All of lore.kernel.org
 help / color / mirror / Atom feed
* Kernel user communication mechanism
@ 2011-12-14 11:44 rohan puri
  2011-12-15 12:24 ` Konstantin Zertsekel
  0 siblings, 1 reply; 5+ messages in thread
From: rohan puri @ 2011-12-14 11:44 UTC (permalink / raw)
  To: kernelnewbies

Hello Guys,

I want to implement a kernel-user communication mechanism in which kernel
can send events to user space (both blocking and non-blocking). Blocking
events would require response which the kernel will interpret. Also
user-space can also send the events to the kernel and get the response.

Was having a look at the netlink sockets. What will be the behaviour for
sending an event which is blocking on the response from user-space the
queue is full. How this case will be handled.

Also fanotify makes use of fsnotify's notification mechanism to send events
and get response from kernel. Is it possible to send event from user-space
to kernel space in -case of fanotify.

I there any other way present which is better than these two above ways.

Any ideas to implement a new mechanism are also appreciated.

Regards,
Rohan Puri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20111214/615f8747/attachment-0001.html 

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Kernel user communication mechanism
  2011-12-14 11:44 Kernel user communication mechanism rohan puri
@ 2011-12-15 12:24 ` Konstantin Zertsekel
  2011-12-16  5:26   ` rohan puri
  0 siblings, 1 reply; 5+ messages in thread
From: Konstantin Zertsekel @ 2011-12-15 12:24 UTC (permalink / raw)
  To: kernelnewbies

> Was having a look at the netlink sockets. What will be the behaviour for
> sending an event which is blocking on the response from user-space the queue
> is full. How this case will be handled.
>
> Also fanotify makes use of fsnotify's notification mechanism to send events
> and get response from kernel. Is it possible to send event from user-space
> to kernel space in -case of fanotify.

Why is it interesting?
What is the story behind?
--- Kosta

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Kernel user communication mechanism
  2011-12-15 12:24 ` Konstantin Zertsekel
@ 2011-12-16  5:26   ` rohan puri
  2011-12-16 15:19     ` Greg KH
  0 siblings, 1 reply; 5+ messages in thread
From: rohan puri @ 2011-12-16  5:26 UTC (permalink / raw)
  To: kernelnewbies

On Thu, Dec 15, 2011 at 5:54 PM, Konstantin Zertsekel
<zertsekel@gmail.com>wrote:

> > Was having a look at the netlink sockets. What will be the behaviour for
> > sending an event which is blocking on the response from user-space the
> queue
> > is full. How this case will be handled.
> >
> > Also fanotify makes use of fsnotify's notification mechanism to send
> events
> > and get response from kernel. Is it possible to send event from
> user-space
> > to kernel space in -case of fanotify.
>
> Why is it interesting?
> What is the story behind?
> --- Kosta
>

I need to move the IP in user-space which currently is as a proprietary
module and release the kernel component as GPL.

So for decision making I require this communication.

Any help in terms of ideas or existing solutions would be appreciable.

Also Greg please if you could help me out :)

Regards,
Rohan Puri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20111216/8e54d5eb/attachment.html 

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Kernel user communication mechanism
  2011-12-16  5:26   ` rohan puri
@ 2011-12-16 15:19     ` Greg KH
  2011-12-16 16:17       ` rohan puri
  0 siblings, 1 reply; 5+ messages in thread
From: Greg KH @ 2011-12-16 15:19 UTC (permalink / raw)
  To: kernelnewbies

On Fri, Dec 16, 2011 at 10:56:16AM +0530, rohan puri wrote:
> 
> 
> On Thu, Dec 15, 2011 at 5:54 PM, Konstantin Zertsekel <zertsekel@gmail.com>
> wrote:
> 
>     > Was having a look at the netlink sockets. What will be the behaviour for
>     > sending an event which is blocking on the response from user-space the
>     queue
>     > is full. How this case will be handled.
>     >
>     > Also fanotify makes use of fsnotify's notification mechanism to send
>     events
>     > and get response from kernel. Is it possible to send event from
>     user-space
>     > to kernel space in -case of fanotify.
> 
>     Why is it interesting?
>     What is the story behind?
>     --- Kosta
> 
> 
> I need to move the IP in user-space which currently is as a proprietary module
> and release the kernel component as GPL.
> 
> So for decision making I require this communication.

Please consult a lawyer as to if this is really a proper separation that
will hold up to the GPL requirements of the kernel.

good luck,

greg k-h

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Kernel user communication mechanism
  2011-12-16 15:19     ` Greg KH
@ 2011-12-16 16:17       ` rohan puri
  0 siblings, 0 replies; 5+ messages in thread
From: rohan puri @ 2011-12-16 16:17 UTC (permalink / raw)
  To: kernelnewbies

On Fri, Dec 16, 2011 at 8:49 PM, Greg KH <greg@kroah.com> wrote:

> On Fri, Dec 16, 2011 at 10:56:16AM +0530, rohan puri wrote:
> >
> >
> > On Thu, Dec 15, 2011 at 5:54 PM, Konstantin Zertsekel <
> zertsekel at gmail.com>
> > wrote:
> >
> >     > Was having a look at the netlink sockets. What will be the
> behaviour for
> >     > sending an event which is blocking on the response from user-space
> the
> >     queue
> >     > is full. How this case will be handled.
> >     >
> >     > Also fanotify makes use of fsnotify's notification mechanism to
> send
> >     events
> >     > and get response from kernel. Is it possible to send event from
> >     user-space
> >     > to kernel space in -case of fanotify.
> >
> >     Why is it interesting?
> >     What is the story behind?
> >     --- Kosta
> >
> >
> > I need to move the IP in user-space which currently is as a proprietary
> module
> > and release the kernel component as GPL.
> >
> > So for decision making I require this communication.
>
> Please consult a lawyer as to if this is really a proper separation that
> will hold up to the GPL requirements of the kernel.
>
> good luck,
>
> greg k-h
>

Thanks greg for the suggestion. Companies manager are looking into those
legal matters. I as a developer have been told to work on U-K & K-U
communication mechanism. Would love to have your suggestion as to how to go
about it?

Thanks & Regards,
Rohan Puri
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20111216/f694bca5/attachment-0001.html 

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2011-12-16 16:17 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-12-14 11:44 Kernel user communication mechanism rohan puri
2011-12-15 12:24 ` Konstantin Zertsekel
2011-12-16  5:26   ` rohan puri
2011-12-16 15:19     ` Greg KH
2011-12-16 16:17       ` rohan puri

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.