Anyone else been having sporadic “invalid memory address” errors since this update? (I think the error message is actually slightly different, but I forgot to cut and paste it last time I hit it) I’m getting them maybe one out of a hundred runs on some analysis stuff I’m doing, and it’s happening with different apps and applets, so I don’t think it’s my fault in any way (I’m not getting an exception, we’re talking a full crash, no logs or anything, just an error message and the program shuts down).
That’s a Really Bad Thing, especially since the code that causes it is very innocuous, just pure Java number crunching stuff (create an array, fill it with numbers, do some math, print some results). If Apple has borked memory management in this update, then the problem goes a lot deeper than just an occasional crash, there are probably some serious security holes just begging to be exploited - one of the most important things about Java is the fact that the JVM shouldn’t screw this stuff up, and it makes me very nervous to see an implementation that occasionally does.
Perfect example of why it’s a real terrible idea for Apple to handle its own JVM implementation…
