[PATCH 1/8] server: Introduce kernel_object struct for generic association between server and kernel objects.

[Jacek Caban]

Signed-off-by: Jacek Caban 
---

This is a version of my previous patchset that was changed to allow 
association only with object types that support it. It adds support for 
event objects as an example, more object types are in my queue.

I hit the problem while debugging MRAC anti-cheat module, but it solves 
a more general problem of kernel object limitations. Bugs 46804, 46205, 
45844 and 29460 should be straightforward on top of this series.

server/async.c      |  2 ++
  server/atom.c       |  1 +
  server/change.c     |  1 +
  server/clipboard.c  |  1 +
  server/completion.c |  1 +
  server/console.c    |  3 +++
  server/debugger.c   |  2 ++
  server/device.c     | 57 ++++++++++++++++++++++++++++++++++++++++++---
  server/directory.c  |  2 ++
  server/event.c      |  2 ++
  server/fd.c         |  4 ++++
  server/file.c       |  1 +
  server/handle.c     |  1 +
  server/hook.c       |  1 +
  server/mailslot.c   |  3 +++
  server/mapping.c    |  3 +++
  server/mutex.c      |  1 +
  server/named_pipe.c |  5 ++++
  server/object.c     |  1 +
  server/object.h     |  4 ++++
  server/process.c    |  3 +++
  server/queue.c      |  2 ++
  server/registry.c   |  1 +
  server/request.c    |  1 +
  server/semaphore.c  |  1 +
  server/serial.c     |  1 +
  server/signal.c     |  1 +
  server/snapshot.c   |  1 +
  server/sock.c       |  2 ++
  server/symlink.c    |  1 +
  server/thread.c     |  2 ++
  server/timer.c      |  1 +
  server/token.c      |  1 +
  server/winstation.c |  2 ++
  34 files changed, 113 insertions(+), 3 deletions(-)

[Marvin]

Hi,

While running your changed tests, I think I found new failures.
Being a bot and all I'm not very good at pattern recognition, so I might be
wrong, but could you please double-check?

Full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=49449

Your paranoid android.

=== debian9b (build log) ===

Task errors:
Unable to set the VM system time: the "nc -q0 '10.42.42.143' '4243'" command returned 1 (settime/connect). Maybe the TestAgentd process is missing the required privileges.

[Alexandre Julliard]

Jacek Caban  writes:
> Signed-off-by: Jacek Caban
>
> This is a version of my previous patchset that was changed to allow
> association only with object types that support it. It adds support
> for event objects as an example, more object types are in my queue.
>
> I hit the problem while debugging MRAC anti-cheat module, but it
> solves a more general problem of kernel object limitations. Bugs
> 46804, 46205, 45844 and 29460 should be straightforward on top of this
> series.

In general the series looks good.
> - /* return list of kernel objects */
> - struct list *(*get_kernel_object)(struct object *);

I found it a bit confusing that get_kernel_object() doesn't return an
object, but a list. Something like get_kernel_obj_list() would be
clearer.

[Jacek Caban]

On 3/25/19 12:02 PM, Alexandre Julliard wrote:
> Jacek Caban  writes:
> > Signed-off-by: Jacek Caban
> >
> > This is a version of my previous patchset that was changed to allow
> > association only with object types that support it. It adds support
> > for event objects as an example, more object types are in my queue.
> >
> > I hit the problem while debugging MRAC anti-cheat module, but it
> > solves a more general problem of kernel object limitations. Bugs
> > 46804, 46205, 45844 and 29460 should be straightforward on top of this
> > series.
>
> In general the series looks good.

Thanks for the review. I will send a new version.

Thanks,

Jacek