Nabble - Micro Kernel

RE: UTCB-Area, Local ThreadIds

2009-12-16T09:36:10Z

> 1: Is there a way, to get the address of a UTCB area of an address space or do
> I have to store/remember it after creation of the address space?
> I've read there is just a hack with MyLocalId() to get the UTCB area of the
> root-task. Is that a technical issue or shouldn't I just get it?

You have to remember it afaik. It's not a technical issue, it's just that user can keep it w/o problems and the kernel doesn't need another interface then.

> 2: In which parts of the memory I am generally allowed to write after startup,
> without accidently overwriting important data? Do I have the entire Virtual
> Memory? Here is a part of my KIP:

So conceptually, you should write into virtual address area backed with normal physical memory, since all other areas have special semantics. Sigma0 helps you a little bit there, since it's implemented in a way that it hands out normal memory on page faults but wants to be asked explicitly (via IPC) for handing out non-memory areas (architecture specific, device mem, ...).

-Jan
--
Jan Stoess
KIT/UKa System Architecture Group
Phone: +49 (721) 608-4056
Fax: +49 (721) 608-7664
eMail: stoess@...

> -----Original Message-----
> From: l4ka-bounces@... [mailto:l4ka-bounces@...-
> karlsruhe.de] On Behalf Of Martin
> Sent: Saturday, December 12, 2009 8:59 PM
> To: l4ka@...
> Subject: UTCB-Area, Local ThreadIds
>
> Hello,
>
> I don't have so much experience with the L4-Kernel yet and got some
> Questions. I am working on IA32 and Kernel-Verion 0.4.0:
>
> 1: Is there a way, to get the address of a UTCB area of an address space or do
> I have to store/remember it after creation of the address space?
> I've read there is just a hack with MyLocalId() to get the UTCB area of the
> root-task. Is that a technical issue or shouldn't I just get it?
>
> 2: In which parts of the memory I am generally allowed to write after startup,
> without accidently overwriting important data? Do I have the entire Virtual
> Memory? Here is a part of my KIP:
>
> Memory regions (17):
> Physical: 0x00000000 - 0xffffffff shared
> 0x00000000 - 0x0009efff conventional
> 0x0009f000 - 0x0009ffff architecture specific (2)
> 0x000e8000 - 0x000fffff architecture specific (2)
> 0x00100000 - 0x07feffff conventional
> 0x07ff0000 - 0x07ffffff architecture specific (3)
> 0xfffbc000 - 0xfffbffff architecture specific (2)
> 0xfffc0000 - 0xffffffff architecture specific (2)
> 0x06ff0000 - 0x07feffff reserved
> 0x000a0000 - 0x000bffff shared
> 0x000c0000 - 0x000effff shared
> 0x00020000 - 0x000297ff bootloader specific (2)
> 0x00300000 - 0x003037ff bootloader specific (2)
> 0x00001000 - 0x00001fff bootloader specific (1)
> 0x00002000 - 0x00002fff bootloader specific (1)
> 0x00100000 - 0x00141fff reserved
> Virtual: 0x00000000 - 0xbfffffff conventional
>
> Memory regions:
> Main mem 0x00000000 - 0x00000000
> Reserved mem0 0x00100200 - 0x00142000
> Reserved mem1 0x00000000 - 0x00000000
> Dedicated mem0 0x00000000 - 0x00000000
> Dedicated mem1 0x00000000 - 0x00000000
> Dedicated mem2 0x00000000 - 0x00000000
> Dedicated mem3 0x00000000 - 0x00000000
> Dedicated mem4 0x00000000 - 0x00000000
>
>
> Where is the root-tasks data? Is it in the Reserved mem0?
>
>
> 3: How do I set/manipulate local ThreadIds? ThreadControl() is just working
> with global IDs. After creation of Threads, all local IDs are 0, this shouldn't be,
> right?
>
>
> Thank you very much,
> Gnuffel
> --
> Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 -
> sicherer, schneller und einfacher! http://portal.gmx.net/de/go/chbrowser

From forum: l4ka@ira.uni-karlsruhe.de

RE: Building a working instance of L4Ka::Pistachio

2009-12-16T09:36:09Z

Hard to tell what the real problem might be. Can you, since this is the simplest check, build pistachio with an older compiler? New compiler and binutils versions sometimes cause headaches, and I’ve not checked build against newer versions for quite some time. You can find a list of known-working compilers here:

http://l4ka.org/projects/pistachio/buildstatus/hg.php

-Jan

Jan Stoess

KIT/UKa System Architecture Group

Phone: +49 (721) 608-4056

Fax: +49 (721) 608-7664

eMail: stoess@...

From: l4ka-bounces@... [mailto:l4ka-bounces@...] On Behalf Of Owen Shepherd
Sent: Thursday, December 03, 2009 7:45 PM
To: l4ka@...
Subject: Building a working instance of L4Ka::Pistachio

OK, I'm currently attempting to build Pistachio for x86 32-bit. I've built a pretty standard kernel (Pentium 2/3, KDB, spinwheels and tracing are the only things I've enabled for now). My aim is an AMD64 kernel with more features, but it's always best to start with a smaller base and get that working first.

I've then built the usermode tasks, and built a grub disk using the included grubdisk.img. Problem is, the kernel crashes on both emulators I've tried it with (Bochs and QEMU).

The last message I get from the kernel before it crashes is "Initializing GDT (Preliminary)". The Bochslog preceding this is

00128039070e[CPU0 ] LTR: doesn't point to an available TSS descriptor!

00128039070e[CPU0 ] interrupt(): gate descriptor is not valid sys seg (vector=0x0d)

00128039070e[CPU0 ] interrupt(): gate descriptor is not valid sys seg (vector=0x08)

00128039070i[CPU0 ] CPU is in protected mode (active)

00128039070i[CPU0 ] CS.d_b = 32 bit

00128039070i[CPU0 ] SS.d_b = 32 bit

00128039070i[CPU0 ] EFER = 0x00000000

00128039070i[CPU0 ] | RAX=000000000000f012 RBX=00000000000000f0

00128039070i[CPU0 ] | RCX=0000000000000f00 RDX=0000000000000000

00128039070i[CPU0 ] | RSP=000000000015648c RBP=0000000000802028

00128039070i[CPU0 ] | RSI=000000000000df00 RDI=000000000003eaa0

00128039070i[CPU0 ] | R8=0000000000000000 R9=0000000000000000

00128039070i[CPU0 ] | R10=0000000000000000 R11=0000000000000000

00128039070i[CPU0 ] | R12=0000000000000000 R13=0000000000000000

00128039070i[CPU0 ] | R14=0000000000000000 R15=0000000000000000

00128039070i[CPU0 ] | IOPL=0 ID vip vif ac vm RF nt of df if tf sf zf af PF cf

00128039070i[CPU0 ] | SEG selector base limit G D

00128039070i[CPU0 ] | SEG sltr(index|ti|rpl) base limit G D

00128039070i[CPU0 ] | CS:0008( 0001| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | DS:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | SS:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | ES:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | FS:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | GS:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | MSR_FS_BASE:0000000000000000

00128039070i[CPU0 ] | MSR_GS_BASE:0000000000000000

00128039070i[CPU0 ] | RIP=00000000001532dc (00000000001532dc)

00128039070i[CPU0 ] | CR0=0xe0010011 CR2=0x0000000000000000

00128039070i[CPU0 ] | CR3=0x00134000 CR4=0x00000090

(0).[128039070] [0x001532dc] 0008:00000000001532dc (unk. ctxt): ltr word ptr ss:[esp+0x10] ; 0f005c2410

00128039070e[CPU0 ] exception(): 3rd (13) exception with no resolution, shutdown status is 00h, resetting

I think we can safely assume that something has gone wrong loading the TSS, a general protection fault has been raised during the LTR, then a double fault has followed that.

