l4-hackers September 2016

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

8 participants
7 discussions

Porting Fiasco + L4Re to Raspberry Pi 2
by Antoine DEBUT 26 Nov '17

26 Nov '17

Hello l4-hackers, I am currently working on porting the Fiasco.OC micro kernel onto the Raspberry pi 2. Therefore, I am struggling with memory mapping, and I encounter the following error when launching the default 'hello' entry: L4 Bootstrapper Build: #2 Tue Feb 23 11:57:19 CET 2016, 4.8.3 20140303 (prerelease) Scanning up to 1024 MB RAM, starting at offset 32MB Memory size is 1024MB (00000000 - 3fffffff) RAM: 0000000000000000 - 000000003fffffff: 1048576kB Total RAM: 1024MB Scanning fiasco Scanning sigma0 Scanning moe Moving up to 5 modules behind 1100000 moving module 02 { 10db000-110c4d7 } -> { 11cc000-11fd4d7 } [201944] moving module 01 { 10c9000-10da373 } -> { 11ba000-11cb373 } [70516] moving module 00 { 1043000-10c889f } -> { 1134000-11b989f } [546976] moving module 04 { 1029000-10425af } -> { 111a000-11335af } [103856] moving module 03 { 100f000-102849b } -> { 1100000-111949b } [103580] Loading fiasco Loading sigma0 Loading moe find kernel info page... found kernel info page at 0x2000 Regions of list 'regions' [ 1000, 1aff] { b00} Kern fiasco [ 2000, 92fff] { 91000} Kern fiasco [ 93000, 930eb] { ec} Root mbi_rt [ c0000, c9c33] { 9c34} Sigma0 sigma0 [ d0000, d6177] { 6178} Sigma0 sigma0 [ 140000, 16905b] { 2905c} Root moe [ 170000, 17b58b] { b58c} Root moe [ 1000000, 100e4ff] { e500} Boot bootstrap [ 1100000, 1133fff] { 34000} Root Module API Version: (87) experimental Sigma0 config ip:000c0100 sp:00000000 Roottask config ip:00140250 sp:00000000 Starting kernel fiasco at 000012b8 Non-HYP kernel detected but running in HYP mode, switching back. Hello from Startup::stage2 Assertion failed at /home/adebut/Documents/snapshotTest/l4re-snapshot-2015123115/src/kernel/fiasco/src/kern/buddy_alloc.cpp:101: (unsigned long)block >= _base Press any key to reboot. I am currently working on the latest snapshot (2015123115), with U-Boot as a bootloader, and testing on a real board (Raspberry Pi 2 model B V1.1). The entry built is the standard 'hello', and I use the ramdisk-arm.rd given in the snapshot. I did add a custom Bsp for the board, mainly based on the first Raspberry Pi except for the Base Address, that is 0x3F000000 instead of the first Pi's 0x20000000. I highly think that it is an address configuration problem, because it first happened when I did expand the RAM size to 1024MB. With further investigation, I found out that it was the kernel memory allocation that causes this error, and suspect an incorrect MMU mapping. This suspicion comes from the fact that the assertion takes the address _base (0xf0400000) and the currently allocated block (0xee100000). I did not modify any parameter in the kern/arm/mem_layout-arm.cpp file on the Fiasco directory, which provides the default ARM addressing. Maybe that's where my error comes from ? How can I correctly configure my memory mapping, besides what I already configure in the Bsp (given in attachment) ? Also, it is the first time that I am assigned for porting a system, do you have some hints, like important steps to follow ? I am pretty sure it is a newbie error, but I cannot figure it out for now :( Thanks in advance for your time, -- Antoine Debut EPITA 2016 - GISTRE

3 7

Graphical desktop, l4con, mag
by teclis High Elf 02 Jun '17

02 Jun '17

Guten Tag, I've got the Hello app runing with fiasco and l4re in qemu, how do I get the graphical UI and console running that I see on the screen shots page? Do I need to use l4con and mag? I don't see an entry with l4con in modules.list. danke * <http://german.about.com/library/media/Audio/gutentag.wav>*

6 15

Benchmarks in l4
by Maxim Baryshnikov 01 Jun '17

01 Jun '17

Hello everybody, Does the L4re contain some kind of simple performance benchmarks which are already prepared to use? I've tried to launch some demos from the ferret package but not succeed... Thank you in advance for any suggestions. Sincerely, Maxim Baryshnikov.

4 11

libfs/ FUSE on L4Re
by Stark, Josef 31 May '17

31 May '17

Hello, yesterday I came across a very interesting paper titled "Porting FUSE to L4Re", accessible here: http://os.inf.tu-dresden.de/papers_ps/pester-beleg.pdf For porting FUSE, the author uses a library called "libfs" which provides filesystem request forwarding and processing, if I understood correctly. It reads as if libfs is already included in L4Re. However, I couldn't find neither FUSE nor libfs in the current L4Re snapshot. Are they by any chance publicly available somewhere else, or are both not to be published? Cheers, Josef

4 7

L4Linux stop at "Calibrating delay loop ..."
by li94575 09 Oct '16

09 Oct '16

Hi，l4hackers I compiled the l4re-snapshot-2014092821 with the default modules.list, and everything seems goes well. However, L4Linux would stop at "Calibrating delay loop ...", because the value of jiffies did not change. Timer thread has not generate soft interrupt? If I entered jdb via ESC at this time, and quit via "g", then l4linux can run again. It makes me confused so much. Any help or suggestion will be very welcome. The log is showing below: L4 Bootstrapper Build: #12 2016年 09月 19日星期一 18:58:15 CST, 4.8.3 20140320 (prerelease) Scanning up to 2048 MB RAM, starting at offset 32MB Memory size is 2048MB (00000000 - 7fffffff) RAM: 0000000000000000 - 000000007fffffff: 2097152kB Total RAM: 2048MB Scanning fiasco Scanning sigma0 Scanning moe Moving up to 10 modules behind 1100000 moving module 02 { 1968000-19a163f } -> { 1a58000-1a9163f } [235072] moving module 01 { 195e000-196738b } -> { 1a4e000-1a5738b } [37772] moving module 00 { 18f8000-195d9db } -> { 19e8000-1a4d9db } [416220] moving module 09 { 15f8000-18f7fff } -> { 16e8000-19e7fff } [3145728] moving module 08 { 11d6000-15f7497 } -> { 12c6000-16e7497 } [4330648] moving module 07 { 11d5000-11d50bc } -> { 12c5000-12c50bc } [189] moving module 06 { 107e000-11d446f } -> { 116e000-12c446f } [1401968] moving module 05 { 107d000-107d19b } -> { 116d000-116d19b } [412] moving module 04 { 102a000-107cc1f } -> { 111a000-116cc1f } [338976] moving module 03 { 1010000-1029473 } -> { 1100000-1119473 } [103540] Loading fiasco Loading sigma0 Loading moe find kernel info page... found kernel info page at 0x2000 Regions of list 'regions' [ 1000, 1aff] { b00} Kern fiasco [ 2000, 72fff] { 71000} Kern fiasco [ 73000, 7315f] { 160} Root mbi_rt [ 90000, 96edf] { 6ee0} Sigma0 sigma0 [ 98000, 9e177] { 6178} Sigma0 sigma0 [ 140000, 171d8f] { 31d90} Root moe [ 178000, 1934ab] { 1b4ac} Root moe [ 1000000, 100f58b] { f58c} Boot bootstrap [ 1100000, 19e7fff] { 8e8000} Root Module API Version: (87) experimental Sigma0 config ip:00090100 sp:00000000 Roottask config ip:00140260 sp:00000000 Starting kernel fiasco at 000012e0 Hello from Startup::stage2 Per_cpu_data_alloc: (orig: 0xf0061720-0xf00620e8) Number of IRQs available at this GIC: 128 FPU0: Arch: Unkn(2), Part: VFPv3(30), r: 5, v: 7, i: 41, t: hard, p: dbl/sngl ARM generic timer: freq=1000000 interval=1000 cnt=-262144 SERIAL ESC: allocated IRQ 89 for serial uart Not using serial hack in slow timer handler. Welcome to Fiasco.OC (arm)! L4/Fiasco.OC arm microkernel (C) 1998-2013 TU Dresden Rev: 8fd26ca-dirty compiled with gcc 4.8.3 for [] Build: #3 Mon Sep 19 18:55:38 CST 2016 Per_cpu_data_alloc: (orig: 0xf0061720-0xf00620e8) Allocate 