l4-hackers

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

2 participants
2880 discussions

using L4-Linux
by edmundo＠rano.demon.co.uk 15 Mar '99

15 Mar '99

The nearest I've got to running L4-Linux so far, apparently, was by dd-ing grub-ext2fs-floppy onto a floppy, adding "main" and "sigma0", which I compiled myself, then booting with the following menu.lst: title = Linux kernel=(fd0)/rmgr -nopentium -sigma0 module=(fd0)/main -nokdb module=(fd0)/sigma0 module=(fd0)/glinux.gz root=/dev/hda2 The first error that I get is: RMGR: ERROR: while parsing config file: syntax error at line 1, col 2 Does this matter? I don't know which config file it's talking about. There's a rmgr.cfg.linux on the floppy, but I'm not booting with it, because it made things worse when I tried using it. At "RMGR: WARNING: there were 1 boot errors - continue?" I press return ... Then I get a mixture of kernel start-up messages and L4 errors which move too quickly for me to note them exactly. There are several screens full of "sigma0 denies mapping", and there's a "Hmm, strange, that shouldn't happen any more", and the last kernel message, before the machine reboots by itself, is "INIT: version 2.75 booting". All quotations may be approximate. I just tried it again, for about the fifth time, and it didn't get as far this time. This time its last words were: Partition check: hda: hda1 hda2 VFS: Mounted root (ext2 filesystem) readonly. wake_up_process: ipc failed wake_up_process: ipc failed wake_up_process: ipc failed ... wake_up_process: ipc failed Can anyone suggest what I can try next? Does anyone have this working? I'm using the files dated 981207 and it's a 486 machine that I'm using. (The other machines that I tried grind to a halt very early on with "space.cc:179: failed assertion `size == PAGE_SIZE'".) Does someone have working binaries of "main" and "sigma0" that I could try, to see if there's something strange about my hardware? The Hello World server does run, by the way. Edmund

1 0

"relocation truncated"
by edmundo＠rano.demon.co.uk 15 Mar '99

15 Mar '99

I wrote to Hohmuth about this, but he might be away or busy ... http://os.inf.tu-dresden.de/fiasco/faq.html#5992 says: [v0.5 981012] Build problem when building oskit/examples/multiboot-smp: "relocation truncated" I cannot reproduce this problem. However, as this binary isn't needed, you can work around this problem by "touch oskit/examples/multiboot-smp". (hohmuth 98/10/24) I got exactly this error when I tried building Fiasco. It was preceded by a warning coming from boot.S, which might be connected, so I'm copying the edited highlights below. Does anyone with knowledge of x86 assembler understand this problem, and can they fix it? How come it only affects some people? Is it caused by different versions of the assembler? Isn't the assembler part of gcc? I'm using standard gcc 2.7.2.3, so which other packages should I check the versions of? I'm using Debian's binutils version 2.9.1.0.19a-2, for example, as released with Debian 2.1. I got the same assembler warning several times when building L4-Linux. Edmund gcc -c -MD -DHAVE_CONFIG_H -I. -I../libsmp/x86 -I- -I.. -I.. -I../flux/c -nostdinc -DASSEMBLER -MD ../libsmp/x86/boot.S /tmp/cca03216: Assembler messages: /tmp/cca03216:538: Warning: warning: missing prefix `*' in absolute indirect address, maybe misassembled! ... ... ... ld -Ttext 100000 -L../lib \ -o multiboot-smp ../lib/multiboot.o multiboot-smp.o \ -lsmp -lkern -lmc -llmm ../lib/crtn.o ../lib/libsmp.a(boot.o): In function `_SMP_TRAMP_START_': boot.o(.text+0x29): relocation truncated to fit: R_386_16 _SMP_TRAMP_32_ENTRY_ make[2]: *** [multiboot-smp] Error 1 make[2]: Leaving directory `/usr/local/tmp/L4/l4/oskit/examples' make[1]: *** [examples/all.MAKE] Error 2 make[1]: Leaving directory `/usr/local/tmp/L4/l4/oskit' make: *** [oskit] Error 2