Is this a bug in the L4 sources, or an issue with my build environment? The important information is

$ uname -a

Linux sidewinder 2.6.31-15-generic #50-Ubuntu SMP Tue Nov 10 14:53:52 UTC 2009 x86_64 GNU/Linux

$ gcc -v

Using built-in specs.

Target: x86_64-linux-gnu

Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.4.1-4ubuntu8' --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 --program-suffix=-4.4 --enable-nls --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc --disable-werror --with-arch-32=i486 --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu

Thread model: posix

gcc version 4.4.1 (Ubuntu 4.4.1-4ubuntu8)

$ ld -v

GNU ld (GNU Binutils for Ubuntu) 2.20

From forum: l4ka@ira.uni-karlsruhe.de

UTCB-Area, Local ThreadIds

2009-12-12T11:58:51Z

Hello,

I don't have so much experience with the L4-Kernel yet and got some Questions. I am working on IA32 and Kernel-Verion 0.4.0:

1: Is there a way, to get the address of a UTCB area of an address space or do I have to store/remember it after creation of the address space?
I've read there is just a hack with MyLocalId() to get the UTCB area of the root-task. Is that a technical issue or shouldn't I just get it?

2: In which parts of the memory I am generally allowed to write after startup, without accidently overwriting important data? Do I have the entire Virtual Memory? Here is a part of my KIP:

Memory regions (17):
Physical: 0x00000000 - 0xffffffff shared
0x00000000 - 0x0009efff conventional
0x0009f000 - 0x0009ffff architecture specific (2)
0x000e8000 - 0x000fffff architecture specific (2)
0x00100000 - 0x07feffff conventional
0x07ff0000 - 0x07ffffff architecture specific (3)
0xfffbc000 - 0xfffbffff architecture specific (2)
0xfffc0000 - 0xffffffff architecture specific (2)
0x06ff0000 - 0x07feffff reserved
0x000a0000 - 0x000bffff shared
0x000c0000 - 0x000effff shared
0x00020000 - 0x000297ff bootloader specific (2)
0x00300000 - 0x003037ff bootloader specific (2)
0x00001000 - 0x00001fff bootloader specific (1)
0x00002000 - 0x00002fff bootloader specific (1)
0x00100000 - 0x00141fff reserved
Virtual: 0x00000000 - 0xbfffffff conventional

Memory regions:
Main mem 0x00000000 - 0x00000000
Reserved mem0 0x00100200 - 0x00142000
Reserved mem1 0x00000000 - 0x00000000
Dedicated mem0 0x00000000 - 0x00000000
Dedicated mem1 0x00000000 - 0x00000000
Dedicated mem2 0x00000000 - 0x00000000
Dedicated mem3 0x00000000 - 0x00000000
Dedicated mem4 0x00000000 - 0x00000000

Where is the root-tasks data? Is it in the Reserved mem0?

3: How do I set/manipulate local ThreadIds? ThreadControl() is just working with global IDs. After creation of Threads, all local IDs are 0, this shouldn't be, right?

Thank you very much,
Gnuffel
--
Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3.5 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/chbrowser

From forum: l4ka@ira.uni-karlsruhe.de

Re: L4 microkernel as a user-level program

2009-12-12T06:51:39Z

I was thinking an interesting project would be to get L4 ported to vx32. vx32 is a really lightweight virtualization environment that runs on top of Mac OS X, Linux and FreeBSD, and one of the authors actually ported the Plan 9 operating system to it, and it's bundled with the vx32 source code.

In fact, I thought it'd be good to port Plan 9 to L4 and L4 to vx32, then L4 linux could supply device drivers to Plan 9 :-).

This would satisfy some very practical goals for a somewhat small community as well as be a nice platform to build and run both L4 and Plan 9 programs without having to use a full fledged emulation environment.

vx32 can be found here: http://pdos.csail.mit.edu/~baford/vm/

(Russ Cox is one of the engineers behind Google's Go programming language, and the primary guy behind the plan 9 userspace port)

Dave

On Dec 11, 2009, at 11:00 PM, Michael FIG wrote:

> Hi,
>
> Norman Feske <norman.feske@...> writes:
>
>> there is a variant of the L4/Fiasco kernel called Fiasco UX that
>> can be executed as Linux program.
>
> Thank you very much for this pointer.
>
> I have some experience with User Mode Linux, ptrace, and other such
> black magic. IMO, it would be too much work and be of limited use to
> me to take the same approach as Fiasco UX... I am targeting multiple
> platforms (such as Windows, *BSD, Linux, and Mac OS X) and have no
> need for binary compatibility with other L4 systems.
>
> So, I'll poke around some more, and see if there is any other advice
> from people on the list. I'm thinking the best approach would be just
> to create a new architecture or two, such as "posix32", "win32", and
> write glue and stubs that is portable to many systems rather than just
> to one.
>
> By using a single-process model and thereby giving up binary
> compatibility, there is a chance of recovering high performance. My
> motivation is developing a VM for a new programming language platform.
> The language can provide some of the safety guarantees that I'll be
> losing with the single-process design.
>
> Of course, once the language gains enough applications to be useful by
> itself, I can recommend that its users install it on a "real" L4
> kernel running on bare hardware.
>
>> As far as I understand, the approach is not tied to Fiasco but could
>> be applied to Pistachio as well. If there existed a user-level version
>> of Pistachio, I'd be one of its users for sure ;-)
>
> Let me know if the above meshes with your needs!
>
> Thanks,
>
> --
> Michael FIG <michael@...> //\
> http://michael.fig.org/ \//
>

From forum: l4ka@ira.uni-karlsruhe.de

Re: L4 microkernel as a user-level program

2009-12-11T23:00:42Z

Hi,

Norman Feske <norman.feske@...> writes:

> there is a variant of the L4/Fiasco kernel called Fiasco UX that
> can be executed as Linux program.

Thank you very much for this pointer.

I have some experience with User Mode Linux, ptrace, and other such
black magic. IMO, it would be too much work and be of limited use to
me to take the same approach as Fiasco UX... I am targeting multiple
platforms (such as Windows, *BSD, Linux, and Mac OS X) and have no
need for binary compatibility with other L4 systems.

So, I'll poke around some more, and see if there is any other advice
from people on the list. I'm thinking the best approach would be just
to create a new architecture or two, such as "posix32", "win32", and
write glue and stubs that is portable to many systems rather than just
to one.

By using a single-process model and thereby giving up binary
compatibility, there is a chance of recovering high performance. My
motivation is developing a VM for a new programming language platform.
The language can provide some of the safety guarantees that I'll be
losing with the single-process design.

Of course, once the language gains enough applications to be useful by
itself, I can recommend that its users install it on a "real" L4
kernel running on bare hardware.

> As far as I understand, the approach is not tied to Fiasco but could
> be applied to Pistachio as well. If there existed a user-level version
> of Pistachio, I'd be one of its users for sure ;-)

Let me know if the above meshes with your needs!

