headers
Kristof Vansant | 1 Sep 2004 02:24
Picon

Re: initial support for video devices and printer ports

On Wed, 2004-09-01 at 00:59, Kay Sievers wrote:
> Shouldn't a printer be a subdevice of the "printer_port" and not merged
> into it? 
> 
It is possible to have a parallel scanner and printer on one parallel
port if I'm not mistaking.
So printer should be a subdevice of the printer_port! and the scanner
too.

> Kay
> 
> _______________________________________________
> hal mailing list
> hal <at> freedesktop.org
> http://freedesktop.org/mailman/listinfo/hal
--

-- 
lupusBE (Kristof Vansant Belgium
Permalink | Reply | 5 comments |
headers
Kay Sievers | 1 Sep 2004 12:25

Re: initial support for video devices and printer ports

On Wed, 2004-09-01 at 02:24 +0200, Kristof Vansant wrote:
> On Wed, 2004-09-01 at 00:59, Kay Sievers wrote:
> > Shouldn't a printer be a subdevice of the "printer_port" and not merged
> > into it? 
> > 
> It is possible to have a parallel scanner and printer on one parallel
> port if I'm not mistaking.
> So printer should be a subdevice of the printer_port! and the scanner
> too.

Yes, there a multi-function devices with a scanner and a printer in the
same box too. Don't know how the actual device is to be selected on the
parport, but they are connected at the same time to the same port.

Today we see such kind of devices all connected through USB, but I still
think the recognized printer should be a child of the parport and not
merged into the interface.

Kay
Permalink | Reply | 1 comment |
headers
Kristof Vansant | 1 Sep 2004 13:05
Picon

Re: initial support for video devices and printer ports

On Wed, 2004-09-01 at 12:25, Kay Sievers wrote:
> On Wed, 2004-09-01 at 02:24 +0200, Kristof Vansant wrote:
> > On Wed, 2004-09-01 at 00:59, Kay Sievers wrote:
> > > Shouldn't a printer be a subdevice of the "printer_port" and not merged
> > > into it? 
> > > 
> > It is possible to have a parallel scanner and printer on one parallel
> > port if I'm not mistaking.
> > So printer should be a subdevice of the printer_port! and the scanner
> > too.
> 
> Yes, there a multi-function devices with a scanner and a printer in the
> same box too. Don't know how the actual device is to be selected on the
> parport, but they are connected at the same time to the same port.
> 
> Today we see such kind of devices all connected through USB, but I still
> think the recognized printer should be a child of the parport and not
> merged into the interface.

ah I thought that child == subdevice :) so yes, best is to have it as
child

> 
> Kay
--

-- 
lupusBE (Kristof Vansant Belgium
Permalink | Reply | 0 comments |
headers
Miek Gieben | 1 Sep 2004 19:36
Picon
Favicon

Re: hal and usb mouse

[On 26 Aug,  <at>  01:40, David wrote in "Re: hal and usb mouse ..."]
> On Tue, 2004-08-24 at 10:56 +0200, Miek Gieben wrote:
> > hal 0.2.97
> > dbus-1 0.22
> > udev 0.030
> > 
> 
> Hmm, this works for me, has done for some time, try killing the hal
> daemon (pkill hald) and run 'hald --daemon=no' as root and post the
> output if it doesn't work.

sorry for my late reply - I was on holiday.

With the debian version of the packages I get:

Remove:
[W] linux/osspec.c:541 remove_class_device() : Removal of class device  <at>  /sys/class/input/mouse0 on
hotplug remove is not yet implemented
[W] linux/osspec.c:501 remove_device() : Couldn't remove device  <at> 
/sys/devices/pci0000:00/0000:00:11.2/usb1/1-1/1-1.3/1-1.3:1.0 on hotplug remove
[W] linux/osspec.c:501 remove_device() : Couldn't remove device  <at> 
/sys/devices/pci0000:00/0000:00:11.2/usb1/1-1/1-1.3 on hotplug remove

Plug in:
=== Long time nothing ==== (mouse is already working btw)
[W] linux/class_device.c:353 class_device_got_sysdevice() : Sysdevice for a class device never appeared!

Looks like the /sys entry is not made/found.

> Also try using hal from CVS and run as root
(Continue reading)

Permalink | Reply | 0 comments |
headers
David Zeuthen | 1 Sep 2004 19:47
Picon
Gravatar

Re: [PATCH] export ieee1394.guid as string


Hi,

Sorry for the delay,

On Fri, 2004-08-27 at 00:45 +0200, Jon Lech Johansen wrote:
> On Thu, Aug 26, 2004 at 02:24:28AM +0200, Jon Lech Johansen (jon <at> nanocrew.net) wrote:
> > I'll send a new patch tomorrow.
> 
> Patch is attached. It breaks hal-device-manager, but lshal works. I
> haven't looked into why as I'm not familiar with python.
> 

Looks good, I've committed this but uncommented the code that use uint64
for now. I also added a property for the mac address for ethernet
devices but this is also uncommented.

The reason h-d-m breaks is that the python dbus bindings doesn't dig
int64 and uint64. I've cooked up a patch for that here

 http://freedesktop.org/pipermail/dbus/2004-September/001543.html

so when this is committed I'll enable the code using uint64 in hal.

