2005-01-03 19:21:35

by René Rebe

[permalink] [raw]
Subject: Please remove hpusbscsi Was: [sane-devel] HP 7450C, hpusbscsi, permissions in Fedora Core 3

Hi,

we should remove hpusbscsi from the Kernel. It is long obsolete and very
unstable. I can send a patch for 2.4 and 2.6 (if needed) ;-)

George Garvey wrote:

> On Sat, Jan 01, 2005 at 10:11:30PM -0800, Thomas Frayne wrote:

...

> As far as I know, Rene has made it pretty clear he doesn't want to
> use hpusbscsi with the avision driver. He prefers libusb.

Yes. Hpusbscsi has many drawbacks. The major ones are:

- does not work with new scanners (that are designed for USB 2.0)
- it is highly instable (e.g. during an i/o error it locks up
quite easily and leaves the (usb sub-)system in a state that
needs a reboot ...

The later problem made me add the user-space i/o code to the
SANE/Avision backend, because I had to reboot my system every 5 minutes
during development ...

Yours,

--
Ren? Rebe - Rubensstr. 64 - 12157 Berlin (Europe / Germany)
http://www.exactcode.de | http://www.exactcode.de/t2
+49 (0)30 255 897 45


2005-01-03 20:04:19

by Oliver Neukum

[permalink] [raw]
Subject: Re: Please remove hpusbscsi Was: [sane-devel] HP 7450C, hpusbscsi, permissions in Fedora Core 3

Am Montag, 3. Januar 2005 20:15 schrieb Rene Rebe:
> Hi,
>
> we should remove hpusbscsi from the Kernel. It is long obsolete and very
> unstable. I can send a patch for 2.4 and 2.6 (if needed) ;-)

I was waiting for 2.7 to do that. The help file for the scanner carries
a big fat warning. Apparently nobody really cares.

Regards
Oliver

2005-01-04 16:47:42

by Till Kamppeter

[permalink] [raw]
Subject: Re: Please remove hpusbscsi Was: [sane-devel] HP 7450C, hpusbscsi, permissions in Fedora Core 3

I have reported this to our Mandrakesoft kernel guys, so next
Mandrakelinux version (10.2) should not have this problem any more.

http://qa.mandrakesoft.com/show_bug.cgi?id=12891

For now, simply remove or rename the module on your system.

Till


Rene Rebe wrote:
> Hi,
>
> we should remove hpusbscsi from the Kernel. It is long obsolete and very
> unstable. I can send a patch for 2.4 and 2.6 (if needed) ;-)
>
> George Garvey wrote:
>
>> On Sat, Jan 01, 2005 at 10:11:30PM -0800, Thomas Frayne wrote:
>
>
> ...
>
>> As far as I know, Rene has made it pretty clear he doesn't want to
>> use hpusbscsi with the avision driver. He prefers libusb.
>
>
> Yes. Hpusbscsi has many drawbacks. The major ones are:
>
> - does not work with new scanners (that are designed for USB 2.0)
> - it is highly instable (e.g. during an i/o error it locks up
> quite easily and leaves the (usb sub-)system in a state that
> needs a reboot ...
>
> The later problem made me add the user-space i/o code to the
> SANE/Avision backend, because I had to reboot my system every 5 minutes
> during development ...
>
> Yours,
>

2005-01-04 17:00:58

by Marcelo Tosatti

[permalink] [raw]
Subject: Re: Please remove hpusbscsi Was: [sane-devel] HP 7450C, hpusbscsi, permissions in Fedora Core 3



I would rather prefer leaving it in the tree marked as broken than
completly removing the source, for 2.4 at least?

On Tue, Jan 04, 2005 at 06:48:10PM +0100, Till Kamppeter wrote:
> I have reported this to our Mandrakesoft kernel guys, so next
> Mandrakelinux version (10.2) should not have this problem any more.
>
> http://qa.mandrakesoft.com/show_bug.cgi?id=12891
>
> For now, simply remove or rename the module on your system.


> Till
>
>
> Rene Rebe wrote:
> >Hi,
> >
> >we should remove hpusbscsi from the Kernel. It is long obsolete and very
> >unstable. I can send a patch for 2.4 and 2.6 (if needed) ;-)
> >
> >George Garvey wrote:
> >
> >>On Sat, Jan 01, 2005 at 10:11:30PM -0800, Thomas Frayne wrote:
> >
> >
> >...
> >
> >> As far as I know, Rene has made it pretty clear he doesn't want to
> >>use hpusbscsi with the avision driver. He prefers libusb.
> >
> >
> >Yes. Hpusbscsi has many drawbacks. The major ones are:
> >
> > - does not work with new scanners (that are designed for USB 2.0)
> > - it is highly instable (e.g. during an i/o error it locks up
> > quite easily and leaves the (usb sub-)system in a state that
> > needs a reboot ...
> >
> >The later problem made me add the user-space i/o code to the
> >SANE/Avision backend, because I had to reboot my system every 5 minutes
> >during development ...
> >
> >Yours,
> >