Thanks,

--
Michael FIG <michael@...> //\
http://michael.fig.org/ \//

From forum: l4ka@ira.uni-karlsruhe.de

Re: L4 microkernel as a user-level program

2009-12-11T22:36:53Z

Hi Michael,

> I am interested in contributing to the L4 project by porting the
> microkernel interfaces to run in a simulation mode (i.e. as a regular
> application) on top of existing popular OS platforms. This would
> allow applications to migrate to using the L4 interfaces as a library
> without having to give up the amenities of their current operating
> system. The point would be to build momentum so that people write
> applications that benefit from being run on L4-based systems.
>
> Are there any people out there who are interested in this?

there is a variant of the L4/Fiasco kernel called Fiasco UX that
can be executed as Linux program. At the Dresden OS Group, it was
(and probably is) used a lot for developing user-level code. You
can find more more information at the project's website:

http://os.inf.tu-dresden.de/fiasco/ux/

I particularly recommend Udo Steinberg's studdy thesis, which offers
a great deal of technical detail:

http://os.inf.tu-dresden.de/papers_ps/steinberg-beleg.pdf

As far as I understand, the approach is not tied to Fiasco but could
be applied to Pistachio as well. If there existed a user-level version
of Pistachio, I'd be one of its users for sure ;-)

Norman

From forum: l4ka@ira.uni-karlsruhe.de

L4 microkernel as a user-level program

2009-12-11T22:16:42Z

Hi,

I'm new to the L4 world, so I'll need to learn my way around a little
before I'll be able to do much. However, I'm writing to ask about an
area that I find important for kernel adoption by application
programmers.

I am interested in contributing to the L4 project by porting the
microkernel interfaces to run in a simulation mode (i.e. as a regular
application) on top of existing popular OS platforms. This would
allow applications to migrate to using the L4 interfaces as a library
without having to give up the amenities of their current operating
system. The point would be to build momentum so that people write
applications that benefit from being run on L4-based systems.

Are there any people out there who are interested in this?

Thanks,

--
Michael FIG <michael@...> //\
http://michael.fig.org/ \//

From forum: l4ka@ira.uni-karlsruhe.de

Codezero Microkernel v0.2 Released

2009-12-07T04:54:38Z

I would like to announce that Codezero Microkernel v0.2 has been released.

On this release, the microkernel is now fully capability checked, and we
introduced the notion of containers to provide isolated execution
environments. Please see below for a brief description.

1.) Containers

In Codezero containers provide the architectural infrastructure for
isolation of execution environments. Each container is allocated with a
set of address spaces, threads, and any other resource that would
otherwise be globally available on the platform, such as virtual and
physical memory.

Using the simple notion of containers, it is possible to build any type
of software design hierarchy. A hierarchical client/server design, or
multi-threaded standalone applications are both possible scenarios.

2.) Capabilities

Capabilities protect all resources maintained by the kernel. Currently,
all system calls are protected by capabilities, except a few trivial
ones that have been left out. Physical and virtual memory, typed memory
pools, and inter-process communication are among other resources that
are protected by capability checking.

Capabilities build upon the foundation of containers, providing a
fine-grained security architecture, inside and among container boundaries.

We also introduced a capability control system call, by which the
capabilities may be shared, granted and modified in different ways,
enabling a highly flexible and configurable security architecture.

3.) Configuration

We attempted to present the above two notions in an easily manageable
kernel configuration system. Containers, their parameters and
capabilities may be adjusted easily at configuration time, using our
CML2-based kernel configuration system.

We are aiming to evolve the L4 API into the future by feedback, hence
this email post. Codezero Project is open with a GPLv3 license.

If you are interested to learn more, download by:

git-clone git://git.l4dev.org/codezero.git

A useful jump-start guide:

http://www.l4dev.org/doku.php?id=getting_started_quick

API reference man pages:

man -M codezero/docs/man pagename

Thanks,

--
Bahadir Balban

From forum: l4ka@ira.uni-karlsruhe.de

Codezero Microkernel v0.2 Released

2009-12-07T04:52:39Z

I would like to announce that Codezero Microkernel v0.2 is released.

On this release, the microkernel is now fully capability checked, and we
introduced the notion of containers to provide isolated execution
environments. Please see below for a brief description.

1.) Containers

In Codezero containers provide the architectural infrastructure for
isolation of execution environments. Each container is allocated with a
set of address spaces, threads, and any other resource that would
otherwise be globally available on the platform, such as virtual and
physical memory.

Using the simple notion of containers, it is possible to build any type
of software design hierarchy. A hierarchical client/server design, or
multi-threaded standalone applications are both possible scenarios.

2.) Capabilities

Capabilities protect all resources maintained by the kernel. Currently,
all system calls are protected by capabilities, except a few trivial
ones that have been left out. Physical and virtual memory, typed memory
pools, and inter-process communication are among other resources that
are protected by capability checking.

Capabilities build upon the foundation of containers, providing a
fine-grained security architecture, inside and among container boundaries.

We also introduced a capability control system call, by which the
capabilities may be shared, granted and modified in different ways,
enabling a highly flexible and configurable security architecture.

3.) Configuration

We attempted to present the above two notions in an easily manageable
kernel configuration system. Containers, their parameters and
capabilities may be adjusted easily at configuration time, using our
CML2-based kernel configuration system.

We are aiming to evolve the L4 API into the future by feedback, hence
this email post. Codezero Project is open with a GPLv3 license.

If you are interested to learn more, download by:

git-clone git://git.l4dev.org/codezero.git

A useful jump-start guide:

http://www.l4dev.org/doku.php?id=getting_started_quick

API reference man pages:

man -M codezero/docs/man pagename

Thanks,

--
Bahadir Balban

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Building a working instance of L4Ka::Pistachio

2009-12-03T10:44:43Z

I've then built the usermode tasks, and built a grub disk using the included grubdisk.img. Problem is, the kernel crashes on both emulators I've tried it with (Bochs and QEMU).

The last message I get from the kernel before it crashes is "Initializing GDT (Preliminary)". The Bochslog preceding this is

00128039070e[CPU0 ] LTR: doesn't point to an available TSS descriptor!

00128039070e[CPU0 ] interrupt(): gate descriptor is not valid sys seg (vector=0x0d)

00128039070e[CPU0 ] interrupt(): gate descriptor is not valid sys seg (vector=0x08)

00128039070i[CPU0 ] CPU is in protected mode (active)

00128039070i[CPU0 ] CS.d_b = 32 bit

00128039070i[CPU0 ] SS.d_b = 32 bit

00128039070i[CPU0 ] EFER = 0x00000000

00128039070i[CPU0 ] | RAX=000000000000f012 RBX=00000000000000f0

00128039070i[CPU0 ] | RCX=0000000000000f00 RDX=0000000000000000

00128039070i[CPU0 ] | RSP=000000000015648c RBP=0000000000802028

00128039070i[CPU0 ] | RSI=000000000000df00 RDI=000000000003eaa0

00128039070i[CPU0 ] | R8=0000000000000000 R9=0000000000000000

00128039070i[CPU0 ] | R10=0000000000000000 R11=0000000000000000

00128039070i[CPU0 ] | R12=0000000000000000 R13=0000000000000000

00128039070i[CPU0 ] | R14=0000000000000000 R15=0000000000000000

00128039070i[CPU0 ] | IOPL=0 ID vip vif ac vm RF nt of df if tf sf zf af PF cf

00128039070i[CPU0 ] | SEG selector base limit G D

00128039070i[CPU0 ] | SEG sltr(index|ti|rpl) base limit G D

00128039070i[CPU0 ] | CS:0008( 0001| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | DS:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | SS:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | ES:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | FS:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | GS:0010( 0002| 0| 0) 00000000 ffffffff 1 1

