Hurd On Linux/Carrier Grade Linux From Scratch

[gui.fortaine@xxxxxxxxx (Guillaume FORTAINE)]

Misters,



Let me introduce myself : Guillaume FORTAINE, French Engineer in 
Computer Science. It's an honor for me to send to you this mail. I am 
currently doing researches about OS engineering.

I am currently involved in the Open Flight Linux project :

http://openflightlinux.org/


Mission Statement

?The Open FlightLinux Project is an open source paragon that strives to 
provide sustainable technology through superior engineering, 
next-generation design concepts, and open standards.?



My partners plan to obtain the Flight Certification Grade for this 
distribution, but I believe it is definitely impossible with a pristine 
Linux Kernel. I follow the HURD project since a long time and I believe 
that it will be the right step to make an HURD on Linux instead of 
L4/Coyotos. For sure, HURD people didn't do their homeworks and the 
following links below show this evidence.
Moreover, there is definitely some similarities between the HURD goals 
and the Carrier Grade Linux ones. So, why not to merge them into a 
common framework ?




http://www.bddebian.com/~wiki/hurd/ng/

CapabilityBasedMicrokernel



*Build Sytem*

-Carrier Grade Linux from scratch

http://www.linuxquestions.org/questions/showthread.php?t=395123

Dear All ,

I want to assemble Carrier Grade Linux from scratch.

As per CGL Requirement Spec 2.1 If I start with Linux Kernel 
<http://www.linuxquestions.org/questions/showthread.php?t=395123#> 2.6 
and glibc 2.3.3 I can meet more than 30% of the requirements as defined 
in the specification.

Other requirements can be met by patching the kernel by different OSS 
projects available on net.

I wanted to know how much time it would take for one person working for 
say 6 hrs in a day 5 days in a week to complete the assembly and comeout 
with a CG Linux distribution.

I hope to get some pointer from the community.

Cheers,
Santosh





*Libc*

http://www.eglibc.org/home

Embedded GLIBC (EGLIBC) is a variant of the GNU C Library (GLIBC) 
<http://www.gnu.org/software/libc/> that is designed to work well on 
embedded systems. EGLIBC strives to be source and binary compatible with 
GLIBC. EGLIBC's goals include reduced footprint, configurable 
components, better support for cross-compilation and cross-testing.




http://www.bddebian.com/~wiki/Hurd/HurdSpecificGlibcApi/


    Hurd-specific glibc API

These functions have meaning only under Hurd. They are needed to get 
port names that are used in native Hurd API (the RPC calls to servers). 
The |.defs| and |.h| files can be found in

|/include/hurd| when all development files are installed (Debian package 
|hurd-dev|.) Note that |.defs| are not included in C programs - they are 
used to produce |.h| files.







*Security*

-Capabilities

http://www.kernel.org/pub/linux/libs/security/linux-privs/README

You can get a copy of the defunct POSIX draft specification for
capabilities, ACLs and some other interesting things here:

  http://wt.xpilot.org/publications/posix.1e/

If you are looking for ACL patches for kernels, take a look here:

  http://acl.bestbits.at/download.html

As for this site, things that can be found here are:

Regular libcap (support for 2.6.x linux capabilities):

  http://www.kernel.org/pub/linux/libs/security/linux-privs/kernel-2.6/

If you want to play with them as they are partially implemented, you
may find it easier if you follow these instructions first:

  http://original.killa.net/infosec/cage/caps.html

Filesystem capability support (for 2.6.23+?) is being actively worked
on. You can obtain source code information here:

  http://www.kernel.org/pub/linux/libs/security/linux-privs/libcap2

Earlier experiments with file capabilities are to be found in:

  http://www.kernel.org/pub/linux/libs/security/linux-privs/old

Cheers

Andrew
2007/07/18






*Microkernel*

-Reliability :

http://ppacc.sourceforge.net/

About PPA
PPA - Precise Process Accounting is a Linux kernel patch for Carrier 
Grade environments.

Precise CPU and scheduling activity accounting aids five-nine 
availability, it hardens systems capacity management, execution 
performance and incident root cause.





-Userspace :

#klibc

http://tree.celinuxforum.org/CelfPubWiki/EarlyUserSpace

"Early User Space" is a phrase which describes work to create a user 
space prior to the "real" root file system coming up.

It involves adding an initramfs initialization step in the kernel boot 
sequence. This will theoretically allow for some current kernel bootup 
code to be moved from the kernel into user space. While it will continue 
to be tightly coupled with the kernel, this will allow some flexibility 
in booting, reduce kernel complexity, and could potentially reduce 
kernel footprint (in the future).



http://linux.derkeiler.com/Mailing-Lists/Kernel/2006-03/msg06641.html


http://git.kernel.org/?p=linux/kernel/git/hpa/linux-2.6-klibc.git;a=summary





#Drivers


http://liquidat.wordpress.com/2007/07/21/linux-kernel-2623-to-have-stable-userspace-driver-api/


    Linux kernel 2.6.23 to have stable userspace driver API

July 21st, 2007 ? liquidat

Tux <http://www.flickr.com/photos/liquidat/163197781/>
/Linus Torvalds included patches into the mainline tree which implement 
a stable userspace driver API into the Linux kernel./

The stable driver API was already announced a year ago 
<http://liquidat.wordpress.com/2006/08/30/new-driver-interface-for-linux-kernel/> 
by Greg Kroah-Hartman. Now the last patches 
<http://permalink.gmane.org/gmane.linux.kernel/557129> where uploaded 
and the API was included 
<http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=fc15bc817eecd5c13581adab2a182c07edededa0> 
in Linus? tree. The idea of the API is to make life easier for driver 
developers:

    This interface allows the ability to write the majority of a driver
    in userspace with only a very small shell of a driver in the kernel
    itself. It uses a char device and sysfs to interact with a userspace
    process to process interrupts and control memory accesses.







#FUSE

http://fuse.sourceforge.net/


  Introduction

With FUSE it is possible to implement a fully functional filesystem in a 
userspace program. Features include:

    * Simple library API
    * Simple installation (no need to patch or recompile the kernel)
    * Secure implementation
    * Userspace - kernel interface is very efficient
    * Usable by non privileged users
    * Runs on Linux kernels 2.4.X and 2.6.X
    * Has proven very stable over time





Welcome into my World ;) !

I look forward to your answer,

Best Regards,

Guillaume FORTAINE

-- 
"I consider life itself instinct for growth, for durability, for accumulation of forces, for power : where the will to power is lacking there is decline"

Friedrich Nietzsche


"Knowledge is Power."

Sir Francis Bacon