[uml-devel] Re: Q: channel file descriptors?

[Heiko Nardmann <h.nardmann@se...>]

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

On Monday 26 November 2001 16:33, Heiko Nardmann wrote:
> This is a technical question concerning the channel mechanism.
>
> Q: during debugging my chipcard reader test application I found that the
> file descriptor in the in channel is 14 and that the file descriptor for
> the out channel is 0. This 0 results in an EPERM (Permission denied) of my
> TIOCMSET ioctl() call to it.
>
> Do I have to pass this ioctl() call to the in file descriptor, too?
>
> The TIOCMGET ioctl() call to the in file descriptor seems to work fine.
>
> BTW: I use 2.4.14-5 UML.

Answering own questions is annoying ...

Okay, I found the CHAN_IN_FD and CHAN_OUT_FD macros ...

- -- 
Heiko Nardmann (Dipl.-Ing.), h.nardmann@..., Software Development
secunet Security Networks AG - Sicherheit in Netzwerken (www.secunet.de),
Weidenauer Str. 223-225, D-57076 Siegen
Tel. : +49 271 48950-13, Fax  : +49 271 48950-50
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE8AmMWpm53PRScYygRAhHEAJ9SH2QXCX26Ble4TpgTDM1tEnzBNwCg4Dcw
jqcaWyq5NCEKdDTUfPtgSys=
=s3LI
-----END PGP SIGNATURE-----

[uml-devel] Q: channel file descriptors?

[Heiko Nardmann <h.nardmann@se...>]

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

This is a technical question concerning the channel mechanism.

Q: during debugging my chipcard reader test application I found that the file 
descriptor in the in channel is 14 and that the file descriptor for the out 
channel is 0. This 0 results in an EPERM (Permission denied) of my TIOCMSET 
ioctl() call to it.

Do I have to pass this ioctl() call to the in file descriptor, too?

The TIOCMGET ioctl() call to the in file descriptor seems to work fine.

BTW: I use 2.4.14-5 UML.

- -- 
Heiko Nardmann (Dipl.-Ing.), h.nardmann@..., Software Development
secunet Security Networks AG - Sicherheit in Netzwerken (www.secunet.de),
Weidenauer Str. 223-225, D-57076 Siegen
Tel. : +49 271 48950-13, Fax  : +49 271 48950-50
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE8AmCxpm53PRScYygRAlEeAJ96ip8aJW6701k0OPhmP/dfVPMwhQCg5lcU
mWG/fKoEDhJgXj0984s5tzc=
=kG3G
-----END PGP SIGNATURE-----

Re: [uml-devel] devfsd for RH root

[William Stearns <wstearns@po...>]

Good day, all,

On Sat, 24 Nov 2001, Jeff Dike wrote:

> mcr@... said:
> > Can someone tell me how these images were generated?
> 
> Bill Stearns did almost all of the RPM-based ones, I assume using mkrootfs.

	Guilty as charged...
	Jeff's right; I use the mkrootfs script.  I keep the master copy 
of the mkrootfs, the functions library it calls, and the calling script 
(updateme) that creates different root filesystems at 
http://www.stearns.org/mkrootfs/
	Likewise, the devfsd rpms (src, i386, and spec) that I create are 
at http://www.stearns.org/devfsd/ .  However, I don't have any record of 
making a devfsd later than 1.3.18.  While rh7.2 doesn't seem to include 
one (it may be renamed or part of some other package), rh7.1 has a 
devfsd-2.4.2-2.i386.rpm.
	Cheers,
	- Bill

---------------------------------------------------------------------------
	"Absence of evidence is not evidence of absence."
	-- SETI, the Search for Extra-Terrestrial Intelligence
--------------------------------------------------------------------------
William Stearns (wstearns@...).  Mason, Buildkernel, named2hosts, 
and ipfwadm2ipchains are at:                http://www.pobox.com/~wstearns
LinuxMonth; articles for Linux Enthusiasts! http://www.linuxmonth.com
--------------------------------------------------------------------------

Re: [uml-devel] devfsd for RH root

[Jeff Dike <jdike@ka...>]

mcr@... said:
> I got devfs turned off, I mean.. it didn't cure /bin/login, but I'll
> try again.

BTW, are you unable to log in to the stock RH 6.2 rootfs, or did you fiddle it
and are now unable to log in to it?