00128039070i[CPU0 ] | MSR_FS_BASE:0000000000000000

00128039070i[CPU0 ] | MSR_GS_BASE:0000000000000000

00128039070i[CPU0 ] | RIP=00000000001532dc (00000000001532dc)

00128039070i[CPU0 ] | CR0=0xe0010011 CR2=0x0000000000000000

00128039070i[CPU0 ] | CR3=0x00134000 CR4=0x00000090

(0).[128039070] [0x001532dc] 0008:00000000001532dc (unk. ctxt): ltr word ptr ss:[esp+0x10] ; 0f005c2410

00128039070e[CPU0 ] exception(): 3rd (13) exception with no resolution, shutdown status is 00h, resetting

I think we can safely assume that something has gone wrong loading the TSS, a general protection fault has been raised during the LTR, then a double fault has followed that.

Is this a bug in the L4 sources, or an issue with my build environment? The important information is

$ uname -a

Linux sidewinder 2.6.31-15-generic #50-Ubuntu SMP Tue Nov 10 14:53:52 UTC 2009 x86_64 GNU/Linux

$ gcc -v

Using built-in specs.

Target: x86_64-linux-gnu

Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.4.1-4ubuntu8' --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 --program-suffix=-4.4 --enable-nls --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc --disable-werror --with-arch-32=i486 --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu

Thread model: posix

gcc version 4.4.1 (Ubuntu 4.4.1-4ubuntu8)

$ ld -v

GNU ld (GNU Binutils for Ubuntu) 2.20

From forum: l4ka@ira.uni-karlsruhe.de

Re: Problem compiling Fiasco Kernel (kip.cc)

2009-12-02T11:28:47Z

On Tue Dec 01, 2009 at 23:43:19 +0100, olivier wrote:

> Adam Lackorzynski a écrit :
> >On Tue Dec 01, 2009 at 00:15:29 +0100, olivier wrote:
> >>Hello,
> >>>>try do Makefile.sub2
> >>>>... Generating version information
> >>>>... Making kip.o
> >>>>cczEXyNz.s: Assembler messages:
> >>>>cczEXyNz.s: 117: Error: junk at end of line, first unrecognized
> >>>>character is ','
> >>>>cczEXyNz.s: 119: 119: unknown pseudo-op: '.previous'
> >>>>etc.
> >>>Anyway, the .previous is the interesting information. This should be
> >>>processed by the assembler. Since this is on Windows maybe the '\n'
> >>>isn't enough and there must be a \n\r or similar. Could you try to add
> >>>the \r in kip.cpp and try again?
> >>No change, I know now problem is in this lines:
> >>
> >>
> >>asm(".section .initkip.version, \"a\", %progbits \n" \
> >> ".string \"" CONFIG_KERNEL_VERSION_STRING "\" \n" \
> >> ".previous \n");
> >>
> >>asm(".section .initkip.features.fini, \"a\", %progbits \n" \
> >> ".string \"\" \n" \
> >> ".previous \n");
> >>
> >>
> >>And more exactly start for compiler with the character ',' after
> >>asm(".section .initkip.version, \"a\"
> >
> >Some more digging reveals that those statements are for ELF targets only
> >and do not work with PE targets. Seems there's no way to tell the
> >assembler to generate ELF instead of PE on cygwin. Sorry but it seems
> >building under Cygwin is not possible.
> So is there any way to compil a Windows version?

The easiest is probably to install Linux in a VM.

Adam
--
Adam adam@...
Lackorzynski http://os.inf.tu-dresden.de/~adam/

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: Problem compiling Fiasco Kernel (kip.cc)

2009-12-01T14:43:19Z

Adam Lackorzynski a écrit :

> On Tue Dec 01, 2009 at 00:15:29 +0100, olivier wrote:
>
>> Hello,
>>
>>>> try do Makefile.sub2
>>>> ... Generating version information
>>>> ... Making kip.o
>>>> cczEXyNz.s: Assembler messages:
>>>> cczEXyNz.s: 117: Error: junk at end of line, first unrecognized
>>>> character is ','
>>>> cczEXyNz.s: 119: 119: unknown pseudo-op: '.previous'
>>>> etc.
>>>>
>>> Anyway, the .previous is the interesting information. This should be
>>> processed by the assembler. Since this is on Windows maybe the '\n'
>>> isn't enough and there must be a \n\r or similar. Could you try to add
>>> the \r in kip.cpp and try again?
>>>
>> No change, I know now problem is in this lines:
>>
>>
>> asm(".section .initkip.version, \"a\", %progbits \n" \
>> ".string \"" CONFIG_KERNEL_VERSION_STRING "\" \n" \
>> ".previous \n");
>>
>> asm(".section .initkip.features.fini, \"a\", %progbits \n" \
>> ".string \"\" \n" \
>> ".previous \n");
>>
>>
>> And more exactly start for compiler with the character ',' after
>> asm(".section .initkip.version, \"a\"
>>
>
> Some more digging reveals that those statements are for ELF targets only
> and do not work with PE targets. Seems there's no way to tell the
> assembler to generate ELF instead of PE on cygwin. Sorry but it seems
> building under Cygwin is not possible.
>

So is there any way to compil a Windows version?

Olivier

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: Problem compiling Fiasco Kernel (kip.cc)

2009-12-01T10:28:15Z

On Tue Dec 01, 2009 at 00:15:29 +0100, olivier wrote:

> Hello,
> >>try do Makefile.sub2
> >>... Generating version information
> >>... Making kip.o
> >>cczEXyNz.s: Assembler messages:
> >>cczEXyNz.s: 117: Error: junk at end of line, first unrecognized
> >>character is ','
> >>cczEXyNz.s: 119: 119: unknown pseudo-op: '.previous'
> >>etc.
> >
> >Anyway, the .previous is the interesting information. This should be
> >processed by the assembler. Since this is on Windows maybe the '\n'
> >isn't enough and there must be a \n\r or similar. Could you try to add
> >the \r in kip.cpp and try again?
>
> No change, I know now problem is in this lines:
>
>
> asm(".section .initkip.version, \"a\", %progbits \n" \
> ".string \"" CONFIG_KERNEL_VERSION_STRING "\" \n" \
> ".previous \n");
>
> asm(".section .initkip.features.fini, \"a\", %progbits \n" \
> ".string \"\" \n" \
> ".previous \n");
>
>
> And more exactly start for compiler with the character ',' after
> asm(".section .initkip.version, \"a\"

Some more digging reveals that those statements are for ELF targets only
and do not work with PE targets. Seems there's no way to tell the
assembler to generate ELF instead of PE on cygwin. Sorry but it seems
building under Cygwin is not possible.