2504 bytes (2KB) for CPU[4] local storage (offset=118a8e0, 0xf11ec000-0xf11ec9c8) cpu1: powered-up Per_cpu_data_alloc: (orig: 0xf0061720-0xf00620e8) cpu2: powered-up Allocate 2504 bytes (2KB) for CPU[1] local storage (offset=118c8e0, 0xf11ee000-0xf11ee9c8) cpu3: powered-up FPU1: Arch: Unkn(2), Part: VFPv3(30), r: 5, v: 7, i: 41, t: hard, p: dbl/sngl Calibrating timer loop... Per_cpu_data_alloc: (orig: 0xf0061720-0xf00620e8) done. Allocate 2504 bytes (2KB) for CPU[2] local storage (offset=118f8e0, 0xf11f1000-0xf11f19c8) Cache config: ON MDB: use page size: 30 FPU2: Arch: Unkn(2), Part: VFPv3(30), r: 5, v: 7, i: 41, t: hard, p: dbl/sngl ID_PFR[01]: 00001131 00011011Per_cpu_data_alloc: (orig: 0xf0061720-0xf00620e8) MDB: use page size: 21 Allocate 2504 bytes (2KB) for CPU[3] local storage (offset=11968e0, 0xf11f8000-0xf11f89c8) ID_[DA]FR0: 02010555 00000000 Cache config: ON MDB: use page size: 12 FPU3: Arch: Unkn(2), Part: VFPv3(30), r: 5, v: 7, i: 41, t: hard, p: dbl/sngl ID_PFR[01]: 00001131 00011011SID_MMFR[04]: 10101105 40000000 01240000 02102211 Cache config: ON ID_[DA]FR0: 02010555 00000000 ID_PFR[01]: 00001131 00011011ID_MMFR[04]: 10101105 40000000 01240000 02102211 ICPU[2]: goes to idle loop ID_[DA]FR0: 02010555 00000000 GID_MMFR[04]: 10101105 40000000 01240000 02102211 CPU[1]: goes to idle loop CPU[3]: goes to idle loop MA0: Hello! KIP @ 2000 allocated 4KB for maintenance structures SIGMA0: Dump of all resource maps RAM:------------------------ [0:0;fff] [4:73000;73fff] [0:74000;8ffff] [0:97000;97fff] [0:9f000;13ffff] [4:140000;171fff] [0:172000;177fff] [4:178000;193fff] [0:194000;10fffff] [4:1100000;19e7fff] [0:19e8000;3effffff] [0:40000000;7fffffff] IOMEM:---------------------- [0:80000000;ffffffff] MOE: Hello world MOE: found 2070816 KByte free memory MOE: found RAM from 73000 to 80000000 MOE: allocated 2047 KByte for the page array @0x194000 MOE: virtual user address space [0-bfffffff] MOE: rom name space cap -> [C:501000] BOOTFS: [1100000-1119474] [C:503000] l4re BOOTFS: [111a000-116cc20] [C:504000] ned BOOTFS: [116d000-116d19c] [C:505000] l4lx.cfg BOOTFS: [116e000-12c4470] [C:506000] io BOOTFS: [12c5000-12c50bd] [C:507000] arm-rv.io BOOTFS: [12c6000-16e7498] [C:508000] vmlinuz.arm BOOTFS: [16e8000-19e8000] [C:509000] ramdisk-arm.rd MOE: cmdline: moe rom/l4lx.cfg MOE: Starting: rom/ned rom/l4lx.cfg MOE: loading 'rom/ned' Ned says: Hi World! Ned: loading file: 'rom/l4lx.cfg' l4linux | libio: Warning: Query of 'vbus' failed! l4linux | PH 0 (t: 1) offs=00008000 vaddr=02000000 vend=024c8dd4 l4linux | phys= 02000000 ephys=024c8dd4 l4linux | f_sz= 00404594 memsz=004c8dd4 flgs=rwx l4linux | PH 1 (t: 4) offs=003b44e0 vaddr=023ac4e0 vend=023ac504 l4linux | phys= 023ac4e0 ephys=023ac504 l4linux | f_sz= 00000024 memsz=00000024 flgs=r-x l4linux | PH 2 (t: 1685382481) offs=00000000 vaddr=00000000 vend=00000000 l4linux | phys= 00000000 ephys=00000000 l4linux | f_sz= 00000000 memsz=00000000 flgs=rwx l4linux | Starting binary at 0x2000360, argc=8 argv=0xafff4f84 *argv=0xb1007ff0 argv0=rom/vmlinuz.arm l4linux | External resolver is at 0xa8000788 l4linux | ======> L4Linux starting... <======== l4linux | Linux version 3.16.0-l4 (li94575@li94575-Lenovo) (gcc version 4.8.3 20140320 (prerelease) (Sourcery CodeBench Lite 2014.05-29) ) #3 SMP Mon Sep 19 18:57:16 CST 2016 l4linux | Binary name: