Xojo Conferences
MBSOct2019CologneDE

Question(s) about Failed Assertion (Real Studio Plugins Mailinglist archive)

Back to the thread list
Previous thread: Callbacks and events on Windows
Next thread: Trying to call a class event via a callback


macosx and unix paths   -   GOLD
  Question(s) about Failed Assertion   -   Alfred Van Hoek
   Re: Question(s) about Failed Assertion   -   Jonathan Johnson

Question(s) about Failed Assertion
Date: 24.04.06 02:20 (Sun, 23 Apr 2006 21:20:46 -0400)
From: Alfred Van Hoek
When a plugin is compiled with Xcode 2.2.1 quitting a test app
compiled for Mach-O leads to:

Runtime Error 4: Failed Assertion
Location: dump.cpp 178
Failure Condition: 0
Failure Message: "overwrote end of block"

After the Destructor of the plugin-control is called.

When a plugin is compiled with CW-10, quitting a test app compiled
for Mach-O does not lead to this failed assertion.

Closing the window, instead of quitting: the same behavior. Making
sure that the destructor is not called by internally giving the
control an extra refcount, does not lead to the Assertion.

How to tackle this?

When the control lives on the window it has a refcount of 2, similar
to a canvas. Could be Xcode 2.2.1, anybody else experience this
Assertion?

Alfred
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>

Re: Question(s) about Failed Assertion
Date: 24.04.06 15:42 (Mon, 24 Apr 2006 09:42:03 -0500)
From: Jonathan Johnson

On Apr 23, 2006, at 8:20 PM, Alfred Van Hoek wrote:

> When a plugin is compiled with Xcode 2.2.1 quitting a test app
> compiled for Mach-O leads to:
>
> Runtime Error 4: Failed Assertion
> Location: dump.cpp 178
> Failure Condition: 0
> Failure Message: "overwrote end of block"
>
> After the Destructor of the plugin-control is called.
>
> When a plugin is compiled with CW-10, quitting a test app compiled
> for Mach-O does not lead to this failed assertion.
>
> Closing the window, instead of quitting: the same behavior. Making
> sure that the destructor is not called by internally giving the
> control an extra refcount, does not lead to the Assertion.
>
> How to tackle this?

You have memory corruption going on. At some point, you're
overwriting the memory allocated to your plugin control. REALbasic
checks some guard words at the beginning and end of memory
allocations, and those guard words were changed, causing this assertion.

> When the control lives on the window it has a refcount of 2,
> similar to a canvas. Could be Xcode 2.2.1, anybody else experience
> this Assertion?

I doubt it's Xcode. It could be that your plugin control's data
structure is different-sized in Xcode than it is in CodeWarrior. Of
course, that would only be an issue if you have both a PEF and Mach-O
part.

HTH,
Jon
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>