- l4-hackers - example.com Site

Data transfer mechanisms
by Lluís Vilanova 20 Aug '13

20 Aug '13

Hi, I'd like to compare the performance of two different synchronous data transfer mechanisms: * Copies * Page maps / unmaps For the second one, it looks like I can use the streammap example (pkg/examples/libs/l4re/streammap), although I'm lacking a mechanism to unmap the page before returning control to the client. It would also be nice to be able to pass the server a read-only region. For the first one, I've only found the use of multiple send/recv operations through the limited amount of data available on the message registers. There is a "Gen_string" class in the "ipc_stream" header, which looks like it could support offline string transfers, but it's disabled at compile time and I could not find the actual code in fiasco that handles these transfers. Thanks, Lluis -- "And it's much the same thing with knowledge, for whenever you learn something new, the whole world becomes that much richer." -- The Princess of Pure Reason, as told by Norton Juster in The Phantom Tollbooth

2 4

Files that Tune Kernel
by Bryan Amato 17 Aug '13

17 Aug '13

Hello everyone, I was wondering what files are the ones that increase the performance of the kernel, besides kernel.hz. Reading in a post by L4, they say that the kernel was not tuned like the Mach Linux port. But I wanted to edit those files to increase performance and to also learn more about what that particular file does. I tried doing this once, but my whole kernel got corrupted from my lack of knowing what is able to be edited and things that had to be there so it could be booted properly. Thanks!

2 1

Desktop Environment
by Bryan Amato 17 Aug '13

17 Aug '13

Hello everyone, I have been using Linux for sometime and only recently tried a special version of BSD, named Dragonfly BSD. What I liked about Dragonfly is that the kernel is not a monolithic kernel. However BSD is not really for me, I prefer the comfort of Linux, so I began searching around the web for a non-monolithic version of the Linux kernel and I came across L4Linux. Looking at the screenshots, I don't see any modern DE's. Does L4L support those desktops like KDE, Gnome and E17? Thanks for all the work you guys are putting into this kernel!

4 3

How to write device driver
by Raja Sekhar 16 Aug '13

16 Aug '13

Hello to all, I want to write a simple character device driver.Can you please give me the procedure for writing device drivers for L4fiasco. * Thanks,* Raja Sekhar