rom/vmlinuz.arm l4linux | This is an AEABI build. l4linux | Linux kernel command line (7 args): mem=64M l4x_cpus=4 l4x_cpus_map=0,1,2,3 console=ttyLv0 l4x_rd=rom/ramdisk-arm.rd root=1:0 ramdisk_size=4000 l4linux | CPU mapping (l:p)[4]: 0:0, 1:1, 2:2, 3:3 l4linux | Image: 020000a0 - 02600000 [6143 KiB]. l4linux | Areas: Text: 020000a0 - 023d6000 [3927kB] (a bit longer) l4linux | Data: 023d6000 - 023ff520 [165kB] l4linux | Initdata: 023ad000 - 023d57c0 [161kB] l4linux | BSS: 02404594 - 024c8dd4 [786kB] l4linux | Device scan: l4linux | Device scan done. l4linux | l4lx_thread_create: Created thread 413 (cpu0) (u:b3000e00, v:b3000c00, sp:023d7fa4) l4linux | main thread will be 413 l4linux | l4x_register_pointer_section: addr = 02000000 size = 5017600 l4linux | section-with-init: Virt: 0x2000000 to 0x24c8dd3 [4899 KiB] l4linux | section-with-init: Phys: 0x0042b000 to 0x008f3dd3, [4899 KiB] l4linux | Main thread running, waiting... l4linux | L4x: Memory size: 64MB l4linux | L4x: Setting superpages for main memory l4linux | L4x: Adjusted memory start: 02000000 l4linux | Main memory: Virt: 0x2600000 to 0x65fffff [65536 KiB] l4linux | Main memory: Phys: 0x01a00000 to 0x059fffff, [65536 KiB] l4linux | l4x: vmalloc area: 06600000 - 0e600000 l4linux | l4x_register_pointer_section: addr = 02000000 size = 5017600 l4linux | text: Virt: 0x2000000 to 0x24c8dd3 [4899 KiB] l4linux | text: Phys: 0x0042b000 to 0x008f3dd3, [4899 KiB] l4linux | Loading: rom/ramdisk-arm.rd l4linux | INITRD: Size of RAMdisk is 3072KiB l4linux | RAMdisk from 00002000 to 00302000 [3072KiB] l4linux | l4lx_thread_create: Created thread 419 (timer0) (u:b3000a00, v:00000000, sp:0245dfa0) Booting Linux on physical CPU 0x0 Linux version 3.16.0-l4 (li94575@li94575-Lenovo) (gcc version 4.8.3 20140320 (prerelease) (Sourcery CodeBench Lite 2014.05-29) ) #3 SMP Mon Sep 19 18:57:16 CST 2016 CPU: Fiasco [410fc075] revision 5 (ARMv7), cr=00000000 CPU: PIPT / VIPT nonaliasing data cache, VIPT nonaliasing instruction cache Machine: L4 Ignoring unrecognised tag 0x00000000 Forcing write-allocate cache policy for SMP Forcing shared mappings for SMP Memory policy: Data cache writealloc INITRD: 00002000 - 00302000 PERCPU: Embedded 5 pages/cpu @06509000 s6080 r0 d14400 u32768 Built 1 zonelists in Zone order, mobility grouping on. Total pages: 17716 Kernel command line: mem=64M l4x_cpus=4 l4x_cpus_map=0,1,2,3 console=ttyLv0 l4x_rd=rom/ramdisk-arm.rd root=1:0 ramdisk_size=4000 PID hash table entries: 512 (order: -1, 2048 bytes) Dentry cache hash table entries: 16384 (order: 4, 65536 bytes) Inode-cache hash table entries: 8192 (order: 3, 32768 bytes) Memory: 64728K/71680K available (3157K kernel code, 165K rwdata, 604K rodata, 161K init, 786K bss, 6952K reserved) Virtual kernel memory layout: vector : 0xbffff000 - 0xc0000000 ( 4 kB) fixmap : 0xffc00000 - 0xffe00000 (2048 kB) vmalloc : 0x06600000 - 0x0e600000 ( 128 MB) lowmem : 0x00000000 - 0x06600000 ( 102 MB) .text : Virtual kernel memory layout: vector : 0xbffff000 - 0xc0000000 ( 4 kB) fixmap : 0xffc00000 - 0xffe00000 (2048 kB) vmalloc : 0x06600000 - 0x0e600000 ( 128 MB) loHierarchical RCU implementation. NR_IRQS:220 l4timer: Using IRQ210 sched_clock: 64 bits at 1000kHz, resolution 1000ns, wraps every 2147483648000ns Console: colour dummy device 80x30 console [ttyLv0] enabled Calibrating delay loop...