Adam
--
Adam adam@...
Lackorzynski http://os.inf.tu-dresden.de/~adam/

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: Problem compiling Fiasco Kernel (kip.cc)

2009-11-30T15:15:29Z

Hello,

>> try do Makefile.sub2
>> ... Generating version information
>> ... Making kip.o
>> cczEXyNz.s: Assembler messages:
>> cczEXyNz.s: 117: Error: junk at end of line, first unrecognized
>> character is ','
>> cczEXyNz.s: 119: 119: unknown pseudo-op: '.previous'
>> etc.
>>
>
> Anyway, the .previous is the interesting information. This should be
> processed by the assembler. Since this is on Windows maybe the '\n'
> isn't enough and there must be a \n\r or similar. Could you try to add
> the \r in kip.cpp and try again?
>

No change, I know now problem is in this lines:

asm(".section .initkip.version, \"a\", %progbits \n" \
".string \"" CONFIG_KERNEL_VERSION_STRING "\" \n" \
".previous \n");

asm(".section .initkip.features.fini, \"a\", %progbits \n" \
".string \"\" \n" \
".previous \n");

And more exactly start for compiler with the character ',' after
asm(".section .initkip.version, \"a\"

I can't said more

Thanks

Olivier
>
> Thanks,
> Adam
>

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: Problem compiling Fiasco Kernel (kip.cc)

2009-11-30T14:57:43Z

On Mon Nov 30, 2009 at 23:42:33 +0100, olivier wrote:

> >On Sat Nov 28, 2009 at 22:41:41 +0100, olivier wrote:
> >>I'm currently trying compil current Fiasco Kernel under Cygwin,
> >>unfortunatly I have a problem of compilation. when compiler try
> >>compil kip.cc I have an assembler message that said: Error: junk at
> >>the end of line, first unrecognized character is ','
> >>
> >>Any Idea?
> >
> >If it's an assembler message you can get the assembly with
> > make kip.S
>
> This is an assembler message but there is no kip.S, problem is when

The kip.S is a generated file (done via 'make kip.S').

> try do Makefile.sub2
> ... Generating version information
> ... Making kip.o
> cczEXyNz.s: Assembler messages:
> cczEXyNz.s: 117: Error: junk at end of line, first unrecognized
> character is ','
> cczEXyNz.s: 119: 119: unknown pseudo-op: '.previous'
> etc.

Anyway, the .previous is the interesting information. This should be
processed by the assembler. Since this is on Windows maybe the '\n'
isn't enough and there must be a \n\r or similar. Could you try to add
the \r in kip.cpp and try again?

Thanks,
Adam
--
Adam adam@...
Lackorzynski http://os.inf.tu-dresden.de/~adam/

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: Problem compiling Fiasco Kernel (kip.cc)

2009-11-30T14:43:39Z

Hello,

> Hi,
>
> On Sat Nov 28, 2009 at 22:41:41 +0100, olivier wrote:
>
>> I'm currently trying compil current Fiasco Kernel under Cygwin,
>> unfortunatly I have a problem of compilation. when compiler try
>> compil kip.cc I have an assembler message that said: Error: junk at
>> the end of line, first unrecognized character is ','
>>
>> Any Idea?
>>
>
> If it's an assembler message you can get the assembly with
> make kip.S
>

This is an assembler message but there is no kip.S, problem is when try
do Makefile.sub2
... Generating version information
... Making kip.o
cczEXyNz.s: Assembler messages:
cczEXyNz.s: 117: Error: junk at end of line, first unrecognized
character is ','
cczEXyNz.s: 119: 119: unknown pseudo-op: '.previous'
etc.

There is inline asm in c++ code
in build\auto\kip.cc and src\abi\kip.cpp

> and look at the line where the error is told to be.
> Which gcc and binutils version does your cygwin have?
>
>
>
gcc 4.3.4

>
> Adam
>

Thanks

Olivier

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: Problem compiling Fiasco Kernel (kip.cc)

2009-11-29T15:22:49Z

Hi,

On Sat Nov 28, 2009 at 22:41:41 +0100, olivier wrote:
> I'm currently trying compil current Fiasco Kernel under Cygwin,
> unfortunatly I have a problem of compilation. when compiler try
> compil kip.cc I have an assembler message that said: Error: junk at
> the end of line, first unrecognized character is ','
>
> Any Idea?

If it's an assembler message you can get the assembly with
make kip.S
and look at the line where the error is told to be.
Which gcc and binutils version does your cygwin have?

Adam
--
Adam adam@...
Lackorzynski http://os.inf.tu-dresden.de/~adam/

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Problem compiling Fiasco Kernel (kip.cc)

2009-11-28T13:41:41Z

Hello all

I'm currently trying compil current Fiasco Kernel under Cygwin,
unfortunatly I have a problem of compilation. when compiler try compil
kip.cc I have an assembler message that said: Error: junk at the end of
line, first unrecognized character is ','

Any Idea?

Thanks

Olivier

PS: my kip.cc

// AUTOMATICALLY GENERATED -- DO NOT EDIT! -*- c++ -*-

#include "kip.h"
#include "kip_i.h"

#line 47 "/tudos/kernel/fiasco/src/abi/kip.cpp"

//============================================================================

#line 182 "/tudos/kernel/fiasco/src/abi/kip.cpp"

//----------------------------------------------------------------------------

#line 188 "/tudos/kernel/fiasco/src/abi/kip.cpp"

Kip *Kip::global_kip;
#line 192 "/tudos/kernel/fiasco/src/abi/kip.cpp"

asm(".section .initkip.version, \"a\", %progbits \n" \
".string \"" CONFIG_KERNEL_VERSION_STRING "\" \n" \
".previous \n");
#line 196 "/tudos/kernel/fiasco/src/abi/kip.cpp"

asm(".section .initkip.features.fini, \"a\", %progbits \n" \
".string \"\" \n" \
".previous \n");

#line 110 "/tudos/kernel/fiasco/src/abi/kip.cpp"

bool Kip::add_mem_region(Mem_desc const &md)
{
Mem_desc *m = mem_descs();
Mem_desc *end = m + num_mem_descs();
for (;m<end;++m)
{
if (m->type() == Mem_desc::Undefined)
{
*m = md;
return true;
}
}

// Add mem region failed -- must be a Fiasco startup problem. Bail out.
assert (0);

return false;
}

#line 130 "/tudos/kernel/fiasco/src/abi/kip.cpp"

Mem_region Kip::last_free() const
{
Mem_desc const *m = mem_descs();
Mem_desc const *const end = m + num_mem_descs();
Mem_region r;
r.start = 0;
r.end = 0;
for (;m<end;++m)
{
Address s = m->start();
Address e = m->end();

// Speep out stupid descriptors (that have the end before the start)
if (s>=e)
{
const_cast<Mem_desc*>(m)->type(Mem_desc::Undefined);
continue;
}

if (m->is_virtual()) // skip virtual memory descriptors
continue;

switch (m->type())
{
case Mem_desc::Conventional:
if (e>r.end)
r.end = e;
if (s>r.start)
r.start = s;
break;
case Mem_desc::Reserved:
case Mem_desc::Dedicated:
case Mem_desc::Shared:
case Mem_desc::Arch:
case Mem_desc::Bootloader:
if (e>=r.start && e<r.end)
{
r.start = e + 1;
}
if (s>=r.start && s<r.end)
{
r.end = s - 1;
}
break;
default:
break;
}
}
return r;
}

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: codezero community

2009-11-24T16:18:48Z

Harold Grovesteen wrote:

> Does one exist? Is there an infrastructure to actually support a
> community? The link from wikipedia gets you to a site. None of the
> links there work for the purported email list or info email. The
> contact email address for the group that appears to be supporting the
> effort, B-labs, also fails. There is supposed to be a git repository
> that I could not clone. The only success I had was downloading the v0.1
> tarball.
>
> I see in the archives an announcement for codezero. I am hoping that
> invidual will at least respond.
>
> Thanks,
> Harold Grovesteen

Harold,

There indeed is a supporting infrastructure. You've just happened to
visit our site in a bad time. We are migrating our server.

