Discussion:
truecrypt
Neal Becker
2007-09-25 16:55:30 UTC
Permalink
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
it says:

Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Jima
2007-09-25 17:07:29 UTC
Permalink
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Don't we BR kernel-devel when building kmods anyway? I don't think this
is a problem at all. (Aside from the possibility of kmod packaging going
anyway, anyway.)

Jima
Eric Sandeen
2007-09-25 17:13:12 UTC
Permalink
Post by Jima
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Don't we BR kernel-devel when building kmods anyway? I don't think this
is a problem at all. (Aside from the possibility of kmod packaging going
anyway, anyway.)
But, drivers/md/dm.h is not in kernel-devel.

-Eric
Jima
2007-09-25 17:17:07 UTC
Permalink
Post by Eric Sandeen
But, drivers/md/dm.h is not in kernel-devel.
...as I was just noticing. Oops, spoke too soon. :-)

Jima
Jima
2007-09-25 17:17:07 UTC
Permalink
Post by Eric Sandeen
But, drivers/md/dm.h is not in kernel-devel.
...as I was just noticing. Oops, spoke too soon. :-)

Jima
Josh Boyer
2007-09-25 17:15:37 UTC
Permalink
On Tue, 25 Sep 2007 12:07:29 -0500 (CDT)
Post by Jima
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Don't we BR kernel-devel when building kmods anyway? I don't think this
is a problem at all. (Aside from the possibility of kmod packaging going
anyway, anyway.)
It's not a possibility. They are going away.

josh
Eric Sandeen
2007-09-25 17:13:12 UTC
Permalink
Post by Jima
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Don't we BR kernel-devel when building kmods anyway? I don't think this
is a problem at all. (Aside from the possibility of kmod packaging going
anyway, anyway.)
But, drivers/md/dm.h is not in kernel-devel.

-Eric
Josh Boyer
2007-09-25 17:15:37 UTC
Permalink
On Tue, 25 Sep 2007 12:07:29 -0500 (CDT)
Post by Jima
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Don't we BR kernel-devel when building kmods anyway? I don't think this
is a problem at all. (Aside from the possibility of kmod packaging going
anyway, anyway.)
It's not a possibility. They are going away.

josh
Luke Macken
2007-09-25 18:39:07 UTC
Permalink
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
hybrid of a bunch of different licenses and has some interesting
distribution clauses. IANAL, so we definitely would want to run it by
fedora-legal-list if we hope to get truecrypt into Fedora. There has also
been some talk about this license in the Debian community:

http://lists.debian.org/debian-legal/2006/06/msg00294.html
http://lists.debian.org/debian-legal/2006/07/msg00008.html

luke

[0]: http://www.truecrypt.org/license.php
Dominik 'Rathann' Mierzejewski
2007-09-25 19:43:36 UTC
Permalink
Post by Luke Macken
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
hybrid of a bunch of different licenses and has some interesting
distribution clauses. IANAL, so we definitely would want to run it by
fedora-legal-list if we hope to get truecrypt into Fedora. There has also
http://lists.debian.org/debian-legal/2006/06/msg00294.html
http://lists.debian.org/debian-legal/2006/07/msg00008.html
luke
[0]: http://www.truecrypt.org/license.php
Here are src.rpms with tools and kmod:
http://rpm.greysector.net/livna/

Feel free to submit them to livna/rpmfusion. I don't have time to maintain
them, but I made those when I was looking for a cryptofs solution for both
Linux and Windows (and settled on LUKS instead).

Regards,
R.
--
Fedora contributor http://fedoraproject.org/wiki/DominikMierzejewski
Livna contributor http://rpm.livna.org MPlayer developer http://mplayerhq.hu
"Faith manages."
-- Delenn to Lennier in Babylon 5:"Confessions and Lamentations"
Farkas Levente
2007-09-25 21:36:17 UTC
Permalink
Post by Luke Macken
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
you can always found the latest kmod in:
http://www.lfarkas.org/linux/packages/fedora/7/
--
Levente "Si vis pacem para bellum!"
Dominik 'Rathann' Mierzejewski
2007-09-26 06:33:33 UTC
Permalink
Post by Farkas Levente
Post by Luke Macken
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
http://www.lfarkas.org/linux/packages/fedora/7/
Talk about duplicated effort. ;) Why don't you submit your package (perhaps
after merging with mine) to rpmfusion?

