I have this bzImage file given to me from a company. They don't want to
give me the .config, but I need it, so I thought I'd try to
- open the bzImage to a vmlinux
- list the .o's in the vmlinux
Is this possible?
Btw.. Does GPL require them to give me the .config file?
roy
--
Roy Sigurd Karlsbakk, MCSE, MCNE, CLS, LCA
Computers are like air conditioners.
They stop working when you open Windows.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
I've never tried this, but could you do something like
bunzip2 -c bzImage > zImage && ar -t zImage
?
--Drew Vogel
Doesn't work
bzcat: dist/images/kernel-nfs is not a bzip2 file.
--Drew Vogel
On Tue, 5 Feb 2002, Roy Sigurd Karlsbakk wrote:
> I have this bzImage file given to me from a company. They don't want to
> give me the .config, but I need it, so I thought I'd try to
>
> - open the bzImage to a vmlinux
> - list the .o's in the vmlinux
>
> Is this possible?
No. Even if you decompress the image you can't list the .o files. The
decompressed image is not an archive (created with ar). It is a raw binary
file generated by objcopy.
Anyway, if you still want to decompress it, the gzip'ed parts does not
start from the beginning. In front of the kernel image might be a boot
sector and some other code for decompressing it, video initialization etc.
Look after the gzip signature (0x1F 0x8B I think) in your bzImage file.
Cut the data before this and use "gzip -d" to decompress it.
> Btw.. Does GPL require them to give me the .config file?
IANAL but I don't think so, there is no code from it to compile and
include in the kernel image. It just defines what the image contains.
--
Catalin
________________________________________________________________________
This email has been scanned for all viruses by the MessageLabs SkyScan
service. For more information on a proactive anti-virus service working
around the clock, around the globe, visit http://www.messagelabs.com
________________________________________________________________________
Yes. To paraphrase, "You may copy and distribute the Program in
executable form provided that you also accompany it with the
complete corresponding machine-readable source code (or a promise
to distribute source upon request, or a copy of the promise given
to you)."
Complete source code for an executable work is defined as "all the
source code for all modules it contains, plus any associated interface
definition files, plus the scripts used to control compilation and
installation of the executable."
IMHO, .config is a "script used to control compilation".
(This is a point in favor of saving .config in the kernel by default.
A competent build system that saves it separately may be desirable, but
that is of no help when faced with incompetence and/or malice.)
> > Btw.. Does GPL require them to give me the .config file?
>
> definition files, plus the scripts used to control compilation and
> installation of the executable."
>
> IMHO, .config is a "script used to control compilation".
I think (though I might be wrong) this is not considered a "script used to
control compilation" in GPL acception, even if it does this. It is not
included in the kernel sources, it is generated by other scripts
(kconfig.tk for example, which is also generated). This is like, for
example, intermediar .s files generated from .c ones. Nobody would bother
to include this in the sources.
I don't think it should be included in the kernel sources because you
can generate it by yourself (of course, you want to know theirs). So, they
might not be forced to give you the .config file.
Regards.
--
Catalin
________________________________________________________________________
This email has been scanned for all viruses by the MessageLabs SkyScan
service. For more information on a proactive anti-virus service working
around the clock, around the globe, visit http://www.messagelabs.com
________________________________________________________________________
Joachim Martillo
IANAL either, but I don't think you're right. Since somebody has distributed
the bzImage to him, that somebody is obligated to provide him the source code
- the full source code - if he requests it. In my opinion, the .config has
to be considered part of the source code since it's needed in order to
compile the binary he received. This issue needs to be clarified.
--
Daniel
I doubt it very much that the FSF will get into this, the kernel is GPL(ish),
but not by the FSF, but by Linus.
What exactly do you want? A bzImage is essentially a stripped executable,
gzipped and then appended to a bootloader. Once I got inside this (forget
the reason) by simply looking for the gzip header (file(1)'s magic is of
help here), dd(1)'ed the compressed tail to a file, and gunzip(1)'ed the
result. Look at your nearest kernel's build process...
As the executable is compiled with -O2 -fomit-frame-pointer and stripped,
it won't be much use by itself anyway. Probably the System.map file (if
extant) is more useful, or at least required to make sense of the kernel.
Or you could futz around in /dev/{,k}mem...