Codezero v0.2 is going to be released very soon. Meanwhile you may reach
the latest repo from git.l4dev.org

--
Bahadir Balban

From forum: l4ka@ira.uni-karlsruhe.de

codezero community

2009-11-24T12:28:57Z

Does one exist? Is there an infrastructure to actually support a
community? The link from wikipedia gets you to a site. None of the
links there work for the purported email list or info email. The
contact email address for the group that appears to be supporting the
effort, B-labs, also fails. There is supposed to be a git repository
that I could not clone. The only success I had was downloading the v0.1
tarball.

I see in the archives an announcement for codezero. I am hoping that
invidual will at least respond.

Thanks,
Harold Grovesteen

From forum: l4ka@ira.uni-karlsruhe.de

Re: ddekit and dde linux2.6

2009-11-21T19:55:04Z

Hello,

2009/11/22 Björn Döbel <doebel@...>:

>>
>> I suppose DDEKit allows us to build and run any Linux drivers without
>> modification. Why do we still need to make some effort to port drivers
>> to DDEKit? Do I miss something?
>
> There is no need to port any driver to DDEKit. In the beginning you have
> the following problem: your driver wants a Linux interface to run and
> your OS is providing a different one:
>
> +--------------+
> | L4 I/O layer |
> +\ +----\ /+
> \-/ \-/
>
>
> /+\
> +-/ \--------+
> | Linux driver |
> +--------------+
>
> To make the Linux driver run on your target OS, you need to provide a
> glue layer that is mapping the Linux driver calls to your target OS.
> This is what the DDE (device driver environment) is good for.
>
> +--------------+
> | L4 I/O layer |
> +\ +----\ /+
> | \-/ \-/ |
> | DDE |
> | |
> | /+\ |
> +-/ \--------+
> | Linux driver |
> +--------------+
>
> Now you have another problem if you want to support not only Linux
> drivers but also the ones from *BSD or any other source. You'd need to
> rewrite the DDE glue library for each of these. The idea of the DDEKit
> is now to provide a generic driver abstraction layer so that
> implementing the higher-level DDE/{Linux,FreeBSD} becomes easier.
>
> +---------------+
> | L4 I/O layer |
> +\ +----\ /-+
> | \-/ \-/ |
> | DDEKit |
> | |
> +---------------+
> ^
> |
> +-----------------------------+
> | |
> +---------------+ +----------------+
> | /+\ DDE/Linux| | DDE/FBSD /\ /-\|
> +-/ \---------+ +---------/ + \
> | Linux driver | | FreeBSD driver |
> +---------------+ +----------------+
>
> Note, that all the DDE/* libs use only those functions provided by the
> DDEKit. This has another advantage. You port DDEKit to a new target OS
> (e.g., the hurd in your case) and you get all the available DDEs to run
> on top for free.

I'm really happy to hear that:-)

>
> So to get back to your question. You don't port any device driver. You
> just link it against the corresponding DDE library and don't do any
> modifications. The plan is that this works for all drivers. However,
> each device driver class uses some different subsystems and when you
> start running a new driver from a not-yet supported class, it may happen
> that you need to add functionality to the DDE. We're pretty sure that we
> have enough support for USB and network drivers and we also have more or
> less working block and sound drivers. If your goal is to run one of
> these drivers, you should be ready to go. If you want something else,
> there may be the need to add functionality to the respective DDE, not to
> DDEKit.
>
> I hope that made this clearer?

Yes, thank you very much for your explanation.
I am also very happy to hear that new functionalities are added to DDE
instead of DDEKit. It really makes the maintenance work in the future
much easier. Good design:-)

Zheng Da

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: ddekit and dde linux2.6

2009-11-21T10:38:05Z

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

> 2009/11/21 Björn Döbel <doebel@...>:
>>> It seems that ddekit is still under active development. What is its current development status?
>> DDE/Linux 2.6 is supporting Linux 2.6.29 drivers at the moment. The
>> DDEKit has been developed in parallel along with our experimental
>> setups, which include network drivers (you might have a look at
>> l4/pkg/ore) and USB (see l4/pkg/usb), as well as some experimental hard
>> disk stuff which is not in the SVN.
>
> I suppose DDEKit allows us to build and run any Linux drivers without
> modification. Why do we still need to make some effort to port drivers
> to DDEKit? Do I miss something?

There is no need to port any driver to DDEKit. In the beginning you have
the following problem: your driver wants a Linux interface to run and
your OS is providing a different one:

+--------------+
| L4 I/O layer |
+\ +----\ /+
\-/ \-/

/+\
+-/ \--------+
| Linux driver |
+--------------+

To make the Linux driver run on your target OS, you need to provide a
glue layer that is mapping the Linux driver calls to your target OS.
This is what the DDE (device driver environment) is good for.

+--------------+
| L4 I/O layer |
+\ +----\ /+
| \-/ \-/ |
| DDE |
| |
| /+\ |
+-/ \--------+
| Linux driver |
+--------------+

Now you have another problem if you want to support not only Linux
drivers but also the ones from *BSD or any other source. You'd need to
rewrite the DDE glue library for each of these. The idea of the DDEKit
is now to provide a generic driver abstraction layer so that
implementing the higher-level DDE/{Linux,FreeBSD} becomes easier.

+---------------+
| L4 I/O layer |
+\ +----\ /-+
| \-/ \-/ |
| DDEKit |
| |
+---------------+
^
|
+-----------------------------+
| |
+---------------+ +----------------+
| /+\ DDE/Linux| | DDE/FBSD /\ /-\|
+-/ \---------+ +---------/ + \
| Linux driver | | FreeBSD driver |
+---------------+ +----------------+

Note, that all the DDE/* libs use only those functions provided by the
DDEKit. This has another advantage. You port DDEKit to a new target OS
(e.g., the hurd in your case) and you get all the available DDEs to run
on top for free.

So to get back to your question. You don't port any device driver. You
just link it against the corresponding DDE library and don't do any
modifications. The plan is that this works for all drivers. However,
each device driver class uses some different subsystems and when you
start running a new driver from a not-yet supported class, it may happen
that you need to add functionality to the DDE. We're pretty sure that we
have enough support for USB and network drivers and we also have more or
less working block and sound drivers. If your goal is to run one of
these drivers, you should be ready to go. If you want something else,
there may be the need to add functionality to the respective DDE, not to
DDEKit.

I hope that made this clearer?

