Re: Stack Trace for Ruby Segfaults

Hi,

In <20050623081541.GA29022@jaa.iki.fi>
  "Re: Stack Trace for Ruby Segfaults" on Thu, 23 Jun 2005 02:15:41 -0600,
  Jani Averbach <jaa@jaa.iki.fi> wrote:

> I guess that this result is ok, I got one failure:

I agree with you.

> > I will remove apr_pool_t from Ruby world soon. (This is
> > already implemented in my working copy.) The change may fix
> > (or hide) the problem. Just a moment, please.
>
> I will keep svntest/ruby turned on in the meantime, if you like?

I committed in r15154.
But I want you to check the situation before update the
revision.

* Whether segmentation fault is occurred or not when ruby's GC
  is disabled?

  To do this, could you add 'GC.disable' to
  subversion/bindings/swig/ruby/test/run-test.rb? For
  example:

    +GC.disable
    
     if Test::Unit::AutoRunner.respond_to?(:standalone?)
       exit Test::Unit::AutoRunner.run($0, File.dirname($0))
     else
       exit Test::Unit::AutoRunner.run(false, File.dirname($0))
     end

--
kou
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org