Regards,
R.
--
Fedora contributor http://fedoraproject.org/wiki/DominikMierzejewski
Livna contributor http://rpm.livna.org MPlayer developer http://mplayerhq.hu
"Faith manages."
-- Delenn to Lennier in Babylon 5:"Confessions and Lamentations"
Farkas Levente
2007-09-26 09:21:16 UTC
Permalink
Post by Dominik 'Rathann' Mierzejewski
Post by Farkas Levente
Post by Luke Macken
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
http://www.lfarkas.org/linux/packages/fedora/7/
Talk about duplicated effort. ;) Why don't you submit your package (perhaps
after merging with mine) to rpmfusion?
i'd be happy if any repo will contain it and i don't have to recompile.
everybody can use my spec in any repo. the main reason that i use it
(ie. i've truecrypt volumes) and i always need the latest kmod for the
latest fedora kernel while most repo is in a few days delay.
--
Levente "Si vis pacem para bellum!"
Farkas Levente
2007-09-26 09:21:16 UTC
Permalink
Post by Dominik 'Rathann' Mierzejewski
Post by Farkas Levente
Post by Luke Macken
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
http://www.lfarkas.org/linux/packages/fedora/7/
Talk about duplicated effort. ;) Why don't you submit your package (perhaps
after merging with mine) to rpmfusion?
i'd be happy if any repo will contain it and i don't have to recompile.
everybody can use my spec in any repo. the main reason that i use it
(ie. i've truecrypt volumes) and i always need the latest kmod for the
latest fedora kernel while most repo is in a few days delay.
--
Levente "Si vis pacem para bellum!"
Dominik 'Rathann' Mierzejewski
2007-09-26 06:33:33 UTC
Permalink
Post by Farkas Levente
Post by Luke Macken
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
http://www.lfarkas.org/linux/packages/fedora/7/
Talk about duplicated effort. ;) Why don't you submit your package (perhaps
after merging with mine) to rpmfusion?

Regards,
R.
--
Fedora contributor http://fedoraproject.org/wiki/DominikMierzejewski
Livna contributor http://rpm.livna.org MPlayer developer http://mplayerhq.hu
"Faith manages."
-- Delenn to Lennier in Babylon 5:"Confessions and Lamentations"
Dominik 'Rathann' Mierzejewski
2007-09-25 19:43:36 UTC
Permalink
Post by Luke Macken
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
hybrid of a bunch of different licenses and has some interesting
distribution clauses. IANAL, so we definitely would want to run it by
fedora-legal-list if we hope to get truecrypt into Fedora. There has also
http://lists.debian.org/debian-legal/2006/06/msg00294.html
http://lists.debian.org/debian-legal/2006/07/msg00008.html
luke
[0]: http://www.truecrypt.org/license.php
Here are src.rpms with tools and kmod:
http://rpm.greysector.net/livna/

Feel free to submit them to livna/rpmfusion. I don't have time to maintain
them, but I made those when I was looking for a cryptofs solution for both
Linux and Windows (and settled on LUKS instead).

Regards,
R.
--
Fedora contributor http://fedoraproject.org/wiki/DominikMierzejewski
Livna contributor http://rpm.livna.org MPlayer developer http://mplayerhq.hu
"Faith manages."
-- Delenn to Lennier in Babylon 5:"Confessions and Lamentations"
Farkas Levente
2007-09-25 21:36:17 UTC
Permalink
Post by Luke Macken
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
you can always found the latest kmod in:
http://www.lfarkas.org/linux/packages/fedora/7/
--
Levente "Si vis pacem para bellum!"
Neal Becker
2007-09-25 16:55:30 UTC
Permalink
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
it says:

Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Jima
2007-09-25 17:07:29 UTC
Permalink
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Don't we BR kernel-devel when building kmods anyway? I don't think this
is a problem at all. (Aside from the possibility of kmod packaging going
anyway, anyway.)

Jima
Luke Macken
2007-09-25 18:39:07 UTC
Permalink
Post by Neal Becker
Has anyone looked at truecrypt? Looks interesting. I D/L the source, but
Note that Linux kernel headers, located in the 'include' directory, are
not sufficient for compilation of the TrueCrypt kernel module. Fields of
'dm_dev' structure must be accessed by TrueCrypt but they are defined only
in
an internal kernel header 'drivers/md/dm.h'. No appropriate accessor
function
is available. The complete source code of the Linux kernel is required for
compilation of the kernel module.
Truecrypt is awesome, but their license[0] is a bit questionable. It's a
hybrid of a bunch of different licenses and has some interesting
distribution clauses. IANAL, so we definitely would want to run it by
fedora-legal-list if we hope to get truecrypt into Fedora. There has also
been some talk about this license in the Debian community:

http://lists.debian.org/debian-legal/2006/06/msg00294.html
http://lists.debian.org/debian-legal/2006/07/msg00008.html

luke

[0]: http://www.truecrypt.org/license.php
Loading...