On Sun, Oct 29, 2006 at 10:27:24AM +0100, Marco d'Itri wrote:
> [EMAIL PROTECTED] wrote:
>
> >potential for all kinds of race conditions. I think that the best way
> >to make it more robust is to go back to having a regular /etc/init.d
> >file, which will both run at boot time and can be called
[EMAIL PROTECTED] wrote:
>potential for all kinds of race conditions. I think that the best way
>to make it more robust is to go back to having a regular /etc/init.d
>file, which will both run at boot time and can be called from modprobe
>(in that case it will just do nothing if the filesystem
On Sun, Oct 22, 2006 at 08:57:25PM +0200, Marco d'Itri wrote:
> [EMAIL PROTECTED] wrote:
>
> >> No, I think modprobe directives are more appropriate for this task.
> >> You need a script anyway to change the sysfs permissions, create the
> >> /var/run/ directory, etc. So it can as well wait for th
[EMAIL PROTECTED] wrote:
>> No, I think modprobe directives are more appropriate for this task.
>> You need a script anyway to change the sysfs permissions, create the
>> /var/run/ directory, etc. So it can as well wait for the sysfs files to
>> appear. Let me know if you have more doubts.
>Ok, it
On Sun, Oct 15, 2006 at 11:41:20AM +0200, Marco d'Itri wrote:
> No, I think modprobe directives are more appropriate for this task.
> You need a script anyway to change the sysfs permissions, create the
> /var/run/ directory, etc. So it can as well wait for the sysfs files to
> appear. Let me know
[EMAIL PROTECTED] wrote:
>Thanks for the comments. As I don't have much experience in it, I've
>asked Bastian Blank for advice, and he suggested that it's better to
>do it with udev rules, since then we can be sure that the needed sysfs
>entries have been created by the time the script runs. Ho
On Sat, Oct 14, 2006 at 11:41:48AM +0200, Marco d'Itri wrote:
> >I would appreciate any comments and suggestions regarding packaging,
> >before I upload it.
> The daemon should be started by modprobe (or udev) as a side-effect of
> loading the driver, not unconditionally at boot time and fail if
On Sat, Oct 14, 2006 at 11:41:48AM +0200, Marco d'Itri wrote:
> The daemon should be started by modprobe (or udev) as a side-effect of
> loading the driver, not unconditionally at boot time and fail if the
> driver has not been loaded yet.
That is exactly what I said.
Bastian
--
No one wants wa
[EMAIL PROTECTED] wrote:
>I would appreciate any comments and suggestions regarding packaging,
>before I upload it.
The daemon should be started by modprobe (or udev) as a side-effect of
loading the driver, not unconditionally at boot time and fail if the
driver has not been loaded yet.
Please im
Hi,
I have prepared the ipw3945-daemon package, following the ITP 391686.
The debian subdirectory can be found in kernel team's svn repo at
svn://svn.debian.org/svn/kernel/people/jurij/ipw3945-daemon
The source and binary packages for i386 are available at
http://www.wooyd.org/debian/ipw3945-d
10 matches
Mail list logo
