xhotplugd ???

xhotplugd ???

[kde]

[-- Attachment #1: Type: text/plain, Size: 313 bytes --]

Hi 
 
Has anywork been started concerning the xhotplugd ?  
I'm starting to integrate something similiar as the proposed thing into 
KDE, but I don't want to reinvent the wheel. Is there already some code I 
code reuse or someone who I could cooperate with ? 
 
Kind regards 
Joseph Wenninger 
 
 
 
 

Re: xhotplugd ???

[Bernhard Kaindl]

> Hi
>
> Has anywork been started concerning the xhotplugd ?
> I'm starting to integrate something similiar as the proposed thing into
> KDE, but I don't want to reinvent the wheel. Is there already some code I
> code reuse or someone who I could cooperate with ?

I don't know anything about the general idea, but I've found GPL code written
by Matthias Elter <elter@kde.org> and Matthias Hoelzer-Kluepfel, both have been
employed at Caldera in Erlangen for some time. Some other distro seems to have
picked it up and an old rpmfind server has an src.rpm saved:

http://at.rpmfind.net/opsys/linux/RPM/lycoris/build44/install/RPMS/khotplug-0.9.7-42rl.i586.html
ftp://at.rpmfind.net/linux/lycoris/build44/rl/install/SRPMS/khotplug-0.9.7-42rl.src.rpm

Bernd



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: xhotplugd ???

[David Brownell]

> http://at.rpmfind.net/opsys/linux/RPM/lycoris/build44/install/RPMS/khotplug-0.9.7-42rl.i586.html
> ftp://at.rpmfind.net/linux/lycoris/build44/rl/install/SRPMS/khotplug-0.9.7-42rl.src.rpm

I took a quick look at this.  Its "hotplugd" seems to be scanning
busses and delivering its own event reports to a UNIX socket that's
partially hidden (/var/run/.hotplugd).  Then it seems like some KDE
program (which I'll think of as "xhotplugd") interprets those reports.

Some key differences from the kernel hotplug support are:

- Since the kernel is already scanning those busses, this is
   duplicating work ... Yet Another Daemon, in this case one
   that's not necessary.  (Having one was a hotplug anti-goal.)

- It doesn't try to use the MODULE_DEVICE_TABLE information
   provided by drivers ... has its own tables to update (or more
   typically, not update) and get out of sync with the kernel.

- It doesn't even try to use the information provided through
   the /sbin/hotplug interface.  It's got its own event model.

While the user mode code might be fine (but it didn't compile for
me on RH7.3), I think it'd be better to just come up with a way
to have the /sbin/hotplug code write more standard events to such
a UNIX domain socket.

I'll be glad to see progress on hooking up hotplug events to
GUIs.  There are (unfortunately) a number of cases where users
need to make choices when devices appear.

- Dave





-------------------------------------------------------
This sf.net email is sponsored by: Jabber - The world's fastest growing 
real-time communications platform! Don't just IM. Build it in! 
http://www.jabber.com/osdn/xim
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: xhotplugd ???

[Joseph Wenninger]

Hi

On Thursday 25 July 2002 17:25, David Brownell wrote:
> > http://at.rpmfind.net/opsys/linux/RPM/lycoris/build44/install/RPMS/khotpl
> >ug-0.9.7-42rl.i586.html
> > ftp://at.rpmfind.net/linux/lycoris/build44/rl/install/SRPMS/khotplug-0.9.
> >7-42rl.src.rpm
>
> I took a quick look at this.  Its "hotplugd" seems to be scanning
> busses and delivering its own event reports to a UNIX socket that's
> partially hidden (/var/run/.hotplugd).  Then it seems like some KDE
> program (which I'll think of as "xhotplugd") interprets those reports.
>

I looked at that application too, but it looks more like a hack than a real 
solution to me

>
> While the user mode code might be fine (but it didn't compile for
> me on RH7.3), I think it'd be better to just come up with a way
> to have the /sbin/hotplug code write more standard events to such
> a UNIX domain socket.
>

I'm installing an old kde2 now to compile this application

>
> I'll be glad to see progress on hooking up hotplug events to
> GUIs.  There are (unfortunately) a number of cases where users
> need to make choices when devices appear.
>
> - Dave

Kind regards
Joseph Wenninger



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel