a Context. If my assumptions are right and that current Context is idle thread, 3. At the last of this function, it calls c->handle_drq(). But when does the server thread's handle_drq() is called? 4. In what situation, the variable "resched" will be set to true? When it is true, 5. c->schedule() will be called, does it mean that we can switch to another thread in the interrupt context? It cannot make sense to me. Thanks again. Best, Yuxin

2 1

how to get or initialize clock in L4Linux
by Yunchuan Geng 08 Aug '14

08 Aug '14

Hi, L4hackers, I am implementing I2C drivers from native linux to L4Linux. I am using Freescale iMX6Q SD board. Kernel version 3.14 in ./drivers/i2c/busses/i2c-imx.c, function i2c_imx_probe(), devm_clk_get() is called. And fails at here. I understand that clock are initialized from function "__init imx6q_clocks_init()". I managed to compile and call the imx6q_clocks_init() function from L4Linux, but encountered many problem. The question is, Is it necessary to initialize clocks in L4Linux and how? Thanks in advance? -Yunchuan

2 3

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

l4-hackers August 2014