				Jeff

Re: [uml-devel] devfsd for RH root

[Michael Richardson <mcr@sa...>]

>>>>> "Jeff" == Jeff Dike <jdike@...> writes:

    Jeff> mcr@... said:
    >> Hmm. that could be. I did try turning devfs off at point, but that
    >> didn't help.

    Jeff> Either build without devfs configured or put 'devfs=nomount' on the
    Jeff> command line...

  I got devfs turned off, I mean.. it didn't cure /bin/login, but I'll try again.

]       ON HUMILITY: to err is human. To moo, bovine.           |  firewalls  [
]   Michael Richardson, Sandelman Software Works, Ottawa, ON    |net architect[
] mcr@... http://www.sandelman.ottawa.on.ca/ |device driver[
] panic("Just another NetBSD/notebook using, kernel hacking, security guy");  [

[uml-devel] 2.4.15-2um

[Jeff Dike <jdike@ka...>]

There is some support for SIGWINCH.  If the main console is stdin/stdout
and you resize that window, whatever is running in it will get a SIGWINCH.
Other things, like other xterms and ptys, won't generate SIGWINCH for UML
because they are not controlling terminals for it.

Ptys are now hung up properly.  So, disconnecting from a console or serial
line on a pty will no longer hang UML.

A couple of problems with rebooting were fixed.  An overlap between errno
and some code, causing that code to be corrupted, was fixed.  The new
channel stuff was accidentally closing stdin and stdout on shutdown, making
the subsequent reboot rather interesting.

The error message from the TUN/TAP backend was fixed.

Fixed a crash caused by the gdb cleanup when gdb hadn't run.

The window IDs in xterms are now gone.

It's available as always at
		http://user-mode-linux.sourceforge.net/dl-sf.html

				Jeff

Re: [uml-devel] devfsd for RH root

[Jeff Dike <jdike@ka...>]

mcr@... said:
> Hmm. that could be. I did try turning devfs off at point, but that
> didn't help.

Either build without devfs configured or put 'devfs=nomount' on the
command line...

				Jeff

Re: [uml-devel] devfsd for RH root

[Michael Richardson <mcr@sa...>]

>>>>> "Jeff" == Jeff Dike <jdike@...> writes:
    Jeff> mcr@... said:
    >> I've never heard of that command.
    >> I'm mostly interested in how the contents were arrived at. 

    Jeff> /usr/bin/mkrootfs

    Jeff> There are RPM lists in there.

  It's a UML thing... aha.. I looked on a box that doesn't not yet UML
installed :-)
  The lists appear 6.2 specific.

    Jeff> If your login is expecting more traditional tty names, then mknod them in
    Jeff> /dev/, fix inittab to use them, and boot UML with devfs off.

  Hmm. that could be. I did try turning devfs off at point, but that didn't
help.

]       ON HUMILITY: to err is human. To moo, bovine.           |  firewalls  [
]   Michael Richardson, Sandelman Software Works, Ottawa, ON    |net architect[
] mcr@... http://www.sandelman.ottawa.on.ca/ |device driver[
] panic("Just another NetBSD/notebook using, kernel hacking, security guy");  [

Re: [uml-devel] devfsd for RH root

[Jeff Dike <jdike@ka...>]

mcr@... said:
>   I've never heard of that command.
>   I'm mostly interested in how the contents were arrived at. 

/usr/bin/mkrootfs

There are RPM lists in there.

If your login is expecting more traditional tty names, then mknod them in
/dev/, fix inittab to use them, and boot UML with devfs off.

				Jeff

Re: [uml-devel] devfsd for RH root

[Michael Richardson <mcr@sa...>]

>>>>> "Jeff" == Jeff Dike <jdike@...> writes:
    Jeff> mcr@... said:
    >> Can someone tell me how these images were generated?

    Jeff> Bill Stearns did almost all of the RPM-based ones, I assume using mkrootfs.

  I've never heard of that command.
  I'm mostly interested in how the contents were arrived at.

  I can boot RH7.2 image single user, but /bin/login is not cooperating.

]       ON HUMILITY: to err is human. To moo, bovine.           |  firewalls  [
]   Michael Richardson, Sandelman Software Works, Ottawa, ON    |net architect[
] mcr@... http://www.sandelman.ottawa.on.ca/ |device driver[
] panic("Just another NetBSD/notebook using, kernel hacking, security guy");  [