2 1

l4linux with ide-harddisk
by Rudolf Weber 15 Aug '13

15 Aug '13

Hello, I prepared a 32 bit ubuntu system on a virtualbox and replaced the linux - Kernel through the l4re-snapshot-2013060718 - L4Linux-mag-x86 Konfiguration. First I made a disk and add it to the vm and I could mount it ! I wonder ( see later). Now I configured the Ubuntu-Linux-Root-Disk as /dev/hda and: ... piix 0000:00:03.0: IDE controller (0x8086:0x7111 rev 0x01) pci 0000:00:03.0: No interrupt pin configured for device 0000:00:03.0 piix 0000:00:03.0: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xd000-0xd007 ide1: BM-DMA at 0xd008-0xd00f hdc: VBOX CD-ROM, ATAPI CD/DVD-ROM drive hdc: UDMA/33 mode selected ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 ESC[37mio | ERROR: binding irq 15, result is -1 (Operation not permitted) ESC[0m ESC[33ml4linux | Bind irq to icu failed with -1ESC[0m l4x-irq: Did not get IRQ 15 from IO service ==> I think, the interupt is not delegated from the IO-Server to l4linux My understanding is, that the io-Server manage all interrupts,ioports. But why I could mount a disk ? I dont see a device in x86-legacy.dev, so I add IDE0 and IDE1 (see attachment), and add a systembus disks => new System_bus() { IDE0 => wrap(hw-root.match("ide0")); IDE1 => wrap(hw-root.match("ide1")); } in l4lx-x86.io (see attachment) in the start-config l4lx-gfx.cfg the disks are granted to l4linux (see attachment) As result, the output is piix 0000:00:03.0: IDE controller (0x8086:0x7111 rev 0x01) pci 0000:00:03.0: No interrupt pin configured for device 0000:00:03.0 piix 0000:00:03.0: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xd000-0xd007 ide1: BM-DMA at 0xd008-0xd00f hda: VBOX HARDDISK, ATA DISK drive hda: UDMA/33 mode selected hdc: VBOX CD-ROM, ATAPI CD/DVD-ROM drive Switching to clocksource tsc hdc: UDMA/33 mode selected ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 ESC[33ml4linux | Bind irq to icu failed with -1ESC[0m l4x-irq: Did not get IRQ 15 from IO service ------------[ cut here ]------------ WARNING: at /home/rw/l4exp/l4re-snapshot-2013060718/src/l4linux/arch/l4/l4lxlib/ l4env/irq_vcpu.c:106 l4lx_irq_dev_startup+0x17a/0x180() Modules linked in: Pid: 1, comm: swapper Not tainted 3.9.0-l4 #1 Call Trace: -> the ERROR "ERROR: binding irq 15, result is -1 (Operation not permitted)" don't appear, but the result is the same. The current output of the serialine is in l4ubuntu.1 The l4-linux I configured with PCI/VPCI see attatchment x86_rw_config like in my last E-mails. Summary: * How to grant the ide-Controllers to l4linux ? * Why l4Linux can't get the interrupt - what is wrong with my configuration ? HW-Question: How are the IDE-Controllers and the PCI related ? Thank you very much Rudolf Weber

2 3

issue in getting vhw_devices....................
by Raja Sekhar 13 Aug '13

13 Aug '13

Hello to all, I've downloaded l4re-core-2013041501.tar.bz2 and extracted and compiled fiasco kernel with following configuration Architecture-> IA-32 processor family, Platform -> PC Platform, CPU -> Intel Pentium Pro After that i compiled L4re env....and I just ran one example ux-vhw(which is the code for display Virtual hardware device). In that I'm getting value *'0'* for the fileld *vhw_offset *from the kernel info structure *struct l4_re_kernel_info_t.* Why It is giving zero and what does it mean....* Thanks*, Raja Sekhar

2 1

Issue in getting Virtual Hardware devices
by Raja Sekhar 10 Aug '13

10 Aug '13

Hello to all, I'm working on L4 fiasco. In that I want to read the virtual hardware devies. for tha I'm running the code from * l4/pkg/example/sys/ux-vhw/main.c *** * 10 #include <l4/sys/ipc.h> 11 #include <l4/sys/vhw.h> 12 #include <l4/util/util.h> 13 #include <l4/util/kip.h> 14 #include <l4/re/env.h> 15 16 #include <stdlib.h> 17 #include <stdio.h> 18 19 static void print_entry(struct l4_vhw_entry *e) 20 { 21 printf("type: %d mem start: %08lx end: %08lx\n" 22 "irq: %d pid %d\n", 23 e->type, e->mem_start, e->mem_size, 24 e->irq_no, e->provider_pid); 25 } 26 27 int main(void) 28 { 29 l4_kernel_info_t *kip = l4re_kip(); 30 struct l4_vhw_descriptor *vhw; 31 int i; 32 33 if (!kip) 34 { 35 36 printf("KIP not available!\n"); 37 return 1; 38 } 39 40 vhw = l4_vhw_get(kip); 41 42 if(vhw) 43 { 44 printf("kip at %p, vhw at %p\n", kip, vhw); 45 printf("magic: %08x, version: %08x, count: %02d\n", 46 vhw->magic, vhw->version, vhw->count); 47 48 for (i = 0; i < vhw->count; i++) 49 print_entry(l4_vhw_get_entry(vhw, i)); 50 } 51 return 0; 52 }** *In the above code at *line no 40* after *calling inline function l4_vhw_get(kip)* I'm getting *vhw=NULL. *and the definition of inline function is from* l4sys/include/vhw.h *is as follows* 80 enum { 81 L4_VHW_MAGIC = 0x56687765, 82 }; 83 84 static inline struct l4_vhw_descriptor * 85 l4_vhw_get(l4_kernel_info_t *kip) L4_NOTHROW 86 { 87 struct l4_vhw_descriptor *v 88 = (struct l4_vhw_descriptor *)(((unsigned long)kip) + kip->vhw_offset); 89 90 if (v->magic == L4_VHW_MAGIC) 91 return v; 92 93 return NULL; 94 } *Here* at line no 90, the v->magic *is having a value of* 0x4BE6344C *which is not equal to* L4_VHW_MAGIC *My qestion is why I'm getting different magic value Instead of 0x56687765, and how can I get correct magic value*. *If there are no virtual hardware device how can i create and integrate into my system........* * * * * *Thanks, Raja Sekhar