3 3

send/receive confusion
by edmundo＠rano.demon.co.uk 15 Mar '99

15 Mar '99

Dear Hackers, I've been studying the descriptions of IPC in the L4 Reference Manual (x86) and the L4 User Manual (MIPS) and would be grateful if someone could confirm to me that I've correctly understood some points which are not explicitly stated in those manuals. Here's are my hypotheses/deductions; please tell me if I'm right or wrong! (1) Both the "send message descriptor" and the "receive message descriptor" allow in-line data, strings and fpages to be transmitted in both directions. In fact, the names are confusing and they might better be referred to as "client message descriptor" and "server message descriptor" as the only differences between them relate to the way the other party is addressed - plus one little optimisation for the case where a "receive message descriptor" wants to receive just an fpage. (2) If you want to transmit some data from thread S to thread T, one way of doing it is to get thread S to do a SEND operation and thread T to do a RECEIVE FROM operation, but you could equally well get thread S to do a RECEIVE FROM and thread T to do a SEND, as the "RECEIVE FROM" and "SEND" IPC types have nothing to do with "receive" and "send", really. (3) When you do a CALL operation, specifying both a "send message descriptor" and a "receive message descriptor" in the same system call, in most cases you probably only want to send data with the former and only receive data with the latter. Since both descriptors have space for both directions, you can use the same message descriptor as both send message descriptor and receive message descriptor, if you want. But you don't have to. Edmund

2 3

Re: send/receive confusion
by gernot＠cse.unsw.edu.au 14 Mar '99

14 Mar '99

Edmundo, I think I now understand the source of your confusion. In the Intel version of the L4 manual (at least in my copy of Sep 96), on p25 where the "rcv descriptor" is explained, the option "mem" has the top 30 bits of the descriptor word as "*snd msg/4". That should, of course, read "*rcv msg/4". Also, further to what I wrote in my previous mail: GH> - one or more page mappings or grants, provided that GH> - the "m" bit is set in the send descriptor, and GH> - the beginning of the sender's direct string (starting with the GH> register part) contains at least one valid "send fpage", and GH> - the receive descriptor either has the form of a valid "receive GH> fpage", or points to a "message" struct containing a valid "receive GH> fpage", and GH> - there is no error The second last bullet should read: - the receive descriptor either has the form of a valid "receive fpage" and the "m" bit is set, or points to a "message" struct containing a valid "receive fpage", and Gernot -- Gernot Heiser ,--_|\ School of Computer Sci. & Engin. Phone: +61 2 9385 5156 / \ The University of NSW Fax: +61 2 9385 5995 \_,--._* Sydney, Australia 2052 E-mail: G.Heiser(a)unsw.edu.au v http://www.cse.unsw.edu.au/~gernot PGP fingerprint: 94 1E B8 28 25 FD 7C 94 20 10 92 E5 0B FF 39 8F

1 0

implementations of L4
by Edmund GRIMLEY EVANS 12 Mar '99

12 Mar '99

I've been trying to find out about existing implementations of L4. Perhaps people could correct or add to what I have discovered: * The original L4 by Jochen Liedtke. Written in assebler. For i486 and i586. NOT FREE and therefore utterly useless :-) * L4/MIPS from Australia. Written in assembler? GNU. * L4/Alpha from Dresden and Australia. Written in assembler? Not yet released but probably GNU? * Fiasco from Dresden. Written in C++, and designed for portabilty rather than efficiency, but only runs on i486 and i586 at present. GNU. * A version of L4 for x86 from Dresden written in C. I saw this mentioned in the list archive. Is it really separate from Fiasco? * L4/ARM in assembler. Unfinished? I saw this in the archive, too. If this list could be corrected and completed, could it be put on the L4 website, with links, as it must be what a lot of casual readers are interested in knowing? Edmund

2 1

486/Pentium version problems
by Edmund GRIMLEY EVANS 12 Mar '99

12 Mar '99

Following the instructions at os.inf.tu-dresden.de/fiasco/use.html I built Fiasco and the Hello World server worked. Excellent! Then I tried building L4-Linux and it didn't work ... Then I noticed that grub-ext2fs-floppy apparently already contains everything you need to run L4-Linux except for L4 itself. So I just copied the fiasco "main" that I had built onto the floppy image as supplied, hacked menu.lst a bit at random, and tried that. That didn't work either, but the thing that caught my attention was a message something like this: RMGR: L4/Pentium ... I was running this on a 486. So my questions are: Do the same executables (main, rmgr, sigma0, glinux.gz) run on both 486 and Pentium, or do different versions have to be built? If the latter, are the versions in grub-ext2fs-floppy supposed to work only on Pentium? If so, how can I build a version for 486 when compiling on a 686 machine? (And also, could the code be made to run on both? It would make life easier ...) Edmund

