Re: [ros-dev] [ros-diffs] [akhaldi] 53527: [WIN32K] * Remove some useless asserts considering cLockObj is DWORD.

[Alex Ionescu]

These asserts are not useless.

A DWORD can also underflow (due to a bug), hence the point of these ASSERTs
-- the only bug is that the ASSERT should cast to LONG.

Best regards,
Alex Ionescu


On Fri, Sep 2, 2011 at 9:23 AM, <akhaldi(a)svn.reactos.org> wrote:

>   Author: akhaldi
>  Date: Fri Sep  2 09:23:20 2011
>  New Revision: 53527
>
>  URL: http://svn.reactos.org/svn/reactos?rev=53527&view=rev
>  Log:
>  [WIN32K]
>  * Remove some useless asserts considering cLockObj is DWORD.
>
>  Modified:
>     trunk/reactos/subsystems/win32/win32k/ntuser/menu.c
>     trunk/reactos/subsystems/win32/win32k/ntuser/monitor.c
>     trunk/reactos/subsystems/win32/win32k/ntuser/window.c
>
>  Modified: trunk/reactos/subsystems/win32/win32k/ntuser/menu.c
>  URL:
>
> http://svn.reactos.org/svn/reactos/trunk/reactos/subsystems/win32/win32k/nt…
>
>  ==============================================================================
>  --- trunk/reactos/subsystems/win32/win32k/ntuser/menu.c [iso-8859-1]
>  (original)
>  +++ trunk/reactos/subsystems/win32/win32k/ntuser/menu.c [iso-8859-1] Fri
>  Sep  2 09:23:20 2011
>  @@ -74,7 +74,6 @@
>        return NULL;
>     }
>
>  -   ASSERT(Menu->head.cLockObj >= 0);
>     return Menu;
>   }
>
>  @@ -137,11 +136,8 @@
>   {
>     PMENU_OBJECT Menu = UserGetMenuObject(hMenu);
>     if (Menu)
>  -   {
>  -      ASSERT(Menu->head.cLockObj >= 0);
>  -
>        Menu->head.cLockObj++;
>  -   }
>  +
>     return Menu;
>   }
>
>
>  Modified: trunk/reactos/subsystems/win32/win32k/ntuser/monitor.c
>  URL:
>
> http://svn.reactos.org/svn/reactos/trunk/reactos/subsystems/win32/win32k/nt…
>
>  ==============================================================================
>  --- trunk/reactos/subsystems/win32/win32k/ntuser/monitor.c [iso-8859-1]
>  (original)
>  +++ trunk/reactos/subsystems/win32/win32k/ntuser/monitor.c [iso-8859-1] Fri
>  Sep  2 09:23:20 2011
>  @@ -93,8 +93,6 @@
>          return NULL;
>      }
>
>  -    ASSERT(Monitor->head.cLockObj >= 0);
>  -
>      return Monitor;
>   }
>
>
>  Modified: trunk/reactos/subsystems/win32/win32k/ntuser/window.c
>  URL:
>
> http://svn.reactos.org/svn/reactos/trunk/reactos/subsystems/win32/win32k/nt…
>
>  ==============================================================================
>  --- trunk/reactos/subsystems/win32/win32k/ntuser/window.c [iso-8859-1]
>  (original)
>  +++ trunk/reactos/subsystems/win32/win32k/ntuser/window.c [iso-8859-1] Fri
>  Sep  2 09:23:20 2011
>  @@ -61,11 +61,8 @@
>
>     Window = UserGetWindowObject(hWnd);
>     if (Window)
>  -   {
>  -      ASSERT(Window->head.cLockObj >= 0);
>  -
>        Window->head.cLockObj++;
>  -   }
>  +
>     return Window;
>   }
>
>  @@ -86,8 +83,6 @@
>        EngSetLastError(ERROR_INVALID_WINDOW_HANDLE);
>        return NULL;
>     }
>  -
>  -   ASSERT(Window->head.cLockObj >= 0);
>
>     return Window;
>   }
>
>
>

[Timo Kreuzer]

Am 02.09.2011 13:05, schrieb Alex Ionescu:
>   These asserts are not useless.
>
>  A DWORD can also underflow (due to a bug), hence the point of these 
>  ASSERTs -- the only bug is that the ASSERT should cast to LONG.
>
>  Best regards,
>  Alex Ionescu
>
>  
They were in the complete wrong place then.

[James Tabor]

It's okay, the check is handled with referencing in object.c.

On Fri, Sep 2, 2011 at 7:22 AM, Timo Kreuzer <timo.kreuzer(a)web.de> wrote:
>   Am 02.09.2011 13:05, schrieb Alex Ionescu:
> >  
> >  These asserts are not useless.
> >
> >  A DWORD can also underflow (due to a bug), hence the point of these
> >  ASSERTs -- the only bug is that the ASSERT should cast to LONG.
> >
> >  Best regards,
> >  Alex Ionescu
> >
> >  
>  They were in the complete wrong place then.
>
>
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev
>