2 1

Setting physical CPU for a task
by Lluís Vilanova 06 Aug '13

06 Aug '13

Hi, I've been looking for a way to run some sort of ping-pong example where each task runs on either the same or different CPUs. I've been unable to see if moe's config files allow to start tasks pinned to a specific CPU, and after running the "thread_migrate" example, looks like only one CPU (CPU00) is always used: $ make -C obj/l4/amd64/ qemu E=thread_migrate QEMU_OPTIONS="-serial stdio -vnc ::59000 -smp 2" [...] Not using serial hack in slow timer handler. CPU[0]: AuthenticAMD (6:2:3:0)[00000623] Model: QEMU Virtual CPU version 1.5.1 at 3092MHz 255/512 Entry I TLB (4K pages) 255 Entry I TLB (4M pages) 255/512 Entry D TLB (4K pages) 255 Entry D TLB (4M pages) 64 KB L1 I Cache (2-way associative, 64 bytes per line) 64 KB L1 D Cache (2-way associative, 64 bytes per line) 512 KB L2 U Cache (8-way associative, 64 bytes per line) Freeing init code/data: 20480 bytes (5 pages) [...] Ned says: Hi World! Ned: loading file: 'rom/thread_migrate.cfg' migrate | Found 4 CPUs. migrate | Created 12 threads. migrate | Migrated Thread00 -> CPU00 migrate | Migrated Thread01 -> CPU00 migrate | Migrated Thread02 -> CPU00 migrate | Migrated Thread03 -> CPU00 migrate | Migrated Thread04 -> CPU00 migrate | Migrated Thread05 -> CPU00 migrate | Migrated Thread06 -> CPU00 migrate | Migrated Thread07 -> CPU00 migrate | Migrated Thread08 -> CPU00 migrate | Migrated Thread09 -> CPU00 migrate | Migrated Thread10 -> CPU00 migrate | Migrated Thread11 -> CPU00 migrate | Migrated Thread00 -> CPU00 migrate | Migrated Thread01 -> CPU00 migrate | Migrated Thread02 -> CPU00 migrate | Migrated Thread03 -> CPU00 migrate | Migrated Thread04 -> CPU00 migrate | Migrated Thread05 -> CPU00 migrate | Migrated Thread06 -> CPU00 migrate | Migrated Thread07 -> CPU00 migrate | Migrated Thread08 -> CPU00 migrate | Migrated Thread09 -> CPU00 migrate | Migrated Thread10 -> CPU00 migrate | Migrated Thread11 -> CPU00 [... same output forever ...] I've modified "src/l4/conf/modules.list" to add the "thread_migrate" test, and I'm using l4re-snapshot-2013060718, with multiprocessing (with a maximum of 4) enabled on fiasco. Looks like the benchmark always detects (L4::Scheduler::info) the number of maximum CPUs configured in fiasco, but not the number of CPUs provided by QEMU (only CPU 0 is online). Am I missing some step to set all CPUs online? Is it possible to have MOE start tasks pinned to a specific CPU? Thanks a lot, Lluis -- "And it's much the same thing with knowledge, for whenever you learn something new, the whole world becomes that much richer." -- The Princess of Pure Reason, as told by Norton Juster in The Phantom Tollbooth

2 5

Re: l4-hackers Digest, Vol 124, Issue 4
by Иван Филиппов 06 Aug '13

06 Aug '13

Hi Lluis, Have pretty much the same as you sad - can't load both qemu for unknown memory address, neither for grub2 for Ubuntu 13.04, but it compiles well. I`ll ask what happens tomorrow on upcoming russian Fiasco/Virtualization school with Bjorn Doebel at al. and tell you what happens. Probably the best choose is Qemu emulator, if you don't have already written software, not to reload system every time you need. I compiled with i386 option both FiascoOC/L4RE and them works well. just as here: http://os.inf.tu-dresden.de/pipermail/l4-hackers/2011/004894.html http://os.inf.tu-dresden.de/pipermail/l4-hackers/2011/<http://os.inf.tu-dresden.de/pipermail/l4-hackers/2011/004894.html> 004894.html<http://os.inf.tu-dresden.de/pipermail/l4-hackers/2011/004894.html> I have 4.7 compiler too. Best wishes, 2013/8/5 <l4-hackers-request(a)os.inf.tu-dresden.de> > Send l4-hackers mailing list submissions to > l4-hackers(a)os.inf.tu-dresden.de > > To subscribe or unsubscribe via the World Wide Web, visit > http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers > or, via email, send a message with subject or body 'help' to > l4-hackers-request(a)os.inf.tu-dresden.de > > You can reach the person managing the list at > l4-hackers-owner(a)os.inf.tu-dresden.de > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of l4-hackers digest..." > > > Today's Topics: > > 1. Unable to boot Fiasco.OC on amd64 (Llu?s Vilanova) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sun, 04 Aug 2013 22:44:24 +0300 > From: Llu?s Vilanova <vilanova(a)ac.upc.edu> > To: l4-hackers(a)os.inf.tu-dresden.de > Subject: Unable to boot Fiasco.OC on amd64 > Message-ID: <87eha95jhj.fsf(a)fimbulvetr.bsc.es> > Content-Type: text/plain; charset="us-ascii" > > Hi there, > > After having to play with the load addresses of both fiasco and some of the > modules [1], I've been able to boot fiasco, but only to find it chokes on > some > assertion, immediately following a pager warning (see error.jpg > screenshot). > > Sorry for using a screenshot, but I've been unable to make the serial > console > work, so no copy&paste is available. > > As I'm new to L4, I don't know if the error is due to my changes or due to > some > other mis-configuration on the boot process. See the attached grub2.txt > for the > GRUB2 entries I've tried (the error corresponds to the first entry). > > I'm using the l4re-core-2013060718 tarball, all compiled on a debian > testing > (gcc 4.7.3). > > Any help on the boot or the serial console support is much appreciated. > > > Thanks a lot, > Lluis > > [1] To avoid some region overlaps, see attached l4.patch, and memory.jpg > for the > memory range configuration of the machine. > >

3 10

mmap2 on L4Linux rev 37
by Sebastian Sumpf 06 Aug '13

06 Aug '13

Hi L4 hackers, recently I started to use L4Linux r37 on ARM and experienced some issues with the 'sys_mmap_pgoff'-system call. When a user-level task tried to map a file the 'pgoff' parameter was usually bogus. Doing a little research I found that it was caused by the 'sys_mmap2' implementation in 'entry.S'. The function overwrites the 'pgoff' argument on the stack with r5 (as it is usually passed from user land), since in L4Linux this code gets called from the dispatch routine, 'pgoff' was set to whatever r5 was, and thus leading to bus errors later on. Cheers, Sebastian -- Sebastian Sumpf 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

2 1