Re: [Gnash-dev] Segmentation fault caused by bug in SafeStack?

[strk]

On Tue, Jul 08, 2008 at 11:16:31AM +0800, Tom Stellard wrote:
> I have fixed the bug that was causing my segmentation fault.  

Do you have a testcase for this specific case ?
I'm preparing a testsuite/server/SafeStackTest.cpp to focus on that.

> but now I am getting a segmentation
> fault in the abc_block::read_instances() function, which might be a
> different bug in SafeStack, but I haven't investigated it yet.

A quick look at the code suggest to me that mDownstop might be the problem.
It's initialized to 1 but .empty() compares it with 0. One of the two must be
broken I guess... Indeed a newly created SafeStack is advertised to have a size
of 1, but it's obvious that the underlying vector is empty!

> I have been using gameday for a few years, and it seems like a good real
> world test case, because they always start to take advantage of new features
> in the flash player as soon as possible.  As far a testing framework, I
> would be able to work on setting something up for AVM2.  Although, I might
> need some help getting started.

For generic classes unit testing is good (testsuite/server/SafeStackTest.cpp
I'll commit shortly, to be improved).
For SWF-specific things we should find/adopt a generator/compiler w/out 
restriction
of use. The ./configure phase would detect it's availability, 'make check' would
create SWF from source and run them, best if self-contained.
This is how we usually do. An alternative is putting the SWF files directly in
the repository, but having the sources greately help adding tests to better 
understand
the model.

--strk;