1 0

ADV: New Daul Mount Copy Holder
by remove＠1vm.com 24 Feb '99

24 Feb '99

Amazing new copy holder can be mounted directly on your computer monitor or it can be used freestanding. Unique design is specifically engineered to increase your efficiency while reducing eye strain, stress, and fatigue. Quality Rubbermaid features include adjustable viewing positions, non-glare surface, spring load paper clamp and detachable line guide. Folds flat for easy out-of-the-way storage. This isn't a cheap clip on, but a full platform copy holder. Special internet savings click here http://www.1vm.com/~rcc and save over $20 from comparable models. Sender: Robert Craig Corp. Phone: 1-800-890-7510 Email: robertcraigcorp(a)prodigy.net Details: http://www.1vm.com/~rcc Remove: REMOVE(a)1vm.com To be removed from future service notifications reply to: REMOVE(a)1vm.com

1 0

Big break!
by Ronald 24 Feb '99

24 Feb '99

Start your own 1-900 business! People are making $$$ week, after week in the 1-900 business. We'll teach you all of our "incredible secrets that will take your new exciting business to a whole new level! It's The Simplest and Most Exciting Business You Could Ever Start! *You'll use our "state" of the art equipment! *You'll use our "Live 1 on 1 Psychics" & "Chat Line" girls! *You'll use our incredible Date Line program(s)! No chargbacks! Quick payouts! No expertise needed! Complete programs start at only $99 (no additional charges) The only thing you'll have to do is advertise! This is an excellent turn key business. We also have excellent turn key programs if you want to own your own "top" of the line adult web site. ACT NOW!!! For a free color brochure: reply to: mailto:megagreen@usa.net?subject=brochure with the following information please: NAME:____________________________ ADDRESS:____________________________ CITY/STATE/ZIP:____________________________ PHONE:____________________________ (optional) remove at: mailto:xpoc@eudoramail.com?subject=remove

1 0

Re: L4 Mapping question
by Michael Hohmuth 23 Feb '99

23 Feb '99

"Adam 'WeirdArms' Wiggins" <awiggins(a)cse.unsw.edu.au> writes: > What TID does the kernel use for its ipc to Sigma0 to request > memory mapping for kernel usage. L4/x86 uses task 1, local thread 0, for communication with Sigma0. The Fiasco kernel doesn't do any IPC operations on behalf of itself -- so no thread id here. Michael -- hohmuth(a)innocent.com, hohmuth(a)sax.de http://www.sax.de/~hohmuth/

1 0

Re: L4 Mapping question
by Michael Hohmuth 23 Feb '99

23 Feb '99

Robert Kaiser <rob(a)sysgo.de> writes: > I was aware that the bookkeeping would have to be in the kernel, otherwise > flush couldn't work as described in the manual. Only, how does it work ? > > Theoretically, each task could map each of it's pages to up to 2047 other > tasks. So, the kernel would need to maintain ~2K entries per mapped page > and task. This can't be! The only explanation I can think of right now > is that the kernel allocates room for entries dynamically, assuming that > the actual number of required entries will always be far less than > the theoretical limit. But that would mean you can break the system > simply by doing a big enough number of mappings. Each mapping of a physical page from virtual address VA1 in a task T1 to VA2 in T2 is kept track of using a kernel data structure called ``mapping entry.'' This data structure is allocated dynamically, but from a fixed-size pool (where the size can be specified at compile time or boot time). It's true that you can break the system by establishing an excessive number of mappings. There are several ways to deal with that, but I don't think any of them has been implemented in the freely-available research kernels; one way is detailed in Jochen Liedtke et al.'s paper ``Preventing Denial-of-Service Attacks on a µ-Kernel for WebOSes'' which is available from <URL:http://os.inf.tu-dresden.de/papers_ps/jochen/Denhot.ps>. > On Mon, 22 Feb 1999, Volkmar Uhlig wrote: > > In the current L4-implementation of Jochen Liedtke, one mapping entry is 16 > > Bytes.The Alpha implementation (done by me) needs 32 Bytes per entry (64 > > bit address space - and a lot of trix). The Fiasco kernel does not use > > lists. They need about 4 Bytes per entry + an some overhead per frame (I > > left bevore it was finished) > > I am getting the distinct feeling that I am missing something > very simple :-(. So, at the risk of looking like a complete fool: > How do you keep track of up to 2K tasks in a 16 Byte entry ? The 16-bytes structure is what I described as a ``mapping entry'' above. Mapping entries belonging to one physical page are linked to each other in a binary-tree structure using pointers. Michael -- hohmuth(a)innocent.com, hohmuth(a)sax.de http://www.sax.de/~hohmuth/

1 0

Jump to page:

2025

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