Bjoern
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEUEARECAAYFAksIM40ACgkQP5ijxgQLUNmBLQCY59ID3ngN0I4kByjuSjcXxK0j
agCeJqwwhcB5iYANDWn/dRGiDhBxHBI=
=WPW7
-----END PGP SIGNATURE-----

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: ddekit and dde linux2.6

2009-11-21T08:36:08Z

Hello

2009/11/21 Björn Döbel <doebel@...>:
>
>> It seems that ddekit is still under active development. What is its current development status?
>
> DDE/Linux 2.6 is supporting Linux 2.6.29 drivers at the moment. The
> DDEKit has been developed in parallel along with our experimental
> setups, which include network drivers (you might have a look at
> l4/pkg/ore) and USB (see l4/pkg/usb), as well as some experimental hard
> disk stuff which is not in the SVN.

I suppose DDEKit allows us to build and run any Linux drivers without
modification. Why do we still need to make some effort to port drivers
to DDEKit? Do I miss something?

Zheng Da

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: ddekit and dde linux2.6

2009-11-21T07:57:38Z

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

> I'm trying to understand ddekit and hopefully to port it to Hurd.

good luck.

> It seems that ddekit is still under active development. What is its current development status?

DDE/Linux 2.6 is supporting Linux 2.6.29 drivers at the moment. The
DDEKit has been developed in parallel along with our experimental
setups, which include network drivers (you might have a look at
l4/pkg/ore) and USB (see l4/pkg/usb), as well as some experimental hard
disk stuff which is not in the SVN.

> I read http://demo.tudos.org/dsweeper_tutorial.html, which said that new features such as hotplugging and power management would be added ddekit but it was in 2006. Have those features been added to ddekit? Is it possible for me to move the Linux USB driver on the top of dde without modification?

USB hotplug worked the last time I looked.

Cheers,
Bjoern
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAksIDfIACgkQP5ijxgQLUNlpawCfcoWyKboQuyNtGcEOn4vPtSn+
g+UAnA06JOPC9ZZlMvDPatTjQiaR80dy
=gl1Z
-----END PGP SIGNATURE-----

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

ddekit and dde linux2.6

2009-11-21T06:40:48Z

Hi,

I'm trying to understand ddekit and hopefully to port it to Hurd.
It seems that ddekit is still under active development. What is its current development status? I read http://demo.tudos.org/dsweeper_tutorial.html, which said that new features such as hotplugging and power management would be added ddekit but it was in 2006. Have those features been added to ddekit? Is it possible for me to move the Linux USB driver on the top of dde without modification?

What is the status of dde linux2.6 (I mean the one in the directory l4/pkg/dde_linux26)? Is it still under development? Does it support the features I mentioned above?

Thank you,
Zheng Da

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: The errors i tried to map iomem to virtual memory

2009-11-06T06:58:59Z

On Thursday 05 November 2009 06:10:04 Adam Lackorzynski wrote:
> On Wed Nov 04, 2009 at 21:11:06 +0800, Guanghui, Cheng wrote:
> > On Wednesday 04 November 2009 07:28:02 Adam Lackorzynski wrote:
> > > Hi,
> > >
> > > On Tue Nov 03, 2009 at 23:37:12 +0800, Guanghui, Cheng wrote:
> > > > I tried to map the hpet iomem to the virtual memory but i got some
> > > > error. The log is as following:
> > > > vtimer | rmgr_pager 4:0
> > > > vtimer | hpet_address 0xfed00000
> > > > vtimer | hpet size 0x400000
> > > > vtimer | hpet_virt_address 0x400000
> > > > ROOT: Cannot map page at fed00000 log2_size=12 failed
> > > > vtimer | IOMem mapping error : No fpage received
> > > > vtimer | __handle_pf(): L4RM: page fault: no entry in region list
> > > > vtimer | L4RM: [PF] read at 0x00400000, ip 0100024e, src 8.02
> > > > It seems that when I map the iomem the pagesize is not permitted but
> > > > superpagesize is ok. But i still get the error as "No fpage
> > > > received". Does anybody tell me what the problem is?
> > >
> > > roottask can map non-RAM as superpages only. For getting device memory
> > > you should use and query l4io which can handle this much better. Did
> > > you try that?
> >
> > Hello, Adam:
> > Now i fixed l4io a little and add the hpet device after the tpm device
> > in the l4io/server/src/static_cfg_x86.c and now i could map the hpet
> > device into the virtual device. I could read/write the hpet device
> > directly. This is a important progress for vtimer server implementation.
> > If i make more i could report to you.
Hello, Adam:
Today I get a small progress about ipet driver for L4/Fiasco because
the hpet driver i mentioned before can't output the correct value and all
fields of hpet driver are 0xFFFFFFFF. Today i find that should be caused by
the Intel buggy chipesets (ICH7 Family) and i also find some code could
fixed it from Linux kernel code that is located in the ich_force_enable_hpet() of arch/x86/kernel/quirks.c.
The solution is like this:
I added two more IO resources named RCBA and HPET in the
l4io/server/src/ static_cfg_x86.c as following:
static l4io_desc_device_t
rcba = { "RCBA", 1, { { L4IO_RESOURCE_MEM, 0xfed1c000, 0xfed1ffff } } };
static l4io_desc_device_t
hpet = {"HPET", 1, { { L4IO_RESOURCE_MEM, 0xfed00000, 0xfed003ff } } };
And then i get the rcba device in the local and do the same behaviour
as the ich_force_enable_hpet() does in the arch/x86/kernel/quirks.c and then
get the hpet device in the local space. This time it could work and display there are 3 timers before it is 16 (F).
From the Linux kernel code this problem should be in the Intel Chipsets
ICH6, ICH7, ICH8, ICH9 and ICH10. But luckily the linux has fixed this problem otherwise i can't get it work.
Thanks.
Cheng Guanghui

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: The errors i tried to map iomem to virtual memory

2009-11-06T02:28:46Z

Hi,

Guanghui, Cheng wrote:
> Now i fixed l4io a little and add the hpet device after the tpm device
> in the l4io/server/src/static_cfg_x86.c and now i could map the hpet
> device
> into the virtual device. I could read/write the hpet device directly.
> This is
> a important progress for vtimer server implementation. If i make more i
> could report to you.
Just out of curiosity, is vtimer a special component related to your
work or is it a general step towards a (working) general-purpose timer
service in DROPS (which would be very nice)?

Regards,
Andre

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: The errors i tried to map iomem to virtual memory

2009-11-04T14:10:04Z

On Wed Nov 04, 2009 at 21:11:06 +0800, Guanghui, Cheng wrote:

> On Wednesday 04 November 2009 07:28:02 Adam Lackorzynski wrote:
> > Hi,
> >
> > On Tue Nov 03, 2009 at 23:37:12 +0800, Guanghui, Cheng wrote:
> > > I tried to map the hpet iomem to the virtual memory but i got some
> > > error. The log is as following:
> > > vtimer | rmgr_pager 4:0
> > > vtimer | hpet_address 0xfed00000
> > > vtimer | hpet size 0x400000
> > > vtimer | hpet_virt_address 0x400000
> > > ROOT: Cannot map page at fed00000 log2_size=12 failed
> > > vtimer | IOMem mapping error : No fpage received
> > > vtimer | __handle_pf(): L4RM: page fault: no entry in region list
> > > vtimer | L4RM: [PF] read at 0x00400000, ip 0100024e, src 8.02
> > > It seems that when I map the iomem the pagesize is not permitted but
> > > superpagesize is ok. But i still get the error as "No fpage received".
> > > Does anybody tell me what the problem is?
> >
> > roottask can map non-RAM as superpages only. For getting device memory
> > you should use and query l4io which can handle this much better. Did you
> > try that?
> Hello, Adam:
> Now i fixed l4io a little and add the hpet device after the tpm device
> in the l4io/server/src/static_cfg_x86.c and now i could map the hpet device
> into the virtual device. I could read/write the hpet device directly. This is
> a important progress for vtimer server implementation. If i make more i
> could report to you.

