[ANN] malloc'd ByteArrays

[Brian Rice]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I don't know. Can there be?

In the BUGS file, Tony mentions the need for a counting semaphore for  
ExternalResource objects. That would certainly help with this, IF  
there is some way for the SDL binding can count itself as a user and  
release only when appropriate.

You can code this up if you like. Right now I'm focussed on fixing  
some bugs with bootstrapping.

With that said, I don't know of any way in which the VM can know that  
no pointers refer to a given memory area. It has to be told so. This  
reminds me of a conservative GC.

On Dec 13, 2005, at 6:10 PM, Todd Fleming wrote:

> Is there going to be a way to prevent the user from closing a  
> ByteArray without first closing Cairo surfaces that refer to it?

- --
- -Brian
http://tunes.org/~water/brice.vcf

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (Darwin)

iD8DBQFDn4IaPVkAvLW1zf4RAmWpAJ9i7N0anJYHjCm2BsW4U5/w1pNlFwCfeSQg
DfyZ2j7I92V4VvSYpySEJgI=
=ptZt
-----END PGP SIGNATURE-----