[erlang-questions] refactoring a very large record

[Matthew Sackman]

On Thu, Oct 20, 2011 at 04:25:28PM +0100, Matthew Sackman wrote:
> Please explain more about this. Why can analysis tell you how many live
> pointers there are to the old record and thus inform the VM whether or
> not it can do update-in-place?

Err, missing negation in there - that was meant to be "why can't..."

Essentially, I'd like to know why live variable analysis isn't used to
permit in-place updates. I would expect that if this were possible then
it would lead to significant speed improvements in large numbers of
places, but I equally suspect there must be something I'm missing which
prevents this approach from working.

Matthew