Great! :)

Adam
--
Adam adam@...
Lackorzynski http://os.inf.tu-dresden.de/~adam/

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: The errors i tried to map iomem to virtual memory

2009-11-04T05:11:06Z

On Wednesday 04 November 2009 07:28:02 Adam Lackorzynski wrote:
> Hi,
>
> On Tue Nov 03, 2009 at 23:37:12 +0800, Guanghui, Cheng wrote:
> > I tried to map the hpet iomem to the virtual memory but i got some
> > error. The log is as following:
> > vtimer | rmgr_pager 4:0
> > vtimer | hpet_address 0xfed00000
> > vtimer | hpet size 0x400000
> > vtimer | hpet_virt_address 0x400000
> > ROOT: Cannot map page at fed00000 log2_size=12 failed
> > vtimer | IOMem mapping error : No fpage received
> > vtimer | __handle_pf(): L4RM: page fault: no entry in region list
> > vtimer | L4RM: [PF] read at 0x00400000, ip 0100024e, src 8.02
> > It seems that when I map the iomem the pagesize is not permitted but
> > superpagesize is ok. But i still get the error as "No fpage received".
> > Does anybody tell me what the problem is?
>
> roottask can map non-RAM as superpages only. For getting device memory
> you should use and query l4io which can handle this much better. Did you
> try that?
Hello, Adam:
Now i fixed l4io a little and add the hpet device after the tpm device
in the l4io/server/src/static_cfg_x86.c and now i could map the hpet device
into the virtual device. I could read/write the hpet device directly. This is
a important progress for vtimer server implementation. If i make more i
could report to you.
Thanks for your kind remind.
Best.
Cheng Guanghui

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: The errors i tried to map iomem to virtual memory

2009-11-03T16:57:27Z

On Wednesday 04 November 2009 07:28:02 Adam Lackorzynski wrote:
> Hi,
>
> On Tue Nov 03, 2009 at 23:37:12 +0800, Guanghui, Cheng wrote:
> > I tried to map the hpet iomem to the virtual memory but i got some
> > error. The log is as following:
> > vtimer | rmgr_pager 4:0
> > vtimer | hpet_address 0xfed00000
> > vtimer | hpet size 0x400000
> > vtimer | hpet_virt_address 0x400000
> > ROOT: Cannot map page at fed00000 log2_size=12 failed
> > vtimer | IOMem mapping error : No fpage received
> > vtimer | __handle_pf(): L4RM: page fault: no entry in region list
> > vtimer | L4RM: [PF] read at 0x00400000, ip 0100024e, src 8.02
> > It seems that when I map the iomem the pagesize is not permitted but
> > superpagesize is ok. But i still get the error as "No fpage received".
> > Does anybody tell me what the problem is?
>
> roottask can map non-RAM as superpages only. For getting device memory
> you should use and query l4io which can handle this much better. Did you
> try that?
i use roottask to map the non-RAM as superpages now but it can't
work correctly. I try l4io to get iomem mapping today. Later i will mail to you about the this part.
Thanks.
Cheng Guanghui

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: The errors i tried to map iomem to virtual memory

2009-11-03T15:28:02Z

Hi,

On Tue Nov 03, 2009 at 23:37:12 +0800, Guanghui, Cheng wrote:

> I tried to map the hpet iomem to the virtual memory but i got some error. The
> log is as following:
> vtimer | rmgr_pager 4:0
> vtimer | hpet_address 0xfed00000
> vtimer | hpet size 0x400000
> vtimer | hpet_virt_address 0x400000
> ROOT: Cannot map page at fed00000 log2_size=12 failed
> vtimer | IOMem mapping error : No fpage received
> vtimer | __handle_pf(): L4RM: page fault: no entry in region list
> vtimer | L4RM: [PF] read at 0x00400000, ip 0100024e, src 8.02
> It seems that when I map the iomem the pagesize is not permitted but
> superpagesize is ok. But i still get the error as "No fpage received". Does
> anybody tell me what the problem is?

roottask can map non-RAM as superpages only. For getting device memory
you should use and query l4io which can handle this much better. Did you
try that?

Adam
--
Adam adam@...
Lackorzynski http://os.inf.tu-dresden.de/~adam/

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

The errors i tried to map iomem to virtual memory

2009-11-03T07:37:12Z

Hello:
I tried to map the hpet iomem to the virtual memory but i got some error. The log is as following:
vtimer | rmgr_pager 4:0
vtimer | hpet_address 0xfed00000
vtimer | hpet size 0x400000
vtimer | hpet_virt_address 0x400000
ROOT: Cannot map page at fed00000 log2_size=12 failed
vtimer | IOMem mapping error : No fpage received
vtimer | __handle_pf(): L4RM: page fault: no entry in region list
vtimer | L4RM: [PF] read at 0x00400000, ip 0100024e, src 8.02
It seems that when I map the iomem the pagesize is not permitted but superpagesize is ok. But i still get the error as "No fpage received". Does anybody tell me what the problem is?
Thanks.
Cheng Guanghui

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: the preempter of one thread

2009-10-29T16:56:32Z

On Wed Oct 28, 2009 at 22:03:44 +0800, Guanghui, Cheng wrote:
> Thread preempter is functional or ignored in the current fiasco.
> i read the lectures about microkernel-based operating systems and there
> are somethings about thread preempter.

It's used for the scheduling extensions.

Adam
--
Adam adam@...
Lackorzynski http://os.inf.tu-dresden.de/~adam/

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

Re: Some Errors in DDE

2009-10-29T16:52:55Z

On Sat Oct 24, 2009 at 12:06:27 +0800, Jingyi WANG wrote:

> I'm porting a sound card(ens1370)to DDE, but found some errors in the
> function of kobj_map whic is in
> map.c(l4/pkg/dde/linux26/contrib.drivers/base/map.c)
> What's more, I check the ELF format of this sound card.The error is in the
> address of 821607. The related messages show as follows:
>
> /home/wangjingyi/l4/tudos/l4/pkg/dde/linux26/contrib/drivers/base/map.c:58
> mutex_lock(domain->lock);
> 821603: 8b 55 08 mov 0x8(%ebp),%edx
> 821606: 58 pop %eax
> 821607: 8b 82 fc 03 00 00 mov 0x3fc(%edx),%eax
> 82160d: 50 push %eax
> 82160e: e8 cd 65 02 00 call 847be0 <mutex_lock>
>
> I was wondering what caused this error and how to fix it.

The first parameter given to kobj_map is wrong, probably 0. The caller
needs to make sure a proper value is given.

Adam
--
Adam adam@...
Lackorzynski http://os.inf.tu-dresden.de/~adam/

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4

the preempter of one thread

2009-10-28T07:03:44Z

Hi:
Thread preempter is functional or ignored in the current fiasco.
i read the lectures about microkernel-based operating systems and there
are somethings about thread preempter.
Thanks.
Cheng Guanghui

_______________________________________________
l4-hackers mailing list
l4-hackers@...
http://os.inf.tu-dresden.de/mailman/listinfo/l4-hackers

From forum: L4