--
Horst von Brand http://counter.li.org # 22616
> I think (though I might be wrong) this is not considered a "script used to
> control compilation" in GPL acception, even if it does this. It is not
> included in the kernel sources, it is generated by other scripts
> (kconfig.tk for example, which is also generated). This is like, for
> example, intermediar .s files generated from .c ones. Nobody would bother
> to include this in the sources.
If you go "make install modules modules_install" it will use that file to
control exactly what is built and installed, and so (IMO, IANAL) it does
influence the resulting binary more than enough so that it has to be
considered a "source file" for that particular binary. That you can come up
with your own shouldn't make a difference (I can write, say, emacs' source
too given sufficient time ;-)
--
Horst von Brand http://counter.li.org # 22616
>Roy Sigurd Karlsbakk <r...@karlsbakk.net> said:
>> No. Not yet. I'm trying to put some pressure on them first. Trying to be a
>> little polite..
>
>I doubt it very much that the FSF will get into this, the kernel is GPL(ish),
>but not by the FSF, but by Linus.
I did not believe they would. They would probably be willing to offer
their interpretation, though.
--Drew Vogel
I beleive it's ok.
They sent me a tar ball with their source code along with their own
patchs.
However ... I beleive some of these patches must be hard linked. Does that
require them to GPL them?
roy
--
Roy Sigurd Karlsbakk, MCSE, MCNE, CLS, LCA
Computers are like air conditioners.
They stop working when you open Windows.
-
>> >I doubt it very much that the FSF will get into this, the kernel is GPL(ish),
>> >but not by the FSF, but by Linus.
>>
>> I did not believe they would. They would probably be willing to offer
>> their interpretation, though.
>
>I beleive it's ok.
>
>They sent me a tar ball with their source code along with their own
>patchs.
Good to hear.
Almost certainly, but at that point you get into the world of derived works,
and probably its best to ask the FSF for an opinion
> However ... I beleive some of these patches must be hard linked. Does
> that require them to GPL them?
I believe they're already GPLed because they linked them into
a GPL work and distributed the resulting binary.
regards,
Rik
--
"Linux holds advantages over the single-vendor commercial OS"
-- Microsoft's "Competing with Linux" document
http://www.surriel.com/ http://distro.conectiva.com/
> On Tue, 5 Feb 2002, Daniel A. Newby wrote:
>
> > > Btw.. Does GPL require them to give me the .config file?
> >
> > definition files, plus the scripts used to control compilation and
> > installation of the executable."
> >
> > IMHO, .config is a "script used to control compilation".
>
> I think (though I might be wrong) this is not considered a "script used to
> control compilation" in GPL acception, even if it does this. It is not
> included in the kernel sources, it is generated by other scripts
> (kconfig.tk for example, which is also generated). This is like, for
> example, intermediar .s files generated from .c ones. Nobody would bother
> to include this in the sources.
I can justify the opposite view: ".config" _IS_ a "script used to control
compilation": just look at how it's used. The "Makefile" (that is shipped with
the sources) does an "include .config" to incorporate the contents of ".config"
into the compilation process, so IMHO it easily qualifies. If fact, some
distributions ;-) include the ".config" files as a matter of course.
--
(Any correlation between the carefully considered options expressed herein
to those of my employer is due entirely to the luck of the latter instead
of the intention of the former.-)
-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- + -- -- -- -- -- -- -- -- -- --
Tommy Reynolds | mailto: <reyn...@redhat.com>
Red Hat, Inc., Embedded Development Services | Phone: +1.256.704.9286
307 Wynn Drive NW, Huntsville, AL 35805 USA | FAX: +1.256.837.3839
Senior Software Developer | Mobile: +1.919.641.2923
> I believe they're already GPLed because they linked them into
> a GPL work and distributed the resulting binary.
IANAL but I believe that this is a myth. Even if they violate the GPL they
would still own their code. Of course, they may be required to stop using it
and to pay some money to Linus if he sues.
Most probably they will GPL it anyway if you ask, because it makes no sense
to them to have bad publictiy, a lawsuit and code they are not allowed to
make use of.
Just my 0.02 euro.
Flo
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: You may find my public key at http://www.galois.de/florian.schmitt.pubkey.asc
iD8DBQE8YbE5H7Gei80C0lQRAtrpAKCvL/XFePqZqY5ioYjgNWnQpGsxcwCgz8xl
1FXBK6f8i9ej1TcgxbCMGA0=
=y91E
-----END PGP SIGNATURE-----