- l4-hackers - example.com Site

L4 on iPAQ question
by Christian Stüble 08 Nov '02

08 Nov '02

Hello, we just got some iPAQ 3870 Handhelds (very nice little devices :-) and I will try to port PERSEUS to it. Obviously, I should start with installing L4 onto it. But I don't know more than that I have installed the bootldr Bootloader that can boot images from Serial or CF (installing plain linux works well). Has someone else tried to install L4 on iPAQ and can give me hints how to do it? If not, what should I do first? What should I read? Next, L4Linux has to be ported to the StrongARM. Is there a L4Linux/ARM available somewhere? Is someone working on it or has a student that can do it? I can try to do so, but I fear here in Saarbruecken it is difficult to find students with experiences in Linux, C and OS in general. Maybe, I have to do it myself, but I will need help from the DROPS and L4KA people... Thanks & Regards, Chris -- Christian Stueble <stueble(a)cs.uni-sb.de> Saarland University, Dept. for Computer Science Security and Cryptography Group @Saarland University/DFKI Im Stadtwald, D-66123 Saarbruecken, Building 45, Room 534 Tel.: +49(681)302-5635 Fax.: +49(681)302 4631 Get public key from http://www-krypt.cs.uni-sb.de Fingerprint 8678 C5D3 CAD9 CD8C F1DD B8EC 202F 116A

1 0

NomadBIOS source availability
by Asger Henriksen 05 Nov '02

05 Nov '02

Hi all, Just wanted to let you know, that a read-only cvs repository has been set up for access to the sourcecode for nomadbios. It is still a little rough, but we will iron out whatever issues can/will arise. The cvs repo contains the source for the linux22 sourcetree, as well as the nomadbios server itself. In the download area of the website patches for Oskit and Fiasco has been added. Just do cvs -d:pserver:anonymous@udenbysvester.dk:/var/lib/cvs login <no password> cvs -d:pserver:anonymous@udenbysvester.dk:/var/lib/cvs co <module> where module is either linux22 or nomadbios. If you check out linux22 you must use the "guest" branch when checking out, otherwise nothing will work. Regards, Asger Henriksen

1 0

NomadBios sources now available
by Jacob Gorm Hansen 05 Nov '02

05 Nov '02

not pretty, but there. I've also put up our experimental fiasco & oskit patches. best, Jacob

1 0

NomadBIOS project
by Jacob Gorm Hansen 02 Nov '02

02 Nov '02

hi all, we have just put up a minimal website about our thesis project, titled 'NomadBIOS', which allows you to run multiple l4linuxes at once, and to hot-migrate them between host machines. The thesis is not exacly written yet, but in the meanwhile you might want to have a go at it. The website and download is at http://www.nomadbios.dk Have fun! Asger Henriksen & Jacob Gorm Hansen

1 0

RE: Re[8]: performance of L4Linux
by Volkmar Uhlig 29 Oct '02

29 Oct '02

> As far as I understand you want to find out whether: > - mklinux is really as bad as published > - L4Linux is really as fast as published > > Start with the first one. > > Then fetch a version of L4 from l4kabuild l4linux 2.0 against it and > try to find out about the second point. (I don't know whether l4ka's > version supports 4Mb pages and small address spaces but you will find > out whether they do or not) Hazelnut supports small spaces. We published a TR about the performance of small spaces for the P4 architecture using an L4Linux environment. See also: http://i30www.ira.uka.de/research/publications/ Actually, the relative syscall overhead using an adopted glibc was lower than the original numbers given by the SOSP paper. We ran L4Linux and the applications in small spaces. (Hazelnut supports different configurations for small spaces -- up to 2GB for the small space area.) - Volkmar

1 0

Re[4]: performance of L4Linux
by Igor Shmukler 29 Oct '02

29 Oct '02

> Well, then you should probably start reading the publications about L3 > and L4 of the last 7 years and you will find an in-detail analysis why > your statement could be taken as offensive rant... ;-) There are pretty > exact numbers of MACH's cache utilization and performance of IPC and the > structural problems of the communication mechanism. Well, I have done that (read papers). Cache utilization is very easy to fix. In fact when I did Mach VM overhauling (couple of years ago) it maybe took me one day to implement coloring. IPC issue was not discovered by l3/4 team, but rather was known issue for very long time. There are much older Utah and UW papers that address that. Anything else l3/4 team has up the sleeve? If I was to say that being tightly architectutre coupled is strange or that since l4 took almost everything out of the kernel, memory management could have been evicted as well (like eros microkernel does), then you may need to get offended. Otherwise, I don't see your point. L4 papers states that since there is no way to prove that Mach only causes 5% degradation statement is not valid. I say same logic applies to MkLinux vs. L4Linux. I am not trying to start am flame war. Nor do I claim that L4 is worthless. I simply want to see if there is a way to validate specific statement. If there is, that means that statement is valid, if there is not, then you tell me ... As simple as that. IS.

2 5

RE: Re[6]: performance of L4Linux
by Volkmar Uhlig 29 Oct '02

29 Oct '02

Just out of curiosity -- where do you want to get the not publicly available L4 x86-assembly kernel from to actually being able to redo the measurements? It is intellectual property of a company and not downloadable anywhere. Furthermore, if you redo the measurements on different hardware and you read the "On mk construction" paper you should know that MKs (at least L4 MKs) are highly optimized for one particular architecture. Hence, even when having the kernel and running it on different hardware your numbers are bogus and do not state anything except that you get a set of more numbers...

3 2

RE: Re[8]: performance of L4Linux
by Volkmar Uhlig 29 Oct '02

29 Oct '02

> > Just out of curiosity -- where do you want to get the not publicly > > available L4 x86-assembly kernel from to actually being > able to redo the > > measurements? It is intellectual property of a company and not > > downloadable anywhere. > I think I already have it. Otherwise, I'll make sure to contact you personally. Which won't help since I don't have it. So I'm surprised that you have an instance. > Well, I was under imperssion that L4 is highly optimized for > a processor (CPU), but not an PCI2ISA bridge or hard drive bandwidth. > Am I missing something? Please let me know, so that I don't > waste my time. So if you have a Pentium 1 and the kernel you should be fine. - Volkmar

1 0

performance of L4Linux
by Igor Shmukler 28 Oct '02

28 Oct '02

I read paper comparing performance of L4Linux of MkLinux (atop OSFMK) on x86 computer. Results are very interesting. I would really like to conduct benchmarks on my own. I know where to download Linux and L4Linux, but not the MkLinux for Intel. Does anyone still have binaries? If you do, please let me know how I could get these. Thank you, IS.

4 8

Local IPC identifier
by Jacob Gorm Hansen 25 Oct '02

25 Oct '02

hi, For some purposes when sending ipc to other threads in my task space, I find the need to explicitly state a taskid problematic. This information is redundant, and redundancy _always_ leads to problems. For me, the problem occurs when I move the l4linux server to a new task number, and another l4linux runs with my old number. All my threads get confused, and there is little I can do to help them (not without C&C at least, but that would still force me to use a taller task hierachy than I would like). What I need is something like 'if partner.id.task==0 then partner.id.task=current.id.task' in the kernel, and I suppose I will make such a hack myself. Is this what the rumoured 'lipc' feature in Pistachio is all about? Or is such a feature already present? thanks, Jacob

3 3