Thanks a lot
David
Permalink | Reply | 0 comments |
headers
Bill Nottingham | 1 Sep 2004 20:30
Picon
Favicon
Gravatar

Re: initial support for video devices and printer ports

Kristof Vansant (de_lupus <at> pandora.be) said: 
> On Wed, 2004-09-01 at 00:59, Kay Sievers wrote:
> > Shouldn't a printer be a subdevice of the "printer_port" and not merged
> > into it? 
> > 
> It is possible to have a parallel scanner and printer on one parallel
> port if I'm not mistaking.

Not to mention parallel port disks. :)

Bill
Permalink | Reply | 2 comments |
headers
Kay Sievers | 1 Sep 2004 20:38

Re: initial support for video devices and printer ports

On Wed, 2004-09-01 at 14:30 -0400, Bill Nottingham wrote:
> Kristof Vansant (de_lupus <at> pandora.be) said: 
> > On Wed, 2004-09-01 at 00:59, Kay Sievers wrote:
> > > Shouldn't a printer be a subdevice of the "printer_port" and not merged
> > > into it? 
> > > 
> > It is possible to have a parallel scanner and printer on one parallel
> > port if I'm not mistaking.
> 
> Not to mention parallel port disks. :)

Yeah, on top of that :)
I just read a small part of the IEEE1284 (parallel port) spec. It is
possible to connect 20 devices to _one_ port by daisy-chaining and
multiplexing the devices. All in line with the spec! And all devices are
adressable.
At least two of the five possible devices in a daisy-chain will appear
sometimes in a HAL environment.
So we should do childs for it, if we find a device?

Kay
Permalink | Reply | 1 comment |
headers
David Zeuthen | 1 Sep 2004 21:04
Picon
Gravatar

Re: initial support for video devices and printer ports


Hey Kay,

sorry for the lag,

On Tue, 2004-08-31 at 21:01 +0200, Kay Sievers wrote:
> Attached is a patch to add:
> 
> Support for v4l devices:
>   http://vrfy.org/projects/hal/v4l.png
>   o it supports v4l and v4l2
>     (the v4l2 headers are missing on Fedora, so I copied the one struct in)
>   o we need to define the capabilities ,it's just "video.FIXME" now

This looks good. Suggest to call it media.* though; 'video' sounds to me
like it should be used for display adapter stuff. Btw, is it possible to
distinguish between capture cards and webcams without using some kind of
heuristic?

Aren't some of the v4l[12] devices also for audio or am I mistaken? If
yes, we probably need to have media.video.* and media.audio.*. Should
probably put tuner related stuff into media.tuner.*; this might include
a link to the i2c device it will have as child (e.g. my Bt878 capture
card is also a i2c-adapter).

Perhaps also call it media.linux.v4l_version instead of video.version.
video. video.FIXME could be hal capabilities media.foobar (or
media.video.foobar) for v4l cap foobar and properties media.foobar.baz
for interesting property baz that a v4l device with cap foobar has.
(Continue reading)

Permalink | Reply | 2 comments |
headers
David Zeuthen | 1 Sep 2004 21:29
Picon
Gravatar

Re: initial support for video devices and printer ports

On Wed, 2004-09-01 at 20:38 +0200, Kay Sievers wrote:
> On Wed, 2004-09-01 at 14:30 -0400, Bill Nottingham wrote:
> > Kristof Vansant (de_lupus <at> pandora.be) said: 
> > > On Wed, 2004-09-01 at 00:59, Kay Sievers wrote:
> > > > Shouldn't a printer be a subdevice of the "printer_port" and not merged
> > > > into it? 
> > > > 
> > > It is possible to have a parallel scanner and printer on one parallel
> > > port if I'm not mistaking.
> > 
> > Not to mention parallel port disks. :)
> 
> Yeah, on top of that :)
> I just read a small part of the IEEE1284 (parallel port) spec. It is
> possible to connect 20 devices to _one_ port by daisy-chaining and
> multiplexing the devices. All in line with the spec! And all devices are
> adressable.
> At least two of the five possible devices in a daisy-chain will appear
> sometimes in a HAL environment.
> So we should do childs for it, if we find a device?
> 

Interesting. One of the things I want to put in hal is to add a method
Rescan() that can be invoked on the org.freedesktop.Hal.Device interface
for a given device object. The semantics of this operation will differ,
depending on the capabilities of the device object. We'll also restrict
access so only certain well defined (e.g. the image for the process must
be /usr/bin/blah or something) processes can invoke it.

One thing I want to use this for is to invoke a poll for media on
(Continue reading)

Permalink | Reply | 0 comments |
headers
David Zeuthen | 1 Sep 2004 22:16
Picon
Gravatar

Re: [aschoenebeck <at> web.de: Bug#269239: "program hald is using a deprecated SCSI ioctl, please convert it to SG_IO"]

On Tue, 2004-08-31 at 18:26 +0200, Sjoerd Simons wrote:
> program hald is using a deprecated SCSI ioctl, please convert it to
> SG_IO

This is also reported here

 https://freedesktop.org/bugzilla/show_bug.cgi?id=1190        

I'll look into fixing this.

David
Permalink | Reply | 0 comments |

Gmane