3 7

Is L4Linux can use its existed Linux driver to access hardware device?
by Zhe Zhao 14 Sep '16

14 Sep '16

Hi, After read the basic ideas of L4Linux I have some questions related to it. Is L4Linux run as a paravirtualization VM on L4Re? Then the traditional Linux Process just still compiled against Linux system and put them together with L4Linux disk image? so it means L4Re works like qemu and Linuxl? or L4Re works as a micro kernel operating system, and L4Linux run as a process on it, all devices handled to L4Linux to reuse the drivers of L4Linux? and the previous Linux process compiled with Linux, but run direct on L4Re as L4 processes which communicate with L4Linux through IPC? it is quite confused after read a little of the codes, in L4Linux, seems there is some injection of L4 task stuffs inside thread_struct of Linux, but also have some vCPU related stuffs, can you help me about what is the real behavior of L4Linux? I'm sorry to ask this basic questions, but after some digging I feel quite confused about how it works. Thanks Br Alex

2 7

Partitioned Scheduling
by Edinger, Stefan 09 Sep '16

09 Sep '16

Hi, is it possible to have partitioned scheduling in the way that one can use different schedulers for different cpu-cores? And if yes, can I do the mapping from scheduler to core on runtime or is it done at compile time? Thanks, Stefan

4 7

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 September 2016